From c2f0c7c356dc9ae15419f00c725a2fcc58eeff58 Mon Sep 17 00:00:00 2001 From: Steve Grubb Date: Fri, 6 May 2005 12:38:39 +0100 Subject: The attached patch addresses the problem with getting the audit daemon shutdown credential information. It creates a new message type AUDIT_TERM_INFO, which is used by the audit daemon to query who issued the shutdown. It requires the placement of a hook function that gathers the information. The hook is after the DAC & MAC checks and before the function returns. Racing threads could overwrite the uid & pid - but they would have to be root and have policy that allows signalling the audit daemon. That should be a manageable risk. The userspace component will be released later in audit 0.7.2. When it receives the TERM signal, it queries the kernel for shutdown information. When it receives it, it writes the message and exits. The message looks like this: type=DAEMON msg=auditd(1114551182.000) auditd normal halt, sending pid=2650 uid=525, auditd pid=1685 Signed-off-by: Steve Grubb Signed-off-by: David Woodhouse --- include/linux/audit.h | 25 +++++++++++++++++-------- 1 file changed, 17 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index 19f04b049798..baa80760824c 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -28,14 +28,16 @@ #include /* Request and reply types */ -#define AUDIT_GET 1000 /* Get status */ -#define AUDIT_SET 1001 /* Set status (enable/disable/auditd) */ -#define AUDIT_LIST 1002 /* List filtering rules */ -#define AUDIT_ADD 1003 /* Add filtering rule */ -#define AUDIT_DEL 1004 /* Delete filtering rule */ -#define AUDIT_USER 1005 /* Send a message from user-space */ -#define AUDIT_LOGIN 1006 /* Define the login id and informaiton */ -#define AUDIT_KERNEL 2000 /* Asynchronous audit record. NOT A REQUEST. */ +#define AUDIT_GET 1000 /* Get status */ +#define AUDIT_SET 1001 /* Set status (enable/disable/auditd) */ +#define AUDIT_LIST 1002 /* List filtering rules */ +#define AUDIT_ADD 1003 /* Add filtering rule */ +#define AUDIT_DEL 1004 /* Delete filtering rule */ +#define AUDIT_USER 1005 /* Send a message from user-space */ +#define AUDIT_LOGIN 1006 /* Define the login id and information */ +#define AUDIT_SIGNAL_INFO 1010 /* Get information about sender of signal*/ + +#define AUDIT_KERNEL 2000 /* Asynchronous audit record. NOT A REQUEST. */ /* Rule flags */ #define AUDIT_PER_TASK 0x01 /* Apply rule at task creation (not syscall) */ @@ -161,6 +163,11 @@ struct audit_rule { /* for AUDIT_LIST, AUDIT_ADD, and AUDIT_DEL */ #ifdef __KERNEL__ +struct audit_sig_info { + uid_t uid; + pid_t pid; +}; + struct audit_buffer; struct audit_context; struct inode; @@ -190,6 +197,7 @@ extern void audit_get_stamp(struct audit_context *ctx, extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); +extern void audit_signal_info(int sig, struct task_struct *t); #else #define audit_alloc(t) ({ 0; }) #define audit_free(t) do { ; } while (0) @@ -200,6 +208,7 @@ extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mo #define audit_inode(n,i) do { ; } while (0) #define audit_get_loginuid(c) ({ -1; }) #define audit_ipc_perms(q,u,g,m) ({ 0; }) +#define audit_signal_info(s,t) do { ; } while (0) #endif #ifdef CONFIG_AUDIT -- cgit v1.2.3 From 197c69c6afd2deb7eec44040ff533d90d26c6161 Mon Sep 17 00:00:00 2001 From: Chris Wright Date: Wed, 11 May 2005 10:54:05 +0100 Subject: Move ifdef CONFIG_AUDITSYSCALL to header Remove code conditionally dependent on CONFIG_AUDITSYSCALL from audit.c. Move these dependencies to audit.h with the rest. Signed-off-by: Chris Wright Signed-off-by: David Woodhouse --- include/linux/audit.h | 4 +++- kernel/audit.c | 12 ++---------- kernel/auditsc.c | 7 +++---- 3 files changed, 8 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index baa80760824c..58c5589b531f 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -192,7 +192,7 @@ extern void audit_inode(const char *name, const struct inode *inode); /* Private API (for audit.c only) */ extern int audit_receive_filter(int type, int pid, int uid, int seq, void *data, uid_t loginuid); -extern void audit_get_stamp(struct audit_context *ctx, +extern int audit_get_stamp(struct audit_context *ctx, struct timespec *t, unsigned int *serial); extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); @@ -206,6 +206,8 @@ extern void audit_signal_info(int sig, struct task_struct *t); #define audit_getname(n) do { ; } while (0) #define audit_putname(n) do { ; } while (0) #define audit_inode(n,i) do { ; } while (0) +#define audit_receive_filter(t,p,u,s,d,l) ({ -EOPNOTSUPP; }) +#define audit_get_stamp(c,t,s) ({ 0; }) #define audit_get_loginuid(c) ({ -1; }) #define audit_ipc_perms(q,u,g,m) ({ 0; }) #define audit_signal_info(s,t) do { ; } while (0) diff --git a/kernel/audit.c b/kernel/audit.c index dc4aba21f30a..c18b769e23a2 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -416,12 +416,8 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) return -EINVAL; /* fallthrough */ case AUDIT_LIST: -#ifdef CONFIG_AUDITSYSCALL err = audit_receive_filter(nlh->nlmsg_type, NETLINK_CB(skb).pid, uid, seq, data, loginuid); -#else - err = -EOPNOTSUPP; -#endif break; case AUDIT_SIGNAL_INFO: sig_data.uid = audit_sig_uid; @@ -636,15 +632,11 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx) return NULL; } -#ifdef CONFIG_AUDITSYSCALL - if (ab->ctx) - audit_get_stamp(ab->ctx, &t, &serial); - else -#endif - { + if (!audit_get_stamp(ab->ctx, &t, &serial)) { t = CURRENT_TIME; serial = 0; } + audit_log_format(ab, "audit(%lu.%03lu:%u): ", t.tv_sec, t.tv_nsec/1000000, serial); return ab; diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 680bb928343b..94338abf76f5 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -992,7 +992,7 @@ void audit_inode(const char *name, const struct inode *inode) context->names[idx].rdev = inode->i_rdev; } -void audit_get_stamp(struct audit_context *ctx, +int audit_get_stamp(struct audit_context *ctx, struct timespec *t, unsigned int *serial) { if (ctx) { @@ -1000,10 +1000,9 @@ void audit_get_stamp(struct audit_context *ctx, t->tv_nsec = ctx->ctime.tv_nsec; *serial = ctx->serial; ctx->auditable = 1; - } else { - *t = CURRENT_TIME; - *serial = 0; + return 1; } + return 0; } extern int audit_set_type(struct audit_buffer *ab, int type); -- cgit v1.2.3 From c1b773d87eadc3972d697444127e89a7291769a2 Mon Sep 17 00:00:00 2001 From: Chris Wright Date: Wed, 11 May 2005 10:55:10 +0100 Subject: Add audit_log_type Add audit_log_type to allow callers to specify type and pid when logging. Convert audit_log to wrapper around audit_log_type. Could have converted all audit_log callers directly, but common case is default of type AUDIT_KERNEL and pid 0. Update audit_log_start to take type and pid values when creating a new audit_buffer. Move sequences that did audit_log_start, audit_log_format, audit_set_type, audit_log_end, to simply call audit_log_type directly. This obsoletes audit_set_type and audit_set_pid, so remove them. Signed-off-by: Chris Wright Signed-off-by: David Woodhouse --- include/linux/audit.h | 16 ++++++++++------ kernel/audit.c | 48 +++++++++++++++--------------------------------- kernel/auditsc.c | 23 +++++++---------------- security/selinux/avc.c | 2 +- 4 files changed, 33 insertions(+), 56 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index 58c5589b531f..405332ebf3c6 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -216,11 +216,14 @@ extern void audit_signal_info(int sig, struct task_struct *t); #ifdef CONFIG_AUDIT /* These are defined in audit.c */ /* Public API */ -extern void audit_log(struct audit_context *ctx, - const char *fmt, ...) - __attribute__((format(printf,2,3))); +#define audit_log(ctx, fmt, args...) \ + audit_log_type(ctx, AUDIT_KERNEL, 0, fmt, ##args) +extern void audit_log_type(struct audit_context *ctx, int type, + int pid, const char *fmt, ...) + __attribute__((format(printf,4,5))); -extern struct audit_buffer *audit_log_start(struct audit_context *ctx); +extern struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, + int pid); extern void audit_log_format(struct audit_buffer *ab, const char *fmt, ...) __attribute__((format(printf,2,3))); @@ -240,8 +243,9 @@ extern void audit_send_reply(int pid, int seq, int type, void *payload, int size); extern void audit_log_lost(const char *message); #else -#define audit_log(t,f,...) do { ; } while (0) -#define audit_log_start(t) ({ NULL; }) +#define audit_log(c,f,...) do { ; } while (0) +#define audit_log_type(c,t,p,f,...) do { ; } while (0) +#define audit_log_start(c,t,p) ({ NULL; }) #define audit_log_vformat(b,f,a) do { ; } while (0) #define audit_log_format(b,f,...) do { ; } while (0) #define audit_log_end(b) do { ; } while (0) diff --git a/kernel/audit.c b/kernel/audit.c index c18b769e23a2..060b554f481e 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -140,18 +140,6 @@ struct audit_buffer { struct audit_context *ctx; /* NULL or associated context */ }; -void audit_set_type(struct audit_buffer *ab, int type) -{ - struct nlmsghdr *nlh = (struct nlmsghdr *)ab->skb->data; - nlh->nlmsg_type = type; -} - -static void audit_set_pid(struct audit_buffer *ab, pid_t pid) -{ - struct nlmsghdr *nlh = (struct nlmsghdr *)ab->skb->data; - nlh->nlmsg_pid = pid; -} - struct audit_entry { struct list_head list; struct audit_rule rule; @@ -344,7 +332,6 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) void *data; struct audit_status *status_get, status_set; int err; - struct audit_buffer *ab; u16 msg_type = nlh->nlmsg_type; uid_t loginuid; /* loginuid of sender */ struct audit_sig_info sig_data; @@ -396,19 +383,13 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) loginuid); break; case AUDIT_USER: - ab = audit_log_start(NULL); - if (!ab) - break; /* audit_panic has been called */ - audit_log_format(ab, + audit_log_type(NULL, AUDIT_USER, pid, "user pid=%d uid=%d length=%d loginuid=%u" " msg='%.1024s'", pid, uid, (int)(nlh->nlmsg_len - ((char *)data - (char *)nlh)), loginuid, (char *)data); - audit_set_type(ab, AUDIT_USER); - audit_set_pid(ab, pid); - audit_log_end(ab); break; case AUDIT_ADD: case AUDIT_DEL: @@ -560,12 +541,10 @@ static void audit_buffer_free(struct audit_buffer *ab) spin_unlock_irqrestore(&audit_freelist_lock, flags); } -static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx, - int gfp_mask) +static struct audit_buffer * audit_buffer_alloc(int gfp_mask) { unsigned long flags; struct audit_buffer *ab = NULL; - struct nlmsghdr *nlh; spin_lock_irqsave(&audit_freelist_lock, flags); if (!list_empty(&audit_freelist)) { @@ -587,12 +566,6 @@ static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx, if (!ab->skb) goto err; - ab->ctx = ctx; - nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0)); - nlh->nlmsg_type = AUDIT_KERNEL; - nlh->nlmsg_flags = 0; - nlh->nlmsg_pid = 0; - nlh->nlmsg_seq = 0; return ab; err: audit_buffer_free(ab); @@ -605,11 +578,12 @@ err: * syscall, then the syscall is marked as auditable and an audit record * will be written at syscall exit. If there is no associated task, tsk * should be NULL. */ -struct audit_buffer *audit_log_start(struct audit_context *ctx) +struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, int pid) { struct audit_buffer *ab = NULL; struct timespec t; unsigned int serial; + struct nlmsghdr *nlh; if (!audit_initialized) return NULL; @@ -626,12 +600,19 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx) return NULL; } - ab = audit_buffer_alloc(ctx, GFP_ATOMIC); + ab = audit_buffer_alloc(GFP_ATOMIC); if (!ab) { audit_log_lost("out of memory in audit_log_start"); return NULL; } + ab->ctx = ctx; + nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0)); + nlh->nlmsg_type = type; + nlh->nlmsg_flags = 0; + nlh->nlmsg_pid = pid; + nlh->nlmsg_seq = 0; + if (!audit_get_stamp(ab->ctx, &t, &serial)) { t = CURRENT_TIME; serial = 0; @@ -828,12 +809,13 @@ void audit_log_end(struct audit_buffer *ab) /* Log an audit record. This is a convenience function that calls * audit_log_start, audit_log_vformat, and audit_log_end. It may be * called in any context. */ -void audit_log(struct audit_context *ctx, const char *fmt, ...) +void audit_log_type(struct audit_context *ctx, int type, int pid, + const char *fmt, ...) { struct audit_buffer *ab; va_list args; - ab = audit_log_start(ctx); + ab = audit_log_start(ctx, type, pid); if (ab) { va_start(args, fmt); audit_log_vformat(ab, fmt, args); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 94338abf76f5..d089263253a7 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -648,7 +648,7 @@ static void audit_log_exit(struct audit_context *context) int i; struct audit_buffer *ab; - ab = audit_log_start(context); + ab = audit_log_start(context, AUDIT_KERNEL, 0); if (!ab) return; /* audit_panic has been called */ audit_log_format(ab, "syscall=%d", context->major); @@ -680,7 +680,7 @@ static void audit_log_exit(struct audit_context *context) while (context->aux) { struct audit_aux_data *aux; - ab = audit_log_start(context); + ab = audit_log_start(context, AUDIT_KERNEL, 0); if (!ab) continue; /* audit_panic has been called */ @@ -701,7 +701,7 @@ static void audit_log_exit(struct audit_context *context) } for (i = 0; i < context->name_count; i++) { - ab = audit_log_start(context); + ab = audit_log_start(context, AUDIT_KERNEL, 0); if (!ab) continue; /* audit_panic has been called */ audit_log_format(ab, "item=%d", i); @@ -1005,22 +1005,13 @@ int audit_get_stamp(struct audit_context *ctx, return 0; } -extern int audit_set_type(struct audit_buffer *ab, int type); - int audit_set_loginuid(struct task_struct *task, uid_t loginuid) { if (task->audit_context) { - struct audit_buffer *ab; - - ab = audit_log_start(NULL); - if (ab) { - audit_log_format(ab, "login pid=%d uid=%u " - "old loginuid=%u new loginuid=%u", - task->pid, task->uid, - task->audit_context->loginuid, loginuid); - audit_set_type(ab, AUDIT_LOGIN); - audit_log_end(ab); - } + audit_log_type(NULL, AUDIT_LOGIN, 0, + "login pid=%d uid=%u old loginuid=%u new loginuid=%u", + task->pid, task->uid, task->audit_context->loginuid, + loginuid); task->audit_context->loginuid = loginuid; } return 0; diff --git a/security/selinux/avc.c b/security/selinux/avc.c index 85a6f66a873f..9e71a1bbe011 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -549,7 +549,7 @@ void avc_audit(u32 ssid, u32 tsid, return; } - ab = audit_log_start(current->audit_context); + ab = audit_log_start(current->audit_context, AUDIT_KERNEL, 0); if (!ab) return; /* audit_panic has been called */ audit_log_format(ab, "avc: %s ", denied ? "denied" : "granted"); -- cgit v1.2.3 From 6f2f38128170814e151cfedf79532e19cd179567 Mon Sep 17 00:00:00 2001 From: Brad Campbell Date: Thu, 12 May 2005 15:07:47 -0400 Subject: [PATCH] libata basic detection and errata for PATA->SATA bridges This patch works around an issue with WD drives (and possibly others) over SiL PATA->SATA Bridges on SATA controllers locking up with transfers > 200 sectors. Signed-off-by: Brad Campbell --- drivers/scsi/libata-core.c | 35 +++++++++++++++++++++++++++++++++-- include/linux/ata.h | 1 + include/linux/libata.h | 1 + 3 files changed, 35 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c index 0b5d3a5b7eda..8b5a3f00083d 100644 --- a/drivers/scsi/libata-core.c +++ b/drivers/scsi/libata-core.c @@ -1186,6 +1186,37 @@ err_out: DPRINTK("EXIT, err\n"); } + +static inline u8 ata_dev_knobble(struct ata_port *ap) +{ + return ((ap->cbl == ATA_CBL_SATA) && (!ata_id_is_sata(ap->device->id))); +} + +/** + * ata_dev_config - Run device specific handlers and check for + * SATA->PATA bridges + * @ap: Bus + * @i: Device + * + * LOCKING: + */ + +void ata_dev_config(struct ata_port *ap, unsigned int i) +{ + /* limit bridge transfers to udma5, 200 sectors */ + if (ata_dev_knobble(ap)) { + printk(KERN_INFO "ata%u(%u): applying bridge limits\n", + ap->id, ap->device->devno); + ap->udma_mask &= ATA_UDMA5; + ap->host->max_sectors = ATA_MAX_SECTORS; + ap->host->hostt->max_sectors = ATA_MAX_SECTORS; + ap->device->flags |= ATA_DFLAG_LOCK_SECTORS; + } + + if (ap->ops->dev_config) + ap->ops->dev_config(ap, &ap->device[i]); +} + /** * ata_bus_probe - Reset and probe ATA bus * @ap: Bus to probe @@ -1208,8 +1239,7 @@ static int ata_bus_probe(struct ata_port *ap) ata_dev_identify(ap, i); if (ata_dev_present(&ap->device[i])) { found = 1; - if (ap->ops->dev_config) - ap->ops->dev_config(ap, &ap->device[i]); + ata_dev_config(ap,i); } } @@ -4014,6 +4044,7 @@ EXPORT_SYMBOL_GPL(ata_scsi_release); EXPORT_SYMBOL_GPL(ata_host_intr); EXPORT_SYMBOL_GPL(ata_dev_classify); EXPORT_SYMBOL_GPL(ata_dev_id_string); +EXPORT_SYMBOL_GPL(ata_dev_config); EXPORT_SYMBOL_GPL(ata_scsi_simulate); #ifdef CONFIG_PCI diff --git a/include/linux/ata.h b/include/linux/ata.h index f178894edd04..ca5fcadf9981 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -224,6 +224,7 @@ struct ata_taskfile { }; #define ata_id_is_ata(id) (((id)[0] & (1 << 15)) == 0) +#define ata_id_is_sata(id) ((id)[93] == 0) #define ata_id_rahead_enabled(id) ((id)[85] & (1 << 6)) #define ata_id_wcache_enabled(id) ((id)[85] & (1 << 5)) #define ata_id_has_flush(id) ((id)[83] & (1 << 12)) diff --git a/include/linux/libata.h b/include/linux/libata.h index 505160ab472b..d33e70361a7d 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -420,6 +420,7 @@ extern void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg, extern unsigned int ata_dev_classify(struct ata_taskfile *tf); extern void ata_dev_id_string(u16 *id, unsigned char *s, unsigned int ofs, unsigned int len); +extern void ata_dev_config(struct ata_port *ap, unsigned int i); extern void ata_bmdma_setup (struct ata_queued_cmd *qc); extern void ata_bmdma_start (struct ata_queued_cmd *qc); extern void ata_bmdma_stop(struct ata_port *ap); -- cgit v1.2.3 From 7d17c1d606f6e89778f05554ddea43791d5c92a0 Mon Sep 17 00:00:00 2001 From: Date: Thu, 12 May 2005 19:45:25 -0400 Subject: [netdrvrs] Use netif_carrier_* instead of IFF_RUNNING --- drivers/net/au1000_eth.c | 10 ++-------- drivers/net/bmac.c | 2 -- drivers/net/sk98lin/skge.c | 8 +++----- drivers/net/tlan.c | 4 ++-- drivers/net/tokenring/ibmtr.c | 11 +++-------- drivers/net/wan/lmc/lmc_main.c | 8 ++------ include/linux/if.h | 2 +- 7 files changed, 13 insertions(+), 32 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/au1000_eth.c b/drivers/net/au1000_eth.c index 5a2efd343db4..c82b9cd1c924 100644 --- a/drivers/net/au1000_eth.c +++ b/drivers/net/au1000_eth.c @@ -1681,10 +1681,6 @@ static int au1000_init(struct net_device *dev) control |= MAC_FULL_DUPLEX; } - /* fix for startup without cable */ - if (!link) - dev->flags &= ~IFF_RUNNING; - aup->mac->control = control; aup->mac->vlan1_tag = 0x8100; /* activate vlan support */ au_sync(); @@ -1709,16 +1705,14 @@ static void au1000_timer(unsigned long data) if_port = dev->if_port; if (aup->phy_ops->phy_status(dev, aup->phy_addr, &link, &speed) == 0) { if (link) { - if (!(dev->flags & IFF_RUNNING)) { + if (!netif_carrier_ok(dev)) { netif_carrier_on(dev); - dev->flags |= IFF_RUNNING; printk(KERN_INFO "%s: link up\n", dev->name); } } else { - if (dev->flags & IFF_RUNNING) { + if (netif_carrier_ok(dev)) { netif_carrier_off(dev); - dev->flags &= ~IFF_RUNNING; dev->if_port = 0; printk(KERN_INFO "%s: link down\n", dev->name); } diff --git a/drivers/net/bmac.c b/drivers/net/bmac.c index 734bd4ee3f9b..00e5257b176f 100644 --- a/drivers/net/bmac.c +++ b/drivers/net/bmac.c @@ -1412,7 +1412,6 @@ static int bmac_open(struct net_device *dev) bp->opened = 1; bmac_reset_and_enable(dev); enable_irq(dev->irq); - dev->flags |= IFF_RUNNING; return 0; } @@ -1425,7 +1424,6 @@ static int bmac_close(struct net_device *dev) int i; bp->sleeping = 1; - dev->flags &= ~(IFF_UP | IFF_RUNNING); /* disable rx and tx */ config = bmread(dev, RXCFG); diff --git a/drivers/net/sk98lin/skge.c b/drivers/net/sk98lin/skge.c index 05b827f79f54..1ccb2989001c 100644 --- a/drivers/net/sk98lin/skge.c +++ b/drivers/net/sk98lin/skge.c @@ -4212,7 +4212,7 @@ SK_BOOL DualNet; Flags); SkGeStopPort(pAC, IoC, FromPort, SK_STOP_ALL, SK_HARD_RST); - pAC->dev[Param.Para32[0]]->flags &= ~IFF_RUNNING; + netif_carrier_off(pAC->dev[Param.Para32[0]]); spin_unlock_irqrestore( &pAC->TxPort[FromPort][TX_PRIO_LOW].TxDesRingLock, Flags); @@ -4355,7 +4355,7 @@ SK_BOOL DualNet; } /* Inform the world that link protocol is up. */ - pAC->dev[Param.Para32[0]]->flags |= IFF_RUNNING; + netif_carrier_on(pAC->dev[Param.Para32[0]]); break; case SK_DRV_NET_DOWN: /* SK_U32 Reason */ @@ -4368,7 +4368,7 @@ SK_BOOL DualNet; } else { DoPrintInterfaceChange = SK_TRUE; } - pAC->dev[Param.Para32[1]]->flags &= ~IFF_RUNNING; + netif_carrier_off(pAC->dev[Param.Para32[1]]); break; case SK_DRV_SWITCH_HARD: /* SK_U32 FromPortIdx SK_U32 ToPortIdx */ SK_DBG_MSG(NULL, SK_DBGMOD_DRV, SK_DBGCAT_DRV_EVENT, @@ -4961,7 +4961,6 @@ static int __devinit skge_probe_one(struct pci_dev *pdev, #ifdef CONFIG_NET_POLL_CONTROLLER dev->poll_controller = &SkGePollController; #endif - dev->flags &= ~IFF_RUNNING; SET_NETDEV_DEV(dev, &pdev->dev); SET_ETHTOOL_OPS(dev, &SkGeEthtoolOps); @@ -5035,7 +5034,6 @@ static int __devinit skge_probe_one(struct pci_dev *pdev, dev->set_mac_address = &SkGeSetMacAddr; dev->do_ioctl = &SkGeIoctl; dev->change_mtu = &SkGeChangeMtu; - dev->flags &= ~IFF_RUNNING; SET_NETDEV_DEV(dev, &pdev->dev); SET_ETHTOOL_OPS(dev, &SkGeEthtoolOps); diff --git a/drivers/net/tlan.c b/drivers/net/tlan.c index a7ffa64502dd..ce116624f136 100644 --- a/drivers/net/tlan.c +++ b/drivers/net/tlan.c @@ -2807,7 +2807,7 @@ void TLan_PhyMonitor( struct net_device *dev ) if (priv->link) { priv->link = 0; printk(KERN_DEBUG "TLAN: %s has lost link\n", dev->name); - dev->flags &= ~IFF_RUNNING; + netif_carrier_off(dev); TLan_SetTimer( dev, (2*HZ), TLAN_TIMER_LINK_BEAT ); return; } @@ -2817,7 +2817,7 @@ void TLan_PhyMonitor( struct net_device *dev ) if ((phy_status & MII_GS_LINK) && !priv->link) { priv->link = 1; printk(KERN_DEBUG "TLAN: %s has reestablished link\n", dev->name); - dev->flags |= IFF_RUNNING; + netif_carrier_on(dev); } /* Setup a new monitor */ diff --git a/drivers/net/tokenring/ibmtr.c b/drivers/net/tokenring/ibmtr.c index c098863bdd9d..3873917a9c22 100644 --- a/drivers/net/tokenring/ibmtr.c +++ b/drivers/net/tokenring/ibmtr.c @@ -888,11 +888,6 @@ static int tok_open(struct net_device *dev) ti->sap_status = CLOSED; /* CLOSED or OPEN */ ti->open_failure = NO; /* NO or YES */ ti->open_mode = MANUAL; /* MANUAL or AUTOMATIC */ - /* 12/2000 not typical Linux, but we can use RUNNING to let us know when - the network has crapped out or cables are disconnected. Useful because - the IFF_UP flag stays up the whole time, until ifconfig tr0 down. - */ - dev->flags &= ~IFF_RUNNING; ti->sram_phys &= ~1; /* to reverse what we do in tok_close */ /* init the spinlock */ @@ -1242,7 +1237,7 @@ irqreturn_t tok_interrupt(int irq, void *dev_id, struct pt_regs *regs) ti->open_status = CLOSED; ti->sap_status = CLOSED; ti->open_mode = AUTOMATIC; - dev->flags &= ~IFF_RUNNING; + netif_carrier_off(dev); netif_stop_queue(dev); ti->open_action = RESTART; outb(0, dev->base_addr + ADAPTRESET); @@ -1323,7 +1318,7 @@ irqreturn_t tok_interrupt(int irq, void *dev_id, struct pt_regs *regs) break; } netif_wake_queue(dev); - dev->flags |= IFF_RUNNING;/*BMS 12/2000*/ + netif_carrier_on(dev); break; case DIR_INTERRUPT: case DIR_MOD_OPEN_PARAMS: @@ -1427,7 +1422,7 @@ irqreturn_t tok_interrupt(int irq, void *dev_id, struct pt_regs *regs) ring_status); if(ring_status& (REMOVE_RECV|AUTO_REMOVAL|LOBE_FAULT)){ netif_stop_queue(dev); - dev->flags &= ~IFF_RUNNING;/*not typical Linux*/ + netif_carrier_off(dev); DPRINTK("Remove received, or Auto-removal error" ", or Lobe fault\n"); DPRINTK("We'll try to reopen the closed adapter" diff --git a/drivers/net/wan/lmc/lmc_main.c b/drivers/net/wan/lmc/lmc_main.c index 15e545f66cd7..2b948ea397d5 100644 --- a/drivers/net/wan/lmc/lmc_main.c +++ b/drivers/net/wan/lmc/lmc_main.c @@ -723,7 +723,7 @@ static void lmc_watchdog (unsigned long data) /*fold00*/ /* lmc_reset (sc); Why reset??? The link can go down ok */ /* Inform the world that link has been lost */ - dev->flags &= ~IFF_RUNNING; + netif_carrier_off(dev); } /* @@ -736,7 +736,7 @@ static void lmc_watchdog (unsigned long data) /*fold00*/ /* lmc_reset (sc); Again why reset??? */ /* Inform the world that link protocol is back up. */ - dev->flags |= IFF_RUNNING; + netif_carrier_on(dev); /* Now we have to tell the syncppp that we had an outage * and that it should deal. Calling sppp_reopen here @@ -1168,8 +1168,6 @@ static void lmc_running_reset (struct net_device *dev) /*fold00*/ sc->lmc_media->set_link_status (sc, 1); sc->lmc_media->set_status (sc, NULL); - //dev->flags |= IFF_RUNNING; - netif_wake_queue(dev); sc->lmc_txfull = 0; @@ -1233,8 +1231,6 @@ static int lmc_ifdown (struct net_device *dev) /*fold00*/ csr6 &= ~LMC_DEC_SR; /* Turn off the Receive bit */ LMC_CSR_WRITE (sc, csr_command, csr6); - dev->flags &= ~IFF_RUNNING; - sc->stats.rx_missed_errors += LMC_CSR_READ (sc, csr_missed_frames) & 0xffff; diff --git a/include/linux/if.h b/include/linux/if.h index d73a9d62f208..ce627d9092ef 100644 --- a/include/linux/if.h +++ b/include/linux/if.h @@ -33,7 +33,7 @@ #define IFF_LOOPBACK 0x8 /* is a loopback net */ #define IFF_POINTOPOINT 0x10 /* interface is has p-p link */ #define IFF_NOTRAILERS 0x20 /* avoid use of trailers */ -#define IFF_RUNNING 0x40 /* resources allocated */ +#define IFF_RUNNING 0x40 /* interface running and carrier ok */ #define IFF_NOARP 0x80 /* no ARP protocol */ #define IFF_PROMISC 0x100 /* receive all packets */ #define IFF_ALLMULTI 0x200 /* receive all multicast packets*/ -- cgit v1.2.3 From fff9cfd99c0f88645c3f50d7476d6c8cef99f140 Mon Sep 17 00:00:00 2001 From: Date: Thu, 12 May 2005 20:24:19 -0400 Subject: [PATCH] Wireless Extensions 18 (aka WPA) This is version 18 of the Wireless Extensions. The main change is that it adds all the necessary APIs for WPA and WPA2 support. This work was entirely done by Jouni Malinen, so let's thank him for both his hard work and deep expertise on the subject ;-) This APIs obviously doesn't do much by itself and works in concert with driver support (Jouni already sent you the HostAP changes) and userspace (Jouni is updating wpa_supplicant). This is also orthogonal with the ongoing work on in-kernel IEEE support (but potentially useful). The patch is attached, tested with 2.6.11. Normally, I would ask you to push that directly in the kernel (99% of the patch has been on my web page for ages and it does not affect non-WPA stuff), but Jouni convinced me that it should bake a few weeks in wireless-2.6 first, so that other driver maintainers can get up to speed with it. Signed-off-by: Jeff Garzik --- include/linux/wireless.h | 283 ++++++++++++++++++++++++++++++++++++++++++++++- net/core/wireless.c | 74 ++++++++++++- 2 files changed, 352 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/wireless.h b/include/linux/wireless.h index 2f51f2b6562e..ae485f9c916e 100644 --- a/include/linux/wireless.h +++ b/include/linux/wireless.h @@ -1,10 +1,10 @@ /* * This file define a set of standard wireless extensions * - * Version : 17 21.6.04 + * Version : 18 12.3.05 * * Authors : Jean Tourrilhes - HPL - - * Copyright (c) 1997-2004 Jean Tourrilhes, All Rights Reserved. + * Copyright (c) 1997-2005 Jean Tourrilhes, All Rights Reserved. */ #ifndef _LINUX_WIRELESS_H @@ -82,7 +82,7 @@ * (there is some stuff that will be added in the future...) * I just plan to increment with each new version. */ -#define WIRELESS_EXT 17 +#define WIRELESS_EXT 18 /* * Changes : @@ -182,6 +182,21 @@ * - Document (struct iw_quality *)->updated, add new flags (INVALID) * - Wireless Event capability in struct iw_range * - Add support for relative TxPower (yick !) + * + * V17 to V18 (From Jouni Malinen ) + * ---------- + * - Add support for WPA/WPA2 + * - Add extended encoding configuration (SIOCSIWENCODEEXT and + * SIOCGIWENCODEEXT) + * - Add SIOCSIWGENIE/SIOCGIWGENIE + * - Add SIOCSIWMLME + * - Add SIOCSIWPMKSA + * - Add struct iw_range bit field for supported encoding capabilities + * - Add optional scan request parameters for SIOCSIWSCAN + * - Add SIOCSIWAUTH/SIOCGIWAUTH for setting authentication and WPA + * related parameters (extensible up to 4096 parameter values) + * - Add wireless events: IWEVGENIE, IWEVMICHAELMICFAILURE, + * IWEVASSOCREQIE, IWEVASSOCRESPIE, IWEVPMKIDCAND */ /**************************** CONSTANTS ****************************/ @@ -256,6 +271,30 @@ #define SIOCSIWPOWER 0x8B2C /* set Power Management settings */ #define SIOCGIWPOWER 0x8B2D /* get Power Management settings */ +/* WPA : Generic IEEE 802.11 informatiom element (e.g., for WPA/RSN/WMM). + * This ioctl uses struct iw_point and data buffer that includes IE id and len + * fields. More than one IE may be included in the request. Setting the generic + * IE to empty buffer (len=0) removes the generic IE from the driver. Drivers + * are allowed to generate their own WPA/RSN IEs, but in these cases, drivers + * are required to report the used IE as a wireless event, e.g., when + * associating with an AP. */ +#define SIOCSIWGENIE 0x8B30 /* set generic IE */ +#define SIOCGIWGENIE 0x8B31 /* get generic IE */ + +/* WPA : IEEE 802.11 MLME requests */ +#define SIOCSIWMLME 0x8B16 /* request MLME operation; uses + * struct iw_mlme */ +/* WPA : Authentication mode parameters */ +#define SIOCSIWAUTH 0x8B32 /* set authentication mode params */ +#define SIOCGIWAUTH 0x8B33 /* get authentication mode params */ + +/* WPA : Extended version of encoding configuration */ +#define SIOCSIWENCODEEXT 0x8B34 /* set encoding token & mode */ +#define SIOCGIWENCODEEXT 0x8B35 /* get encoding token & mode */ + +/* WPA2 : PMKSA cache management */ +#define SIOCSIWPMKSA 0x8B36 /* PMKSA cache operation */ + /* -------------------- DEV PRIVATE IOCTL LIST -------------------- */ /* These 32 ioctl are wireless device private, for 16 commands. @@ -297,6 +336,34 @@ #define IWEVCUSTOM 0x8C02 /* Driver specific ascii string */ #define IWEVREGISTERED 0x8C03 /* Discovered a new node (AP mode) */ #define IWEVEXPIRED 0x8C04 /* Expired a node (AP mode) */ +#define IWEVGENIE 0x8C05 /* Generic IE (WPA, RSN, WMM, ..) + * (scan results); This includes id and + * length fields. One IWEVGENIE may + * contain more than one IE. Scan + * results may contain one or more + * IWEVGENIE events. */ +#define IWEVMICHAELMICFAILURE 0x8C06 /* Michael MIC failure + * (struct iw_michaelmicfailure) + */ +#define IWEVASSOCREQIE 0x8C07 /* IEs used in (Re)Association Request. + * The data includes id and length + * fields and may contain more than one + * IE. This event is required in + * Managed mode if the driver + * generates its own WPA/RSN IE. This + * should be sent just before + * IWEVREGISTERED event for the + * association. */ +#define IWEVASSOCRESPIE 0x8C08 /* IEs used in (Re)Association + * Response. The data includes id and + * length fields and may contain more + * than one IE. This may be sent + * between IWEVASSOCREQIE and + * IWEVREGISTERED events for the + * association. */ +#define IWEVPMKIDCAND 0x8C09 /* PMKID candidate for RSN + * pre-authentication + * (struct iw_pmkid_cand) */ #define IWEVFIRST 0x8C00 @@ -432,12 +499,94 @@ #define IW_SCAN_THIS_MODE 0x0020 /* Scan only this Mode */ #define IW_SCAN_ALL_RATE 0x0040 /* Scan all Bit-Rates */ #define IW_SCAN_THIS_RATE 0x0080 /* Scan only this Bit-Rate */ +/* struct iw_scan_req scan_type */ +#define IW_SCAN_TYPE_ACTIVE 0 +#define IW_SCAN_TYPE_PASSIVE 1 /* Maximum size of returned data */ #define IW_SCAN_MAX_DATA 4096 /* In bytes */ /* Max number of char in custom event - use multiple of them if needed */ #define IW_CUSTOM_MAX 256 /* In bytes */ +/* Generic information element */ +#define IW_GENERIC_IE_MAX 1024 + +/* MLME requests (SIOCSIWMLME / struct iw_mlme) */ +#define IW_MLME_DEAUTH 0 +#define IW_MLME_DISASSOC 1 + +/* SIOCSIWAUTH/SIOCGIWAUTH struct iw_param flags */ +#define IW_AUTH_INDEX 0x0FFF +#define IW_AUTH_FLAGS 0xF000 +/* SIOCSIWAUTH/SIOCGIWAUTH parameters (0 .. 4095) + * (IW_AUTH_INDEX mask in struct iw_param flags; this is the index of the + * parameter that is being set/get to; value will be read/written to + * struct iw_param value field) */ +#define IW_AUTH_WPA_VERSION 0 +#define IW_AUTH_CIPHER_PAIRWISE 1 +#define IW_AUTH_CIPHER_GROUP 2 +#define IW_AUTH_KEY_MGMT 3 +#define IW_AUTH_TKIP_COUNTERMEASURES 4 +#define IW_AUTH_DROP_UNENCRYPTED 5 +#define IW_AUTH_80211_AUTH_ALG 6 +#define IW_AUTH_WPA_ENABLED 7 +#define IW_AUTH_RX_UNENCRYPTED_EAPOL 8 +#define IW_AUTH_ROAMING_CONTROL 9 +#define IW_AUTH_PRIVACY_INVOKED 10 + +/* IW_AUTH_WPA_VERSION values (bit field) */ +#define IW_AUTH_WPA_VERSION_DISABLED 0x00000001 +#define IW_AUTH_WPA_VERSION_WPA 0x00000002 +#define IW_AUTH_WPA_VERSION_WPA2 0x00000004 + +/* IW_AUTH_PAIRWISE_CIPHER and IW_AUTH_GROUP_CIPHER values (bit field) */ +#define IW_AUTH_CIPHER_NONE 0x00000001 +#define IW_AUTH_CIPHER_WEP40 0x00000002 +#define IW_AUTH_CIPHER_TKIP 0x00000004 +#define IW_AUTH_CIPHER_CCMP 0x00000008 +#define IW_AUTH_CIPHER_WEP104 0x00000010 + +/* IW_AUTH_KEY_MGMT values (bit field) */ +#define IW_AUTH_KEY_MGMT_802_1X 1 +#define IW_AUTH_KEY_MGMT_PSK 2 + +/* IW_AUTH_80211_AUTH_ALG values (bit field) */ +#define IW_AUTH_ALG_OPEN_SYSTEM 0x00000001 +#define IW_AUTH_ALG_SHARED_KEY 0x00000002 +#define IW_AUTH_ALG_LEAP 0x00000004 + +/* IW_AUTH_ROAMING_CONTROL values */ +#define IW_AUTH_ROAMING_ENABLE 0 /* driver/firmware based roaming */ +#define IW_AUTH_ROAMING_DISABLE 1 /* user space program used for roaming + * control */ + +/* SIOCSIWENCODEEXT definitions */ +#define IW_ENCODE_SEQ_MAX_SIZE 8 +/* struct iw_encode_ext ->alg */ +#define IW_ENCODE_ALG_NONE 0 +#define IW_ENCODE_ALG_WEP 1 +#define IW_ENCODE_ALG_TKIP 2 +#define IW_ENCODE_ALG_CCMP 3 +/* struct iw_encode_ext ->ext_flags */ +#define IW_ENCODE_EXT_TX_SEQ_VALID 0x00000001 +#define IW_ENCODE_EXT_RX_SEQ_VALID 0x00000002 +#define IW_ENCODE_EXT_GROUP_KEY 0x00000004 +#define IW_ENCODE_EXT_SET_TX_KEY 0x00000008 + +/* IWEVMICHAELMICFAILURE : struct iw_michaelmicfailure ->flags */ +#define IW_MICFAILURE_KEY_ID 0x00000003 /* Key ID 0..3 */ +#define IW_MICFAILURE_GROUP 0x00000004 +#define IW_MICFAILURE_PAIRWISE 0x00000008 +#define IW_MICFAILURE_STAKEY 0x00000010 +#define IW_MICFAILURE_COUNT 0x00000060 /* 1 or 2 (0 = count not supported) + */ + +/* Bit field values for enc_capa in struct iw_range */ +#define IW_ENC_CAPA_WPA 0x00000001 +#define IW_ENC_CAPA_WPA2 0x00000002 +#define IW_ENC_CAPA_CIPHER_TKIP 0x00000004 +#define IW_ENC_CAPA_CIPHER_CCMP 0x00000008 + /* Event capability macros - in (struct iw_range *)->event_capa * Because we have more than 32 possible events, we use an array of * 32 bit bitmasks. Note : 32 bits = 0x20 = 2^5. */ @@ -546,6 +695,132 @@ struct iw_thrspy struct iw_quality high; /* High threshold */ }; +/* + * Optional data for scan request + * + * Note: these optional parameters are controlling parameters for the + * scanning behavior, these do not apply to getting scan results + * (SIOCGIWSCAN). Drivers are expected to keep a local BSS table and + * provide a merged results with all BSSes even if the previous scan + * request limited scanning to a subset, e.g., by specifying an SSID. + * Especially, scan results are required to include an entry for the + * current BSS if the driver is in Managed mode and associated with an AP. + */ +struct iw_scan_req +{ + __u8 scan_type; /* IW_SCAN_TYPE_{ACTIVE,PASSIVE} */ + __u8 essid_len; + __u8 num_channels; /* num entries in channel_list; + * 0 = scan all allowed channels */ + __u8 flags; /* reserved as padding; use zero, this may + * be used in the future for adding flags + * to request different scan behavior */ + struct sockaddr bssid; /* ff:ff:ff:ff:ff:ff for broadcast BSSID or + * individual address of a specific BSS */ + + /* + * Use this ESSID if IW_SCAN_THIS_ESSID flag is used instead of using + * the current ESSID. This allows scan requests for specific ESSID + * without having to change the current ESSID and potentially breaking + * the current association. + */ + __u8 essid[IW_ESSID_MAX_SIZE]; + + /* + * Optional parameters for changing the default scanning behavior. + * These are based on the MLME-SCAN.request from IEEE Std 802.11. + * TU is 1.024 ms. If these are set to 0, driver is expected to use + * reasonable default values. min_channel_time defines the time that + * will be used to wait for the first reply on each channel. If no + * replies are received, next channel will be scanned after this. If + * replies are received, total time waited on the channel is defined by + * max_channel_time. + */ + __u32 min_channel_time; /* in TU */ + __u32 max_channel_time; /* in TU */ + + struct iw_freq channel_list[IW_MAX_FREQUENCIES]; +}; + +/* ------------------------- WPA SUPPORT ------------------------- */ + +/* + * Extended data structure for get/set encoding (this is used with + * SIOCSIWENCODEEXT/SIOCGIWENCODEEXT. struct iw_point and IW_ENCODE_* + * flags are used in the same way as with SIOCSIWENCODE/SIOCGIWENCODE and + * only the data contents changes (key data -> this structure, including + * key data). + * + * If the new key is the first group key, it will be set as the default + * TX key. Otherwise, default TX key index is only changed if + * IW_ENCODE_EXT_SET_TX_KEY flag is set. + * + * Key will be changed with SIOCSIWENCODEEXT in all cases except for + * special "change TX key index" operation which is indicated by setting + * key_len = 0 and ext_flags |= IW_ENCODE_EXT_SET_TX_KEY. + * + * tx_seq/rx_seq are only used when respective + * IW_ENCODE_EXT_{TX,RX}_SEQ_VALID flag is set in ext_flags. Normal + * TKIP/CCMP operation is to set RX seq with SIOCSIWENCODEEXT and start + * TX seq from zero whenever key is changed. SIOCGIWENCODEEXT is normally + * used only by an Authenticator (AP or an IBSS station) to get the + * current TX sequence number. Using TX_SEQ_VALID for SIOCSIWENCODEEXT and + * RX_SEQ_VALID for SIOCGIWENCODEEXT are optional, but can be useful for + * debugging/testing. + */ +struct iw_encode_ext +{ + __u32 ext_flags; /* IW_ENCODE_EXT_* */ + __u8 tx_seq[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ + __u8 rx_seq[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ + struct sockaddr addr; /* ff:ff:ff:ff:ff:ff for broadcast/multicast + * (group) keys or unicast address for + * individual keys */ + __u16 alg; /* IW_ENCODE_ALG_* */ + __u16 key_len; + __u8 key[0]; +}; + +/* SIOCSIWMLME data */ +struct iw_mlme +{ + __u16 cmd; /* IW_MLME_* */ + __u16 reason_code; + struct sockaddr addr; +}; + +/* SIOCSIWPMKSA data */ +#define IW_PMKSA_ADD 1 +#define IW_PMKSA_REMOVE 2 +#define IW_PMKSA_FLUSH 3 + +#define IW_PMKID_LEN 16 + +struct iw_pmksa +{ + __u32 cmd; /* IW_PMKSA_* */ + struct sockaddr bssid; + __u8 pmkid[IW_PMKID_LEN]; +}; + +/* IWEVMICHAELMICFAILURE data */ +struct iw_michaelmicfailure +{ + __u32 flags; + struct sockaddr src_addr; + __u8 tsc[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ +}; + +/* IWEVPMKIDCAND data */ +#define IW_PMKID_CAND_PREAUTH 0x00000001 /* RNS pre-authentication enabled */ +struct iw_pmkid_cand +{ + __u32 flags; /* IW_PMKID_CAND_* */ + __u32 index; /* the smaller the index, the higher the + * priority */ + struct sockaddr bssid; +}; + /* ------------------------ WIRELESS STATS ------------------------ */ /* * Wireless statistics (used for /proc/net/wireless) @@ -725,6 +1000,8 @@ struct iw_range struct iw_freq freq[IW_MAX_FREQUENCIES]; /* list */ /* Note : this frequency list doesn't need to fit channel numbers, * because each entry contain its channel index */ + + __u32 enc_capa; /* IW_ENC_CAPA_* bit field */ }; /* diff --git a/net/core/wireless.c b/net/core/wireless.c index 750cc5daeb03..b2fe378dfbf8 100644 --- a/net/core/wireless.c +++ b/net/core/wireless.c @@ -2,7 +2,7 @@ * This file implement the Wireless Extensions APIs. * * Authors : Jean Tourrilhes - HPL - - * Copyright (c) 1997-2004 Jean Tourrilhes, All Rights Reserved. + * Copyright (c) 1997-2005 Jean Tourrilhes, All Rights Reserved. * * (As all part of the Linux kernel, this file is GPL) */ @@ -187,6 +187,12 @@ static const struct iw_ioctl_description standard_ioctl[] = { .header_type = IW_HEADER_TYPE_ADDR, .flags = IW_DESCR_FLAG_DUMP, }, + [SIOCSIWMLME - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_mlme), + .max_tokens = sizeof(struct iw_mlme), + }, [SIOCGIWAPLIST - SIOCIWFIRST] = { .header_type = IW_HEADER_TYPE_POINT, .token_size = sizeof(struct sockaddr) + @@ -195,7 +201,10 @@ static const struct iw_ioctl_description standard_ioctl[] = { .flags = IW_DESCR_FLAG_NOMAX, }, [SIOCSIWSCAN - SIOCIWFIRST] = { - .header_type = IW_HEADER_TYPE_PARAM, + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = 0, + .max_tokens = sizeof(struct iw_scan_req), }, [SIOCGIWSCAN - SIOCIWFIRST] = { .header_type = IW_HEADER_TYPE_POINT, @@ -273,6 +282,42 @@ static const struct iw_ioctl_description standard_ioctl[] = { [SIOCGIWPOWER - SIOCIWFIRST] = { .header_type = IW_HEADER_TYPE_PARAM, }, + [SIOCSIWGENIE - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [SIOCGIWGENIE - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [SIOCSIWAUTH - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_PARAM, + }, + [SIOCGIWAUTH - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_PARAM, + }, + [SIOCSIWENCODEEXT - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_encode_ext), + .max_tokens = sizeof(struct iw_encode_ext) + + IW_ENCODING_TOKEN_MAX, + }, + [SIOCGIWENCODEEXT - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_encode_ext), + .max_tokens = sizeof(struct iw_encode_ext) + + IW_ENCODING_TOKEN_MAX, + }, + [SIOCSIWPMKSA - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_pmksa), + .max_tokens = sizeof(struct iw_pmksa), + }, }; static const int standard_ioctl_num = (sizeof(standard_ioctl) / sizeof(struct iw_ioctl_description)); @@ -299,6 +344,31 @@ static const struct iw_ioctl_description standard_event[] = { [IWEVEXPIRED - IWEVFIRST] = { .header_type = IW_HEADER_TYPE_ADDR, }, + [IWEVGENIE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [IWEVMICHAELMICFAILURE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = sizeof(struct iw_michaelmicfailure), + }, + [IWEVASSOCREQIE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [IWEVASSOCRESPIE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [IWEVPMKIDCAND - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = sizeof(struct iw_pmkid_cand), + }, }; static const int standard_event_num = (sizeof(standard_event) / sizeof(struct iw_ioctl_description)); -- cgit v1.2.3 From c04049939f88b29e235d2da217bce6e8ead44f32 Mon Sep 17 00:00:00 2001 From: Steve Grubb Date: Fri, 13 May 2005 18:17:42 +0100 Subject: AUDIT: Add message types to audit records This patch adds more messages types to the audit subsystem so that audit analysis is quicker, intuitive, and more useful. Signed-off-by: Steve Grubb --- I forgot one type in the big patch. I need to add one for user space originating SE Linux avc messages. This is used by dbus and nscd. -Steve --- Updated to 2.6.12-rc4-mm1. -dwmw2 Signed-off-by: David Woodhouse --- include/linux/audit.h | 66 ++++++++++++++++++++++++++--------- kernel/audit.c | 78 +++++++++++++++++++++++++++++------------- kernel/auditsc.c | 42 ++++++++++++++--------- security/selinux/avc.c | 4 +-- security/selinux/hooks.c | 2 +- security/selinux/nlmsgtab.c | 8 +++++ security/selinux/ss/services.c | 4 +-- 7 files changed, 143 insertions(+), 61 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index 405332ebf3c6..1a15ba38c660 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -27,15 +27,53 @@ #include #include -/* Request and reply types */ +/* The netlink messages for the audit system is divided into blocks: + * 1000 - 1099 are for commanding the audit system + * 1100 - 1199 user space trusted application messages + * 1200 - 1299 messages internal to the audit daemon + * 1300 - 1399 audit event messages + * 1400 - 1499 SE Linux use + * 1500 - 1999 future use + * 2000 is for otherwise unclassified kernel audit messages + * + * Messages from 1000-1199 are bi-directional. 1200-1299 are exclusively user + * space. Anything over that is kernel --> user space communication. + */ #define AUDIT_GET 1000 /* Get status */ #define AUDIT_SET 1001 /* Set status (enable/disable/auditd) */ -#define AUDIT_LIST 1002 /* List filtering rules */ -#define AUDIT_ADD 1003 /* Add filtering rule */ -#define AUDIT_DEL 1004 /* Delete filtering rule */ -#define AUDIT_USER 1005 /* Send a message from user-space */ +#define AUDIT_LIST 1002 /* List syscall filtering rules */ +#define AUDIT_ADD 1003 /* Add syscall filtering rule */ +#define AUDIT_DEL 1004 /* Delete syscall filtering rule */ +#define AUDIT_USER 1005 /* Message from userspace -- deprecated */ #define AUDIT_LOGIN 1006 /* Define the login id and information */ -#define AUDIT_SIGNAL_INFO 1010 /* Get information about sender of signal*/ +#define AUDIT_WATCH_INS 1007 /* Insert file/dir watch entry */ +#define AUDIT_WATCH_REM 1008 /* Remove file/dir watch entry */ +#define AUDIT_WATCH_LIST 1009 /* List all file/dir watches */ +#define AUDIT_SIGNAL_INFO 1010 /* Get info about sender of signal to auditd */ + +#define AUDIT_USER_AUTH 1100 /* User space authentication */ +#define AUDIT_USER_ACCT 1101 /* User space acct change */ +#define AUDIT_USER_MGMT 1102 /* User space acct management */ +#define AUDIT_CRED_ACQ 1103 /* User space credential acquired */ +#define AUDIT_CRED_DISP 1104 /* User space credential disposed */ +#define AUDIT_USER_START 1105 /* User space session start */ +#define AUDIT_USER_END 1106 /* User space session end */ +#define AUDIT_USER_AVC 1107 /* User space avc message */ + +#define AUDIT_DAEMON_START 1200 /* Daemon startup record */ +#define AUDIT_DAEMON_END 1201 /* Daemon normal stop record */ +#define AUDIT_DAEMON_ABORT 1202 /* Daemon error stop record */ +#define AUDIT_DAEMON_CONFIG 1203 /* Daemon config change */ + +#define AUDIT_SYSCALL 1300 /* Syscall event */ +#define AUDIT_FS_WATCH 1301 /* Filesystem watch event */ +#define AUDIT_PATH 1302 /* Filname path information */ +#define AUDIT_IPC 1303 /* IPC record */ +#define AUDIT_SOCKET 1304 /* Socket record */ +#define AUDIT_CONFIG_CHANGE 1305 /* Audit system configuration change */ + +#define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ +#define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ #define AUDIT_KERNEL 2000 /* Asynchronous audit record. NOT A REQUEST. */ @@ -216,14 +254,11 @@ extern void audit_signal_info(int sig, struct task_struct *t); #ifdef CONFIG_AUDIT /* These are defined in audit.c */ /* Public API */ -#define audit_log(ctx, fmt, args...) \ - audit_log_type(ctx, AUDIT_KERNEL, 0, fmt, ##args) -extern void audit_log_type(struct audit_context *ctx, int type, - int pid, const char *fmt, ...) - __attribute__((format(printf,4,5))); +extern void audit_log(struct audit_context *ctx, int type, + const char *fmt, ...) + __attribute__((format(printf,3,4))); -extern struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, - int pid); +extern struct audit_buffer *audit_log_start(struct audit_context *ctx,int type); extern void audit_log_format(struct audit_buffer *ab, const char *fmt, ...) __attribute__((format(printf,2,3))); @@ -243,9 +278,8 @@ extern void audit_send_reply(int pid, int seq, int type, void *payload, int size); extern void audit_log_lost(const char *message); #else -#define audit_log(c,f,...) do { ; } while (0) -#define audit_log_type(c,t,p,f,...) do { ; } while (0) -#define audit_log_start(c,t,p) ({ NULL; }) +#define audit_log(c,t,f,...) do { ; } while (0) +#define audit_log_start(c,t) ({ NULL; }) #define audit_log_vformat(b,f,a) do { ; } while (0) #define audit_log_format(b,f,...) do { ; } while (0) #define audit_log_end(b) do { ; } while (0) diff --git a/kernel/audit.c b/kernel/audit.c index 187164572bd0..4e940c05ede8 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -140,6 +140,12 @@ struct audit_buffer { struct audit_context *ctx; /* NULL or associated context */ }; +static void audit_set_pid(struct audit_buffer *ab, pid_t pid) +{ + struct nlmsghdr *nlh = (struct nlmsghdr *)ab->skb->data; + nlh->nlmsg_pid = pid; +} + struct audit_entry { struct list_head list; struct audit_rule rule; @@ -233,7 +239,8 @@ static int audit_set_rate_limit(int limit, uid_t loginuid) { int old = audit_rate_limit; audit_rate_limit = limit; - audit_log(NULL, "audit_rate_limit=%d old=%d by auid %u", + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_rate_limit=%d old=%d by auid %u", audit_rate_limit, old, loginuid); return old; } @@ -242,7 +249,8 @@ static int audit_set_backlog_limit(int limit, uid_t loginuid) { int old = audit_backlog_limit; audit_backlog_limit = limit; - audit_log(NULL, "audit_backlog_limit=%d old=%d by auid %u", + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_backlog_limit=%d old=%d by auid %u", audit_backlog_limit, old, loginuid); return old; } @@ -253,8 +261,9 @@ static int audit_set_enabled(int state, uid_t loginuid) if (state != 0 && state != 1) return -EINVAL; audit_enabled = state; - audit_log(NULL, "audit_enabled=%d old=%d by auid %u", - audit_enabled, old, loginuid); + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_enabled=%d old=%d by auid %u", + audit_enabled, old, loginuid); return old; } @@ -266,8 +275,9 @@ static int audit_set_failure(int state, uid_t loginuid) && state != AUDIT_FAIL_PANIC) return -EINVAL; audit_failure = state; - audit_log(NULL, "audit_failure=%d old=%d by auid %u", - audit_failure, old, loginuid); + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_failure=%d old=%d by auid %u", + audit_failure, old, loginuid); return old; } @@ -316,6 +326,14 @@ static int audit_netlink_ok(kernel_cap_t eff_cap, u16 msg_type) err = -EPERM; break; case AUDIT_USER: + case AUDIT_USER_AUTH: + case AUDIT_USER_ACCT: + case AUDIT_USER_MGMT: + case AUDIT_CRED_ACQ: + case AUDIT_CRED_DISP: + case AUDIT_USER_START: + case AUDIT_USER_END: + case AUDIT_USER_AVC: if (!cap_raised(eff_cap, CAP_AUDIT_WRITE)) err = -EPERM; break; @@ -332,6 +350,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) void *data; struct audit_status *status_get, status_set; int err; + struct audit_buffer *ab; u16 msg_type = nlh->nlmsg_type; uid_t loginuid; /* loginuid of sender */ struct audit_sig_info sig_data; @@ -373,7 +392,8 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) if (status_get->mask & AUDIT_STATUS_PID) { int old = audit_pid; audit_pid = status_get->pid; - audit_log(NULL, "audit_pid=%d old=%d by auid %u", + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_pid=%d old=%d by auid %u", audit_pid, old, loginuid); } if (status_get->mask & AUDIT_STATUS_RATE_LIMIT) @@ -383,13 +403,26 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) loginuid); break; case AUDIT_USER: - audit_log_type(NULL, AUDIT_USER, pid, + case AUDIT_USER_AUTH: + case AUDIT_USER_ACCT: + case AUDIT_USER_MGMT: + case AUDIT_CRED_ACQ: + case AUDIT_CRED_DISP: + case AUDIT_USER_START: + case AUDIT_USER_END: + case AUDIT_USER_AVC: + ab = audit_log_start(NULL, msg_type); + if (!ab) + break; /* audit_panic has been called */ + audit_log_format(ab, "user pid=%d uid=%d length=%d loginuid=%u" " msg='%.1024s'", pid, uid, (int)(nlh->nlmsg_len - ((char *)data - (char *)nlh)), loginuid, (char *)data); + audit_set_pid(ab, pid); + audit_log_end(ab); break; case AUDIT_ADD: case AUDIT_DEL: @@ -504,7 +537,7 @@ static int __init audit_init(void) audit_initialized = 1; audit_enabled = audit_default; - audit_log(NULL, "initialized"); + audit_log(NULL, AUDIT_KERNEL, "initialized"); return 0; } __initcall(audit_init); @@ -541,10 +574,12 @@ static void audit_buffer_free(struct audit_buffer *ab) spin_unlock_irqrestore(&audit_freelist_lock, flags); } -static struct audit_buffer * audit_buffer_alloc(int gfp_mask) +static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx, + int gfp_mask, int type) { unsigned long flags; struct audit_buffer *ab = NULL; + struct nlmsghdr *nlh; spin_lock_irqsave(&audit_freelist_lock, flags); if (!list_empty(&audit_freelist)) { @@ -566,6 +601,12 @@ static struct audit_buffer * audit_buffer_alloc(int gfp_mask) if (!ab->skb) goto err; + ab->ctx = ctx; + nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0)); + nlh->nlmsg_type = type; + nlh->nlmsg_flags = 0; + nlh->nlmsg_pid = 0; + nlh->nlmsg_seq = 0; return ab; err: audit_buffer_free(ab); @@ -578,12 +619,11 @@ err: * syscall, then the syscall is marked as auditable and an audit record * will be written at syscall exit. If there is no associated task, tsk * should be NULL. */ -struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, int pid) +struct audit_buffer *audit_log_start(struct audit_context *ctx, int type) { struct audit_buffer *ab = NULL; struct timespec t; unsigned int serial; - struct nlmsghdr *nlh; if (!audit_initialized) return NULL; @@ -600,19 +640,12 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, int pi return NULL; } - ab = audit_buffer_alloc(GFP_ATOMIC); + ab = audit_buffer_alloc(ctx, GFP_ATOMIC, type); if (!ab) { audit_log_lost("out of memory in audit_log_start"); return NULL; } - ab->ctx = ctx; - nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0)); - nlh->nlmsg_type = type; - nlh->nlmsg_flags = 0; - nlh->nlmsg_pid = pid; - nlh->nlmsg_seq = 0; - if (!audit_get_stamp(ab->ctx, &t, &serial)) { t = CURRENT_TIME; serial = 0; @@ -809,13 +842,12 @@ void audit_log_end(struct audit_buffer *ab) /* Log an audit record. This is a convenience function that calls * audit_log_start, audit_log_vformat, and audit_log_end. It may be * called in any context. */ -void audit_log_type(struct audit_context *ctx, int type, int pid, - const char *fmt, ...) +void audit_log(struct audit_context *ctx, int type, const char *fmt, ...) { struct audit_buffer *ab; va_list args; - ab = audit_log_start(ctx, type, pid); + ab = audit_log_start(ctx, type); if (ab) { va_start(args, fmt); audit_log_vformat(ab, fmt, args); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index d089263253a7..1b7c91f9d5ff 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -286,7 +286,8 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data, err = audit_add_rule(entry, &audit_entlist); if (!err && (flags & AUDIT_AT_EXIT)) err = audit_add_rule(entry, &audit_extlist); - audit_log(NULL, "auid %u added an audit rule\n", loginuid); + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "auid %u added an audit rule\n", loginuid); break; case AUDIT_DEL: flags =((struct audit_rule *)data)->flags; @@ -296,7 +297,8 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data, err = audit_del_rule(data, &audit_entlist); if (!err && (flags & AUDIT_AT_EXIT)) err = audit_del_rule(data, &audit_extlist); - audit_log(NULL, "auid %u removed an audit rule\n", loginuid); + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "auid %u removed an audit rule\n", loginuid); break; default: return -EINVAL; @@ -648,7 +650,7 @@ static void audit_log_exit(struct audit_context *context) int i; struct audit_buffer *ab; - ab = audit_log_start(context, AUDIT_KERNEL, 0); + ab = audit_log_start(context, AUDIT_SYSCALL); if (!ab) return; /* audit_panic has been called */ audit_log_format(ab, "syscall=%d", context->major); @@ -680,28 +682,28 @@ static void audit_log_exit(struct audit_context *context) while (context->aux) { struct audit_aux_data *aux; - ab = audit_log_start(context, AUDIT_KERNEL, 0); + aux = context->aux; + + ab = audit_log_start(context, aux->type); if (!ab) continue; /* audit_panic has been called */ - aux = context->aux; - context->aux = aux->next; - - audit_log_format(ab, "auxitem=%d", aux->type); switch (aux->type) { - case AUDIT_AUX_IPCPERM: { + case AUDIT_IPC: { struct audit_aux_data_ipcctl *axi = (void *)aux; audit_log_format(ab, - " qbytes=%lx uid=%d gid=%d mode=%x", + " qbytes=%lx iuid=%d igid=%d mode=%x", axi->qbytes, axi->uid, axi->gid, axi->mode); } } audit_log_end(ab); + + context->aux = aux->next; kfree(aux); } for (i = 0; i < context->name_count; i++) { - ab = audit_log_start(context, AUDIT_KERNEL, 0); + ab = audit_log_start(context, AUDIT_PATH); if (!ab) continue; /* audit_panic has been called */ audit_log_format(ab, "item=%d", i); @@ -711,7 +713,7 @@ static void audit_log_exit(struct audit_context *context) } if (context->names[i].ino != (unsigned long)-1) audit_log_format(ab, " inode=%lu dev=%02x:%02x mode=%#o" - " uid=%d gid=%d rdev=%02x:%02x", + " ouid=%d ogid=%d rdev=%02x:%02x", context->names[i].ino, MAJOR(context->names[i].dev), MINOR(context->names[i].dev), @@ -1008,10 +1010,16 @@ int audit_get_stamp(struct audit_context *ctx, int audit_set_loginuid(struct task_struct *task, uid_t loginuid) { if (task->audit_context) { - audit_log_type(NULL, AUDIT_LOGIN, 0, - "login pid=%d uid=%u old loginuid=%u new loginuid=%u", - task->pid, task->uid, task->audit_context->loginuid, - loginuid); + struct audit_buffer *ab; + + ab = audit_log_start(NULL, AUDIT_LOGIN); + if (ab) { + audit_log_format(ab, "login pid=%d uid=%u " + "old loginuid=%u new loginuid=%u", + task->pid, task->uid, + task->audit_context->loginuid, loginuid); + audit_log_end(ab); + } task->audit_context->loginuid = loginuid; } return 0; @@ -1039,7 +1047,7 @@ int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode) ax->gid = gid; ax->mode = mode; - ax->d.type = AUDIT_AUX_IPCPERM; + ax->d.type = AUDIT_IPC; ax->d.next = context->aux; context->aux = (void *)ax; return 0; diff --git a/security/selinux/avc.c b/security/selinux/avc.c index 9e71a1bbe011..042f91e9f9d2 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -242,7 +242,7 @@ void __init avc_init(void) avc_node_cachep = kmem_cache_create("avc_node", sizeof(struct avc_node), 0, SLAB_PANIC, NULL, NULL); - audit_log(current->audit_context, "AVC INITIALIZED\n"); + audit_log(current->audit_context, AUDIT_KERNEL, "AVC INITIALIZED\n"); } int avc_get_hash_stats(char *page) @@ -549,7 +549,7 @@ void avc_audit(u32 ssid, u32 tsid, return; } - ab = audit_log_start(current->audit_context, AUDIT_KERNEL, 0); + ab = audit_log_start(current->audit_context, AUDIT_AVC); if (!ab) return; /* audit_panic has been called */ audit_log_format(ab, "avc: %s ", denied ? "denied" : "granted"); diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index aae1e794fe48..db845cbd5841 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -3419,7 +3419,7 @@ static int selinux_nlmsg_perm(struct sock *sk, struct sk_buff *skb) err = selinux_nlmsg_lookup(isec->sclass, nlh->nlmsg_type, &perm); if (err) { if (err == -EINVAL) { - audit_log(current->audit_context, + audit_log(current->audit_context, AUDIT_SELINUX_ERR, "SELinux: unrecognized netlink message" " type=%hu for sclass=%hu\n", nlh->nlmsg_type, isec->sclass); diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index deac14367d43..67e77acc4795 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -98,6 +98,14 @@ static struct nlmsg_perm nlmsg_audit_perms[] = { AUDIT_DEL, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, { AUDIT_USER, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, { AUDIT_SIGNAL_INFO, NETLINK_AUDIT_SOCKET__NLMSG_READ }, + { AUDIT_USER_AUTH, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_ACCT, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_MGMT, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_CRED_ACQ, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_CRED_DISP, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_START, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_END, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_AVC, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, }; diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 5a820cf88c9c..07fdf6ee6148 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -365,7 +365,7 @@ static int security_validtrans_handle_fail(struct context *ocontext, goto out; if (context_struct_to_string(tcontext, &t, &tlen) < 0) goto out; - audit_log(current->audit_context, + audit_log(current->audit_context, AUDIT_SELINUX_ERR, "security_validate_transition: denied for" " oldcontext=%s newcontext=%s taskcontext=%s tclass=%s", o, n, t, policydb.p_class_val_to_name[tclass-1]); @@ -742,7 +742,7 @@ static int compute_sid_handle_invalid_context( goto out; if (context_struct_to_string(newcontext, &n, &nlen) < 0) goto out; - audit_log(current->audit_context, + audit_log(current->audit_context, AUDIT_SELINUX_ERR, "security_compute_sid: invalid context %s" " for scontext=%s" " tcontext=%s" -- cgit v1.2.3 From 23f32d18aa589e228c5a9e12e0d0c67c9b5bcdce Mon Sep 17 00:00:00 2001 From: Steve Grubb Date: Fri, 13 May 2005 18:35:15 +0100 Subject: AUDIT: Fix some spelling errors I'm going through the kernel code and have a patch that corrects several spelling errors in comments. From: Steve Grubb Signed-off-by: David Woodhouse --- include/linux/audit.h | 2 +- kernel/audit.c | 4 ++-- kernel/auditsc.c | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index 1a15ba38c660..51e5879af7fc 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -181,7 +181,7 @@ struct audit_message { struct audit_status { __u32 mask; /* Bit mask for valid entries */ - __u32 enabled; /* 1 = enabled, 0 = disbaled */ + __u32 enabled; /* 1 = enabled, 0 = disabled */ __u32 failure; /* Failure-to-log action */ __u32 pid; /* pid of auditd process */ __u32 rate_limit; /* messages rate limit (per second) */ diff --git a/kernel/audit.c b/kernel/audit.c index 4e940c05ede8..74779d3769fa 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -116,7 +116,7 @@ static LIST_HEAD(audit_entlist); static LIST_HEAD(audit_extlist); /* The netlink socket is only to be read by 1 CPU, which lets us assume - * that list additions and deletions never happen simultaneiously in + * that list additions and deletions never happen simultaneously in * auditsc.c */ static DECLARE_MUTEX(audit_netlink_sem); @@ -775,7 +775,7 @@ void audit_log_d_path(struct audit_buffer *ab, const char *prefix, } } -/* Remove queued messages from the audit_txlist and send them to userspace. */ +/* Remove queued messages from the audit_txlist and send them to user space. */ static void audit_tasklet_handler(unsigned long arg) { LIST_HEAD(list); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 1b7c91f9d5ff..773d28a3f701 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -444,7 +444,7 @@ static enum audit_state audit_filter_task(struct task_struct *tsk) /* At syscall entry and exit time, this filter is called if the * audit_state is not low enough that auditing cannot take place, but is - * also not high enough that we already know we have to write and audit + * also not high enough that we already know we have to write an audit * record (i.e., the state is AUDIT_SETUP_CONTEXT or AUDIT_BUILD_CONTEXT). */ static enum audit_state audit_filter_syscall(struct task_struct *tsk, @@ -750,7 +750,7 @@ void audit_free(struct task_struct *tsk) /* Compute a serial number for the audit record. Audit records are * written to user-space as soon as they are generated, so a complete * audit record may be written in several pieces. The timestamp of the - * record and this serial number are used by the user-space daemon to + * record and this serial number are used by the user-space tools to * determine which pieces belong to the same audit record. The * (timestamp,serial) tuple is unique for each syscall and is live from * syscall entry to syscall exit. -- cgit v1.2.3 From a1365275e745bb0a173c918a52bcdfa6ce122f7e Mon Sep 17 00:00:00 2001 From: Sascha Hauer Date: Thu, 5 May 2005 15:14:15 -0700 Subject: [PATCH] DM9000 network driver This patch adds support for the davicom dm9000 network driver. The dm9000 is found on some embedded arm boards such as the pimx1 or the scb9328. Signed-off-by: Sascha Hauer Signed-off-by: Ben Dooks Signed-off-by: Andrew Morton diff -puN /dev/null drivers/net/dm9000.c --- drivers/net/Kconfig | 12 + drivers/net/Makefile | 1 + drivers/net/dm9000.c | 1219 ++++++++++++++++++++++++++++++++++++++++++++++++ drivers/net/dm9000.h | 135 ++++++ include/linux/dm9000.h | 36 ++ 5 files changed, 1403 insertions(+) create mode 100644 drivers/net/dm9000.c create mode 100644 drivers/net/dm9000.h create mode 100644 include/linux/dm9000.h (limited to 'include/linux') diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 3a0a55b62aaf..4d0c1fb15c7c 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -824,6 +824,18 @@ config SMC9194 . The module will be called smc9194. +config DM9000 + tristate "DM9000 support" + depends on ARM && NET_ETHERNET + select CRC32 + select MII + ---help--- + Support for DM9000 chipset. + + To compile this driver as a module, choose M here and read + . The module will be + called dm9000. + config NET_VENDOR_RACAL bool "Racal-Interlan (Micom) NI cards" depends on NET_ETHERNET && ISA diff --git a/drivers/net/Makefile b/drivers/net/Makefile index 6202b10dbb4d..cbd11688608e 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -179,6 +179,7 @@ obj-$(CONFIG_AMD8111_ETH) += amd8111e.o obj-$(CONFIG_IBMVETH) += ibmveth.o obj-$(CONFIG_S2IO) += s2io.o obj-$(CONFIG_SMC91X) += smc91x.o +obj-$(CONFIG_DM9000) += dm9000.o obj-$(CONFIG_FEC_8XX) += fec_8xx/ obj-$(CONFIG_ARM) += arm/ diff --git a/drivers/net/dm9000.c b/drivers/net/dm9000.c new file mode 100644 index 000000000000..f4ba0ffb8637 --- /dev/null +++ b/drivers/net/dm9000.c @@ -0,0 +1,1219 @@ +/* + * dm9000.c: Version 1.2 03/18/2003 + * + * A Davicom DM9000 ISA NIC fast Ethernet driver for Linux. + * Copyright (C) 1997 Sten Wang + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * (C)Copyright 1997-1998 DAVICOM Semiconductor,Inc. All Rights Reserved. + * + * V0.11 06/20/2001 REG_0A bit3=1, default enable BP with DA match + * 06/22/2001 Support DM9801 progrmming + * E3: R25 = ((R24 + NF) & 0x00ff) | 0xf000 + * E4: R25 = ((R24 + NF) & 0x00ff) | 0xc200 + * R17 = (R17 & 0xfff0) | NF + 3 + * E5: R25 = ((R24 + NF - 3) & 0x00ff) | 0xc200 + * R17 = (R17 & 0xfff0) | NF + * + * v1.00 modify by simon 2001.9.5 + * change for kernel 2.4.x + * + * v1.1 11/09/2001 fix force mode bug + * + * v1.2 03/18/2003 Weilun Huang : + * Fixed phy reset. + * Added tx/rx 32 bit mode. + * Cleaned up for kernel merge. + * + * 03/03/2004 Sascha Hauer + * Port to 2.6 kernel + * + * 24-Sep-2004 Ben Dooks + * Cleanup of code to remove ifdefs + * Allowed platform device data to influence access width + * Reformatting areas of code + * + * 17-Mar-2005 Sascha Hauer + * * removed 2.4 style module parameters + * * removed removed unused stat counter and fixed + * net_device_stats + * * introduced tx_timeout function + * * reworked locking + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "dm9000.h" + +/* Board/System/Debug information/definition ---------------- */ + +#define DM9000_PHY 0x40 /* PHY address 0x01 */ + +#define TRUE 1 +#define FALSE 0 + +#define CARDNAME "dm9000" +#define PFX CARDNAME ": " + +#define DM9000_TIMER_WUT jiffies+(HZ*2) /* timer wakeup time : 2 second */ + +#define DM9000_DEBUG 0 + +#if DM9000_DEBUG > 2 +#define PRINTK3(args...) printk(CARDNAME ": " args) +#else +#define PRINTK3(args...) do { } while(0) +#endif + +#if DM9000_DEBUG > 1 +#define PRINTK2(args...) printk(CARDNAME ": " args) +#else +#define PRINTK2(args...) do { } while(0) +#endif + +#if DM9000_DEBUG > 0 +#define PRINTK1(args...) printk(CARDNAME ": " args) +#define PRINTK(args...) printk(CARDNAME ": " args) +#else +#define PRINTK1(args...) do { } while(0) +#define PRINTK(args...) printk(KERN_DEBUG args) +#endif + +/* + * Transmit timeout, default 5 seconds. + */ +static int watchdog = 5000; +module_param(watchdog, int, 0400); +MODULE_PARM_DESC(watchdog, "transmit timeout in milliseconds"); + +/* Structure/enum declaration ------------------------------- */ +typedef struct board_info { + + void __iomem *io_addr; /* Register I/O base address */ + void __iomem *io_data; /* Data I/O address */ + u16 irq; /* IRQ */ + + u16 tx_pkt_cnt; + u16 queue_pkt_len; + u16 queue_start_addr; + u16 dbug_cnt; + u8 io_mode; /* 0:word, 2:byte */ + u8 phy_addr; + + void (*inblk)(void __iomem *port, void *data, int length); + void (*outblk)(void __iomem *port, void *data, int length); + void (*dumpblk)(void __iomem *port, int length); + + struct resource *addr_res; /* resources found */ + struct resource *data_res; + struct resource *addr_req; /* resources requested */ + struct resource *data_req; + struct resource *irq_res; + + struct timer_list timer; + struct net_device_stats stats; + unsigned char srom[128]; + spinlock_t lock; + + struct mii_if_info mii; + u32 msg_enable; +} board_info_t; + +/* function declaration ------------------------------------- */ +static int dm9000_probe(struct device *); +static int dm9000_open(struct net_device *); +static int dm9000_start_xmit(struct sk_buff *, struct net_device *); +static int dm9000_stop(struct net_device *); +static int dm9000_do_ioctl(struct net_device *, struct ifreq *, int); + + +static void dm9000_timer(unsigned long); +static void dm9000_init_dm9000(struct net_device *); + +static struct net_device_stats *dm9000_get_stats(struct net_device *); + +static irqreturn_t dm9000_interrupt(int, void *, struct pt_regs *); + +static int dm9000_phy_read(struct net_device *dev, int phyaddr_unsused, int reg); +static void dm9000_phy_write(struct net_device *dev, int phyaddr_unused, int reg, + int value); +static u16 read_srom_word(board_info_t *, int); +static void dm9000_rx(struct net_device *); +static void dm9000_hash_table(struct net_device *); + +//#define DM9000_PROGRAM_EEPROM +#ifdef DM9000_PROGRAM_EEPROM +static void program_eeprom(board_info_t * db); +#endif +/* DM9000 network board routine ---------------------------- */ + +static void +dm9000_reset(board_info_t * db) +{ + PRINTK1("dm9000x: resetting\n"); + /* RESET device */ + writeb(DM9000_NCR, db->io_addr); + udelay(200); + writeb(NCR_RST, db->io_data); + udelay(200); +} + +/* + * Read a byte from I/O port + */ +static u8 +ior(board_info_t * db, int reg) +{ + writeb(reg, db->io_addr); + return readb(db->io_data); +} + +/* + * Write a byte to I/O port + */ + +static void +iow(board_info_t * db, int reg, int value) +{ + writeb(reg, db->io_addr); + writeb(value, db->io_data); +} + +/* routines for sending block to chip */ + +static void dm9000_outblk_8bit(void __iomem *reg, void *data, int count) +{ + writesb(reg, data, count); +} + +static void dm9000_outblk_16bit(void __iomem *reg, void *data, int count) +{ + writesw(reg, data, (count+1) >> 1); +} + +static void dm9000_outblk_32bit(void __iomem *reg, void *data, int count) +{ + writesl(reg, data, (count+3) >> 2); +} + +/* input block from chip to memory */ + +static void dm9000_inblk_8bit(void __iomem *reg, void *data, int count) +{ + readsb(reg, data, count+1); +} + + +static void dm9000_inblk_16bit(void __iomem *reg, void *data, int count) +{ + readsw(reg, data, (count+1) >> 1); +} + +static void dm9000_inblk_32bit(void __iomem *reg, void *data, int count) +{ + readsl(reg, data, (count+3) >> 2); +} + +/* dump block from chip to null */ + +static void dm9000_dumpblk_8bit(void __iomem *reg, int count) +{ + int i; + int tmp; + + for (i = 0; i < count; i++) + tmp = readb(reg); +} + +static void dm9000_dumpblk_16bit(void __iomem *reg, int count) +{ + int i; + int tmp; + + count = (count + 1) >> 1; + + for (i = 0; i < count; i++) + tmp = readw(reg); +} + +static void dm9000_dumpblk_32bit(void __iomem *reg, int count) +{ + int i; + int tmp; + + count = (count + 3) >> 2; + + for (i = 0; i < count; i++) + tmp = readl(reg); +} + +/* dm9000_set_io + * + * select the specified set of io routines to use with the + * device + */ + +static void dm9000_set_io(struct board_info *db, int byte_width) +{ + /* use the size of the data resource to work out what IO + * routines we want to use + */ + + switch (byte_width) { + case 1: + db->dumpblk = dm9000_dumpblk_8bit; + db->outblk = dm9000_outblk_8bit; + db->inblk = dm9000_inblk_8bit; + break; + + case 2: + db->dumpblk = dm9000_dumpblk_16bit; + db->outblk = dm9000_outblk_16bit; + db->inblk = dm9000_inblk_16bit; + break; + + case 3: + printk(KERN_ERR PFX ": 3 byte IO, falling back to 16bit\n"); + db->dumpblk = dm9000_dumpblk_16bit; + db->outblk = dm9000_outblk_16bit; + db->inblk = dm9000_inblk_16bit; + break; + + case 4: + default: + db->dumpblk = dm9000_dumpblk_32bit; + db->outblk = dm9000_outblk_32bit; + db->inblk = dm9000_inblk_32bit; + break; + } +} + + +/* Our watchdog timed out. Called by the networking layer */ +static void dm9000_timeout(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + u8 reg_save; + unsigned long flags; + + /* Save previous register address */ + reg_save = readb(db->io_addr); + spin_lock_irqsave(db->lock,flags); + + netif_stop_queue(dev); + dm9000_reset(db); + dm9000_init_dm9000(dev); + /* We can accept TX packets again */ + dev->trans_start = jiffies; + netif_wake_queue(dev); + + /* Restore previous register address */ + writeb(reg_save, db->io_addr); + spin_unlock_irqrestore(db->lock,flags); +} + + +/* dm9000_release_board + * + * release a board, and any mapped resources + */ + +static void +dm9000_release_board(struct platform_device *pdev, struct board_info *db) +{ + if (db->data_res == NULL) { + if (db->addr_res != NULL) + release_mem_region((unsigned long)db->io_addr, 4); + return; + } + + /* unmap our resources */ + + iounmap(db->io_addr); + iounmap(db->io_data); + + /* release the resources */ + + if (db->data_req != NULL) { + release_resource(db->data_req); + kfree(db->data_req); + } + + if (db->addr_res != NULL) { + release_resource(db->data_req); + kfree(db->addr_req); + } +} + +#define res_size(_r) (((_r)->end - (_r)->start) + 1) + +/* + * Search DM9000 board, allocate space and register it + */ +static int +dm9000_probe(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct dm9000_plat_data *pdata = pdev->dev.platform_data; + struct board_info *db; /* Point a board information structure */ + struct net_device *ndev; + unsigned long base; + int ret = 0; + int iosize; + int i; + u32 id_val; + + printk(KERN_INFO "%s Ethernet Driver\n", CARDNAME); + + /* Init network device */ + ndev = alloc_etherdev(sizeof (struct board_info)); + if (!ndev) { + printk("%s: could not allocate device.\n", CARDNAME); + return -ENOMEM; + } + + SET_MODULE_OWNER(ndev); + SET_NETDEV_DEV(ndev, dev); + + PRINTK2("dm9000_probe()"); + + /* setup board info structure */ + db = (struct board_info *) ndev->priv; + memset(db, 0, sizeof (*db)); + + if (pdev->num_resources < 2) { + ret = -ENODEV; + goto out; + } + + switch (pdev->num_resources) { + case 2: + base = pdev->resource[0].start; + + if (!request_mem_region(base, 4, ndev->name)) { + ret = -EBUSY; + goto out; + } + + ndev->base_addr = base; + ndev->irq = pdev->resource[1].start; + db->io_addr = (void *)base; + db->io_data = (void *)(base + 4); + + break; + + case 3: + db->addr_res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + db->data_res = platform_get_resource(pdev, IORESOURCE_MEM, 1); + db->irq_res = platform_get_resource(pdev, IORESOURCE_IRQ, 0); + + if (db->addr_res == NULL || db->data_res == NULL) { + printk(KERN_ERR PFX "insufficient resources\n"); + ret = -ENOENT; + goto out; + } + + i = res_size(db->addr_res); + db->addr_req = request_mem_region(db->addr_res->start, i, + pdev->name); + + if (db->addr_req == NULL) { + printk(KERN_ERR PFX "cannot claim address reg area\n"); + ret = -EIO; + goto out; + } + + db->io_addr = ioremap(db->addr_res->start, i); + + if (db->io_addr == NULL) { + printk(KERN_ERR "failed to ioremap address reg\n"); + ret = -EINVAL; + goto out; + } + + iosize = res_size(db->data_res); + db->data_req = request_mem_region(db->data_res->start, iosize, + pdev->name); + + if (db->data_req == NULL) { + printk(KERN_ERR PFX "cannot claim data reg area\n"); + ret = -EIO; + goto out; + } + + db->io_data = ioremap(db->data_res->start, iosize); + + if (db->io_data == NULL) { + printk(KERN_ERR "failed to ioremap data reg\n"); + ret = -EINVAL; + goto out; + } + + /* fill in parameters for net-dev structure */ + + ndev->base_addr = (unsigned long)db->io_addr; + ndev->irq = db->irq_res->start; + + /* ensure at least we have a default set of IO routines */ + dm9000_set_io(db, iosize); + + } + + /* check to see if anything is being over-ridden */ + if (pdata != NULL) { + /* check to see if the driver wants to over-ride the + * default IO width */ + + if (pdata->flags & DM9000_PLATF_8BITONLY) + dm9000_set_io(db, 1); + + if (pdata->flags & DM9000_PLATF_16BITONLY) + dm9000_set_io(db, 2); + + if (pdata->flags & DM9000_PLATF_32BITONLY) + dm9000_set_io(db, 4); + + /* check to see if there are any IO routine + * over-rides */ + + if (pdata->inblk != NULL) + db->inblk = pdata->inblk; + + if (pdata->outblk != NULL) + db->outblk = pdata->outblk; + + if (pdata->dumpblk != NULL) + db->dumpblk = pdata->dumpblk; + } + + dm9000_reset(db); + + /* try two times, DM9000 sometimes gets the first read wrong */ + for (i = 0; i < 2; i++) { + id_val = ior(db, DM9000_VIDL); + id_val |= (u32)ior(db, DM9000_VIDH) << 8; + id_val |= (u32)ior(db, DM9000_PIDL) << 16; + id_val |= (u32)ior(db, DM9000_PIDH) << 24; + + if (id_val == DM9000_ID) + break; + printk("%s: read wrong id 0x%08x\n", CARDNAME, id_val); + } + + if (id_val != DM9000_ID) { + printk("%s: wrong id: 0x%08x\n", CARDNAME, id_val); + goto release; + } + + /* from this point we assume that we have found a DM9000 */ + + /* driver system function */ + ether_setup(ndev); + + ndev->open = &dm9000_open; + ndev->hard_start_xmit = &dm9000_start_xmit; + ndev->tx_timeout = &dm9000_timeout; + ndev->watchdog_timeo = msecs_to_jiffies(watchdog); + ndev->stop = &dm9000_stop; + ndev->get_stats = &dm9000_get_stats; + ndev->set_multicast_list = &dm9000_hash_table; + ndev->do_ioctl = &dm9000_do_ioctl; + +#ifdef DM9000_PROGRAM_EEPROM + program_eeprom(db); +#endif + db->msg_enable = NETIF_MSG_LINK; + db->mii.phy_id_mask = 0x1f; + db->mii.reg_num_mask = 0x1f; + db->mii.force_media = 0; + db->mii.full_duplex = 0; + db->mii.dev = ndev; + db->mii.mdio_read = dm9000_phy_read; + db->mii.mdio_write = dm9000_phy_write; + + /* Read SROM content */ + for (i = 0; i < 64; i++) + ((u16 *) db->srom)[i] = read_srom_word(db, i); + + /* Set Node Address */ + for (i = 0; i < 6; i++) + ndev->dev_addr[i] = db->srom[i]; + + if (!is_valid_ether_addr(ndev->dev_addr)) + printk("%s: Invalid ethernet MAC address. Please " + "set using ifconfig\n", ndev->name); + + dev_set_drvdata(dev, ndev); + ret = register_netdev(ndev); + + if (ret == 0) { + printk("%s: dm9000 at %p,%p IRQ %d MAC: ", + ndev->name, db->io_addr, db->io_data, ndev->irq); + for (i = 0; i < 5; i++) + printk("%02x:", ndev->dev_addr[i]); + printk("%02x\n", ndev->dev_addr[5]); + } + return 0; + + release: + out: + printk("%s: not found (%d).\n", CARDNAME, ret); + + dm9000_release_board(pdev, db); + kfree(ndev); + + return ret; +} + +/* + * Open the interface. + * The interface is opened whenever "ifconfig" actives it. + */ +static int +dm9000_open(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + + PRINTK2("entering dm9000_open\n"); + + if (request_irq(dev->irq, &dm9000_interrupt, SA_SHIRQ, dev->name, dev)) + return -EAGAIN; + + /* Initialize DM9000 board */ + dm9000_reset(db); + dm9000_init_dm9000(dev); + + /* Init driver variable */ + db->dbug_cnt = 0; + + /* set and active a timer process */ + init_timer(&db->timer); + db->timer.expires = DM9000_TIMER_WUT * 2; + db->timer.data = (unsigned long) dev; + db->timer.function = &dm9000_timer; + add_timer(&db->timer); + + mii_check_media(&db->mii, netif_msg_link(db), 1); + netif_start_queue(dev); + + return 0; +} + +/* + * Initilize dm9000 board + */ +static void +dm9000_init_dm9000(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + + PRINTK1("entering %s\n",__FUNCTION__); + + /* I/O mode */ + db->io_mode = ior(db, DM9000_ISR) >> 6; /* ISR bit7:6 keeps I/O mode */ + + /* GPIO0 on pre-activate PHY */ + iow(db, DM9000_GPR, 0); /* REG_1F bit0 activate phyxcer */ + iow(db, DM9000_GPCR, GPCR_GEP_CNTL); /* Let GPIO0 output */ + iow(db, DM9000_GPR, 0); /* Enable PHY */ + + /* Program operating register */ + iow(db, DM9000_TCR, 0); /* TX Polling clear */ + iow(db, DM9000_BPTR, 0x3f); /* Less 3Kb, 200us */ + iow(db, DM9000_FCR, 0xff); /* Flow Control */ + iow(db, DM9000_SMCR, 0); /* Special Mode */ + /* clear TX status */ + iow(db, DM9000_NSR, NSR_WAKEST | NSR_TX2END | NSR_TX1END); + iow(db, DM9000_ISR, ISR_CLR_STATUS); /* Clear interrupt status */ + + /* Set address filter table */ + dm9000_hash_table(dev); + + /* Activate DM9000 */ + iow(db, DM9000_RCR, RCR_DIS_LONG | RCR_DIS_CRC | RCR_RXEN); + /* Enable TX/RX interrupt mask */ + iow(db, DM9000_IMR, IMR_PAR | IMR_PTM | IMR_PRM); + + /* Init Driver variable */ + db->tx_pkt_cnt = 0; + db->queue_pkt_len = 0; + dev->trans_start = 0; + spin_lock_init(&db->lock); +} + +/* + * Hardware start transmission. + * Send a packet to media from the upper layer. + */ +static int +dm9000_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + + PRINTK3("dm9000_start_xmit\n"); + + if (db->tx_pkt_cnt > 1) + return 1; + + netif_stop_queue(dev); + + /* Disable all interrupts */ + iow(db, DM9000_IMR, IMR_PAR); + + /* Move data to DM9000 TX RAM */ + writeb(DM9000_MWCMD, db->io_addr); + + (db->outblk)(db->io_data, skb->data, skb->len); + db->stats.tx_bytes += skb->len; + + /* TX control: First packet immediately send, second packet queue */ + if (db->tx_pkt_cnt == 0) { + + /* First Packet */ + db->tx_pkt_cnt++; + + /* Set TX length to DM9000 */ + iow(db, DM9000_TXPLL, skb->len & 0xff); + iow(db, DM9000_TXPLH, (skb->len >> 8) & 0xff); + + /* Issue TX polling command */ + iow(db, DM9000_TCR, TCR_TXREQ); /* Cleared after TX complete */ + + dev->trans_start = jiffies; /* save the time stamp */ + + } else { + /* Second packet */ + db->tx_pkt_cnt++; + db->queue_pkt_len = skb->len; + } + + /* free this SKB */ + dev_kfree_skb(skb); + + /* Re-enable resource check */ + if (db->tx_pkt_cnt == 1) + netif_wake_queue(dev); + + /* Re-enable interrupt */ + iow(db, DM9000_IMR, IMR_PAR | IMR_PTM | IMR_PRM); + + return 0; +} + +static void +dm9000_shutdown(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + + /* RESET device */ + dm9000_phy_write(dev, 0, MII_BMCR, BMCR_RESET); /* PHY RESET */ + iow(db, DM9000_GPR, 0x01); /* Power-Down PHY */ + iow(db, DM9000_IMR, IMR_PAR); /* Disable all interrupt */ + iow(db, DM9000_RCR, 0x00); /* Disable RX */ +} + +/* + * Stop the interface. + * The interface is stopped when it is brought. + */ +static int +dm9000_stop(struct net_device *ndev) +{ + board_info_t *db = (board_info_t *) ndev->priv; + + PRINTK1("entering %s\n",__FUNCTION__); + + /* deleted timer */ + del_timer(&db->timer); + + netif_stop_queue(ndev); + netif_carrier_off(ndev); + + /* free interrupt */ + free_irq(ndev->irq, ndev); + + dm9000_shutdown(ndev); + + return 0; +} + +/* + * DM9000 interrupt handler + * receive the packet to upper layer, free the transmitted packet + */ + +void +dm9000_tx_done(struct net_device *dev, board_info_t * db) +{ + int tx_status = ior(db, DM9000_NSR); /* Got TX status */ + + if (tx_status & (NSR_TX2END | NSR_TX1END)) { + /* One packet sent complete */ + db->tx_pkt_cnt--; + db->stats.tx_packets++; + + /* Queue packet check & send */ + if (db->tx_pkt_cnt > 0) { + iow(db, DM9000_TXPLL, db->queue_pkt_len & 0xff); + iow(db, DM9000_TXPLH, (db->queue_pkt_len >> 8) & 0xff); + iow(db, DM9000_TCR, TCR_TXREQ); + dev->trans_start = jiffies; + } + netif_wake_queue(dev); + } +} + +static irqreturn_t +dm9000_interrupt(int irq, void *dev_id, struct pt_regs *regs) +{ + struct net_device *dev = dev_id; + board_info_t *db; + int int_status; + u8 reg_save; + + PRINTK3("entering %s\n",__FUNCTION__); + + if (!dev) { + PRINTK1("dm9000_interrupt() without DEVICE arg\n"); + return IRQ_HANDLED; + } + + /* A real interrupt coming */ + db = (board_info_t *) dev->priv; + spin_lock(&db->lock); + + /* Save previous register address */ + reg_save = readb(db->io_addr); + + /* Disable all interrupts */ + iow(db, DM9000_IMR, IMR_PAR); + + /* Got DM9000 interrupt status */ + int_status = ior(db, DM9000_ISR); /* Got ISR */ + iow(db, DM9000_ISR, int_status); /* Clear ISR status */ + + /* Received the coming packet */ + if (int_status & ISR_PRS) + dm9000_rx(dev); + + /* Trnasmit Interrupt check */ + if (int_status & ISR_PTS) + dm9000_tx_done(dev, db); + + /* Re-enable interrupt mask */ + iow(db, DM9000_IMR, IMR_PAR | IMR_PTM | IMR_PRM); + + /* Restore previous register address */ + writeb(reg_save, db->io_addr); + + spin_unlock(&db->lock); + + return IRQ_HANDLED; +} + +/* + * Get statistics from driver. + */ +static struct net_device_stats * +dm9000_get_stats(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + return &db->stats; +} + +/* + * Process the upper socket ioctl command + */ +static int +dm9000_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +{ + PRINTK1("entering %s\n",__FUNCTION__); + return 0; +} + +/* + * A periodic timer routine + * Dynamic media sense, allocated Rx buffer... + */ +static void +dm9000_timer(unsigned long data) +{ + struct net_device *dev = (struct net_device *) data; + board_info_t *db = (board_info_t *) dev->priv; + u8 reg_save; + unsigned long flags; + + PRINTK3("dm9000_timer()\n"); + + spin_lock_irqsave(db->lock,flags); + /* Save previous register address */ + reg_save = readb(db->io_addr); + + mii_check_media(&db->mii, netif_msg_link(db), 0); + + /* Restore previous register address */ + writeb(reg_save, db->io_addr); + spin_unlock_irqrestore(db->lock,flags); + + /* Set timer again */ + db->timer.expires = DM9000_TIMER_WUT; + add_timer(&db->timer); +} + +struct dm9000_rxhdr { + u16 RxStatus; + u16 RxLen; +} __attribute__((__packed__)); + +/* + * Received a packet and pass to upper layer + */ +static void +dm9000_rx(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + struct dm9000_rxhdr rxhdr; + struct sk_buff *skb; + u8 rxbyte, *rdptr; + int GoodPacket; + int RxLen; + + /* Check packet ready or not */ + do { + ior(db, DM9000_MRCMDX); /* Dummy read */ + + /* Get most updated data */ + rxbyte = readb(db->io_data); + + /* Status check: this byte must be 0 or 1 */ + if (rxbyte > DM9000_PKT_RDY) { + printk("status check failed: %d\n", rxbyte); + iow(db, DM9000_RCR, 0x00); /* Stop Device */ + iow(db, DM9000_ISR, IMR_PAR); /* Stop INT request */ + return; + } + + if (rxbyte != DM9000_PKT_RDY) + return; + + /* A packet ready now & Get status/length */ + GoodPacket = TRUE; + writeb(DM9000_MRCMD, db->io_addr); + + (db->inblk)(db->io_data, &rxhdr, sizeof(rxhdr)); + + RxLen = rxhdr.RxLen; + + /* Packet Status check */ + if (RxLen < 0x40) { + GoodPacket = FALSE; + PRINTK1("Bad Packet received (runt)\n"); + } + + if (RxLen > DM9000_PKT_MAX) { + PRINTK1("RST: RX Len:%x\n", RxLen); + } + + if (rxhdr.RxStatus & 0xbf00) { + GoodPacket = FALSE; + if (rxhdr.RxStatus & 0x100) { + PRINTK1("fifo error\n"); + db->stats.rx_fifo_errors++; + } + if (rxhdr.RxStatus & 0x200) { + PRINTK1("crc error\n"); + db->stats.rx_crc_errors++; + } + if (rxhdr.RxStatus & 0x8000) { + PRINTK1("length error\n"); + db->stats.rx_length_errors++; + } + } + + /* Move data from DM9000 */ + if (GoodPacket + && ((skb = dev_alloc_skb(RxLen + 4)) != NULL)) { + skb->dev = dev; + skb_reserve(skb, 2); + rdptr = (u8 *) skb_put(skb, RxLen - 4); + + /* Read received packet from RX SRAM */ + + (db->inblk)(db->io_data, rdptr, RxLen); + db->stats.rx_bytes += RxLen; + + /* Pass to upper layer */ + skb->protocol = eth_type_trans(skb, dev); + netif_rx(skb); + db->stats.rx_packets++; + + } else { + /* need to dump the packet's data */ + + (db->dumpblk)(db->io_data, RxLen); + } + } while (rxbyte == DM9000_PKT_RDY); +} + +/* + * Read a word data from SROM + */ +static u16 +read_srom_word(board_info_t * db, int offset) +{ + iow(db, DM9000_EPAR, offset); + iow(db, DM9000_EPCR, EPCR_ERPRR); + mdelay(8); /* according to the datasheet 200us should be enough, + but it doesn't work */ + iow(db, DM9000_EPCR, 0x0); + return (ior(db, DM9000_EPDRL) + (ior(db, DM9000_EPDRH) << 8)); +} + +#ifdef DM9000_PROGRAM_EEPROM +/* + * Write a word data to SROM + */ +static void +write_srom_word(board_info_t * db, int offset, u16 val) +{ + iow(db, DM9000_EPAR, offset); + iow(db, DM9000_EPDRH, ((val >> 8) & 0xff)); + iow(db, DM9000_EPDRL, (val & 0xff)); + iow(db, DM9000_EPCR, EPCR_WEP | EPCR_ERPRW); + mdelay(8); /* same shit */ + iow(db, DM9000_EPCR, 0); +} + +/* + * Only for development: + * Here we write static data to the eeprom in case + * we don't have valid content on a new board + */ +static void +program_eeprom(board_info_t * db) +{ + u16 eeprom[] = { 0x0c00, 0x007f, 0x1300, /* MAC Address */ + 0x0000, /* Autoload: accept nothing */ + 0x0a46, 0x9000, /* Vendor / Product ID */ + 0x0000, /* pin control */ + 0x0000, + }; /* Wake-up mode control */ + int i; + for (i = 0; i < 8; i++) + write_srom_word(db, i, eeprom[i]); +} +#endif + + +/* + * Calculate the CRC valude of the Rx packet + * flag = 1 : return the reverse CRC (for the received packet CRC) + * 0 : return the normal CRC (for Hash Table index) + */ + +static unsigned long +cal_CRC(unsigned char *Data, unsigned int Len, u8 flag) +{ + + u32 crc = ether_crc_le(Len, Data); + + if (flag) + return ~crc; + + return crc; +} + +/* + * Set DM9000 multicast address + */ +static void +dm9000_hash_table(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + struct dev_mc_list *mcptr = dev->mc_list; + int mc_cnt = dev->mc_count; + u32 hash_val; + u16 i, oft, hash_table[4]; + unsigned long flags; + + PRINTK2("dm9000_hash_table()\n"); + + spin_lock_irqsave(&db->lock,flags); + + for (i = 0, oft = 0x10; i < 6; i++, oft++) + iow(db, oft, dev->dev_addr[i]); + + /* Clear Hash Table */ + for (i = 0; i < 4; i++) + hash_table[i] = 0x0; + + /* broadcast address */ + hash_table[3] = 0x8000; + + /* the multicast address in Hash Table : 64 bits */ + for (i = 0; i < mc_cnt; i++, mcptr = mcptr->next) { + hash_val = cal_CRC((char *) mcptr->dmi_addr, 6, 0) & 0x3f; + hash_table[hash_val / 16] |= (u16) 1 << (hash_val % 16); + } + + /* Write the hash table to MAC MD table */ + for (i = 0, oft = 0x16; i < 4; i++) { + iow(db, oft++, hash_table[i] & 0xff); + iow(db, oft++, (hash_table[i] >> 8) & 0xff); + } + + spin_unlock_irqrestore(&db->lock,flags); +} + + +/* + * Read a word from phyxcer + */ +static int +dm9000_phy_read(struct net_device *dev, int phy_reg_unused, int reg) +{ + board_info_t *db = (board_info_t *) dev->priv; + unsigned long flags; + int ret; + + spin_lock_irqsave(&db->lock,flags); + /* Fill the phyxcer register into REG_0C */ + iow(db, DM9000_EPAR, DM9000_PHY | reg); + + iow(db, DM9000_EPCR, 0xc); /* Issue phyxcer read command */ + udelay(100); /* Wait read complete */ + iow(db, DM9000_EPCR, 0x0); /* Clear phyxcer read command */ + + /* The read data keeps on REG_0D & REG_0E */ + ret = (ior(db, DM9000_EPDRH) << 8) | ior(db, DM9000_EPDRL); + + spin_unlock_irqrestore(&db->lock,flags); + + return ret; +} + +/* + * Write a word to phyxcer + */ +static void +dm9000_phy_write(struct net_device *dev, int phyaddr_unused, int reg, int value) +{ + board_info_t *db = (board_info_t *) dev->priv; + unsigned long flags; + + spin_lock_irqsave(&db->lock,flags); + + /* Fill the phyxcer register into REG_0C */ + iow(db, DM9000_EPAR, DM9000_PHY | reg); + + /* Fill the written data into REG_0D & REG_0E */ + iow(db, DM9000_EPDRL, (value & 0xff)); + iow(db, DM9000_EPDRH, ((value >> 8) & 0xff)); + + iow(db, DM9000_EPCR, 0xa); /* Issue phyxcer write command */ + udelay(500); /* Wait write complete */ + iow(db, DM9000_EPCR, 0x0); /* Clear phyxcer write command */ + + spin_unlock_irqrestore(&db->lock,flags); +} + +static int +dm9000_drv_suspend(struct device *dev, u32 state, u32 level) +{ + struct net_device *ndev = dev_get_drvdata(dev); + + if (ndev && level == SUSPEND_DISABLE) { + if (netif_running(ndev)) { + netif_device_detach(ndev); + dm9000_shutdown(ndev); + } + } + return 0; +} + +static int +dm9000_drv_resume(struct device *dev, u32 level) +{ + struct net_device *ndev = dev_get_drvdata(dev); + board_info_t *db = (board_info_t *) ndev->priv; + + if (ndev && level == RESUME_ENABLE) { + + if (netif_running(ndev)) { + dm9000_reset(db); + dm9000_init_dm9000(ndev); + + netif_device_attach(ndev); + } + } + return 0; +} + +static int +dm9000_drv_remove(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct net_device *ndev = dev_get_drvdata(dev); + + dev_set_drvdata(dev, NULL); + + unregister_netdev(ndev); + dm9000_release_board(pdev, (board_info_t *) ndev->priv); + kfree(ndev); /* free device structure */ + + PRINTK1("clean_module() exit\n"); + + return 0; +} + +static struct device_driver dm9000_driver = { + .name = "dm9000", + .bus = &platform_bus_type, + .probe = dm9000_probe, + .remove = dm9000_drv_remove, + .suspend = dm9000_drv_suspend, + .resume = dm9000_drv_resume, +}; + +static int __init +dm9000_init(void) +{ + return driver_register(&dm9000_driver); /* search board and register */ +} + +static void __exit +dm9000_cleanup(void) +{ + driver_unregister(&dm9000_driver); +} + +module_init(dm9000_init); +module_exit(dm9000_cleanup); + +MODULE_AUTHOR("Sascha Hauer, Ben Dooks"); +MODULE_DESCRIPTION("Davicom DM9000 network driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/dm9000.h b/drivers/net/dm9000.h new file mode 100644 index 000000000000..82cad360bafc --- /dev/null +++ b/drivers/net/dm9000.h @@ -0,0 +1,135 @@ +/* + * dm9000 Ethernet + */ + +#ifndef _DM9000X_H_ +#define _DM9000X_H_ + +#define DM9000_ID 0x90000A46 + +/* although the registers are 16 bit, they are 32-bit aligned. + */ + +#define DM9000_NCR 0x00 +#define DM9000_NSR 0x01 +#define DM9000_TCR 0x02 +#define DM9000_TSR1 0x03 +#define DM9000_TSR2 0x04 +#define DM9000_RCR 0x05 +#define DM9000_RSR 0x06 +#define DM9000_ROCR 0x07 +#define DM9000_BPTR 0x08 +#define DM9000_FCTR 0x09 +#define DM9000_FCR 0x0A +#define DM9000_EPCR 0x0B +#define DM9000_EPAR 0x0C +#define DM9000_EPDRL 0x0D +#define DM9000_EPDRH 0x0E +#define DM9000_WCR 0x0F + +#define DM9000_PAR 0x10 +#define DM9000_MAR 0x16 + +#define DM9000_GPCR 0x1e +#define DM9000_GPR 0x1f +#define DM9000_TRPAL 0x22 +#define DM9000_TRPAH 0x23 +#define DM9000_RWPAL 0x24 +#define DM9000_RWPAH 0x25 + +#define DM9000_VIDL 0x28 +#define DM9000_VIDH 0x29 +#define DM9000_PIDL 0x2A +#define DM9000_PIDH 0x2B + +#define DM9000_CHIPR 0x2C +#define DM9000_SMCR 0x2F + +#define DM9000_MRCMDX 0xF0 +#define DM9000_MRCMD 0xF2 +#define DM9000_MRRL 0xF4 +#define DM9000_MRRH 0xF5 +#define DM9000_MWCMDX 0xF6 +#define DM9000_MWCMD 0xF8 +#define DM9000_MWRL 0xFA +#define DM9000_MWRH 0xFB +#define DM9000_TXPLL 0xFC +#define DM9000_TXPLH 0xFD +#define DM9000_ISR 0xFE +#define DM9000_IMR 0xFF + +#define NCR_EXT_PHY (1<<7) +#define NCR_WAKEEN (1<<6) +#define NCR_FCOL (1<<4) +#define NCR_FDX (1<<3) +#define NCR_LBK (3<<1) +#define NCR_RST (1<<0) + +#define NSR_SPEED (1<<7) +#define NSR_LINKST (1<<6) +#define NSR_WAKEST (1<<5) +#define NSR_TX2END (1<<3) +#define NSR_TX1END (1<<2) +#define NSR_RXOV (1<<1) + +#define TCR_TJDIS (1<<6) +#define TCR_EXCECM (1<<5) +#define TCR_PAD_DIS2 (1<<4) +#define TCR_CRC_DIS2 (1<<3) +#define TCR_PAD_DIS1 (1<<2) +#define TCR_CRC_DIS1 (1<<1) +#define TCR_TXREQ (1<<0) + +#define TSR_TJTO (1<<7) +#define TSR_LC (1<<6) +#define TSR_NC (1<<5) +#define TSR_LCOL (1<<4) +#define TSR_COL (1<<3) +#define TSR_EC (1<<2) + +#define RCR_WTDIS (1<<6) +#define RCR_DIS_LONG (1<<5) +#define RCR_DIS_CRC (1<<4) +#define RCR_ALL (1<<3) +#define RCR_RUNT (1<<2) +#define RCR_PRMSC (1<<1) +#define RCR_RXEN (1<<0) + +#define RSR_RF (1<<7) +#define RSR_MF (1<<6) +#define RSR_LCS (1<<5) +#define RSR_RWTO (1<<4) +#define RSR_PLE (1<<3) +#define RSR_AE (1<<2) +#define RSR_CE (1<<1) +#define RSR_FOE (1<<0) + +#define FCTR_HWOT(ot) (( ot & 0xf ) << 4 ) +#define FCTR_LWOT(ot) ( ot & 0xf ) + +#define IMR_PAR (1<<7) +#define IMR_ROOM (1<<3) +#define IMR_ROM (1<<2) +#define IMR_PTM (1<<1) +#define IMR_PRM (1<<0) + +#define ISR_ROOS (1<<3) +#define ISR_ROS (1<<2) +#define ISR_PTS (1<<1) +#define ISR_PRS (1<<0) +#define ISR_CLR_STATUS (ISR_ROOS | ISR_ROS | ISR_PTS | ISR_PRS) + +#define EPCR_REEP (1<<5) +#define EPCR_WEP (1<<4) +#define EPCR_EPOS (1<<3) +#define EPCR_ERPRR (1<<2) +#define EPCR_ERPRW (1<<1) +#define EPCR_ERRE (1<<0) + +#define GPCR_GEP_CNTL (1<<0) + +#define DM9000_PKT_RDY 0x01 /* Packet ready to receive */ +#define DM9000_PKT_MAX 1536 /* Received packet max size */ + +#endif /* _DM9000X_H_ */ + diff --git a/include/linux/dm9000.h b/include/linux/dm9000.h new file mode 100644 index 000000000000..0008e2ad0c9f --- /dev/null +++ b/include/linux/dm9000.h @@ -0,0 +1,36 @@ +/* include/linux/dm9000.h + * + * Copyright (c) 2004 Simtec Electronics + * Ben Dooks + * + * Header file for dm9000 platform data + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * +*/ + +#ifndef __DM9000_PLATFORM_DATA +#define __DM9000_PLATFORM_DATA __FILE__ + +/* IO control flags */ + +#define DM9000_PLATF_8BITONLY (0x0001) +#define DM9000_PLATF_16BITONLY (0x0002) +#define DM9000_PLATF_32BITONLY (0x0004) + +/* platfrom data for platfrom device structure's platfrom_data field */ + +struct dm9000_plat_data { + unsigned int flags; + + /* allow replacement IO routines */ + + void (*inblk)(void __iomem *reg, void *data, int len); + void (*outblk)(void __iomem *reg, void *data, int len); + void (*dumpblk)(void __iomem *reg, int len); +}; + +#endif /* __DM9000_PLATFORM_DATA */ + -- cgit v1.2.3 From b3dd65f958354226275522b5a64157834bdc5415 Mon Sep 17 00:00:00 2001 From: Krzysztof Halasa Date: Thu, 21 Apr 2005 15:57:25 +0200 Subject: [PATCH] Generic HDLC update The attached patch updates generic HDLC to version 1.18. FR Cisco LMI production-tested. Please apply to Linux 2.6. Thanks. Changes: - doc updates - added Cisco LMI support to Frame-Relay code - cleaned hdlc_fr.c a bit, removed some orphaned #defines etc. - fixed a problem with non-functional LMI in FR DCE mode. - changed diagnostic messages to better conform to FR standards - all protocols: information about carrier changes (DCD line) is now printed to kernel logs. Signed-Off-By: Krzysztof Halasa --- Documentation/networking/generic-hdlc.txt | 51 ++--- drivers/net/wan/hdlc_fr.c | 320 ++++++++++++++++-------------- drivers/net/wan/hdlc_generic.c | 16 +- include/linux/hdlc.h | 4 +- 4 files changed, 216 insertions(+), 175 deletions(-) (limited to 'include/linux') diff --git a/Documentation/networking/generic-hdlc.txt b/Documentation/networking/generic-hdlc.txt index 7d1dc6b884f3..31bc8b759b75 100644 --- a/Documentation/networking/generic-hdlc.txt +++ b/Documentation/networking/generic-hdlc.txt @@ -1,21 +1,21 @@ Generic HDLC layer Krzysztof Halasa -January, 2003 Generic HDLC layer currently supports: -- Frame Relay (ANSI, CCITT and no LMI), with ARP support (no InARP). - Normal (routed) and Ethernet-bridged (Ethernet device emulation) - interfaces can share a single PVC. -- raw HDLC - either IP (IPv4) interface or Ethernet device emulation. -- Cisco HDLC, -- PPP (uses syncppp.c), -- X.25 (uses X.25 routines). - -There are hardware drivers for the following cards: -- C101 by Moxa Technologies Co., Ltd. -- RISCom/N2 by SDL Communications Inc. -- and others, some not in the official kernel. +1. Frame Relay (ANSI, CCITT, Cisco and no LMI). + - Normal (routed) and Ethernet-bridged (Ethernet device emulation) + interfaces can share a single PVC. + - ARP support (no InARP support in the kernel - there is an + experimental InARP user-space daemon available on: + http://www.kernel.org/pub/linux/utils/net/hdlc/). +2. raw HDLC - either IP (IPv4) interface or Ethernet device emulation. +3. Cisco HDLC. +4. PPP (uses syncppp.c). +5. X.25 (uses X.25 routines). + +Generic HDLC is a protocol driver only - it needs a low-level driver +for your particular hardware. Ethernet device emulation (using HDLC or Frame-Relay PVC) is compatible with IEEE 802.1Q (VLANs) and 802.1D (Ethernet bridging). @@ -24,7 +24,7 @@ with IEEE 802.1Q (VLANs) and 802.1D (Ethernet bridging). Make sure the hdlc.o and the hardware driver are loaded. It should create a number of "hdlc" (hdlc0 etc) network devices, one for each WAN port. You'll need the "sethdlc" utility, get it from: - http://hq.pm.waw.pl/hdlc/ + http://www.kernel.org/pub/linux/utils/net/hdlc/ Compile sethdlc.c utility: gcc -O2 -Wall -o sethdlc sethdlc.c @@ -52,12 +52,12 @@ Setting interface: * v35 | rs232 | x21 | t1 | e1 - sets physical interface for a given port if the card has software-selectable interfaces loopback - activate hardware loopback (for testing only) -* clock ext - external clock (uses DTE RX and TX clock) -* clock int - internal clock (provides clock signal on DCE clock output) -* clock txint - TX internal, RX external (provides TX clock on DCE output) -* clock txfromrx - TX clock derived from RX clock (TX clock on DCE output) -* rate - sets clock rate in bps (not required for external clock or - for txfromrx) +* clock ext - both RX clock and TX clock external +* clock int - both RX clock and TX clock internal +* clock txint - RX clock external, TX clock internal +* clock txfromrx - RX clock external, TX clock derived from RX clock +* rate - sets clock rate in bps (for "int" or "txint" clock only) + Setting protocol: @@ -79,7 +79,7 @@ Setting protocol: * x25 - sets X.25 mode * fr - Frame Relay mode - lmi ansi / ccitt / none - LMI (link management) type + lmi ansi / ccitt / cisco / none - LMI (link management) type dce - Frame Relay DCE (network) side LMI instead of default DTE (user). It has nothing to do with clocks! t391 - link integrity verification polling timer (in seconds) - user @@ -119,13 +119,14 @@ or -If you have a problem with N2 or C101 card, you can issue the "private" -command to see port's packet descriptor rings (in kernel logs): +If you have a problem with N2, C101 or PLX200SYN card, you can issue the +"private" command to see port's packet descriptor rings (in kernel logs): sethdlc hdlc0 private -The hardware driver has to be build with CONFIG_HDLC_DEBUG_RINGS. +The hardware driver has to be build with #define DEBUG_RINGS. Attaching this info to bug reports would be helpful. Anyway, let me know if you have problems using this. -For patches and other info look at http://hq.pm.waw.pl/hdlc/ +For patches and other info look at: +. diff --git a/drivers/net/wan/hdlc_fr.c b/drivers/net/wan/hdlc_fr.c index 7f450b51a6cb..a5d6891c9d4c 100644 --- a/drivers/net/wan/hdlc_fr.c +++ b/drivers/net/wan/hdlc_fr.c @@ -2,7 +2,7 @@ * Generic HDLC support routines for Linux * Frame Relay support * - * Copyright (C) 1999 - 2003 Krzysztof Halasa + * Copyright (C) 1999 - 2005 Krzysztof Halasa * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License @@ -27,6 +27,10 @@ active = open and "link reliable" exist = new = not used + CCITT LMI: ITU-T Q.933 Annex A + ANSI LMI: ANSI T1.617 Annex D + CISCO LMI: the original, aka "Gang of Four" LMI + */ #include @@ -49,45 +53,41 @@ #undef DEBUG_ECN #undef DEBUG_LINK -#define MAXLEN_LMISTAT 20 /* max size of status enquiry frame */ - -#define PVC_STATE_NEW 0x01 -#define PVC_STATE_ACTIVE 0x02 -#define PVC_STATE_FECN 0x08 /* FECN condition */ -#define PVC_STATE_BECN 0x10 /* BECN condition */ - - -#define FR_UI 0x03 -#define FR_PAD 0x00 - -#define NLPID_IP 0xCC -#define NLPID_IPV6 0x8E -#define NLPID_SNAP 0x80 -#define NLPID_PAD 0x00 -#define NLPID_Q933 0x08 - - -#define LMI_DLCI 0 /* LMI DLCI */ -#define LMI_PROTO 0x08 -#define LMI_CALLREF 0x00 /* Call Reference */ -#define LMI_ANSI_LOCKSHIFT 0x95 /* ANSI lockshift */ -#define LMI_REPTYPE 1 /* report type */ -#define LMI_CCITT_REPTYPE 0x51 -#define LMI_ALIVE 3 /* keep alive */ -#define LMI_CCITT_ALIVE 0x53 -#define LMI_PVCSTAT 7 /* pvc status */ -#define LMI_CCITT_PVCSTAT 0x57 -#define LMI_FULLREP 0 /* full report */ -#define LMI_INTEGRITY 1 /* link integrity report */ -#define LMI_SINGLE 2 /* single pvc report */ +#define FR_UI 0x03 +#define FR_PAD 0x00 + +#define NLPID_IP 0xCC +#define NLPID_IPV6 0x8E +#define NLPID_SNAP 0x80 +#define NLPID_PAD 0x00 +#define NLPID_CCITT_ANSI_LMI 0x08 +#define NLPID_CISCO_LMI 0x09 + + +#define LMI_CCITT_ANSI_DLCI 0 /* LMI DLCI */ +#define LMI_CISCO_DLCI 1023 + +#define LMI_CALLREF 0x00 /* Call Reference */ +#define LMI_ANSI_LOCKSHIFT 0x95 /* ANSI locking shift */ +#define LMI_ANSI_CISCO_REPTYPE 0x01 /* report type */ +#define LMI_CCITT_REPTYPE 0x51 +#define LMI_ANSI_CISCO_ALIVE 0x03 /* keep alive */ +#define LMI_CCITT_ALIVE 0x53 +#define LMI_ANSI_CISCO_PVCSTAT 0x07 /* PVC status */ +#define LMI_CCITT_PVCSTAT 0x57 + +#define LMI_FULLREP 0x00 /* full report */ +#define LMI_INTEGRITY 0x01 /* link integrity report */ +#define LMI_SINGLE 0x02 /* single PVC report */ + #define LMI_STATUS_ENQUIRY 0x75 #define LMI_STATUS 0x7D /* reply */ #define LMI_REPT_LEN 1 /* report type element length */ #define LMI_INTEG_LEN 2 /* link integrity element length */ -#define LMI_LENGTH 13 /* standard LMI frame length */ -#define LMI_ANSI_LENGTH 14 +#define LMI_CCITT_CISCO_LENGTH 13 /* LMI frame lengths */ +#define LMI_ANSI_LENGTH 14 typedef struct { @@ -223,51 +223,34 @@ static inline struct net_device** get_dev_p(pvc_device *pvc, int type) } -static inline u16 status_to_dlci(u8 *status, int *active, int *new) -{ - *new = (status[2] & 0x08) ? 1 : 0; - *active = (status[2] & 0x02) ? 1 : 0; - - return ((status[0] & 0x3F) << 4) | ((status[1] & 0x78) >> 3); -} - - -static inline void dlci_to_status(u16 dlci, u8 *status, int active, int new) -{ - status[0] = (dlci >> 4) & 0x3F; - status[1] = ((dlci << 3) & 0x78) | 0x80; - status[2] = 0x80; - - if (new) - status[2] |= 0x08; - else if (active) - status[2] |= 0x02; -} - - - static int fr_hard_header(struct sk_buff **skb_p, u16 dlci) { u16 head_len; struct sk_buff *skb = *skb_p; switch (skb->protocol) { - case __constant_ntohs(ETH_P_IP): + case __constant_ntohs(NLPID_CCITT_ANSI_LMI): head_len = 4; skb_push(skb, head_len); - skb->data[3] = NLPID_IP; + skb->data[3] = NLPID_CCITT_ANSI_LMI; break; - case __constant_ntohs(ETH_P_IPV6): + case __constant_ntohs(NLPID_CISCO_LMI): head_len = 4; skb_push(skb, head_len); - skb->data[3] = NLPID_IPV6; + skb->data[3] = NLPID_CISCO_LMI; break; - case __constant_ntohs(LMI_PROTO): + case __constant_ntohs(ETH_P_IP): + head_len = 4; + skb_push(skb, head_len); + skb->data[3] = NLPID_IP; + break; + + case __constant_ntohs(ETH_P_IPV6): head_len = 4; skb_push(skb, head_len); - skb->data[3] = LMI_PROTO; + skb->data[3] = NLPID_IPV6; break; case __constant_ntohs(ETH_P_802_3): @@ -461,13 +444,14 @@ static void fr_lmi_send(struct net_device *dev, int fullrep) hdlc_device *hdlc = dev_to_hdlc(dev); struct sk_buff *skb; pvc_device *pvc = hdlc->state.fr.first_pvc; - int len = (hdlc->state.fr.settings.lmi == LMI_ANSI) ? LMI_ANSI_LENGTH - : LMI_LENGTH; - int stat_len = 3; + int lmi = hdlc->state.fr.settings.lmi; + int dce = hdlc->state.fr.settings.dce; + int len = lmi == LMI_ANSI ? LMI_ANSI_LENGTH : LMI_CCITT_CISCO_LENGTH; + int stat_len = (lmi == LMI_CISCO) ? 6 : 3; u8 *data; int i = 0; - if (hdlc->state.fr.settings.dce && fullrep) { + if (dce && fullrep) { len += hdlc->state.fr.dce_pvc_count * (2 + stat_len); if (len > HDLC_MAX_MRU) { printk(KERN_WARNING "%s: Too many PVCs while sending " @@ -484,29 +468,31 @@ static void fr_lmi_send(struct net_device *dev, int fullrep) } memset(skb->data, 0, len); skb_reserve(skb, 4); - skb->protocol = __constant_htons(LMI_PROTO); - fr_hard_header(&skb, LMI_DLCI); + if (lmi == LMI_CISCO) { + skb->protocol = __constant_htons(NLPID_CISCO_LMI); + fr_hard_header(&skb, LMI_CISCO_DLCI); + } else { + skb->protocol = __constant_htons(NLPID_CCITT_ANSI_LMI); + fr_hard_header(&skb, LMI_CCITT_ANSI_DLCI); + } data = skb->tail; data[i++] = LMI_CALLREF; - data[i++] = hdlc->state.fr.settings.dce - ? LMI_STATUS : LMI_STATUS_ENQUIRY; - if (hdlc->state.fr.settings.lmi == LMI_ANSI) + data[i++] = dce ? LMI_STATUS : LMI_STATUS_ENQUIRY; + if (lmi == LMI_ANSI) data[i++] = LMI_ANSI_LOCKSHIFT; - data[i++] = (hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_REPTYPE : LMI_REPTYPE; + data[i++] = lmi == LMI_CCITT ? LMI_CCITT_REPTYPE : + LMI_ANSI_CISCO_REPTYPE; data[i++] = LMI_REPT_LEN; data[i++] = fullrep ? LMI_FULLREP : LMI_INTEGRITY; - - data[i++] = (hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_ALIVE : LMI_ALIVE; + data[i++] = lmi == LMI_CCITT ? LMI_CCITT_ALIVE : LMI_ANSI_CISCO_ALIVE; data[i++] = LMI_INTEG_LEN; data[i++] = hdlc->state.fr.txseq =fr_lmi_nextseq(hdlc->state.fr.txseq); data[i++] = hdlc->state.fr.rxseq; - if (hdlc->state.fr.settings.dce && fullrep) { + if (dce && fullrep) { while (pvc) { - data[i++] = (hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_PVCSTAT : LMI_PVCSTAT; + data[i++] = lmi == LMI_CCITT ? LMI_CCITT_PVCSTAT : + LMI_ANSI_CISCO_PVCSTAT; data[i++] = stat_len; /* LMI start/restart */ @@ -523,8 +509,20 @@ static void fr_lmi_send(struct net_device *dev, int fullrep) fr_log_dlci_active(pvc); } - dlci_to_status(pvc->dlci, data + i, - pvc->state.active, pvc->state.new); + if (lmi == LMI_CISCO) { + data[i] = pvc->dlci >> 8; + data[i + 1] = pvc->dlci & 0xFF; + } else { + data[i] = (pvc->dlci >> 4) & 0x3F; + data[i + 1] = ((pvc->dlci << 3) & 0x78) | 0x80; + data[i + 2] = 0x80; + } + + if (pvc->state.new) + data[i + 2] |= 0x08; + else if (pvc->state.active) + data[i + 2] |= 0x02; + i += stat_len; pvc = pvc->next; } @@ -569,6 +567,8 @@ static void fr_set_link_state(int reliable, struct net_device *dev) pvc_carrier(0, pvc); pvc->state.exist = pvc->state.active = 0; pvc->state.new = 0; + if (!hdlc->state.fr.settings.dce) + pvc->state.bandwidth = 0; pvc = pvc->next; } } @@ -583,11 +583,12 @@ static void fr_timer(unsigned long arg) int i, cnt = 0, reliable; u32 list; - if (hdlc->state.fr.settings.dce) + if (hdlc->state.fr.settings.dce) { reliable = hdlc->state.fr.request && time_before(jiffies, hdlc->state.fr.last_poll + hdlc->state.fr.settings.t392 * HZ); - else { + hdlc->state.fr.request = 0; + } else { hdlc->state.fr.last_errors <<= 1; /* Shift the list */ if (hdlc->state.fr.request) { if (hdlc->state.fr.reliable) @@ -634,65 +635,88 @@ static void fr_timer(unsigned long arg) static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) { hdlc_device *hdlc = dev_to_hdlc(dev); - int stat_len; pvc_device *pvc; - int reptype = -1, error, no_ram; u8 rxseq, txseq; - int i; + int lmi = hdlc->state.fr.settings.lmi; + int dce = hdlc->state.fr.settings.dce; + int stat_len = (lmi == LMI_CISCO) ? 6 : 3, reptype, error, no_ram, i; - if (skb->len < ((hdlc->state.fr.settings.lmi == LMI_ANSI) - ? LMI_ANSI_LENGTH : LMI_LENGTH)) { + if (skb->len < (lmi == LMI_ANSI ? LMI_ANSI_LENGTH : + LMI_CCITT_CISCO_LENGTH)) { printk(KERN_INFO "%s: Short LMI frame\n", dev->name); return 1; } - if (skb->data[5] != (!hdlc->state.fr.settings.dce ? - LMI_STATUS : LMI_STATUS_ENQUIRY)) { - printk(KERN_INFO "%s: LMI msgtype=%x, Not LMI status %s\n", - dev->name, skb->data[2], - hdlc->state.fr.settings.dce ? "enquiry" : "reply"); + if (skb->data[3] != (lmi == LMI_CISCO ? NLPID_CISCO_LMI : + NLPID_CCITT_ANSI_LMI)) { + printk(KERN_INFO "%s: Received non-LMI frame with LMI" + " DLCI\n", dev->name); + return 1; + } + + if (skb->data[4] != LMI_CALLREF) { + printk(KERN_INFO "%s: Invalid LMI Call reference (0x%02X)\n", + dev->name, skb->data[4]); + return 1; + } + + if (skb->data[5] != (dce ? LMI_STATUS_ENQUIRY : LMI_STATUS)) { + printk(KERN_INFO "%s: Invalid LMI Message type (0x%02X)\n", + dev->name, skb->data[5]); return 1; } - i = (hdlc->state.fr.settings.lmi == LMI_ANSI) ? 7 : 6; + if (lmi == LMI_ANSI) { + if (skb->data[6] != LMI_ANSI_LOCKSHIFT) { + printk(KERN_INFO "%s: Not ANSI locking shift in LMI" + " message (0x%02X)\n", dev->name, skb->data[6]); + return 1; + } + i = 7; + } else + i = 6; - if (skb->data[i] != - ((hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_REPTYPE : LMI_REPTYPE)) { - printk(KERN_INFO "%s: Not a report type=%x\n", + if (skb->data[i] != (lmi == LMI_CCITT ? LMI_CCITT_REPTYPE : + LMI_ANSI_CISCO_REPTYPE)) { + printk(KERN_INFO "%s: Not an LMI Report type IE (0x%02X)\n", dev->name, skb->data[i]); return 1; } - i++; - i++; /* Skip length field */ + if (skb->data[++i] != LMI_REPT_LEN) { + printk(KERN_INFO "%s: Invalid LMI Report type IE length" + " (%u)\n", dev->name, skb->data[i]); + return 1; + } - reptype = skb->data[i++]; + reptype = skb->data[++i]; + if (reptype != LMI_INTEGRITY && reptype != LMI_FULLREP) { + printk(KERN_INFO "%s: Unsupported LMI Report type (0x%02X)\n", + dev->name, reptype); + return 1; + } - if (skb->data[i]!= - ((hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_ALIVE : LMI_ALIVE)) { - printk(KERN_INFO "%s: Unsupported status element=%x\n", - dev->name, skb->data[i]); + if (skb->data[++i] != (lmi == LMI_CCITT ? LMI_CCITT_ALIVE : + LMI_ANSI_CISCO_ALIVE)) { + printk(KERN_INFO "%s: Not an LMI Link integrity verification" + " IE (0x%02X)\n", dev->name, skb->data[i]); return 1; } - i++; - i++; /* Skip length field */ + if (skb->data[++i] != LMI_INTEG_LEN) { + printk(KERN_INFO "%s: Invalid LMI Link integrity verification" + " IE length (%u)\n", dev->name, skb->data[i]); + return 1; + } + i++; hdlc->state.fr.rxseq = skb->data[i++]; /* TX sequence from peer */ rxseq = skb->data[i++]; /* Should confirm our sequence */ txseq = hdlc->state.fr.txseq; - if (hdlc->state.fr.settings.dce) { - if (reptype != LMI_FULLREP && reptype != LMI_INTEGRITY) { - printk(KERN_INFO "%s: Unsupported report type=%x\n", - dev->name, reptype); - return 1; - } + if (dce) hdlc->state.fr.last_poll = jiffies; - } error = 0; if (!hdlc->state.fr.reliable) @@ -703,7 +727,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) error = 1; } - if (hdlc->state.fr.settings.dce) { + if (dce) { if (hdlc->state.fr.fullrep_sent && !error) { /* Stop sending full report - the last one has been confirmed by DTE */ hdlc->state.fr.fullrep_sent = 0; @@ -725,6 +749,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) hdlc->state.fr.dce_changed = 0; } + hdlc->state.fr.request = 1; /* got request */ fr_lmi_send(dev, reptype == LMI_FULLREP ? 1 : 0); return 0; } @@ -739,7 +764,6 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) if (reptype != LMI_FULLREP) return 0; - stat_len = 3; pvc = hdlc->state.fr.first_pvc; while (pvc) { @@ -750,24 +774,35 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) no_ram = 0; while (skb->len >= i + 2 + stat_len) { u16 dlci; + u32 bw; unsigned int active, new; - if (skb->data[i] != ((hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_PVCSTAT : LMI_PVCSTAT)) { - printk(KERN_WARNING "%s: Invalid PVCSTAT ID: %x\n", - dev->name, skb->data[i]); + if (skb->data[i] != (lmi == LMI_CCITT ? LMI_CCITT_PVCSTAT : + LMI_ANSI_CISCO_PVCSTAT)) { + printk(KERN_INFO "%s: Not an LMI PVC status IE" + " (0x%02X)\n", dev->name, skb->data[i]); return 1; } - i++; - if (skb->data[i] != stat_len) { - printk(KERN_WARNING "%s: Invalid PVCSTAT length: %x\n", - dev->name, skb->data[i]); + if (skb->data[++i] != stat_len) { + printk(KERN_INFO "%s: Invalid LMI PVC status IE length" + " (%u)\n", dev->name, skb->data[i]); return 1; } i++; - dlci = status_to_dlci(skb->data + i, &active, &new); + new = !! (skb->data[i + 2] & 0x08); + active = !! (skb->data[i + 2] & 0x02); + if (lmi == LMI_CISCO) { + dlci = (skb->data[i] << 8) | skb->data[i + 1]; + bw = (skb->data[i + 3] << 16) | + (skb->data[i + 4] << 8) | + (skb->data[i + 5]); + } else { + dlci = ((skb->data[i] & 0x3F) << 4) | + ((skb->data[i + 1] & 0x78) >> 3); + bw = 0; + } pvc = add_pvc(dev, dlci); @@ -783,9 +818,11 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) pvc->state.deleted = 0; if (active != pvc->state.active || new != pvc->state.new || + bw != pvc->state.bandwidth || !pvc->state.exist) { pvc->state.new = new; pvc->state.active = active; + pvc->state.bandwidth = bw; pvc_carrier(active, pvc); fr_log_dlci_active(pvc); } @@ -801,6 +838,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) pvc_carrier(0, pvc); pvc->state.active = pvc->state.new = 0; pvc->state.exist = 0; + pvc->state.bandwidth = 0; fr_log_dlci_active(pvc); } pvc = pvc->next; @@ -829,22 +867,15 @@ static int fr_rx(struct sk_buff *skb) dlci = q922_to_dlci(skb->data); - if (dlci == LMI_DLCI) { - if (hdlc->state.fr.settings.lmi == LMI_NONE) - goto rx_error; /* LMI packet with no LMI? */ - - if (data[3] == LMI_PROTO) { - if (fr_lmi_recv(ndev, skb)) - goto rx_error; - else { - dev_kfree_skb_any(skb); - return NET_RX_SUCCESS; - } - } - - printk(KERN_INFO "%s: Received non-LMI frame with LMI DLCI\n", - ndev->name); - goto rx_error; + if ((dlci == LMI_CCITT_ANSI_DLCI && + (hdlc->state.fr.settings.lmi == LMI_ANSI || + hdlc->state.fr.settings.lmi == LMI_CCITT)) || + (dlci == LMI_CISCO_DLCI && + hdlc->state.fr.settings.lmi == LMI_CISCO)) { + if (fr_lmi_recv(ndev, skb)) + goto rx_error; + dev_kfree_skb_any(skb); + return NET_RX_SUCCESS; } pvc = find_pvc(hdlc, dlci); @@ -1170,7 +1201,8 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr) if ((new_settings.lmi != LMI_NONE && new_settings.lmi != LMI_ANSI && - new_settings.lmi != LMI_CCITT) || + new_settings.lmi != LMI_CCITT && + new_settings.lmi != LMI_CISCO) || new_settings.t391 < 1 || new_settings.t392 < 2 || new_settings.n391 < 1 || diff --git a/drivers/net/wan/hdlc_generic.c b/drivers/net/wan/hdlc_generic.c index 6ed064cb4469..a63f6a2cc4f7 100644 --- a/drivers/net/wan/hdlc_generic.c +++ b/drivers/net/wan/hdlc_generic.c @@ -1,7 +1,7 @@ /* * Generic HDLC support routines for Linux * - * Copyright (C) 1999 - 2003 Krzysztof Halasa + * Copyright (C) 1999 - 2005 Krzysztof Halasa * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License @@ -38,7 +38,7 @@ #include -static const char* version = "HDLC support module revision 1.17"; +static const char* version = "HDLC support module revision 1.18"; #undef DEBUG_LINK @@ -126,10 +126,13 @@ void hdlc_set_carrier(int on, struct net_device *dev) if (!hdlc->open) goto carrier_exit; - if (hdlc->carrier) + if (hdlc->carrier) { + printk(KERN_INFO "%s: Carrier detected\n", dev->name); __hdlc_set_carrier_on(dev); - else + } else { + printk(KERN_INFO "%s: Carrier lost\n", dev->name); __hdlc_set_carrier_off(dev); + } carrier_exit: spin_unlock_irqrestore(&hdlc->state_lock, flags); @@ -157,8 +160,11 @@ int hdlc_open(struct net_device *dev) spin_lock_irq(&hdlc->state_lock); - if (hdlc->carrier) + if (hdlc->carrier) { + printk(KERN_INFO "%s: Carrier detected\n", dev->name); __hdlc_set_carrier_on(dev); + } else + printk(KERN_INFO "%s: No carrier\n", dev->name); hdlc->open = 1; diff --git a/include/linux/hdlc.h b/include/linux/hdlc.h index 503194e62fe1..ed2927ef1ff7 100644 --- a/include/linux/hdlc.h +++ b/include/linux/hdlc.h @@ -1,7 +1,7 @@ /* * Generic HDLC support routines for Linux * - * Copyright (C) 1999-2003 Krzysztof Halasa + * Copyright (C) 1999-2005 Krzysztof Halasa * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License @@ -41,6 +41,7 @@ #define LMI_NONE 1 /* No LMI, all PVCs are static */ #define LMI_ANSI 2 /* ANSI Annex D */ #define LMI_CCITT 3 /* ITU-T Annex A */ +#define LMI_CISCO 4 /* The "original" LMI, aka Gang of Four */ #define HDLC_MAX_MTU 1500 /* Ethernet 1500 bytes */ #define HDLC_MAX_MRU (HDLC_MAX_MTU + 10 + 14 + 4) /* for ETH+VLAN over FR */ @@ -89,6 +90,7 @@ typedef struct pvc_device_struct { unsigned int deleted: 1; unsigned int fecn: 1; unsigned int becn: 1; + unsigned int bandwidth; /* Cisco LMI reporting only */ }state; }pvc_device; -- cgit v1.2.3 From 3ec3b2fba526ead2fa3f3d7c91924f39a0733749 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Tue, 17 May 2005 12:08:48 +0100 Subject: AUDIT: Capture sys_socketcall arguments and sockaddrs Signed-off-by: David Woodhouse --- include/linux/audit.h | 7 ++++- kernel/auditsc.c | 73 +++++++++++++++++++++++++++++++++++++++++++++++++-- net/socket.c | 9 +++++-- 3 files changed, 84 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index 51e5879af7fc..2f5dc60f8bbd 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -69,8 +69,9 @@ #define AUDIT_FS_WATCH 1301 /* Filesystem watch event */ #define AUDIT_PATH 1302 /* Filname path information */ #define AUDIT_IPC 1303 /* IPC record */ -#define AUDIT_SOCKET 1304 /* Socket record */ +#define AUDIT_SOCKETCALL 1304 /* sys_socketcall arguments */ #define AUDIT_CONFIG_CHANGE 1305 /* Audit system configuration change */ +#define AUDIT_SOCKADDR 1306 /* sockaddr copied as syscall arg */ #define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ #define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ @@ -235,6 +236,8 @@ extern int audit_get_stamp(struct audit_context *ctx, extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); +extern int audit_socketcall(int nargs, unsigned long *args); +extern int audit_sockaddr(int len, void *addr); extern void audit_signal_info(int sig, struct task_struct *t); #else #define audit_alloc(t) ({ 0; }) @@ -248,6 +251,8 @@ extern void audit_signal_info(int sig, struct task_struct *t); #define audit_get_stamp(c,t,s) ({ 0; }) #define audit_get_loginuid(c) ({ -1; }) #define audit_ipc_perms(q,u,g,m) ({ 0; }) +#define audit_socketcall(n,a) ({ 0; }) +#define audit_sockaddr(len, addr) ({ 0; }) #define audit_signal_info(s,t) do { ; } while (0) #endif diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 773d28a3f701..818778d5b6ad 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -34,7 +34,7 @@ #include #include #include - +#include #include #include #include @@ -112,6 +112,18 @@ struct audit_aux_data_ipcctl { mode_t mode; }; +struct audit_aux_data_socketcall { + struct audit_aux_data d; + int nargs; + unsigned long args[0]; +}; + +struct audit_aux_data_sockaddr { + struct audit_aux_data d; + int len; + char a[0]; +}; + /* The per-task audit context. */ struct audit_context { @@ -694,7 +706,22 @@ static void audit_log_exit(struct audit_context *context) audit_log_format(ab, " qbytes=%lx iuid=%d igid=%d mode=%x", axi->qbytes, axi->uid, axi->gid, axi->mode); - } + break; } + + case AUDIT_SOCKETCALL: { + int i; + struct audit_aux_data_socketcall *axs = (void *)aux; + audit_log_format(ab, "nargs=%d", axs->nargs); + for (i=0; inargs; i++) + audit_log_format(ab, " a%d=%lx", i, axs->args[i]); + break; } + + case AUDIT_SOCKADDR: { + struct audit_aux_data_sockaddr *axs = (void *)aux; + + audit_log_format(ab, "saddr="); + audit_log_hex(ab, axs->a, axs->len); + break; } } audit_log_end(ab); @@ -1053,6 +1080,48 @@ int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode) return 0; } +int audit_socketcall(int nargs, unsigned long *args) +{ + struct audit_aux_data_socketcall *ax; + struct audit_context *context = current->audit_context; + + if (likely(!context)) + return 0; + + ax = kmalloc(sizeof(*ax) + nargs * sizeof(unsigned long), GFP_KERNEL); + if (!ax) + return -ENOMEM; + + ax->nargs = nargs; + memcpy(ax->args, args, nargs * sizeof(unsigned long)); + + ax->d.type = AUDIT_SOCKETCALL; + ax->d.next = context->aux; + context->aux = (void *)ax; + return 0; +} + +int audit_sockaddr(int len, void *a) +{ + struct audit_aux_data_sockaddr *ax; + struct audit_context *context = current->audit_context; + + if (likely(!context)) + return 0; + + ax = kmalloc(sizeof(*ax) + len, GFP_KERNEL); + if (!ax) + return -ENOMEM; + + ax->len = len; + memcpy(ax->a, a, len); + + ax->d.type = AUDIT_SOCKADDR; + ax->d.next = context->aux; + context->aux = (void *)ax; + return 0; +} + void audit_signal_info(int sig, struct task_struct *t) { extern pid_t audit_sig_pid; diff --git a/net/socket.c b/net/socket.c index cec0cb38b9ce..6b7c3b51a7c1 100644 --- a/net/socket.c +++ b/net/socket.c @@ -81,6 +81,7 @@ #include #include #include +#include #ifdef CONFIG_NET_RADIO #include /* Note : will define WIRELESS_EXT */ @@ -226,7 +227,7 @@ int move_addr_to_kernel(void __user *uaddr, int ulen, void *kaddr) return 0; if(copy_from_user(kaddr,uaddr,ulen)) return -EFAULT; - return 0; + return audit_sockaddr(ulen, kaddr); } /** @@ -1906,7 +1907,11 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) /* copy_from_user should be SMP safe. */ if (copy_from_user(a, args, nargs[call])) return -EFAULT; - + + err = audit_socketcall(nargs[call]/sizeof(unsigned long), args); + if (err) + return err; + a0=a[0]; a1=a[1]; -- cgit v1.2.3 From 209aba03243ee42a22f8df8d08aa9963f62aec64 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 18 May 2005 10:21:07 +0100 Subject: AUDIT: Treat all user messages identically. It's silly to have to add explicit entries for new userspace messages as we invent them. Just treat all messages in the user range the same. Signed-off-by: David Woodhouse --- include/linux/audit.h | 17 ++--------------- kernel/audit.c | 20 ++------------------ security/selinux/nlmsgtab.c | 17 +++++++---------- 3 files changed, 11 insertions(+), 43 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index 2f5dc60f8bbd..17ea5d522d81 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -51,14 +51,8 @@ #define AUDIT_WATCH_LIST 1009 /* List all file/dir watches */ #define AUDIT_SIGNAL_INFO 1010 /* Get info about sender of signal to auditd */ -#define AUDIT_USER_AUTH 1100 /* User space authentication */ -#define AUDIT_USER_ACCT 1101 /* User space acct change */ -#define AUDIT_USER_MGMT 1102 /* User space acct management */ -#define AUDIT_CRED_ACQ 1103 /* User space credential acquired */ -#define AUDIT_CRED_DISP 1104 /* User space credential disposed */ -#define AUDIT_USER_START 1105 /* User space session start */ -#define AUDIT_USER_END 1106 /* User space session end */ -#define AUDIT_USER_AVC 1107 /* User space avc message */ +#define AUDIT_FIRST_USER_MSG 1100 /* Userspace messages uninteresting to kernel */ +#define AUDIT_LAST_USER_MSG 1199 #define AUDIT_DAEMON_START 1200 /* Daemon startup record */ #define AUDIT_DAEMON_END 1201 /* Daemon normal stop record */ @@ -173,13 +167,6 @@ #define AUDIT_ARCH_V850 (EM_V850|__AUDIT_ARCH_LE) #define AUDIT_ARCH_X86_64 (EM_X86_64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) -#ifndef __KERNEL__ -struct audit_message { - struct nlmsghdr nlh; - char data[1200]; -}; -#endif - struct audit_status { __u32 mask; /* Bit mask for valid entries */ __u32 enabled; /* 1 = enabled, 0 = disabled */ diff --git a/kernel/audit.c b/kernel/audit.c index a0e33b6897d7..e6d88635032c 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -325,15 +325,7 @@ static int audit_netlink_ok(kernel_cap_t eff_cap, u16 msg_type) if (!cap_raised(eff_cap, CAP_AUDIT_CONTROL)) err = -EPERM; break; - case AUDIT_USER: - case AUDIT_USER_AUTH: - case AUDIT_USER_ACCT: - case AUDIT_USER_MGMT: - case AUDIT_CRED_ACQ: - case AUDIT_CRED_DISP: - case AUDIT_USER_START: - case AUDIT_USER_END: - case AUDIT_USER_AVC: + case AUDIT_FIRST_USER_MSG...AUDIT_LAST_USER_MSG: if (!cap_raised(eff_cap, CAP_AUDIT_WRITE)) err = -EPERM; break; @@ -402,15 +394,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) audit_set_backlog_limit(status_get->backlog_limit, loginuid); break; - case AUDIT_USER: - case AUDIT_USER_AUTH: - case AUDIT_USER_ACCT: - case AUDIT_USER_MGMT: - case AUDIT_CRED_ACQ: - case AUDIT_CRED_DISP: - case AUDIT_USER_START: - case AUDIT_USER_END: - case AUDIT_USER_AVC: + case AUDIT_FIRST_USER_MSG...AUDIT_LAST_USER_MSG: ab = audit_log_start(NULL, msg_type); if (!ab) break; /* audit_panic has been called */ diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index 67e77acc4795..f0fb6d76f7c5 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -98,14 +98,6 @@ static struct nlmsg_perm nlmsg_audit_perms[] = { AUDIT_DEL, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, { AUDIT_USER, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, { AUDIT_SIGNAL_INFO, NETLINK_AUDIT_SOCKET__NLMSG_READ }, - { AUDIT_USER_AUTH, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_ACCT, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_MGMT, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_CRED_ACQ, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_CRED_DISP, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_START, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_END, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_AVC, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, }; @@ -150,8 +142,13 @@ int selinux_nlmsg_lookup(u16 sclass, u16 nlmsg_type, u32 *perm) break; case SECCLASS_NETLINK_AUDIT_SOCKET: - err = nlmsg_perm(nlmsg_type, perm, nlmsg_audit_perms, - sizeof(nlmsg_audit_perms)); + if (nlmsg_type >= AUDIT_FIRST_USER_MSG && + nlmsg_type <= AUDIT_LAST_USER_MSG) { + *perm = NETLINK_AUDIT_SOCKET__NLMSG_RELAY; + } else { + err = nlmsg_perm(nlmsg_type, perm, nlmsg_audit_perms, + sizeof(nlmsg_audit_perms)); + } break; /* No messaging from userspace, or class unknown/unhandled */ -- cgit v1.2.3 From 867d1191fca388a79e4bb500dd85a9e871c96b99 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sun, 24 Apr 2005 02:06:05 -0500 Subject: [SCSI] remove requeue feature from blk_insert_request() blk_insert_request() has a unobivous feature of requeuing a request setting REQ_SPECIAL|REQ_SOFTBARRIER. SCSI midlayer was the only user and as previous patches removed the usage, remove the feature from blk_insert_request(). Only special requests should be queued with blk_insert_request(). All requeueing should go through blk_requeue_request(). Signed-off-by: Tejun Heo Signed-off-by: James Bottomley --- drivers/block/ll_rw_blk.c | 20 ++++++-------------- drivers/block/paride/pd.c | 2 +- drivers/block/sx8.c | 4 ++-- drivers/scsi/scsi_lib.c | 2 +- include/linux/blkdev.h | 2 +- 5 files changed, 11 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c index 11ef9d9ea139..f20eba22b14b 100644 --- a/drivers/block/ll_rw_blk.c +++ b/drivers/block/ll_rw_blk.c @@ -2038,7 +2038,6 @@ EXPORT_SYMBOL(blk_requeue_request); * @rq: request to be inserted * @at_head: insert request at head or tail of queue * @data: private data - * @reinsert: true if request it a reinsertion of previously processed one * * Description: * Many block devices need to execute commands asynchronously, so they don't @@ -2053,8 +2052,9 @@ EXPORT_SYMBOL(blk_requeue_request); * host that is unable to accept a particular command. */ void blk_insert_request(request_queue_t *q, struct request *rq, - int at_head, void *data, int reinsert) + int at_head, void *data) { + int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK; unsigned long flags; /* @@ -2071,20 +2071,12 @@ void blk_insert_request(request_queue_t *q, struct request *rq, /* * If command is tagged, release the tag */ - if (reinsert) - blk_requeue_request(q, rq); - else { - int where = ELEVATOR_INSERT_BACK; - - if (at_head) - where = ELEVATOR_INSERT_FRONT; + if (blk_rq_tagged(rq)) + blk_queue_end_tag(q, rq); - if (blk_rq_tagged(rq)) - blk_queue_end_tag(q, rq); + drive_stat_acct(rq, rq->nr_sectors, 1); + __elv_add_request(q, rq, where, 0); - drive_stat_acct(rq, rq->nr_sectors, 1); - __elv_add_request(q, rq, where, 0); - } if (blk_queue_plugged(q)) __generic_unplug_device(q); else diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c index 202a5a74ad37..fa49d62626ba 100644 --- a/drivers/block/paride/pd.c +++ b/drivers/block/paride/pd.c @@ -723,7 +723,7 @@ static int pd_special_command(struct pd_unit *disk, rq.ref_count = 1; rq.waiting = &wait; rq.end_io = blk_end_sync_rq; - blk_insert_request(disk->gd->queue, &rq, 0, func, 0); + blk_insert_request(disk->gd->queue, &rq, 0, func); wait_for_completion(&wait); rq.waiting = NULL; if (rq.errors) diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index 797f5988c2b5..5ed3a6379452 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -614,7 +614,7 @@ static int carm_array_info (struct carm_host *host, unsigned int array_idx) spin_unlock_irq(&host->lock); DPRINTK("blk_insert_request, tag == %u\n", idx); - blk_insert_request(host->oob_q, crq->rq, 1, crq, 0); + blk_insert_request(host->oob_q, crq->rq, 1, crq); return 0; @@ -653,7 +653,7 @@ static int carm_send_special (struct carm_host *host, carm_sspc_t func) crq->msg_bucket = (u32) rc; DPRINTK("blk_insert_request, tag == %u\n", idx); - blk_insert_request(host->oob_q, crq->rq, 1, crq, 0); + blk_insert_request(host->oob_q, crq->rq, 1, crq); return 0; } diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 861d5f5c9722..47a4ad40bf4e 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -92,7 +92,7 @@ int scsi_insert_special_req(struct scsi_request *sreq, int at_head) */ sreq->sr_request->flags &= ~REQ_DONTPREP; blk_insert_request(sreq->sr_device->request_queue, sreq->sr_request, - at_head, sreq, 0); + at_head, sreq); return 0; } diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index ef1afc178c0a..4a99b76c5a33 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -544,7 +544,7 @@ extern void blk_end_sync_rq(struct request *rq); extern void blk_attempt_remerge(request_queue_t *, struct request *); extern void __blk_attempt_remerge(request_queue_t *, struct request *); extern struct request *blk_get_request(request_queue_t *, int, int); -extern void blk_insert_request(request_queue_t *, struct request *, int, void *, int); +extern void blk_insert_request(request_queue_t *, struct request *, int, void *); extern void blk_requeue_request(request_queue_t *, struct request *); extern void blk_plug_device(request_queue_t *); extern int blk_remove_plug(request_queue_t *); -- cgit v1.2.3 From daa6eda65a53e5addf86c6bc829129ff51b08bda Mon Sep 17 00:00:00 2001 From: Gerd Knorr Date: Tue, 10 May 2005 10:59:13 +0200 Subject: [SCSI] add scsi changer driver This patch adds a device driver for scsi media changer devices. Signed-off-by: Gerd Knorr Signed-off-by: James Bottomley --- Documentation/scsi/scsi-changer.txt | 180 ++++++ drivers/scsi/Kconfig | 18 + drivers/scsi/Makefile | 1 + drivers/scsi/ch.c | 1025 +++++++++++++++++++++++++++++++++++ include/linux/chio.h | 168 ++++++ include/linux/major.h | 1 + include/scsi/scsi.h | 3 + 7 files changed, 1396 insertions(+) create mode 100644 Documentation/scsi/scsi-changer.txt create mode 100644 drivers/scsi/ch.c create mode 100644 include/linux/chio.h (limited to 'include/linux') diff --git a/Documentation/scsi/scsi-changer.txt b/Documentation/scsi/scsi-changer.txt new file mode 100644 index 000000000000..c132687b017a --- /dev/null +++ b/Documentation/scsi/scsi-changer.txt @@ -0,0 +1,180 @@ + +README for the SCSI media changer driver +======================================== + +This is a driver for SCSI Medium Changer devices, which are listed +with "Type: Medium Changer" in /proc/scsi/scsi. + +This is for *real* Jukeboxes. It is *not* supported to work with +common small CD-ROM changers, neither one-lun-per-slot SCSI changers +nor IDE drives. + +Userland tools available from here: + http://linux.bytesex.org/misc/changer.html + + +General Information +------------------- + +First some words about how changers work: A changer has 2 (possibly +more) SCSI ID's. One for the changer device which controls the robot, +and one for the device which actually reads and writes the data. The +later may be anything, a MOD, a CD-ROM, a tape or whatever. For the +changer device this is a "don't care", he *only* shuffles around the +media, nothing else. + + +The SCSI changer model is complex, compared to - for example - IDE-CD +changers. But it allows to handle nearly all possible cases. It knows +4 different types of changer elements: + + media transport - this one shuffles around the media, i.e. the + transport arm. Also known as "picker". + storage - a slot which can hold a media. + import/export - the same as above, but is accessable from outside, + i.e. there the operator (you !) can use this to + fill in and remove media from the changer. + Sometimes named "mailslot". + data transfer - this is the device which reads/writes, i.e. the + CD-ROM / Tape / whatever drive. + +None of these is limited to one: A huge Jukebox could have slots for +123 CD-ROM's, 5 CD-ROM readers (and therefore 6 SCSI ID's: the changer +and each CD-ROM) and 2 transport arms. No problem to handle. + + +How it is implemented +--------------------- + +I implemented the driver as character device driver with a NetBSD-like +ioctl interface. Just grabbed NetBSD's header file and one of the +other linux SCSI device drivers as starting point. The interface +should be source code compatible with NetBSD. So if there is any +software (anybody knows ???) which supports a BSDish changer driver, +it should work with this driver too. + +Over time a few more ioctls where added, volume tag support for example +wasn't covered by the NetBSD ioctl API. + + +Current State +------------- + +Support for more than one transport arm is not implemented yet (and +nobody asked for it so far...). + +I test and use the driver myself with a 35 slot cdrom jukebox from +Grundig. I got some reports telling it works ok with tape autoloaders +(Exabyte, HP and DEC). Some People use this driver with amanda. It +works fine with small (11 slots) and a huge (4 MOs, 88 slots) +magneto-optical Jukebox. Probably with lots of other changers too, most +(but not all :-) people mail me only if it does *not* work... + +I don't have any device lists, neither black-list nor white-list. Thus +it is quite useless to ask me whenever a specific device is supported or +not. In theory every changer device which supports the SCSI-2 media +changer command set should work out-of-the-box with this driver. If it +doesn't, it is a bug. Either within the driver or within the firmware +of the changer device. + + +Using it +-------- + +This is a character device with major number is 86, so use +"mknod /dev/sch0 c 86 0" to create the special file for the driver. + +If the module finds the changer, it prints some messages about the +device [ try "dmesg" if you don't see anything ] and should show up in +/proc/devices. If not.... some changers use ID ? / LUN 0 for the +device and ID ? / LUN 1 for the robot mechanism. But Linux does *not* +look for LUN's other than 0 as default, becauce there are to many +broken devices. So you can try: + + 1) echo "scsi add-single-device 0 0 ID 1" > /proc/scsi/scsi + (replace ID with the SCSI-ID of the device) + 2) boot the kernel with "max_scsi_luns=1" on the command line + (append="max_scsi_luns=1" in lilo.conf should do the trick) + + +Trouble? +-------- + +If you insmod the driver with "insmod debug=1", it will be verbose and +prints a lot of stuff to the syslog. Compiling the kernel with +CONFIG_SCSI_CONSTANTS=y improves the quality of the error messages alot +because the kernel will translate the error codes into human-readable +strings then. + +You can display these messages with the dmesg command (or check the +logfiles). If you email me some question becauce of a problem with the +driver, please include these messages. + + +Insmod options +-------------- + +debug=0/1 + Enable debug messages (see above, default: 0). + +verbose=0/1 + Be verbose (default: 1). + +init=0/1 + Send INITIALIZE ELEMENT STATUS command to the changer + at insmod time (default: 1). + +timeout_init= + timeout for the INITIALIZE ELEMENT STATUS command + (default: 3600). + +timeout_move= + timeout for all other commands (default: 120). + +dt_id=,,... +dt_lun=,,... + These two allow to specify the SCSI ID and LUN for the data + transfer elements. You likely don't need this as the jukebox + should provide this information. But some devices don't ... + +vendor_firsts= +vendor_counts= +vendor_labels= + These insmod options can be used to tell the driver that there + are some vendor-specific element types. Grundig for example + does this. Some jukeboxes have a printer to label fresh burned + CDs, which is addressed as element 0xc000 (type 5). To tell the + driver about this vendor-specific element, use this: + $ insmod ch \ + vendor_firsts=0xc000 \ + vendor_counts=1 \ + vendor_labels=printer + All three insmod options accept up to four comma-separated + values, this way you can configure the element types 5-8. + You likely need the SCSI specs for the device in question to + find the correct values as they are not covered by the SCSI-2 + standard. + + +Credits +------- + +I wrote this driver using the famous mailing-patches-around-the-world +method. With (more or less) help from: + + Daniel Moehwald + Dane Jasper + R. Scott Bailey + Jonathan Corbet + +Special thanks go to + Martin Kuehne +for a old, second-hand (but full functional) cdrom jukebox which I use +to develop/test driver and tools now. + +Have fun, + + Gerd + +-- +Gerd Knorr diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index 2ef5aee86b29..ba88be399a59 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -137,6 +137,24 @@ config CHR_DEV_SG If unsure, say N. +config CHR_DEV_SCH + tristate "SCSI media changer support" + depends on SCSI + ---help--- + This is a driver for SCSI media changers. Most common devices are + tape libraries and MOD/CDROM jukeboxes. *Real* jukeboxes, you + don't need this for those tiny 6-slot cdrom changers. Media + changers are listed as "Type: Medium Changer" in /proc/scsi/scsi. + If you have such hardware and want to use it with linux, say Y + here. Check for details. + + If you want to compile this as a module ( = code which can be + inserted in and removed from the running kernel whenever you want), + say M here and read and + . The module will be called ch.o. + If unsure, say N. + + comment "Some SCSI devices (e.g. CD jukebox) support multiple LUNs" depends on SCSI diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile index 51d9c1e1884b..3746fb9fa2f5 100644 --- a/drivers/scsi/Makefile +++ b/drivers/scsi/Makefile @@ -140,6 +140,7 @@ obj-$(CONFIG_CHR_DEV_OSST) += osst.o obj-$(CONFIG_BLK_DEV_SD) += sd_mod.o obj-$(CONFIG_BLK_DEV_SR) += sr_mod.o obj-$(CONFIG_CHR_DEV_SG) += sg.o +obj-$(CONFIG_CHR_DEV_SCH) += ch.o scsi_mod-y += scsi.o hosts.o scsi_ioctl.o constants.o \ scsicam.o scsi_error.o scsi_lib.o \ diff --git a/drivers/scsi/ch.c b/drivers/scsi/ch.c new file mode 100644 index 000000000000..44f5a71ec34a --- /dev/null +++ b/drivers/scsi/ch.c @@ -0,0 +1,1025 @@ +/* + * SCSI Media Changer device driver for Linux 2.6 + * + * (c) 1996-2003 Gerd Knorr + * + */ + +#define VERSION "0.25" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include /* here are all the ioctls */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#define CH_DT_MAX 16 +#define CH_TYPES 8 + +MODULE_DESCRIPTION("device driver for scsi media changer devices"); +MODULE_AUTHOR("Gerd Knorr "); +MODULE_LICENSE("GPL"); + +static int init = 1; +module_param(init, int, 0444); +MODULE_PARM_DESC(init, \ + "initialize element status on driver load (default: on)"); + +static int timeout_move = 300; +module_param(timeout_move, int, 0644); +MODULE_PARM_DESC(timeout_move,"timeout for move commands " + "(default: 300 seconds)"); + +static int timeout_init = 3600; +module_param(timeout_init, int, 0644); +MODULE_PARM_DESC(timeout_init,"timeout for INITIALIZE ELEMENT STATUS " + "(default: 3600 seconds)"); + +static int verbose = 1; +module_param(verbose, int, 0644); +MODULE_PARM_DESC(verbose,"be verbose (default: on)"); + +static int debug = 0; +module_param(debug, int, 0644); +MODULE_PARM_DESC(debug,"enable/disable debug messages, also prints more " + "detailed sense codes on scsi errors (default: off)"); + +static int dt_id[CH_DT_MAX] = { [ 0 ... (CH_DT_MAX-1) ] = -1 }; +static int dt_lun[CH_DT_MAX]; +module_param_array(dt_id, int, NULL, 0444); +module_param_array(dt_lun, int, NULL, 0444); + +/* tell the driver about vendor-specific slots */ +static int vendor_firsts[CH_TYPES-4]; +static int vendor_counts[CH_TYPES-4]; +module_param_array(vendor_firsts, int, NULL, 0444); +module_param_array(vendor_counts, int, NULL, 0444); + +static char *vendor_labels[CH_TYPES-4] = { + "v0", "v1", "v2", "v3" +}; +// module_param_string_array(vendor_labels, NULL, 0444); + +#define dprintk(fmt, arg...) if (debug) \ + printk(KERN_DEBUG "%s: " fmt, ch->name , ## arg) +#define vprintk(fmt, arg...) if (verbose) \ + printk(KERN_INFO "%s: " fmt, ch->name , ## arg) + +/* ------------------------------------------------------------------- */ + +#define MAX_RETRIES 1 + +static int ch_probe(struct device *); +static int ch_remove(struct device *); +static int ch_open(struct inode * inode, struct file * filp); +static int ch_release(struct inode * inode, struct file * filp); +static int ch_ioctl(struct inode * inode, struct file * filp, + unsigned int cmd, unsigned long arg); +#ifdef CONFIG_COMPAT +static long ch_ioctl_compat(struct file * filp, + unsigned int cmd, unsigned long arg); +#endif + +static struct class_simple * ch_sysfs_class; + +typedef struct { + struct list_head list; + int minor; + char name[8]; + struct scsi_device *device; + struct scsi_device **dt; /* ptrs to data transfer elements */ + u_int firsts[CH_TYPES]; + u_int counts[CH_TYPES]; + u_int unit_attention; + u_int voltags; + struct semaphore lock; +} scsi_changer; + +static LIST_HEAD(ch_devlist); +static spinlock_t ch_devlist_lock = SPIN_LOCK_UNLOCKED; +static int ch_devcount; + +static struct scsi_driver ch_template = +{ + .owner = THIS_MODULE, + .gendrv = { + .name = "ch", + .probe = ch_probe, + .remove = ch_remove, + }, +}; + +static struct file_operations changer_fops = +{ + .owner = THIS_MODULE, + .open = ch_open, + .release = ch_release, + .ioctl = ch_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = ch_ioctl_compat, +#endif +}; + +static struct { + unsigned char sense; + unsigned char asc; + unsigned char ascq; + int errno; +} err[] = { +/* Just filled in what looks right. Hav'nt checked any standard paper for + these errno assignments, so they may be wrong... */ + { + .sense = ILLEGAL_REQUEST, + .asc = 0x21, + .ascq = 0x01, + .errno = EBADSLT, /* Invalid element address */ + },{ + .sense = ILLEGAL_REQUEST, + .asc = 0x28, + .ascq = 0x01, + .errno = EBADE, /* Import or export element accessed */ + },{ + .sense = ILLEGAL_REQUEST, + .asc = 0x3B, + .ascq = 0x0D, + .errno = EXFULL, /* Medium destination element full */ + },{ + .sense = ILLEGAL_REQUEST, + .asc = 0x3B, + .ascq = 0x0E, + .errno = EBADE, /* Medium source element empty */ + },{ + .sense = ILLEGAL_REQUEST, + .asc = 0x20, + .ascq = 0x00, + .errno = EBADRQC, /* Invalid command operation code */ + },{ + /* end of list */ + } +}; + +/* ------------------------------------------------------------------- */ + +static int ch_find_errno(unsigned char *sense_buffer) +{ + int i,errno = 0; + + /* Check to see if additional sense information is available */ + if (sense_buffer[7] > 5 && + sense_buffer[12] != 0) { + for (i = 0; err[i].errno != 0; i++) { + if (err[i].sense == sense_buffer[ 2] && + err[i].asc == sense_buffer[12] && + err[i].ascq == sense_buffer[13]) { + errno = -err[i].errno; + break; + } + } + } + if (errno == 0) + errno = -EIO; + return errno; +} + +static int +ch_do_scsi(scsi_changer *ch, unsigned char *cmd, + void *buffer, unsigned buflength, + enum dma_data_direction direction) +{ + int errno, retries = 0, timeout; + struct scsi_request *sr; + + sr = scsi_allocate_request(ch->device, GFP_KERNEL); + if (NULL == sr) + return -ENOMEM; + + timeout = (cmd[0] == INITIALIZE_ELEMENT_STATUS) + ? timeout_init : timeout_move; + + retry: + errno = 0; + if (debug) { + dprintk("command: "); + __scsi_print_command(cmd); + } + + scsi_wait_req(sr, cmd, buffer, buflength, + timeout * HZ, MAX_RETRIES); + + dprintk("result: 0x%x\n",sr->sr_result); + if (driver_byte(sr->sr_result) & DRIVER_SENSE) { + if (debug) + scsi_print_req_sense(ch->name, sr); + errno = ch_find_errno(sr->sr_sense_buffer); + + switch(sr->sr_sense_buffer[2] & 0xf) { + case UNIT_ATTENTION: + ch->unit_attention = 1; + if (retries++ < 3) + goto retry; + break; + } + } + scsi_release_request(sr); + return errno; +} + +/* ------------------------------------------------------------------------ */ + +static int +ch_elem_to_typecode(scsi_changer *ch, u_int elem) +{ + int i; + + for (i = 0; i < CH_TYPES; i++) { + if (elem >= ch->firsts[i] && + elem < ch->firsts[i] + + ch->counts[i]) + return i+1; + } + return 0; +} + +static int +ch_read_element_status(scsi_changer *ch, u_int elem, char *data) +{ + u_char cmd[12]; + u_char *buffer; + int result; + + buffer = kmalloc(512, GFP_KERNEL | GFP_DMA); + if(!buffer) + return -ENOMEM; + + retry: + memset(cmd,0,sizeof(cmd)); + cmd[0] = READ_ELEMENT_STATUS; + cmd[1] = (ch->device->lun << 5) | + (ch->voltags ? 0x10 : 0) | + ch_elem_to_typecode(ch,elem); + cmd[2] = (elem >> 8) & 0xff; + cmd[3] = elem & 0xff; + cmd[5] = 1; + cmd[9] = 255; + if (0 == (result = ch_do_scsi(ch, cmd, buffer, 256, DMA_FROM_DEVICE))) { + if (((buffer[16] << 8) | buffer[17]) != elem) { + dprintk("asked for element 0x%02x, got 0x%02x\n", + elem,(buffer[16] << 8) | buffer[17]); + kfree(buffer); + return -EIO; + } + memcpy(data,buffer+16,16); + } else { + if (ch->voltags) { + ch->voltags = 0; + vprintk("device has no volume tag support\n"); + goto retry; + } + dprintk("READ ELEMENT STATUS for element 0x%x failed\n",elem); + } + kfree(buffer); + return result; +} + +static int +ch_init_elem(scsi_changer *ch) +{ + int err; + u_char cmd[6]; + + vprintk("INITIALIZE ELEMENT STATUS, may take some time ...\n"); + memset(cmd,0,sizeof(cmd)); + cmd[0] = INITIALIZE_ELEMENT_STATUS; + cmd[1] = ch->device->lun << 5; + err = ch_do_scsi(ch, cmd, NULL, 0, DMA_NONE); + vprintk("... finished\n"); + return err; +} + +static int +ch_readconfig(scsi_changer *ch) +{ + u_char cmd[10], data[16]; + u_char *buffer; + int result,id,lun,i; + u_int elem; + + buffer = kmalloc(512, GFP_KERNEL | GFP_DMA); + if (!buffer) + return -ENOMEM; + memset(buffer,0,512); + + memset(cmd,0,sizeof(cmd)); + cmd[0] = MODE_SENSE; + cmd[1] = ch->device->lun << 5; + cmd[2] = 0x1d; + cmd[4] = 255; + result = ch_do_scsi(ch, cmd, buffer, 255, DMA_FROM_DEVICE); + if (0 != result) { + cmd[1] |= (1<<3); + result = ch_do_scsi(ch, cmd, buffer, 255, DMA_FROM_DEVICE); + } + if (0 == result) { + ch->firsts[CHET_MT] = + (buffer[buffer[3]+ 6] << 8) | buffer[buffer[3]+ 7]; + ch->counts[CHET_MT] = + (buffer[buffer[3]+ 8] << 8) | buffer[buffer[3]+ 9]; + ch->firsts[CHET_ST] = + (buffer[buffer[3]+10] << 8) | buffer[buffer[3]+11]; + ch->counts[CHET_ST] = + (buffer[buffer[3]+12] << 8) | buffer[buffer[3]+13]; + ch->firsts[CHET_IE] = + (buffer[buffer[3]+14] << 8) | buffer[buffer[3]+15]; + ch->counts[CHET_IE] = + (buffer[buffer[3]+16] << 8) | buffer[buffer[3]+17]; + ch->firsts[CHET_DT] = + (buffer[buffer[3]+18] << 8) | buffer[buffer[3]+19]; + ch->counts[CHET_DT] = + (buffer[buffer[3]+20] << 8) | buffer[buffer[3]+21]; + vprintk("type #1 (mt): 0x%x+%d [medium transport]\n", + ch->firsts[CHET_MT], + ch->counts[CHET_MT]); + vprintk("type #2 (st): 0x%x+%d [storage]\n", + ch->firsts[CHET_ST], + ch->counts[CHET_ST]); + vprintk("type #3 (ie): 0x%x+%d [import/export]\n", + ch->firsts[CHET_IE], + ch->counts[CHET_IE]); + vprintk("type #4 (dt): 0x%x+%d [data transfer]\n", + ch->firsts[CHET_DT], + ch->counts[CHET_DT]); + } else { + vprintk("reading element address assigment page failed!\n"); + } + + /* vendor specific element types */ + for (i = 0; i < 4; i++) { + if (0 == vendor_counts[i]) + continue; + if (NULL == vendor_labels[i]) + continue; + ch->firsts[CHET_V1+i] = vendor_firsts[i]; + ch->counts[CHET_V1+i] = vendor_counts[i]; + vprintk("type #%d (v%d): 0x%x+%d [%s, vendor specific]\n", + i+5,i+1,vendor_firsts[i],vendor_counts[i], + vendor_labels[i]); + } + + /* look up the devices of the data transfer elements */ + ch->dt = kmalloc(ch->counts[CHET_DT]*sizeof(struct scsi_device), + GFP_KERNEL); + for (elem = 0; elem < ch->counts[CHET_DT]; elem++) { + id = -1; + lun = 0; + if (elem < CH_DT_MAX && -1 != dt_id[elem]) { + id = dt_id[elem]; + lun = dt_lun[elem]; + vprintk("dt 0x%x: [insmod option] ", + elem+ch->firsts[CHET_DT]); + } else if (0 != ch_read_element_status + (ch,elem+ch->firsts[CHET_DT],data)) { + vprintk("dt 0x%x: READ ELEMENT STATUS failed\n", + elem+ch->firsts[CHET_DT]); + } else { + vprintk("dt 0x%x: ",elem+ch->firsts[CHET_DT]); + if (data[6] & 0x80) { + if (verbose) + printk("not this SCSI bus\n"); + ch->dt[elem] = NULL; + } else if (0 == (data[6] & 0x30)) { + if (verbose) + printk("ID/LUN unknown\n"); + ch->dt[elem] = NULL; + } else { + id = ch->device->id; + lun = 0; + if (data[6] & 0x20) id = data[7]; + if (data[6] & 0x10) lun = data[6] & 7; + } + } + if (-1 != id) { + if (verbose) + printk("ID %i, LUN %i, ",id,lun); + ch->dt[elem] = + scsi_device_lookup(ch->device->host, + ch->device->channel, + id,lun); + if (!ch->dt[elem]) { + /* should not happen */ + if (verbose) + printk("Huh? device not found!\n"); + } else { + if (verbose) + printk("name: %8.8s %16.16s %4.4s\n", + ch->dt[elem]->vendor, + ch->dt[elem]->model, + ch->dt[elem]->rev); + } + } + } + ch->voltags = 1; + kfree(buffer); + + return 0; +} + +/* ------------------------------------------------------------------------ */ + +static int +ch_position(scsi_changer *ch, u_int trans, u_int elem, int rotate) +{ + u_char cmd[10]; + + dprintk("position: 0x%x\n",elem); + if (0 == trans) + trans = ch->firsts[CHET_MT]; + memset(cmd,0,sizeof(cmd)); + cmd[0] = POSITION_TO_ELEMENT; + cmd[1] = ch->device->lun << 5; + cmd[2] = (trans >> 8) & 0xff; + cmd[3] = trans & 0xff; + cmd[4] = (elem >> 8) & 0xff; + cmd[5] = elem & 0xff; + cmd[8] = rotate ? 1 : 0; + return ch_do_scsi(ch, cmd, NULL, 0, DMA_NONE); +} + +static int +ch_move(scsi_changer *ch, u_int trans, u_int src, u_int dest, int rotate) +{ + u_char cmd[12]; + + dprintk("move: 0x%x => 0x%x\n",src,dest); + if (0 == trans) + trans = ch->firsts[CHET_MT]; + memset(cmd,0,sizeof(cmd)); + cmd[0] = MOVE_MEDIUM; + cmd[1] = ch->device->lun << 5; + cmd[2] = (trans >> 8) & 0xff; + cmd[3] = trans & 0xff; + cmd[4] = (src >> 8) & 0xff; + cmd[5] = src & 0xff; + cmd[6] = (dest >> 8) & 0xff; + cmd[7] = dest & 0xff; + cmd[10] = rotate ? 1 : 0; + return ch_do_scsi(ch, cmd, NULL,0, DMA_NONE); +} + +static int +ch_exchange(scsi_changer *ch, u_int trans, u_int src, + u_int dest1, u_int dest2, int rotate1, int rotate2) +{ + u_char cmd[12]; + + dprintk("exchange: 0x%x => 0x%x => 0x%x\n", + src,dest1,dest2); + if (0 == trans) + trans = ch->firsts[CHET_MT]; + memset(cmd,0,sizeof(cmd)); + cmd[0] = EXCHANGE_MEDIUM; + cmd[1] = ch->device->lun << 5; + cmd[2] = (trans >> 8) & 0xff; + cmd[3] = trans & 0xff; + cmd[4] = (src >> 8) & 0xff; + cmd[5] = src & 0xff; + cmd[6] = (dest1 >> 8) & 0xff; + cmd[7] = dest1 & 0xff; + cmd[8] = (dest2 >> 8) & 0xff; + cmd[9] = dest2 & 0xff; + cmd[10] = (rotate1 ? 1 : 0) | (rotate2 ? 2 : 0); + + return ch_do_scsi(ch, cmd, NULL,0, DMA_NONE); +} + +static void +ch_check_voltag(char *tag) +{ + int i; + + for (i = 0; i < 32; i++) { + /* restrict to ascii */ + if (tag[i] >= 0x7f || tag[i] < 0x20) + tag[i] = ' '; + /* don't allow search wildcards */ + if (tag[i] == '?' || + tag[i] == '*') + tag[i] = ' '; + } +} + +static int +ch_set_voltag(scsi_changer *ch, u_int elem, + int alternate, int clear, u_char *tag) +{ + u_char cmd[12]; + u_char *buffer; + int result; + + buffer = kmalloc(512, GFP_KERNEL); + if (!buffer) + return -ENOMEM; + memset(buffer,0,512); + + dprintk("%s %s voltag: 0x%x => \"%s\"\n", + clear ? "clear" : "set", + alternate ? "alternate" : "primary", + elem, tag); + memset(cmd,0,sizeof(cmd)); + cmd[0] = SEND_VOLUME_TAG; + cmd[1] = (ch->device->lun << 5) | + ch_elem_to_typecode(ch,elem); + cmd[2] = (elem >> 8) & 0xff; + cmd[3] = elem & 0xff; + cmd[5] = clear + ? (alternate ? 0x0d : 0x0c) + : (alternate ? 0x0b : 0x0a); + + cmd[9] = 255; + + memcpy(buffer,tag,32); + ch_check_voltag(buffer); + + result = ch_do_scsi(ch, cmd, buffer, 256, DMA_TO_DEVICE); + kfree(buffer); + return result; +} + +static int ch_gstatus(scsi_changer *ch, int type, unsigned char *dest) +{ + int retval = 0; + u_char data[16]; + unsigned int i; + + down(&ch->lock); + for (i = 0; i < ch->counts[type]; i++) { + if (0 != ch_read_element_status + (ch, ch->firsts[type]+i,data)) { + retval = -EIO; + break; + } + put_user(data[2], dest+i); + if (data[2] & CESTATUS_EXCEPT) + vprintk("element 0x%x: asc=0x%x, ascq=0x%x\n", + ch->firsts[type]+i, + (int)data[4],(int)data[5]); + retval = ch_read_element_status + (ch, ch->firsts[type]+i,data); + if (0 != retval) + break; + } + up(&ch->lock); + return retval; +} + +/* ------------------------------------------------------------------------ */ + +static int +ch_release(struct inode *inode, struct file *file) +{ + scsi_changer *ch = file->private_data; + + scsi_device_put(ch->device); + file->private_data = NULL; + return 0; +} + +static int +ch_open(struct inode *inode, struct file *file) +{ + scsi_changer *tmp, *ch; + int minor = iminor(inode); + + spin_lock(&ch_devlist_lock); + ch = NULL; + list_for_each_entry(tmp,&ch_devlist,list) { + if (tmp->minor == minor) + ch = tmp; + } + if (NULL == ch || scsi_device_get(ch->device)) { + spin_unlock(&ch_devlist_lock); + return -ENXIO; + } + spin_unlock(&ch_devlist_lock); + + file->private_data = ch; + return 0; +} + +static int +ch_checkrange(scsi_changer *ch, unsigned int type, unsigned int unit) +{ + if (type >= CH_TYPES || unit >= ch->counts[type]) + return -1; + return 0; +} + +static int ch_ioctl(struct inode * inode, struct file * file, + unsigned int cmd, unsigned long arg) +{ + scsi_changer *ch = file->private_data; + int retval; + + switch (cmd) { + case CHIOGPARAMS: + { + struct changer_params params; + + params.cp_curpicker = 0; + params.cp_npickers = ch->counts[CHET_MT]; + params.cp_nslots = ch->counts[CHET_ST]; + params.cp_nportals = ch->counts[CHET_IE]; + params.cp_ndrives = ch->counts[CHET_DT]; + + if (copy_to_user((void *) arg, ¶ms, sizeof(params))) + return -EFAULT; + return 0; + } + case CHIOGVPARAMS: + { + struct changer_vendor_params vparams; + + memset(&vparams,0,sizeof(vparams)); + if (ch->counts[CHET_V1]) { + vparams.cvp_n1 = ch->counts[CHET_V1]; + strncpy(vparams.cvp_label1,vendor_labels[0],16); + } + if (ch->counts[CHET_V2]) { + vparams.cvp_n2 = ch->counts[CHET_V2]; + strncpy(vparams.cvp_label2,vendor_labels[1],16); + } + if (ch->counts[CHET_V3]) { + vparams.cvp_n3 = ch->counts[CHET_V3]; + strncpy(vparams.cvp_label3,vendor_labels[2],16); + } + if (ch->counts[CHET_V4]) { + vparams.cvp_n4 = ch->counts[CHET_V4]; + strncpy(vparams.cvp_label4,vendor_labels[3],16); + } + if (copy_to_user((void *) arg, &vparams, sizeof(vparams))) + return -EFAULT; + return 0; + } + + case CHIOPOSITION: + { + struct changer_position pos; + + if (copy_from_user(&pos, (void*)arg, sizeof (pos))) + return -EFAULT; + + if (0 != ch_checkrange(ch, pos.cp_type, pos.cp_unit)) { + dprintk("CHIOPOSITION: invalid parameter\n"); + return -EBADSLT; + } + down(&ch->lock); + retval = ch_position(ch,0, + ch->firsts[pos.cp_type] + pos.cp_unit, + pos.cp_flags & CP_INVERT); + up(&ch->lock); + return retval; + } + + case CHIOMOVE: + { + struct changer_move mv; + + if (copy_from_user(&mv, (void*)arg, sizeof (mv))) + return -EFAULT; + + if (0 != ch_checkrange(ch, mv.cm_fromtype, mv.cm_fromunit) || + 0 != ch_checkrange(ch, mv.cm_totype, mv.cm_tounit )) { + dprintk("CHIOMOVE: invalid parameter\n"); + return -EBADSLT; + } + + down(&ch->lock); + retval = ch_move(ch,0, + ch->firsts[mv.cm_fromtype] + mv.cm_fromunit, + ch->firsts[mv.cm_totype] + mv.cm_tounit, + mv.cm_flags & CM_INVERT); + up(&ch->lock); + return retval; + } + + case CHIOEXCHANGE: + { + struct changer_exchange mv; + + if (copy_from_user(&mv, (void*)arg, sizeof (mv))) + return -EFAULT; + + if (0 != ch_checkrange(ch, mv.ce_srctype, mv.ce_srcunit ) || + 0 != ch_checkrange(ch, mv.ce_fdsttype, mv.ce_fdstunit) || + 0 != ch_checkrange(ch, mv.ce_sdsttype, mv.ce_sdstunit)) { + dprintk("CHIOEXCHANGE: invalid parameter\n"); + return -EBADSLT; + } + + down(&ch->lock); + retval = ch_exchange + (ch,0, + ch->firsts[mv.ce_srctype] + mv.ce_srcunit, + ch->firsts[mv.ce_fdsttype] + mv.ce_fdstunit, + ch->firsts[mv.ce_sdsttype] + mv.ce_sdstunit, + mv.ce_flags & CE_INVERT1, mv.ce_flags & CE_INVERT2); + up(&ch->lock); + return retval; + } + + case CHIOGSTATUS: + { + struct changer_element_status ces; + + if (copy_from_user(&ces, (void*)arg, sizeof (ces))) + return -EFAULT; + if (ces.ces_type < 0 || ces.ces_type >= CH_TYPES) + return -EINVAL; + + return ch_gstatus(ch, ces.ces_type, ces.ces_data); + } + + case CHIOGELEM: + { + struct changer_get_element cge; + u_char cmd[12]; + u_char *buffer; + unsigned int elem; + int result,i; + + if (copy_from_user(&cge, (void*)arg, sizeof (cge))) + return -EFAULT; + + if (0 != ch_checkrange(ch, cge.cge_type, cge.cge_unit)) + return -EINVAL; + elem = ch->firsts[cge.cge_type] + cge.cge_unit; + + buffer = kmalloc(512, GFP_KERNEL | GFP_DMA); + if (!buffer) + return -ENOMEM; + down(&ch->lock); + + voltag_retry: + memset(cmd,0,sizeof(cmd)); + cmd[0] = READ_ELEMENT_STATUS; + cmd[1] = (ch->device->lun << 5) | + (ch->voltags ? 0x10 : 0) | + ch_elem_to_typecode(ch,elem); + cmd[2] = (elem >> 8) & 0xff; + cmd[3] = elem & 0xff; + cmd[5] = 1; + cmd[9] = 255; + + if (0 == (result = ch_do_scsi(ch, cmd, buffer, 256, DMA_FROM_DEVICE))) { + cge.cge_status = buffer[18]; + cge.cge_flags = 0; + if (buffer[18] & CESTATUS_EXCEPT) { + cge.cge_errno = EIO; + } + if (buffer[25] & 0x80) { + cge.cge_flags |= CGE_SRC; + if (buffer[25] & 0x40) + cge.cge_flags |= CGE_INVERT; + elem = (buffer[26]<<8) | buffer[27]; + for (i = 0; i < 4; i++) { + if (elem >= ch->firsts[i] && + elem < ch->firsts[i] + ch->counts[i]) { + cge.cge_srctype = i; + cge.cge_srcunit = elem-ch->firsts[i]; + } + } + } + if ((buffer[22] & 0x30) == 0x30) { + cge.cge_flags |= CGE_IDLUN; + cge.cge_id = buffer[23]; + cge.cge_lun = buffer[22] & 7; + } + if (buffer[9] & 0x80) { + cge.cge_flags |= CGE_PVOLTAG; + memcpy(cge.cge_pvoltag,buffer+28,36); + } + if (buffer[9] & 0x40) { + cge.cge_flags |= CGE_AVOLTAG; + memcpy(cge.cge_avoltag,buffer+64,36); + } + } else if (ch->voltags) { + ch->voltags = 0; + vprintk("device has no volume tag support\n"); + goto voltag_retry; + } + kfree(buffer); + up(&ch->lock); + + if (copy_to_user((void*)arg, &cge, sizeof (cge))) + return -EFAULT; + return result; + } + + case CHIOINITELEM: + { + down(&ch->lock); + retval = ch_init_elem(ch); + up(&ch->lock); + return retval; + } + + case CHIOSVOLTAG: + { + struct changer_set_voltag csv; + int elem; + + if (copy_from_user(&csv, (void*)arg, sizeof(csv))) + return -EFAULT; + + if (0 != ch_checkrange(ch, csv.csv_type, csv.csv_unit)) { + dprintk("CHIOSVOLTAG: invalid parameter\n"); + return -EBADSLT; + } + elem = ch->firsts[csv.csv_type] + csv.csv_unit; + down(&ch->lock); + retval = ch_set_voltag(ch, elem, + csv.csv_flags & CSV_AVOLTAG, + csv.csv_flags & CSV_CLEARTAG, + csv.csv_voltag); + up(&ch->lock); + return retval; + } + + default: + return scsi_ioctl(ch->device, cmd, (void*)arg); + + } +} + +#ifdef CONFIG_COMPAT + +struct changer_element_status32 { + int ces_type; + compat_uptr_t ces_data; +}; +#define CHIOGSTATUS32 _IOW('c', 8,struct changer_element_status32) + +static long ch_ioctl_compat(struct file * file, + unsigned int cmd, unsigned long arg) +{ + scsi_changer *ch = file->private_data; + + switch (cmd) { + case CHIOGPARAMS: + case CHIOGVPARAMS: + case CHIOPOSITION: + case CHIOMOVE: + case CHIOEXCHANGE: + case CHIOGELEM: + case CHIOINITELEM: + case CHIOSVOLTAG: + /* compatible */ + return ch_ioctl(NULL /* inode, unused */, + file, cmd, arg); + case CHIOGSTATUS32: + { + struct changer_element_status32 ces32; + unsigned char *data; + + if (copy_from_user(&ces32, (void*)arg, sizeof (ces32))) + return -EFAULT; + if (ces32.ces_type < 0 || ces32.ces_type >= CH_TYPES) + return -EINVAL; + + data = compat_ptr(ces32.ces_data); + return ch_gstatus(ch, ces32.ces_type, data); + } + default: + // return scsi_ioctl_compat(ch->device, cmd, (void*)arg); + return -ENOIOCTLCMD; + + } +} +#endif + +/* ------------------------------------------------------------------------ */ + +static int ch_probe(struct device *dev) +{ + struct scsi_device *sd = to_scsi_device(dev); + scsi_changer *ch; + + if (sd->type != TYPE_MEDIUM_CHANGER) + return -ENODEV; + + ch = kmalloc(sizeof(*ch), GFP_KERNEL); + if (NULL == ch) + return -ENOMEM; + + memset(ch,0,sizeof(*ch)); + ch->minor = ch_devcount; + sprintf(ch->name,"ch%d",ch->minor); + init_MUTEX(&ch->lock); + ch->device = sd; + ch_readconfig(ch); + if (init) + ch_init_elem(ch); + + devfs_mk_cdev(MKDEV(SCSI_CHANGER_MAJOR,ch->minor), + S_IFCHR | S_IRUGO | S_IWUGO, ch->name); + class_simple_device_add(ch_sysfs_class, + MKDEV(SCSI_CHANGER_MAJOR,ch->minor), + dev, "s%s", ch->name); + + printk(KERN_INFO "Attached scsi changer %s " + "at scsi%d, channel %d, id %d, lun %d\n", + ch->name, sd->host->host_no, sd->channel, sd->id, sd->lun); + + spin_lock(&ch_devlist_lock); + list_add_tail(&ch->list,&ch_devlist); + ch_devcount++; + spin_unlock(&ch_devlist_lock); + return 0; +} + +static int ch_remove(struct device *dev) +{ + struct scsi_device *sd = to_scsi_device(dev); + scsi_changer *tmp, *ch; + + spin_lock(&ch_devlist_lock); + ch = NULL; + list_for_each_entry(tmp,&ch_devlist,list) { + if (tmp->device == sd) + ch = tmp; + } + BUG_ON(NULL == ch); + list_del(&ch->list); + spin_unlock(&ch_devlist_lock); + + class_simple_device_remove(MKDEV(SCSI_CHANGER_MAJOR,ch->minor)); + devfs_remove(ch->name); + kfree(ch->dt); + kfree(ch); + ch_devcount--; + return 0; +} + +static int __init init_ch_module(void) +{ + int rc; + + printk(KERN_INFO "SCSI Media Changer driver v" VERSION " \n"); + ch_sysfs_class = class_simple_create(THIS_MODULE, "scsi_changer"); + if (IS_ERR(ch_sysfs_class)) { + rc = PTR_ERR(ch_sysfs_class); + return rc; + } + rc = register_chrdev(SCSI_CHANGER_MAJOR,"ch",&changer_fops); + if (rc < 0) { + printk("Unable to get major %d for SCSI-Changer\n", + SCSI_CHANGER_MAJOR); + goto fail1; + } + rc = scsi_register_driver(&ch_template.gendrv); + if (rc < 0) + goto fail2; + return 0; + + fail2: + unregister_chrdev(SCSI_CHANGER_MAJOR, "ch"); + fail1: + class_simple_destroy(ch_sysfs_class); + return rc; +} + +static void __exit exit_ch_module(void) +{ + scsi_unregister_driver(&ch_template.gendrv); + unregister_chrdev(SCSI_CHANGER_MAJOR, "ch"); + class_simple_destroy(ch_sysfs_class); +} + +module_init(init_ch_module); +module_exit(exit_ch_module); + +/* + * Local variables: + * c-basic-offset: 8 + * End: + */ diff --git a/include/linux/chio.h b/include/linux/chio.h new file mode 100644 index 000000000000..63035ae67e63 --- /dev/null +++ b/include/linux/chio.h @@ -0,0 +1,168 @@ +/* + * ioctl interface for the scsi media changer driver + */ + +/* changer element types */ +#define CHET_MT 0 /* media transport element (robot) */ +#define CHET_ST 1 /* storage element (media slots) */ +#define CHET_IE 2 /* import/export element */ +#define CHET_DT 3 /* data transfer element (tape/cdrom/whatever) */ +#define CHET_V1 4 /* vendor specific #1 */ +#define CHET_V2 5 /* vendor specific #2 */ +#define CHET_V3 6 /* vendor specific #3 */ +#define CHET_V4 7 /* vendor specific #4 */ + + +/* + * CHIOGPARAMS + * query changer properties + * + * CHIOVGPARAMS + * query vendor-specific element types + * + * accessing elements works by specifing type and unit of the element. + * for eample, storage elements are addressed with type = CHET_ST and + * unit = 0 .. cp_nslots-1 + * + */ +struct changer_params { + int cp_curpicker; /* current transport element */ + int cp_npickers; /* number of transport elements (CHET_MT) */ + int cp_nslots; /* number of storage elements (CHET_ST) */ + int cp_nportals; /* number of import/export elements (CHET_IE) */ + int cp_ndrives; /* number of data transfer elements (CHET_DT) */ +}; +struct changer_vendor_params { + int cvp_n1; /* number of vendor specific elems (CHET_V1) */ + char cvp_label1[16]; + int cvp_n2; /* number of vendor specific elems (CHET_V2) */ + char cvp_label2[16]; + int cvp_n3; /* number of vendor specific elems (CHET_V3) */ + char cvp_label3[16]; + int cvp_n4; /* number of vendor specific elems (CHET_V4) */ + char cvp_label4[16]; + int reserved[8]; +}; + + +/* + * CHIOMOVE + * move a medium from one element to another + */ +struct changer_move { + int cm_fromtype; /* type/unit of source element */ + int cm_fromunit; + int cm_totype; /* type/unit of destination element */ + int cm_tounit; + int cm_flags; +}; +#define CM_INVERT 1 /* flag: rotate media (for double-sided like MOD) */ + + +/* + * CHIOEXCHANGE + * move one medium from element #1 to element #2, + * and another one from element #2 to element #3. + * element #1 and #3 are allowed to be identical. + */ +struct changer_exchange { + int ce_srctype; /* type/unit of element #1 */ + int ce_srcunit; + int ce_fdsttype; /* type/unit of element #2 */ + int ce_fdstunit; + int ce_sdsttype; /* type/unit of element #3 */ + int ce_sdstunit; + int ce_flags; +}; +#define CE_INVERT1 1 +#define CE_INVERT2 2 + + +/* + * CHIOPOSITION + * move the transport element (robot arm) to a specific element. + */ +struct changer_position { + int cp_type; + int cp_unit; + int cp_flags; +}; +#define CP_INVERT 1 + + +/* + * CHIOGSTATUS + * get element status for all elements of a specific type + */ +struct changer_element_status { + int ces_type; + unsigned char *ces_data; +}; +#define CESTATUS_FULL 0x01 /* full */ +#define CESTATUS_IMPEXP 0x02 /* media was imported (inserted by sysop) */ +#define CESTATUS_EXCEPT 0x04 /* error condition */ +#define CESTATUS_ACCESS 0x08 /* access allowed */ +#define CESTATUS_EXENAB 0x10 /* element can export media */ +#define CESTATUS_INENAB 0x20 /* element can import media */ + + +/* + * CHIOGELEM + * get more detailed status informtion for a single element + */ +struct changer_get_element { + int cge_type; /* type/unit */ + int cge_unit; + int cge_status; /* status */ + int cge_errno; /* errno */ + int cge_srctype; /* source element of the last move/exchange */ + int cge_srcunit; + int cge_id; /* scsi id (for data transfer elements) */ + int cge_lun; /* scsi lun (for data transfer elements) */ + char cge_pvoltag[36]; /* primary volume tag */ + char cge_avoltag[36]; /* alternate volume tag */ + int cge_flags; +}; +/* flags */ +#define CGE_ERRNO 0x01 /* errno available */ +#define CGE_INVERT 0x02 /* media inverted */ +#define CGE_SRC 0x04 /* media src available */ +#define CGE_IDLUN 0x08 /* ID+LUN available */ +#define CGE_PVOLTAG 0x10 /* primary volume tag available */ +#define CGE_AVOLTAG 0x20 /* alternate volume tag available */ + + +/* + * CHIOSVOLTAG + * set volume tag + */ +struct changer_set_voltag { + int csv_type; /* type/unit */ + int csv_unit; + char csv_voltag[36]; /* volume tag */ + int csv_flags; +}; +#define CSV_PVOLTAG 0x01 /* primary volume tag */ +#define CSV_AVOLTAG 0x02 /* alternate volume tag */ +#define CSV_CLEARTAG 0x04 /* clear volume tag */ + +/* ioctls */ +#define CHIOMOVE _IOW('c', 1,struct changer_move) +#define CHIOEXCHANGE _IOW('c', 2,struct changer_exchange) +#define CHIOPOSITION _IOW('c', 3,struct changer_position) +#define CHIOGPICKER _IOR('c', 4,int) /* not impl. */ +#define CHIOSPICKER _IOW('c', 5,int) /* not impl. */ +#define CHIOGPARAMS _IOR('c', 6,struct changer_params) +#define CHIOGSTATUS _IOW('c', 8,struct changer_element_status) +#define CHIOGELEM _IOW('c',16,struct changer_get_element) +#define CHIOINITELEM _IO('c',17) +#define CHIOSVOLTAG _IOW('c',18,struct changer_set_voltag) +#define CHIOGVPARAMS _IOR('c',19,struct changer_vendor_params) + +/* ---------------------------------------------------------------------- */ + +/* + * Local variables: + * c-basic-offset: 8 + * End: + */ diff --git a/include/linux/major.h b/include/linux/major.h index 4b62c42b842c..e36a46702d94 100644 --- a/include/linux/major.h +++ b/include/linux/major.h @@ -100,6 +100,7 @@ #define I2O_MAJOR 80 /* 80->87 */ #define SHMIQ_MAJOR 85 /* Linux/mips, SGI /dev/shmiq */ +#define SCSI_CHANGER_MAJOR 86 #define IDE6_MAJOR 88 #define IDE7_MAJOR 89 diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h index 659ecf48fb4a..ca1e3b4a3183 100644 --- a/include/scsi/scsi.h +++ b/include/scsi/scsi.h @@ -41,6 +41,7 @@ extern const char *const scsi_device_types[MAX_SCSI_DEVICE_CODE]; #define FORMAT_UNIT 0x04 #define READ_BLOCK_LIMITS 0x05 #define REASSIGN_BLOCKS 0x07 +#define INITIALIZE_ELEMENT_STATUS 0x07 #define READ_6 0x08 #define WRITE_6 0x0a #define SEEK_6 0x0b @@ -65,6 +66,7 @@ extern const char *const scsi_device_types[MAX_SCSI_DEVICE_CODE]; #define READ_10 0x28 #define WRITE_10 0x2a #define SEEK_10 0x2b +#define POSITION_TO_ELEMENT 0x2b #define WRITE_VERIFY 0x2e #define VERIFY 0x2f #define SEARCH_HIGH 0x30 @@ -97,6 +99,7 @@ extern const char *const scsi_device_types[MAX_SCSI_DEVICE_CODE]; #define PERSISTENT_RESERVE_OUT 0x5f #define REPORT_LUNS 0xa0 #define MOVE_MEDIUM 0xa5 +#define EXCHANGE_MEDIUM 0xa6 #define READ_12 0xa8 #define WRITE_12 0xaa #define WRITE_VERIFY_12 0xae -- cgit v1.2.3 From 011161051bbc25f7f8b7df059dbd934c534443f0 Mon Sep 17 00:00:00 2001 From: Stephen Smalley Date: Sat, 21 May 2005 00:15:52 +0100 Subject: AUDIT: Avoid sleeping function in SElinux AVC audit. This patch changes the SELinux AVC to defer logging of paths to the audit framework upon syscall exit, by saving a reference to the (dentry,vfsmount) pair in an auxiliary audit item on the current audit context for processing by audit_log_exit. Signed-off-by: Stephen Smalley Signed-off-by: David Woodhouse --- include/linux/audit.h | 3 +++ kernel/auditsc.c | 40 ++++++++++++++++++++++++++++++++++++++++ security/selinux/avc.c | 17 ++++++++--------- 3 files changed, 51 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index 17ea5d522d81..4b7caf0c6e10 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -69,6 +69,7 @@ #define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ #define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ +#define AUDIT_AVC_PATH 1402 /* dentry, vfsmount pair from avc */ #define AUDIT_KERNEL 2000 /* Asynchronous audit record. NOT A REQUEST. */ @@ -225,6 +226,7 @@ extern uid_t audit_get_loginuid(struct audit_context *ctx); extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); extern int audit_socketcall(int nargs, unsigned long *args); extern int audit_sockaddr(int len, void *addr); +extern int audit_avc_path(struct dentry *dentry, struct vfsmount *mnt); extern void audit_signal_info(int sig, struct task_struct *t); #else #define audit_alloc(t) ({ 0; }) @@ -240,6 +242,7 @@ extern void audit_signal_info(int sig, struct task_struct *t); #define audit_ipc_perms(q,u,g,m) ({ 0; }) #define audit_socketcall(n,a) ({ 0; }) #define audit_sockaddr(len, addr) ({ 0; }) +#define audit_avc_path(dentry, mnt) ({ 0; }) #define audit_signal_info(s,t) do { ; } while (0) #endif diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 78d7a13fc86f..8dc5b2767145 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include #include @@ -124,6 +125,11 @@ struct audit_aux_data_sockaddr { char a[0]; }; +struct audit_aux_data_path { + struct audit_aux_data d; + struct dentry *dentry; + struct vfsmount *mnt; +}; /* The per-task audit context. */ struct audit_context { @@ -553,6 +559,11 @@ static inline void audit_free_aux(struct audit_context *context) struct audit_aux_data *aux; while ((aux = context->aux)) { + if (aux->type == AUDIT_AVC_PATH) { + struct audit_aux_data_path *axi = (void *)aux; + dput(axi->dentry); + mntput(axi->mnt); + } context->aux = aux->next; kfree(aux); } @@ -724,6 +735,14 @@ static void audit_log_exit(struct audit_context *context) audit_log_format(ab, "saddr="); audit_log_hex(ab, axs->a, axs->len); break; } + + case AUDIT_AVC_PATH: { + struct audit_aux_data_path *axi = (void *)aux; + audit_log_d_path(ab, "path=", axi->dentry, axi->mnt); + dput(axi->dentry); + mntput(axi->mnt); + break; } + } audit_log_end(ab); @@ -1124,6 +1143,27 @@ int audit_sockaddr(int len, void *a) return 0; } +int audit_avc_path(struct dentry *dentry, struct vfsmount *mnt) +{ + struct audit_aux_data_path *ax; + struct audit_context *context = current->audit_context; + + if (likely(!context)) + return 0; + + ax = kmalloc(sizeof(*ax), GFP_ATOMIC); + if (!ax) + return -ENOMEM; + + ax->dentry = dget(dentry); + ax->mnt = mntget(mnt); + + ax->d.type = AUDIT_AVC_PATH; + ax->d.next = context->aux; + context->aux = (void *)ax; + return 0; +} + void audit_signal_info(int sig, struct task_struct *t) { extern pid_t audit_sig_pid; diff --git a/security/selinux/avc.c b/security/selinux/avc.c index 62b963aca275..0fbc3e98c5ea 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -573,13 +573,10 @@ void avc_audit(u32 ssid, u32 tsid, case AVC_AUDIT_DATA_FS: if (a->u.fs.dentry) { struct dentry *dentry = a->u.fs.dentry; - if (a->u.fs.mnt) { - audit_log_d_path(ab, "path=", dentry, - a->u.fs.mnt); - } else { - audit_log_format(ab, " name=%s", - dentry->d_name.name); - } + if (a->u.fs.mnt) + audit_avc_path(dentry, a->u.fs.mnt); + audit_log_format(ab, " name=%s", + dentry->d_name.name); inode = dentry->d_inode; } else if (a->u.fs.inode) { struct dentry *dentry; @@ -630,8 +627,10 @@ void avc_audit(u32 ssid, u32 tsid, case AF_UNIX: u = unix_sk(sk); if (u->dentry) { - audit_log_d_path(ab, "path=", - u->dentry, u->mnt); + audit_avc_path(u->dentry, u->mnt); + audit_log_format(ab, " name=%s", + u->dentry->d_name.name); + break; } if (!u->addr) -- cgit v1.2.3 From bfb4496e7239c9132d732a65cdcf3d6a7431ad1a Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Sat, 21 May 2005 21:08:09 +0100 Subject: AUDIT: Assign serial number to non-syscall messages Move audit_serial() into audit.c and use it to generate serial numbers on messages even when there is no audit context from syscall auditing. This allows us to disambiguate audit records when more than one is generated in the same millisecond. Based on a patch by Steve Grubb after he observed the problem. Signed-off-by: David Woodhouse --- include/linux/audit.h | 7 ++++--- kernel/audit.c | 46 ++++++++++++++++++++++++++++++++++++++++++---- kernel/auditsc.c | 46 ++++++---------------------------------------- 3 files changed, 52 insertions(+), 47 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index 4b7caf0c6e10..3278ddf41ce6 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -219,8 +219,9 @@ extern void audit_inode(const char *name, const struct inode *inode); /* Private API (for audit.c only) */ extern int audit_receive_filter(int type, int pid, int uid, int seq, void *data, uid_t loginuid); -extern int audit_get_stamp(struct audit_context *ctx, - struct timespec *t, unsigned int *serial); +extern unsigned int audit_serial(void); +extern void auditsc_get_stamp(struct audit_context *ctx, + struct timespec *t, unsigned int *serial); extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); @@ -237,7 +238,7 @@ extern void audit_signal_info(int sig, struct task_struct *t); #define audit_putname(n) do { ; } while (0) #define audit_inode(n,i) do { ; } while (0) #define audit_receive_filter(t,p,u,s,d,l) ({ -EOPNOTSUPP; }) -#define audit_get_stamp(c,t,s) ({ 0; }) +#define auditsc_get_stamp(c,t,s) do { BUG(); } while (0) #define audit_get_loginuid(c) ({ -1; }) #define audit_ipc_perms(q,u,g,m) ({ 0; }) #define audit_socketcall(n,a) ({ 0; }) diff --git a/kernel/audit.c b/kernel/audit.c index f0a003acf621..35306f4369e7 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -597,6 +597,47 @@ err: return NULL; } +/* Compute a serial number for the audit record. Audit records are + * written to user-space as soon as they are generated, so a complete + * audit record may be written in several pieces. The timestamp of the + * record and this serial number are used by the user-space tools to + * determine which pieces belong to the same audit record. The + * (timestamp,serial) tuple is unique for each syscall and is live from + * syscall entry to syscall exit. + * + * Atomic values are only guaranteed to be 24-bit, so we count down. + * + * NOTE: Another possibility is to store the formatted records off the + * audit context (for those records that have a context), and emit them + * all at syscall exit. However, this could delay the reporting of + * significant errors until syscall exit (or never, if the system + * halts). */ +unsigned int audit_serial(void) +{ + static atomic_t serial = ATOMIC_INIT(0xffffff); + unsigned int a, b; + + do { + a = atomic_read(&serial); + if (atomic_dec_and_test(&serial)) + atomic_set(&serial, 0xffffff); + b = atomic_read(&serial); + } while (b != a - 1); + + return 0xffffff - b; +} + +static inline void audit_get_stamp(struct audit_context *ctx, + struct timespec *t, unsigned int *serial) +{ + if (ctx) + auditsc_get_stamp(ctx, t, serial); + else { + *t = CURRENT_TIME; + *serial = audit_serial(); + } +} + /* Obtain an audit buffer. This routine does locking to obtain the * audit buffer, but then no locking is required for calls to * audit_log_*format. If the tsk is a task that is currently in a @@ -630,10 +671,7 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, int type) return NULL; } - if (!audit_get_stamp(ab->ctx, &t, &serial)) { - t = CURRENT_TIME; - serial = 0; - } + audit_get_stamp(ab->ctx, &t, &serial); audit_log_format(ab, "audit(%lu.%03lu:%u): ", t.tv_sec, t.tv_nsec/1000000, serial); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 4193811d4fe1..74c2ae804ca8 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -795,36 +795,6 @@ void audit_free(struct task_struct *tsk) audit_free_context(context); } -/* Compute a serial number for the audit record. Audit records are - * written to user-space as soon as they are generated, so a complete - * audit record may be written in several pieces. The timestamp of the - * record and this serial number are used by the user-space tools to - * determine which pieces belong to the same audit record. The - * (timestamp,serial) tuple is unique for each syscall and is live from - * syscall entry to syscall exit. - * - * Atomic values are only guaranteed to be 24-bit, so we count down. - * - * NOTE: Another possibility is to store the formatted records off the - * audit context (for those records that have a context), and emit them - * all at syscall exit. However, this could delay the reporting of - * significant errors until syscall exit (or never, if the system - * halts). */ -static inline unsigned int audit_serial(void) -{ - static atomic_t serial = ATOMIC_INIT(0xffffff); - unsigned int a, b; - - do { - a = atomic_read(&serial); - if (atomic_dec_and_test(&serial)) - atomic_set(&serial, 0xffffff); - b = atomic_read(&serial); - } while (b != a - 1); - - return 0xffffff - b; -} - /* Fill in audit context at syscall entry. This only happens if the * audit context was created when the task was created and the state or * filters demand the audit context be built. If the state from the @@ -1042,17 +1012,13 @@ void audit_inode(const char *name, const struct inode *inode) context->names[idx].rdev = inode->i_rdev; } -int audit_get_stamp(struct audit_context *ctx, - struct timespec *t, unsigned int *serial) +void auditsc_get_stamp(struct audit_context *ctx, + struct timespec *t, unsigned int *serial) { - if (ctx) { - t->tv_sec = ctx->ctime.tv_sec; - t->tv_nsec = ctx->ctime.tv_nsec; - *serial = ctx->serial; - ctx->auditable = 1; - return 1; - } - return 0; + t->tv_sec = ctx->ctime.tv_sec; + t->tv_nsec = ctx->ctime.tv_nsec; + *serial = ctx->serial; + ctx->auditable = 1; } int audit_set_loginuid(struct task_struct *task, uid_t loginuid) -- cgit v1.2.3 From deaf2622bdc74270381873c7936c91500c683853 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 23 May 2005 12:45:26 -0700 Subject: [NET]: Kill stray reference to sock->passcred. That struct member was deleted, but a comment was not updated to reflect this. Signed-off-by: David S. Miller --- include/linux/net.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/net.h b/include/linux/net.h index 6d997ff3f103..20cb226b2268 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -101,7 +101,6 @@ enum sock_type { * @sk: internal networking protocol agnostic socket representation * @wait: wait queue for several uses * @type: socket type (%SOCK_STREAM, etc) - * @passcred: credentials (used only in Unix Sockets (aka PF_LOCAL)) */ struct socket { socket_state state; -- cgit v1.2.3 From 82d1542c1df6b3a0105843deddebeff03068d1dc Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Mon, 23 May 2005 12:53:14 -0700 Subject: [NET]: include/linux/if_tr.h clean-up I removed the ethernet definitions (which were commented out) and cleaned up the tabs. Signed-off-by: Jon Mason --- include/linux/if_tr.h | 39 +++++++++++++++------------------------ 1 file changed, 15 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/include/linux/if_tr.h b/include/linux/if_tr.h index 4fd451f81ccb..a2b01e1e72f2 100644 --- a/include/linux/if_tr.h +++ b/include/linux/if_tr.h @@ -9,7 +9,7 @@ * * Author: Fred N. van Kempen, * Donald Becker, - * Peter De Schrijver, + * Peter De Schrijver, * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -19,24 +19,16 @@ #ifndef _LINUX_IF_TR_H #define _LINUX_IF_TR_H - /* IEEE 802.5 Token-Ring magic constants. The frame sizes omit the preamble and FCS/CRC (frame check sequence). */ -#define TR_ALEN 6 /* Octets in one ethernet addr */ -#define TR_HLEN (sizeof(struct trh_hdr)+sizeof(struct trllc)) -#define AC 0x10 -#define LLC_FRAME 0x40 -#if 0 -#define ETH_HLEN 14 /* Total octets in header. */ -#define ETH_ZLEN 60 /* Min. octets in frame sans FCS */ -#define ETH_DATA_LEN 1500 /* Max. octets in payload */ -#define ETH_FRAME_LEN 1514 /* Max. octets in frame sans FCS */ -#endif - +#define TR_ALEN 6 /* Octets in one token-ring addr */ +#define TR_HLEN (sizeof(struct trh_hdr)+sizeof(struct trllc)) +#define AC 0x10 +#define LLC_FRAME 0x40 /* LLC and SNAP constants */ -#define EXTENDED_SAP 0xAA -#define UI_CMD 0x03 +#define EXTENDED_SAP 0xAA +#define UI_CMD 0x03 /* This is an Token-Ring frame header. */ struct trh_hdr { @@ -96,14 +88,13 @@ struct tr_statistics { }; /* source routing stuff */ - -#define TR_RII 0x80 -#define TR_RCF_DIR_BIT 0x80 -#define TR_RCF_LEN_MASK 0x1f00 -#define TR_RCF_BROADCAST 0x8000 /* all-routes broadcast */ -#define TR_RCF_LIMITED_BROADCAST 0xC000 /* single-route broadcast */ -#define TR_RCF_FRAME2K 0x20 -#define TR_RCF_BROADCAST_MASK 0xC000 -#define TR_MAXRIFLEN 18 +#define TR_RII 0x80 +#define TR_RCF_DIR_BIT 0x80 +#define TR_RCF_LEN_MASK 0x1f00 +#define TR_RCF_BROADCAST 0x8000 /* all-routes broadcast */ +#define TR_RCF_LIMITED_BROADCAST 0xC000 /* single-route broadcast */ +#define TR_RCF_FRAME2K 0x20 +#define TR_RCF_BROADCAST_MASK 0xC000 +#define TR_MAXRIFLEN 18 #endif /* _LINUX_IF_TR_H */ -- cgit v1.2.3 From d8659255c599ba497bcfb8b4abc187d9e0951d99 Mon Sep 17 00:00:00 2001 From: Xose Vazquez Perez Date: Mon, 23 May 2005 12:54:51 -0700 Subject: [TG3]: Add 5752M device ID. Add 0x1601 as 5752M, it's a 5752 but for mobile PCs. Stolen from Broadcom bcm5700-8.1.55 driver. Someone forgot to add it to tg3 ;-) Signed-off-by: David S. Miller --- drivers/net/tg3.c | 2 ++ include/linux/pci_ids.h | 1 + 2 files changed, 3 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 4d2bdbdd34e8..ebd882fbc705 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -206,6 +206,8 @@ static struct pci_device_id tg3_pci_tbl[] = { PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5752, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5752M, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5753, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5753M, diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index ae27792b5aa4..7b9720e35361 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2064,6 +2064,7 @@ #define PCI_VENDOR_ID_BROADCOM 0x14e4 #define PCI_DEVICE_ID_TIGON3_5752 0x1600 +#define PCI_DEVICE_ID_TIGON3_5752M 0x1601 #define PCI_DEVICE_ID_TIGON3_5700 0x1644 #define PCI_DEVICE_ID_TIGON3_5701 0x1645 #define PCI_DEVICE_ID_TIGON3_5702 0x1646 -- cgit v1.2.3 From 205902ecc7bcb2a007ed27bb3518e8f70e5d113e Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 23 May 2005 13:09:19 -0700 Subject: [NET]: netdevice.h: be'ify packet_type Everybody does struct packet_type foo_packet_type = { .type = __constant_htons(ETH_P_FOO); }; 5 introduced warnings will be properly fixed later. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ac11d73be4ce..b25bd02720d3 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -503,7 +503,7 @@ static inline void *netdev_priv(struct net_device *dev) #define SET_NETDEV_DEV(net, pdev) ((net)->class_dev.dev = (pdev)) struct packet_type { - unsigned short type; /* This is really htons(ether_type). */ + __be16 type; /* This is really htons(ether_type). */ struct net_device *dev; /* NULL is wildcarded here */ int (*func) (struct sk_buff *, struct net_device *, struct packet_type *); -- cgit v1.2.3 From 7e43c84e3e5423ae72ae31e4cd7bbccfd1605194 Mon Sep 17 00:00:00 2001 From: "blaisorblade@yahoo.it" Date: Wed, 25 May 2005 01:31:42 +0200 Subject: [PATCH] Cleanup DEFINE_WAIT Use LIST_HEAD_INIT rather than doing it by hand in DEFINE_WAIT. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- include/linux/wait.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/wait.h b/include/linux/wait.h index 17c874a8eb3f..c9486c3efb4a 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -386,9 +386,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key); wait_queue_t name = { \ .task = current, \ .func = autoremove_wake_function, \ - .task_list = { .next = &(name).task_list, \ - .prev = &(name).task_list, \ - }, \ + .task_list = LIST_HEAD_INIT((name).task_list), \ } #define DEFINE_WAIT_BIT(name, word, bit) \ -- cgit v1.2.3 From 8604affde9d4f52f04342d6a37c77d95fa167e7a Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Thu, 26 May 2005 14:55:34 +0200 Subject: [PATCH] convert IDE device drivers to driver-model * add ide_bus_match() and export ide_bus_type * split ide_remove_driver_from_hwgroup() out of ide_unregister() * move device cleanup from ide_unregister() to drive_release_dev() * convert ide_driver_t->name to driver->name * convert ide_driver_t->{attach,cleanup} to driver->{probe,remove} * remove ide_driver_t->busy as ide_bus_type->subsys.rwsem protects against concurrent ->{probe,remove} calls * make ide_{un}register_driver() void as it cannot fail now * use driver_{un}register() directly, remove ide_{un}register_driver() * use device_register() instead of ata_attach(), remove ata_attach() * add proc_print_driver() and ide_drivers_show(), remove ide_drivers_op * fix ide_replace_subdriver() and move it to ide-proc.c * remove ide_driver_t->drives, ide_drives and drives_lock * remove ide_driver_t->drivers, drivers and drivers_lock * remove ide_drive_t->driver and DRIVER() macro Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-cd.c | 47 +++----- drivers/ide/ide-disk.c | 41 +++---- drivers/ide/ide-floppy.c | 42 +++---- drivers/ide/ide-probe.c | 47 +++++++- drivers/ide/ide-proc.c | 52 +++++++- drivers/ide/ide-tape.c | 51 +++----- drivers/ide/ide.c | 307 ++++------------------------------------------- drivers/scsi/ide-scsi.c | 58 ++++----- include/linux/ide.h | 20 +-- 9 files changed, 216 insertions(+), 449 deletions(-) (limited to 'include/linux') diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index 4f7ce7056228..f0bd242e030f 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -3255,16 +3255,12 @@ sector_t ide_cdrom_capacity (ide_drive_t *drive) return capacity * sectors_per_frame; } -static -int ide_cdrom_cleanup(ide_drive_t *drive) +static int ide_cd_remove(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); struct cdrom_info *info = drive->driver_data; - if (ide_unregister_subdriver(drive)) { - printk(KERN_ERR "%s: %s: failed to ide_unregister_subdriver\n", - __FUNCTION__, drive->name); - return 1; - } + ide_unregister_subdriver(drive, info->driver); del_gendisk(info->disk); @@ -3297,7 +3293,7 @@ static void ide_cd_release(struct kref *kref) kfree(info); } -static int ide_cdrom_attach (ide_drive_t *drive); +static int ide_cd_probe(struct device *); #ifdef CONFIG_PROC_FS static int proc_idecd_read_capacity @@ -3320,19 +3316,20 @@ static ide_proc_entry_t idecd_proc[] = { static ide_driver_t ide_cdrom_driver = { .owner = THIS_MODULE, - .name = "ide-cdrom", + .gen_driver = { + .name = "ide-cdrom", + .bus = &ide_bus_type, + .probe = ide_cd_probe, + .remove = ide_cd_remove, + }, .version = IDECD_VERSION, .media = ide_cdrom, - .busy = 0, .supports_dsc_overlap = 1, - .cleanup = ide_cdrom_cleanup, .do_request = ide_do_rw_cdrom, .end_request = ide_end_request, .error = __ide_error, .abort = __ide_abort, .proc = idecd_proc, - .attach = ide_cdrom_attach, - .drives = LIST_HEAD_INIT(ide_cdrom_driver.drives), }; static int idecd_open(struct inode * inode, struct file * file) @@ -3418,8 +3415,9 @@ static char *ignore = NULL; module_param(ignore, charp, 0400); MODULE_DESCRIPTION("ATAPI CD-ROM Driver"); -static int ide_cdrom_attach (ide_drive_t *drive) +static int ide_cd_probe(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); struct cdrom_info *info; struct gendisk *g; struct request_sense sense; @@ -3453,11 +3451,8 @@ static int ide_cdrom_attach (ide_drive_t *drive) ide_init_disk(g, drive); - if (ide_register_subdriver(drive, &ide_cdrom_driver)) { - printk(KERN_ERR "%s: Failed to register the driver with ide.c\n", - drive->name); - goto out_put_disk; - } + ide_register_subdriver(drive, &ide_cdrom_driver); + memset(info, 0, sizeof (struct cdrom_info)); kref_init(&info->kref); @@ -3470,7 +3465,6 @@ static int ide_cdrom_attach (ide_drive_t *drive) drive->driver_data = info; - DRIVER(drive)->busy++; g->minors = 1; snprintf(g->devfs_name, sizeof(g->devfs_name), "%s/cd", drive->devfs_name); @@ -3478,8 +3472,7 @@ static int ide_cdrom_attach (ide_drive_t *drive) g->flags = GENHD_FL_CD | GENHD_FL_REMOVABLE; if (ide_cdrom_setup(drive)) { struct cdrom_device_info *devinfo = &info->devinfo; - DRIVER(drive)->busy--; - ide_unregister_subdriver(drive); + ide_unregister_subdriver(drive, &ide_cdrom_driver); if (info->buffer != NULL) kfree(info->buffer); if (info->toc != NULL) @@ -3492,7 +3485,6 @@ static int ide_cdrom_attach (ide_drive_t *drive) drive->driver_data = NULL; goto failed; } - DRIVER(drive)->busy--; cdrom_read_toc(drive, &sense); g->fops = &idecd_ops; @@ -3500,23 +3492,20 @@ static int ide_cdrom_attach (ide_drive_t *drive) add_disk(g); return 0; -out_put_disk: - put_disk(g); out_free_cd: kfree(info); failed: - return 1; + return -ENODEV; } static void __exit ide_cdrom_exit(void) { - ide_unregister_driver(&ide_cdrom_driver); + driver_unregister(&ide_cdrom_driver.gen_driver); } static int ide_cdrom_init(void) { - ide_register_driver(&ide_cdrom_driver); - return 0; + return driver_register(&ide_cdrom_driver.gen_driver); } module_init(ide_cdrom_init); diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c index 5d54f7756100..3302cd8eab4c 100644 --- a/drivers/ide/ide-disk.c +++ b/drivers/ide/ide-disk.c @@ -1024,14 +1024,16 @@ static void ide_cacheflush_p(ide_drive_t *drive) printk(KERN_INFO "%s: wcache flush failed!\n", drive->name); } -static int idedisk_cleanup (ide_drive_t *drive) +static int ide_disk_remove(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); struct ide_disk_obj *idkp = drive->driver_data; struct gendisk *g = idkp->disk; ide_cacheflush_p(drive); - if (ide_unregister_subdriver(drive)) - return 1; + + ide_unregister_subdriver(drive, idkp->driver); + del_gendisk(g); ide_disk_put(idkp); @@ -1052,7 +1054,7 @@ static void ide_disk_release(struct kref *kref) kfree(idkp); } -static int idedisk_attach(ide_drive_t *drive); +static int ide_disk_probe(struct device *dev); static void ide_device_shutdown(struct device *dev) { @@ -1082,27 +1084,23 @@ static void ide_device_shutdown(struct device *dev) dev->bus->suspend(dev, PMSG_SUSPEND); } -/* - * IDE subdriver functions, registered with ide.c - */ static ide_driver_t idedisk_driver = { .owner = THIS_MODULE, .gen_driver = { + .name = "ide-disk", + .bus = &ide_bus_type, + .probe = ide_disk_probe, + .remove = ide_disk_remove, .shutdown = ide_device_shutdown, }, - .name = "ide-disk", .version = IDEDISK_VERSION, .media = ide_disk, - .busy = 0, .supports_dsc_overlap = 0, - .cleanup = idedisk_cleanup, .do_request = ide_do_rw_disk, .end_request = ide_end_request, .error = __ide_error, .abort = __ide_abort, .proc = idedisk_proc, - .attach = idedisk_attach, - .drives = LIST_HEAD_INIT(idedisk_driver.drives), }; static int idedisk_open(struct inode *inode, struct file *filp) @@ -1199,8 +1197,9 @@ static struct block_device_operations idedisk_ops = { MODULE_DESCRIPTION("ATA DISK Driver"); -static int idedisk_attach(ide_drive_t *drive) +static int ide_disk_probe(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); struct ide_disk_obj *idkp; struct gendisk *g; @@ -1222,10 +1221,7 @@ static int idedisk_attach(ide_drive_t *drive) ide_init_disk(g, drive); - if (ide_register_subdriver(drive, &idedisk_driver)) { - printk (KERN_ERR "ide-disk: %s: Failed to register the driver with ide.c\n", drive->name); - goto out_put_disk; - } + ide_register_subdriver(drive, &idedisk_driver); memset(idkp, 0, sizeof(*idkp)); @@ -1239,7 +1235,6 @@ static int idedisk_attach(ide_drive_t *drive) drive->driver_data = idkp; - DRIVER(drive)->busy++; idedisk_setup(drive); if ((!drive->head || drive->head > 16) && !drive->select.b.lba) { printk(KERN_ERR "%s: INVALID GEOMETRY: %d PHYSICAL HEADS?\n", @@ -1247,7 +1242,7 @@ static int idedisk_attach(ide_drive_t *drive) drive->attach = 0; } else drive->attach = 1; - DRIVER(drive)->busy--; + g->minors = 1 << PARTN_BITS; strcpy(g->devfs_name, drive->devfs_name); g->driverfs_dev = &drive->gendev; @@ -1257,22 +1252,20 @@ static int idedisk_attach(ide_drive_t *drive) add_disk(g); return 0; -out_put_disk: - put_disk(g); out_free_idkp: kfree(idkp); failed: - return 1; + return -ENODEV; } static void __exit idedisk_exit (void) { - ide_unregister_driver(&idedisk_driver); + driver_unregister(&idedisk_driver.gen_driver); } static int idedisk_init (void) { - return ide_register_driver(&idedisk_driver); + return driver_register(&idedisk_driver.gen_driver); } module_init(idedisk_init); diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c index 36c0b74a4e45..c949e98df4b6 100644 --- a/drivers/ide/ide-floppy.c +++ b/drivers/ide/ide-floppy.c @@ -1865,13 +1865,13 @@ static void idefloppy_setup (ide_drive_t *drive, idefloppy_floppy_t *floppy) idefloppy_add_settings(drive); } -static int idefloppy_cleanup (ide_drive_t *drive) +static int ide_floppy_remove(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); idefloppy_floppy_t *floppy = drive->driver_data; struct gendisk *g = floppy->disk; - if (ide_unregister_subdriver(drive)) - return 1; + ide_unregister_subdriver(drive, floppy->driver); del_gendisk(g); @@ -1916,26 +1916,24 @@ static ide_proc_entry_t idefloppy_proc[] = { #endif /* CONFIG_PROC_FS */ -static int idefloppy_attach(ide_drive_t *drive); +static int ide_floppy_probe(struct device *); -/* - * IDE subdriver functions, registered with ide.c - */ static ide_driver_t idefloppy_driver = { .owner = THIS_MODULE, - .name = "ide-floppy", + .gen_driver = { + .name = "ide-floppy", + .bus = &ide_bus_type, + .probe = ide_floppy_probe, + .remove = ide_floppy_remove, + }, .version = IDEFLOPPY_VERSION, .media = ide_floppy, - .busy = 0, .supports_dsc_overlap = 0, - .cleanup = idefloppy_cleanup, .do_request = idefloppy_do_request, .end_request = idefloppy_do_end_request, .error = __ide_error, .abort = __ide_abort, .proc = idefloppy_proc, - .attach = idefloppy_attach, - .drives = LIST_HEAD_INIT(idefloppy_driver.drives), }; static int idefloppy_open(struct inode *inode, struct file *filp) @@ -2122,8 +2120,9 @@ static struct block_device_operations idefloppy_ops = { .revalidate_disk= idefloppy_revalidate_disk }; -static int idefloppy_attach (ide_drive_t *drive) +static int ide_floppy_probe(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); idefloppy_floppy_t *floppy; struct gendisk *g; @@ -2152,10 +2151,7 @@ static int idefloppy_attach (ide_drive_t *drive) ide_init_disk(g, drive); - if (ide_register_subdriver(drive, &idefloppy_driver)) { - printk (KERN_ERR "ide-floppy: %s: Failed to register the driver with ide.c\n", drive->name); - goto out_put_disk; - } + ide_register_subdriver(drive, &idefloppy_driver); memset(floppy, 0, sizeof(*floppy)); @@ -2169,9 +2165,8 @@ static int idefloppy_attach (ide_drive_t *drive) drive->driver_data = floppy; - DRIVER(drive)->busy++; idefloppy_setup (drive, floppy); - DRIVER(drive)->busy--; + g->minors = 1 << PARTN_BITS; g->driverfs_dev = &drive->gendev; strcpy(g->devfs_name, drive->devfs_name); @@ -2181,19 +2176,17 @@ static int idefloppy_attach (ide_drive_t *drive) add_disk(g); return 0; -out_put_disk: - put_disk(g); out_free_floppy: kfree(floppy); failed: - return 1; + return -ENODEV; } MODULE_DESCRIPTION("ATAPI FLOPPY Driver"); static void __exit idefloppy_exit (void) { - ide_unregister_driver(&idefloppy_driver); + driver_unregister(&idefloppy_driver.gen_driver); } /* @@ -2202,8 +2195,7 @@ static void __exit idefloppy_exit (void) static int idefloppy_init (void) { printk("ide-floppy driver " IDEFLOPPY_VERSION "\n"); - ide_register_driver(&idefloppy_driver); - return 0; + return driver_register(&idefloppy_driver.gen_driver); } module_init(idefloppy_init); diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c index 554473a95cf7..53b84a84f6cb 100644 --- a/drivers/ide/ide-probe.c +++ b/drivers/ide/ide-probe.c @@ -47,6 +47,7 @@ #include #include #include +#include #include #include #include @@ -918,7 +919,7 @@ int probe_hwif_init_with_fixup(ide_hwif_t *hwif, void (*fixup)(ide_hwif_t *hwif) want them on default or a new "empty" class for hotplug reprobing ? */ if (drive->present) { - ata_attach(drive); + device_register(&drive->gendev); } } } @@ -1279,10 +1280,51 @@ void ide_init_disk(struct gendisk *disk, ide_drive_t *drive) EXPORT_SYMBOL_GPL(ide_init_disk); +static void ide_remove_drive_from_hwgroup(ide_drive_t *drive) +{ + ide_hwgroup_t *hwgroup = drive->hwif->hwgroup; + + if (drive == drive->next) { + /* special case: last drive from hwgroup. */ + BUG_ON(hwgroup->drive != drive); + hwgroup->drive = NULL; + } else { + ide_drive_t *walk; + + walk = hwgroup->drive; + while (walk->next != drive) + walk = walk->next; + walk->next = drive->next; + if (hwgroup->drive == drive) { + hwgroup->drive = drive->next; + hwgroup->hwif = hwgroup->drive->hwif; + } + } + BUG_ON(hwgroup->drive == drive); +} + static void drive_release_dev (struct device *dev) { ide_drive_t *drive = container_of(dev, ide_drive_t, gendev); + spin_lock_irq(&ide_lock); + if (drive->devfs_name[0] != '\0') { + devfs_remove(drive->devfs_name); + drive->devfs_name[0] = '\0'; + } + ide_remove_drive_from_hwgroup(drive); + if (drive->id != NULL) { + kfree(drive->id); + drive->id = NULL; + } + drive->present = 0; + /* Messed up locking ... */ + spin_unlock_irq(&ide_lock); + blk_cleanup_queue(drive->queue); + spin_lock_irq(&ide_lock); + drive->queue = NULL; + spin_unlock_irq(&ide_lock); + up(&drive->gendev_rel_sem); } @@ -1306,7 +1348,6 @@ static void init_gendisk (ide_hwif_t *hwif) drive->gendev.driver_data = drive; drive->gendev.release = drive_release_dev; if (drive->present) { - device_register(&drive->gendev); sprintf(drive->devfs_name, "ide/host%d/bus%d/target%d/lun%d", (hwif->channel && hwif->mate) ? hwif->mate->index : hwif->index, @@ -1412,7 +1453,7 @@ int ideprobe_init (void) hwif->chipset = ide_generic; for (unit = 0; unit < MAX_DRIVES; ++unit) if (hwif->drives[unit].present) - ata_attach(&hwif->drives[unit]); + device_register(&hwif->drives[unit].gendev); } } return 0; diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c index 4b1e43b4118b..4063d2c34e3d 100644 --- a/drivers/ide/ide-proc.c +++ b/drivers/ide/ide-proc.c @@ -307,17 +307,41 @@ static int proc_ide_read_driver (char *page, char **start, off_t off, int count, int *eof, void *data) { ide_drive_t *drive = (ide_drive_t *) data; - ide_driver_t *driver = drive->driver; + struct device *dev = &drive->gendev; + ide_driver_t *ide_drv; int len; - if (driver) { + down_read(&dev->bus->subsys.rwsem); + if (dev->driver) { + ide_drv = container_of(dev->driver, ide_driver_t, gen_driver); len = sprintf(page, "%s version %s\n", - driver->name, driver->version); + dev->driver->name, ide_drv->version); } else len = sprintf(page, "ide-default version 0.9.newide\n"); + up_read(&dev->bus->subsys.rwsem); PROC_IDE_READ_RETURN(page,start,off,count,eof,len); } +static int ide_replace_subdriver(ide_drive_t *drive, const char *driver) +{ + struct device *dev = &drive->gendev; + int ret = 1; + + down_write(&dev->bus->subsys.rwsem); + device_release_driver(dev); + /* FIXME: device can still be in use by previous driver */ + strlcpy(drive->driver_req, driver, sizeof(drive->driver_req)); + device_attach(dev); + drive->driver_req[0] = 0; + if (dev->driver == NULL) + device_attach(dev); + if (dev->driver && !strcmp(dev->driver->name, driver)) + ret = 0; + up_write(&dev->bus->subsys.rwsem); + + return ret; +} + static int proc_ide_write_driver (struct file *file, const char __user *buffer, unsigned long count, void *data) { @@ -488,16 +512,32 @@ void destroy_proc_ide_interface(ide_hwif_t *hwif) } } -extern struct seq_operations ide_drivers_op; +static int proc_print_driver(struct device_driver *drv, void *data) +{ + ide_driver_t *ide_drv = container_of(drv, ide_driver_t, gen_driver); + struct seq_file *s = data; + + seq_printf(s, "%s version %s\n", drv->name, ide_drv->version); + + return 0; +} + +static int ide_drivers_show(struct seq_file *s, void *p) +{ + bus_for_each_drv(&ide_bus_type, NULL, s, proc_print_driver); + return 0; +} + static int ide_drivers_open(struct inode *inode, struct file *file) { - return seq_open(file, &ide_drivers_op); + return single_open(file, &ide_drivers_show, NULL); } + static struct file_operations ide_drivers_operations = { .open = ide_drivers_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = single_release, }; void proc_ide_create(void) diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c index 482544854985..5a3dc46008e6 100644 --- a/drivers/ide/ide-tape.c +++ b/drivers/ide/ide-tape.c @@ -4681,21 +4681,12 @@ static void idetape_setup (ide_drive_t *drive, idetape_tape_t *tape, int minor) idetape_add_settings(drive); } -static int idetape_cleanup (ide_drive_t *drive) +static int ide_tape_remove(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); idetape_tape_t *tape = drive->driver_data; - unsigned long flags; - - spin_lock_irqsave(&ide_lock, flags); - if (test_bit(IDETAPE_BUSY, &tape->flags) || drive->usage || - tape->first_stage != NULL || tape->merge_stage_size) { - spin_unlock_irqrestore(&ide_lock, flags); - return 1; - } - spin_unlock_irqrestore(&ide_lock, flags); - DRIVER(drive)->busy = 0; - (void) ide_unregister_subdriver(drive); + ide_unregister_subdriver(drive, tape->driver); ide_unregister_region(tape->disk); @@ -4710,6 +4701,8 @@ static void ide_tape_release(struct kref *kref) ide_drive_t *drive = tape->drive; struct gendisk *g = tape->disk; + BUG_ON(tape->first_stage != NULL || tape->merge_stage_size); + drive->dsc_overlap = 0; drive->driver_data = NULL; devfs_remove("%s/mt", drive->devfs_name); @@ -4747,26 +4740,24 @@ static ide_proc_entry_t idetape_proc[] = { #endif -static int idetape_attach(ide_drive_t *drive); +static int ide_tape_probe(struct device *); -/* - * IDE subdriver functions, registered with ide.c - */ static ide_driver_t idetape_driver = { .owner = THIS_MODULE, - .name = "ide-tape", + .gen_driver = { + .name = "ide-tape", + .bus = &ide_bus_type, + .probe = ide_tape_probe, + .remove = ide_tape_remove, + }, .version = IDETAPE_VERSION, .media = ide_tape, - .busy = 1, .supports_dsc_overlap = 1, - .cleanup = idetape_cleanup, .do_request = idetape_do_request, .end_request = idetape_end_request, .error = __ide_error, .abort = __ide_abort, .proc = idetape_proc, - .attach = idetape_attach, - .drives = LIST_HEAD_INIT(idetape_driver.drives), }; /* @@ -4829,8 +4820,9 @@ static struct block_device_operations idetape_block_ops = { .ioctl = idetape_ioctl, }; -static int idetape_attach (ide_drive_t *drive) +static int ide_tape_probe(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); idetape_tape_t *tape; struct gendisk *g; int minor; @@ -4865,10 +4857,7 @@ static int idetape_attach (ide_drive_t *drive) ide_init_disk(g, drive); - if (ide_register_subdriver(drive, &idetape_driver)) { - printk(KERN_ERR "ide-tape: %s: Failed to register the driver with ide.c\n", drive->name); - goto out_put_disk; - } + ide_register_subdriver(drive, &idetape_driver); memset(tape, 0, sizeof(*tape)); @@ -4902,12 +4891,11 @@ static int idetape_attach (ide_drive_t *drive) ide_register_region(g); return 0; -out_put_disk: - put_disk(g); + out_free_tape: kfree(tape); failed: - return 1; + return -ENODEV; } MODULE_DESCRIPTION("ATAPI Streaming TAPE Driver"); @@ -4915,7 +4903,7 @@ MODULE_LICENSE("GPL"); static void __exit idetape_exit (void) { - ide_unregister_driver(&idetape_driver); + driver_unregister(&idetape_driver.gen_driver); unregister_chrdev(IDETAPE_MAJOR, "ht"); } @@ -4928,8 +4916,7 @@ static int idetape_init (void) printk(KERN_ERR "ide-tape: Failed to register character device interface\n"); return -EBUSY; } - ide_register_driver(&idetape_driver); - return 0; + return driver_register(&idetape_driver.gen_driver); } module_init(idetape_init); diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index 973dec799b5c..dae1bd5b8c3e 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -196,8 +196,6 @@ ide_hwif_t ide_hwifs[MAX_HWIFS]; /* master data repository */ EXPORT_SYMBOL(ide_hwifs); -static struct list_head ide_drives = LIST_HEAD_INIT(ide_drives); - /* * Do not even *think* about calling this! */ @@ -358,54 +356,6 @@ static int ide_system_bus_speed(void) return system_bus_speed; } -/* - * drives_lock protects the list of drives, drivers_lock the - * list of drivers. Currently nobody takes both at once. - */ - -static DEFINE_SPINLOCK(drives_lock); -static DEFINE_SPINLOCK(drivers_lock); -static LIST_HEAD(drivers); - -/* Iterator for the driver list. */ - -static void *m_start(struct seq_file *m, loff_t *pos) -{ - struct list_head *p; - loff_t l = *pos; - spin_lock(&drivers_lock); - list_for_each(p, &drivers) - if (!l--) - return list_entry(p, ide_driver_t, drivers); - return NULL; -} - -static void *m_next(struct seq_file *m, void *v, loff_t *pos) -{ - struct list_head *p = ((ide_driver_t *)v)->drivers.next; - (*pos)++; - return p==&drivers ? NULL : list_entry(p, ide_driver_t, drivers); -} - -static void m_stop(struct seq_file *m, void *v) -{ - spin_unlock(&drivers_lock); -} - -static int show_driver(struct seq_file *m, void *v) -{ - ide_driver_t *driver = v; - seq_printf(m, "%s version %s\n", driver->name, driver->version); - return 0; -} - -struct seq_operations ide_drivers_op = { - .start = m_start, - .next = m_next, - .stop = m_stop, - .show = show_driver -}; - #ifdef CONFIG_PROC_FS struct proc_dir_entry *proc_ide_root; #endif @@ -630,7 +580,7 @@ void ide_unregister(unsigned int index) ide_hwif_t *hwif, *g; static ide_hwif_t tmp_hwif; /* protected by ide_cfg_sem */ ide_hwgroup_t *hwgroup; - int irq_count = 0, unit, i; + int irq_count = 0, unit; BUG_ON(index >= MAX_HWIFS); @@ -643,23 +593,22 @@ void ide_unregister(unsigned int index) goto abort; for (unit = 0; unit < MAX_DRIVES; ++unit) { drive = &hwif->drives[unit]; - if (!drive->present) + if (!drive->present) { + if (drive->devfs_name[0] != '\0') { + devfs_remove(drive->devfs_name); + drive->devfs_name[0] = '\0'; + } continue; - if (drive->usage || DRIVER(drive)->busy) - goto abort; - drive->dead = 1; + } + spin_unlock_irq(&ide_lock); + device_unregister(&drive->gendev); + down(&drive->gendev_rel_sem); + spin_lock_irq(&ide_lock); } hwif->present = 0; spin_unlock_irq(&ide_lock); - for (unit = 0; unit < MAX_DRIVES; ++unit) { - drive = &hwif->drives[unit]; - if (!drive->present) - continue; - DRIVER(drive)->cleanup(drive); - } - destroy_proc_ide_interface(hwif); hwgroup = hwif->hwgroup; @@ -687,44 +636,6 @@ void ide_unregister(unsigned int index) * Remove us from the hwgroup, and free * the hwgroup if we were the only member */ - for (i = 0; i < MAX_DRIVES; ++i) { - drive = &hwif->drives[i]; - if (drive->devfs_name[0] != '\0') { - devfs_remove(drive->devfs_name); - drive->devfs_name[0] = '\0'; - } - if (!drive->present) - continue; - if (drive == drive->next) { - /* special case: last drive from hwgroup. */ - BUG_ON(hwgroup->drive != drive); - hwgroup->drive = NULL; - } else { - ide_drive_t *walk; - - walk = hwgroup->drive; - while (walk->next != drive) - walk = walk->next; - walk->next = drive->next; - if (hwgroup->drive == drive) { - hwgroup->drive = drive->next; - hwgroup->hwif = HWIF(hwgroup->drive); - } - } - BUG_ON(hwgroup->drive == drive); - if (drive->id != NULL) { - kfree(drive->id); - drive->id = NULL; - } - drive->present = 0; - /* Messed up locking ... */ - spin_unlock_irq(&ide_lock); - blk_cleanup_queue(drive->queue); - device_unregister(&drive->gendev); - down(&drive->gendev_rel_sem); - spin_lock_irq(&ide_lock); - drive->queue = NULL; - } if (hwif->next == hwif) { BUG_ON(hwgroup->hwif != hwif); kfree(hwgroup); @@ -1304,73 +1215,6 @@ int system_bus_clock (void) EXPORT_SYMBOL(system_bus_clock); -/* - * Locking is badly broken here - since way back. That sucker is - * root-only, but that's not an excuse... The real question is what - * exclusion rules do we want here. - */ -int ide_replace_subdriver (ide_drive_t *drive, const char *driver) -{ - if (!drive->present || drive->usage || drive->dead) - goto abort; - if (DRIVER(drive)->cleanup(drive)) - goto abort; - strlcpy(drive->driver_req, driver, sizeof(drive->driver_req)); - if (ata_attach(drive)) { - spin_lock(&drives_lock); - list_del_init(&drive->list); - spin_unlock(&drives_lock); - drive->driver_req[0] = 0; - ata_attach(drive); - } else { - drive->driver_req[0] = 0; - } - if (drive->driver && !strcmp(drive->driver->name, driver)) - return 0; -abort: - return 1; -} - -/** - * ata_attach - attach an ATA/ATAPI device - * @drive: drive to attach - * - * Takes a drive that is as yet not assigned to any midlayer IDE - * driver (or is assigned to the default driver) and figures out - * which driver would like to own it. If nobody claims the drive - * then it is automatically attached to the default driver used for - * unclaimed objects. - * - * A return of zero indicates attachment to a driver, of one - * attachment to the default driver. - * - * Takes drivers_lock. - */ - -int ata_attach(ide_drive_t *drive) -{ - struct list_head *p; - spin_lock(&drivers_lock); - list_for_each(p, &drivers) { - ide_driver_t *driver = list_entry(p, ide_driver_t, drivers); - if (!try_module_get(driver->owner)) - continue; - spin_unlock(&drivers_lock); - if (driver->attach(drive) == 0) { - module_put(driver->owner); - drive->gendev.driver = &driver->gen_driver; - return 0; - } - spin_lock(&drivers_lock); - module_put(driver->owner); - } - drive->gendev.driver = NULL; - spin_unlock(&drivers_lock); - if (ide_register_subdriver(drive, NULL)) - panic("ide: default attach failed"); - return 1; -} - static int generic_ide_suspend(struct device *dev, pm_message_t state) { ide_drive_t *drive = dev->driver_data; @@ -2013,27 +1857,11 @@ static void __init probe_for_hwifs (void) #endif } -int ide_register_subdriver(ide_drive_t *drive, ide_driver_t *driver) +void ide_register_subdriver(ide_drive_t *drive, ide_driver_t *driver) { - unsigned long flags; - - spin_lock_irqsave(&ide_lock, flags); - if (!drive->present || drive->driver != NULL || - drive->usage || drive->dead) { - spin_unlock_irqrestore(&ide_lock, flags); - return 1; - } - drive->driver = driver; - spin_unlock_irqrestore(&ide_lock, flags); - spin_lock(&drives_lock); - list_add_tail(&drive->list, driver ? &driver->drives : &ide_drives); - spin_unlock(&drives_lock); -// printk(KERN_INFO "%s: attached %s driver.\n", drive->name, driver->name); #ifdef CONFIG_PROC_FS - if (driver) - ide_add_proc_entries(drive->proc, driver->proc, drive); + ide_add_proc_entries(drive->proc, driver->proc, drive); #endif - return 0; } EXPORT_SYMBOL(ide_register_subdriver); @@ -2041,136 +1869,51 @@ EXPORT_SYMBOL(ide_register_subdriver); /** * ide_unregister_subdriver - disconnect drive from driver * @drive: drive to unplug + * @driver: driver * * Disconnect a drive from the driver it was attached to and then * clean up the various proc files and other objects attached to it. * - * Takes ide_setting_sem, ide_lock and drives_lock. + * Takes ide_setting_sem and ide_lock. * Caller must hold none of the locks. - * - * No locking versus subdriver unload because we are moving to the - * default driver anyway. Wants double checking. */ -int ide_unregister_subdriver (ide_drive_t *drive) +void ide_unregister_subdriver(ide_drive_t *drive, ide_driver_t *driver) { unsigned long flags; down(&ide_setting_sem); spin_lock_irqsave(&ide_lock, flags); - if (drive->usage || drive->driver == NULL || DRIVER(drive)->busy) { - spin_unlock_irqrestore(&ide_lock, flags); - up(&ide_setting_sem); - return 1; - } #ifdef CONFIG_PROC_FS - ide_remove_proc_entries(drive->proc, DRIVER(drive)->proc); + ide_remove_proc_entries(drive->proc, driver->proc); #endif auto_remove_settings(drive); - drive->driver = NULL; spin_unlock_irqrestore(&ide_lock, flags); up(&ide_setting_sem); - spin_lock(&drives_lock); - list_del_init(&drive->list); - spin_unlock(&drives_lock); - /* drive will be added to &ide_drives in ata_attach() */ - return 0; } EXPORT_SYMBOL(ide_unregister_subdriver); -static int ide_drive_remove(struct device * dev) -{ - ide_drive_t * drive = container_of(dev,ide_drive_t,gendev); - DRIVER(drive)->cleanup(drive); - return 0; -} - -/** - * ide_register_driver - register IDE device driver - * @driver: the IDE device driver - * - * Register a new device driver and then scan the devices - * on the IDE bus in case any should be attached to the - * driver we have just registered. If so attach them. - * - * Takes drivers_lock and drives_lock. - */ - -int ide_register_driver(ide_driver_t *driver) -{ - struct list_head list; - struct list_head *list_loop; - struct list_head *tmp_storage; - - spin_lock(&drivers_lock); - list_add(&driver->drivers, &drivers); - spin_unlock(&drivers_lock); - - INIT_LIST_HEAD(&list); - spin_lock(&drives_lock); - list_splice_init(&ide_drives, &list); - spin_unlock(&drives_lock); - - list_for_each_safe(list_loop, tmp_storage, &list) { - ide_drive_t *drive = container_of(list_loop, ide_drive_t, list); - list_del_init(&drive->list); - if (drive->present) - ata_attach(drive); - } - driver->gen_driver.name = (char *) driver->name; - driver->gen_driver.bus = &ide_bus_type; - driver->gen_driver.remove = ide_drive_remove; - return driver_register(&driver->gen_driver); -} - -EXPORT_SYMBOL(ide_register_driver); - -/** - * ide_unregister_driver - unregister IDE device driver - * @driver: the IDE device driver - * - * Called when a driver module is being unloaded. We reattach any - * devices to whatever driver claims them next (typically the default - * driver). - * - * Takes drivers_lock and called functions will take ide_setting_sem. - */ - -void ide_unregister_driver(ide_driver_t *driver) -{ - ide_drive_t *drive; - - spin_lock(&drivers_lock); - list_del(&driver->drivers); - spin_unlock(&drivers_lock); - - driver_unregister(&driver->gen_driver); - - while(!list_empty(&driver->drives)) { - drive = list_entry(driver->drives.next, ide_drive_t, list); - if (driver->cleanup(drive)) { - printk(KERN_ERR "%s: cleanup_module() called while still busy\n", drive->name); - BUG(); - } - ata_attach(drive); - } -} - -EXPORT_SYMBOL(ide_unregister_driver); - /* * Probe module */ EXPORT_SYMBOL(ide_lock); +static int ide_bus_match(struct device *dev, struct device_driver *drv) +{ + return 1; +} + struct bus_type ide_bus_type = { .name = "ide", + .match = ide_bus_match, .suspend = generic_ide_suspend, .resume = generic_ide_resume, }; +EXPORT_SYMBOL_GPL(ide_bus_type); + /* * This is gets invoked once during initialization, to set *everything* up */ diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c index 2e2486b035dd..d80c4c9d5a63 100644 --- a/drivers/scsi/ide-scsi.c +++ b/drivers/scsi/ide-scsi.c @@ -713,7 +713,6 @@ static void idescsi_add_settings(ide_drive_t *drive) */ static void idescsi_setup (ide_drive_t *drive, idescsi_scsi_t *scsi) { - DRIVER(drive)->busy++; if (drive->id && (drive->id->config & 0x0060) == 0x20) set_bit (IDESCSI_DRQ_INTERRUPT, &scsi->flags); set_bit(IDESCSI_TRANSFORM, &scsi->transform); @@ -722,17 +721,16 @@ static void idescsi_setup (ide_drive_t *drive, idescsi_scsi_t *scsi) set_bit(IDESCSI_LOG_CMD, &scsi->log); #endif /* IDESCSI_DEBUG_LOG */ idescsi_add_settings(drive); - DRIVER(drive)->busy--; } -static int idescsi_cleanup (ide_drive_t *drive) +static int ide_scsi_remove(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); struct Scsi_Host *scsihost = drive->driver_data; struct ide_scsi_obj *scsi = scsihost_to_idescsi(scsihost); struct gendisk *g = scsi->disk; - if (ide_unregister_subdriver(drive)) - return 1; + ide_unregister_subdriver(drive, scsi->driver); ide_unregister_region(g); @@ -746,7 +744,7 @@ static int idescsi_cleanup (ide_drive_t *drive) return 0; } -static int idescsi_attach(ide_drive_t *drive); +static int ide_scsi_probe(struct device *); #ifdef CONFIG_PROC_FS static ide_proc_entry_t idescsi_proc[] = { @@ -757,24 +755,22 @@ static ide_proc_entry_t idescsi_proc[] = { # define idescsi_proc NULL #endif -/* - * IDE subdriver functions, registered with ide.c - */ static ide_driver_t idescsi_driver = { .owner = THIS_MODULE, - .name = "ide-scsi", + .gen_driver = { + .name = "ide-scsi", + .bus = &ide_bus_type, + .probe = ide_scsi_probe, + .remove = ide_scsi_remove, + }, .version = IDESCSI_VERSION, .media = ide_scsi, - .busy = 0, .supports_dsc_overlap = 0, .proc = idescsi_proc, - .attach = idescsi_attach, - .cleanup = idescsi_cleanup, .do_request = idescsi_do_request, .end_request = idescsi_end_request, .error = idescsi_atapi_error, .abort = idescsi_atapi_abort, - .drives = LIST_HEAD_INIT(idescsi_driver.drives), }; static int idescsi_ide_open(struct inode *inode, struct file *filp) @@ -821,8 +817,6 @@ static struct block_device_operations idescsi_ops = { .ioctl = idescsi_ide_ioctl, }; -static int idescsi_attach(ide_drive_t *drive); - static int idescsi_slave_configure(struct scsi_device * sdp) { /* Configure detected device */ @@ -1095,8 +1089,9 @@ static struct scsi_host_template idescsi_template = { .proc_name = "ide-scsi", }; -static int idescsi_attach(ide_drive_t *drive) +static int ide_scsi_probe(struct device *dev) { + ide_drive_t *drive = to_ide_device(dev); idescsi_scsi_t *idescsi; struct Scsi_Host *host; struct gendisk *g; @@ -1112,7 +1107,7 @@ static int idescsi_attach(ide_drive_t *drive) !drive->present || drive->media == ide_disk || !(host = scsi_host_alloc(&idescsi_template,sizeof(idescsi_scsi_t)))) - return 1; + return -ENODEV; g = alloc_disk(1 << PARTN_BITS); if (!g) @@ -1138,20 +1133,19 @@ static int idescsi_attach(ide_drive_t *drive) idescsi->host = host; idescsi->disk = g; g->private_data = &idescsi->driver; - err = ide_register_subdriver(drive, &idescsi_driver); + ide_register_subdriver(drive, &idescsi_driver); + err = 0; + idescsi_setup(drive, idescsi); + g->fops = &idescsi_ops; + ide_register_region(g); + err = scsi_add_host(host, &drive->gendev); if (!err) { - idescsi_setup (drive, idescsi); - g->fops = &idescsi_ops; - ide_register_region(g); - err = scsi_add_host(host, &drive->gendev); - if (!err) { - scsi_scan_host(host); - return 0; - } - /* fall through on error */ - ide_unregister_region(g); - ide_unregister_subdriver(drive); + scsi_scan_host(host); + return 0; } + /* fall through on error */ + ide_unregister_region(g); + ide_unregister_subdriver(drive, &idescsi_driver); put_disk(g); out_host_put: @@ -1161,12 +1155,12 @@ out_host_put: static int __init init_idescsi_module(void) { - return ide_register_driver(&idescsi_driver); + return driver_register(&idescsi_driver.gen_driver); } static void __exit exit_idescsi_module(void) { - ide_unregister_driver(&idescsi_driver); + driver_unregister(&idescsi_driver.gen_driver); } module_init(init_idescsi_module); diff --git a/include/linux/ide.h b/include/linux/ide.h index 9cfc0999becb..336d6e509f59 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -664,7 +664,6 @@ typedef struct ide_drive_s { struct request *rq; /* current request */ struct ide_drive_s *next; /* circular list of hwgroup drives */ - struct ide_driver_s *driver;/* (ide_driver_t *) */ void *driver_data; /* extra driver data */ struct hd_driveid *id; /* drive model identification info */ struct proc_dir_entry *proc; /* /proc/ide/ directory entry */ @@ -758,6 +757,8 @@ typedef struct ide_drive_s { struct semaphore gendev_rel_sem; /* to deal with device release() */ } ide_drive_t; +#define to_ide_device(dev)container_of(dev, ide_drive_t, gendev) + #define IDE_CHIPSET_PCI_MASK \ ((1<> (c)) & 1) @@ -1086,28 +1087,20 @@ enum { */ typedef struct ide_driver_s { struct module *owner; - const char *name; const char *version; u8 media; - unsigned busy : 1; unsigned supports_dsc_overlap : 1; - int (*cleanup)(ide_drive_t *); ide_startstop_t (*do_request)(ide_drive_t *, struct request *, sector_t); int (*end_request)(ide_drive_t *, int, int); ide_startstop_t (*error)(ide_drive_t *, struct request *rq, u8, u8); ide_startstop_t (*abort)(ide_drive_t *, struct request *rq); int (*ioctl)(ide_drive_t *, struct inode *, struct file *, unsigned int, unsigned long); ide_proc_entry_t *proc; - int (*attach)(ide_drive_t *); void (*ata_prebuilder)(ide_drive_t *); void (*atapi_prebuilder)(ide_drive_t *); struct device_driver gen_driver; - struct list_head drives; - struct list_head drivers; } ide_driver_t; -#define DRIVER(drive) ((drive)->driver) - int generic_ide_ioctl(ide_drive_t *, struct file *, struct block_device *, unsigned, unsigned long); /* @@ -1328,8 +1321,6 @@ extern void ide_init_subdrivers(void); void ide_init_disk(struct gendisk *, ide_drive_t *); -extern int ata_attach(ide_drive_t *); - extern int ideprobe_init(void); extern void ide_scan_pcibus(int scan_direction) __init; @@ -1342,11 +1333,8 @@ extern void default_hwif_iops(ide_hwif_t *); extern void default_hwif_mmiops(ide_hwif_t *); extern void default_hwif_transport(ide_hwif_t *); -int ide_register_driver(ide_driver_t *driver); -void ide_unregister_driver(ide_driver_t *driver); -int ide_register_subdriver(ide_drive_t *, ide_driver_t *); -int ide_unregister_subdriver (ide_drive_t *drive); -int ide_replace_subdriver(ide_drive_t *drive, const char *driver); +void ide_register_subdriver(ide_drive_t *, ide_driver_t *); +void ide_unregister_subdriver(ide_drive_t *, ide_driver_t *); #define ON_BOARD 1 #define NEVER_BOARD 0 -- cgit v1.2.3 From 0dca51d362b8e4af6b0dbc9e54d1e5165341918a Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 May 2005 12:55:48 -0700 Subject: [PKT_SCHED] netem: allow random reordering (with fix) Here is a fixed up version of the reorder feature of netem. It is the same as the earlier patch plus with the bugfix from Julio merged in. Has expected backwards compatibility behaviour. Go ahead and merge this one, the TCP strangeness I was seeing was due to the reordering bug, and previous version of TSO patch. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/pkt_sched.h | 9 +++++++- net/sched/sch_netem.c | 54 ++++++++++++++++++++++++++++++++++++----------- 2 files changed, 50 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h index 73d84c071cb1..1d9da36eb9db 100644 --- a/include/linux/pkt_sched.h +++ b/include/linux/pkt_sched.h @@ -427,6 +427,7 @@ enum TCA_NETEM_UNSPEC, TCA_NETEM_CORR, TCA_NETEM_DELAY_DIST, + TCA_NETEM_REORDER, __TCA_NETEM_MAX, }; @@ -437,7 +438,7 @@ struct tc_netem_qopt __u32 latency; /* added delay (us) */ __u32 limit; /* fifo limit (packets) */ __u32 loss; /* random packet loss (0=none ~0=100%) */ - __u32 gap; /* re-ordering gap (0 for delay all) */ + __u32 gap; /* re-ordering gap (0 for none) */ __u32 duplicate; /* random packet dup (0=none ~0=100%) */ __u32 jitter; /* random jitter in latency (us) */ }; @@ -449,6 +450,12 @@ struct tc_netem_corr __u32 dup_corr; /* duplicate correlation */ }; +struct tc_netem_reorder +{ + __u32 probability; + __u32 correlation; +}; + #define NETEM_DIST_SCALE 8192 #endif diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 48360f7eec5d..bb9bf8d5003c 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -62,11 +62,12 @@ struct netem_sched_data { u32 gap; u32 jitter; u32 duplicate; + u32 reorder; struct crndstate { unsigned long last; unsigned long rho; - } delay_cor, loss_cor, dup_cor; + } delay_cor, loss_cor, dup_cor, reorder_cor; struct disttable { u32 size; @@ -180,23 +181,23 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) q->duplicate = dupsave; } - /* - * Do re-ordering by putting one out of N packets at the front - * of the queue. - * gap == 0 is special case for no-reordering. - */ - if (q->gap == 0 || q->counter != q->gap) { + if (q->gap == 0 /* not doing reordering */ + || q->counter < q->gap /* inside last reordering gap */ + || q->reorder < get_crandom(&q->reorder_cor)) { psched_time_t now; PSCHED_GET_TIME(now); - PSCHED_TADD2(now, - tabledist(q->latency, q->jitter, &q->delay_cor, q->delay_dist), + PSCHED_TADD2(now, tabledist(q->latency, q->jitter, + &q->delay_cor, q->delay_dist), cb->time_to_send); - ++q->counter; ret = q->qdisc->enqueue(skb, q->qdisc); } else { - q->counter = 0; + /* + * Do re-ordering by putting one out of N packets at the front + * of the queue. + */ PSCHED_GET_TIME(cb->time_to_send); + q->counter = 0; ret = q->qdisc->ops->requeue(skb, q->qdisc); } @@ -351,6 +352,19 @@ static int get_correlation(struct Qdisc *sch, const struct rtattr *attr) return 0; } +static int get_reorder(struct Qdisc *sch, const struct rtattr *attr) +{ + struct netem_sched_data *q = qdisc_priv(sch); + const struct tc_netem_reorder *r = RTA_DATA(attr); + + if (RTA_PAYLOAD(attr) != sizeof(*r)) + return -EINVAL; + + q->reorder = r->probability; + init_crandom(&q->reorder_cor, r->correlation); + return 0; +} + static int netem_change(struct Qdisc *sch, struct rtattr *opt) { struct netem_sched_data *q = qdisc_priv(sch); @@ -371,9 +385,15 @@ static int netem_change(struct Qdisc *sch, struct rtattr *opt) q->jitter = qopt->jitter; q->limit = qopt->limit; q->gap = qopt->gap; + q->counter = 0; q->loss = qopt->loss; q->duplicate = qopt->duplicate; + /* for compatiablity with earlier versions. + * if gap is set, need to assume 100% probablity + */ + q->reorder = ~0; + /* Handle nested options after initial queue options. * Should have put all options in nested format but too late now. */ @@ -395,6 +415,11 @@ static int netem_change(struct Qdisc *sch, struct rtattr *opt) if (ret) return ret; } + if (tb[TCA_NETEM_REORDER-1]) { + ret = get_reorder(sch, tb[TCA_NETEM_REORDER-1]); + if (ret) + return ret; + } } @@ -412,7 +437,6 @@ static int netem_init(struct Qdisc *sch, struct rtattr *opt) init_timer(&q->timer); q->timer.function = netem_watchdog; q->timer.data = (unsigned long) sch; - q->counter = 0; q->qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); if (!q->qdisc) { @@ -444,6 +468,7 @@ static int netem_dump(struct Qdisc *sch, struct sk_buff *skb) struct rtattr *rta = (struct rtattr *) b; struct tc_netem_qopt qopt; struct tc_netem_corr cor; + struct tc_netem_reorder reorder; qopt.latency = q->latency; qopt.jitter = q->jitter; @@ -457,6 +482,11 @@ static int netem_dump(struct Qdisc *sch, struct sk_buff *skb) cor.loss_corr = q->loss_cor.rho; cor.dup_corr = q->dup_cor.rho; RTA_PUT(skb, TCA_NETEM_CORR, sizeof(cor), &cor); + + reorder.probability = q->reorder; + reorder.correlation = q->reorder_cor.rho; + RTA_PUT(skb, TCA_NETEM_REORDER, sizeof(reorder), &reorder); + rta->rta_len = skb->tail - b; return skb->len; -- cgit v1.2.3 From c6b3365391c626206f6789354794a81a010cb7a1 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 26 May 2005 12:59:05 -0700 Subject: [TOKENRING]: be'ify trh_hdr, trllc, rif_cache_s Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/linux/if_tr.h | 6 +++--- net/802/tr.c | 4 ++-- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/if_tr.h b/include/linux/if_tr.h index a2b01e1e72f2..6688b414c529 100644 --- a/include/linux/if_tr.h +++ b/include/linux/if_tr.h @@ -36,8 +36,8 @@ struct trh_hdr { __u8 fc; /* frame control field */ __u8 daddr[TR_ALEN]; /* destination address */ __u8 saddr[TR_ALEN]; /* source address */ - __u16 rcf; /* route control field */ - __u16 rseg[8]; /* routing registers */ + __be16 rcf; /* route control field */ + __be16 rseg[8]; /* routing registers */ }; #ifdef __KERNEL__ @@ -55,7 +55,7 @@ struct trllc { __u8 ssap; /* source SAP */ __u8 llc; /* LLC control field */ __u8 protid[3]; /* protocol id */ - __u16 ethertype; /* ether type field */ + __be16 ethertype; /* ether type field */ }; /* Token-Ring statistics collection data. */ diff --git a/net/802/tr.c b/net/802/tr.c index 85293ccf7efc..3bfb86260f27 100644 --- a/net/802/tr.c +++ b/net/802/tr.c @@ -50,8 +50,8 @@ static void rif_check_expire(unsigned long dummy); struct rif_cache_s { unsigned char addr[TR_ALEN]; int iface; - __u16 rcf; - __u16 rseg[8]; + __be16 rcf; + __be16 rseg[8]; struct rif_cache_s *next; unsigned long last_used; unsigned char local_ring; -- cgit v1.2.3 From b6016b767397258b58163494a869f8f1199e6897 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 26 May 2005 13:03:09 -0700 Subject: [BNX2]: New Broadcom gigabit network driver. A new driver bnx2 for Broadcom bcm5706 is available. The patch also includes new 1000BASE-X advertisement bit definitions in mii.h Thanks to David Miller and Jeff Garzik for reviewing and their valuable feedback. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/Kconfig | 9 + drivers/net/Makefile | 1 + drivers/net/bnx2.c | 5530 +++++++++++++++++++++++++++++++++++++++++++++++ drivers/net/bnx2.h | 4352 +++++++++++++++++++++++++++++++++++++ drivers/net/bnx2_fw.h | 2468 +++++++++++++++++++++ include/linux/mii.h | 8 + include/linux/pci_ids.h | 2 + 7 files changed, 12370 insertions(+) create mode 100644 drivers/net/bnx2.c create mode 100644 drivers/net/bnx2.h create mode 100644 drivers/net/bnx2_fw.h (limited to 'include/linux') diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 3a0a55b62aaf..9e6501d98817 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -2031,6 +2031,15 @@ config TIGON3 To compile this driver as a module, choose M here: the module will be called tg3. This is recommended. +config BNX2 + tristate "Broadcom NetXtremeII support" + depends on PCI + help + This driver supports Broadcom NetXtremeII gigabit Ethernet cards. + + To compile this driver as a module, choose M here: the module + will be called bnx2. This is recommended. + config GIANFAR tristate "Gianfar Ethernet" depends on 85xx || 83xx diff --git a/drivers/net/Makefile b/drivers/net/Makefile index e038d55e4f6f..30c7567001fe 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -51,6 +51,7 @@ obj-$(CONFIG_NS83820) += ns83820.o obj-$(CONFIG_STNIC) += stnic.o 8390.o obj-$(CONFIG_FEALNX) += fealnx.o obj-$(CONFIG_TIGON3) += tg3.o +obj-$(CONFIG_BNX2) += bnx2.o obj-$(CONFIG_TC35815) += tc35815.o obj-$(CONFIG_SK98LIN) += sk98lin/ obj-$(CONFIG_SKFP) += skfp/ diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c new file mode 100644 index 000000000000..8acc655ec1e8 --- /dev/null +++ b/drivers/net/bnx2.c @@ -0,0 +1,5530 @@ +/* bnx2.c: Broadcom NX2 network driver. + * + * Copyright (c) 2004, 2005 Broadcom Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation. + * + * Written by: Michael Chan (mchan@broadcom.com) + */ + +#include "bnx2.h" +#include "bnx2_fw.h" + +#define DRV_MODULE_NAME "bnx2" +#define PFX DRV_MODULE_NAME ": " +#define DRV_MODULE_VERSION "1.2.19" +#define DRV_MODULE_RELDATE "May 23, 2005" + +#define RUN_AT(x) (jiffies + (x)) + +/* Time in jiffies before concluding the transmitter is hung. */ +#define TX_TIMEOUT (5*HZ) + +static char version[] __devinitdata = + "Broadcom NetXtreme II Gigabit Ethernet Driver " DRV_MODULE_NAME " v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n"; + +MODULE_AUTHOR("Michael Chan "); +MODULE_DESCRIPTION("Broadcom NetXtreme II BCM5706 Driver"); +MODULE_LICENSE("GPL"); +MODULE_VERSION(DRV_MODULE_VERSION); + +static int disable_msi = 0; + +module_param(disable_msi, int, 0); +MODULE_PARM_DESC(disable_msi, "Disable Message Signaled Interrupt (MSI)"); + +typedef enum { + BCM5706 = 0, + NC370T, + NC370I, + BCM5706S, + NC370F, +} board_t; + +/* indexed by board_t, above */ +static struct { + char *name; +} board_info[] __devinitdata = { + { "Broadcom NetXtreme II BCM5706 1000Base-T" }, + { "HP NC370T Multifunction Gigabit Server Adapter" }, + { "HP NC370i Multifunction Gigabit Server Adapter" }, + { "Broadcom NetXtreme II BCM5706 1000Base-SX" }, + { "HP NC370F Multifunction Gigabit Server Adapter" }, + { 0 }, + }; + +static struct pci_device_id bnx2_pci_tbl[] = { + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706, + PCI_VENDOR_ID_HP, 0x3101, 0, 0, NC370T }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706, + PCI_VENDOR_ID_HP, 0x3106, 0, 0, NC370I }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5706 }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706S, + PCI_VENDOR_ID_HP, 0x3102, 0, 0, NC370F }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706S, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5706S }, + { 0, } +}; + +static struct flash_spec flash_table[] = +{ + /* Slow EEPROM */ + {0x00000000, 0x40030380, 0x009f0081, 0xa184a053, 0xaf000400, + 1, SEEPROM_PAGE_BITS, SEEPROM_PAGE_SIZE, + SEEPROM_BYTE_ADDR_MASK, SEEPROM_TOTAL_SIZE, + "EEPROM - slow"}, + /* Fast EEPROM */ + {0x02000000, 0x62008380, 0x009f0081, 0xa184a053, 0xaf000400, + 1, SEEPROM_PAGE_BITS, SEEPROM_PAGE_SIZE, + SEEPROM_BYTE_ADDR_MASK, SEEPROM_TOTAL_SIZE, + "EEPROM - fast"}, + /* ATMEL AT45DB011B (buffered flash) */ + {0x02000003, 0x6e008173, 0x00570081, 0x68848353, 0xaf000400, + 1, BUFFERED_FLASH_PAGE_BITS, BUFFERED_FLASH_PAGE_SIZE, + BUFFERED_FLASH_BYTE_ADDR_MASK, BUFFERED_FLASH_TOTAL_SIZE, + "Buffered flash"}, + /* Saifun SA25F005 (non-buffered flash) */ + /* strap, cfg1, & write1 need updates */ + {0x01000003, 0x5f008081, 0x00050081, 0x03840253, 0xaf020406, + 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, + SAIFUN_FLASH_BYTE_ADDR_MASK, SAIFUN_FLASH_BASE_TOTAL_SIZE, + "Non-buffered flash (64kB)"}, + /* Saifun SA25F010 (non-buffered flash) */ + /* strap, cfg1, & write1 need updates */ + {0x00000001, 0x47008081, 0x00050081, 0x03840253, 0xaf020406, + 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, + SAIFUN_FLASH_BYTE_ADDR_MASK, SAIFUN_FLASH_BASE_TOTAL_SIZE*2, + "Non-buffered flash (128kB)"}, + /* Saifun SA25F020 (non-buffered flash) */ + /* strap, cfg1, & write1 need updates */ + {0x00000003, 0x4f008081, 0x00050081, 0x03840253, 0xaf020406, + 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, + SAIFUN_FLASH_BYTE_ADDR_MASK, SAIFUN_FLASH_BASE_TOTAL_SIZE*4, + "Non-buffered flash (256kB)"}, +}; + +MODULE_DEVICE_TABLE(pci, bnx2_pci_tbl); + +static u32 +bnx2_reg_rd_ind(struct bnx2 *bp, u32 offset) +{ + REG_WR(bp, BNX2_PCICFG_REG_WINDOW_ADDRESS, offset); + return (REG_RD(bp, BNX2_PCICFG_REG_WINDOW)); +} + +static void +bnx2_reg_wr_ind(struct bnx2 *bp, u32 offset, u32 val) +{ + REG_WR(bp, BNX2_PCICFG_REG_WINDOW_ADDRESS, offset); + REG_WR(bp, BNX2_PCICFG_REG_WINDOW, val); +} + +static void +bnx2_ctx_wr(struct bnx2 *bp, u32 cid_addr, u32 offset, u32 val) +{ + offset += cid_addr; + REG_WR(bp, BNX2_CTX_DATA_ADR, offset); + REG_WR(bp, BNX2_CTX_DATA, val); +} + +static int +bnx2_read_phy(struct bnx2 *bp, u32 reg, u32 *val) +{ + u32 val1; + int i, ret; + + if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) { + val1 = REG_RD(bp, BNX2_EMAC_MDIO_MODE); + val1 &= ~BNX2_EMAC_MDIO_MODE_AUTO_POLL; + + REG_WR(bp, BNX2_EMAC_MDIO_MODE, val1); + REG_RD(bp, BNX2_EMAC_MDIO_MODE); + + udelay(40); + } + + val1 = (bp->phy_addr << 21) | (reg << 16) | + BNX2_EMAC_MDIO_COMM_COMMAND_READ | BNX2_EMAC_MDIO_COMM_DISEXT | + BNX2_EMAC_MDIO_COMM_START_BUSY; + REG_WR(bp, BNX2_EMAC_MDIO_COMM, val1); + + for (i = 0; i < 50; i++) { + udelay(10); + + val1 = REG_RD(bp, BNX2_EMAC_MDIO_COMM); + if (!(val1 & BNX2_EMAC_MDIO_COMM_START_BUSY)) { + udelay(5); + + val1 = REG_RD(bp, BNX2_EMAC_MDIO_COMM); + val1 &= BNX2_EMAC_MDIO_COMM_DATA; + + break; + } + } + + if (val1 & BNX2_EMAC_MDIO_COMM_START_BUSY) { + *val = 0x0; + ret = -EBUSY; + } + else { + *val = val1; + ret = 0; + } + + if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) { + val1 = REG_RD(bp, BNX2_EMAC_MDIO_MODE); + val1 |= BNX2_EMAC_MDIO_MODE_AUTO_POLL; + + REG_WR(bp, BNX2_EMAC_MDIO_MODE, val1); + REG_RD(bp, BNX2_EMAC_MDIO_MODE); + + udelay(40); + } + + return ret; +} + +static int +bnx2_write_phy(struct bnx2 *bp, u32 reg, u32 val) +{ + u32 val1; + int i, ret; + + if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) { + val1 = REG_RD(bp, BNX2_EMAC_MDIO_MODE); + val1 &= ~BNX2_EMAC_MDIO_MODE_AUTO_POLL; + + REG_WR(bp, BNX2_EMAC_MDIO_MODE, val1); + REG_RD(bp, BNX2_EMAC_MDIO_MODE); + + udelay(40); + } + + val1 = (bp->phy_addr << 21) | (reg << 16) | val | + BNX2_EMAC_MDIO_COMM_COMMAND_WRITE | + BNX2_EMAC_MDIO_COMM_START_BUSY | BNX2_EMAC_MDIO_COMM_DISEXT; + REG_WR(bp, BNX2_EMAC_MDIO_COMM, val1); + + for (i = 0; i < 50; i++) { + udelay(10); + + val1 = REG_RD(bp, BNX2_EMAC_MDIO_COMM); + if (!(val1 & BNX2_EMAC_MDIO_COMM_START_BUSY)) { + udelay(5); + break; + } + } + + if (val1 & BNX2_EMAC_MDIO_COMM_START_BUSY) + ret = -EBUSY; + else + ret = 0; + + if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) { + val1 = REG_RD(bp, BNX2_EMAC_MDIO_MODE); + val1 |= BNX2_EMAC_MDIO_MODE_AUTO_POLL; + + REG_WR(bp, BNX2_EMAC_MDIO_MODE, val1); + REG_RD(bp, BNX2_EMAC_MDIO_MODE); + + udelay(40); + } + + return ret; +} + +static void +bnx2_disable_int(struct bnx2 *bp) +{ + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, + BNX2_PCICFG_INT_ACK_CMD_MASK_INT); + REG_RD(bp, BNX2_PCICFG_INT_ACK_CMD); +} + +static void +bnx2_enable_int(struct bnx2 *bp) +{ + u32 val; + + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, + BNX2_PCICFG_INT_ACK_CMD_INDEX_VALID | bp->last_status_idx); + + val = REG_RD(bp, BNX2_HC_COMMAND); + REG_WR(bp, BNX2_HC_COMMAND, val | BNX2_HC_COMMAND_COAL_NOW); +} + +static void +bnx2_disable_int_sync(struct bnx2 *bp) +{ + atomic_inc(&bp->intr_sem); + bnx2_disable_int(bp); + synchronize_irq(bp->pdev->irq); +} + +static void +bnx2_netif_stop(struct bnx2 *bp) +{ + bnx2_disable_int_sync(bp); + if (netif_running(bp->dev)) { + netif_poll_disable(bp->dev); + netif_tx_disable(bp->dev); + bp->dev->trans_start = jiffies; /* prevent tx timeout */ + } +} + +static void +bnx2_netif_start(struct bnx2 *bp) +{ + if (atomic_dec_and_test(&bp->intr_sem)) { + if (netif_running(bp->dev)) { + netif_wake_queue(bp->dev); + netif_poll_enable(bp->dev); + bnx2_enable_int(bp); + } + } +} + +static void +bnx2_free_mem(struct bnx2 *bp) +{ + if (bp->stats_blk) { + pci_free_consistent(bp->pdev, sizeof(struct statistics_block), + bp->stats_blk, bp->stats_blk_mapping); + bp->stats_blk = NULL; + } + if (bp->status_blk) { + pci_free_consistent(bp->pdev, sizeof(struct status_block), + bp->status_blk, bp->status_blk_mapping); + bp->status_blk = NULL; + } + if (bp->tx_desc_ring) { + pci_free_consistent(bp->pdev, + sizeof(struct tx_bd) * TX_DESC_CNT, + bp->tx_desc_ring, bp->tx_desc_mapping); + bp->tx_desc_ring = NULL; + } + if (bp->tx_buf_ring) { + kfree(bp->tx_buf_ring); + bp->tx_buf_ring = NULL; + } + if (bp->rx_desc_ring) { + pci_free_consistent(bp->pdev, + sizeof(struct rx_bd) * RX_DESC_CNT, + bp->rx_desc_ring, bp->rx_desc_mapping); + bp->rx_desc_ring = NULL; + } + if (bp->rx_buf_ring) { + kfree(bp->rx_buf_ring); + bp->rx_buf_ring = NULL; + } +} + +static int +bnx2_alloc_mem(struct bnx2 *bp) +{ + bp->tx_buf_ring = kmalloc(sizeof(struct sw_bd) * TX_DESC_CNT, + GFP_KERNEL); + if (bp->tx_buf_ring == NULL) + return -ENOMEM; + + memset(bp->tx_buf_ring, 0, sizeof(struct sw_bd) * TX_DESC_CNT); + bp->tx_desc_ring = pci_alloc_consistent(bp->pdev, + sizeof(struct tx_bd) * + TX_DESC_CNT, + &bp->tx_desc_mapping); + if (bp->tx_desc_ring == NULL) + goto alloc_mem_err; + + bp->rx_buf_ring = kmalloc(sizeof(struct sw_bd) * RX_DESC_CNT, + GFP_KERNEL); + if (bp->rx_buf_ring == NULL) + goto alloc_mem_err; + + memset(bp->rx_buf_ring, 0, sizeof(struct sw_bd) * RX_DESC_CNT); + bp->rx_desc_ring = pci_alloc_consistent(bp->pdev, + sizeof(struct rx_bd) * + RX_DESC_CNT, + &bp->rx_desc_mapping); + if (bp->rx_desc_ring == NULL) + goto alloc_mem_err; + + bp->status_blk = pci_alloc_consistent(bp->pdev, + sizeof(struct status_block), + &bp->status_blk_mapping); + if (bp->status_blk == NULL) + goto alloc_mem_err; + + memset(bp->status_blk, 0, sizeof(struct status_block)); + + bp->stats_blk = pci_alloc_consistent(bp->pdev, + sizeof(struct statistics_block), + &bp->stats_blk_mapping); + if (bp->stats_blk == NULL) + goto alloc_mem_err; + + memset(bp->stats_blk, 0, sizeof(struct statistics_block)); + + return 0; + +alloc_mem_err: + bnx2_free_mem(bp); + return -ENOMEM; +} + +static void +bnx2_report_link(struct bnx2 *bp) +{ + if (bp->link_up) { + netif_carrier_on(bp->dev); + printk(KERN_INFO PFX "%s NIC Link is Up, ", bp->dev->name); + + printk("%d Mbps ", bp->line_speed); + + if (bp->duplex == DUPLEX_FULL) + printk("full duplex"); + else + printk("half duplex"); + + if (bp->flow_ctrl) { + if (bp->flow_ctrl & FLOW_CTRL_RX) { + printk(", receive "); + if (bp->flow_ctrl & FLOW_CTRL_TX) + printk("& transmit "); + } + else { + printk(", transmit "); + } + printk("flow control ON"); + } + printk("\n"); + } + else { + netif_carrier_off(bp->dev); + printk(KERN_ERR PFX "%s NIC Link is Down\n", bp->dev->name); + } +} + +static void +bnx2_resolve_flow_ctrl(struct bnx2 *bp) +{ + u32 local_adv, remote_adv; + + bp->flow_ctrl = 0; + if ((bp->autoneg & (AUTONEG_SPEED | AUTONEG_FLOW_CTRL)) != + (AUTONEG_SPEED | AUTONEG_FLOW_CTRL)) { + + if (bp->duplex == DUPLEX_FULL) { + bp->flow_ctrl = bp->req_flow_ctrl; + } + return; + } + + if (bp->duplex != DUPLEX_FULL) { + return; + } + + bnx2_read_phy(bp, MII_ADVERTISE, &local_adv); + bnx2_read_phy(bp, MII_LPA, &remote_adv); + + if (bp->phy_flags & PHY_SERDES_FLAG) { + u32 new_local_adv = 0; + u32 new_remote_adv = 0; + + if (local_adv & ADVERTISE_1000XPAUSE) + new_local_adv |= ADVERTISE_PAUSE_CAP; + if (local_adv & ADVERTISE_1000XPSE_ASYM) + new_local_adv |= ADVERTISE_PAUSE_ASYM; + if (remote_adv & ADVERTISE_1000XPAUSE) + new_remote_adv |= ADVERTISE_PAUSE_CAP; + if (remote_adv & ADVERTISE_1000XPSE_ASYM) + new_remote_adv |= ADVERTISE_PAUSE_ASYM; + + local_adv = new_local_adv; + remote_adv = new_remote_adv; + } + + /* See Table 28B-3 of 802.3ab-1999 spec. */ + if (local_adv & ADVERTISE_PAUSE_CAP) { + if(local_adv & ADVERTISE_PAUSE_ASYM) { + if (remote_adv & ADVERTISE_PAUSE_CAP) { + bp->flow_ctrl = FLOW_CTRL_TX | FLOW_CTRL_RX; + } + else if (remote_adv & ADVERTISE_PAUSE_ASYM) { + bp->flow_ctrl = FLOW_CTRL_RX; + } + } + else { + if (remote_adv & ADVERTISE_PAUSE_CAP) { + bp->flow_ctrl = FLOW_CTRL_TX | FLOW_CTRL_RX; + } + } + } + else if (local_adv & ADVERTISE_PAUSE_ASYM) { + if ((remote_adv & ADVERTISE_PAUSE_CAP) && + (remote_adv & ADVERTISE_PAUSE_ASYM)) { + + bp->flow_ctrl = FLOW_CTRL_TX; + } + } +} + +static int +bnx2_serdes_linkup(struct bnx2 *bp) +{ + u32 bmcr, local_adv, remote_adv, common; + + bp->link_up = 1; + bp->line_speed = SPEED_1000; + + bnx2_read_phy(bp, MII_BMCR, &bmcr); + if (bmcr & BMCR_FULLDPLX) { + bp->duplex = DUPLEX_FULL; + } + else { + bp->duplex = DUPLEX_HALF; + } + + if (!(bmcr & BMCR_ANENABLE)) { + return 0; + } + + bnx2_read_phy(bp, MII_ADVERTISE, &local_adv); + bnx2_read_phy(bp, MII_LPA, &remote_adv); + + common = local_adv & remote_adv; + if (common & (ADVERTISE_1000XHALF | ADVERTISE_1000XFULL)) { + + if (common & ADVERTISE_1000XFULL) { + bp->duplex = DUPLEX_FULL; + } + else { + bp->duplex = DUPLEX_HALF; + } + } + + return 0; +} + +static int +bnx2_copper_linkup(struct bnx2 *bp) +{ + u32 bmcr; + + bnx2_read_phy(bp, MII_BMCR, &bmcr); + if (bmcr & BMCR_ANENABLE) { + u32 local_adv, remote_adv, common; + + bnx2_read_phy(bp, MII_CTRL1000, &local_adv); + bnx2_read_phy(bp, MII_STAT1000, &remote_adv); + + common = local_adv & (remote_adv >> 2); + if (common & ADVERTISE_1000FULL) { + bp->line_speed = SPEED_1000; + bp->duplex = DUPLEX_FULL; + } + else if (common & ADVERTISE_1000HALF) { + bp->line_speed = SPEED_1000; + bp->duplex = DUPLEX_HALF; + } + else { + bnx2_read_phy(bp, MII_ADVERTISE, &local_adv); + bnx2_read_phy(bp, MII_LPA, &remote_adv); + + common = local_adv & remote_adv; + if (common & ADVERTISE_100FULL) { + bp->line_speed = SPEED_100; + bp->duplex = DUPLEX_FULL; + } + else if (common & ADVERTISE_100HALF) { + bp->line_speed = SPEED_100; + bp->duplex = DUPLEX_HALF; + } + else if (common & ADVERTISE_10FULL) { + bp->line_speed = SPEED_10; + bp->duplex = DUPLEX_FULL; + } + else if (common & ADVERTISE_10HALF) { + bp->line_speed = SPEED_10; + bp->duplex = DUPLEX_HALF; + } + else { + bp->line_speed = 0; + bp->link_up = 0; + } + } + } + else { + if (bmcr & BMCR_SPEED100) { + bp->line_speed = SPEED_100; + } + else { + bp->line_speed = SPEED_10; + } + if (bmcr & BMCR_FULLDPLX) { + bp->duplex = DUPLEX_FULL; + } + else { + bp->duplex = DUPLEX_HALF; + } + } + + return 0; +} + +static int +bnx2_set_mac_link(struct bnx2 *bp) +{ + u32 val; + + REG_WR(bp, BNX2_EMAC_TX_LENGTHS, 0x2620); + if (bp->link_up && (bp->line_speed == SPEED_1000) && + (bp->duplex == DUPLEX_HALF)) { + REG_WR(bp, BNX2_EMAC_TX_LENGTHS, 0x26ff); + } + + /* Configure the EMAC mode register. */ + val = REG_RD(bp, BNX2_EMAC_MODE); + + val &= ~(BNX2_EMAC_MODE_PORT | BNX2_EMAC_MODE_HALF_DUPLEX | + BNX2_EMAC_MODE_MAC_LOOP | BNX2_EMAC_MODE_FORCE_LINK); + + if (bp->link_up) { + if (bp->line_speed != SPEED_1000) + val |= BNX2_EMAC_MODE_PORT_MII; + else + val |= BNX2_EMAC_MODE_PORT_GMII; + } + else { + val |= BNX2_EMAC_MODE_PORT_GMII; + } + + /* Set the MAC to operate in the appropriate duplex mode. */ + if (bp->duplex == DUPLEX_HALF) + val |= BNX2_EMAC_MODE_HALF_DUPLEX; + REG_WR(bp, BNX2_EMAC_MODE, val); + + /* Enable/disable rx PAUSE. */ + bp->rx_mode &= ~BNX2_EMAC_RX_MODE_FLOW_EN; + + if (bp->flow_ctrl & FLOW_CTRL_RX) + bp->rx_mode |= BNX2_EMAC_RX_MODE_FLOW_EN; + REG_WR(bp, BNX2_EMAC_RX_MODE, bp->rx_mode); + + /* Enable/disable tx PAUSE. */ + val = REG_RD(bp, BNX2_EMAC_TX_MODE); + val &= ~BNX2_EMAC_TX_MODE_FLOW_EN; + + if (bp->flow_ctrl & FLOW_CTRL_TX) + val |= BNX2_EMAC_TX_MODE_FLOW_EN; + REG_WR(bp, BNX2_EMAC_TX_MODE, val); + + /* Acknowledge the interrupt. */ + REG_WR(bp, BNX2_EMAC_STATUS, BNX2_EMAC_STATUS_LINK_CHANGE); + + return 0; +} + +static int +bnx2_set_link(struct bnx2 *bp) +{ + u32 bmsr; + u8 link_up; + + if (bp->loopback == MAC_LOOPBACK) { + bp->link_up = 1; + return 0; + } + + link_up = bp->link_up; + + bnx2_read_phy(bp, MII_BMSR, &bmsr); + bnx2_read_phy(bp, MII_BMSR, &bmsr); + + if ((bp->phy_flags & PHY_SERDES_FLAG) && + (CHIP_NUM(bp) == CHIP_NUM_5706)) { + u32 val; + + val = REG_RD(bp, BNX2_EMAC_STATUS); + if (val & BNX2_EMAC_STATUS_LINK) + bmsr |= BMSR_LSTATUS; + else + bmsr &= ~BMSR_LSTATUS; + } + + if (bmsr & BMSR_LSTATUS) { + bp->link_up = 1; + + if (bp->phy_flags & PHY_SERDES_FLAG) { + bnx2_serdes_linkup(bp); + } + else { + bnx2_copper_linkup(bp); + } + bnx2_resolve_flow_ctrl(bp); + } + else { + if ((bp->phy_flags & PHY_SERDES_FLAG) && + (bp->autoneg & AUTONEG_SPEED)) { + + u32 bmcr; + + bnx2_read_phy(bp, MII_BMCR, &bmcr); + if (!(bmcr & BMCR_ANENABLE)) { + bnx2_write_phy(bp, MII_BMCR, bmcr | + BMCR_ANENABLE); + } + } + bp->phy_flags &= ~PHY_PARALLEL_DETECT_FLAG; + bp->link_up = 0; + } + + if (bp->link_up != link_up) { + bnx2_report_link(bp); + } + + bnx2_set_mac_link(bp); + + return 0; +} + +static int +bnx2_reset_phy(struct bnx2 *bp) +{ + int i; + u32 reg; + + bnx2_write_phy(bp, MII_BMCR, BMCR_RESET); + +#define PHY_RESET_MAX_WAIT 100 + for (i = 0; i < PHY_RESET_MAX_WAIT; i++) { + udelay(10); + + bnx2_read_phy(bp, MII_BMCR, ®); + if (!(reg & BMCR_RESET)) { + udelay(20); + break; + } + } + if (i == PHY_RESET_MAX_WAIT) { + return -EBUSY; + } + return 0; +} + +static u32 +bnx2_phy_get_pause_adv(struct bnx2 *bp) +{ + u32 adv = 0; + + if ((bp->req_flow_ctrl & (FLOW_CTRL_RX | FLOW_CTRL_TX)) == + (FLOW_CTRL_RX | FLOW_CTRL_TX)) { + + if (bp->phy_flags & PHY_SERDES_FLAG) { + adv = ADVERTISE_1000XPAUSE; + } + else { + adv = ADVERTISE_PAUSE_CAP; + } + } + else if (bp->req_flow_ctrl & FLOW_CTRL_TX) { + if (bp->phy_flags & PHY_SERDES_FLAG) { + adv = ADVERTISE_1000XPSE_ASYM; + } + else { + adv = ADVERTISE_PAUSE_ASYM; + } + } + else if (bp->req_flow_ctrl & FLOW_CTRL_RX) { + if (bp->phy_flags & PHY_SERDES_FLAG) { + adv = ADVERTISE_1000XPAUSE | ADVERTISE_1000XPSE_ASYM; + } + else { + adv = ADVERTISE_PAUSE_CAP | ADVERTISE_PAUSE_ASYM; + } + } + return adv; +} + +static int +bnx2_setup_serdes_phy(struct bnx2 *bp) +{ + u32 adv, bmcr; + u32 new_adv = 0; + + if (!(bp->autoneg & AUTONEG_SPEED)) { + u32 new_bmcr; + + bnx2_read_phy(bp, MII_BMCR, &bmcr); + new_bmcr = bmcr & ~BMCR_ANENABLE; + new_bmcr |= BMCR_SPEED1000; + if (bp->req_duplex == DUPLEX_FULL) { + new_bmcr |= BMCR_FULLDPLX; + } + else { + new_bmcr &= ~BMCR_FULLDPLX; + } + if (new_bmcr != bmcr) { + /* Force a link down visible on the other side */ + if (bp->link_up) { + bnx2_read_phy(bp, MII_ADVERTISE, &adv); + adv &= ~(ADVERTISE_1000XFULL | + ADVERTISE_1000XHALF); + bnx2_write_phy(bp, MII_ADVERTISE, adv); + bnx2_write_phy(bp, MII_BMCR, bmcr | + BMCR_ANRESTART | BMCR_ANENABLE); + + bp->link_up = 0; + netif_carrier_off(bp->dev); + } + bnx2_write_phy(bp, MII_BMCR, new_bmcr); + } + return 0; + } + + if (bp->advertising & ADVERTISED_1000baseT_Full) + new_adv |= ADVERTISE_1000XFULL; + + new_adv |= bnx2_phy_get_pause_adv(bp); + + bnx2_read_phy(bp, MII_ADVERTISE, &adv); + bnx2_read_phy(bp, MII_BMCR, &bmcr); + + bp->serdes_an_pending = 0; + if ((adv != new_adv) || ((bmcr & BMCR_ANENABLE) == 0)) { + /* Force a link down visible on the other side */ + if (bp->link_up) { + int i; + + bnx2_write_phy(bp, MII_BMCR, BMCR_LOOPBACK); + for (i = 0; i < 110; i++) { + udelay(100); + } + } + + bnx2_write_phy(bp, MII_ADVERTISE, new_adv); + bnx2_write_phy(bp, MII_BMCR, bmcr | BMCR_ANRESTART | + BMCR_ANENABLE); + bp->serdes_an_pending = SERDES_AN_TIMEOUT / bp->timer_interval; + } + + return 0; +} + +#define ETHTOOL_ALL_FIBRE_SPEED \ + (ADVERTISED_1000baseT_Full) + +#define ETHTOOL_ALL_COPPER_SPEED \ + (ADVERTISED_10baseT_Half | ADVERTISED_10baseT_Full | \ + ADVERTISED_100baseT_Half | ADVERTISED_100baseT_Full | \ + ADVERTISED_1000baseT_Full) + +#define PHY_ALL_10_100_SPEED (ADVERTISE_10HALF | ADVERTISE_10FULL | \ + ADVERTISE_100HALF | ADVERTISE_100FULL | ADVERTISE_CSMA) + +#define PHY_ALL_1000_SPEED (ADVERTISE_1000HALF | ADVERTISE_1000FULL) + +static int +bnx2_setup_copper_phy(struct bnx2 *bp) +{ + u32 bmcr; + u32 new_bmcr; + + bnx2_read_phy(bp, MII_BMCR, &bmcr); + + if (bp->autoneg & AUTONEG_SPEED) { + u32 adv_reg, adv1000_reg; + u32 new_adv_reg = 0; + u32 new_adv1000_reg = 0; + + bnx2_read_phy(bp, MII_ADVERTISE, &adv_reg); + adv_reg &= (PHY_ALL_10_100_SPEED | ADVERTISE_PAUSE_CAP | + ADVERTISE_PAUSE_ASYM); + + bnx2_read_phy(bp, MII_CTRL1000, &adv1000_reg); + adv1000_reg &= PHY_ALL_1000_SPEED; + + if (bp->advertising & ADVERTISED_10baseT_Half) + new_adv_reg |= ADVERTISE_10HALF; + if (bp->advertising & ADVERTISED_10baseT_Full) + new_adv_reg |= ADVERTISE_10FULL; + if (bp->advertising & ADVERTISED_100baseT_Half) + new_adv_reg |= ADVERTISE_100HALF; + if (bp->advertising & ADVERTISED_100baseT_Full) + new_adv_reg |= ADVERTISE_100FULL; + if (bp->advertising & ADVERTISED_1000baseT_Full) + new_adv1000_reg |= ADVERTISE_1000FULL; + + new_adv_reg |= ADVERTISE_CSMA; + + new_adv_reg |= bnx2_phy_get_pause_adv(bp); + + if ((adv1000_reg != new_adv1000_reg) || + (adv_reg != new_adv_reg) || + ((bmcr & BMCR_ANENABLE) == 0)) { + + bnx2_write_phy(bp, MII_ADVERTISE, new_adv_reg); + bnx2_write_phy(bp, MII_CTRL1000, new_adv1000_reg); + bnx2_write_phy(bp, MII_BMCR, BMCR_ANRESTART | + BMCR_ANENABLE); + } + else if (bp->link_up) { + /* Flow ctrl may have changed from auto to forced */ + /* or vice-versa. */ + + bnx2_resolve_flow_ctrl(bp); + bnx2_set_mac_link(bp); + } + return 0; + } + + new_bmcr = 0; + if (bp->req_line_speed == SPEED_100) { + new_bmcr |= BMCR_SPEED100; + } + if (bp->req_duplex == DUPLEX_FULL) { + new_bmcr |= BMCR_FULLDPLX; + } + if (new_bmcr != bmcr) { + u32 bmsr; + int i = 0; + + bnx2_read_phy(bp, MII_BMSR, &bmsr); + bnx2_read_phy(bp, MII_BMSR, &bmsr); + + if (bmsr & BMSR_LSTATUS) { + /* Force link down */ + bnx2_write_phy(bp, MII_BMCR, BMCR_LOOPBACK); + do { + udelay(100); + bnx2_read_phy(bp, MII_BMSR, &bmsr); + bnx2_read_phy(bp, MII_BMSR, &bmsr); + i++; + } while ((bmsr & BMSR_LSTATUS) && (i < 620)); + } + + bnx2_write_phy(bp, MII_BMCR, new_bmcr); + + /* Normally, the new speed is setup after the link has + * gone down and up again. In some cases, link will not go + * down so we need to set up the new speed here. + */ + if (bmsr & BMSR_LSTATUS) { + bp->line_speed = bp->req_line_speed; + bp->duplex = bp->req_duplex; + bnx2_resolve_flow_ctrl(bp); + bnx2_set_mac_link(bp); + } + } + return 0; +} + +static int +bnx2_setup_phy(struct bnx2 *bp) +{ + if (bp->loopback == MAC_LOOPBACK) + return 0; + + if (bp->phy_flags & PHY_SERDES_FLAG) { + return (bnx2_setup_serdes_phy(bp)); + } + else { + return (bnx2_setup_copper_phy(bp)); + } +} + +static int +bnx2_init_serdes_phy(struct bnx2 *bp) +{ + bp->phy_flags &= ~PHY_PARALLEL_DETECT_FLAG; + + if (CHIP_NUM(bp) == CHIP_NUM_5706) { + REG_WR(bp, BNX2_MISC_UNUSED0, 0x300); + } + + if (bp->dev->mtu > 1500) { + u32 val; + + /* Set extended packet length bit */ + bnx2_write_phy(bp, 0x18, 0x7); + bnx2_read_phy(bp, 0x18, &val); + bnx2_write_phy(bp, 0x18, (val & 0xfff8) | 0x4000); + + bnx2_write_phy(bp, 0x1c, 0x6c00); + bnx2_read_phy(bp, 0x1c, &val); + bnx2_write_phy(bp, 0x1c, (val & 0x3ff) | 0xec02); + } + else { + u32 val; + + bnx2_write_phy(bp, 0x18, 0x7); + bnx2_read_phy(bp, 0x18, &val); + bnx2_write_phy(bp, 0x18, val & ~0x4007); + + bnx2_write_phy(bp, 0x1c, 0x6c00); + bnx2_read_phy(bp, 0x1c, &val); + bnx2_write_phy(bp, 0x1c, (val & 0x3fd) | 0xec00); + } + + return 0; +} + +static int +bnx2_init_copper_phy(struct bnx2 *bp) +{ + bp->phy_flags |= PHY_CRC_FIX_FLAG; + + if (bp->phy_flags & PHY_CRC_FIX_FLAG) { + bnx2_write_phy(bp, 0x18, 0x0c00); + bnx2_write_phy(bp, 0x17, 0x000a); + bnx2_write_phy(bp, 0x15, 0x310b); + bnx2_write_phy(bp, 0x17, 0x201f); + bnx2_write_phy(bp, 0x15, 0x9506); + bnx2_write_phy(bp, 0x17, 0x401f); + bnx2_write_phy(bp, 0x15, 0x14e2); + bnx2_write_phy(bp, 0x18, 0x0400); + } + + if (bp->dev->mtu > 1500) { + u32 val; + + /* Set extended packet length bit */ + bnx2_write_phy(bp, 0x18, 0x7); + bnx2_read_phy(bp, 0x18, &val); + bnx2_write_phy(bp, 0x18, val | 0x4000); + + bnx2_read_phy(bp, 0x10, &val); + bnx2_write_phy(bp, 0x10, val | 0x1); + } + else { + u32 val; + + bnx2_write_phy(bp, 0x18, 0x7); + bnx2_read_phy(bp, 0x18, &val); + bnx2_write_phy(bp, 0x18, val & ~0x4007); + + bnx2_read_phy(bp, 0x10, &val); + bnx2_write_phy(bp, 0x10, val & ~0x1); + } + + return 0; +} + + +static int +bnx2_init_phy(struct bnx2 *bp) +{ + u32 val; + int rc = 0; + + bp->phy_flags &= ~PHY_INT_MODE_MASK_FLAG; + bp->phy_flags |= PHY_INT_MODE_LINK_READY_FLAG; + + REG_WR(bp, BNX2_EMAC_ATTENTION_ENA, BNX2_EMAC_ATTENTION_ENA_LINK); + + bnx2_reset_phy(bp); + + bnx2_read_phy(bp, MII_PHYSID1, &val); + bp->phy_id = val << 16; + bnx2_read_phy(bp, MII_PHYSID2, &val); + bp->phy_id |= val & 0xffff; + + if (bp->phy_flags & PHY_SERDES_FLAG) { + rc = bnx2_init_serdes_phy(bp); + } + else { + rc = bnx2_init_copper_phy(bp); + } + + bnx2_setup_phy(bp); + + return rc; +} + +static int +bnx2_set_mac_loopback(struct bnx2 *bp) +{ + u32 mac_mode; + + mac_mode = REG_RD(bp, BNX2_EMAC_MODE); + mac_mode &= ~BNX2_EMAC_MODE_PORT; + mac_mode |= BNX2_EMAC_MODE_MAC_LOOP | BNX2_EMAC_MODE_FORCE_LINK; + REG_WR(bp, BNX2_EMAC_MODE, mac_mode); + bp->link_up = 1; + return 0; +} + +static int +bnx2_fw_sync(struct bnx2 *bp, u32 msg_data) +{ + int i; + u32 val; + + if (bp->fw_timed_out) + return -EBUSY; + + bp->fw_wr_seq++; + msg_data |= bp->fw_wr_seq; + + REG_WR_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DRV_MB, msg_data); + + /* wait for an acknowledgement. */ + for (i = 0; i < (FW_ACK_TIME_OUT_MS * 1000)/5; i++) { + udelay(5); + + val = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_FW_MB); + + if ((val & BNX2_FW_MSG_ACK) == (msg_data & BNX2_DRV_MSG_SEQ)) + break; + } + + /* If we timed out, inform the firmware that this is the case. */ + if (((val & BNX2_FW_MSG_ACK) != (msg_data & BNX2_DRV_MSG_SEQ)) && + ((msg_data & BNX2_DRV_MSG_DATA) != BNX2_DRV_MSG_DATA_WAIT0)) { + + msg_data &= ~BNX2_DRV_MSG_CODE; + msg_data |= BNX2_DRV_MSG_CODE_FW_TIMEOUT; + + REG_WR_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DRV_MB, msg_data); + + bp->fw_timed_out = 1; + + return -EBUSY; + } + + return 0; +} + +static void +bnx2_init_context(struct bnx2 *bp) +{ + u32 vcid; + + vcid = 96; + while (vcid) { + u32 vcid_addr, pcid_addr, offset; + + vcid--; + + if (CHIP_ID(bp) == CHIP_ID_5706_A0) { + u32 new_vcid; + + vcid_addr = GET_PCID_ADDR(vcid); + if (vcid & 0x8) { + new_vcid = 0x60 + (vcid & 0xf0) + (vcid & 0x7); + } + else { + new_vcid = vcid; + } + pcid_addr = GET_PCID_ADDR(new_vcid); + } + else { + vcid_addr = GET_CID_ADDR(vcid); + pcid_addr = vcid_addr; + } + + REG_WR(bp, BNX2_CTX_VIRT_ADDR, 0x00); + REG_WR(bp, BNX2_CTX_PAGE_TBL, pcid_addr); + + /* Zero out the context. */ + for (offset = 0; offset < PHY_CTX_SIZE; offset += 4) { + CTX_WR(bp, 0x00, offset, 0); + } + + REG_WR(bp, BNX2_CTX_VIRT_ADDR, vcid_addr); + REG_WR(bp, BNX2_CTX_PAGE_TBL, pcid_addr); + } +} + +static int +bnx2_alloc_bad_rbuf(struct bnx2 *bp) +{ + u16 *good_mbuf; + u32 good_mbuf_cnt; + u32 val; + + good_mbuf = kmalloc(512 * sizeof(u16), GFP_KERNEL); + if (good_mbuf == NULL) { + printk(KERN_ERR PFX "Failed to allocate memory in " + "bnx2_alloc_bad_rbuf\n"); + return -ENOMEM; + } + + REG_WR(bp, BNX2_MISC_ENABLE_SET_BITS, + BNX2_MISC_ENABLE_SET_BITS_RX_MBUF_ENABLE); + + good_mbuf_cnt = 0; + + /* Allocate a bunch of mbufs and save the good ones in an array. */ + val = REG_RD_IND(bp, BNX2_RBUF_STATUS1); + while (val & BNX2_RBUF_STATUS1_FREE_COUNT) { + REG_WR_IND(bp, BNX2_RBUF_COMMAND, BNX2_RBUF_COMMAND_ALLOC_REQ); + + val = REG_RD_IND(bp, BNX2_RBUF_FW_BUF_ALLOC); + + val &= BNX2_RBUF_FW_BUF_ALLOC_VALUE; + + /* The addresses with Bit 9 set are bad memory blocks. */ + if (!(val & (1 << 9))) { + good_mbuf[good_mbuf_cnt] = (u16) val; + good_mbuf_cnt++; + } + + val = REG_RD_IND(bp, BNX2_RBUF_STATUS1); + } + + /* Free the good ones back to the mbuf pool thus discarding + * all the bad ones. */ + while (good_mbuf_cnt) { + good_mbuf_cnt--; + + val = good_mbuf[good_mbuf_cnt]; + val = (val << 9) | val | 1; + + REG_WR_IND(bp, BNX2_RBUF_FW_BUF_FREE, val); + } + kfree(good_mbuf); + return 0; +} + +static void +bnx2_set_mac_addr(struct bnx2 *bp) +{ + u32 val; + u8 *mac_addr = bp->dev->dev_addr; + + val = (mac_addr[0] << 8) | mac_addr[1]; + + REG_WR(bp, BNX2_EMAC_MAC_MATCH0, val); + + val = (mac_addr[2] << 24) | (mac_addr[3] << 16) | + (mac_addr[4] << 8) | mac_addr[5]; + + REG_WR(bp, BNX2_EMAC_MAC_MATCH1, val); +} + +static inline int +bnx2_alloc_rx_skb(struct bnx2 *bp, u16 index) +{ + struct sk_buff *skb; + struct sw_bd *rx_buf = &bp->rx_buf_ring[index]; + dma_addr_t mapping; + struct rx_bd *rxbd = &bp->rx_desc_ring[index]; + unsigned long align; + + skb = dev_alloc_skb(bp->rx_buf_size); + if (skb == NULL) { + return -ENOMEM; + } + + if (unlikely((align = (unsigned long) skb->data & 0x7))) { + skb_reserve(skb, 8 - align); + } + + skb->dev = bp->dev; + mapping = pci_map_single(bp->pdev, skb->data, bp->rx_buf_use_size, + PCI_DMA_FROMDEVICE); + + rx_buf->skb = skb; + pci_unmap_addr_set(rx_buf, mapping, mapping); + + rxbd->rx_bd_haddr_hi = (u64) mapping >> 32; + rxbd->rx_bd_haddr_lo = (u64) mapping & 0xffffffff; + + bp->rx_prod_bseq += bp->rx_buf_use_size; + + return 0; +} + +static void +bnx2_phy_int(struct bnx2 *bp) +{ + u32 new_link_state, old_link_state; + + new_link_state = bp->status_blk->status_attn_bits & + STATUS_ATTN_BITS_LINK_STATE; + old_link_state = bp->status_blk->status_attn_bits_ack & + STATUS_ATTN_BITS_LINK_STATE; + if (new_link_state != old_link_state) { + if (new_link_state) { + REG_WR(bp, BNX2_PCICFG_STATUS_BIT_SET_CMD, + STATUS_ATTN_BITS_LINK_STATE); + } + else { + REG_WR(bp, BNX2_PCICFG_STATUS_BIT_CLEAR_CMD, + STATUS_ATTN_BITS_LINK_STATE); + } + bnx2_set_link(bp); + } +} + +static void +bnx2_tx_int(struct bnx2 *bp) +{ + u16 hw_cons, sw_cons, sw_ring_cons; + int tx_free_bd = 0; + + hw_cons = bp->status_blk->status_tx_quick_consumer_index0; + if ((hw_cons & MAX_TX_DESC_CNT) == MAX_TX_DESC_CNT) { + hw_cons++; + } + sw_cons = bp->tx_cons; + + while (sw_cons != hw_cons) { + struct sw_bd *tx_buf; + struct sk_buff *skb; + int i, last; + + sw_ring_cons = TX_RING_IDX(sw_cons); + + tx_buf = &bp->tx_buf_ring[sw_ring_cons]; + skb = tx_buf->skb; +#ifdef BCM_TSO + /* partial BD completions possible with TSO packets */ + if (skb_shinfo(skb)->tso_size) { + u16 last_idx, last_ring_idx; + + last_idx = sw_cons + + skb_shinfo(skb)->nr_frags + 1; + last_ring_idx = sw_ring_cons + + skb_shinfo(skb)->nr_frags + 1; + if (unlikely(last_ring_idx >= MAX_TX_DESC_CNT)) { + last_idx++; + } + if (((s16) ((s16) last_idx - (s16) hw_cons)) > 0) { + break; + } + } +#endif + pci_unmap_single(bp->pdev, pci_unmap_addr(tx_buf, mapping), + skb_headlen(skb), PCI_DMA_TODEVICE); + + tx_buf->skb = NULL; + last = skb_shinfo(skb)->nr_frags; + + for (i = 0; i < last; i++) { + sw_cons = NEXT_TX_BD(sw_cons); + + pci_unmap_page(bp->pdev, + pci_unmap_addr( + &bp->tx_buf_ring[TX_RING_IDX(sw_cons)], + mapping), + skb_shinfo(skb)->frags[i].size, + PCI_DMA_TODEVICE); + } + + sw_cons = NEXT_TX_BD(sw_cons); + + tx_free_bd += last + 1; + + dev_kfree_skb_irq(skb); + + hw_cons = bp->status_blk->status_tx_quick_consumer_index0; + if ((hw_cons & MAX_TX_DESC_CNT) == MAX_TX_DESC_CNT) { + hw_cons++; + } + } + + atomic_add(tx_free_bd, &bp->tx_avail_bd); + + if (unlikely(netif_queue_stopped(bp->dev))) { + unsigned long flags; + + spin_lock_irqsave(&bp->tx_lock, flags); + if ((netif_queue_stopped(bp->dev)) && + (atomic_read(&bp->tx_avail_bd) > MAX_SKB_FRAGS)) { + + netif_wake_queue(bp->dev); + } + spin_unlock_irqrestore(&bp->tx_lock, flags); + } + + bp->tx_cons = sw_cons; + +} + +static inline void +bnx2_reuse_rx_skb(struct bnx2 *bp, struct sk_buff *skb, + u16 cons, u16 prod) +{ + struct sw_bd *cons_rx_buf = &bp->rx_buf_ring[cons]; + struct sw_bd *prod_rx_buf = &bp->rx_buf_ring[prod]; + struct rx_bd *cons_bd = &bp->rx_desc_ring[cons]; + struct rx_bd *prod_bd = &bp->rx_desc_ring[prod]; + + pci_dma_sync_single_for_device(bp->pdev, + pci_unmap_addr(cons_rx_buf, mapping), + bp->rx_offset + RX_COPY_THRESH, PCI_DMA_FROMDEVICE); + + prod_rx_buf->skb = cons_rx_buf->skb; + pci_unmap_addr_set(prod_rx_buf, mapping, + pci_unmap_addr(cons_rx_buf, mapping)); + + memcpy(prod_bd, cons_bd, 8); + + bp->rx_prod_bseq += bp->rx_buf_use_size; + +} + +static int +bnx2_rx_int(struct bnx2 *bp, int budget) +{ + u16 hw_cons, sw_cons, sw_ring_cons, sw_prod, sw_ring_prod; + struct l2_fhdr *rx_hdr; + int rx_pkt = 0; + + hw_cons = bp->status_blk->status_rx_quick_consumer_index0; + if ((hw_cons & MAX_RX_DESC_CNT) == MAX_RX_DESC_CNT) { + hw_cons++; + } + sw_cons = bp->rx_cons; + sw_prod = bp->rx_prod; + + /* Memory barrier necessary as speculative reads of the rx + * buffer can be ahead of the index in the status block + */ + rmb(); + while (sw_cons != hw_cons) { + unsigned int len; + u16 status; + struct sw_bd *rx_buf; + struct sk_buff *skb; + + sw_ring_cons = RX_RING_IDX(sw_cons); + sw_ring_prod = RX_RING_IDX(sw_prod); + + rx_buf = &bp->rx_buf_ring[sw_ring_cons]; + skb = rx_buf->skb; + pci_dma_sync_single_for_cpu(bp->pdev, + pci_unmap_addr(rx_buf, mapping), + bp->rx_offset + RX_COPY_THRESH, PCI_DMA_FROMDEVICE); + + rx_hdr = (struct l2_fhdr *) skb->data; + len = rx_hdr->l2_fhdr_pkt_len - 4; + + if (rx_hdr->l2_fhdr_errors & + (L2_FHDR_ERRORS_BAD_CRC | + L2_FHDR_ERRORS_PHY_DECODE | + L2_FHDR_ERRORS_ALIGNMENT | + L2_FHDR_ERRORS_TOO_SHORT | + L2_FHDR_ERRORS_GIANT_FRAME)) { + + goto reuse_rx; + } + + /* Since we don't have a jumbo ring, copy small packets + * if mtu > 1500 + */ + if ((bp->dev->mtu > 1500) && (len <= RX_COPY_THRESH)) { + struct sk_buff *new_skb; + + new_skb = dev_alloc_skb(len + 2); + if (new_skb == NULL) + goto reuse_rx; + + /* aligned copy */ + memcpy(new_skb->data, + skb->data + bp->rx_offset - 2, + len + 2); + + skb_reserve(new_skb, 2); + skb_put(new_skb, len); + new_skb->dev = bp->dev; + + bnx2_reuse_rx_skb(bp, skb, + sw_ring_cons, sw_ring_prod); + + skb = new_skb; + } + else if (bnx2_alloc_rx_skb(bp, sw_ring_prod) == 0) { + pci_unmap_single(bp->pdev, + pci_unmap_addr(rx_buf, mapping), + bp->rx_buf_use_size, PCI_DMA_FROMDEVICE); + + skb_reserve(skb, bp->rx_offset); + skb_put(skb, len); + } + else { +reuse_rx: + bnx2_reuse_rx_skb(bp, skb, + sw_ring_cons, sw_ring_prod); + goto next_rx; + } + + skb->protocol = eth_type_trans(skb, bp->dev); + + if ((len > (bp->dev->mtu + ETH_HLEN)) && + (htons(skb->protocol) != 0x8100)) { + + dev_kfree_skb_irq(skb); + goto next_rx; + + } + + status = rx_hdr->l2_fhdr_status; + skb->ip_summed = CHECKSUM_NONE; + if (bp->rx_csum && + (status & (L2_FHDR_STATUS_TCP_SEGMENT | + L2_FHDR_STATUS_UDP_DATAGRAM))) { + + u16 cksum = rx_hdr->l2_fhdr_tcp_udp_xsum; + + if (cksum == 0xffff) + skb->ip_summed = CHECKSUM_UNNECESSARY; + } + +#ifdef BCM_VLAN + if ((status & L2_FHDR_STATUS_L2_VLAN_TAG) && (bp->vlgrp != 0)) { + vlan_hwaccel_receive_skb(skb, bp->vlgrp, + rx_hdr->l2_fhdr_vlan_tag); + } + else +#endif + netif_receive_skb(skb); + + bp->dev->last_rx = jiffies; + rx_pkt++; + +next_rx: + rx_buf->skb = NULL; + + sw_cons = NEXT_RX_BD(sw_cons); + sw_prod = NEXT_RX_BD(sw_prod); + + if ((rx_pkt == budget)) + break; + } + bp->rx_cons = sw_cons; + bp->rx_prod = sw_prod; + + REG_WR16(bp, MB_RX_CID_ADDR + BNX2_L2CTX_HOST_BDIDX, sw_prod); + + REG_WR(bp, MB_RX_CID_ADDR + BNX2_L2CTX_HOST_BSEQ, bp->rx_prod_bseq); + + mmiowb(); + + return rx_pkt; + +} + +/* MSI ISR - The only difference between this and the INTx ISR + * is that the MSI interrupt is always serviced. + */ +static irqreturn_t +bnx2_msi(int irq, void *dev_instance, struct pt_regs *regs) +{ + struct net_device *dev = dev_instance; + struct bnx2 *bp = dev->priv; + + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, + BNX2_PCICFG_INT_ACK_CMD_USE_INT_HC_PARAM | + BNX2_PCICFG_INT_ACK_CMD_MASK_INT); + + /* Return here if interrupt is disabled. */ + if (unlikely(atomic_read(&bp->intr_sem) != 0)) { + return IRQ_RETVAL(1); + } + + if (netif_rx_schedule_prep(dev)) { + __netif_rx_schedule(dev); + } + + return IRQ_RETVAL(1); +} + +static irqreturn_t +bnx2_interrupt(int irq, void *dev_instance, struct pt_regs *regs) +{ + struct net_device *dev = dev_instance; + struct bnx2 *bp = dev->priv; + + /* When using INTx, it is possible for the interrupt to arrive + * at the CPU before the status block posted prior to the + * interrupt. Reading a register will flush the status block. + * When using MSI, the MSI message will always complete after + * the status block write. + */ + if ((bp->status_blk->status_idx == bp->last_status_idx) || + (REG_RD(bp, BNX2_PCICFG_MISC_STATUS) & + BNX2_PCICFG_MISC_STATUS_INTA_VALUE)) + return IRQ_RETVAL(0); + + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, + BNX2_PCICFG_INT_ACK_CMD_USE_INT_HC_PARAM | + BNX2_PCICFG_INT_ACK_CMD_MASK_INT); + + /* Return here if interrupt is shared and is disabled. */ + if (unlikely(atomic_read(&bp->intr_sem) != 0)) { + return IRQ_RETVAL(1); + } + + if (netif_rx_schedule_prep(dev)) { + __netif_rx_schedule(dev); + } + + return IRQ_RETVAL(1); +} + +static int +bnx2_poll(struct net_device *dev, int *budget) +{ + struct bnx2 *bp = dev->priv; + int rx_done = 1; + + bp->last_status_idx = bp->status_blk->status_idx; + + rmb(); + if ((bp->status_blk->status_attn_bits & + STATUS_ATTN_BITS_LINK_STATE) != + (bp->status_blk->status_attn_bits_ack & + STATUS_ATTN_BITS_LINK_STATE)) { + + unsigned long flags; + + spin_lock_irqsave(&bp->phy_lock, flags); + bnx2_phy_int(bp); + spin_unlock_irqrestore(&bp->phy_lock, flags); + } + + if (bp->status_blk->status_tx_quick_consumer_index0 != bp->tx_cons) { + bnx2_tx_int(bp); + } + + if (bp->status_blk->status_rx_quick_consumer_index0 != bp->rx_cons) { + int orig_budget = *budget; + int work_done; + + if (orig_budget > dev->quota) + orig_budget = dev->quota; + + work_done = bnx2_rx_int(bp, orig_budget); + *budget -= work_done; + dev->quota -= work_done; + + if (work_done >= orig_budget) { + rx_done = 0; + } + } + + if (rx_done) { + netif_rx_complete(dev); + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, + BNX2_PCICFG_INT_ACK_CMD_INDEX_VALID | + bp->last_status_idx); + return 0; + } + + return 1; +} + +/* Called with rtnl_lock from vlan functions and also dev->xmit_lock + * from set_multicast. + */ +static void +bnx2_set_rx_mode(struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + u32 rx_mode, sort_mode; + int i; + unsigned long flags; + + spin_lock_irqsave(&bp->phy_lock, flags); + + rx_mode = bp->rx_mode & ~(BNX2_EMAC_RX_MODE_PROMISCUOUS | + BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG); + sort_mode = 1 | BNX2_RPM_SORT_USER0_BC_EN; +#ifdef BCM_VLAN + if (!bp->vlgrp) { + rx_mode |= BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG; + } +#else + rx_mode |= BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG; +#endif + if (dev->flags & IFF_PROMISC) { + /* Promiscuous mode. */ + rx_mode |= BNX2_EMAC_RX_MODE_PROMISCUOUS; + sort_mode |= BNX2_RPM_SORT_USER0_PROM_EN; + } + else if (dev->flags & IFF_ALLMULTI) { + for (i = 0; i < NUM_MC_HASH_REGISTERS; i++) { + REG_WR(bp, BNX2_EMAC_MULTICAST_HASH0 + (i * 4), + 0xffffffff); + } + sort_mode |= BNX2_RPM_SORT_USER0_MC_EN; + } + else { + /* Accept one or more multicast(s). */ + struct dev_mc_list *mclist; + u32 mc_filter[NUM_MC_HASH_REGISTERS]; + u32 regidx; + u32 bit; + u32 crc; + + memset(mc_filter, 0, 4 * NUM_MC_HASH_REGISTERS); + + for (i = 0, mclist = dev->mc_list; mclist && i < dev->mc_count; + i++, mclist = mclist->next) { + + crc = ether_crc_le(ETH_ALEN, mclist->dmi_addr); + bit = crc & 0xff; + regidx = (bit & 0xe0) >> 5; + bit &= 0x1f; + mc_filter[regidx] |= (1 << bit); + } + + for (i = 0; i < NUM_MC_HASH_REGISTERS; i++) { + REG_WR(bp, BNX2_EMAC_MULTICAST_HASH0 + (i * 4), + mc_filter[i]); + } + + sort_mode |= BNX2_RPM_SORT_USER0_MC_HSH_EN; + } + + if (rx_mode != bp->rx_mode) { + bp->rx_mode = rx_mode; + REG_WR(bp, BNX2_EMAC_RX_MODE, rx_mode); + } + + REG_WR(bp, BNX2_RPM_SORT_USER0, 0x0); + REG_WR(bp, BNX2_RPM_SORT_USER0, sort_mode); + REG_WR(bp, BNX2_RPM_SORT_USER0, sort_mode | BNX2_RPM_SORT_USER0_ENA); + + spin_unlock_irqrestore(&bp->phy_lock, flags); +} + +static void +load_rv2p_fw(struct bnx2 *bp, u32 *rv2p_code, u32 rv2p_code_len, + u32 rv2p_proc) +{ + int i; + u32 val; + + + for (i = 0; i < rv2p_code_len; i += 8) { + REG_WR(bp, BNX2_RV2P_INSTR_HIGH, *rv2p_code); + rv2p_code++; + REG_WR(bp, BNX2_RV2P_INSTR_LOW, *rv2p_code); + rv2p_code++; + + if (rv2p_proc == RV2P_PROC1) { + val = (i / 8) | BNX2_RV2P_PROC1_ADDR_CMD_RDWR; + REG_WR(bp, BNX2_RV2P_PROC1_ADDR_CMD, val); + } + else { + val = (i / 8) | BNX2_RV2P_PROC2_ADDR_CMD_RDWR; + REG_WR(bp, BNX2_RV2P_PROC2_ADDR_CMD, val); + } + } + + /* Reset the processor, un-stall is done later. */ + if (rv2p_proc == RV2P_PROC1) { + REG_WR(bp, BNX2_RV2P_COMMAND, BNX2_RV2P_COMMAND_PROC1_RESET); + } + else { + REG_WR(bp, BNX2_RV2P_COMMAND, BNX2_RV2P_COMMAND_PROC2_RESET); + } +} + +static void +load_cpu_fw(struct bnx2 *bp, struct cpu_reg *cpu_reg, struct fw_info *fw) +{ + u32 offset; + u32 val; + + /* Halt the CPU. */ + val = REG_RD_IND(bp, cpu_reg->mode); + val |= cpu_reg->mode_value_halt; + REG_WR_IND(bp, cpu_reg->mode, val); + REG_WR_IND(bp, cpu_reg->state, cpu_reg->state_value_clear); + + /* Load the Text area. */ + offset = cpu_reg->spad_base + (fw->text_addr - cpu_reg->mips_view_base); + if (fw->text) { + int j; + + for (j = 0; j < (fw->text_len / 4); j++, offset += 4) { + REG_WR_IND(bp, offset, fw->text[j]); + } + } + + /* Load the Data area. */ + offset = cpu_reg->spad_base + (fw->data_addr - cpu_reg->mips_view_base); + if (fw->data) { + int j; + + for (j = 0; j < (fw->data_len / 4); j++, offset += 4) { + REG_WR_IND(bp, offset, fw->data[j]); + } + } + + /* Load the SBSS area. */ + offset = cpu_reg->spad_base + (fw->sbss_addr - cpu_reg->mips_view_base); + if (fw->sbss) { + int j; + + for (j = 0; j < (fw->sbss_len / 4); j++, offset += 4) { + REG_WR_IND(bp, offset, fw->sbss[j]); + } + } + + /* Load the BSS area. */ + offset = cpu_reg->spad_base + (fw->bss_addr - cpu_reg->mips_view_base); + if (fw->bss) { + int j; + + for (j = 0; j < (fw->bss_len/4); j++, offset += 4) { + REG_WR_IND(bp, offset, fw->bss[j]); + } + } + + /* Load the Read-Only area. */ + offset = cpu_reg->spad_base + + (fw->rodata_addr - cpu_reg->mips_view_base); + if (fw->rodata) { + int j; + + for (j = 0; j < (fw->rodata_len / 4); j++, offset += 4) { + REG_WR_IND(bp, offset, fw->rodata[j]); + } + } + + /* Clear the pre-fetch instruction. */ + REG_WR_IND(bp, cpu_reg->inst, 0); + REG_WR_IND(bp, cpu_reg->pc, fw->start_addr); + + /* Start the CPU. */ + val = REG_RD_IND(bp, cpu_reg->mode); + val &= ~cpu_reg->mode_value_halt; + REG_WR_IND(bp, cpu_reg->state, cpu_reg->state_value_clear); + REG_WR_IND(bp, cpu_reg->mode, val); +} + +static void +bnx2_init_cpus(struct bnx2 *bp) +{ + struct cpu_reg cpu_reg; + struct fw_info fw; + + /* Initialize the RV2P processor. */ + load_rv2p_fw(bp, bnx2_rv2p_proc1, sizeof(bnx2_rv2p_proc1), RV2P_PROC1); + load_rv2p_fw(bp, bnx2_rv2p_proc2, sizeof(bnx2_rv2p_proc2), RV2P_PROC2); + + /* Initialize the RX Processor. */ + cpu_reg.mode = BNX2_RXP_CPU_MODE; + cpu_reg.mode_value_halt = BNX2_RXP_CPU_MODE_SOFT_HALT; + cpu_reg.mode_value_sstep = BNX2_RXP_CPU_MODE_STEP_ENA; + cpu_reg.state = BNX2_RXP_CPU_STATE; + cpu_reg.state_value_clear = 0xffffff; + cpu_reg.gpr0 = BNX2_RXP_CPU_REG_FILE; + cpu_reg.evmask = BNX2_RXP_CPU_EVENT_MASK; + cpu_reg.pc = BNX2_RXP_CPU_PROGRAM_COUNTER; + cpu_reg.inst = BNX2_RXP_CPU_INSTRUCTION; + cpu_reg.bp = BNX2_RXP_CPU_HW_BREAKPOINT; + cpu_reg.spad_base = BNX2_RXP_SCRATCH; + cpu_reg.mips_view_base = 0x8000000; + + fw.ver_major = bnx2_RXP_b06FwReleaseMajor; + fw.ver_minor = bnx2_RXP_b06FwReleaseMinor; + fw.ver_fix = bnx2_RXP_b06FwReleaseFix; + fw.start_addr = bnx2_RXP_b06FwStartAddr; + + fw.text_addr = bnx2_RXP_b06FwTextAddr; + fw.text_len = bnx2_RXP_b06FwTextLen; + fw.text_index = 0; + fw.text = bnx2_RXP_b06FwText; + + fw.data_addr = bnx2_RXP_b06FwDataAddr; + fw.data_len = bnx2_RXP_b06FwDataLen; + fw.data_index = 0; + fw.data = bnx2_RXP_b06FwData; + + fw.sbss_addr = bnx2_RXP_b06FwSbssAddr; + fw.sbss_len = bnx2_RXP_b06FwSbssLen; + fw.sbss_index = 0; + fw.sbss = bnx2_RXP_b06FwSbss; + + fw.bss_addr = bnx2_RXP_b06FwBssAddr; + fw.bss_len = bnx2_RXP_b06FwBssLen; + fw.bss_index = 0; + fw.bss = bnx2_RXP_b06FwBss; + + fw.rodata_addr = bnx2_RXP_b06FwRodataAddr; + fw.rodata_len = bnx2_RXP_b06FwRodataLen; + fw.rodata_index = 0; + fw.rodata = bnx2_RXP_b06FwRodata; + + load_cpu_fw(bp, &cpu_reg, &fw); + + /* Initialize the TX Processor. */ + cpu_reg.mode = BNX2_TXP_CPU_MODE; + cpu_reg.mode_value_halt = BNX2_TXP_CPU_MODE_SOFT_HALT; + cpu_reg.mode_value_sstep = BNX2_TXP_CPU_MODE_STEP_ENA; + cpu_reg.state = BNX2_TXP_CPU_STATE; + cpu_reg.state_value_clear = 0xffffff; + cpu_reg.gpr0 = BNX2_TXP_CPU_REG_FILE; + cpu_reg.evmask = BNX2_TXP_CPU_EVENT_MASK; + cpu_reg.pc = BNX2_TXP_CPU_PROGRAM_COUNTER; + cpu_reg.inst = BNX2_TXP_CPU_INSTRUCTION; + cpu_reg.bp = BNX2_TXP_CPU_HW_BREAKPOINT; + cpu_reg.spad_base = BNX2_TXP_SCRATCH; + cpu_reg.mips_view_base = 0x8000000; + + fw.ver_major = bnx2_TXP_b06FwReleaseMajor; + fw.ver_minor = bnx2_TXP_b06FwReleaseMinor; + fw.ver_fix = bnx2_TXP_b06FwReleaseFix; + fw.start_addr = bnx2_TXP_b06FwStartAddr; + + fw.text_addr = bnx2_TXP_b06FwTextAddr; + fw.text_len = bnx2_TXP_b06FwTextLen; + fw.text_index = 0; + fw.text = bnx2_TXP_b06FwText; + + fw.data_addr = bnx2_TXP_b06FwDataAddr; + fw.data_len = bnx2_TXP_b06FwDataLen; + fw.data_index = 0; + fw.data = bnx2_TXP_b06FwData; + + fw.sbss_addr = bnx2_TXP_b06FwSbssAddr; + fw.sbss_len = bnx2_TXP_b06FwSbssLen; + fw.sbss_index = 0; + fw.sbss = bnx2_TXP_b06FwSbss; + + fw.bss_addr = bnx2_TXP_b06FwBssAddr; + fw.bss_len = bnx2_TXP_b06FwBssLen; + fw.bss_index = 0; + fw.bss = bnx2_TXP_b06FwBss; + + fw.rodata_addr = bnx2_TXP_b06FwRodataAddr; + fw.rodata_len = bnx2_TXP_b06FwRodataLen; + fw.rodata_index = 0; + fw.rodata = bnx2_TXP_b06FwRodata; + + load_cpu_fw(bp, &cpu_reg, &fw); + + /* Initialize the TX Patch-up Processor. */ + cpu_reg.mode = BNX2_TPAT_CPU_MODE; + cpu_reg.mode_value_halt = BNX2_TPAT_CPU_MODE_SOFT_HALT; + cpu_reg.mode_value_sstep = BNX2_TPAT_CPU_MODE_STEP_ENA; + cpu_reg.state = BNX2_TPAT_CPU_STATE; + cpu_reg.state_value_clear = 0xffffff; + cpu_reg.gpr0 = BNX2_TPAT_CPU_REG_FILE; + cpu_reg.evmask = BNX2_TPAT_CPU_EVENT_MASK; + cpu_reg.pc = BNX2_TPAT_CPU_PROGRAM_COUNTER; + cpu_reg.inst = BNX2_TPAT_CPU_INSTRUCTION; + cpu_reg.bp = BNX2_TPAT_CPU_HW_BREAKPOINT; + cpu_reg.spad_base = BNX2_TPAT_SCRATCH; + cpu_reg.mips_view_base = 0x8000000; + + fw.ver_major = bnx2_TPAT_b06FwReleaseMajor; + fw.ver_minor = bnx2_TPAT_b06FwReleaseMinor; + fw.ver_fix = bnx2_TPAT_b06FwReleaseFix; + fw.start_addr = bnx2_TPAT_b06FwStartAddr; + + fw.text_addr = bnx2_TPAT_b06FwTextAddr; + fw.text_len = bnx2_TPAT_b06FwTextLen; + fw.text_index = 0; + fw.text = bnx2_TPAT_b06FwText; + + fw.data_addr = bnx2_TPAT_b06FwDataAddr; + fw.data_len = bnx2_TPAT_b06FwDataLen; + fw.data_index = 0; + fw.data = bnx2_TPAT_b06FwData; + + fw.sbss_addr = bnx2_TPAT_b06FwSbssAddr; + fw.sbss_len = bnx2_TPAT_b06FwSbssLen; + fw.sbss_index = 0; + fw.sbss = bnx2_TPAT_b06FwSbss; + + fw.bss_addr = bnx2_TPAT_b06FwBssAddr; + fw.bss_len = bnx2_TPAT_b06FwBssLen; + fw.bss_index = 0; + fw.bss = bnx2_TPAT_b06FwBss; + + fw.rodata_addr = bnx2_TPAT_b06FwRodataAddr; + fw.rodata_len = bnx2_TPAT_b06FwRodataLen; + fw.rodata_index = 0; + fw.rodata = bnx2_TPAT_b06FwRodata; + + load_cpu_fw(bp, &cpu_reg, &fw); + + /* Initialize the Completion Processor. */ + cpu_reg.mode = BNX2_COM_CPU_MODE; + cpu_reg.mode_value_halt = BNX2_COM_CPU_MODE_SOFT_HALT; + cpu_reg.mode_value_sstep = BNX2_COM_CPU_MODE_STEP_ENA; + cpu_reg.state = BNX2_COM_CPU_STATE; + cpu_reg.state_value_clear = 0xffffff; + cpu_reg.gpr0 = BNX2_COM_CPU_REG_FILE; + cpu_reg.evmask = BNX2_COM_CPU_EVENT_MASK; + cpu_reg.pc = BNX2_COM_CPU_PROGRAM_COUNTER; + cpu_reg.inst = BNX2_COM_CPU_INSTRUCTION; + cpu_reg.bp = BNX2_COM_CPU_HW_BREAKPOINT; + cpu_reg.spad_base = BNX2_COM_SCRATCH; + cpu_reg.mips_view_base = 0x8000000; + + fw.ver_major = bnx2_COM_b06FwReleaseMajor; + fw.ver_minor = bnx2_COM_b06FwReleaseMinor; + fw.ver_fix = bnx2_COM_b06FwReleaseFix; + fw.start_addr = bnx2_COM_b06FwStartAddr; + + fw.text_addr = bnx2_COM_b06FwTextAddr; + fw.text_len = bnx2_COM_b06FwTextLen; + fw.text_index = 0; + fw.text = bnx2_COM_b06FwText; + + fw.data_addr = bnx2_COM_b06FwDataAddr; + fw.data_len = bnx2_COM_b06FwDataLen; + fw.data_index = 0; + fw.data = bnx2_COM_b06FwData; + + fw.sbss_addr = bnx2_COM_b06FwSbssAddr; + fw.sbss_len = bnx2_COM_b06FwSbssLen; + fw.sbss_index = 0; + fw.sbss = bnx2_COM_b06FwSbss; + + fw.bss_addr = bnx2_COM_b06FwBssAddr; + fw.bss_len = bnx2_COM_b06FwBssLen; + fw.bss_index = 0; + fw.bss = bnx2_COM_b06FwBss; + + fw.rodata_addr = bnx2_COM_b06FwRodataAddr; + fw.rodata_len = bnx2_COM_b06FwRodataLen; + fw.rodata_index = 0; + fw.rodata = bnx2_COM_b06FwRodata; + + load_cpu_fw(bp, &cpu_reg, &fw); + +} + +static int +bnx2_set_power_state(struct bnx2 *bp, int state) +{ + u16 pmcsr; + + pci_read_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL, &pmcsr); + + switch (state) { + case 0: { + u32 val; + + pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL, + (pmcsr & ~PCI_PM_CTRL_STATE_MASK) | + PCI_PM_CTRL_PME_STATUS); + + if (pmcsr & PCI_PM_CTRL_STATE_MASK) + /* delay required during transition out of D3hot */ + msleep(20); + + val = REG_RD(bp, BNX2_EMAC_MODE); + val |= BNX2_EMAC_MODE_MPKT_RCVD | BNX2_EMAC_MODE_ACPI_RCVD; + val &= ~BNX2_EMAC_MODE_MPKT; + REG_WR(bp, BNX2_EMAC_MODE, val); + + val = REG_RD(bp, BNX2_RPM_CONFIG); + val &= ~BNX2_RPM_CONFIG_ACPI_ENA; + REG_WR(bp, BNX2_RPM_CONFIG, val); + break; + } + case 3: { + int i; + u32 val, wol_msg; + + if (bp->wol) { + u32 advertising; + u8 autoneg; + + autoneg = bp->autoneg; + advertising = bp->advertising; + + bp->autoneg = AUTONEG_SPEED; + bp->advertising = ADVERTISED_10baseT_Half | + ADVERTISED_10baseT_Full | + ADVERTISED_100baseT_Half | + ADVERTISED_100baseT_Full | + ADVERTISED_Autoneg; + + bnx2_setup_copper_phy(bp); + + bp->autoneg = autoneg; + bp->advertising = advertising; + + bnx2_set_mac_addr(bp); + + val = REG_RD(bp, BNX2_EMAC_MODE); + + /* Enable port mode. */ + val &= ~BNX2_EMAC_MODE_PORT; + val |= BNX2_EMAC_MODE_PORT_MII | + BNX2_EMAC_MODE_MPKT_RCVD | + BNX2_EMAC_MODE_ACPI_RCVD | + BNX2_EMAC_MODE_FORCE_LINK | + BNX2_EMAC_MODE_MPKT; + + REG_WR(bp, BNX2_EMAC_MODE, val); + + /* receive all multicast */ + for (i = 0; i < NUM_MC_HASH_REGISTERS; i++) { + REG_WR(bp, BNX2_EMAC_MULTICAST_HASH0 + (i * 4), + 0xffffffff); + } + REG_WR(bp, BNX2_EMAC_RX_MODE, + BNX2_EMAC_RX_MODE_SORT_MODE); + + val = 1 | BNX2_RPM_SORT_USER0_BC_EN | + BNX2_RPM_SORT_USER0_MC_EN; + REG_WR(bp, BNX2_RPM_SORT_USER0, 0x0); + REG_WR(bp, BNX2_RPM_SORT_USER0, val); + REG_WR(bp, BNX2_RPM_SORT_USER0, val | + BNX2_RPM_SORT_USER0_ENA); + + /* Need to enable EMAC and RPM for WOL. */ + REG_WR(bp, BNX2_MISC_ENABLE_SET_BITS, + BNX2_MISC_ENABLE_SET_BITS_RX_PARSER_MAC_ENABLE | + BNX2_MISC_ENABLE_SET_BITS_TX_HEADER_Q_ENABLE | + BNX2_MISC_ENABLE_SET_BITS_EMAC_ENABLE); + + val = REG_RD(bp, BNX2_RPM_CONFIG); + val &= ~BNX2_RPM_CONFIG_ACPI_ENA; + REG_WR(bp, BNX2_RPM_CONFIG, val); + + wol_msg = BNX2_DRV_MSG_CODE_SUSPEND_WOL; + } + else { + wol_msg = BNX2_DRV_MSG_CODE_SUSPEND_NO_WOL; + } + + bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT3 | wol_msg); + + pmcsr &= ~PCI_PM_CTRL_STATE_MASK; + if ((CHIP_ID(bp) == CHIP_ID_5706_A0) || + (CHIP_ID(bp) == CHIP_ID_5706_A1)) { + + if (bp->wol) + pmcsr |= 3; + } + else { + pmcsr |= 3; + } + if (bp->wol) { + pmcsr |= PCI_PM_CTRL_PME_ENABLE; + } + pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL, + pmcsr); + + /* No more memory access after this point until + * device is brought back to D0. + */ + udelay(50); + break; + } + default: + return -EINVAL; + } + return 0; +} + +static int +bnx2_acquire_nvram_lock(struct bnx2 *bp) +{ + u32 val; + int j; + + /* Request access to the flash interface. */ + REG_WR(bp, BNX2_NVM_SW_ARB, BNX2_NVM_SW_ARB_ARB_REQ_SET2); + for (j = 0; j < NVRAM_TIMEOUT_COUNT; j++) { + val = REG_RD(bp, BNX2_NVM_SW_ARB); + if (val & BNX2_NVM_SW_ARB_ARB_ARB2) + break; + + udelay(5); + } + + if (j >= NVRAM_TIMEOUT_COUNT) + return -EBUSY; + + return 0; +} + +static int +bnx2_release_nvram_lock(struct bnx2 *bp) +{ + int j; + u32 val; + + /* Relinquish nvram interface. */ + REG_WR(bp, BNX2_NVM_SW_ARB, BNX2_NVM_SW_ARB_ARB_REQ_CLR2); + + for (j = 0; j < NVRAM_TIMEOUT_COUNT; j++) { + val = REG_RD(bp, BNX2_NVM_SW_ARB); + if (!(val & BNX2_NVM_SW_ARB_ARB_ARB2)) + break; + + udelay(5); + } + + if (j >= NVRAM_TIMEOUT_COUNT) + return -EBUSY; + + return 0; +} + + +static int +bnx2_enable_nvram_write(struct bnx2 *bp) +{ + u32 val; + + val = REG_RD(bp, BNX2_MISC_CFG); + REG_WR(bp, BNX2_MISC_CFG, val | BNX2_MISC_CFG_NVM_WR_EN_PCI); + + if (!bp->flash_info->buffered) { + int j; + + REG_WR(bp, BNX2_NVM_COMMAND, BNX2_NVM_COMMAND_DONE); + REG_WR(bp, BNX2_NVM_COMMAND, + BNX2_NVM_COMMAND_WREN | BNX2_NVM_COMMAND_DOIT); + + for (j = 0; j < NVRAM_TIMEOUT_COUNT; j++) { + udelay(5); + + val = REG_RD(bp, BNX2_NVM_COMMAND); + if (val & BNX2_NVM_COMMAND_DONE) + break; + } + + if (j >= NVRAM_TIMEOUT_COUNT) + return -EBUSY; + } + return 0; +} + +static void +bnx2_disable_nvram_write(struct bnx2 *bp) +{ + u32 val; + + val = REG_RD(bp, BNX2_MISC_CFG); + REG_WR(bp, BNX2_MISC_CFG, val & ~BNX2_MISC_CFG_NVM_WR_EN); +} + + +static void +bnx2_enable_nvram_access(struct bnx2 *bp) +{ + u32 val; + + val = REG_RD(bp, BNX2_NVM_ACCESS_ENABLE); + /* Enable both bits, even on read. */ + REG_WR(bp, BNX2_NVM_ACCESS_ENABLE, + val | BNX2_NVM_ACCESS_ENABLE_EN | BNX2_NVM_ACCESS_ENABLE_WR_EN); +} + +static void +bnx2_disable_nvram_access(struct bnx2 *bp) +{ + u32 val; + + val = REG_RD(bp, BNX2_NVM_ACCESS_ENABLE); + /* Disable both bits, even after read. */ + REG_WR(bp, BNX2_NVM_ACCESS_ENABLE, + val & ~(BNX2_NVM_ACCESS_ENABLE_EN | + BNX2_NVM_ACCESS_ENABLE_WR_EN)); +} + +static int +bnx2_nvram_erase_page(struct bnx2 *bp, u32 offset) +{ + u32 cmd; + int j; + + if (bp->flash_info->buffered) + /* Buffered flash, no erase needed */ + return 0; + + /* Build an erase command */ + cmd = BNX2_NVM_COMMAND_ERASE | BNX2_NVM_COMMAND_WR | + BNX2_NVM_COMMAND_DOIT; + + /* Need to clear DONE bit separately. */ + REG_WR(bp, BNX2_NVM_COMMAND, BNX2_NVM_COMMAND_DONE); + + /* Address of the NVRAM to read from. */ + REG_WR(bp, BNX2_NVM_ADDR, offset & BNX2_NVM_ADDR_NVM_ADDR_VALUE); + + /* Issue an erase command. */ + REG_WR(bp, BNX2_NVM_COMMAND, cmd); + + /* Wait for completion. */ + for (j = 0; j < NVRAM_TIMEOUT_COUNT; j++) { + u32 val; + + udelay(5); + + val = REG_RD(bp, BNX2_NVM_COMMAND); + if (val & BNX2_NVM_COMMAND_DONE) + break; + } + + if (j >= NVRAM_TIMEOUT_COUNT) + return -EBUSY; + + return 0; +} + +static int +bnx2_nvram_read_dword(struct bnx2 *bp, u32 offset, u8 *ret_val, u32 cmd_flags) +{ + u32 cmd; + int j; + + /* Build the command word. */ + cmd = BNX2_NVM_COMMAND_DOIT | cmd_flags; + + /* Calculate an offset of a buffered flash. */ + if (bp->flash_info->buffered) { + offset = ((offset / bp->flash_info->page_size) << + bp->flash_info->page_bits) + + (offset % bp->flash_info->page_size); + } + + /* Need to clear DONE bit separately. */ + REG_WR(bp, BNX2_NVM_COMMAND, BNX2_NVM_COMMAND_DONE); + + /* Address of the NVRAM to read from. */ + REG_WR(bp, BNX2_NVM_ADDR, offset & BNX2_NVM_ADDR_NVM_ADDR_VALUE); + + /* Issue a read command. */ + REG_WR(bp, BNX2_NVM_COMMAND, cmd); + + /* Wait for completion. */ + for (j = 0; j < NVRAM_TIMEOUT_COUNT; j++) { + u32 val; + + udelay(5); + + val = REG_RD(bp, BNX2_NVM_COMMAND); + if (val & BNX2_NVM_COMMAND_DONE) { + val = REG_RD(bp, BNX2_NVM_READ); + + val = be32_to_cpu(val); + memcpy(ret_val, &val, 4); + break; + } + } + if (j >= NVRAM_TIMEOUT_COUNT) + return -EBUSY; + + return 0; +} + + +static int +bnx2_nvram_write_dword(struct bnx2 *bp, u32 offset, u8 *val, u32 cmd_flags) +{ + u32 cmd, val32; + int j; + + /* Build the command word. */ + cmd = BNX2_NVM_COMMAND_DOIT | BNX2_NVM_COMMAND_WR | cmd_flags; + + /* Calculate an offset of a buffered flash. */ + if (bp->flash_info->buffered) { + offset = ((offset / bp->flash_info->page_size) << + bp->flash_info->page_bits) + + (offset % bp->flash_info->page_size); + } + + /* Need to clear DONE bit separately. */ + REG_WR(bp, BNX2_NVM_COMMAND, BNX2_NVM_COMMAND_DONE); + + memcpy(&val32, val, 4); + val32 = cpu_to_be32(val32); + + /* Write the data. */ + REG_WR(bp, BNX2_NVM_WRITE, val32); + + /* Address of the NVRAM to write to. */ + REG_WR(bp, BNX2_NVM_ADDR, offset & BNX2_NVM_ADDR_NVM_ADDR_VALUE); + + /* Issue the write command. */ + REG_WR(bp, BNX2_NVM_COMMAND, cmd); + + /* Wait for completion. */ + for (j = 0; j < NVRAM_TIMEOUT_COUNT; j++) { + udelay(5); + + if (REG_RD(bp, BNX2_NVM_COMMAND) & BNX2_NVM_COMMAND_DONE) + break; + } + if (j >= NVRAM_TIMEOUT_COUNT) + return -EBUSY; + + return 0; +} + +static int +bnx2_init_nvram(struct bnx2 *bp) +{ + u32 val; + int j, entry_count, rc; + struct flash_spec *flash; + + /* Determine the selected interface. */ + val = REG_RD(bp, BNX2_NVM_CFG1); + + entry_count = sizeof(flash_table) / sizeof(struct flash_spec); + + rc = 0; + if (val & 0x40000000) { + + /* Flash interface has been reconfigured */ + for (j = 0, flash = &flash_table[0]; j < entry_count; + j++, flash++) { + + if (val == flash->config1) { + bp->flash_info = flash; + break; + } + } + } + else { + /* Not yet been reconfigured */ + + for (j = 0, flash = &flash_table[0]; j < entry_count; + j++, flash++) { + + if ((val & FLASH_STRAP_MASK) == flash->strapping) { + bp->flash_info = flash; + + /* Request access to the flash interface. */ + if ((rc = bnx2_acquire_nvram_lock(bp)) != 0) + return rc; + + /* Enable access to flash interface */ + bnx2_enable_nvram_access(bp); + + /* Reconfigure the flash interface */ + REG_WR(bp, BNX2_NVM_CFG1, flash->config1); + REG_WR(bp, BNX2_NVM_CFG2, flash->config2); + REG_WR(bp, BNX2_NVM_CFG3, flash->config3); + REG_WR(bp, BNX2_NVM_WRITE1, flash->write1); + + /* Disable access to flash interface */ + bnx2_disable_nvram_access(bp); + bnx2_release_nvram_lock(bp); + + break; + } + } + } /* if (val & 0x40000000) */ + + if (j == entry_count) { + bp->flash_info = NULL; + printk(KERN_ALERT "Unknown flash/EEPROM type.\n"); + rc = -ENODEV; + } + + return rc; +} + +static int +bnx2_nvram_read(struct bnx2 *bp, u32 offset, u8 *ret_buf, + int buf_size) +{ + int rc = 0; + u32 cmd_flags, offset32, len32, extra; + + if (buf_size == 0) + return 0; + + /* Request access to the flash interface. */ + if ((rc = bnx2_acquire_nvram_lock(bp)) != 0) + return rc; + + /* Enable access to flash interface */ + bnx2_enable_nvram_access(bp); + + len32 = buf_size; + offset32 = offset; + extra = 0; + + cmd_flags = 0; + + if (offset32 & 3) { + u8 buf[4]; + u32 pre_len; + + offset32 &= ~3; + pre_len = 4 - (offset & 3); + + if (pre_len >= len32) { + pre_len = len32; + cmd_flags = BNX2_NVM_COMMAND_FIRST | + BNX2_NVM_COMMAND_LAST; + } + else { + cmd_flags = BNX2_NVM_COMMAND_FIRST; + } + + rc = bnx2_nvram_read_dword(bp, offset32, buf, cmd_flags); + + if (rc) + return rc; + + memcpy(ret_buf, buf + (offset & 3), pre_len); + + offset32 += 4; + ret_buf += pre_len; + len32 -= pre_len; + } + if (len32 & 3) { + extra = 4 - (len32 & 3); + len32 = (len32 + 4) & ~3; + } + + if (len32 == 4) { + u8 buf[4]; + + if (cmd_flags) + cmd_flags = BNX2_NVM_COMMAND_LAST; + else + cmd_flags = BNX2_NVM_COMMAND_FIRST | + BNX2_NVM_COMMAND_LAST; + + rc = bnx2_nvram_read_dword(bp, offset32, buf, cmd_flags); + + memcpy(ret_buf, buf, 4 - extra); + } + else if (len32 > 0) { + u8 buf[4]; + + /* Read the first word. */ + if (cmd_flags) + cmd_flags = 0; + else + cmd_flags = BNX2_NVM_COMMAND_FIRST; + + rc = bnx2_nvram_read_dword(bp, offset32, ret_buf, cmd_flags); + + /* Advance to the next dword. */ + offset32 += 4; + ret_buf += 4; + len32 -= 4; + + while (len32 > 4 && rc == 0) { + rc = bnx2_nvram_read_dword(bp, offset32, ret_buf, 0); + + /* Advance to the next dword. */ + offset32 += 4; + ret_buf += 4; + len32 -= 4; + } + + if (rc) + return rc; + + cmd_flags = BNX2_NVM_COMMAND_LAST; + rc = bnx2_nvram_read_dword(bp, offset32, buf, cmd_flags); + + memcpy(ret_buf, buf, 4 - extra); + } + + /* Disable access to flash interface */ + bnx2_disable_nvram_access(bp); + + bnx2_release_nvram_lock(bp); + + return rc; +} + +static int +bnx2_nvram_write(struct bnx2 *bp, u32 offset, u8 *data_buf, + int buf_size) +{ + u32 written, offset32, len32; + u8 *buf, start[4], end[4]; + int rc = 0; + int align_start, align_end; + + buf = data_buf; + offset32 = offset; + len32 = buf_size; + align_start = align_end = 0; + + if ((align_start = (offset32 & 3))) { + offset32 &= ~3; + len32 += align_start; + if ((rc = bnx2_nvram_read(bp, offset32, start, 4))) + return rc; + } + + if (len32 & 3) { + if ((len32 > 4) || !align_start) { + align_end = 4 - (len32 & 3); + len32 += align_end; + if ((rc = bnx2_nvram_read(bp, offset32 + len32 - 4, + end, 4))) { + return rc; + } + } + } + + if (align_start || align_end) { + buf = kmalloc(len32, GFP_KERNEL); + if (buf == 0) + return -ENOMEM; + if (align_start) { + memcpy(buf, start, 4); + } + if (align_end) { + memcpy(buf + len32 - 4, end, 4); + } + memcpy(buf + align_start, data_buf, buf_size); + } + + written = 0; + while ((written < len32) && (rc == 0)) { + u32 page_start, page_end, data_start, data_end; + u32 addr, cmd_flags; + int i; + u8 flash_buffer[264]; + + /* Find the page_start addr */ + page_start = offset32 + written; + page_start -= (page_start % bp->flash_info->page_size); + /* Find the page_end addr */ + page_end = page_start + bp->flash_info->page_size; + /* Find the data_start addr */ + data_start = (written == 0) ? offset32 : page_start; + /* Find the data_end addr */ + data_end = (page_end > offset32 + len32) ? + (offset32 + len32) : page_end; + + /* Request access to the flash interface. */ + if ((rc = bnx2_acquire_nvram_lock(bp)) != 0) + goto nvram_write_end; + + /* Enable access to flash interface */ + bnx2_enable_nvram_access(bp); + + cmd_flags = BNX2_NVM_COMMAND_FIRST; + if (bp->flash_info->buffered == 0) { + int j; + + /* Read the whole page into the buffer + * (non-buffer flash only) */ + for (j = 0; j < bp->flash_info->page_size; j += 4) { + if (j == (bp->flash_info->page_size - 4)) { + cmd_flags |= BNX2_NVM_COMMAND_LAST; + } + rc = bnx2_nvram_read_dword(bp, + page_start + j, + &flash_buffer[j], + cmd_flags); + + if (rc) + goto nvram_write_end; + + cmd_flags = 0; + } + } + + /* Enable writes to flash interface (unlock write-protect) */ + if ((rc = bnx2_enable_nvram_write(bp)) != 0) + goto nvram_write_end; + + /* Erase the page */ + if ((rc = bnx2_nvram_erase_page(bp, page_start)) != 0) + goto nvram_write_end; + + /* Re-enable the write again for the actual write */ + bnx2_enable_nvram_write(bp); + + /* Loop to write back the buffer data from page_start to + * data_start */ + i = 0; + if (bp->flash_info->buffered == 0) { + for (addr = page_start; addr < data_start; + addr += 4, i += 4) { + + rc = bnx2_nvram_write_dword(bp, addr, + &flash_buffer[i], cmd_flags); + + if (rc != 0) + goto nvram_write_end; + + cmd_flags = 0; + } + } + + /* Loop to write the new data from data_start to data_end */ + for (addr = data_start; addr < data_end; addr += 4, i++) { + if ((addr == page_end - 4) || + ((bp->flash_info->buffered) && + (addr == data_end - 4))) { + + cmd_flags |= BNX2_NVM_COMMAND_LAST; + } + rc = bnx2_nvram_write_dword(bp, addr, buf, + cmd_flags); + + if (rc != 0) + goto nvram_write_end; + + cmd_flags = 0; + buf += 4; + } + + /* Loop to write back the buffer data from data_end + * to page_end */ + if (bp->flash_info->buffered == 0) { + for (addr = data_end; addr < page_end; + addr += 4, i += 4) { + + if (addr == page_end-4) { + cmd_flags = BNX2_NVM_COMMAND_LAST; + } + rc = bnx2_nvram_write_dword(bp, addr, + &flash_buffer[i], cmd_flags); + + if (rc != 0) + goto nvram_write_end; + + cmd_flags = 0; + } + } + + /* Disable writes to flash interface (lock write-protect) */ + bnx2_disable_nvram_write(bp); + + /* Disable access to flash interface */ + bnx2_disable_nvram_access(bp); + bnx2_release_nvram_lock(bp); + + /* Increment written */ + written += data_end - data_start; + } + +nvram_write_end: + if (align_start || align_end) + kfree(buf); + return rc; +} + +static int +bnx2_reset_chip(struct bnx2 *bp, u32 reset_code) +{ + u32 val; + int i, rc = 0; + + /* Wait for the current PCI transaction to complete before + * issuing a reset. */ + REG_WR(bp, BNX2_MISC_ENABLE_CLR_BITS, + BNX2_MISC_ENABLE_CLR_BITS_TX_DMA_ENABLE | + BNX2_MISC_ENABLE_CLR_BITS_DMA_ENGINE_ENABLE | + BNX2_MISC_ENABLE_CLR_BITS_RX_DMA_ENABLE | + BNX2_MISC_ENABLE_CLR_BITS_HOST_COALESCE_ENABLE); + val = REG_RD(bp, BNX2_MISC_ENABLE_CLR_BITS); + udelay(5); + + /* Deposit a driver reset signature so the firmware knows that + * this is a soft reset. */ + REG_WR_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DRV_RESET_SIGNATURE, + BNX2_DRV_RESET_SIGNATURE_MAGIC); + + bp->fw_timed_out = 0; + + /* Wait for the firmware to tell us it is ok to issue a reset. */ + bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT0 | reset_code); + + /* Do a dummy read to force the chip to complete all current transaction + * before we issue a reset. */ + val = REG_RD(bp, BNX2_MISC_ID); + + val = BNX2_PCICFG_MISC_CONFIG_CORE_RST_REQ | + BNX2_PCICFG_MISC_CONFIG_REG_WINDOW_ENA | + BNX2_PCICFG_MISC_CONFIG_TARGET_MB_WORD_SWAP; + + /* Chip reset. */ + REG_WR(bp, BNX2_PCICFG_MISC_CONFIG, val); + + if ((CHIP_ID(bp) == CHIP_ID_5706_A0) || + (CHIP_ID(bp) == CHIP_ID_5706_A1)) + msleep(15); + + /* Reset takes approximate 30 usec */ + for (i = 0; i < 10; i++) { + val = REG_RD(bp, BNX2_PCICFG_MISC_CONFIG); + if ((val & (BNX2_PCICFG_MISC_CONFIG_CORE_RST_REQ | + BNX2_PCICFG_MISC_CONFIG_CORE_RST_BSY)) == 0) { + break; + } + udelay(10); + } + + if (val & (BNX2_PCICFG_MISC_CONFIG_CORE_RST_REQ | + BNX2_PCICFG_MISC_CONFIG_CORE_RST_BSY)) { + printk(KERN_ERR PFX "Chip reset did not complete\n"); + return -EBUSY; + } + + /* Make sure byte swapping is properly configured. */ + val = REG_RD(bp, BNX2_PCI_SWAP_DIAG0); + if (val != 0x01020304) { + printk(KERN_ERR PFX "Chip not in correct endian mode\n"); + return -ENODEV; + } + + bp->fw_timed_out = 0; + + /* Wait for the firmware to finish its initialization. */ + bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT1 | reset_code); + + if (CHIP_ID(bp) == CHIP_ID_5706_A0) { + /* Adjust the voltage regular to two steps lower. The default + * of this register is 0x0000000e. */ + REG_WR(bp, BNX2_MISC_VREG_CONTROL, 0x000000fa); + + /* Remove bad rbuf memory from the free pool. */ + rc = bnx2_alloc_bad_rbuf(bp); + } + + return rc; +} + +static int +bnx2_init_chip(struct bnx2 *bp) +{ + u32 val; + + /* Make sure the interrupt is not active. */ + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, BNX2_PCICFG_INT_ACK_CMD_MASK_INT); + + val = BNX2_DMA_CONFIG_DATA_BYTE_SWAP | + BNX2_DMA_CONFIG_DATA_WORD_SWAP | +#ifdef __BIG_ENDIAN + BNX2_DMA_CONFIG_CNTL_BYTE_SWAP | +#endif + BNX2_DMA_CONFIG_CNTL_WORD_SWAP | + DMA_READ_CHANS << 12 | + DMA_WRITE_CHANS << 16; + + val |= (0x2 << 20) | (1 << 11); + + if ((bp->flags & PCIX_FLAG) && (bp->bus_speed_mhz = 133)) + val |= (1 << 23); + + if ((CHIP_NUM(bp) == CHIP_NUM_5706) && + (CHIP_ID(bp) != CHIP_ID_5706_A0) && !(bp->flags & PCIX_FLAG)) + val |= BNX2_DMA_CONFIG_CNTL_PING_PONG_DMA; + + REG_WR(bp, BNX2_DMA_CONFIG, val); + + if (CHIP_ID(bp) == CHIP_ID_5706_A0) { + val = REG_RD(bp, BNX2_TDMA_CONFIG); + val |= BNX2_TDMA_CONFIG_ONE_DMA; + REG_WR(bp, BNX2_TDMA_CONFIG, val); + } + + if (bp->flags & PCIX_FLAG) { + u16 val16; + + pci_read_config_word(bp->pdev, bp->pcix_cap + PCI_X_CMD, + &val16); + pci_write_config_word(bp->pdev, bp->pcix_cap + PCI_X_CMD, + val16 & ~PCI_X_CMD_ERO); + } + + REG_WR(bp, BNX2_MISC_ENABLE_SET_BITS, + BNX2_MISC_ENABLE_SET_BITS_HOST_COALESCE_ENABLE | + BNX2_MISC_ENABLE_STATUS_BITS_RX_V2P_ENABLE | + BNX2_MISC_ENABLE_STATUS_BITS_CONTEXT_ENABLE); + + /* Initialize context mapping and zero out the quick contexts. The + * context block must have already been enabled. */ + bnx2_init_context(bp); + + bnx2_init_cpus(bp); + bnx2_init_nvram(bp); + + bnx2_set_mac_addr(bp); + + val = REG_RD(bp, BNX2_MQ_CONFIG); + val &= ~BNX2_MQ_CONFIG_KNL_BYP_BLK_SIZE; + val |= BNX2_MQ_CONFIG_KNL_BYP_BLK_SIZE_256; + REG_WR(bp, BNX2_MQ_CONFIG, val); + + val = 0x10000 + (MAX_CID_CNT * MB_KERNEL_CTX_SIZE); + REG_WR(bp, BNX2_MQ_KNL_BYP_WIND_START, val); + REG_WR(bp, BNX2_MQ_KNL_WIND_END, val); + + val = (BCM_PAGE_BITS - 8) << 24; + REG_WR(bp, BNX2_RV2P_CONFIG, val); + + /* Configure page size. */ + val = REG_RD(bp, BNX2_TBDR_CONFIG); + val &= ~BNX2_TBDR_CONFIG_PAGE_SIZE; + val |= (BCM_PAGE_BITS - 8) << 24 | 0x40; + REG_WR(bp, BNX2_TBDR_CONFIG, val); + + val = bp->mac_addr[0] + + (bp->mac_addr[1] << 8) + + (bp->mac_addr[2] << 16) + + bp->mac_addr[3] + + (bp->mac_addr[4] << 8) + + (bp->mac_addr[5] << 16); + REG_WR(bp, BNX2_EMAC_BACKOFF_SEED, val); + + /* Program the MTU. Also include 4 bytes for CRC32. */ + val = bp->dev->mtu + ETH_HLEN + 4; + if (val > (MAX_ETHERNET_PACKET_SIZE + 4)) + val |= BNX2_EMAC_RX_MTU_SIZE_JUMBO_ENA; + REG_WR(bp, BNX2_EMAC_RX_MTU_SIZE, val); + + bp->last_status_idx = 0; + bp->rx_mode = BNX2_EMAC_RX_MODE_SORT_MODE; + + /* Set up how to generate a link change interrupt. */ + REG_WR(bp, BNX2_EMAC_ATTENTION_ENA, BNX2_EMAC_ATTENTION_ENA_LINK); + + REG_WR(bp, BNX2_HC_STATUS_ADDR_L, + (u64) bp->status_blk_mapping & 0xffffffff); + REG_WR(bp, BNX2_HC_STATUS_ADDR_H, (u64) bp->status_blk_mapping >> 32); + + REG_WR(bp, BNX2_HC_STATISTICS_ADDR_L, + (u64) bp->stats_blk_mapping & 0xffffffff); + REG_WR(bp, BNX2_HC_STATISTICS_ADDR_H, + (u64) bp->stats_blk_mapping >> 32); + + REG_WR(bp, BNX2_HC_TX_QUICK_CONS_TRIP, + (bp->tx_quick_cons_trip_int << 16) | bp->tx_quick_cons_trip); + + REG_WR(bp, BNX2_HC_RX_QUICK_CONS_TRIP, + (bp->rx_quick_cons_trip_int << 16) | bp->rx_quick_cons_trip); + + REG_WR(bp, BNX2_HC_COMP_PROD_TRIP, + (bp->comp_prod_trip_int << 16) | bp->comp_prod_trip); + + REG_WR(bp, BNX2_HC_TX_TICKS, (bp->tx_ticks_int << 16) | bp->tx_ticks); + + REG_WR(bp, BNX2_HC_RX_TICKS, (bp->rx_ticks_int << 16) | bp->rx_ticks); + + REG_WR(bp, BNX2_HC_COM_TICKS, + (bp->com_ticks_int << 16) | bp->com_ticks); + + REG_WR(bp, BNX2_HC_CMD_TICKS, + (bp->cmd_ticks_int << 16) | bp->cmd_ticks); + + REG_WR(bp, BNX2_HC_STATS_TICKS, bp->stats_ticks & 0xffff00); + REG_WR(bp, BNX2_HC_STAT_COLLECT_TICKS, 0xbb8); /* 3ms */ + + if (CHIP_ID(bp) == CHIP_ID_5706_A1) + REG_WR(bp, BNX2_HC_CONFIG, BNX2_HC_CONFIG_COLLECT_STATS); + else { + REG_WR(bp, BNX2_HC_CONFIG, BNX2_HC_CONFIG_RX_TMR_MODE | + BNX2_HC_CONFIG_TX_TMR_MODE | + BNX2_HC_CONFIG_COLLECT_STATS); + } + + /* Clear internal stats counters. */ + REG_WR(bp, BNX2_HC_COMMAND, BNX2_HC_COMMAND_CLR_STAT_NOW); + + REG_WR(bp, BNX2_HC_ATTN_BITS_ENABLE, STATUS_ATTN_BITS_LINK_STATE); + + /* Initialize the receive filter. */ + bnx2_set_rx_mode(bp->dev); + + bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT2 | BNX2_DRV_MSG_CODE_RESET); + + REG_WR(bp, BNX2_MISC_ENABLE_SET_BITS, 0x5ffffff); + REG_RD(bp, BNX2_MISC_ENABLE_SET_BITS); + + udelay(20); + + return 0; +} + + +static void +bnx2_init_tx_ring(struct bnx2 *bp) +{ + struct tx_bd *txbd; + u32 val; + + txbd = &bp->tx_desc_ring[MAX_TX_DESC_CNT]; + + txbd->tx_bd_haddr_hi = (u64) bp->tx_desc_mapping >> 32; + txbd->tx_bd_haddr_lo = (u64) bp->tx_desc_mapping & 0xffffffff; + + bp->tx_prod = 0; + bp->tx_cons = 0; + bp->tx_prod_bseq = 0; + atomic_set(&bp->tx_avail_bd, bp->tx_ring_size); + + val = BNX2_L2CTX_TYPE_TYPE_L2; + val |= BNX2_L2CTX_TYPE_SIZE_L2; + CTX_WR(bp, GET_CID_ADDR(TX_CID), BNX2_L2CTX_TYPE, val); + + val = BNX2_L2CTX_CMD_TYPE_TYPE_L2; + val |= 8 << 16; + CTX_WR(bp, GET_CID_ADDR(TX_CID), BNX2_L2CTX_CMD_TYPE, val); + + val = (u64) bp->tx_desc_mapping >> 32; + CTX_WR(bp, GET_CID_ADDR(TX_CID), BNX2_L2CTX_TBDR_BHADDR_HI, val); + + val = (u64) bp->tx_desc_mapping & 0xffffffff; + CTX_WR(bp, GET_CID_ADDR(TX_CID), BNX2_L2CTX_TBDR_BHADDR_LO, val); +} + +static void +bnx2_init_rx_ring(struct bnx2 *bp) +{ + struct rx_bd *rxbd; + int i; + u16 prod, ring_prod; + u32 val; + + /* 8 for CRC and VLAN */ + bp->rx_buf_use_size = bp->dev->mtu + ETH_HLEN + bp->rx_offset + 8; + /* 8 for alignment */ + bp->rx_buf_size = bp->rx_buf_use_size + 8; + + ring_prod = prod = bp->rx_prod = 0; + bp->rx_cons = 0; + bp->rx_prod_bseq = 0; + + rxbd = &bp->rx_desc_ring[0]; + for (i = 0; i < MAX_RX_DESC_CNT; i++, rxbd++) { + rxbd->rx_bd_len = bp->rx_buf_use_size; + rxbd->rx_bd_flags = RX_BD_FLAGS_START | RX_BD_FLAGS_END; + } + + rxbd->rx_bd_haddr_hi = (u64) bp->rx_desc_mapping >> 32; + rxbd->rx_bd_haddr_lo = (u64) bp->rx_desc_mapping & 0xffffffff; + + val = BNX2_L2CTX_CTX_TYPE_CTX_BD_CHN_TYPE_VALUE; + val |= BNX2_L2CTX_CTX_TYPE_SIZE_L2; + val |= 0x02 << 8; + CTX_WR(bp, GET_CID_ADDR(RX_CID), BNX2_L2CTX_CTX_TYPE, val); + + val = (u64) bp->rx_desc_mapping >> 32; + CTX_WR(bp, GET_CID_ADDR(RX_CID), BNX2_L2CTX_NX_BDHADDR_HI, val); + + val = (u64) bp->rx_desc_mapping & 0xffffffff; + CTX_WR(bp, GET_CID_ADDR(RX_CID), BNX2_L2CTX_NX_BDHADDR_LO, val); + + for ( ;ring_prod < bp->rx_ring_size; ) { + if (bnx2_alloc_rx_skb(bp, ring_prod) < 0) { + break; + } + prod = NEXT_RX_BD(prod); + ring_prod = RX_RING_IDX(prod); + } + bp->rx_prod = prod; + + REG_WR16(bp, MB_RX_CID_ADDR + BNX2_L2CTX_HOST_BDIDX, prod); + + REG_WR(bp, MB_RX_CID_ADDR + BNX2_L2CTX_HOST_BSEQ, bp->rx_prod_bseq); +} + +static void +bnx2_free_tx_skbs(struct bnx2 *bp) +{ + int i; + + if (bp->tx_buf_ring == NULL) + return; + + for (i = 0; i < TX_DESC_CNT; ) { + struct sw_bd *tx_buf = &bp->tx_buf_ring[i]; + struct sk_buff *skb = tx_buf->skb; + int j, last; + + if (skb == NULL) { + i++; + continue; + } + + pci_unmap_single(bp->pdev, pci_unmap_addr(tx_buf, mapping), + skb_headlen(skb), PCI_DMA_TODEVICE); + + tx_buf->skb = NULL; + + last = skb_shinfo(skb)->nr_frags; + for (j = 0; j < last; j++) { + tx_buf = &bp->tx_buf_ring[i + j + 1]; + pci_unmap_page(bp->pdev, + pci_unmap_addr(tx_buf, mapping), + skb_shinfo(skb)->frags[j].size, + PCI_DMA_TODEVICE); + } + dev_kfree_skb_any(skb); + i += j + 1; + } + +} + +static void +bnx2_free_rx_skbs(struct bnx2 *bp) +{ + int i; + + if (bp->rx_buf_ring == NULL) + return; + + for (i = 0; i < RX_DESC_CNT; i++) { + struct sw_bd *rx_buf = &bp->rx_buf_ring[i]; + struct sk_buff *skb = rx_buf->skb; + + if (skb == 0) + continue; + + pci_unmap_single(bp->pdev, pci_unmap_addr(rx_buf, mapping), + bp->rx_buf_use_size, PCI_DMA_FROMDEVICE); + + rx_buf->skb = NULL; + + dev_kfree_skb_any(skb); + } +} + +static void +bnx2_free_skbs(struct bnx2 *bp) +{ + bnx2_free_tx_skbs(bp); + bnx2_free_rx_skbs(bp); +} + +static int +bnx2_reset_nic(struct bnx2 *bp, u32 reset_code) +{ + int rc; + + rc = bnx2_reset_chip(bp, reset_code); + bnx2_free_skbs(bp); + if (rc) + return rc; + + bnx2_init_chip(bp); + bnx2_init_tx_ring(bp); + bnx2_init_rx_ring(bp); + return 0; +} + +static int +bnx2_init_nic(struct bnx2 *bp) +{ + int rc; + + if ((rc = bnx2_reset_nic(bp, BNX2_DRV_MSG_CODE_RESET)) != 0) + return rc; + + bnx2_init_phy(bp); + bnx2_set_link(bp); + return 0; +} + +static int +bnx2_test_registers(struct bnx2 *bp) +{ + int ret; + int i; + static struct { + u16 offset; + u16 flags; + u32 rw_mask; + u32 ro_mask; + } reg_tbl[] = { + { 0x006c, 0, 0x00000000, 0x0000003f }, + { 0x0090, 0, 0xffffffff, 0x00000000 }, + { 0x0094, 0, 0x00000000, 0x00000000 }, + + { 0x0404, 0, 0x00003f00, 0x00000000 }, + { 0x0418, 0, 0x00000000, 0xffffffff }, + { 0x041c, 0, 0x00000000, 0xffffffff }, + { 0x0420, 0, 0x00000000, 0x80ffffff }, + { 0x0424, 0, 0x00000000, 0x00000000 }, + { 0x0428, 0, 0x00000000, 0x00000001 }, + { 0x0450, 0, 0x00000000, 0x0000ffff }, + { 0x0454, 0, 0x00000000, 0xffffffff }, + { 0x0458, 0, 0x00000000, 0xffffffff }, + + { 0x0808, 0, 0x00000000, 0xffffffff }, + { 0x0854, 0, 0x00000000, 0xffffffff }, + { 0x0868, 0, 0x00000000, 0x77777777 }, + { 0x086c, 0, 0x00000000, 0x77777777 }, + { 0x0870, 0, 0x00000000, 0x77777777 }, + { 0x0874, 0, 0x00000000, 0x77777777 }, + + { 0x0c00, 0, 0x00000000, 0x00000001 }, + { 0x0c04, 0, 0x00000000, 0x03ff0001 }, + { 0x0c08, 0, 0x0f0ff073, 0x00000000 }, + { 0x0c0c, 0, 0x00ffffff, 0x00000000 }, + { 0x0c30, 0, 0x00000000, 0xffffffff }, + { 0x0c34, 0, 0x00000000, 0xffffffff }, + { 0x0c38, 0, 0x00000000, 0xffffffff }, + { 0x0c3c, 0, 0x00000000, 0xffffffff }, + { 0x0c40, 0, 0x00000000, 0xffffffff }, + { 0x0c44, 0, 0x00000000, 0xffffffff }, + { 0x0c48, 0, 0x00000000, 0x0007ffff }, + { 0x0c4c, 0, 0x00000000, 0xffffffff }, + { 0x0c50, 0, 0x00000000, 0xffffffff }, + { 0x0c54, 0, 0x00000000, 0xffffffff }, + { 0x0c58, 0, 0x00000000, 0xffffffff }, + { 0x0c5c, 0, 0x00000000, 0xffffffff }, + { 0x0c60, 0, 0x00000000, 0xffffffff }, + { 0x0c64, 0, 0x00000000, 0xffffffff }, + { 0x0c68, 0, 0x00000000, 0xffffffff }, + { 0x0c6c, 0, 0x00000000, 0xffffffff }, + { 0x0c70, 0, 0x00000000, 0xffffffff }, + { 0x0c74, 0, 0x00000000, 0xffffffff }, + { 0x0c78, 0, 0x00000000, 0xffffffff }, + { 0x0c7c, 0, 0x00000000, 0xffffffff }, + { 0x0c80, 0, 0x00000000, 0xffffffff }, + { 0x0c84, 0, 0x00000000, 0xffffffff }, + { 0x0c88, 0, 0x00000000, 0xffffffff }, + { 0x0c8c, 0, 0x00000000, 0xffffffff }, + { 0x0c90, 0, 0x00000000, 0xffffffff }, + { 0x0c94, 0, 0x00000000, 0xffffffff }, + { 0x0c98, 0, 0x00000000, 0xffffffff }, + { 0x0c9c, 0, 0x00000000, 0xffffffff }, + { 0x0ca0, 0, 0x00000000, 0xffffffff }, + { 0x0ca4, 0, 0x00000000, 0xffffffff }, + { 0x0ca8, 0, 0x00000000, 0x0007ffff }, + { 0x0cac, 0, 0x00000000, 0xffffffff }, + { 0x0cb0, 0, 0x00000000, 0xffffffff }, + { 0x0cb4, 0, 0x00000000, 0xffffffff }, + { 0x0cb8, 0, 0x00000000, 0xffffffff }, + { 0x0cbc, 0, 0x00000000, 0xffffffff }, + { 0x0cc0, 0, 0x00000000, 0xffffffff }, + { 0x0cc4, 0, 0x00000000, 0xffffffff }, + { 0x0cc8, 0, 0x00000000, 0xffffffff }, + { 0x0ccc, 0, 0x00000000, 0xffffffff }, + { 0x0cd0, 0, 0x00000000, 0xffffffff }, + { 0x0cd4, 0, 0x00000000, 0xffffffff }, + { 0x0cd8, 0, 0x00000000, 0xffffffff }, + { 0x0cdc, 0, 0x00000000, 0xffffffff }, + { 0x0ce0, 0, 0x00000000, 0xffffffff }, + { 0x0ce4, 0, 0x00000000, 0xffffffff }, + { 0x0ce8, 0, 0x00000000, 0xffffffff }, + { 0x0cec, 0, 0x00000000, 0xffffffff }, + { 0x0cf0, 0, 0x00000000, 0xffffffff }, + { 0x0cf4, 0, 0x00000000, 0xffffffff }, + { 0x0cf8, 0, 0x00000000, 0xffffffff }, + { 0x0cfc, 0, 0x00000000, 0xffffffff }, + { 0x0d00, 0, 0x00000000, 0xffffffff }, + { 0x0d04, 0, 0x00000000, 0xffffffff }, + + { 0x1000, 0, 0x00000000, 0x00000001 }, + { 0x1004, 0, 0x00000000, 0x000f0001 }, + { 0x1044, 0, 0x00000000, 0xffc003ff }, + { 0x1080, 0, 0x00000000, 0x0001ffff }, + { 0x1084, 0, 0x00000000, 0xffffffff }, + { 0x1088, 0, 0x00000000, 0xffffffff }, + { 0x108c, 0, 0x00000000, 0xffffffff }, + { 0x1090, 0, 0x00000000, 0xffffffff }, + { 0x1094, 0, 0x00000000, 0xffffffff }, + { 0x1098, 0, 0x00000000, 0xffffffff }, + { 0x109c, 0, 0x00000000, 0xffffffff }, + { 0x10a0, 0, 0x00000000, 0xffffffff }, + + { 0x1408, 0, 0x01c00800, 0x00000000 }, + { 0x149c, 0, 0x8000ffff, 0x00000000 }, + { 0x14a8, 0, 0x00000000, 0x000001ff }, + { 0x14ac, 0, 0x4fffffff, 0x10000000 }, + { 0x14b0, 0, 0x00000002, 0x00000001 }, + { 0x14b8, 0, 0x00000000, 0x00000000 }, + { 0x14c0, 0, 0x00000000, 0x00000009 }, + { 0x14c4, 0, 0x00003fff, 0x00000000 }, + { 0x14cc, 0, 0x00000000, 0x00000001 }, + { 0x14d0, 0, 0xffffffff, 0x00000000 }, + { 0x1500, 0, 0x00000000, 0xffffffff }, + { 0x1504, 0, 0x00000000, 0xffffffff }, + { 0x1508, 0, 0x00000000, 0xffffffff }, + { 0x150c, 0, 0x00000000, 0xffffffff }, + { 0x1510, 0, 0x00000000, 0xffffffff }, + { 0x1514, 0, 0x00000000, 0xffffffff }, + { 0x1518, 0, 0x00000000, 0xffffffff }, + { 0x151c, 0, 0x00000000, 0xffffffff }, + { 0x1520, 0, 0x00000000, 0xffffffff }, + { 0x1524, 0, 0x00000000, 0xffffffff }, + { 0x1528, 0, 0x00000000, 0xffffffff }, + { 0x152c, 0, 0x00000000, 0xffffffff }, + { 0x1530, 0, 0x00000000, 0xffffffff }, + { 0x1534, 0, 0x00000000, 0xffffffff }, + { 0x1538, 0, 0x00000000, 0xffffffff }, + { 0x153c, 0, 0x00000000, 0xffffffff }, + { 0x1540, 0, 0x00000000, 0xffffffff }, + { 0x1544, 0, 0x00000000, 0xffffffff }, + { 0x1548, 0, 0x00000000, 0xffffffff }, + { 0x154c, 0, 0x00000000, 0xffffffff }, + { 0x1550, 0, 0x00000000, 0xffffffff }, + { 0x1554, 0, 0x00000000, 0xffffffff }, + { 0x1558, 0, 0x00000000, 0xffffffff }, + { 0x1600, 0, 0x00000000, 0xffffffff }, + { 0x1604, 0, 0x00000000, 0xffffffff }, + { 0x1608, 0, 0x00000000, 0xffffffff }, + { 0x160c, 0, 0x00000000, 0xffffffff }, + { 0x1610, 0, 0x00000000, 0xffffffff }, + { 0x1614, 0, 0x00000000, 0xffffffff }, + { 0x1618, 0, 0x00000000, 0xffffffff }, + { 0x161c, 0, 0x00000000, 0xffffffff }, + { 0x1620, 0, 0x00000000, 0xffffffff }, + { 0x1624, 0, 0x00000000, 0xffffffff }, + { 0x1628, 0, 0x00000000, 0xffffffff }, + { 0x162c, 0, 0x00000000, 0xffffffff }, + { 0x1630, 0, 0x00000000, 0xffffffff }, + { 0x1634, 0, 0x00000000, 0xffffffff }, + { 0x1638, 0, 0x00000000, 0xffffffff }, + { 0x163c, 0, 0x00000000, 0xffffffff }, + { 0x1640, 0, 0x00000000, 0xffffffff }, + { 0x1644, 0, 0x00000000, 0xffffffff }, + { 0x1648, 0, 0x00000000, 0xffffffff }, + { 0x164c, 0, 0x00000000, 0xffffffff }, + { 0x1650, 0, 0x00000000, 0xffffffff }, + { 0x1654, 0, 0x00000000, 0xffffffff }, + + { 0x1800, 0, 0x00000000, 0x00000001 }, + { 0x1804, 0, 0x00000000, 0x00000003 }, + { 0x1840, 0, 0x00000000, 0xffffffff }, + { 0x1844, 0, 0x00000000, 0xffffffff }, + { 0x1848, 0, 0x00000000, 0xffffffff }, + { 0x184c, 0, 0x00000000, 0xffffffff }, + { 0x1850, 0, 0x00000000, 0xffffffff }, + { 0x1900, 0, 0x7ffbffff, 0x00000000 }, + { 0x1904, 0, 0xffffffff, 0x00000000 }, + { 0x190c, 0, 0xffffffff, 0x00000000 }, + { 0x1914, 0, 0xffffffff, 0x00000000 }, + { 0x191c, 0, 0xffffffff, 0x00000000 }, + { 0x1924, 0, 0xffffffff, 0x00000000 }, + { 0x192c, 0, 0xffffffff, 0x00000000 }, + { 0x1934, 0, 0xffffffff, 0x00000000 }, + { 0x193c, 0, 0xffffffff, 0x00000000 }, + { 0x1944, 0, 0xffffffff, 0x00000000 }, + { 0x194c, 0, 0xffffffff, 0x00000000 }, + { 0x1954, 0, 0xffffffff, 0x00000000 }, + { 0x195c, 0, 0xffffffff, 0x00000000 }, + { 0x1964, 0, 0xffffffff, 0x00000000 }, + { 0x196c, 0, 0xffffffff, 0x00000000 }, + { 0x1974, 0, 0xffffffff, 0x00000000 }, + { 0x197c, 0, 0xffffffff, 0x00000000 }, + { 0x1980, 0, 0x0700ffff, 0x00000000 }, + + { 0x1c00, 0, 0x00000000, 0x00000001 }, + { 0x1c04, 0, 0x00000000, 0x00000003 }, + { 0x1c08, 0, 0x0000000f, 0x00000000 }, + { 0x1c40, 0, 0x00000000, 0xffffffff }, + { 0x1c44, 0, 0x00000000, 0xffffffff }, + { 0x1c48, 0, 0x00000000, 0xffffffff }, + { 0x1c4c, 0, 0x00000000, 0xffffffff }, + { 0x1c50, 0, 0x00000000, 0xffffffff }, + { 0x1d00, 0, 0x7ffbffff, 0x00000000 }, + { 0x1d04, 0, 0xffffffff, 0x00000000 }, + { 0x1d0c, 0, 0xffffffff, 0x00000000 }, + { 0x1d14, 0, 0xffffffff, 0x00000000 }, + { 0x1d1c, 0, 0xffffffff, 0x00000000 }, + { 0x1d24, 0, 0xffffffff, 0x00000000 }, + { 0x1d2c, 0, 0xffffffff, 0x00000000 }, + { 0x1d34, 0, 0xffffffff, 0x00000000 }, + { 0x1d3c, 0, 0xffffffff, 0x00000000 }, + { 0x1d44, 0, 0xffffffff, 0x00000000 }, + { 0x1d4c, 0, 0xffffffff, 0x00000000 }, + { 0x1d54, 0, 0xffffffff, 0x00000000 }, + { 0x1d5c, 0, 0xffffffff, 0x00000000 }, + { 0x1d64, 0, 0xffffffff, 0x00000000 }, + { 0x1d6c, 0, 0xffffffff, 0x00000000 }, + { 0x1d74, 0, 0xffffffff, 0x00000000 }, + { 0x1d7c, 0, 0xffffffff, 0x00000000 }, + { 0x1d80, 0, 0x0700ffff, 0x00000000 }, + + { 0x2004, 0, 0x00000000, 0x0337000f }, + { 0x2008, 0, 0xffffffff, 0x00000000 }, + { 0x200c, 0, 0xffffffff, 0x00000000 }, + { 0x2010, 0, 0xffffffff, 0x00000000 }, + { 0x2014, 0, 0x801fff80, 0x00000000 }, + { 0x2018, 0, 0x000003ff, 0x00000000 }, + + { 0x2800, 0, 0x00000000, 0x00000001 }, + { 0x2804, 0, 0x00000000, 0x00003f01 }, + { 0x2808, 0, 0x0f3f3f03, 0x00000000 }, + { 0x2810, 0, 0xffff0000, 0x00000000 }, + { 0x2814, 0, 0xffff0000, 0x00000000 }, + { 0x2818, 0, 0xffff0000, 0x00000000 }, + { 0x281c, 0, 0xffff0000, 0x00000000 }, + { 0x2834, 0, 0xffffffff, 0x00000000 }, + { 0x2840, 0, 0x00000000, 0xffffffff }, + { 0x2844, 0, 0x00000000, 0xffffffff }, + { 0x2848, 0, 0xffffffff, 0x00000000 }, + { 0x284c, 0, 0xf800f800, 0x07ff07ff }, + + { 0x2c00, 0, 0x00000000, 0x00000011 }, + { 0x2c04, 0, 0x00000000, 0x00030007 }, + + { 0x3000, 0, 0x00000000, 0x00000001 }, + { 0x3004, 0, 0x00000000, 0x007007ff }, + { 0x3008, 0, 0x00000003, 0x00000000 }, + { 0x300c, 0, 0xffffffff, 0x00000000 }, + { 0x3010, 0, 0xffffffff, 0x00000000 }, + { 0x3014, 0, 0xffffffff, 0x00000000 }, + { 0x3034, 0, 0xffffffff, 0x00000000 }, + { 0x3038, 0, 0xffffffff, 0x00000000 }, + { 0x3050, 0, 0x00000001, 0x00000000 }, + + { 0x3c00, 0, 0x00000000, 0x00000001 }, + { 0x3c04, 0, 0x00000000, 0x00070000 }, + { 0x3c08, 0, 0x00007f71, 0x07f00000 }, + { 0x3c0c, 0, 0x1f3ffffc, 0x00000000 }, + { 0x3c10, 0, 0xffffffff, 0x00000000 }, + { 0x3c14, 0, 0x00000000, 0xffffffff }, + { 0x3c18, 0, 0x00000000, 0xffffffff }, + { 0x3c1c, 0, 0xfffff000, 0x00000000 }, + { 0x3c20, 0, 0xffffff00, 0x00000000 }, + { 0x3c24, 0, 0xffffffff, 0x00000000 }, + { 0x3c28, 0, 0xffffffff, 0x00000000 }, + { 0x3c2c, 0, 0xffffffff, 0x00000000 }, + { 0x3c30, 0, 0xffffffff, 0x00000000 }, + { 0x3c34, 0, 0xffffffff, 0x00000000 }, + { 0x3c38, 0, 0xffffffff, 0x00000000 }, + { 0x3c3c, 0, 0xffffffff, 0x00000000 }, + { 0x3c40, 0, 0xffffffff, 0x00000000 }, + { 0x3c44, 0, 0xffffffff, 0x00000000 }, + { 0x3c48, 0, 0xffffffff, 0x00000000 }, + { 0x3c4c, 0, 0xffffffff, 0x00000000 }, + { 0x3c50, 0, 0xffffffff, 0x00000000 }, + { 0x3c54, 0, 0xffffffff, 0x00000000 }, + { 0x3c58, 0, 0xffffffff, 0x00000000 }, + { 0x3c5c, 0, 0xffffffff, 0x00000000 }, + { 0x3c60, 0, 0xffffffff, 0x00000000 }, + { 0x3c64, 0, 0xffffffff, 0x00000000 }, + { 0x3c68, 0, 0xffffffff, 0x00000000 }, + { 0x3c6c, 0, 0xffffffff, 0x00000000 }, + { 0x3c70, 0, 0xffffffff, 0x00000000 }, + { 0x3c74, 0, 0x0000003f, 0x00000000 }, + { 0x3c78, 0, 0x00000000, 0x00000000 }, + { 0x3c7c, 0, 0x00000000, 0x00000000 }, + { 0x3c80, 0, 0x3fffffff, 0x00000000 }, + { 0x3c84, 0, 0x0000003f, 0x00000000 }, + { 0x3c88, 0, 0x00000000, 0xffffffff }, + { 0x3c8c, 0, 0x00000000, 0xffffffff }, + + { 0x4000, 0, 0x00000000, 0x00000001 }, + { 0x4004, 0, 0x00000000, 0x00030000 }, + { 0x4008, 0, 0x00000ff0, 0x00000000 }, + { 0x400c, 0, 0xffffffff, 0x00000000 }, + { 0x4088, 0, 0x00000000, 0x00070303 }, + + { 0x4400, 0, 0x00000000, 0x00000001 }, + { 0x4404, 0, 0x00000000, 0x00003f01 }, + { 0x4408, 0, 0x7fff00ff, 0x00000000 }, + { 0x440c, 0, 0xffffffff, 0x00000000 }, + { 0x4410, 0, 0xffff, 0x0000 }, + { 0x4414, 0, 0xffff, 0x0000 }, + { 0x4418, 0, 0xffff, 0x0000 }, + { 0x441c, 0, 0xffff, 0x0000 }, + { 0x4428, 0, 0xffffffff, 0x00000000 }, + { 0x442c, 0, 0xffffffff, 0x00000000 }, + { 0x4430, 0, 0xffffffff, 0x00000000 }, + { 0x4434, 0, 0xffffffff, 0x00000000 }, + { 0x4438, 0, 0xffffffff, 0x00000000 }, + { 0x443c, 0, 0xffffffff, 0x00000000 }, + { 0x4440, 0, 0xffffffff, 0x00000000 }, + { 0x4444, 0, 0xffffffff, 0x00000000 }, + + { 0x4c00, 0, 0x00000000, 0x00000001 }, + { 0x4c04, 0, 0x00000000, 0x0000003f }, + { 0x4c08, 0, 0xffffffff, 0x00000000 }, + { 0x4c0c, 0, 0x0007fc00, 0x00000000 }, + { 0x4c10, 0, 0x80003fe0, 0x00000000 }, + { 0x4c14, 0, 0xffffffff, 0x00000000 }, + { 0x4c44, 0, 0x00000000, 0x9fff9fff }, + { 0x4c48, 0, 0x00000000, 0xb3009fff }, + { 0x4c4c, 0, 0x00000000, 0x77f33b30 }, + { 0x4c50, 0, 0x00000000, 0xffffffff }, + + { 0x5004, 0, 0x00000000, 0x0000007f }, + { 0x5008, 0, 0x0f0007ff, 0x00000000 }, + { 0x500c, 0, 0xf800f800, 0x07ff07ff }, + + { 0x5400, 0, 0x00000008, 0x00000001 }, + { 0x5404, 0, 0x00000000, 0x0000003f }, + { 0x5408, 0, 0x0000001f, 0x00000000 }, + { 0x540c, 0, 0xffffffff, 0x00000000 }, + { 0x5410, 0, 0xffffffff, 0x00000000 }, + { 0x5414, 0, 0x0000ffff, 0x00000000 }, + { 0x5418, 0, 0x0000ffff, 0x00000000 }, + { 0x541c, 0, 0x0000ffff, 0x00000000 }, + { 0x5420, 0, 0x0000ffff, 0x00000000 }, + { 0x5428, 0, 0x000000ff, 0x00000000 }, + { 0x542c, 0, 0xff00ffff, 0x00000000 }, + { 0x5430, 0, 0x001fff80, 0x00000000 }, + { 0x5438, 0, 0xffffffff, 0x00000000 }, + { 0x543c, 0, 0xffffffff, 0x00000000 }, + { 0x5440, 0, 0xf800f800, 0x07ff07ff }, + + { 0x5c00, 0, 0x00000000, 0x00000001 }, + { 0x5c04, 0, 0x00000000, 0x0003000f }, + { 0x5c08, 0, 0x00000003, 0x00000000 }, + { 0x5c0c, 0, 0x0000fff8, 0x00000000 }, + { 0x5c10, 0, 0x00000000, 0xffffffff }, + { 0x5c80, 0, 0x00000000, 0x0f7113f1 }, + { 0x5c84, 0, 0x00000000, 0x0000f333 }, + { 0x5c88, 0, 0x00000000, 0x00077373 }, + { 0x5c8c, 0, 0x00000000, 0x0007f737 }, + + { 0x6808, 0, 0x0000ff7f, 0x00000000 }, + { 0x680c, 0, 0xffffffff, 0x00000000 }, + { 0x6810, 0, 0xffffffff, 0x00000000 }, + { 0x6814, 0, 0xffffffff, 0x00000000 }, + { 0x6818, 0, 0xffffffff, 0x00000000 }, + { 0x681c, 0, 0xffffffff, 0x00000000 }, + { 0x6820, 0, 0x00ff00ff, 0x00000000 }, + { 0x6824, 0, 0x00ff00ff, 0x00000000 }, + { 0x6828, 0, 0x00ff00ff, 0x00000000 }, + { 0x682c, 0, 0x03ff03ff, 0x00000000 }, + { 0x6830, 0, 0x03ff03ff, 0x00000000 }, + { 0x6834, 0, 0x03ff03ff, 0x00000000 }, + { 0x6838, 0, 0x03ff03ff, 0x00000000 }, + { 0x683c, 0, 0x0000ffff, 0x00000000 }, + { 0x6840, 0, 0x00000ff0, 0x00000000 }, + { 0x6844, 0, 0x00ffff00, 0x00000000 }, + { 0x684c, 0, 0xffffffff, 0x00000000 }, + { 0x6850, 0, 0x7f7f7f7f, 0x00000000 }, + { 0x6854, 0, 0x7f7f7f7f, 0x00000000 }, + { 0x6858, 0, 0x7f7f7f7f, 0x00000000 }, + { 0x685c, 0, 0x7f7f7f7f, 0x00000000 }, + { 0x6908, 0, 0x00000000, 0x0001ff0f }, + { 0x690c, 0, 0x00000000, 0x0ffe00f0 }, + + { 0xffff, 0, 0x00000000, 0x00000000 }, + }; + + ret = 0; + for (i = 0; reg_tbl[i].offset != 0xffff; i++) { + u32 offset, rw_mask, ro_mask, save_val, val; + + offset = (u32) reg_tbl[i].offset; + rw_mask = reg_tbl[i].rw_mask; + ro_mask = reg_tbl[i].ro_mask; + + save_val = readl((u8 *) bp->regview + offset); + + writel(0, (u8 *) bp->regview + offset); + + val = readl((u8 *) bp->regview + offset); + if ((val & rw_mask) != 0) { + goto reg_test_err; + } + + if ((val & ro_mask) != (save_val & ro_mask)) { + goto reg_test_err; + } + + writel(0xffffffff, (u8 *) bp->regview + offset); + + val = readl((u8 *) bp->regview + offset); + if ((val & rw_mask) != rw_mask) { + goto reg_test_err; + } + + if ((val & ro_mask) != (save_val & ro_mask)) { + goto reg_test_err; + } + + writel(save_val, (u8 *) bp->regview + offset); + continue; + +reg_test_err: + writel(save_val, (u8 *) bp->regview + offset); + ret = -ENODEV; + break; + } + return ret; +} + +static int +bnx2_do_mem_test(struct bnx2 *bp, u32 start, u32 size) +{ + static u32 test_pattern[] = { 0x00000000, 0xffffffff, 0x55555555, + 0xaaaaaaaa , 0xaa55aa55, 0x55aa55aa }; + int i; + + for (i = 0; i < sizeof(test_pattern) / 4; i++) { + u32 offset; + + for (offset = 0; offset < size; offset += 4) { + + REG_WR_IND(bp, start + offset, test_pattern[i]); + + if (REG_RD_IND(bp, start + offset) != + test_pattern[i]) { + return -ENODEV; + } + } + } + return 0; +} + +static int +bnx2_test_memory(struct bnx2 *bp) +{ + int ret = 0; + int i; + static struct { + u32 offset; + u32 len; + } mem_tbl[] = { + { 0x60000, 0x4000 }, + { 0xa0000, 0x4000 }, + { 0xe0000, 0x4000 }, + { 0x120000, 0x4000 }, + { 0x1a0000, 0x4000 }, + { 0x160000, 0x4000 }, + { 0xffffffff, 0 }, + }; + + for (i = 0; mem_tbl[i].offset != 0xffffffff; i++) { + if ((ret = bnx2_do_mem_test(bp, mem_tbl[i].offset, + mem_tbl[i].len)) != 0) { + return ret; + } + } + + return ret; +} + +static int +bnx2_test_loopback(struct bnx2 *bp) +{ + unsigned int pkt_size, num_pkts, i; + struct sk_buff *skb, *rx_skb; + unsigned char *packet; + u16 rx_start_idx, rx_idx, send_idx; + u32 send_bseq, val; + dma_addr_t map; + struct tx_bd *txbd; + struct sw_bd *rx_buf; + struct l2_fhdr *rx_hdr; + int ret = -ENODEV; + + if (!netif_running(bp->dev)) + return -ENODEV; + + bp->loopback = MAC_LOOPBACK; + bnx2_reset_nic(bp, BNX2_DRV_MSG_CODE_DIAG); + bnx2_set_mac_loopback(bp); + + pkt_size = 1514; + skb = dev_alloc_skb(pkt_size); + packet = skb_put(skb, pkt_size); + memcpy(packet, bp->mac_addr, 6); + memset(packet + 6, 0x0, 8); + for (i = 14; i < pkt_size; i++) + packet[i] = (unsigned char) (i & 0xff); + + map = pci_map_single(bp->pdev, skb->data, pkt_size, + PCI_DMA_TODEVICE); + + val = REG_RD(bp, BNX2_HC_COMMAND); + REG_WR(bp, BNX2_HC_COMMAND, val | BNX2_HC_COMMAND_COAL_NOW_WO_INT); + REG_RD(bp, BNX2_HC_COMMAND); + + udelay(5); + rx_start_idx = bp->status_blk->status_rx_quick_consumer_index0; + + send_idx = 0; + send_bseq = 0; + num_pkts = 0; + + txbd = &bp->tx_desc_ring[send_idx]; + + txbd->tx_bd_haddr_hi = (u64) map >> 32; + txbd->tx_bd_haddr_lo = (u64) map & 0xffffffff; + txbd->tx_bd_mss_nbytes = pkt_size; + txbd->tx_bd_vlan_tag_flags = TX_BD_FLAGS_START | TX_BD_FLAGS_END; + + num_pkts++; + send_idx = NEXT_TX_BD(send_idx); + + send_bseq += pkt_size; + + REG_WR16(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BIDX, send_idx); + REG_WR(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BSEQ, send_bseq); + + + udelay(100); + + val = REG_RD(bp, BNX2_HC_COMMAND); + REG_WR(bp, BNX2_HC_COMMAND, val | BNX2_HC_COMMAND_COAL_NOW_WO_INT); + REG_RD(bp, BNX2_HC_COMMAND); + + udelay(5); + + pci_unmap_single(bp->pdev, map, pkt_size, PCI_DMA_TODEVICE); + dev_kfree_skb_irq(skb); + + if (bp->status_blk->status_tx_quick_consumer_index0 != send_idx) { + goto loopback_test_done; + } + + rx_idx = bp->status_blk->status_rx_quick_consumer_index0; + if (rx_idx != rx_start_idx + num_pkts) { + goto loopback_test_done; + } + + rx_buf = &bp->rx_buf_ring[rx_start_idx]; + rx_skb = rx_buf->skb; + + rx_hdr = (struct l2_fhdr *) rx_skb->data; + skb_reserve(rx_skb, bp->rx_offset); + + pci_dma_sync_single_for_cpu(bp->pdev, + pci_unmap_addr(rx_buf, mapping), + bp->rx_buf_size, PCI_DMA_FROMDEVICE); + + if (rx_hdr->l2_fhdr_errors & + (L2_FHDR_ERRORS_BAD_CRC | + L2_FHDR_ERRORS_PHY_DECODE | + L2_FHDR_ERRORS_ALIGNMENT | + L2_FHDR_ERRORS_TOO_SHORT | + L2_FHDR_ERRORS_GIANT_FRAME)) { + + goto loopback_test_done; + } + + if ((rx_hdr->l2_fhdr_pkt_len - 4) != pkt_size) { + goto loopback_test_done; + } + + for (i = 14; i < pkt_size; i++) { + if (*(rx_skb->data + i) != (unsigned char) (i & 0xff)) { + goto loopback_test_done; + } + } + + ret = 0; + +loopback_test_done: + bp->loopback = 0; + return ret; +} + +#define NVRAM_SIZE 0x200 +#define CRC32_RESIDUAL 0xdebb20e3 + +static int +bnx2_test_nvram(struct bnx2 *bp) +{ + u32 buf[NVRAM_SIZE / 4]; + u8 *data = (u8 *) buf; + int rc = 0; + u32 magic, csum; + + if ((rc = bnx2_nvram_read(bp, 0, data, 4)) != 0) + goto test_nvram_done; + + magic = be32_to_cpu(buf[0]); + if (magic != 0x669955aa) { + rc = -ENODEV; + goto test_nvram_done; + } + + if ((rc = bnx2_nvram_read(bp, 0x100, data, NVRAM_SIZE)) != 0) + goto test_nvram_done; + + csum = ether_crc_le(0x100, data); + if (csum != CRC32_RESIDUAL) { + rc = -ENODEV; + goto test_nvram_done; + } + + csum = ether_crc_le(0x100, data + 0x100); + if (csum != CRC32_RESIDUAL) { + rc = -ENODEV; + } + +test_nvram_done: + return rc; +} + +static int +bnx2_test_link(struct bnx2 *bp) +{ + u32 bmsr; + + spin_lock_irq(&bp->phy_lock); + bnx2_read_phy(bp, MII_BMSR, &bmsr); + bnx2_read_phy(bp, MII_BMSR, &bmsr); + spin_unlock_irq(&bp->phy_lock); + + if (bmsr & BMSR_LSTATUS) { + return 0; + } + return -ENODEV; +} + +static int +bnx2_test_intr(struct bnx2 *bp) +{ + int i; + u32 val; + u16 status_idx; + + if (!netif_running(bp->dev)) + return -ENODEV; + + status_idx = REG_RD(bp, BNX2_PCICFG_INT_ACK_CMD) & 0xffff; + + /* This register is not touched during run-time. */ + val = REG_RD(bp, BNX2_HC_COMMAND); + REG_WR(bp, BNX2_HC_COMMAND, val | BNX2_HC_COMMAND_COAL_NOW); + REG_RD(bp, BNX2_HC_COMMAND); + + for (i = 0; i < 10; i++) { + if ((REG_RD(bp, BNX2_PCICFG_INT_ACK_CMD) & 0xffff) != + status_idx) { + + break; + } + + msleep_interruptible(10); + } + if (i < 10) + return 0; + + return -ENODEV; +} + +static void +bnx2_timer(unsigned long data) +{ + struct bnx2 *bp = (struct bnx2 *) data; + u32 msg; + + if (atomic_read(&bp->intr_sem) != 0) + goto bnx2_restart_timer; + + msg = (u32) ++bp->fw_drv_pulse_wr_seq; + REG_WR_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DRV_PULSE_MB, msg); + + if ((bp->phy_flags & PHY_SERDES_FLAG) && + (CHIP_NUM(bp) == CHIP_NUM_5706)) { + unsigned long flags; + + spin_lock_irqsave(&bp->phy_lock, flags); + if (bp->serdes_an_pending) { + bp->serdes_an_pending--; + } + else if ((bp->link_up == 0) && (bp->autoneg & AUTONEG_SPEED)) { + u32 bmcr; + + bnx2_read_phy(bp, MII_BMCR, &bmcr); + + if (bmcr & BMCR_ANENABLE) { + u32 phy1, phy2; + + bnx2_write_phy(bp, 0x1c, 0x7c00); + bnx2_read_phy(bp, 0x1c, &phy1); + + bnx2_write_phy(bp, 0x17, 0x0f01); + bnx2_read_phy(bp, 0x15, &phy2); + bnx2_write_phy(bp, 0x17, 0x0f01); + bnx2_read_phy(bp, 0x15, &phy2); + + if ((phy1 & 0x10) && /* SIGNAL DETECT */ + !(phy2 & 0x20)) { /* no CONFIG */ + + bmcr &= ~BMCR_ANENABLE; + bmcr |= BMCR_SPEED1000 | + BMCR_FULLDPLX; + bnx2_write_phy(bp, MII_BMCR, bmcr); + bp->phy_flags |= + PHY_PARALLEL_DETECT_FLAG; + } + } + } + else if ((bp->link_up) && (bp->autoneg & AUTONEG_SPEED) && + (bp->phy_flags & PHY_PARALLEL_DETECT_FLAG)) { + u32 phy2; + + bnx2_write_phy(bp, 0x17, 0x0f01); + bnx2_read_phy(bp, 0x15, &phy2); + if (phy2 & 0x20) { + u32 bmcr; + + bnx2_read_phy(bp, MII_BMCR, &bmcr); + bmcr |= BMCR_ANENABLE; + bnx2_write_phy(bp, MII_BMCR, bmcr); + + bp->phy_flags &= ~PHY_PARALLEL_DETECT_FLAG; + + } + } + + spin_unlock_irqrestore(&bp->phy_lock, flags); + } + +bnx2_restart_timer: + bp->timer.expires = RUN_AT(bp->timer_interval); + + add_timer(&bp->timer); +} + +/* Called with rtnl_lock */ +static int +bnx2_open(struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + int rc; + + bnx2_set_power_state(bp, 0); + bnx2_disable_int(bp); + + rc = bnx2_alloc_mem(bp); + if (rc) + return rc; + + if ((CHIP_ID(bp) != CHIP_ID_5706_A0) && + (CHIP_ID(bp) != CHIP_ID_5706_A1) && + !disable_msi) { + + if (pci_enable_msi(bp->pdev) == 0) { + bp->flags |= USING_MSI_FLAG; + rc = request_irq(bp->pdev->irq, bnx2_msi, 0, dev->name, + dev); + } + else { + rc = request_irq(bp->pdev->irq, bnx2_interrupt, + SA_SHIRQ, dev->name, dev); + } + } + else { + rc = request_irq(bp->pdev->irq, bnx2_interrupt, SA_SHIRQ, + dev->name, dev); + } + if (rc) { + bnx2_free_mem(bp); + return rc; + } + + rc = bnx2_init_nic(bp); + + if (rc) { + free_irq(bp->pdev->irq, dev); + if (bp->flags & USING_MSI_FLAG) { + pci_disable_msi(bp->pdev); + bp->flags &= ~USING_MSI_FLAG; + } + bnx2_free_skbs(bp); + bnx2_free_mem(bp); + return rc; + } + + init_timer(&bp->timer); + + bp->timer.expires = RUN_AT(bp->timer_interval); + bp->timer.data = (unsigned long) bp; + bp->timer.function = bnx2_timer; + add_timer(&bp->timer); + + atomic_set(&bp->intr_sem, 0); + + bnx2_enable_int(bp); + + if (bp->flags & USING_MSI_FLAG) { + /* Test MSI to make sure it is working + * If MSI test fails, go back to INTx mode + */ + if (bnx2_test_intr(bp) != 0) { + printk(KERN_WARNING PFX "%s: No interrupt was generated" + " using MSI, switching to INTx mode. Please" + " report this failure to the PCI maintainer" + " and include system chipset information.\n", + bp->dev->name); + + bnx2_disable_int(bp); + free_irq(bp->pdev->irq, dev); + pci_disable_msi(bp->pdev); + bp->flags &= ~USING_MSI_FLAG; + + rc = bnx2_init_nic(bp); + + if (!rc) { + rc = request_irq(bp->pdev->irq, bnx2_interrupt, + SA_SHIRQ, dev->name, dev); + } + if (rc) { + bnx2_free_skbs(bp); + bnx2_free_mem(bp); + del_timer_sync(&bp->timer); + return rc; + } + bnx2_enable_int(bp); + } + } + if (bp->flags & USING_MSI_FLAG) { + printk(KERN_INFO PFX "%s: using MSI\n", dev->name); + } + + netif_start_queue(dev); + + return 0; +} + +static void +bnx2_reset_task(void *data) +{ + struct bnx2 *bp = data; + + bnx2_netif_stop(bp); + + bnx2_init_nic(bp); + + atomic_set(&bp->intr_sem, 1); + bnx2_netif_start(bp); +} + +static void +bnx2_tx_timeout(struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + + /* This allows the netif to be shutdown gracefully before resetting */ + schedule_work(&bp->reset_task); +} + +#ifdef BCM_VLAN +/* Called with rtnl_lock */ +static void +bnx2_vlan_rx_register(struct net_device *dev, struct vlan_group *vlgrp) +{ + struct bnx2 *bp = dev->priv; + + bnx2_netif_stop(bp); + + bp->vlgrp = vlgrp; + bnx2_set_rx_mode(dev); + + bnx2_netif_start(bp); +} + +/* Called with rtnl_lock */ +static void +bnx2_vlan_rx_kill_vid(struct net_device *dev, uint16_t vid) +{ + struct bnx2 *bp = dev->priv; + + bnx2_netif_stop(bp); + + if (bp->vlgrp) + bp->vlgrp->vlan_devices[vid] = NULL; + bnx2_set_rx_mode(dev); + + bnx2_netif_start(bp); +} +#endif + +/* Called with dev->xmit_lock. + * hard_start_xmit is pseudo-lockless - a lock is only required when + * the tx queue is full. This way, we get the benefit of lockless + * operations most of the time without the complexities to handle + * netif_stop_queue/wake_queue race conditions. + */ +static int +bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + dma_addr_t mapping; + struct tx_bd *txbd; + struct sw_bd *tx_buf; + u32 len, vlan_tag_flags, last_frag, mss; + u16 prod, ring_prod; + int i; + + if (unlikely(atomic_read(&bp->tx_avail_bd) < + (skb_shinfo(skb)->nr_frags + 1))) { + + netif_stop_queue(dev); + printk(KERN_ERR PFX "%s: BUG! Tx ring full when queue awake!\n", + dev->name); + + return NETDEV_TX_BUSY; + } + len = skb_headlen(skb); + prod = bp->tx_prod; + ring_prod = TX_RING_IDX(prod); + + vlan_tag_flags = 0; + if (skb->ip_summed == CHECKSUM_HW) { + vlan_tag_flags |= TX_BD_FLAGS_TCP_UDP_CKSUM; + } + + if (bp->vlgrp != 0 && vlan_tx_tag_present(skb)) { + vlan_tag_flags |= + (TX_BD_FLAGS_VLAN_TAG | (vlan_tx_tag_get(skb) << 16)); + } +#ifdef BCM_TSO + if ((mss = skb_shinfo(skb)->tso_size) && + (skb->len > (bp->dev->mtu + ETH_HLEN))) { + u32 tcp_opt_len, ip_tcp_len; + + if (skb_header_cloned(skb) && + pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) { + dev_kfree_skb(skb); + return NETDEV_TX_OK; + } + + tcp_opt_len = ((skb->h.th->doff - 5) * 4); + vlan_tag_flags |= TX_BD_FLAGS_SW_LSO; + + tcp_opt_len = 0; + if (skb->h.th->doff > 5) { + tcp_opt_len = (skb->h.th->doff - 5) << 2; + } + ip_tcp_len = (skb->nh.iph->ihl << 2) + sizeof(struct tcphdr); + + skb->nh.iph->check = 0; + skb->nh.iph->tot_len = ntohs(mss + ip_tcp_len + tcp_opt_len); + skb->h.th->check = + ~csum_tcpudp_magic(skb->nh.iph->saddr, + skb->nh.iph->daddr, + 0, IPPROTO_TCP, 0); + + if (tcp_opt_len || (skb->nh.iph->ihl > 5)) { + vlan_tag_flags |= ((skb->nh.iph->ihl - 5) + + (tcp_opt_len >> 2)) << 8; + } + } + else +#endif + { + mss = 0; + } + + mapping = pci_map_single(bp->pdev, skb->data, len, PCI_DMA_TODEVICE); + + tx_buf = &bp->tx_buf_ring[ring_prod]; + tx_buf->skb = skb; + pci_unmap_addr_set(tx_buf, mapping, mapping); + + txbd = &bp->tx_desc_ring[ring_prod]; + + txbd->tx_bd_haddr_hi = (u64) mapping >> 32; + txbd->tx_bd_haddr_lo = (u64) mapping & 0xffffffff; + txbd->tx_bd_mss_nbytes = len | (mss << 16); + txbd->tx_bd_vlan_tag_flags = vlan_tag_flags | TX_BD_FLAGS_START; + + last_frag = skb_shinfo(skb)->nr_frags; + + for (i = 0; i < last_frag; i++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + + prod = NEXT_TX_BD(prod); + ring_prod = TX_RING_IDX(prod); + txbd = &bp->tx_desc_ring[ring_prod]; + + len = frag->size; + mapping = pci_map_page(bp->pdev, frag->page, frag->page_offset, + len, PCI_DMA_TODEVICE); + pci_unmap_addr_set(&bp->tx_buf_ring[ring_prod], + mapping, mapping); + + txbd->tx_bd_haddr_hi = (u64) mapping >> 32; + txbd->tx_bd_haddr_lo = (u64) mapping & 0xffffffff; + txbd->tx_bd_mss_nbytes = len | (mss << 16); + txbd->tx_bd_vlan_tag_flags = vlan_tag_flags; + + } + txbd->tx_bd_vlan_tag_flags |= TX_BD_FLAGS_END; + + prod = NEXT_TX_BD(prod); + bp->tx_prod_bseq += skb->len; + + atomic_sub(last_frag + 1, &bp->tx_avail_bd); + + REG_WR16(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BIDX, prod); + REG_WR(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BSEQ, bp->tx_prod_bseq); + + mmiowb(); + + bp->tx_prod = prod; + dev->trans_start = jiffies; + + if (unlikely(atomic_read(&bp->tx_avail_bd) <= MAX_SKB_FRAGS)) { + unsigned long flags; + + spin_lock_irqsave(&bp->tx_lock, flags); + if (atomic_read(&bp->tx_avail_bd) <= MAX_SKB_FRAGS) { + netif_stop_queue(dev); + + if (atomic_read(&bp->tx_avail_bd) > MAX_SKB_FRAGS) + netif_wake_queue(dev); + } + spin_unlock_irqrestore(&bp->tx_lock, flags); + } + + return NETDEV_TX_OK; +} + +/* Called with rtnl_lock */ +static int +bnx2_close(struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + u32 reset_code; + + flush_scheduled_work(); + bnx2_netif_stop(bp); + del_timer_sync(&bp->timer); + if (bp->wol) + reset_code = BNX2_DRV_MSG_CODE_SUSPEND_WOL; + else + reset_code = BNX2_DRV_MSG_CODE_SUSPEND_NO_WOL; + bnx2_reset_chip(bp, reset_code); + free_irq(bp->pdev->irq, dev); + if (bp->flags & USING_MSI_FLAG) { + pci_disable_msi(bp->pdev); + bp->flags &= ~USING_MSI_FLAG; + } + bnx2_free_skbs(bp); + bnx2_free_mem(bp); + bp->link_up = 0; + netif_carrier_off(bp->dev); + bnx2_set_power_state(bp, 3); + return 0; +} + +#define GET_NET_STATS64(ctr) \ + (unsigned long) ((unsigned long) (ctr##_hi) << 32) + \ + (unsigned long) (ctr##_lo) + +#define GET_NET_STATS32(ctr) \ + (ctr##_lo) + +#if (BITS_PER_LONG == 64) +#define GET_NET_STATS GET_NET_STATS64 +#else +#define GET_NET_STATS GET_NET_STATS32 +#endif + +static struct net_device_stats * +bnx2_get_stats(struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + struct statistics_block *stats_blk = bp->stats_blk; + struct net_device_stats *net_stats = &bp->net_stats; + + if (bp->stats_blk == NULL) { + return net_stats; + } + net_stats->rx_packets = + GET_NET_STATS(stats_blk->stat_IfHCInUcastPkts) + + GET_NET_STATS(stats_blk->stat_IfHCInMulticastPkts) + + GET_NET_STATS(stats_blk->stat_IfHCInBroadcastPkts); + + net_stats->tx_packets = + GET_NET_STATS(stats_blk->stat_IfHCOutUcastPkts) + + GET_NET_STATS(stats_blk->stat_IfHCOutMulticastPkts) + + GET_NET_STATS(stats_blk->stat_IfHCOutBroadcastPkts); + + net_stats->rx_bytes = + GET_NET_STATS(stats_blk->stat_IfHCInOctets); + + net_stats->tx_bytes = + GET_NET_STATS(stats_blk->stat_IfHCOutOctets); + + net_stats->multicast = + GET_NET_STATS(stats_blk->stat_IfHCOutMulticastPkts); + + net_stats->collisions = + (unsigned long) stats_blk->stat_EtherStatsCollisions; + + net_stats->rx_length_errors = + (unsigned long) (stats_blk->stat_EtherStatsUndersizePkts + + stats_blk->stat_EtherStatsOverrsizePkts); + + net_stats->rx_over_errors = + (unsigned long) stats_blk->stat_IfInMBUFDiscards; + + net_stats->rx_frame_errors = + (unsigned long) stats_blk->stat_Dot3StatsAlignmentErrors; + + net_stats->rx_crc_errors = + (unsigned long) stats_blk->stat_Dot3StatsFCSErrors; + + net_stats->rx_errors = net_stats->rx_length_errors + + net_stats->rx_over_errors + net_stats->rx_frame_errors + + net_stats->rx_crc_errors; + + net_stats->tx_aborted_errors = + (unsigned long) (stats_blk->stat_Dot3StatsExcessiveCollisions + + stats_blk->stat_Dot3StatsLateCollisions); + + if (CHIP_NUM(bp) == CHIP_NUM_5706) + net_stats->tx_carrier_errors = 0; + else { + net_stats->tx_carrier_errors = + (unsigned long) + stats_blk->stat_Dot3StatsCarrierSenseErrors; + } + + net_stats->tx_errors = + (unsigned long) + stats_blk->stat_emac_tx_stat_dot3statsinternalmactransmiterrors + + + net_stats->tx_aborted_errors + + net_stats->tx_carrier_errors; + + return net_stats; +} + +/* All ethtool functions called with rtnl_lock */ + +static int +bnx2_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) +{ + struct bnx2 *bp = dev->priv; + + cmd->supported = SUPPORTED_Autoneg; + if (bp->phy_flags & PHY_SERDES_FLAG) { + cmd->supported |= SUPPORTED_1000baseT_Full | + SUPPORTED_FIBRE; + + cmd->port = PORT_FIBRE; + } + else { + cmd->supported |= SUPPORTED_10baseT_Half | + SUPPORTED_10baseT_Full | + SUPPORTED_100baseT_Half | + SUPPORTED_100baseT_Full | + SUPPORTED_1000baseT_Full | + SUPPORTED_TP; + + cmd->port = PORT_TP; + } + + cmd->advertising = bp->advertising; + + if (bp->autoneg & AUTONEG_SPEED) { + cmd->autoneg = AUTONEG_ENABLE; + } + else { + cmd->autoneg = AUTONEG_DISABLE; + } + + if (netif_carrier_ok(dev)) { + cmd->speed = bp->line_speed; + cmd->duplex = bp->duplex; + } + else { + cmd->speed = -1; + cmd->duplex = -1; + } + + cmd->transceiver = XCVR_INTERNAL; + cmd->phy_address = bp->phy_addr; + + return 0; +} + +static int +bnx2_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) +{ + struct bnx2 *bp = dev->priv; + u8 autoneg = bp->autoneg; + u8 req_duplex = bp->req_duplex; + u16 req_line_speed = bp->req_line_speed; + u32 advertising = bp->advertising; + + if (cmd->autoneg == AUTONEG_ENABLE) { + autoneg |= AUTONEG_SPEED; + + cmd->advertising &= ETHTOOL_ALL_COPPER_SPEED; + + /* allow advertising 1 speed */ + if ((cmd->advertising == ADVERTISED_10baseT_Half) || + (cmd->advertising == ADVERTISED_10baseT_Full) || + (cmd->advertising == ADVERTISED_100baseT_Half) || + (cmd->advertising == ADVERTISED_100baseT_Full)) { + + if (bp->phy_flags & PHY_SERDES_FLAG) + return -EINVAL; + + advertising = cmd->advertising; + + } + else if (cmd->advertising == ADVERTISED_1000baseT_Full) { + advertising = cmd->advertising; + } + else if (cmd->advertising == ADVERTISED_1000baseT_Half) { + return -EINVAL; + } + else { + if (bp->phy_flags & PHY_SERDES_FLAG) { + advertising = ETHTOOL_ALL_FIBRE_SPEED; + } + else { + advertising = ETHTOOL_ALL_COPPER_SPEED; + } + } + advertising |= ADVERTISED_Autoneg; + } + else { + if (bp->phy_flags & PHY_SERDES_FLAG) { + if ((cmd->speed != SPEED_1000) || + (cmd->duplex != DUPLEX_FULL)) { + return -EINVAL; + } + } + else if (cmd->speed == SPEED_1000) { + return -EINVAL; + } + autoneg &= ~AUTONEG_SPEED; + req_line_speed = cmd->speed; + req_duplex = cmd->duplex; + advertising = 0; + } + + bp->autoneg = autoneg; + bp->advertising = advertising; + bp->req_line_speed = req_line_speed; + bp->req_duplex = req_duplex; + + spin_lock_irq(&bp->phy_lock); + + bnx2_setup_phy(bp); + + spin_unlock_irq(&bp->phy_lock); + + return 0; +} + +static void +bnx2_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) +{ + struct bnx2 *bp = dev->priv; + + strcpy(info->driver, DRV_MODULE_NAME); + strcpy(info->version, DRV_MODULE_VERSION); + strcpy(info->bus_info, pci_name(bp->pdev)); + info->fw_version[0] = ((bp->fw_ver & 0xff000000) >> 24) + '0'; + info->fw_version[2] = ((bp->fw_ver & 0xff0000) >> 16) + '0'; + info->fw_version[4] = ((bp->fw_ver & 0xff00) >> 8) + '0'; + info->fw_version[6] = (bp->fw_ver & 0xff) + '0'; + info->fw_version[1] = info->fw_version[3] = info->fw_version[5] = '.'; + info->fw_version[7] = 0; +} + +static void +bnx2_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) +{ + struct bnx2 *bp = dev->priv; + + if (bp->flags & NO_WOL_FLAG) { + wol->supported = 0; + wol->wolopts = 0; + } + else { + wol->supported = WAKE_MAGIC; + if (bp->wol) + wol->wolopts = WAKE_MAGIC; + else + wol->wolopts = 0; + } + memset(&wol->sopass, 0, sizeof(wol->sopass)); +} + +static int +bnx2_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) +{ + struct bnx2 *bp = dev->priv; + + if (wol->wolopts & ~WAKE_MAGIC) + return -EINVAL; + + if (wol->wolopts & WAKE_MAGIC) { + if (bp->flags & NO_WOL_FLAG) + return -EINVAL; + + bp->wol = 1; + } + else { + bp->wol = 0; + } + return 0; +} + +static int +bnx2_nway_reset(struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + u32 bmcr; + + if (!(bp->autoneg & AUTONEG_SPEED)) { + return -EINVAL; + } + + spin_lock_irq(&bp->phy_lock); + + /* Force a link down visible on the other side */ + if (bp->phy_flags & PHY_SERDES_FLAG) { + bnx2_write_phy(bp, MII_BMCR, BMCR_LOOPBACK); + spin_unlock_irq(&bp->phy_lock); + + msleep(20); + + spin_lock_irq(&bp->phy_lock); + if (CHIP_NUM(bp) == CHIP_NUM_5706) { + bp->serdes_an_pending = SERDES_AN_TIMEOUT / + bp->timer_interval; + } + } + + bnx2_read_phy(bp, MII_BMCR, &bmcr); + bmcr &= ~BMCR_LOOPBACK; + bnx2_write_phy(bp, MII_BMCR, bmcr | BMCR_ANRESTART | BMCR_ANENABLE); + + spin_unlock_irq(&bp->phy_lock); + + return 0; +} + +static int +bnx2_get_eeprom_len(struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + + if (bp->flash_info == 0) + return 0; + + return (int) bp->flash_info->total_size; +} + +static int +bnx2_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, + u8 *eebuf) +{ + struct bnx2 *bp = dev->priv; + int rc; + + if (eeprom->offset > bp->flash_info->total_size) + return -EINVAL; + + if ((eeprom->offset + eeprom->len) > bp->flash_info->total_size) + eeprom->len = bp->flash_info->total_size - eeprom->offset; + + rc = bnx2_nvram_read(bp, eeprom->offset, eebuf, eeprom->len); + + return rc; +} + +static int +bnx2_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, + u8 *eebuf) +{ + struct bnx2 *bp = dev->priv; + int rc; + + if (eeprom->offset > bp->flash_info->total_size) + return -EINVAL; + + if ((eeprom->offset + eeprom->len) > bp->flash_info->total_size) + eeprom->len = bp->flash_info->total_size - eeprom->offset; + + rc = bnx2_nvram_write(bp, eeprom->offset, eebuf, eeprom->len); + + return rc; +} + +static int +bnx2_get_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) +{ + struct bnx2 *bp = dev->priv; + + memset(coal, 0, sizeof(struct ethtool_coalesce)); + + coal->rx_coalesce_usecs = bp->rx_ticks; + coal->rx_max_coalesced_frames = bp->rx_quick_cons_trip; + coal->rx_coalesce_usecs_irq = bp->rx_ticks_int; + coal->rx_max_coalesced_frames_irq = bp->rx_quick_cons_trip_int; + + coal->tx_coalesce_usecs = bp->tx_ticks; + coal->tx_max_coalesced_frames = bp->tx_quick_cons_trip; + coal->tx_coalesce_usecs_irq = bp->tx_ticks_int; + coal->tx_max_coalesced_frames_irq = bp->tx_quick_cons_trip_int; + + coal->stats_block_coalesce_usecs = bp->stats_ticks; + + return 0; +} + +static int +bnx2_set_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) +{ + struct bnx2 *bp = dev->priv; + + bp->rx_ticks = (u16) coal->rx_coalesce_usecs; + if (bp->rx_ticks > 0x3ff) bp->rx_ticks = 0x3ff; + + bp->rx_quick_cons_trip = (u16) coal->rx_max_coalesced_frames; + if (bp->rx_quick_cons_trip > 0xff) bp->rx_quick_cons_trip = 0xff; + + bp->rx_ticks_int = (u16) coal->rx_coalesce_usecs_irq; + if (bp->rx_ticks_int > 0x3ff) bp->rx_ticks_int = 0x3ff; + + bp->rx_quick_cons_trip_int = (u16) coal->rx_max_coalesced_frames_irq; + if (bp->rx_quick_cons_trip_int > 0xff) + bp->rx_quick_cons_trip_int = 0xff; + + bp->tx_ticks = (u16) coal->tx_coalesce_usecs; + if (bp->tx_ticks > 0x3ff) bp->tx_ticks = 0x3ff; + + bp->tx_quick_cons_trip = (u16) coal->tx_max_coalesced_frames; + if (bp->tx_quick_cons_trip > 0xff) bp->tx_quick_cons_trip = 0xff; + + bp->tx_ticks_int = (u16) coal->tx_coalesce_usecs_irq; + if (bp->tx_ticks_int > 0x3ff) bp->tx_ticks_int = 0x3ff; + + bp->tx_quick_cons_trip_int = (u16) coal->tx_max_coalesced_frames_irq; + if (bp->tx_quick_cons_trip_int > 0xff) bp->tx_quick_cons_trip_int = + 0xff; + + bp->stats_ticks = coal->stats_block_coalesce_usecs; + if (bp->stats_ticks > 0xffff00) bp->stats_ticks = 0xffff00; + bp->stats_ticks &= 0xffff00; + + if (netif_running(bp->dev)) { + bnx2_netif_stop(bp); + bnx2_init_nic(bp); + bnx2_netif_start(bp); + } + + return 0; +} + +static void +bnx2_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering) +{ + struct bnx2 *bp = dev->priv; + + ering->rx_max_pending = MAX_RX_DESC_CNT; + ering->rx_mini_max_pending = 0; + ering->rx_jumbo_max_pending = 0; + + ering->rx_pending = bp->rx_ring_size; + ering->rx_mini_pending = 0; + ering->rx_jumbo_pending = 0; + + ering->tx_max_pending = MAX_TX_DESC_CNT; + ering->tx_pending = bp->tx_ring_size; +} + +static int +bnx2_set_ringparam(struct net_device *dev, struct ethtool_ringparam *ering) +{ + struct bnx2 *bp = dev->priv; + + if ((ering->rx_pending > MAX_RX_DESC_CNT) || + (ering->tx_pending > MAX_TX_DESC_CNT) || + (ering->tx_pending <= MAX_SKB_FRAGS)) { + + return -EINVAL; + } + bp->rx_ring_size = ering->rx_pending; + bp->tx_ring_size = ering->tx_pending; + + if (netif_running(bp->dev)) { + bnx2_netif_stop(bp); + bnx2_init_nic(bp); + bnx2_netif_start(bp); + } + + return 0; +} + +static void +bnx2_get_pauseparam(struct net_device *dev, struct ethtool_pauseparam *epause) +{ + struct bnx2 *bp = dev->priv; + + epause->autoneg = ((bp->autoneg & AUTONEG_FLOW_CTRL) != 0); + epause->rx_pause = ((bp->flow_ctrl & FLOW_CTRL_RX) != 0); + epause->tx_pause = ((bp->flow_ctrl & FLOW_CTRL_TX) != 0); +} + +static int +bnx2_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *epause) +{ + struct bnx2 *bp = dev->priv; + + bp->req_flow_ctrl = 0; + if (epause->rx_pause) + bp->req_flow_ctrl |= FLOW_CTRL_RX; + if (epause->tx_pause) + bp->req_flow_ctrl |= FLOW_CTRL_TX; + + if (epause->autoneg) { + bp->autoneg |= AUTONEG_FLOW_CTRL; + } + else { + bp->autoneg &= ~AUTONEG_FLOW_CTRL; + } + + spin_lock_irq(&bp->phy_lock); + + bnx2_setup_phy(bp); + + spin_unlock_irq(&bp->phy_lock); + + return 0; +} + +static u32 +bnx2_get_rx_csum(struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + + return bp->rx_csum; +} + +static int +bnx2_set_rx_csum(struct net_device *dev, u32 data) +{ + struct bnx2 *bp = dev->priv; + + bp->rx_csum = data; + return 0; +} + +#define BNX2_NUM_STATS 45 + +struct { + char string[ETH_GSTRING_LEN]; +} bnx2_stats_str_arr[BNX2_NUM_STATS] = { + { "rx_bytes" }, + { "rx_error_bytes" }, + { "tx_bytes" }, + { "tx_error_bytes" }, + { "rx_ucast_packets" }, + { "rx_mcast_packets" }, + { "rx_bcast_packets" }, + { "tx_ucast_packets" }, + { "tx_mcast_packets" }, + { "tx_bcast_packets" }, + { "tx_mac_errors" }, + { "tx_carrier_errors" }, + { "rx_crc_errors" }, + { "rx_align_errors" }, + { "tx_single_collisions" }, + { "tx_multi_collisions" }, + { "tx_deferred" }, + { "tx_excess_collisions" }, + { "tx_late_collisions" }, + { "tx_total_collisions" }, + { "rx_fragments" }, + { "rx_jabbers" }, + { "rx_undersize_packets" }, + { "rx_oversize_packets" }, + { "rx_64_byte_packets" }, + { "rx_65_to_127_byte_packets" }, + { "rx_128_to_255_byte_packets" }, + { "rx_256_to_511_byte_packets" }, + { "rx_512_to_1023_byte_packets" }, + { "rx_1024_to_1522_byte_packets" }, + { "rx_1523_to_9022_byte_packets" }, + { "tx_64_byte_packets" }, + { "tx_65_to_127_byte_packets" }, + { "tx_128_to_255_byte_packets" }, + { "tx_256_to_511_byte_packets" }, + { "tx_512_to_1023_byte_packets" }, + { "tx_1024_to_1522_byte_packets" }, + { "tx_1523_to_9022_byte_packets" }, + { "rx_xon_frames" }, + { "rx_xoff_frames" }, + { "tx_xon_frames" }, + { "tx_xoff_frames" }, + { "rx_mac_ctrl_frames" }, + { "rx_filtered_packets" }, + { "rx_discards" }, +}; + +#define STATS_OFFSET32(offset_name) (offsetof(struct statistics_block, offset_name) / 4) + +unsigned long bnx2_stats_offset_arr[BNX2_NUM_STATS] = { + STATS_OFFSET32(stat_IfHCInOctets_hi), + STATS_OFFSET32(stat_IfHCInBadOctets_hi), + STATS_OFFSET32(stat_IfHCOutOctets_hi), + STATS_OFFSET32(stat_IfHCOutBadOctets_hi), + STATS_OFFSET32(stat_IfHCInUcastPkts_hi), + STATS_OFFSET32(stat_IfHCInMulticastPkts_hi), + STATS_OFFSET32(stat_IfHCInBroadcastPkts_hi), + STATS_OFFSET32(stat_IfHCOutUcastPkts_hi), + STATS_OFFSET32(stat_IfHCOutMulticastPkts_hi), + STATS_OFFSET32(stat_IfHCOutBroadcastPkts_hi), + STATS_OFFSET32(stat_emac_tx_stat_dot3statsinternalmactransmiterrors), + STATS_OFFSET32(stat_Dot3StatsCarrierSenseErrors), + STATS_OFFSET32(stat_Dot3StatsFCSErrors), + STATS_OFFSET32(stat_Dot3StatsAlignmentErrors), + STATS_OFFSET32(stat_Dot3StatsSingleCollisionFrames), + STATS_OFFSET32(stat_Dot3StatsMultipleCollisionFrames), + STATS_OFFSET32(stat_Dot3StatsDeferredTransmissions), + STATS_OFFSET32(stat_Dot3StatsExcessiveCollisions), + STATS_OFFSET32(stat_Dot3StatsLateCollisions), + STATS_OFFSET32(stat_EtherStatsCollisions), + STATS_OFFSET32(stat_EtherStatsFragments), + STATS_OFFSET32(stat_EtherStatsJabbers), + STATS_OFFSET32(stat_EtherStatsUndersizePkts), + STATS_OFFSET32(stat_EtherStatsOverrsizePkts), + STATS_OFFSET32(stat_EtherStatsPktsRx64Octets), + STATS_OFFSET32(stat_EtherStatsPktsRx65Octetsto127Octets), + STATS_OFFSET32(stat_EtherStatsPktsRx128Octetsto255Octets), + STATS_OFFSET32(stat_EtherStatsPktsRx256Octetsto511Octets), + STATS_OFFSET32(stat_EtherStatsPktsRx512Octetsto1023Octets), + STATS_OFFSET32(stat_EtherStatsPktsRx1024Octetsto1522Octets), + STATS_OFFSET32(stat_EtherStatsPktsRx1523Octetsto9022Octets), + STATS_OFFSET32(stat_EtherStatsPktsTx64Octets), + STATS_OFFSET32(stat_EtherStatsPktsTx65Octetsto127Octets), + STATS_OFFSET32(stat_EtherStatsPktsTx128Octetsto255Octets), + STATS_OFFSET32(stat_EtherStatsPktsTx256Octetsto511Octets), + STATS_OFFSET32(stat_EtherStatsPktsTx512Octetsto1023Octets), + STATS_OFFSET32(stat_EtherStatsPktsTx1024Octetsto1522Octets), + STATS_OFFSET32(stat_EtherStatsPktsTx1523Octetsto9022Octets), + STATS_OFFSET32(stat_XonPauseFramesReceived), + STATS_OFFSET32(stat_XoffPauseFramesReceived), + STATS_OFFSET32(stat_OutXonSent), + STATS_OFFSET32(stat_OutXoffSent), + STATS_OFFSET32(stat_MacControlFramesReceived), + STATS_OFFSET32(stat_IfInFramesL2FilterDiscards), + STATS_OFFSET32(stat_IfInMBUFDiscards), +}; + +/* stat_IfHCInBadOctets and stat_Dot3StatsCarrierSenseErrors are + * skipped because of errata. + */ +u8 bnx2_5706_stats_len_arr[BNX2_NUM_STATS] = { + 8,0,8,8,8,8,8,8,8,8, + 4,0,4,4,4,4,4,4,4,4, + 4,4,4,4,4,4,4,4,4,4, + 4,4,4,4,4,4,4,4,4,4, + 4,4,4,4,4, +}; + +#define BNX2_NUM_TESTS 6 + +struct { + char string[ETH_GSTRING_LEN]; +} bnx2_tests_str_arr[BNX2_NUM_TESTS] = { + { "register_test (offline)" }, + { "memory_test (offline)" }, + { "loopback_test (offline)" }, + { "nvram_test (online)" }, + { "interrupt_test (online)" }, + { "link_test (online)" }, +}; + +static int +bnx2_self_test_count(struct net_device *dev) +{ + return BNX2_NUM_TESTS; +} + +static void +bnx2_self_test(struct net_device *dev, struct ethtool_test *etest, u64 *buf) +{ + struct bnx2 *bp = dev->priv; + + memset(buf, 0, sizeof(u64) * BNX2_NUM_TESTS); + if (etest->flags & ETH_TEST_FL_OFFLINE) { + bnx2_netif_stop(bp); + bnx2_reset_chip(bp, BNX2_DRV_MSG_CODE_DIAG); + bnx2_free_skbs(bp); + + if (bnx2_test_registers(bp) != 0) { + buf[0] = 1; + etest->flags |= ETH_TEST_FL_FAILED; + } + if (bnx2_test_memory(bp) != 0) { + buf[1] = 1; + etest->flags |= ETH_TEST_FL_FAILED; + } + if (bnx2_test_loopback(bp) != 0) { + buf[2] = 1; + etest->flags |= ETH_TEST_FL_FAILED; + } + + if (!netif_running(bp->dev)) { + bnx2_reset_chip(bp, BNX2_DRV_MSG_CODE_RESET); + } + else { + bnx2_init_nic(bp); + bnx2_netif_start(bp); + } + + /* wait for link up */ + msleep_interruptible(3000); + if ((!bp->link_up) && !(bp->phy_flags & PHY_SERDES_FLAG)) + msleep_interruptible(4000); + } + + if (bnx2_test_nvram(bp) != 0) { + buf[3] = 1; + etest->flags |= ETH_TEST_FL_FAILED; + } + if (bnx2_test_intr(bp) != 0) { + buf[4] = 1; + etest->flags |= ETH_TEST_FL_FAILED; + } + + if (bnx2_test_link(bp) != 0) { + buf[5] = 1; + etest->flags |= ETH_TEST_FL_FAILED; + + } +} + +static void +bnx2_get_strings(struct net_device *dev, u32 stringset, u8 *buf) +{ + switch (stringset) { + case ETH_SS_STATS: + memcpy(buf, bnx2_stats_str_arr, + sizeof(bnx2_stats_str_arr)); + break; + case ETH_SS_TEST: + memcpy(buf, bnx2_tests_str_arr, + sizeof(bnx2_tests_str_arr)); + break; + } +} + +static int +bnx2_get_stats_count(struct net_device *dev) +{ + return BNX2_NUM_STATS; +} + +static void +bnx2_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *stats, u64 *buf) +{ + struct bnx2 *bp = dev->priv; + int i; + u32 *hw_stats = (u32 *) bp->stats_blk; + u8 *stats_len_arr = 0; + + if (hw_stats == NULL) { + memset(buf, 0, sizeof(u64) * BNX2_NUM_STATS); + return; + } + + if (CHIP_NUM(bp) == CHIP_NUM_5706) + stats_len_arr = bnx2_5706_stats_len_arr; + + for (i = 0; i < BNX2_NUM_STATS; i++) { + if (stats_len_arr[i] == 0) { + /* skip this counter */ + buf[i] = 0; + continue; + } + if (stats_len_arr[i] == 4) { + /* 4-byte counter */ + buf[i] = (u64) + *(hw_stats + bnx2_stats_offset_arr[i]); + continue; + } + /* 8-byte counter */ + buf[i] = (((u64) *(hw_stats + + bnx2_stats_offset_arr[i])) << 32) + + *(hw_stats + bnx2_stats_offset_arr[i] + 1); + } +} + +static int +bnx2_phys_id(struct net_device *dev, u32 data) +{ + struct bnx2 *bp = dev->priv; + int i; + u32 save; + + if (data == 0) + data = 2; + + save = REG_RD(bp, BNX2_MISC_CFG); + REG_WR(bp, BNX2_MISC_CFG, BNX2_MISC_CFG_LEDMODE_MAC); + + for (i = 0; i < (data * 2); i++) { + if ((i % 2) == 0) { + REG_WR(bp, BNX2_EMAC_LED, BNX2_EMAC_LED_OVERRIDE); + } + else { + REG_WR(bp, BNX2_EMAC_LED, BNX2_EMAC_LED_OVERRIDE | + BNX2_EMAC_LED_1000MB_OVERRIDE | + BNX2_EMAC_LED_100MB_OVERRIDE | + BNX2_EMAC_LED_10MB_OVERRIDE | + BNX2_EMAC_LED_TRAFFIC_OVERRIDE | + BNX2_EMAC_LED_TRAFFIC); + } + msleep_interruptible(500); + if (signal_pending(current)) + break; + } + REG_WR(bp, BNX2_EMAC_LED, 0); + REG_WR(bp, BNX2_MISC_CFG, save); + return 0; +} + +static struct ethtool_ops bnx2_ethtool_ops = { + .get_settings = bnx2_get_settings, + .set_settings = bnx2_set_settings, + .get_drvinfo = bnx2_get_drvinfo, + .get_wol = bnx2_get_wol, + .set_wol = bnx2_set_wol, + .nway_reset = bnx2_nway_reset, + .get_link = ethtool_op_get_link, + .get_eeprom_len = bnx2_get_eeprom_len, + .get_eeprom = bnx2_get_eeprom, + .set_eeprom = bnx2_set_eeprom, + .get_coalesce = bnx2_get_coalesce, + .set_coalesce = bnx2_set_coalesce, + .get_ringparam = bnx2_get_ringparam, + .set_ringparam = bnx2_set_ringparam, + .get_pauseparam = bnx2_get_pauseparam, + .set_pauseparam = bnx2_set_pauseparam, + .get_rx_csum = bnx2_get_rx_csum, + .set_rx_csum = bnx2_set_rx_csum, + .get_tx_csum = ethtool_op_get_tx_csum, + .set_tx_csum = ethtool_op_set_tx_csum, + .get_sg = ethtool_op_get_sg, + .set_sg = ethtool_op_set_sg, +#ifdef BCM_TSO + .get_tso = ethtool_op_get_tso, + .set_tso = ethtool_op_set_tso, +#endif + .self_test_count = bnx2_self_test_count, + .self_test = bnx2_self_test, + .get_strings = bnx2_get_strings, + .phys_id = bnx2_phys_id, + .get_stats_count = bnx2_get_stats_count, + .get_ethtool_stats = bnx2_get_ethtool_stats, +}; + +/* Called with rtnl_lock */ +static int +bnx2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +{ + struct mii_ioctl_data *data = (struct mii_ioctl_data *)&ifr->ifr_data; + struct bnx2 *bp = dev->priv; + int err; + + switch(cmd) { + case SIOCGMIIPHY: + data->phy_id = bp->phy_addr; + + /* fallthru */ + case SIOCGMIIREG: { + u32 mii_regval; + + spin_lock_irq(&bp->phy_lock); + err = bnx2_read_phy(bp, data->reg_num & 0x1f, &mii_regval); + spin_unlock_irq(&bp->phy_lock); + + data->val_out = mii_regval; + + return err; + } + + case SIOCSMIIREG: + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + + spin_lock_irq(&bp->phy_lock); + err = bnx2_write_phy(bp, data->reg_num & 0x1f, data->val_in); + spin_unlock_irq(&bp->phy_lock); + + return err; + + default: + /* do nothing */ + break; + } + return -EOPNOTSUPP; +} + +/* Called with rtnl_lock */ +static int +bnx2_change_mac_addr(struct net_device *dev, void *p) +{ + struct sockaddr *addr = p; + struct bnx2 *bp = dev->priv; + + memcpy(dev->dev_addr, addr->sa_data, dev->addr_len); + if (netif_running(dev)) + bnx2_set_mac_addr(bp); + + return 0; +} + +/* Called with rtnl_lock */ +static int +bnx2_change_mtu(struct net_device *dev, int new_mtu) +{ + struct bnx2 *bp = dev->priv; + + if (((new_mtu + ETH_HLEN) > MAX_ETHERNET_JUMBO_PACKET_SIZE) || + ((new_mtu + ETH_HLEN) < MIN_ETHERNET_PACKET_SIZE)) + return -EINVAL; + + dev->mtu = new_mtu; + if (netif_running(dev)) { + bnx2_netif_stop(bp); + + bnx2_init_nic(bp); + + bnx2_netif_start(bp); + } + return 0; +} + +#if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER) +static void +poll_bnx2(struct net_device *dev) +{ + struct bnx2 *bp = dev->priv; + + disable_irq(bp->pdev->irq); + bnx2_interrupt(bp->pdev->irq, dev, NULL); + enable_irq(bp->pdev->irq); +} +#endif + +static int __devinit +bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) +{ + struct bnx2 *bp; + unsigned long mem_len; + int rc; + u32 reg; + + SET_MODULE_OWNER(dev); + SET_NETDEV_DEV(dev, &pdev->dev); + bp = dev->priv; + + bp->flags = 0; + bp->phy_flags = 0; + + /* enable device (incl. PCI PM wakeup), and bus-mastering */ + rc = pci_enable_device(pdev); + if (rc) { + printk(KERN_ERR PFX "Cannot enable PCI device, aborting."); + goto err_out; + } + + if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) { + printk(KERN_ERR PFX "Cannot find PCI device base address, " + "aborting.\n"); + rc = -ENODEV; + goto err_out_disable; + } + + rc = pci_request_regions(pdev, DRV_MODULE_NAME); + if (rc) { + printk(KERN_ERR PFX "Cannot obtain PCI resources, aborting.\n"); + goto err_out_disable; + } + + pci_set_master(pdev); + + bp->pm_cap = pci_find_capability(pdev, PCI_CAP_ID_PM); + if (bp->pm_cap == 0) { + printk(KERN_ERR PFX "Cannot find power management capability, " + "aborting.\n"); + rc = -EIO; + goto err_out_release; + } + + bp->pcix_cap = pci_find_capability(pdev, PCI_CAP_ID_PCIX); + if (bp->pcix_cap == 0) { + printk(KERN_ERR PFX "Cannot find PCIX capability, aborting.\n"); + rc = -EIO; + goto err_out_release; + } + + if (pci_set_dma_mask(pdev, DMA_64BIT_MASK) == 0) { + bp->flags |= USING_DAC_FLAG; + if (pci_set_consistent_dma_mask(pdev, DMA_64BIT_MASK) != 0) { + printk(KERN_ERR PFX "pci_set_consistent_dma_mask " + "failed, aborting.\n"); + rc = -EIO; + goto err_out_release; + } + } + else if (pci_set_dma_mask(pdev, DMA_32BIT_MASK) != 0) { + printk(KERN_ERR PFX "System does not support DMA, aborting.\n"); + rc = -EIO; + goto err_out_release; + } + + bp->dev = dev; + bp->pdev = pdev; + + spin_lock_init(&bp->phy_lock); + spin_lock_init(&bp->tx_lock); + INIT_WORK(&bp->reset_task, bnx2_reset_task, bp); + + dev->base_addr = dev->mem_start = pci_resource_start(pdev, 0); + mem_len = MB_GET_CID_ADDR(17); + dev->mem_end = dev->mem_start + mem_len; + dev->irq = pdev->irq; + + bp->regview = ioremap_nocache(dev->base_addr, mem_len); + + if (!bp->regview) { + printk(KERN_ERR PFX "Cannot map register space, aborting.\n"); + rc = -ENOMEM; + goto err_out_release; + } + + /* Configure byte swap and enable write to the reg_window registers. + * Rely on CPU to do target byte swapping on big endian systems + * The chip's target access swapping will not swap all accesses + */ + pci_write_config_dword(bp->pdev, BNX2_PCICFG_MISC_CONFIG, + BNX2_PCICFG_MISC_CONFIG_REG_WINDOW_ENA | + BNX2_PCICFG_MISC_CONFIG_TARGET_MB_WORD_SWAP); + + bnx2_set_power_state(bp, 0); + + bp->chip_id = REG_RD(bp, BNX2_MISC_ID); + + bp->phy_addr = 1; + + /* Get bus information. */ + reg = REG_RD(bp, BNX2_PCICFG_MISC_STATUS); + if (reg & BNX2_PCICFG_MISC_STATUS_PCIX_DET) { + u32 clkreg; + + bp->flags |= PCIX_FLAG; + + clkreg = REG_RD(bp, BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS); + + clkreg &= BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET; + switch (clkreg) { + case BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_133MHZ: + bp->bus_speed_mhz = 133; + break; + + case BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_95MHZ: + bp->bus_speed_mhz = 100; + break; + + case BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_66MHZ: + case BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_80MHZ: + bp->bus_speed_mhz = 66; + break; + + case BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_48MHZ: + case BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_55MHZ: + bp->bus_speed_mhz = 50; + break; + + case BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_LOW: + case BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_32MHZ: + case BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_38MHZ: + bp->bus_speed_mhz = 33; + break; + } + } + else { + if (reg & BNX2_PCICFG_MISC_STATUS_M66EN) + bp->bus_speed_mhz = 66; + else + bp->bus_speed_mhz = 33; + } + + if (reg & BNX2_PCICFG_MISC_STATUS_32BIT_DET) + bp->flags |= PCI_32BIT_FLAG; + + /* 5706A0 may falsely detect SERR and PERR. */ + if (CHIP_ID(bp) == CHIP_ID_5706_A0) { + reg = REG_RD(bp, PCI_COMMAND); + reg &= ~(PCI_COMMAND_SERR | PCI_COMMAND_PARITY); + REG_WR(bp, PCI_COMMAND, reg); + } + else if ((CHIP_ID(bp) == CHIP_ID_5706_A1) && + !(bp->flags & PCIX_FLAG)) { + + printk(KERN_ERR PFX "5706 A1 can only be used in a PCIX bus, " + "aborting.\n"); + goto err_out_unmap; + } + + bnx2_init_nvram(bp); + + /* Get the permanent MAC address. First we need to make sure the + * firmware is actually running. + */ + reg = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DEV_INFO_SIGNATURE); + + if ((reg & BNX2_DEV_INFO_SIGNATURE_MAGIC_MASK) != + BNX2_DEV_INFO_SIGNATURE_MAGIC) { + printk(KERN_ERR PFX "Firmware not running, aborting.\n"); + rc = -ENODEV; + goto err_out_unmap; + } + + bp->fw_ver = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + + BNX2_DEV_INFO_BC_REV); + + reg = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_PORT_HW_CFG_MAC_UPPER); + bp->mac_addr[0] = (u8) (reg >> 8); + bp->mac_addr[1] = (u8) reg; + + reg = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_PORT_HW_CFG_MAC_LOWER); + bp->mac_addr[2] = (u8) (reg >> 24); + bp->mac_addr[3] = (u8) (reg >> 16); + bp->mac_addr[4] = (u8) (reg >> 8); + bp->mac_addr[5] = (u8) reg; + + bp->tx_ring_size = MAX_TX_DESC_CNT; + bp->rx_ring_size = 100; + + bp->rx_csum = 1; + + bp->rx_offset = sizeof(struct l2_fhdr) + 2; + + bp->tx_quick_cons_trip_int = 20; + bp->tx_quick_cons_trip = 20; + bp->tx_ticks_int = 80; + bp->tx_ticks = 80; + + bp->rx_quick_cons_trip_int = 6; + bp->rx_quick_cons_trip = 6; + bp->rx_ticks_int = 18; + bp->rx_ticks = 18; + + bp->stats_ticks = 1000000 & 0xffff00; + + bp->timer_interval = HZ; + + /* Disable WOL support if we are running on a SERDES chip. */ + if (CHIP_BOND_ID(bp) & CHIP_BOND_ID_SERDES_BIT) { + bp->phy_flags |= PHY_SERDES_FLAG; + bp->flags |= NO_WOL_FLAG; + } + + if (CHIP_ID(bp) == CHIP_ID_5706_A0) { + bp->tx_quick_cons_trip_int = + bp->tx_quick_cons_trip; + bp->tx_ticks_int = bp->tx_ticks; + bp->rx_quick_cons_trip_int = + bp->rx_quick_cons_trip; + bp->rx_ticks_int = bp->rx_ticks; + bp->comp_prod_trip_int = bp->comp_prod_trip; + bp->com_ticks_int = bp->com_ticks; + bp->cmd_ticks_int = bp->cmd_ticks; + } + + bp->autoneg = AUTONEG_SPEED | AUTONEG_FLOW_CTRL; + bp->req_line_speed = 0; + if (bp->phy_flags & PHY_SERDES_FLAG) { + bp->advertising = ETHTOOL_ALL_FIBRE_SPEED | ADVERTISED_Autoneg; + } + else { + bp->advertising = ETHTOOL_ALL_COPPER_SPEED | ADVERTISED_Autoneg; + } + + bp->req_flow_ctrl = FLOW_CTRL_RX | FLOW_CTRL_TX; + + return 0; + +err_out_unmap: + if (bp->regview) { + iounmap(bp->regview); + } + +err_out_release: + pci_release_regions(pdev); + +err_out_disable: + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); + +err_out: + return rc; +} + +static int __devinit +bnx2_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + static int version_printed = 0; + struct net_device *dev = NULL; + struct bnx2 *bp; + int rc, i; + + if (version_printed++ == 0) + printk(KERN_INFO "%s", version); + + /* dev zeroed in init_etherdev */ + dev = alloc_etherdev(sizeof(*bp)); + + if (!dev) + return -ENOMEM; + + rc = bnx2_init_board(pdev, dev); + if (rc < 0) { + free_netdev(dev); + return rc; + } + + dev->open = bnx2_open; + dev->hard_start_xmit = bnx2_start_xmit; + dev->stop = bnx2_close; + dev->get_stats = bnx2_get_stats; + dev->set_multicast_list = bnx2_set_rx_mode; + dev->do_ioctl = bnx2_ioctl; + dev->set_mac_address = bnx2_change_mac_addr; + dev->change_mtu = bnx2_change_mtu; + dev->tx_timeout = bnx2_tx_timeout; + dev->watchdog_timeo = TX_TIMEOUT; +#ifdef BCM_VLAN + dev->vlan_rx_register = bnx2_vlan_rx_register; + dev->vlan_rx_kill_vid = bnx2_vlan_rx_kill_vid; +#endif + dev->poll = bnx2_poll; + dev->ethtool_ops = &bnx2_ethtool_ops; + dev->weight = 64; + + bp = dev->priv; + +#if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER) + dev->poll_controller = poll_bnx2; +#endif + + if ((rc = register_netdev(dev))) { + printk(KERN_ERR PFX "Cannot register net device\n"); + if (bp->regview) + iounmap(bp->regview); + pci_release_regions(pdev); + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); + free_netdev(dev); + return rc; + } + + pci_set_drvdata(pdev, dev); + + memcpy(dev->dev_addr, bp->mac_addr, 6); + bp->name = board_info[ent->driver_data].name, + printk(KERN_INFO "%s: %s (%c%d) PCI%s %s %dMHz found at mem %lx, " + "IRQ %d, ", + dev->name, + bp->name, + ((CHIP_ID(bp) & 0xf000) >> 12) + 'A', + ((CHIP_ID(bp) & 0x0ff0) >> 4), + ((bp->flags & PCIX_FLAG) ? "-X" : ""), + ((bp->flags & PCI_32BIT_FLAG) ? "32-bit" : "64-bit"), + bp->bus_speed_mhz, + dev->base_addr, + bp->pdev->irq); + + printk("node addr "); + for (i = 0; i < 6; i++) + printk("%2.2x", dev->dev_addr[i]); + printk("\n"); + + dev->features |= NETIF_F_SG; + if (bp->flags & USING_DAC_FLAG) + dev->features |= NETIF_F_HIGHDMA; + dev->features |= NETIF_F_IP_CSUM; +#ifdef BCM_VLAN + dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX; +#endif +#ifdef BCM_TSO + dev->features |= NETIF_F_TSO; +#endif + + netif_carrier_off(bp->dev); + + return 0; +} + +static void __devexit +bnx2_remove_one(struct pci_dev *pdev) +{ + struct net_device *dev = pci_get_drvdata(pdev); + struct bnx2 *bp = dev->priv; + + unregister_netdev(dev); + + if (bp->regview) + iounmap(bp->regview); + + free_netdev(dev); + pci_release_regions(pdev); + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); +} + +static int +bnx2_suspend(struct pci_dev *pdev, u32 state) +{ + struct net_device *dev = pci_get_drvdata(pdev); + struct bnx2 *bp = dev->priv; + u32 reset_code; + + if (!netif_running(dev)) + return 0; + + bnx2_netif_stop(bp); + netif_device_detach(dev); + del_timer_sync(&bp->timer); + if (bp->wol) + reset_code = BNX2_DRV_MSG_CODE_SUSPEND_WOL; + else + reset_code = BNX2_DRV_MSG_CODE_SUSPEND_NO_WOL; + bnx2_reset_chip(bp, reset_code); + bnx2_free_skbs(bp); + bnx2_set_power_state(bp, state); + return 0; +} + +static int +bnx2_resume(struct pci_dev *pdev) +{ + struct net_device *dev = pci_get_drvdata(pdev); + struct bnx2 *bp = dev->priv; + + if (!netif_running(dev)) + return 0; + + bnx2_set_power_state(bp, 0); + netif_device_attach(dev); + bnx2_init_nic(bp); + bnx2_netif_start(bp); + return 0; +} + +static struct pci_driver bnx2_pci_driver = { + name: DRV_MODULE_NAME, + id_table: bnx2_pci_tbl, + probe: bnx2_init_one, + remove: __devexit_p(bnx2_remove_one), + suspend: bnx2_suspend, + resume: bnx2_resume, +}; + +static int __init bnx2_init(void) +{ + return pci_module_init(&bnx2_pci_driver); +} + +static void __exit bnx2_cleanup(void) +{ + pci_unregister_driver(&bnx2_pci_driver); +} + +module_init(bnx2_init); +module_exit(bnx2_cleanup); + + + diff --git a/drivers/net/bnx2.h b/drivers/net/bnx2.h new file mode 100644 index 000000000000..8214a2853d0d --- /dev/null +++ b/drivers/net/bnx2.h @@ -0,0 +1,4352 @@ +/* bnx2.h: Broadcom NX2 network driver. + * + * Copyright (c) 2004, 2005 Broadcom Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation. + * + * Written by: Michael Chan (mchan@broadcom.com) + */ + + +#ifndef BNX2_H +#define BNX2_H + +#include + +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#ifdef NETIF_F_HW_VLAN_TX +#include +#define BCM_VLAN 1 +#endif +#ifdef NETIF_F_TSO +#include +#include +#include +#define BCM_TSO 1 +#endif +#include +#include + +/* Hardware data structures and register definitions automatically + * generated from RTL code. Do not modify. + */ + +/* + * tx_bd definition + */ +struct tx_bd { + u32 tx_bd_haddr_hi; + u32 tx_bd_haddr_lo; + u32 tx_bd_mss_nbytes; + u32 tx_bd_vlan_tag_flags; + #define TX_BD_FLAGS_CONN_FAULT (1<<0) + #define TX_BD_FLAGS_TCP_UDP_CKSUM (1<<1) + #define TX_BD_FLAGS_IP_CKSUM (1<<2) + #define TX_BD_FLAGS_VLAN_TAG (1<<3) + #define TX_BD_FLAGS_COAL_NOW (1<<4) + #define TX_BD_FLAGS_DONT_GEN_CRC (1<<5) + #define TX_BD_FLAGS_END (1<<6) + #define TX_BD_FLAGS_START (1<<7) + #define TX_BD_FLAGS_SW_OPTION_WORD (0x1f<<8) + #define TX_BD_FLAGS_SW_FLAGS (1<<13) + #define TX_BD_FLAGS_SW_SNAP (1<<14) + #define TX_BD_FLAGS_SW_LSO (1<<15) + +}; + + +/* + * rx_bd definition + */ +struct rx_bd { + u32 rx_bd_haddr_hi; + u32 rx_bd_haddr_lo; + u32 rx_bd_len; + u32 rx_bd_flags; + #define RX_BD_FLAGS_NOPUSH (1<<0) + #define RX_BD_FLAGS_DUMMY (1<<1) + #define RX_BD_FLAGS_END (1<<2) + #define RX_BD_FLAGS_START (1<<3) + +}; + + +/* + * status_block definition + */ +struct status_block { + u32 status_attn_bits; + #define STATUS_ATTN_BITS_LINK_STATE (1L<<0) + #define STATUS_ATTN_BITS_TX_SCHEDULER_ABORT (1L<<1) + #define STATUS_ATTN_BITS_TX_BD_READ_ABORT (1L<<2) + #define STATUS_ATTN_BITS_TX_BD_CACHE_ABORT (1L<<3) + #define STATUS_ATTN_BITS_TX_PROCESSOR_ABORT (1L<<4) + #define STATUS_ATTN_BITS_TX_DMA_ABORT (1L<<5) + #define STATUS_ATTN_BITS_TX_PATCHUP_ABORT (1L<<6) + #define STATUS_ATTN_BITS_TX_ASSEMBLER_ABORT (1L<<7) + #define STATUS_ATTN_BITS_RX_PARSER_MAC_ABORT (1L<<8) + #define STATUS_ATTN_BITS_RX_PARSER_CATCHUP_ABORT (1L<<9) + #define STATUS_ATTN_BITS_RX_MBUF_ABORT (1L<<10) + #define STATUS_ATTN_BITS_RX_LOOKUP_ABORT (1L<<11) + #define STATUS_ATTN_BITS_RX_PROCESSOR_ABORT (1L<<12) + #define STATUS_ATTN_BITS_RX_V2P_ABORT (1L<<13) + #define STATUS_ATTN_BITS_RX_BD_CACHE_ABORT (1L<<14) + #define STATUS_ATTN_BITS_RX_DMA_ABORT (1L<<15) + #define STATUS_ATTN_BITS_COMPLETION_ABORT (1L<<16) + #define STATUS_ATTN_BITS_HOST_COALESCE_ABORT (1L<<17) + #define STATUS_ATTN_BITS_MAILBOX_QUEUE_ABORT (1L<<18) + #define STATUS_ATTN_BITS_CONTEXT_ABORT (1L<<19) + #define STATUS_ATTN_BITS_CMD_SCHEDULER_ABORT (1L<<20) + #define STATUS_ATTN_BITS_CMD_PROCESSOR_ABORT (1L<<21) + #define STATUS_ATTN_BITS_MGMT_PROCESSOR_ABORT (1L<<22) + #define STATUS_ATTN_BITS_MAC_ABORT (1L<<23) + #define STATUS_ATTN_BITS_TIMER_ABORT (1L<<24) + #define STATUS_ATTN_BITS_DMAE_ABORT (1L<<25) + #define STATUS_ATTN_BITS_FLSH_ABORT (1L<<26) + #define STATUS_ATTN_BITS_GRC_ABORT (1L<<27) + #define STATUS_ATTN_BITS_PARITY_ERROR (1L<<31) + + u32 status_attn_bits_ack; +#if defined(__BIG_ENDIAN) + u16 status_tx_quick_consumer_index0; + u16 status_tx_quick_consumer_index1; + u16 status_tx_quick_consumer_index2; + u16 status_tx_quick_consumer_index3; + u16 status_rx_quick_consumer_index0; + u16 status_rx_quick_consumer_index1; + u16 status_rx_quick_consumer_index2; + u16 status_rx_quick_consumer_index3; + u16 status_rx_quick_consumer_index4; + u16 status_rx_quick_consumer_index5; + u16 status_rx_quick_consumer_index6; + u16 status_rx_quick_consumer_index7; + u16 status_rx_quick_consumer_index8; + u16 status_rx_quick_consumer_index9; + u16 status_rx_quick_consumer_index10; + u16 status_rx_quick_consumer_index11; + u16 status_rx_quick_consumer_index12; + u16 status_rx_quick_consumer_index13; + u16 status_rx_quick_consumer_index14; + u16 status_rx_quick_consumer_index15; + u16 status_completion_producer_index; + u16 status_cmd_consumer_index; + u16 status_idx; + u16 status_unused; +#elif defined(__LITTLE_ENDIAN) + u16 status_tx_quick_consumer_index1; + u16 status_tx_quick_consumer_index0; + u16 status_tx_quick_consumer_index3; + u16 status_tx_quick_consumer_index2; + u16 status_rx_quick_consumer_index1; + u16 status_rx_quick_consumer_index0; + u16 status_rx_quick_consumer_index3; + u16 status_rx_quick_consumer_index2; + u16 status_rx_quick_consumer_index5; + u16 status_rx_quick_consumer_index4; + u16 status_rx_quick_consumer_index7; + u16 status_rx_quick_consumer_index6; + u16 status_rx_quick_consumer_index9; + u16 status_rx_quick_consumer_index8; + u16 status_rx_quick_consumer_index11; + u16 status_rx_quick_consumer_index10; + u16 status_rx_quick_consumer_index13; + u16 status_rx_quick_consumer_index12; + u16 status_rx_quick_consumer_index15; + u16 status_rx_quick_consumer_index14; + u16 status_cmd_consumer_index; + u16 status_completion_producer_index; + u16 status_unused; + u16 status_idx; +#endif +}; + + +/* + * statistics_block definition + */ +struct statistics_block { + u32 stat_IfHCInOctets_hi; + u32 stat_IfHCInOctets_lo; + u32 stat_IfHCInBadOctets_hi; + u32 stat_IfHCInBadOctets_lo; + u32 stat_IfHCOutOctets_hi; + u32 stat_IfHCOutOctets_lo; + u32 stat_IfHCOutBadOctets_hi; + u32 stat_IfHCOutBadOctets_lo; + u32 stat_IfHCInUcastPkts_hi; + u32 stat_IfHCInUcastPkts_lo; + u32 stat_IfHCInMulticastPkts_hi; + u32 stat_IfHCInMulticastPkts_lo; + u32 stat_IfHCInBroadcastPkts_hi; + u32 stat_IfHCInBroadcastPkts_lo; + u32 stat_IfHCOutUcastPkts_hi; + u32 stat_IfHCOutUcastPkts_lo; + u32 stat_IfHCOutMulticastPkts_hi; + u32 stat_IfHCOutMulticastPkts_lo; + u32 stat_IfHCOutBroadcastPkts_hi; + u32 stat_IfHCOutBroadcastPkts_lo; + u32 stat_emac_tx_stat_dot3statsinternalmactransmiterrors; + u32 stat_Dot3StatsCarrierSenseErrors; + u32 stat_Dot3StatsFCSErrors; + u32 stat_Dot3StatsAlignmentErrors; + u32 stat_Dot3StatsSingleCollisionFrames; + u32 stat_Dot3StatsMultipleCollisionFrames; + u32 stat_Dot3StatsDeferredTransmissions; + u32 stat_Dot3StatsExcessiveCollisions; + u32 stat_Dot3StatsLateCollisions; + u32 stat_EtherStatsCollisions; + u32 stat_EtherStatsFragments; + u32 stat_EtherStatsJabbers; + u32 stat_EtherStatsUndersizePkts; + u32 stat_EtherStatsOverrsizePkts; + u32 stat_EtherStatsPktsRx64Octets; + u32 stat_EtherStatsPktsRx65Octetsto127Octets; + u32 stat_EtherStatsPktsRx128Octetsto255Octets; + u32 stat_EtherStatsPktsRx256Octetsto511Octets; + u32 stat_EtherStatsPktsRx512Octetsto1023Octets; + u32 stat_EtherStatsPktsRx1024Octetsto1522Octets; + u32 stat_EtherStatsPktsRx1523Octetsto9022Octets; + u32 stat_EtherStatsPktsTx64Octets; + u32 stat_EtherStatsPktsTx65Octetsto127Octets; + u32 stat_EtherStatsPktsTx128Octetsto255Octets; + u32 stat_EtherStatsPktsTx256Octetsto511Octets; + u32 stat_EtherStatsPktsTx512Octetsto1023Octets; + u32 stat_EtherStatsPktsTx1024Octetsto1522Octets; + u32 stat_EtherStatsPktsTx1523Octetsto9022Octets; + u32 stat_XonPauseFramesReceived; + u32 stat_XoffPauseFramesReceived; + u32 stat_OutXonSent; + u32 stat_OutXoffSent; + u32 stat_FlowControlDone; + u32 stat_MacControlFramesReceived; + u32 stat_XoffStateEntered; + u32 stat_IfInFramesL2FilterDiscards; + u32 stat_IfInRuleCheckerDiscards; + u32 stat_IfInFTQDiscards; + u32 stat_IfInMBUFDiscards; + u32 stat_IfInRuleCheckerP4Hit; + u32 stat_CatchupInRuleCheckerDiscards; + u32 stat_CatchupInFTQDiscards; + u32 stat_CatchupInMBUFDiscards; + u32 stat_CatchupInRuleCheckerP4Hit; + u32 stat_GenStat00; + u32 stat_GenStat01; + u32 stat_GenStat02; + u32 stat_GenStat03; + u32 stat_GenStat04; + u32 stat_GenStat05; + u32 stat_GenStat06; + u32 stat_GenStat07; + u32 stat_GenStat08; + u32 stat_GenStat09; + u32 stat_GenStat10; + u32 stat_GenStat11; + u32 stat_GenStat12; + u32 stat_GenStat13; + u32 stat_GenStat14; + u32 stat_GenStat15; +}; + + +/* + * l2_fhdr definition + */ +struct l2_fhdr { +#if defined(__BIG_ENDIAN) + u16 l2_fhdr_errors; + u16 l2_fhdr_status; +#elif defined(__LITTLE_ENDIAN) + u16 l2_fhdr_status; + u16 l2_fhdr_errors; +#endif + #define L2_FHDR_ERRORS_BAD_CRC (1<<1) + #define L2_FHDR_ERRORS_PHY_DECODE (1<<2) + #define L2_FHDR_ERRORS_ALIGNMENT (1<<3) + #define L2_FHDR_ERRORS_TOO_SHORT (1<<4) + #define L2_FHDR_ERRORS_GIANT_FRAME (1<<5) + + #define L2_FHDR_STATUS_RULE_CLASS (0x7<<0) + #define L2_FHDR_STATUS_RULE_P2 (1<<3) + #define L2_FHDR_STATUS_RULE_P3 (1<<4) + #define L2_FHDR_STATUS_RULE_P4 (1<<5) + #define L2_FHDR_STATUS_L2_VLAN_TAG (1<<6) + #define L2_FHDR_STATUS_L2_LLC_SNAP (1<<7) + #define L2_FHDR_STATUS_RSS_HASH (1<<8) + #define L2_FHDR_STATUS_IP_DATAGRAM (1<<13) + #define L2_FHDR_STATUS_TCP_SEGMENT (1<<14) + #define L2_FHDR_STATUS_UDP_DATAGRAM (1<<15) + + u32 l2_fhdr_hash; +#if defined(__BIG_ENDIAN) + u16 l2_fhdr_pkt_len; + u16 l2_fhdr_vlan_tag; + u16 l2_fhdr_ip_xsum; + u16 l2_fhdr_tcp_udp_xsum; +#elif defined(__LITTLE_ENDIAN) + u16 l2_fhdr_vlan_tag; + u16 l2_fhdr_pkt_len; + u16 l2_fhdr_tcp_udp_xsum; + u16 l2_fhdr_ip_xsum; +#endif +}; + + +/* + * l2_context definition + */ +#define BNX2_L2CTX_TYPE 0x00000000 +#define BNX2_L2CTX_TYPE_SIZE_L2 ((0xc0/0x20)<<16) +#define BNX2_L2CTX_TYPE_TYPE (0xf<<28) +#define BNX2_L2CTX_TYPE_TYPE_EMPTY (0<<28) +#define BNX2_L2CTX_TYPE_TYPE_L2 (1<<28) + +#define BNX2_L2CTX_TX_HOST_BIDX 0x00000088 +#define BNX2_L2CTX_EST_NBD 0x00000088 +#define BNX2_L2CTX_CMD_TYPE 0x00000088 +#define BNX2_L2CTX_CMD_TYPE_TYPE (0xf<<24) +#define BNX2_L2CTX_CMD_TYPE_TYPE_L2 (0<<24) +#define BNX2_L2CTX_CMD_TYPE_TYPE_TCP (1<<24) + +#define BNX2_L2CTX_TX_HOST_BSEQ 0x00000090 +#define BNX2_L2CTX_TSCH_BSEQ 0x00000094 +#define BNX2_L2CTX_TBDR_BSEQ 0x00000098 +#define BNX2_L2CTX_TBDR_BOFF 0x0000009c +#define BNX2_L2CTX_TBDR_BIDX 0x0000009c +#define BNX2_L2CTX_TBDR_BHADDR_HI 0x000000a0 +#define BNX2_L2CTX_TBDR_BHADDR_LO 0x000000a4 +#define BNX2_L2CTX_TXP_BOFF 0x000000a8 +#define BNX2_L2CTX_TXP_BIDX 0x000000a8 +#define BNX2_L2CTX_TXP_BSEQ 0x000000ac + + +/* + * l2_bd_chain_context definition + */ +#define BNX2_L2CTX_BD_PRE_READ 0x00000000 +#define BNX2_L2CTX_CTX_SIZE 0x00000000 +#define BNX2_L2CTX_CTX_TYPE 0x00000000 +#define BNX2_L2CTX_CTX_TYPE_SIZE_L2 ((0x20/20)<<16) +#define BNX2_L2CTX_CTX_TYPE_CTX_BD_CHN_TYPE (0xf<<28) +#define BNX2_L2CTX_CTX_TYPE_CTX_BD_CHN_TYPE_UNDEFINED (0<<28) +#define BNX2_L2CTX_CTX_TYPE_CTX_BD_CHN_TYPE_VALUE (1<<28) + +#define BNX2_L2CTX_HOST_BDIDX 0x00000004 +#define BNX2_L2CTX_HOST_BSEQ 0x00000008 +#define BNX2_L2CTX_NX_BSEQ 0x0000000c +#define BNX2_L2CTX_NX_BDHADDR_HI 0x00000010 +#define BNX2_L2CTX_NX_BDHADDR_LO 0x00000014 +#define BNX2_L2CTX_NX_BDIDX 0x00000018 + + +/* + * pci_config_l definition + * offset: 0000 + */ +#define BNX2_PCICFG_MISC_CONFIG 0x00000068 +#define BNX2_PCICFG_MISC_CONFIG_TARGET_BYTE_SWAP (1L<<2) +#define BNX2_PCICFG_MISC_CONFIG_TARGET_MB_WORD_SWAP (1L<<3) +#define BNX2_PCICFG_MISC_CONFIG_CLOCK_CTL_ENA (1L<<5) +#define BNX2_PCICFG_MISC_CONFIG_TARGET_GRC_WORD_SWAP (1L<<6) +#define BNX2_PCICFG_MISC_CONFIG_REG_WINDOW_ENA (1L<<7) +#define BNX2_PCICFG_MISC_CONFIG_CORE_RST_REQ (1L<<8) +#define BNX2_PCICFG_MISC_CONFIG_CORE_RST_BSY (1L<<9) +#define BNX2_PCICFG_MISC_CONFIG_ASIC_METAL_REV (0xffL<<16) +#define BNX2_PCICFG_MISC_CONFIG_ASIC_BASE_REV (0xfL<<24) +#define BNX2_PCICFG_MISC_CONFIG_ASIC_ID (0xfL<<28) + +#define BNX2_PCICFG_MISC_STATUS 0x0000006c +#define BNX2_PCICFG_MISC_STATUS_INTA_VALUE (1L<<0) +#define BNX2_PCICFG_MISC_STATUS_32BIT_DET (1L<<1) +#define BNX2_PCICFG_MISC_STATUS_M66EN (1L<<2) +#define BNX2_PCICFG_MISC_STATUS_PCIX_DET (1L<<3) +#define BNX2_PCICFG_MISC_STATUS_PCIX_SPEED (0x3L<<4) +#define BNX2_PCICFG_MISC_STATUS_PCIX_SPEED_66 (0L<<4) +#define BNX2_PCICFG_MISC_STATUS_PCIX_SPEED_100 (1L<<4) +#define BNX2_PCICFG_MISC_STATUS_PCIX_SPEED_133 (2L<<4) +#define BNX2_PCICFG_MISC_STATUS_PCIX_SPEED_PCI_MODE (3L<<4) + +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS 0x00000070 +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET (0xfL<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_32MHZ (0L<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_38MHZ (1L<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_48MHZ (2L<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_55MHZ (3L<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_66MHZ (4L<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_80MHZ (5L<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_95MHZ (6L<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_133MHZ (7L<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_LOW (0xfL<<0) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_DISABLE (1L<<6) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_ALT (1L<<7) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC (0x7L<<8) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC_UNDEF (0L<<8) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC_12 (1L<<8) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC_6 (2L<<8) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC_62 (4L<<8) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PLAY_DEAD (1L<<11) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED (0xfL<<12) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_100 (0L<<12) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_80 (1L<<12) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_50 (2L<<12) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_40 (4L<<12) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_25 (8L<<12) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_CORE_CLK_PLL_STOP (1L<<16) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_PCI_PLL_STOP (1L<<17) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_RESERVED_18 (1L<<18) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_USE_SPD_DET (1L<<19) +#define BNX2_PCICFG_PCI_CLOCK_CONTROL_BITS_RESERVED (0xfffL<<20) + +#define BNX2_PCICFG_REG_WINDOW_ADDRESS 0x00000078 +#define BNX2_PCICFG_REG_WINDOW 0x00000080 +#define BNX2_PCICFG_INT_ACK_CMD 0x00000084 +#define BNX2_PCICFG_INT_ACK_CMD_INDEX (0xffffL<<0) +#define BNX2_PCICFG_INT_ACK_CMD_INDEX_VALID (1L<<16) +#define BNX2_PCICFG_INT_ACK_CMD_USE_INT_HC_PARAM (1L<<17) +#define BNX2_PCICFG_INT_ACK_CMD_MASK_INT (1L<<18) + +#define BNX2_PCICFG_STATUS_BIT_SET_CMD 0x00000088 +#define BNX2_PCICFG_STATUS_BIT_CLEAR_CMD 0x0000008c +#define BNX2_PCICFG_MAILBOX_QUEUE_ADDR 0x00000090 +#define BNX2_PCICFG_MAILBOX_QUEUE_DATA 0x00000094 + + +/* + * pci_reg definition + * offset: 0x400 + */ +#define BNX2_PCI_GRC_WINDOW_ADDR 0x00000400 +#define BNX2_PCI_GRC_WINDOW_ADDR_PCI_GRC_WINDOW_ADDR_VALUE (0x3ffffL<<8) + +#define BNX2_PCI_CONFIG_1 0x00000404 +#define BNX2_PCI_CONFIG_1_READ_BOUNDARY (0x7L<<8) +#define BNX2_PCI_CONFIG_1_READ_BOUNDARY_OFF (0L<<8) +#define BNX2_PCI_CONFIG_1_READ_BOUNDARY_16 (1L<<8) +#define BNX2_PCI_CONFIG_1_READ_BOUNDARY_32 (2L<<8) +#define BNX2_PCI_CONFIG_1_READ_BOUNDARY_64 (3L<<8) +#define BNX2_PCI_CONFIG_1_READ_BOUNDARY_128 (4L<<8) +#define BNX2_PCI_CONFIG_1_READ_BOUNDARY_256 (5L<<8) +#define BNX2_PCI_CONFIG_1_READ_BOUNDARY_512 (6L<<8) +#define BNX2_PCI_CONFIG_1_READ_BOUNDARY_1024 (7L<<8) +#define BNX2_PCI_CONFIG_1_WRITE_BOUNDARY (0x7L<<11) +#define BNX2_PCI_CONFIG_1_WRITE_BOUNDARY_OFF (0L<<11) +#define BNX2_PCI_CONFIG_1_WRITE_BOUNDARY_16 (1L<<11) +#define BNX2_PCI_CONFIG_1_WRITE_BOUNDARY_32 (2L<<11) +#define BNX2_PCI_CONFIG_1_WRITE_BOUNDARY_64 (3L<<11) +#define BNX2_PCI_CONFIG_1_WRITE_BOUNDARY_128 (4L<<11) +#define BNX2_PCI_CONFIG_1_WRITE_BOUNDARY_256 (5L<<11) +#define BNX2_PCI_CONFIG_1_WRITE_BOUNDARY_512 (6L<<11) +#define BNX2_PCI_CONFIG_1_WRITE_BOUNDARY_1024 (7L<<11) + +#define BNX2_PCI_CONFIG_2 0x00000408 +#define BNX2_PCI_CONFIG_2_BAR1_SIZE (0xfL<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_DISABLED (0L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_64K (1L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_128K (2L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_256K (3L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_512K (4L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_1M (5L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_2M (6L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_4M (7L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_8M (8L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_16M (9L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_32M (10L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_64M (11L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_128M (12L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_256M (13L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_512M (14L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_SIZE_1G (15L<<0) +#define BNX2_PCI_CONFIG_2_BAR1_64ENA (1L<<4) +#define BNX2_PCI_CONFIG_2_EXP_ROM_RETRY (1L<<5) +#define BNX2_PCI_CONFIG_2_CFG_CYCLE_RETRY (1L<<6) +#define BNX2_PCI_CONFIG_2_FIRST_CFG_DONE (1L<<7) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE (0xffL<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_DISABLED (0L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_1K (1L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_2K (2L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_4K (3L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_8K (4L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_16K (5L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_32K (6L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_64K (7L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_128K (8L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_256K (9L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_512K (10L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_1M (11L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_2M (12L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_4M (13L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_8M (14L<<8) +#define BNX2_PCI_CONFIG_2_EXP_ROM_SIZE_16M (15L<<8) +#define BNX2_PCI_CONFIG_2_MAX_SPLIT_LIMIT (0x1fL<<16) +#define BNX2_PCI_CONFIG_2_MAX_READ_LIMIT (0x3L<<21) +#define BNX2_PCI_CONFIG_2_MAX_READ_LIMIT_512 (0L<<21) +#define BNX2_PCI_CONFIG_2_MAX_READ_LIMIT_1K (1L<<21) +#define BNX2_PCI_CONFIG_2_MAX_READ_LIMIT_2K (2L<<21) +#define BNX2_PCI_CONFIG_2_MAX_READ_LIMIT_4K (3L<<21) +#define BNX2_PCI_CONFIG_2_FORCE_32_BIT_MSTR (1L<<23) +#define BNX2_PCI_CONFIG_2_FORCE_32_BIT_TGT (1L<<24) +#define BNX2_PCI_CONFIG_2_KEEP_REQ_ASSERT (1L<<25) + +#define BNX2_PCI_CONFIG_3 0x0000040c +#define BNX2_PCI_CONFIG_3_STICKY_BYTE (0xffL<<0) +#define BNX2_PCI_CONFIG_3_FORCE_PME (1L<<24) +#define BNX2_PCI_CONFIG_3_PME_STATUS (1L<<25) +#define BNX2_PCI_CONFIG_3_PME_ENABLE (1L<<26) +#define BNX2_PCI_CONFIG_3_PM_STATE (0x3L<<27) +#define BNX2_PCI_CONFIG_3_VAUX_PRESET (1L<<30) +#define BNX2_PCI_CONFIG_3_PCI_POWER (1L<<31) + +#define BNX2_PCI_PM_DATA_A 0x00000410 +#define BNX2_PCI_PM_DATA_A_PM_DATA_0_PRG (0xffL<<0) +#define BNX2_PCI_PM_DATA_A_PM_DATA_1_PRG (0xffL<<8) +#define BNX2_PCI_PM_DATA_A_PM_DATA_2_PRG (0xffL<<16) +#define BNX2_PCI_PM_DATA_A_PM_DATA_3_PRG (0xffL<<24) + +#define BNX2_PCI_PM_DATA_B 0x00000414 +#define BNX2_PCI_PM_DATA_B_PM_DATA_4_PRG (0xffL<<0) +#define BNX2_PCI_PM_DATA_B_PM_DATA_5_PRG (0xffL<<8) +#define BNX2_PCI_PM_DATA_B_PM_DATA_6_PRG (0xffL<<16) +#define BNX2_PCI_PM_DATA_B_PM_DATA_7_PRG (0xffL<<24) + +#define BNX2_PCI_SWAP_DIAG0 0x00000418 +#define BNX2_PCI_SWAP_DIAG1 0x0000041c +#define BNX2_PCI_EXP_ROM_ADDR 0x00000420 +#define BNX2_PCI_EXP_ROM_ADDR_ADDRESS (0x3fffffL<<2) +#define BNX2_PCI_EXP_ROM_ADDR_REQ (1L<<31) + +#define BNX2_PCI_EXP_ROM_DATA 0x00000424 +#define BNX2_PCI_VPD_INTF 0x00000428 +#define BNX2_PCI_VPD_INTF_INTF_REQ (1L<<0) + +#define BNX2_PCI_VPD_ADDR_FLAG 0x0000042c +#define BNX2_PCI_VPD_ADDR_FLAG_ADDRESS (0x1fff<<2) +#define BNX2_PCI_VPD_ADDR_FLAG_WR (1<<15) + +#define BNX2_PCI_VPD_DATA 0x00000430 +#define BNX2_PCI_ID_VAL1 0x00000434 +#define BNX2_PCI_ID_VAL1_DEVICE_ID (0xffffL<<0) +#define BNX2_PCI_ID_VAL1_VENDOR_ID (0xffffL<<16) + +#define BNX2_PCI_ID_VAL2 0x00000438 +#define BNX2_PCI_ID_VAL2_SUBSYSTEM_VENDOR_ID (0xffffL<<0) +#define BNX2_PCI_ID_VAL2_SUBSYSTEM_ID (0xffffL<<16) + +#define BNX2_PCI_ID_VAL3 0x0000043c +#define BNX2_PCI_ID_VAL3_CLASS_CODE (0xffffffL<<0) +#define BNX2_PCI_ID_VAL3_REVISION_ID (0xffL<<24) + +#define BNX2_PCI_ID_VAL4 0x00000440 +#define BNX2_PCI_ID_VAL4_CAP_ENA (0xfL<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_0 (0L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_1 (1L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_2 (2L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_3 (3L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_4 (4L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_5 (5L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_6 (6L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_7 (7L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_8 (8L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_9 (9L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_10 (10L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_11 (11L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_12 (12L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_13 (13L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_14 (14L<<0) +#define BNX2_PCI_ID_VAL4_CAP_ENA_15 (15L<<0) +#define BNX2_PCI_ID_VAL4_PM_SCALE_PRG (0x3L<<6) +#define BNX2_PCI_ID_VAL4_PM_SCALE_PRG_0 (0L<<6) +#define BNX2_PCI_ID_VAL4_PM_SCALE_PRG_1 (1L<<6) +#define BNX2_PCI_ID_VAL4_PM_SCALE_PRG_2 (2L<<6) +#define BNX2_PCI_ID_VAL4_PM_SCALE_PRG_3 (3L<<6) +#define BNX2_PCI_ID_VAL4_MSI_LIMIT (0x7L<<9) +#define BNX2_PCI_ID_VAL4_MSI_ADVERTIZE (0x7L<<12) +#define BNX2_PCI_ID_VAL4_MSI_ENABLE (1L<<15) +#define BNX2_PCI_ID_VAL4_MAX_64_ADVERTIZE (1L<<16) +#define BNX2_PCI_ID_VAL4_MAX_133_ADVERTIZE (1L<<17) +#define BNX2_PCI_ID_VAL4_MAX_MEM_READ_SIZE (0x3L<<21) +#define BNX2_PCI_ID_VAL4_MAX_SPLIT_SIZE (0x7L<<23) +#define BNX2_PCI_ID_VAL4_MAX_CUMULATIVE_SIZE (0x7L<<26) + +#define BNX2_PCI_ID_VAL5 0x00000444 +#define BNX2_PCI_ID_VAL5_D1_SUPPORT (1L<<0) +#define BNX2_PCI_ID_VAL5_D2_SUPPORT (1L<<1) +#define BNX2_PCI_ID_VAL5_PME_IN_D0 (1L<<2) +#define BNX2_PCI_ID_VAL5_PME_IN_D1 (1L<<3) +#define BNX2_PCI_ID_VAL5_PME_IN_D2 (1L<<4) +#define BNX2_PCI_ID_VAL5_PME_IN_D3_HOT (1L<<5) + +#define BNX2_PCI_PCIX_EXTENDED_STATUS 0x00000448 +#define BNX2_PCI_PCIX_EXTENDED_STATUS_NO_SNOOP (1L<<8) +#define BNX2_PCI_PCIX_EXTENDED_STATUS_LONG_BURST (1L<<9) +#define BNX2_PCI_PCIX_EXTENDED_STATUS_SPLIT_COMP_MSG_CLASS (0xfL<<16) +#define BNX2_PCI_PCIX_EXTENDED_STATUS_SPLIT_COMP_MSG_IDX (0xffL<<24) + +#define BNX2_PCI_ID_VAL6 0x0000044c +#define BNX2_PCI_ID_VAL6_MAX_LAT (0xffL<<0) +#define BNX2_PCI_ID_VAL6_MIN_GNT (0xffL<<8) +#define BNX2_PCI_ID_VAL6_BIST (0xffL<<16) + +#define BNX2_PCI_MSI_DATA 0x00000450 +#define BNX2_PCI_MSI_DATA_PCI_MSI_DATA (0xffffL<<0) + +#define BNX2_PCI_MSI_ADDR_H 0x00000454 +#define BNX2_PCI_MSI_ADDR_L 0x00000458 + + +/* + * misc_reg definition + * offset: 0x800 + */ +#define BNX2_MISC_COMMAND 0x00000800 +#define BNX2_MISC_COMMAND_ENABLE_ALL (1L<<0) +#define BNX2_MISC_COMMAND_DISABLE_ALL (1L<<1) +#define BNX2_MISC_COMMAND_CORE_RESET (1L<<4) +#define BNX2_MISC_COMMAND_HARD_RESET (1L<<5) +#define BNX2_MISC_COMMAND_PAR_ERROR (1L<<8) +#define BNX2_MISC_COMMAND_PAR_ERR_RAM (0x7fL<<16) + +#define BNX2_MISC_CFG 0x00000804 +#define BNX2_MISC_CFG_PCI_GRC_TMOUT (1L<<0) +#define BNX2_MISC_CFG_NVM_WR_EN (0x3L<<1) +#define BNX2_MISC_CFG_NVM_WR_EN_PROTECT (0L<<1) +#define BNX2_MISC_CFG_NVM_WR_EN_PCI (1L<<1) +#define BNX2_MISC_CFG_NVM_WR_EN_ALLOW (2L<<1) +#define BNX2_MISC_CFG_NVM_WR_EN_ALLOW2 (3L<<1) +#define BNX2_MISC_CFG_BIST_EN (1L<<3) +#define BNX2_MISC_CFG_CK25_OUT_ALT_SRC (1L<<4) +#define BNX2_MISC_CFG_BYPASS_BSCAN (1L<<5) +#define BNX2_MISC_CFG_BYPASS_EJTAG (1L<<6) +#define BNX2_MISC_CFG_CLK_CTL_OVERRIDE (1L<<7) +#define BNX2_MISC_CFG_LEDMODE (0x3L<<8) +#define BNX2_MISC_CFG_LEDMODE_MAC (0L<<8) +#define BNX2_MISC_CFG_LEDMODE_GPHY1 (1L<<8) +#define BNX2_MISC_CFG_LEDMODE_GPHY2 (2L<<8) + +#define BNX2_MISC_ID 0x00000808 +#define BNX2_MISC_ID_BOND_ID (0xfL<<0) +#define BNX2_MISC_ID_CHIP_METAL (0xffL<<4) +#define BNX2_MISC_ID_CHIP_REV (0xfL<<12) +#define BNX2_MISC_ID_CHIP_NUM (0xffffL<<16) + +#define BNX2_MISC_ENABLE_STATUS_BITS 0x0000080c +#define BNX2_MISC_ENABLE_STATUS_BITS_TX_SCHEDULER_ENABLE (1L<<0) +#define BNX2_MISC_ENABLE_STATUS_BITS_TX_BD_READ_ENABLE (1L<<1) +#define BNX2_MISC_ENABLE_STATUS_BITS_TX_BD_CACHE_ENABLE (1L<<2) +#define BNX2_MISC_ENABLE_STATUS_BITS_TX_PROCESSOR_ENABLE (1L<<3) +#define BNX2_MISC_ENABLE_STATUS_BITS_TX_DMA_ENABLE (1L<<4) +#define BNX2_MISC_ENABLE_STATUS_BITS_TX_PATCHUP_ENABLE (1L<<5) +#define BNX2_MISC_ENABLE_STATUS_BITS_TX_PAYLOAD_Q_ENABLE (1L<<6) +#define BNX2_MISC_ENABLE_STATUS_BITS_TX_HEADER_Q_ENABLE (1L<<7) +#define BNX2_MISC_ENABLE_STATUS_BITS_TX_ASSEMBLER_ENABLE (1L<<8) +#define BNX2_MISC_ENABLE_STATUS_BITS_EMAC_ENABLE (1L<<9) +#define BNX2_MISC_ENABLE_STATUS_BITS_RX_PARSER_MAC_ENABLE (1L<<10) +#define BNX2_MISC_ENABLE_STATUS_BITS_RX_PARSER_CATCHUP_ENABLE (1L<<11) +#define BNX2_MISC_ENABLE_STATUS_BITS_RX_MBUF_ENABLE (1L<<12) +#define BNX2_MISC_ENABLE_STATUS_BITS_RX_LOOKUP_ENABLE (1L<<13) +#define BNX2_MISC_ENABLE_STATUS_BITS_RX_PROCESSOR_ENABLE (1L<<14) +#define BNX2_MISC_ENABLE_STATUS_BITS_RX_V2P_ENABLE (1L<<15) +#define BNX2_MISC_ENABLE_STATUS_BITS_RX_BD_CACHE_ENABLE (1L<<16) +#define BNX2_MISC_ENABLE_STATUS_BITS_RX_DMA_ENABLE (1L<<17) +#define BNX2_MISC_ENABLE_STATUS_BITS_COMPLETION_ENABLE (1L<<18) +#define BNX2_MISC_ENABLE_STATUS_BITS_HOST_COALESCE_ENABLE (1L<<19) +#define BNX2_MISC_ENABLE_STATUS_BITS_MAILBOX_QUEUE_ENABLE (1L<<20) +#define BNX2_MISC_ENABLE_STATUS_BITS_CONTEXT_ENABLE (1L<<21) +#define BNX2_MISC_ENABLE_STATUS_BITS_CMD_SCHEDULER_ENABLE (1L<<22) +#define BNX2_MISC_ENABLE_STATUS_BITS_CMD_PROCESSOR_ENABLE (1L<<23) +#define BNX2_MISC_ENABLE_STATUS_BITS_MGMT_PROCESSOR_ENABLE (1L<<24) +#define BNX2_MISC_ENABLE_STATUS_BITS_TIMER_ENABLE (1L<<25) +#define BNX2_MISC_ENABLE_STATUS_BITS_DMA_ENGINE_ENABLE (1L<<26) +#define BNX2_MISC_ENABLE_STATUS_BITS_UMP_ENABLE (1L<<27) + +#define BNX2_MISC_ENABLE_SET_BITS 0x00000810 +#define BNX2_MISC_ENABLE_SET_BITS_TX_SCHEDULER_ENABLE (1L<<0) +#define BNX2_MISC_ENABLE_SET_BITS_TX_BD_READ_ENABLE (1L<<1) +#define BNX2_MISC_ENABLE_SET_BITS_TX_BD_CACHE_ENABLE (1L<<2) +#define BNX2_MISC_ENABLE_SET_BITS_TX_PROCESSOR_ENABLE (1L<<3) +#define BNX2_MISC_ENABLE_SET_BITS_TX_DMA_ENABLE (1L<<4) +#define BNX2_MISC_ENABLE_SET_BITS_TX_PATCHUP_ENABLE (1L<<5) +#define BNX2_MISC_ENABLE_SET_BITS_TX_PAYLOAD_Q_ENABLE (1L<<6) +#define BNX2_MISC_ENABLE_SET_BITS_TX_HEADER_Q_ENABLE (1L<<7) +#define BNX2_MISC_ENABLE_SET_BITS_TX_ASSEMBLER_ENABLE (1L<<8) +#define BNX2_MISC_ENABLE_SET_BITS_EMAC_ENABLE (1L<<9) +#define BNX2_MISC_ENABLE_SET_BITS_RX_PARSER_MAC_ENABLE (1L<<10) +#define BNX2_MISC_ENABLE_SET_BITS_RX_PARSER_CATCHUP_ENABLE (1L<<11) +#define BNX2_MISC_ENABLE_SET_BITS_RX_MBUF_ENABLE (1L<<12) +#define BNX2_MISC_ENABLE_SET_BITS_RX_LOOKUP_ENABLE (1L<<13) +#define BNX2_MISC_ENABLE_SET_BITS_RX_PROCESSOR_ENABLE (1L<<14) +#define BNX2_MISC_ENABLE_SET_BITS_RX_V2P_ENABLE (1L<<15) +#define BNX2_MISC_ENABLE_SET_BITS_RX_BD_CACHE_ENABLE (1L<<16) +#define BNX2_MISC_ENABLE_SET_BITS_RX_DMA_ENABLE (1L<<17) +#define BNX2_MISC_ENABLE_SET_BITS_COMPLETION_ENABLE (1L<<18) +#define BNX2_MISC_ENABLE_SET_BITS_HOST_COALESCE_ENABLE (1L<<19) +#define BNX2_MISC_ENABLE_SET_BITS_MAILBOX_QUEUE_ENABLE (1L<<20) +#define BNX2_MISC_ENABLE_SET_BITS_CONTEXT_ENABLE (1L<<21) +#define BNX2_MISC_ENABLE_SET_BITS_CMD_SCHEDULER_ENABLE (1L<<22) +#define BNX2_MISC_ENABLE_SET_BITS_CMD_PROCESSOR_ENABLE (1L<<23) +#define BNX2_MISC_ENABLE_SET_BITS_MGMT_PROCESSOR_ENABLE (1L<<24) +#define BNX2_MISC_ENABLE_SET_BITS_TIMER_ENABLE (1L<<25) +#define BNX2_MISC_ENABLE_SET_BITS_DMA_ENGINE_ENABLE (1L<<26) +#define BNX2_MISC_ENABLE_SET_BITS_UMP_ENABLE (1L<<27) + +#define BNX2_MISC_ENABLE_CLR_BITS 0x00000814 +#define BNX2_MISC_ENABLE_CLR_BITS_TX_SCHEDULER_ENABLE (1L<<0) +#define BNX2_MISC_ENABLE_CLR_BITS_TX_BD_READ_ENABLE (1L<<1) +#define BNX2_MISC_ENABLE_CLR_BITS_TX_BD_CACHE_ENABLE (1L<<2) +#define BNX2_MISC_ENABLE_CLR_BITS_TX_PROCESSOR_ENABLE (1L<<3) +#define BNX2_MISC_ENABLE_CLR_BITS_TX_DMA_ENABLE (1L<<4) +#define BNX2_MISC_ENABLE_CLR_BITS_TX_PATCHUP_ENABLE (1L<<5) +#define BNX2_MISC_ENABLE_CLR_BITS_TX_PAYLOAD_Q_ENABLE (1L<<6) +#define BNX2_MISC_ENABLE_CLR_BITS_TX_HEADER_Q_ENABLE (1L<<7) +#define BNX2_MISC_ENABLE_CLR_BITS_TX_ASSEMBLER_ENABLE (1L<<8) +#define BNX2_MISC_ENABLE_CLR_BITS_EMAC_ENABLE (1L<<9) +#define BNX2_MISC_ENABLE_CLR_BITS_RX_PARSER_MAC_ENABLE (1L<<10) +#define BNX2_MISC_ENABLE_CLR_BITS_RX_PARSER_CATCHUP_ENABLE (1L<<11) +#define BNX2_MISC_ENABLE_CLR_BITS_RX_MBUF_ENABLE (1L<<12) +#define BNX2_MISC_ENABLE_CLR_BITS_RX_LOOKUP_ENABLE (1L<<13) +#define BNX2_MISC_ENABLE_CLR_BITS_RX_PROCESSOR_ENABLE (1L<<14) +#define BNX2_MISC_ENABLE_CLR_BITS_RX_V2P_ENABLE (1L<<15) +#define BNX2_MISC_ENABLE_CLR_BITS_RX_BD_CACHE_ENABLE (1L<<16) +#define BNX2_MISC_ENABLE_CLR_BITS_RX_DMA_ENABLE (1L<<17) +#define BNX2_MISC_ENABLE_CLR_BITS_COMPLETION_ENABLE (1L<<18) +#define BNX2_MISC_ENABLE_CLR_BITS_HOST_COALESCE_ENABLE (1L<<19) +#define BNX2_MISC_ENABLE_CLR_BITS_MAILBOX_QUEUE_ENABLE (1L<<20) +#define BNX2_MISC_ENABLE_CLR_BITS_CONTEXT_ENABLE (1L<<21) +#define BNX2_MISC_ENABLE_CLR_BITS_CMD_SCHEDULER_ENABLE (1L<<22) +#define BNX2_MISC_ENABLE_CLR_BITS_CMD_PROCESSOR_ENABLE (1L<<23) +#define BNX2_MISC_ENABLE_CLR_BITS_MGMT_PROCESSOR_ENABLE (1L<<24) +#define BNX2_MISC_ENABLE_CLR_BITS_TIMER_ENABLE (1L<<25) +#define BNX2_MISC_ENABLE_CLR_BITS_DMA_ENGINE_ENABLE (1L<<26) +#define BNX2_MISC_ENABLE_CLR_BITS_UMP_ENABLE (1L<<27) + +#define BNX2_MISC_CLOCK_CONTROL_BITS 0x00000818 +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET (0xfL<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_32MHZ (0L<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_38MHZ (1L<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_48MHZ (2L<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_55MHZ (3L<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_66MHZ (4L<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_80MHZ (5L<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_95MHZ (6L<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_133MHZ (7L<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_CLK_SPD_DET_LOW (0xfL<<0) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_DISABLE (1L<<6) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_ALT (1L<<7) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC (0x7L<<8) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC_UNDEF (0L<<8) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC_12 (1L<<8) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC_6 (2L<<8) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_ALT_SRC_62 (4L<<8) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PLAY_DEAD (1L<<11) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED (0xfL<<12) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_100 (0L<<12) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_80 (1L<<12) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_50 (2L<<12) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_40 (4L<<12) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_PLL_SPEED_25 (8L<<12) +#define BNX2_MISC_CLOCK_CONTROL_BITS_CORE_CLK_PLL_STOP (1L<<16) +#define BNX2_MISC_CLOCK_CONTROL_BITS_PCI_PLL_STOP (1L<<17) +#define BNX2_MISC_CLOCK_CONTROL_BITS_RESERVED_18 (1L<<18) +#define BNX2_MISC_CLOCK_CONTROL_BITS_USE_SPD_DET (1L<<19) +#define BNX2_MISC_CLOCK_CONTROL_BITS_RESERVED (0xfffL<<20) + +#define BNX2_MISC_GPIO 0x0000081c +#define BNX2_MISC_GPIO_VALUE (0xffL<<0) +#define BNX2_MISC_GPIO_SET (0xffL<<8) +#define BNX2_MISC_GPIO_CLR (0xffL<<16) +#define BNX2_MISC_GPIO_FLOAT (0xffL<<24) + +#define BNX2_MISC_GPIO_INT 0x00000820 +#define BNX2_MISC_GPIO_INT_INT_STATE (0xfL<<0) +#define BNX2_MISC_GPIO_INT_OLD_VALUE (0xfL<<8) +#define BNX2_MISC_GPIO_INT_OLD_SET (0xfL<<16) +#define BNX2_MISC_GPIO_INT_OLD_CLR (0xfL<<24) + +#define BNX2_MISC_CONFIG_LFSR 0x00000824 +#define BNX2_MISC_CONFIG_LFSR_DIV (0xffffL<<0) + +#define BNX2_MISC_LFSR_MASK_BITS 0x00000828 +#define BNX2_MISC_LFSR_MASK_BITS_TX_SCHEDULER_ENABLE (1L<<0) +#define BNX2_MISC_LFSR_MASK_BITS_TX_BD_READ_ENABLE (1L<<1) +#define BNX2_MISC_LFSR_MASK_BITS_TX_BD_CACHE_ENABLE (1L<<2) +#define BNX2_MISC_LFSR_MASK_BITS_TX_PROCESSOR_ENABLE (1L<<3) +#define BNX2_MISC_LFSR_MASK_BITS_TX_DMA_ENABLE (1L<<4) +#define BNX2_MISC_LFSR_MASK_BITS_TX_PATCHUP_ENABLE (1L<<5) +#define BNX2_MISC_LFSR_MASK_BITS_TX_PAYLOAD_Q_ENABLE (1L<<6) +#define BNX2_MISC_LFSR_MASK_BITS_TX_HEADER_Q_ENABLE (1L<<7) +#define BNX2_MISC_LFSR_MASK_BITS_TX_ASSEMBLER_ENABLE (1L<<8) +#define BNX2_MISC_LFSR_MASK_BITS_EMAC_ENABLE (1L<<9) +#define BNX2_MISC_LFSR_MASK_BITS_RX_PARSER_MAC_ENABLE (1L<<10) +#define BNX2_MISC_LFSR_MASK_BITS_RX_PARSER_CATCHUP_ENABLE (1L<<11) +#define BNX2_MISC_LFSR_MASK_BITS_RX_MBUF_ENABLE (1L<<12) +#define BNX2_MISC_LFSR_MASK_BITS_RX_LOOKUP_ENABLE (1L<<13) +#define BNX2_MISC_LFSR_MASK_BITS_RX_PROCESSOR_ENABLE (1L<<14) +#define BNX2_MISC_LFSR_MASK_BITS_RX_V2P_ENABLE (1L<<15) +#define BNX2_MISC_LFSR_MASK_BITS_RX_BD_CACHE_ENABLE (1L<<16) +#define BNX2_MISC_LFSR_MASK_BITS_RX_DMA_ENABLE (1L<<17) +#define BNX2_MISC_LFSR_MASK_BITS_COMPLETION_ENABLE (1L<<18) +#define BNX2_MISC_LFSR_MASK_BITS_HOST_COALESCE_ENABLE (1L<<19) +#define BNX2_MISC_LFSR_MASK_BITS_MAILBOX_QUEUE_ENABLE (1L<<20) +#define BNX2_MISC_LFSR_MASK_BITS_CONTEXT_ENABLE (1L<<21) +#define BNX2_MISC_LFSR_MASK_BITS_CMD_SCHEDULER_ENABLE (1L<<22) +#define BNX2_MISC_LFSR_MASK_BITS_CMD_PROCESSOR_ENABLE (1L<<23) +#define BNX2_MISC_LFSR_MASK_BITS_MGMT_PROCESSOR_ENABLE (1L<<24) +#define BNX2_MISC_LFSR_MASK_BITS_TIMER_ENABLE (1L<<25) +#define BNX2_MISC_LFSR_MASK_BITS_DMA_ENGINE_ENABLE (1L<<26) +#define BNX2_MISC_LFSR_MASK_BITS_UMP_ENABLE (1L<<27) + +#define BNX2_MISC_ARB_REQ0 0x0000082c +#define BNX2_MISC_ARB_REQ1 0x00000830 +#define BNX2_MISC_ARB_REQ2 0x00000834 +#define BNX2_MISC_ARB_REQ3 0x00000838 +#define BNX2_MISC_ARB_REQ4 0x0000083c +#define BNX2_MISC_ARB_FREE0 0x00000840 +#define BNX2_MISC_ARB_FREE1 0x00000844 +#define BNX2_MISC_ARB_FREE2 0x00000848 +#define BNX2_MISC_ARB_FREE3 0x0000084c +#define BNX2_MISC_ARB_FREE4 0x00000850 +#define BNX2_MISC_ARB_REQ_STATUS0 0x00000854 +#define BNX2_MISC_ARB_REQ_STATUS1 0x00000858 +#define BNX2_MISC_ARB_REQ_STATUS2 0x0000085c +#define BNX2_MISC_ARB_REQ_STATUS3 0x00000860 +#define BNX2_MISC_ARB_REQ_STATUS4 0x00000864 +#define BNX2_MISC_ARB_GNT0 0x00000868 +#define BNX2_MISC_ARB_GNT0_0 (0x7L<<0) +#define BNX2_MISC_ARB_GNT0_1 (0x7L<<4) +#define BNX2_MISC_ARB_GNT0_2 (0x7L<<8) +#define BNX2_MISC_ARB_GNT0_3 (0x7L<<12) +#define BNX2_MISC_ARB_GNT0_4 (0x7L<<16) +#define BNX2_MISC_ARB_GNT0_5 (0x7L<<20) +#define BNX2_MISC_ARB_GNT0_6 (0x7L<<24) +#define BNX2_MISC_ARB_GNT0_7 (0x7L<<28) + +#define BNX2_MISC_ARB_GNT1 0x0000086c +#define BNX2_MISC_ARB_GNT1_8 (0x7L<<0) +#define BNX2_MISC_ARB_GNT1_9 (0x7L<<4) +#define BNX2_MISC_ARB_GNT1_10 (0x7L<<8) +#define BNX2_MISC_ARB_GNT1_11 (0x7L<<12) +#define BNX2_MISC_ARB_GNT1_12 (0x7L<<16) +#define BNX2_MISC_ARB_GNT1_13 (0x7L<<20) +#define BNX2_MISC_ARB_GNT1_14 (0x7L<<24) +#define BNX2_MISC_ARB_GNT1_15 (0x7L<<28) + +#define BNX2_MISC_ARB_GNT2 0x00000870 +#define BNX2_MISC_ARB_GNT2_16 (0x7L<<0) +#define BNX2_MISC_ARB_GNT2_17 (0x7L<<4) +#define BNX2_MISC_ARB_GNT2_18 (0x7L<<8) +#define BNX2_MISC_ARB_GNT2_19 (0x7L<<12) +#define BNX2_MISC_ARB_GNT2_20 (0x7L<<16) +#define BNX2_MISC_ARB_GNT2_21 (0x7L<<20) +#define BNX2_MISC_ARB_GNT2_22 (0x7L<<24) +#define BNX2_MISC_ARB_GNT2_23 (0x7L<<28) + +#define BNX2_MISC_ARB_GNT3 0x00000874 +#define BNX2_MISC_ARB_GNT3_24 (0x7L<<0) +#define BNX2_MISC_ARB_GNT3_25 (0x7L<<4) +#define BNX2_MISC_ARB_GNT3_26 (0x7L<<8) +#define BNX2_MISC_ARB_GNT3_27 (0x7L<<12) +#define BNX2_MISC_ARB_GNT3_28 (0x7L<<16) +#define BNX2_MISC_ARB_GNT3_29 (0x7L<<20) +#define BNX2_MISC_ARB_GNT3_30 (0x7L<<24) +#define BNX2_MISC_ARB_GNT3_31 (0x7L<<28) + +#define BNX2_MISC_PRBS_CONTROL 0x00000878 +#define BNX2_MISC_PRBS_CONTROL_EN (1L<<0) +#define BNX2_MISC_PRBS_CONTROL_RSTB (1L<<1) +#define BNX2_MISC_PRBS_CONTROL_INV (1L<<2) +#define BNX2_MISC_PRBS_CONTROL_ERR_CLR (1L<<3) +#define BNX2_MISC_PRBS_CONTROL_ORDER (0x3L<<4) +#define BNX2_MISC_PRBS_CONTROL_ORDER_7TH (0L<<4) +#define BNX2_MISC_PRBS_CONTROL_ORDER_15TH (1L<<4) +#define BNX2_MISC_PRBS_CONTROL_ORDER_23RD (2L<<4) +#define BNX2_MISC_PRBS_CONTROL_ORDER_31ST (3L<<4) + +#define BNX2_MISC_PRBS_STATUS 0x0000087c +#define BNX2_MISC_PRBS_STATUS_LOCK (1L<<0) +#define BNX2_MISC_PRBS_STATUS_STKY (1L<<1) +#define BNX2_MISC_PRBS_STATUS_ERRORS (0x3fffL<<2) +#define BNX2_MISC_PRBS_STATUS_STATE (0xfL<<16) + +#define BNX2_MISC_SM_ASF_CONTROL 0x00000880 +#define BNX2_MISC_SM_ASF_CONTROL_ASF_RST (1L<<0) +#define BNX2_MISC_SM_ASF_CONTROL_TSC_EN (1L<<1) +#define BNX2_MISC_SM_ASF_CONTROL_WG_TO (1L<<2) +#define BNX2_MISC_SM_ASF_CONTROL_HB_TO (1L<<3) +#define BNX2_MISC_SM_ASF_CONTROL_PA_TO (1L<<4) +#define BNX2_MISC_SM_ASF_CONTROL_PL_TO (1L<<5) +#define BNX2_MISC_SM_ASF_CONTROL_RT_TO (1L<<6) +#define BNX2_MISC_SM_ASF_CONTROL_SMB_EVENT (1L<<7) +#define BNX2_MISC_SM_ASF_CONTROL_RES (0xfL<<8) +#define BNX2_MISC_SM_ASF_CONTROL_SMB_EN (1L<<12) +#define BNX2_MISC_SM_ASF_CONTROL_SMB_BB_EN (1L<<13) +#define BNX2_MISC_SM_ASF_CONTROL_SMB_NO_ADDR_FILT (1L<<14) +#define BNX2_MISC_SM_ASF_CONTROL_SMB_AUTOREAD (1L<<15) +#define BNX2_MISC_SM_ASF_CONTROL_NIC_SMB_ADDR1 (0x3fL<<16) +#define BNX2_MISC_SM_ASF_CONTROL_NIC_SMB_ADDR2 (0x3fL<<24) +#define BNX2_MISC_SM_ASF_CONTROL_EN_NIC_SMB_ADDR_0 (1L<<30) +#define BNX2_MISC_SM_ASF_CONTROL_SMB_EARLY_ATTN (1L<<31) + +#define BNX2_MISC_SMB_IN 0x00000884 +#define BNX2_MISC_SMB_IN_DAT_IN (0xffL<<0) +#define BNX2_MISC_SMB_IN_RDY (1L<<8) +#define BNX2_MISC_SMB_IN_DONE (1L<<9) +#define BNX2_MISC_SMB_IN_FIRSTBYTE (1L<<10) +#define BNX2_MISC_SMB_IN_STATUS (0x7L<<11) +#define BNX2_MISC_SMB_IN_STATUS_OK (0x0L<<11) +#define BNX2_MISC_SMB_IN_STATUS_PEC (0x1L<<11) +#define BNX2_MISC_SMB_IN_STATUS_OFLOW (0x2L<<11) +#define BNX2_MISC_SMB_IN_STATUS_STOP (0x3L<<11) +#define BNX2_MISC_SMB_IN_STATUS_TIMEOUT (0x4L<<11) + +#define BNX2_MISC_SMB_OUT 0x00000888 +#define BNX2_MISC_SMB_OUT_DAT_OUT (0xffL<<0) +#define BNX2_MISC_SMB_OUT_RDY (1L<<8) +#define BNX2_MISC_SMB_OUT_START (1L<<9) +#define BNX2_MISC_SMB_OUT_LAST (1L<<10) +#define BNX2_MISC_SMB_OUT_ACC_TYPE (1L<<11) +#define BNX2_MISC_SMB_OUT_ENB_PEC (1L<<12) +#define BNX2_MISC_SMB_OUT_GET_RX_LEN (1L<<13) +#define BNX2_MISC_SMB_OUT_SMB_READ_LEN (0x3fL<<14) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS (0xfL<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS_OK (0L<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS_FIRST_NACK (1L<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS_SUB_NACK (9L<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS_UFLOW (2L<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS_STOP (3L<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS_TIMEOUT (4L<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS_FIRST_LOST (5L<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS_SUB_LOST (0xdL<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_STATUS_BADACK (0x6L<<20) +#define BNX2_MISC_SMB_OUT_SMB_OUT_SLAVEMODE (1L<<24) +#define BNX2_MISC_SMB_OUT_SMB_OUT_DAT_EN (1L<<25) +#define BNX2_MISC_SMB_OUT_SMB_OUT_DAT_IN (1L<<26) +#define BNX2_MISC_SMB_OUT_SMB_OUT_CLK_EN (1L<<27) +#define BNX2_MISC_SMB_OUT_SMB_OUT_CLK_IN (1L<<28) + +#define BNX2_MISC_SMB_WATCHDOG 0x0000088c +#define BNX2_MISC_SMB_WATCHDOG_WATCHDOG (0xffffL<<0) + +#define BNX2_MISC_SMB_HEARTBEAT 0x00000890 +#define BNX2_MISC_SMB_HEARTBEAT_HEARTBEAT (0xffffL<<0) + +#define BNX2_MISC_SMB_POLL_ASF 0x00000894 +#define BNX2_MISC_SMB_POLL_ASF_POLL_ASF (0xffffL<<0) + +#define BNX2_MISC_SMB_POLL_LEGACY 0x00000898 +#define BNX2_MISC_SMB_POLL_LEGACY_POLL_LEGACY (0xffffL<<0) + +#define BNX2_MISC_SMB_RETRAN 0x0000089c +#define BNX2_MISC_SMB_RETRAN_RETRAN (0xffL<<0) + +#define BNX2_MISC_SMB_TIMESTAMP 0x000008a0 +#define BNX2_MISC_SMB_TIMESTAMP_TIMESTAMP (0xffffffffL<<0) + +#define BNX2_MISC_PERR_ENA0 0x000008a4 +#define BNX2_MISC_PERR_ENA0_COM_MISC_CTXC (1L<<0) +#define BNX2_MISC_PERR_ENA0_COM_MISC_REGF (1L<<1) +#define BNX2_MISC_PERR_ENA0_COM_MISC_SCPAD (1L<<2) +#define BNX2_MISC_PERR_ENA0_CP_MISC_CTXC (1L<<3) +#define BNX2_MISC_PERR_ENA0_CP_MISC_REGF (1L<<4) +#define BNX2_MISC_PERR_ENA0_CP_MISC_SCPAD (1L<<5) +#define BNX2_MISC_PERR_ENA0_CS_MISC_TMEM (1L<<6) +#define BNX2_MISC_PERR_ENA0_CTX_MISC_ACCM0 (1L<<7) +#define BNX2_MISC_PERR_ENA0_CTX_MISC_ACCM1 (1L<<8) +#define BNX2_MISC_PERR_ENA0_CTX_MISC_ACCM2 (1L<<9) +#define BNX2_MISC_PERR_ENA0_CTX_MISC_ACCM3 (1L<<10) +#define BNX2_MISC_PERR_ENA0_CTX_MISC_ACCM4 (1L<<11) +#define BNX2_MISC_PERR_ENA0_CTX_MISC_ACCM5 (1L<<12) +#define BNX2_MISC_PERR_ENA0_CTX_MISC_PGTBL (1L<<13) +#define BNX2_MISC_PERR_ENA0_DMAE_MISC_DR0 (1L<<14) +#define BNX2_MISC_PERR_ENA0_DMAE_MISC_DR1 (1L<<15) +#define BNX2_MISC_PERR_ENA0_DMAE_MISC_DR2 (1L<<16) +#define BNX2_MISC_PERR_ENA0_DMAE_MISC_DR3 (1L<<17) +#define BNX2_MISC_PERR_ENA0_DMAE_MISC_DR4 (1L<<18) +#define BNX2_MISC_PERR_ENA0_DMAE_MISC_DW0 (1L<<19) +#define BNX2_MISC_PERR_ENA0_DMAE_MISC_DW1 (1L<<20) +#define BNX2_MISC_PERR_ENA0_DMAE_MISC_DW2 (1L<<21) +#define BNX2_MISC_PERR_ENA0_HC_MISC_DMA (1L<<22) +#define BNX2_MISC_PERR_ENA0_MCP_MISC_REGF (1L<<23) +#define BNX2_MISC_PERR_ENA0_MCP_MISC_SCPAD (1L<<24) +#define BNX2_MISC_PERR_ENA0_MQ_MISC_CTX (1L<<25) +#define BNX2_MISC_PERR_ENA0_RBDC_MISC (1L<<26) +#define BNX2_MISC_PERR_ENA0_RBUF_MISC_MB (1L<<27) +#define BNX2_MISC_PERR_ENA0_RBUF_MISC_PTR (1L<<28) +#define BNX2_MISC_PERR_ENA0_RDE_MISC_RPC (1L<<29) +#define BNX2_MISC_PERR_ENA0_RDE_MISC_RPM (1L<<30) +#define BNX2_MISC_PERR_ENA0_RV2P_MISC_CB0REGS (1L<<31) + +#define BNX2_MISC_PERR_ENA1 0x000008a8 +#define BNX2_MISC_PERR_ENA1_RV2P_MISC_CB1REGS (1L<<0) +#define BNX2_MISC_PERR_ENA1_RV2P_MISC_P1IRAM (1L<<1) +#define BNX2_MISC_PERR_ENA1_RV2P_MISC_P2IRAM (1L<<2) +#define BNX2_MISC_PERR_ENA1_RXP_MISC_CTXC (1L<<3) +#define BNX2_MISC_PERR_ENA1_RXP_MISC_REGF (1L<<4) +#define BNX2_MISC_PERR_ENA1_RXP_MISC_SCPAD (1L<<5) +#define BNX2_MISC_PERR_ENA1_RXP_MISC_RBUFC (1L<<6) +#define BNX2_MISC_PERR_ENA1_TBDC_MISC (1L<<7) +#define BNX2_MISC_PERR_ENA1_TDMA_MISC (1L<<8) +#define BNX2_MISC_PERR_ENA1_THBUF_MISC_MB0 (1L<<9) +#define BNX2_MISC_PERR_ENA1_THBUF_MISC_MB1 (1L<<10) +#define BNX2_MISC_PERR_ENA1_TPAT_MISC_REGF (1L<<11) +#define BNX2_MISC_PERR_ENA1_TPAT_MISC_SCPAD (1L<<12) +#define BNX2_MISC_PERR_ENA1_TPBUF_MISC_MB (1L<<13) +#define BNX2_MISC_PERR_ENA1_TSCH_MISC_LR (1L<<14) +#define BNX2_MISC_PERR_ENA1_TXP_MISC_CTXC (1L<<15) +#define BNX2_MISC_PERR_ENA1_TXP_MISC_REGF (1L<<16) +#define BNX2_MISC_PERR_ENA1_TXP_MISC_SCPAD (1L<<17) +#define BNX2_MISC_PERR_ENA1_UMP_MISC_FIORX (1L<<18) +#define BNX2_MISC_PERR_ENA1_UMP_MISC_FIOTX (1L<<19) +#define BNX2_MISC_PERR_ENA1_UMP_MISC_RX (1L<<20) +#define BNX2_MISC_PERR_ENA1_UMP_MISC_TX (1L<<21) +#define BNX2_MISC_PERR_ENA1_RDMAQ_MISC (1L<<22) +#define BNX2_MISC_PERR_ENA1_CSQ_MISC (1L<<23) +#define BNX2_MISC_PERR_ENA1_CPQ_MISC (1L<<24) +#define BNX2_MISC_PERR_ENA1_MCPQ_MISC (1L<<25) +#define BNX2_MISC_PERR_ENA1_RV2PMQ_MISC (1L<<26) +#define BNX2_MISC_PERR_ENA1_RV2PPQ_MISC (1L<<27) +#define BNX2_MISC_PERR_ENA1_RV2PTQ_MISC (1L<<28) +#define BNX2_MISC_PERR_ENA1_RXPQ_MISC (1L<<29) +#define BNX2_MISC_PERR_ENA1_RXPCQ_MISC (1L<<30) +#define BNX2_MISC_PERR_ENA1_RLUPQ_MISC (1L<<31) + +#define BNX2_MISC_PERR_ENA2 0x000008ac +#define BNX2_MISC_PERR_ENA2_COMQ_MISC (1L<<0) +#define BNX2_MISC_PERR_ENA2_COMXQ_MISC (1L<<1) +#define BNX2_MISC_PERR_ENA2_COMTQ_MISC (1L<<2) +#define BNX2_MISC_PERR_ENA2_TSCHQ_MISC (1L<<3) +#define BNX2_MISC_PERR_ENA2_TBDRQ_MISC (1L<<4) +#define BNX2_MISC_PERR_ENA2_TXPQ_MISC (1L<<5) +#define BNX2_MISC_PERR_ENA2_TDMAQ_MISC (1L<<6) +#define BNX2_MISC_PERR_ENA2_TPATQ_MISC (1L<<7) +#define BNX2_MISC_PERR_ENA2_TASQ_MISC (1L<<8) + +#define BNX2_MISC_DEBUG_VECTOR_SEL 0x000008b0 +#define BNX2_MISC_DEBUG_VECTOR_SEL_0 (0xfffL<<0) +#define BNX2_MISC_DEBUG_VECTOR_SEL_1 (0xfffL<<12) + +#define BNX2_MISC_VREG_CONTROL 0x000008b4 +#define BNX2_MISC_VREG_CONTROL_1_2 (0xfL<<0) +#define BNX2_MISC_VREG_CONTROL_2_5 (0xfL<<4) + +#define BNX2_MISC_FINAL_CLK_CTL_VAL 0x000008b8 +#define BNX2_MISC_FINAL_CLK_CTL_VAL_MISC_FINAL_CLK_CTL_VAL (0x3ffffffL<<6) + +#define BNX2_MISC_UNUSED0 0x000008bc + + +/* + * nvm_reg definition + * offset: 0x6400 + */ +#define BNX2_NVM_COMMAND 0x00006400 +#define BNX2_NVM_COMMAND_RST (1L<<0) +#define BNX2_NVM_COMMAND_DONE (1L<<3) +#define BNX2_NVM_COMMAND_DOIT (1L<<4) +#define BNX2_NVM_COMMAND_WR (1L<<5) +#define BNX2_NVM_COMMAND_ERASE (1L<<6) +#define BNX2_NVM_COMMAND_FIRST (1L<<7) +#define BNX2_NVM_COMMAND_LAST (1L<<8) +#define BNX2_NVM_COMMAND_WREN (1L<<16) +#define BNX2_NVM_COMMAND_WRDI (1L<<17) +#define BNX2_NVM_COMMAND_EWSR (1L<<18) +#define BNX2_NVM_COMMAND_WRSR (1L<<19) + +#define BNX2_NVM_STATUS 0x00006404 +#define BNX2_NVM_STATUS_PI_FSM_STATE (0xfL<<0) +#define BNX2_NVM_STATUS_EE_FSM_STATE (0xfL<<4) +#define BNX2_NVM_STATUS_EQ_FSM_STATE (0xfL<<8) + +#define BNX2_NVM_WRITE 0x00006408 +#define BNX2_NVM_WRITE_NVM_WRITE_VALUE (0xffffffffL<<0) +#define BNX2_NVM_WRITE_NVM_WRITE_VALUE_BIT_BANG (0L<<0) +#define BNX2_NVM_WRITE_NVM_WRITE_VALUE_EECLK (1L<<0) +#define BNX2_NVM_WRITE_NVM_WRITE_VALUE_EEDATA (2L<<0) +#define BNX2_NVM_WRITE_NVM_WRITE_VALUE_SCLK (4L<<0) +#define BNX2_NVM_WRITE_NVM_WRITE_VALUE_CS_B (8L<<0) +#define BNX2_NVM_WRITE_NVM_WRITE_VALUE_SO (16L<<0) +#define BNX2_NVM_WRITE_NVM_WRITE_VALUE_SI (32L<<0) + +#define BNX2_NVM_ADDR 0x0000640c +#define BNX2_NVM_ADDR_NVM_ADDR_VALUE (0xffffffL<<0) +#define BNX2_NVM_ADDR_NVM_ADDR_VALUE_BIT_BANG (0L<<0) +#define BNX2_NVM_ADDR_NVM_ADDR_VALUE_EECLK (1L<<0) +#define BNX2_NVM_ADDR_NVM_ADDR_VALUE_EEDATA (2L<<0) +#define BNX2_NVM_ADDR_NVM_ADDR_VALUE_SCLK (4L<<0) +#define BNX2_NVM_ADDR_NVM_ADDR_VALUE_CS_B (8L<<0) +#define BNX2_NVM_ADDR_NVM_ADDR_VALUE_SO (16L<<0) +#define BNX2_NVM_ADDR_NVM_ADDR_VALUE_SI (32L<<0) + +#define BNX2_NVM_READ 0x00006410 +#define BNX2_NVM_READ_NVM_READ_VALUE (0xffffffffL<<0) +#define BNX2_NVM_READ_NVM_READ_VALUE_BIT_BANG (0L<<0) +#define BNX2_NVM_READ_NVM_READ_VALUE_EECLK (1L<<0) +#define BNX2_NVM_READ_NVM_READ_VALUE_EEDATA (2L<<0) +#define BNX2_NVM_READ_NVM_READ_VALUE_SCLK (4L<<0) +#define BNX2_NVM_READ_NVM_READ_VALUE_CS_B (8L<<0) +#define BNX2_NVM_READ_NVM_READ_VALUE_SO (16L<<0) +#define BNX2_NVM_READ_NVM_READ_VALUE_SI (32L<<0) + +#define BNX2_NVM_CFG1 0x00006414 +#define BNX2_NVM_CFG1_FLASH_MODE (1L<<0) +#define BNX2_NVM_CFG1_BUFFER_MODE (1L<<1) +#define BNX2_NVM_CFG1_PASS_MODE (1L<<2) +#define BNX2_NVM_CFG1_BITBANG_MODE (1L<<3) +#define BNX2_NVM_CFG1_STATUS_BIT (0x7L<<4) +#define BNX2_NVM_CFG1_STATUS_BIT_FLASH_RDY (0L<<4) +#define BNX2_NVM_CFG1_STATUS_BIT_BUFFER_RDY (7L<<4) +#define BNX2_NVM_CFG1_SPI_CLK_DIV (0xfL<<7) +#define BNX2_NVM_CFG1_SEE_CLK_DIV (0x7ffL<<11) +#define BNX2_NVM_CFG1_PROTECT_MODE (1L<<24) +#define BNX2_NVM_CFG1_FLASH_SIZE (1L<<25) +#define BNX2_NVM_CFG1_COMPAT_BYPASSS (1L<<31) + +#define BNX2_NVM_CFG2 0x00006418 +#define BNX2_NVM_CFG2_ERASE_CMD (0xffL<<0) +#define BNX2_NVM_CFG2_DUMMY (0xffL<<8) +#define BNX2_NVM_CFG2_STATUS_CMD (0xffL<<16) + +#define BNX2_NVM_CFG3 0x0000641c +#define BNX2_NVM_CFG3_BUFFER_RD_CMD (0xffL<<0) +#define BNX2_NVM_CFG3_WRITE_CMD (0xffL<<8) +#define BNX2_NVM_CFG3_BUFFER_WRITE_CMD (0xffL<<16) +#define BNX2_NVM_CFG3_READ_CMD (0xffL<<24) + +#define BNX2_NVM_SW_ARB 0x00006420 +#define BNX2_NVM_SW_ARB_ARB_REQ_SET0 (1L<<0) +#define BNX2_NVM_SW_ARB_ARB_REQ_SET1 (1L<<1) +#define BNX2_NVM_SW_ARB_ARB_REQ_SET2 (1L<<2) +#define BNX2_NVM_SW_ARB_ARB_REQ_SET3 (1L<<3) +#define BNX2_NVM_SW_ARB_ARB_REQ_CLR0 (1L<<4) +#define BNX2_NVM_SW_ARB_ARB_REQ_CLR1 (1L<<5) +#define BNX2_NVM_SW_ARB_ARB_REQ_CLR2 (1L<<6) +#define BNX2_NVM_SW_ARB_ARB_REQ_CLR3 (1L<<7) +#define BNX2_NVM_SW_ARB_ARB_ARB0 (1L<<8) +#define BNX2_NVM_SW_ARB_ARB_ARB1 (1L<<9) +#define BNX2_NVM_SW_ARB_ARB_ARB2 (1L<<10) +#define BNX2_NVM_SW_ARB_ARB_ARB3 (1L<<11) +#define BNX2_NVM_SW_ARB_REQ0 (1L<<12) +#define BNX2_NVM_SW_ARB_REQ1 (1L<<13) +#define BNX2_NVM_SW_ARB_REQ2 (1L<<14) +#define BNX2_NVM_SW_ARB_REQ3 (1L<<15) + +#define BNX2_NVM_ACCESS_ENABLE 0x00006424 +#define BNX2_NVM_ACCESS_ENABLE_EN (1L<<0) +#define BNX2_NVM_ACCESS_ENABLE_WR_EN (1L<<1) + +#define BNX2_NVM_WRITE1 0x00006428 +#define BNX2_NVM_WRITE1_WREN_CMD (0xffL<<0) +#define BNX2_NVM_WRITE1_WRDI_CMD (0xffL<<8) +#define BNX2_NVM_WRITE1_SR_DATA (0xffL<<16) + + + +/* + * dma_reg definition + * offset: 0xc00 + */ +#define BNX2_DMA_COMMAND 0x00000c00 +#define BNX2_DMA_COMMAND_ENABLE (1L<<0) + +#define BNX2_DMA_STATUS 0x00000c04 +#define BNX2_DMA_STATUS_PAR_ERROR_STATE (1L<<0) +#define BNX2_DMA_STATUS_READ_TRANSFERS_STAT (1L<<16) +#define BNX2_DMA_STATUS_READ_DELAY_PCI_CLKS_STAT (1L<<17) +#define BNX2_DMA_STATUS_BIG_READ_TRANSFERS_STAT (1L<<18) +#define BNX2_DMA_STATUS_BIG_READ_DELAY_PCI_CLKS_STAT (1L<<19) +#define BNX2_DMA_STATUS_BIG_READ_RETRY_AFTER_DATA_STAT (1L<<20) +#define BNX2_DMA_STATUS_WRITE_TRANSFERS_STAT (1L<<21) +#define BNX2_DMA_STATUS_WRITE_DELAY_PCI_CLKS_STAT (1L<<22) +#define BNX2_DMA_STATUS_BIG_WRITE_TRANSFERS_STAT (1L<<23) +#define BNX2_DMA_STATUS_BIG_WRITE_DELAY_PCI_CLKS_STAT (1L<<24) +#define BNX2_DMA_STATUS_BIG_WRITE_RETRY_AFTER_DATA_STAT (1L<<25) + +#define BNX2_DMA_CONFIG 0x00000c08 +#define BNX2_DMA_CONFIG_DATA_BYTE_SWAP (1L<<0) +#define BNX2_DMA_CONFIG_DATA_WORD_SWAP (1L<<1) +#define BNX2_DMA_CONFIG_CNTL_BYTE_SWAP (1L<<4) +#define BNX2_DMA_CONFIG_CNTL_WORD_SWAP (1L<<5) +#define BNX2_DMA_CONFIG_ONE_DMA (1L<<6) +#define BNX2_DMA_CONFIG_CNTL_TWO_DMA (1L<<7) +#define BNX2_DMA_CONFIG_CNTL_FPGA_MODE (1L<<8) +#define BNX2_DMA_CONFIG_CNTL_PING_PONG_DMA (1L<<10) +#define BNX2_DMA_CONFIG_CNTL_PCI_COMP_DLY (1L<<11) +#define BNX2_DMA_CONFIG_NO_RCHANS_IN_USE (0xfL<<12) +#define BNX2_DMA_CONFIG_NO_WCHANS_IN_USE (0xfL<<16) +#define BNX2_DMA_CONFIG_PCI_CLK_CMP_BITS (0x7L<<20) +#define BNX2_DMA_CONFIG_PCI_FAST_CLK_CMP (1L<<23) +#define BNX2_DMA_CONFIG_BIG_SIZE (0xfL<<24) +#define BNX2_DMA_CONFIG_BIG_SIZE_NONE (0x0L<<24) +#define BNX2_DMA_CONFIG_BIG_SIZE_64 (0x1L<<24) +#define BNX2_DMA_CONFIG_BIG_SIZE_128 (0x2L<<24) +#define BNX2_DMA_CONFIG_BIG_SIZE_256 (0x4L<<24) +#define BNX2_DMA_CONFIG_BIG_SIZE_512 (0x8L<<24) + +#define BNX2_DMA_BLACKOUT 0x00000c0c +#define BNX2_DMA_BLACKOUT_RD_RETRY_BLACKOUT (0xffL<<0) +#define BNX2_DMA_BLACKOUT_2ND_RD_RETRY_BLACKOUT (0xffL<<8) +#define BNX2_DMA_BLACKOUT_WR_RETRY_BLACKOUT (0xffL<<16) + +#define BNX2_DMA_RCHAN_STAT 0x00000c30 +#define BNX2_DMA_RCHAN_STAT_COMP_CODE_0 (0x7L<<0) +#define BNX2_DMA_RCHAN_STAT_PAR_ERR_0 (1L<<3) +#define BNX2_DMA_RCHAN_STAT_COMP_CODE_1 (0x7L<<4) +#define BNX2_DMA_RCHAN_STAT_PAR_ERR_1 (1L<<7) +#define BNX2_DMA_RCHAN_STAT_COMP_CODE_2 (0x7L<<8) +#define BNX2_DMA_RCHAN_STAT_PAR_ERR_2 (1L<<11) +#define BNX2_DMA_RCHAN_STAT_COMP_CODE_3 (0x7L<<12) +#define BNX2_DMA_RCHAN_STAT_PAR_ERR_3 (1L<<15) +#define BNX2_DMA_RCHAN_STAT_COMP_CODE_4 (0x7L<<16) +#define BNX2_DMA_RCHAN_STAT_PAR_ERR_4 (1L<<19) +#define BNX2_DMA_RCHAN_STAT_COMP_CODE_5 (0x7L<<20) +#define BNX2_DMA_RCHAN_STAT_PAR_ERR_5 (1L<<23) +#define BNX2_DMA_RCHAN_STAT_COMP_CODE_6 (0x7L<<24) +#define BNX2_DMA_RCHAN_STAT_PAR_ERR_6 (1L<<27) +#define BNX2_DMA_RCHAN_STAT_COMP_CODE_7 (0x7L<<28) +#define BNX2_DMA_RCHAN_STAT_PAR_ERR_7 (1L<<31) + +#define BNX2_DMA_WCHAN_STAT 0x00000c34 +#define BNX2_DMA_WCHAN_STAT_COMP_CODE_0 (0x7L<<0) +#define BNX2_DMA_WCHAN_STAT_PAR_ERR_0 (1L<<3) +#define BNX2_DMA_WCHAN_STAT_COMP_CODE_1 (0x7L<<4) +#define BNX2_DMA_WCHAN_STAT_PAR_ERR_1 (1L<<7) +#define BNX2_DMA_WCHAN_STAT_COMP_CODE_2 (0x7L<<8) +#define BNX2_DMA_WCHAN_STAT_PAR_ERR_2 (1L<<11) +#define BNX2_DMA_WCHAN_STAT_COMP_CODE_3 (0x7L<<12) +#define BNX2_DMA_WCHAN_STAT_PAR_ERR_3 (1L<<15) +#define BNX2_DMA_WCHAN_STAT_COMP_CODE_4 (0x7L<<16) +#define BNX2_DMA_WCHAN_STAT_PAR_ERR_4 (1L<<19) +#define BNX2_DMA_WCHAN_STAT_COMP_CODE_5 (0x7L<<20) +#define BNX2_DMA_WCHAN_STAT_PAR_ERR_5 (1L<<23) +#define BNX2_DMA_WCHAN_STAT_COMP_CODE_6 (0x7L<<24) +#define BNX2_DMA_WCHAN_STAT_PAR_ERR_6 (1L<<27) +#define BNX2_DMA_WCHAN_STAT_COMP_CODE_7 (0x7L<<28) +#define BNX2_DMA_WCHAN_STAT_PAR_ERR_7 (1L<<31) + +#define BNX2_DMA_RCHAN_ASSIGNMENT 0x00000c38 +#define BNX2_DMA_RCHAN_ASSIGNMENT_0 (0xfL<<0) +#define BNX2_DMA_RCHAN_ASSIGNMENT_1 (0xfL<<4) +#define BNX2_DMA_RCHAN_ASSIGNMENT_2 (0xfL<<8) +#define BNX2_DMA_RCHAN_ASSIGNMENT_3 (0xfL<<12) +#define BNX2_DMA_RCHAN_ASSIGNMENT_4 (0xfL<<16) +#define BNX2_DMA_RCHAN_ASSIGNMENT_5 (0xfL<<20) +#define BNX2_DMA_RCHAN_ASSIGNMENT_6 (0xfL<<24) +#define BNX2_DMA_RCHAN_ASSIGNMENT_7 (0xfL<<28) + +#define BNX2_DMA_WCHAN_ASSIGNMENT 0x00000c3c +#define BNX2_DMA_WCHAN_ASSIGNMENT_0 (0xfL<<0) +#define BNX2_DMA_WCHAN_ASSIGNMENT_1 (0xfL<<4) +#define BNX2_DMA_WCHAN_ASSIGNMENT_2 (0xfL<<8) +#define BNX2_DMA_WCHAN_ASSIGNMENT_3 (0xfL<<12) +#define BNX2_DMA_WCHAN_ASSIGNMENT_4 (0xfL<<16) +#define BNX2_DMA_WCHAN_ASSIGNMENT_5 (0xfL<<20) +#define BNX2_DMA_WCHAN_ASSIGNMENT_6 (0xfL<<24) +#define BNX2_DMA_WCHAN_ASSIGNMENT_7 (0xfL<<28) + +#define BNX2_DMA_RCHAN_STAT_00 0x00000c40 +#define BNX2_DMA_RCHAN_STAT_00_RCHAN_STA_HOST_ADDR_LOW (0xffffffffL<<0) + +#define BNX2_DMA_RCHAN_STAT_01 0x00000c44 +#define BNX2_DMA_RCHAN_STAT_01_RCHAN_STA_HOST_ADDR_HIGH (0xffffffffL<<0) + +#define BNX2_DMA_RCHAN_STAT_02 0x00000c48 +#define BNX2_DMA_RCHAN_STAT_02_LENGTH (0xffffL<<0) +#define BNX2_DMA_RCHAN_STAT_02_WORD_SWAP (1L<<16) +#define BNX2_DMA_RCHAN_STAT_02_BYTE_SWAP (1L<<17) +#define BNX2_DMA_RCHAN_STAT_02_PRIORITY_LVL (1L<<18) + +#define BNX2_DMA_RCHAN_STAT_10 0x00000c4c +#define BNX2_DMA_RCHAN_STAT_11 0x00000c50 +#define BNX2_DMA_RCHAN_STAT_12 0x00000c54 +#define BNX2_DMA_RCHAN_STAT_20 0x00000c58 +#define BNX2_DMA_RCHAN_STAT_21 0x00000c5c +#define BNX2_DMA_RCHAN_STAT_22 0x00000c60 +#define BNX2_DMA_RCHAN_STAT_30 0x00000c64 +#define BNX2_DMA_RCHAN_STAT_31 0x00000c68 +#define BNX2_DMA_RCHAN_STAT_32 0x00000c6c +#define BNX2_DMA_RCHAN_STAT_40 0x00000c70 +#define BNX2_DMA_RCHAN_STAT_41 0x00000c74 +#define BNX2_DMA_RCHAN_STAT_42 0x00000c78 +#define BNX2_DMA_RCHAN_STAT_50 0x00000c7c +#define BNX2_DMA_RCHAN_STAT_51 0x00000c80 +#define BNX2_DMA_RCHAN_STAT_52 0x00000c84 +#define BNX2_DMA_RCHAN_STAT_60 0x00000c88 +#define BNX2_DMA_RCHAN_STAT_61 0x00000c8c +#define BNX2_DMA_RCHAN_STAT_62 0x00000c90 +#define BNX2_DMA_RCHAN_STAT_70 0x00000c94 +#define BNX2_DMA_RCHAN_STAT_71 0x00000c98 +#define BNX2_DMA_RCHAN_STAT_72 0x00000c9c +#define BNX2_DMA_WCHAN_STAT_00 0x00000ca0 +#define BNX2_DMA_WCHAN_STAT_00_WCHAN_STA_HOST_ADDR_LOW (0xffffffffL<<0) + +#define BNX2_DMA_WCHAN_STAT_01 0x00000ca4 +#define BNX2_DMA_WCHAN_STAT_01_WCHAN_STA_HOST_ADDR_HIGH (0xffffffffL<<0) + +#define BNX2_DMA_WCHAN_STAT_02 0x00000ca8 +#define BNX2_DMA_WCHAN_STAT_02_LENGTH (0xffffL<<0) +#define BNX2_DMA_WCHAN_STAT_02_WORD_SWAP (1L<<16) +#define BNX2_DMA_WCHAN_STAT_02_BYTE_SWAP (1L<<17) +#define BNX2_DMA_WCHAN_STAT_02_PRIORITY_LVL (1L<<18) + +#define BNX2_DMA_WCHAN_STAT_10 0x00000cac +#define BNX2_DMA_WCHAN_STAT_11 0x00000cb0 +#define BNX2_DMA_WCHAN_STAT_12 0x00000cb4 +#define BNX2_DMA_WCHAN_STAT_20 0x00000cb8 +#define BNX2_DMA_WCHAN_STAT_21 0x00000cbc +#define BNX2_DMA_WCHAN_STAT_22 0x00000cc0 +#define BNX2_DMA_WCHAN_STAT_30 0x00000cc4 +#define BNX2_DMA_WCHAN_STAT_31 0x00000cc8 +#define BNX2_DMA_WCHAN_STAT_32 0x00000ccc +#define BNX2_DMA_WCHAN_STAT_40 0x00000cd0 +#define BNX2_DMA_WCHAN_STAT_41 0x00000cd4 +#define BNX2_DMA_WCHAN_STAT_42 0x00000cd8 +#define BNX2_DMA_WCHAN_STAT_50 0x00000cdc +#define BNX2_DMA_WCHAN_STAT_51 0x00000ce0 +#define BNX2_DMA_WCHAN_STAT_52 0x00000ce4 +#define BNX2_DMA_WCHAN_STAT_60 0x00000ce8 +#define BNX2_DMA_WCHAN_STAT_61 0x00000cec +#define BNX2_DMA_WCHAN_STAT_62 0x00000cf0 +#define BNX2_DMA_WCHAN_STAT_70 0x00000cf4 +#define BNX2_DMA_WCHAN_STAT_71 0x00000cf8 +#define BNX2_DMA_WCHAN_STAT_72 0x00000cfc +#define BNX2_DMA_ARB_STAT_00 0x00000d00 +#define BNX2_DMA_ARB_STAT_00_MASTER (0xffffL<<0) +#define BNX2_DMA_ARB_STAT_00_MASTER_ENC (0xffL<<16) +#define BNX2_DMA_ARB_STAT_00_CUR_BINMSTR (0xffL<<24) + +#define BNX2_DMA_ARB_STAT_01 0x00000d04 +#define BNX2_DMA_ARB_STAT_01_LPR_RPTR (0xfL<<0) +#define BNX2_DMA_ARB_STAT_01_LPR_WPTR (0xfL<<4) +#define BNX2_DMA_ARB_STAT_01_LPB_RPTR (0xfL<<8) +#define BNX2_DMA_ARB_STAT_01_LPB_WPTR (0xfL<<12) +#define BNX2_DMA_ARB_STAT_01_HPR_RPTR (0xfL<<16) +#define BNX2_DMA_ARB_STAT_01_HPR_WPTR (0xfL<<20) +#define BNX2_DMA_ARB_STAT_01_HPB_RPTR (0xfL<<24) +#define BNX2_DMA_ARB_STAT_01_HPB_WPTR (0xfL<<28) + +#define BNX2_DMA_FUSE_CTRL0_CMD 0x00000f00 +#define BNX2_DMA_FUSE_CTRL0_CMD_PWRUP_DONE (1L<<0) +#define BNX2_DMA_FUSE_CTRL0_CMD_SHIFT_DONE (1L<<1) +#define BNX2_DMA_FUSE_CTRL0_CMD_SHIFT (1L<<2) +#define BNX2_DMA_FUSE_CTRL0_CMD_LOAD (1L<<3) +#define BNX2_DMA_FUSE_CTRL0_CMD_SEL (0xfL<<8) + +#define BNX2_DMA_FUSE_CTRL0_DATA 0x00000f04 +#define BNX2_DMA_FUSE_CTRL1_CMD 0x00000f08 +#define BNX2_DMA_FUSE_CTRL1_CMD_PWRUP_DONE (1L<<0) +#define BNX2_DMA_FUSE_CTRL1_CMD_SHIFT_DONE (1L<<1) +#define BNX2_DMA_FUSE_CTRL1_CMD_SHIFT (1L<<2) +#define BNX2_DMA_FUSE_CTRL1_CMD_LOAD (1L<<3) +#define BNX2_DMA_FUSE_CTRL1_CMD_SEL (0xfL<<8) + +#define BNX2_DMA_FUSE_CTRL1_DATA 0x00000f0c +#define BNX2_DMA_FUSE_CTRL2_CMD 0x00000f10 +#define BNX2_DMA_FUSE_CTRL2_CMD_PWRUP_DONE (1L<<0) +#define BNX2_DMA_FUSE_CTRL2_CMD_SHIFT_DONE (1L<<1) +#define BNX2_DMA_FUSE_CTRL2_CMD_SHIFT (1L<<2) +#define BNX2_DMA_FUSE_CTRL2_CMD_LOAD (1L<<3) +#define BNX2_DMA_FUSE_CTRL2_CMD_SEL (0xfL<<8) + +#define BNX2_DMA_FUSE_CTRL2_DATA 0x00000f14 + + +/* + * context_reg definition + * offset: 0x1000 + */ +#define BNX2_CTX_COMMAND 0x00001000 +#define BNX2_CTX_COMMAND_ENABLED (1L<<0) + +#define BNX2_CTX_STATUS 0x00001004 +#define BNX2_CTX_STATUS_LOCK_WAIT (1L<<0) +#define BNX2_CTX_STATUS_READ_STAT (1L<<16) +#define BNX2_CTX_STATUS_WRITE_STAT (1L<<17) +#define BNX2_CTX_STATUS_ACC_STALL_STAT (1L<<18) +#define BNX2_CTX_STATUS_LOCK_STALL_STAT (1L<<19) + +#define BNX2_CTX_VIRT_ADDR 0x00001008 +#define BNX2_CTX_VIRT_ADDR_VIRT_ADDR (0x7fffL<<6) + +#define BNX2_CTX_PAGE_TBL 0x0000100c +#define BNX2_CTX_PAGE_TBL_PAGE_TBL (0x3fffL<<6) + +#define BNX2_CTX_DATA_ADR 0x00001010 +#define BNX2_CTX_DATA_ADR_DATA_ADR (0x7ffffL<<2) + +#define BNX2_CTX_DATA 0x00001014 +#define BNX2_CTX_LOCK 0x00001018 +#define BNX2_CTX_LOCK_TYPE (0x7L<<0) +#define BNX2_CTX_LOCK_TYPE_LOCK_TYPE_VOID (0x0L<<0) +#define BNX2_CTX_LOCK_TYPE_LOCK_TYPE_COMPLETE (0x7L<<0) +#define BNX2_CTX_LOCK_TYPE_LOCK_TYPE_PROTOCOL (0x1L<<0) +#define BNX2_CTX_LOCK_TYPE_LOCK_TYPE_TX (0x2L<<0) +#define BNX2_CTX_LOCK_TYPE_LOCK_TYPE_TIMER (0x4L<<0) +#define BNX2_CTX_LOCK_CID_VALUE (0x3fffL<<7) +#define BNX2_CTX_LOCK_GRANTED (1L<<26) +#define BNX2_CTX_LOCK_MODE (0x7L<<27) +#define BNX2_CTX_LOCK_MODE_UNLOCK (0x0L<<27) +#define BNX2_CTX_LOCK_MODE_IMMEDIATE (0x1L<<27) +#define BNX2_CTX_LOCK_MODE_SURE (0x2L<<27) +#define BNX2_CTX_LOCK_STATUS (1L<<30) +#define BNX2_CTX_LOCK_REQ (1L<<31) + +#define BNX2_CTX_ACCESS_STATUS 0x00001040 +#define BNX2_CTX_ACCESS_STATUS_MASTERENCODED (0xfL<<0) +#define BNX2_CTX_ACCESS_STATUS_ACCESSMEMORYSM (0x3L<<10) +#define BNX2_CTX_ACCESS_STATUS_PAGETABLEINITSM (0x3L<<12) +#define BNX2_CTX_ACCESS_STATUS_ACCESSMEMORYINITSM (0x3L<<14) +#define BNX2_CTX_ACCESS_STATUS_QUALIFIED_REQUEST (0x7ffL<<17) + +#define BNX2_CTX_DBG_LOCK_STATUS 0x00001044 +#define BNX2_CTX_DBG_LOCK_STATUS_SM (0x3ffL<<0) +#define BNX2_CTX_DBG_LOCK_STATUS_MATCH (0x3ffL<<22) + +#define BNX2_CTX_CHNL_LOCK_STATUS_0 0x00001080 +#define BNX2_CTX_CHNL_LOCK_STATUS_0_CID (0x3fffL<<0) +#define BNX2_CTX_CHNL_LOCK_STATUS_0_TYPE (0x3L<<14) +#define BNX2_CTX_CHNL_LOCK_STATUS_0_MODE (1L<<16) + +#define BNX2_CTX_CHNL_LOCK_STATUS_1 0x00001084 +#define BNX2_CTX_CHNL_LOCK_STATUS_2 0x00001088 +#define BNX2_CTX_CHNL_LOCK_STATUS_3 0x0000108c +#define BNX2_CTX_CHNL_LOCK_STATUS_4 0x00001090 +#define BNX2_CTX_CHNL_LOCK_STATUS_5 0x00001094 +#define BNX2_CTX_CHNL_LOCK_STATUS_6 0x00001098 +#define BNX2_CTX_CHNL_LOCK_STATUS_7 0x0000109c +#define BNX2_CTX_CHNL_LOCK_STATUS_8 0x000010a0 + + +/* + * emac_reg definition + * offset: 0x1400 + */ +#define BNX2_EMAC_MODE 0x00001400 +#define BNX2_EMAC_MODE_RESET (1L<<0) +#define BNX2_EMAC_MODE_HALF_DUPLEX (1L<<1) +#define BNX2_EMAC_MODE_PORT (0x3L<<2) +#define BNX2_EMAC_MODE_PORT_NONE (0L<<2) +#define BNX2_EMAC_MODE_PORT_MII (1L<<2) +#define BNX2_EMAC_MODE_PORT_GMII (2L<<2) +#define BNX2_EMAC_MODE_PORT_UNDEF (3L<<2) +#define BNX2_EMAC_MODE_MAC_LOOP (1L<<4) +#define BNX2_EMAC_MODE_TAGGED_MAC_CTL (1L<<7) +#define BNX2_EMAC_MODE_TX_BURST (1L<<8) +#define BNX2_EMAC_MODE_MAX_DEFER_DROP_ENA (1L<<9) +#define BNX2_EMAC_MODE_EXT_LINK_POL (1L<<10) +#define BNX2_EMAC_MODE_FORCE_LINK (1L<<11) +#define BNX2_EMAC_MODE_MPKT (1L<<18) +#define BNX2_EMAC_MODE_MPKT_RCVD (1L<<19) +#define BNX2_EMAC_MODE_ACPI_RCVD (1L<<20) + +#define BNX2_EMAC_STATUS 0x00001404 +#define BNX2_EMAC_STATUS_LINK (1L<<11) +#define BNX2_EMAC_STATUS_LINK_CHANGE (1L<<12) +#define BNX2_EMAC_STATUS_MI_COMPLETE (1L<<22) +#define BNX2_EMAC_STATUS_MI_INT (1L<<23) +#define BNX2_EMAC_STATUS_AP_ERROR (1L<<24) +#define BNX2_EMAC_STATUS_PARITY_ERROR_STATE (1L<<31) + +#define BNX2_EMAC_ATTENTION_ENA 0x00001408 +#define BNX2_EMAC_ATTENTION_ENA_LINK (1L<<11) +#define BNX2_EMAC_ATTENTION_ENA_MI_COMPLETE (1L<<22) +#define BNX2_EMAC_ATTENTION_ENA_MI_INT (1L<<23) +#define BNX2_EMAC_ATTENTION_ENA_AP_ERROR (1L<<24) + +#define BNX2_EMAC_LED 0x0000140c +#define BNX2_EMAC_LED_OVERRIDE (1L<<0) +#define BNX2_EMAC_LED_1000MB_OVERRIDE (1L<<1) +#define BNX2_EMAC_LED_100MB_OVERRIDE (1L<<2) +#define BNX2_EMAC_LED_10MB_OVERRIDE (1L<<3) +#define BNX2_EMAC_LED_TRAFFIC_OVERRIDE (1L<<4) +#define BNX2_EMAC_LED_BLNK_TRAFFIC (1L<<5) +#define BNX2_EMAC_LED_TRAFFIC (1L<<6) +#define BNX2_EMAC_LED_1000MB (1L<<7) +#define BNX2_EMAC_LED_100MB (1L<<8) +#define BNX2_EMAC_LED_10MB (1L<<9) +#define BNX2_EMAC_LED_TRAFFIC_STAT (1L<<10) +#define BNX2_EMAC_LED_BLNK_RATE (0xfffL<<19) +#define BNX2_EMAC_LED_BLNK_RATE_ENA (1L<<31) + +#define BNX2_EMAC_MAC_MATCH0 0x00001410 +#define BNX2_EMAC_MAC_MATCH1 0x00001414 +#define BNX2_EMAC_MAC_MATCH2 0x00001418 +#define BNX2_EMAC_MAC_MATCH3 0x0000141c +#define BNX2_EMAC_MAC_MATCH4 0x00001420 +#define BNX2_EMAC_MAC_MATCH5 0x00001424 +#define BNX2_EMAC_MAC_MATCH6 0x00001428 +#define BNX2_EMAC_MAC_MATCH7 0x0000142c +#define BNX2_EMAC_MAC_MATCH8 0x00001430 +#define BNX2_EMAC_MAC_MATCH9 0x00001434 +#define BNX2_EMAC_MAC_MATCH10 0x00001438 +#define BNX2_EMAC_MAC_MATCH11 0x0000143c +#define BNX2_EMAC_MAC_MATCH12 0x00001440 +#define BNX2_EMAC_MAC_MATCH13 0x00001444 +#define BNX2_EMAC_MAC_MATCH14 0x00001448 +#define BNX2_EMAC_MAC_MATCH15 0x0000144c +#define BNX2_EMAC_MAC_MATCH16 0x00001450 +#define BNX2_EMAC_MAC_MATCH17 0x00001454 +#define BNX2_EMAC_MAC_MATCH18 0x00001458 +#define BNX2_EMAC_MAC_MATCH19 0x0000145c +#define BNX2_EMAC_MAC_MATCH20 0x00001460 +#define BNX2_EMAC_MAC_MATCH21 0x00001464 +#define BNX2_EMAC_MAC_MATCH22 0x00001468 +#define BNX2_EMAC_MAC_MATCH23 0x0000146c +#define BNX2_EMAC_MAC_MATCH24 0x00001470 +#define BNX2_EMAC_MAC_MATCH25 0x00001474 +#define BNX2_EMAC_MAC_MATCH26 0x00001478 +#define BNX2_EMAC_MAC_MATCH27 0x0000147c +#define BNX2_EMAC_MAC_MATCH28 0x00001480 +#define BNX2_EMAC_MAC_MATCH29 0x00001484 +#define BNX2_EMAC_MAC_MATCH30 0x00001488 +#define BNX2_EMAC_MAC_MATCH31 0x0000148c +#define BNX2_EMAC_BACKOFF_SEED 0x00001498 +#define BNX2_EMAC_BACKOFF_SEED_EMAC_BACKOFF_SEED (0x3ffL<<0) + +#define BNX2_EMAC_RX_MTU_SIZE 0x0000149c +#define BNX2_EMAC_RX_MTU_SIZE_MTU_SIZE (0xffffL<<0) +#define BNX2_EMAC_RX_MTU_SIZE_JUMBO_ENA (1L<<31) + +#define BNX2_EMAC_SERDES_CNTL 0x000014a4 +#define BNX2_EMAC_SERDES_CNTL_RXR (0x7L<<0) +#define BNX2_EMAC_SERDES_CNTL_RXG (0x3L<<3) +#define BNX2_EMAC_SERDES_CNTL_RXCKSEL (1L<<6) +#define BNX2_EMAC_SERDES_CNTL_TXBIAS (0x7L<<7) +#define BNX2_EMAC_SERDES_CNTL_BGMAX (1L<<10) +#define BNX2_EMAC_SERDES_CNTL_BGMIN (1L<<11) +#define BNX2_EMAC_SERDES_CNTL_TXMODE (1L<<12) +#define BNX2_EMAC_SERDES_CNTL_TXEDGE (1L<<13) +#define BNX2_EMAC_SERDES_CNTL_SERDES_MODE (1L<<14) +#define BNX2_EMAC_SERDES_CNTL_PLLTEST (1L<<15) +#define BNX2_EMAC_SERDES_CNTL_CDET_EN (1L<<16) +#define BNX2_EMAC_SERDES_CNTL_TBI_LBK (1L<<17) +#define BNX2_EMAC_SERDES_CNTL_REMOTE_LBK (1L<<18) +#define BNX2_EMAC_SERDES_CNTL_REV_PHASE (1L<<19) +#define BNX2_EMAC_SERDES_CNTL_REGCTL12 (0x3L<<20) +#define BNX2_EMAC_SERDES_CNTL_REGCTL25 (0x3L<<22) + +#define BNX2_EMAC_SERDES_STATUS 0x000014a8 +#define BNX2_EMAC_SERDES_STATUS_RX_STAT (0xffL<<0) +#define BNX2_EMAC_SERDES_STATUS_COMMA_DET (1L<<8) + +#define BNX2_EMAC_MDIO_COMM 0x000014ac +#define BNX2_EMAC_MDIO_COMM_DATA (0xffffL<<0) +#define BNX2_EMAC_MDIO_COMM_REG_ADDR (0x1fL<<16) +#define BNX2_EMAC_MDIO_COMM_PHY_ADDR (0x1fL<<21) +#define BNX2_EMAC_MDIO_COMM_COMMAND (0x3L<<26) +#define BNX2_EMAC_MDIO_COMM_COMMAND_UNDEFINED_0 (0L<<26) +#define BNX2_EMAC_MDIO_COMM_COMMAND_WRITE (1L<<26) +#define BNX2_EMAC_MDIO_COMM_COMMAND_READ (2L<<26) +#define BNX2_EMAC_MDIO_COMM_COMMAND_UNDEFINED_3 (3L<<26) +#define BNX2_EMAC_MDIO_COMM_FAIL (1L<<28) +#define BNX2_EMAC_MDIO_COMM_START_BUSY (1L<<29) +#define BNX2_EMAC_MDIO_COMM_DISEXT (1L<<30) + +#define BNX2_EMAC_MDIO_STATUS 0x000014b0 +#define BNX2_EMAC_MDIO_STATUS_LINK (1L<<0) +#define BNX2_EMAC_MDIO_STATUS_10MB (1L<<1) + +#define BNX2_EMAC_MDIO_MODE 0x000014b4 +#define BNX2_EMAC_MDIO_MODE_SHORT_PREAMBLE (1L<<1) +#define BNX2_EMAC_MDIO_MODE_AUTO_POLL (1L<<4) +#define BNX2_EMAC_MDIO_MODE_BIT_BANG (1L<<8) +#define BNX2_EMAC_MDIO_MODE_MDIO (1L<<9) +#define BNX2_EMAC_MDIO_MODE_MDIO_OE (1L<<10) +#define BNX2_EMAC_MDIO_MODE_MDC (1L<<11) +#define BNX2_EMAC_MDIO_MODE_MDINT (1L<<12) +#define BNX2_EMAC_MDIO_MODE_CLOCK_CNT (0x1fL<<16) + +#define BNX2_EMAC_MDIO_AUTO_STATUS 0x000014b8 +#define BNX2_EMAC_MDIO_AUTO_STATUS_AUTO_ERR (1L<<0) + +#define BNX2_EMAC_TX_MODE 0x000014bc +#define BNX2_EMAC_TX_MODE_RESET (1L<<0) +#define BNX2_EMAC_TX_MODE_EXT_PAUSE_EN (1L<<3) +#define BNX2_EMAC_TX_MODE_FLOW_EN (1L<<4) +#define BNX2_EMAC_TX_MODE_BIG_BACKOFF (1L<<5) +#define BNX2_EMAC_TX_MODE_LONG_PAUSE (1L<<6) +#define BNX2_EMAC_TX_MODE_LINK_AWARE (1L<<7) + +#define BNX2_EMAC_TX_STATUS 0x000014c0 +#define BNX2_EMAC_TX_STATUS_XOFFED (1L<<0) +#define BNX2_EMAC_TX_STATUS_XOFF_SENT (1L<<1) +#define BNX2_EMAC_TX_STATUS_XON_SENT (1L<<2) +#define BNX2_EMAC_TX_STATUS_LINK_UP (1L<<3) +#define BNX2_EMAC_TX_STATUS_UNDERRUN (1L<<4) + +#define BNX2_EMAC_TX_LENGTHS 0x000014c4 +#define BNX2_EMAC_TX_LENGTHS_SLOT (0xffL<<0) +#define BNX2_EMAC_TX_LENGTHS_IPG (0xfL<<8) +#define BNX2_EMAC_TX_LENGTHS_IPG_CRS (0x3L<<12) + +#define BNX2_EMAC_RX_MODE 0x000014c8 +#define BNX2_EMAC_RX_MODE_RESET (1L<<0) +#define BNX2_EMAC_RX_MODE_FLOW_EN (1L<<2) +#define BNX2_EMAC_RX_MODE_KEEP_MAC_CONTROL (1L<<3) +#define BNX2_EMAC_RX_MODE_KEEP_PAUSE (1L<<4) +#define BNX2_EMAC_RX_MODE_ACCEPT_OVERSIZE (1L<<5) +#define BNX2_EMAC_RX_MODE_ACCEPT_RUNTS (1L<<6) +#define BNX2_EMAC_RX_MODE_LLC_CHK (1L<<7) +#define BNX2_EMAC_RX_MODE_PROMISCUOUS (1L<<8) +#define BNX2_EMAC_RX_MODE_NO_CRC_CHK (1L<<9) +#define BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG (1L<<10) +#define BNX2_EMAC_RX_MODE_FILT_BROADCAST (1L<<11) +#define BNX2_EMAC_RX_MODE_SORT_MODE (1L<<12) + +#define BNX2_EMAC_RX_STATUS 0x000014cc +#define BNX2_EMAC_RX_STATUS_FFED (1L<<0) +#define BNX2_EMAC_RX_STATUS_FF_RECEIVED (1L<<1) +#define BNX2_EMAC_RX_STATUS_N_RECEIVED (1L<<2) + +#define BNX2_EMAC_MULTICAST_HASH0 0x000014d0 +#define BNX2_EMAC_MULTICAST_HASH1 0x000014d4 +#define BNX2_EMAC_MULTICAST_HASH2 0x000014d8 +#define BNX2_EMAC_MULTICAST_HASH3 0x000014dc +#define BNX2_EMAC_MULTICAST_HASH4 0x000014e0 +#define BNX2_EMAC_MULTICAST_HASH5 0x000014e4 +#define BNX2_EMAC_MULTICAST_HASH6 0x000014e8 +#define BNX2_EMAC_MULTICAST_HASH7 0x000014ec +#define BNX2_EMAC_RX_STAT_IFHCINOCTETS 0x00001500 +#define BNX2_EMAC_RX_STAT_IFHCINBADOCTETS 0x00001504 +#define BNX2_EMAC_RX_STAT_ETHERSTATSFRAGMENTS 0x00001508 +#define BNX2_EMAC_RX_STAT_IFHCINUCASTPKTS 0x0000150c +#define BNX2_EMAC_RX_STAT_IFHCINMULTICASTPKTS 0x00001510 +#define BNX2_EMAC_RX_STAT_IFHCINBROADCASTPKTS 0x00001514 +#define BNX2_EMAC_RX_STAT_DOT3STATSFCSERRORS 0x00001518 +#define BNX2_EMAC_RX_STAT_DOT3STATSALIGNMENTERRORS 0x0000151c +#define BNX2_EMAC_RX_STAT_DOT3STATSCARRIERSENSEERRORS 0x00001520 +#define BNX2_EMAC_RX_STAT_XONPAUSEFRAMESRECEIVED 0x00001524 +#define BNX2_EMAC_RX_STAT_XOFFPAUSEFRAMESRECEIVED 0x00001528 +#define BNX2_EMAC_RX_STAT_MACCONTROLFRAMESRECEIVED 0x0000152c +#define BNX2_EMAC_RX_STAT_XOFFSTATEENTERED 0x00001530 +#define BNX2_EMAC_RX_STAT_DOT3STATSFRAMESTOOLONG 0x00001534 +#define BNX2_EMAC_RX_STAT_ETHERSTATSJABBERS 0x00001538 +#define BNX2_EMAC_RX_STAT_ETHERSTATSUNDERSIZEPKTS 0x0000153c +#define BNX2_EMAC_RX_STAT_ETHERSTATSPKTS64OCTETS 0x00001540 +#define BNX2_EMAC_RX_STAT_ETHERSTATSPKTS65OCTETSTO127OCTETS 0x00001544 +#define BNX2_EMAC_RX_STAT_ETHERSTATSPKTS128OCTETSTO255OCTETS 0x00001548 +#define BNX2_EMAC_RX_STAT_ETHERSTATSPKTS256OCTETSTO511OCTETS 0x0000154c +#define BNX2_EMAC_RX_STAT_ETHERSTATSPKTS512OCTETSTO1023OCTETS 0x00001550 +#define BNX2_EMAC_RX_STAT_ETHERSTATSPKTS1024OCTETSTO1522OCTETS 0x00001554 +#define BNX2_EMAC_RX_STAT_ETHERSTATSPKTS1523OCTETSTO9022OCTETS 0x00001558 +#define BNX2_EMAC_RXMAC_DEBUG0 0x0000155c +#define BNX2_EMAC_RXMAC_DEBUG1 0x00001560 +#define BNX2_EMAC_RXMAC_DEBUG1_LENGTH_NE_BYTE_COUNT (1L<<0) +#define BNX2_EMAC_RXMAC_DEBUG1_LENGTH_OUT_RANGE (1L<<1) +#define BNX2_EMAC_RXMAC_DEBUG1_BAD_CRC (1L<<2) +#define BNX2_EMAC_RXMAC_DEBUG1_RX_ERROR (1L<<3) +#define BNX2_EMAC_RXMAC_DEBUG1_ALIGN_ERROR (1L<<4) +#define BNX2_EMAC_RXMAC_DEBUG1_LAST_DATA (1L<<5) +#define BNX2_EMAC_RXMAC_DEBUG1_ODD_BYTE_START (1L<<6) +#define BNX2_EMAC_RXMAC_DEBUG1_BYTE_COUNT (0xffffL<<7) +#define BNX2_EMAC_RXMAC_DEBUG1_SLOT_TIME (0xffL<<23) + +#define BNX2_EMAC_RXMAC_DEBUG2 0x00001564 +#define BNX2_EMAC_RXMAC_DEBUG2_SM_STATE (0x7L<<0) +#define BNX2_EMAC_RXMAC_DEBUG2_SM_STATE_IDLE (0x0L<<0) +#define BNX2_EMAC_RXMAC_DEBUG2_SM_STATE_SFD (0x1L<<0) +#define BNX2_EMAC_RXMAC_DEBUG2_SM_STATE_DATA (0x2L<<0) +#define BNX2_EMAC_RXMAC_DEBUG2_SM_STATE_SKEEP (0x3L<<0) +#define BNX2_EMAC_RXMAC_DEBUG2_SM_STATE_EXT (0x4L<<0) +#define BNX2_EMAC_RXMAC_DEBUG2_SM_STATE_DROP (0x5L<<0) +#define BNX2_EMAC_RXMAC_DEBUG2_SM_STATE_SDROP (0x6L<<0) +#define BNX2_EMAC_RXMAC_DEBUG2_SM_STATE_FC (0x7L<<0) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE (0xfL<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE_IDLE (0x0L<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE_DATA0 (0x1L<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE_DATA1 (0x2L<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE_DATA2 (0x3L<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE_DATA3 (0x4L<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE_ABORT (0x5L<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE_WAIT (0x6L<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE_STATUS (0x7L<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_IDI_STATE_LAST (0x8L<<3) +#define BNX2_EMAC_RXMAC_DEBUG2_BYTE_IN (0xffL<<7) +#define BNX2_EMAC_RXMAC_DEBUG2_FALSEC (1L<<15) +#define BNX2_EMAC_RXMAC_DEBUG2_TAGGED (1L<<16) +#define BNX2_EMAC_RXMAC_DEBUG2_PAUSE_STATE (1L<<18) +#define BNX2_EMAC_RXMAC_DEBUG2_PAUSE_STATE_IDLE (0L<<18) +#define BNX2_EMAC_RXMAC_DEBUG2_PAUSE_STATE_PAUSED (1L<<18) +#define BNX2_EMAC_RXMAC_DEBUG2_SE_COUNTER (0xfL<<19) +#define BNX2_EMAC_RXMAC_DEBUG2_QUANTA (0x1fL<<23) + +#define BNX2_EMAC_RXMAC_DEBUG3 0x00001568 +#define BNX2_EMAC_RXMAC_DEBUG3_PAUSE_CTR (0xffffL<<0) +#define BNX2_EMAC_RXMAC_DEBUG3_TMP_PAUSE_CTR (0xffffL<<16) + +#define BNX2_EMAC_RXMAC_DEBUG4 0x0000156c +#define BNX2_EMAC_RXMAC_DEBUG4_TYPE_FIELD (0xffffL<<0) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE (0x3fL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_IDLE (0x0L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_UMAC2 (0x1L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_UMAC3 (0x2L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_UNI (0x3L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MMAC2 (0x7L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MMAC3 (0x5L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_PSA1 (0x6L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_PSA2 (0x7L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_PSA3 (0x8L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MC2 (0x9L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MC3 (0xaL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MWAIT1 (0xeL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MWAIT2 (0xfL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MCHECK (0x10L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MC (0x11L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_BC2 (0x12L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_BC3 (0x13L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_BSA1 (0x14L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_BSA2 (0x15L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_BSA3 (0x16L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_BTYPE (0x17L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_BC (0x18L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_PTYPE (0x19L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_CMD (0x1aL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MAC (0x1bL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_LATCH (0x1cL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_XOFF (0x1dL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_XON (0x1eL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_PAUSED (0x1fL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_NPAUSED (0x20L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_TTYPE (0x21L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_TVAL (0x22L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_USA1 (0x23L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_USA2 (0x24L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_USA3 (0x25L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_UTYPE (0x26L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_UTTYPE (0x27L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_UTVAL (0x28L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_MTYPE (0x29L<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_FILT_STATE_DROP (0x2aL<<16) +#define BNX2_EMAC_RXMAC_DEBUG4_DROP_PKT (1L<<22) +#define BNX2_EMAC_RXMAC_DEBUG4_SLOT_FILLED (1L<<23) +#define BNX2_EMAC_RXMAC_DEBUG4_FALSE_CARRIER (1L<<24) +#define BNX2_EMAC_RXMAC_DEBUG4_LAST_DATA (1L<<25) +#define BNX2_EMAC_RXMAC_DEBUG4_sfd_FOUND (1L<<26) +#define BNX2_EMAC_RXMAC_DEBUG4_ADVANCE (1L<<27) +#define BNX2_EMAC_RXMAC_DEBUG4_START (1L<<28) + +#define BNX2_EMAC_RXMAC_DEBUG5 0x00001570 +#define BNX2_EMAC_RXMAC_DEBUG5_PS_IDISM (0x7L<<0) +#define BNX2_EMAC_RXMAC_DEBUG5_PS_IDISM_IDLE (0L<<0) +#define BNX2_EMAC_RXMAC_DEBUG5_PS_IDISM_WAIT_EOF (1L<<0) +#define BNX2_EMAC_RXMAC_DEBUG5_PS_IDISM_WAIT_STAT (2L<<0) +#define BNX2_EMAC_RXMAC_DEBUG5_PS_IDISM_SET_EOF4FCRC (3L<<0) +#define BNX2_EMAC_RXMAC_DEBUG5_PS_IDISM_SET_EOF4RDE (4L<<0) +#define BNX2_EMAC_RXMAC_DEBUG5_PS_IDISM_SET_EOF4ALL (5L<<0) +#define BNX2_EMAC_RXMAC_DEBUG5_PS_IDISM_1WD_WAIT_STAT (6L<<0) +#define BNX2_EMAC_RXMAC_DEBUG5_CCODE_BUF1 (0x7L<<4) +#define BNX2_EMAC_RXMAC_DEBUG5_CCODE_BUF1_VDW (0x0L<<4) +#define BNX2_EMAC_RXMAC_DEBUG5_CCODE_BUF1_STAT (0x1L<<4) +#define BNX2_EMAC_RXMAC_DEBUG5_CCODE_BUF1_AEOF (0x2L<<4) +#define BNX2_EMAC_RXMAC_DEBUG5_CCODE_BUF1_NEOF (0x3L<<4) +#define BNX2_EMAC_RXMAC_DEBUG5_CCODE_BUF1_SOF (0x4L<<4) +#define BNX2_EMAC_RXMAC_DEBUG5_CCODE_BUF1_SAEOF (0x6L<<4) +#define BNX2_EMAC_RXMAC_DEBUG5_CCODE_BUF1_SNEOF (0x7L<<4) +#define BNX2_EMAC_RXMAC_DEBUG5_EOF_DETECTED (1L<<7) +#define BNX2_EMAC_RXMAC_DEBUG5_CCODE_BUF0 (0x7L<<8) +#define BNX2_EMAC_RXMAC_DEBUG5_RPM_IDI_FIFO_FULL (1L<<11) +#define BNX2_EMAC_RXMAC_DEBUG5_LOAD_CCODE (1L<<12) +#define BNX2_EMAC_RXMAC_DEBUG5_LOAD_DATA (1L<<13) +#define BNX2_EMAC_RXMAC_DEBUG5_LOAD_STAT (1L<<14) +#define BNX2_EMAC_RXMAC_DEBUG5_CLR_STAT (1L<<15) +#define BNX2_EMAC_RXMAC_DEBUG5_IDI_RPM_CCODE (0x3L<<16) +#define BNX2_EMAC_RXMAC_DEBUG5_IDI_RPM_ACCEPT (1L<<19) +#define BNX2_EMAC_RXMAC_DEBUG5_FMLEN (0xfffL<<20) + +#define BNX2_EMAC_RX_STAT_AC0 0x00001580 +#define BNX2_EMAC_RX_STAT_AC1 0x00001584 +#define BNX2_EMAC_RX_STAT_AC2 0x00001588 +#define BNX2_EMAC_RX_STAT_AC3 0x0000158c +#define BNX2_EMAC_RX_STAT_AC4 0x00001590 +#define BNX2_EMAC_RX_STAT_AC5 0x00001594 +#define BNX2_EMAC_RX_STAT_AC6 0x00001598 +#define BNX2_EMAC_RX_STAT_AC7 0x0000159c +#define BNX2_EMAC_RX_STAT_AC8 0x000015a0 +#define BNX2_EMAC_RX_STAT_AC9 0x000015a4 +#define BNX2_EMAC_RX_STAT_AC10 0x000015a8 +#define BNX2_EMAC_RX_STAT_AC11 0x000015ac +#define BNX2_EMAC_RX_STAT_AC12 0x000015b0 +#define BNX2_EMAC_RX_STAT_AC13 0x000015b4 +#define BNX2_EMAC_RX_STAT_AC14 0x000015b8 +#define BNX2_EMAC_RX_STAT_AC15 0x000015bc +#define BNX2_EMAC_RX_STAT_AC16 0x000015c0 +#define BNX2_EMAC_RX_STAT_AC17 0x000015c4 +#define BNX2_EMAC_RX_STAT_AC18 0x000015c8 +#define BNX2_EMAC_RX_STAT_AC19 0x000015cc +#define BNX2_EMAC_RX_STAT_AC20 0x000015d0 +#define BNX2_EMAC_RX_STAT_AC21 0x000015d4 +#define BNX2_EMAC_RX_STAT_AC22 0x000015d8 +#define BNX2_EMAC_RXMAC_SUC_DBG_OVERRUNVEC 0x000015dc +#define BNX2_EMAC_TX_STAT_IFHCOUTOCTETS 0x00001600 +#define BNX2_EMAC_TX_STAT_IFHCOUTBADOCTETS 0x00001604 +#define BNX2_EMAC_TX_STAT_ETHERSTATSCOLLISIONS 0x00001608 +#define BNX2_EMAC_TX_STAT_OUTXONSENT 0x0000160c +#define BNX2_EMAC_TX_STAT_OUTXOFFSENT 0x00001610 +#define BNX2_EMAC_TX_STAT_FLOWCONTROLDONE 0x00001614 +#define BNX2_EMAC_TX_STAT_DOT3STATSSINGLECOLLISIONFRAMES 0x00001618 +#define BNX2_EMAC_TX_STAT_DOT3STATSMULTIPLECOLLISIONFRAMES 0x0000161c +#define BNX2_EMAC_TX_STAT_DOT3STATSDEFERREDTRANSMISSIONS 0x00001620 +#define BNX2_EMAC_TX_STAT_DOT3STATSEXCESSIVECOLLISIONS 0x00001624 +#define BNX2_EMAC_TX_STAT_DOT3STATSLATECOLLISIONS 0x00001628 +#define BNX2_EMAC_TX_STAT_IFHCOUTUCASTPKTS 0x0000162c +#define BNX2_EMAC_TX_STAT_IFHCOUTMULTICASTPKTS 0x00001630 +#define BNX2_EMAC_TX_STAT_IFHCOUTBROADCASTPKTS 0x00001634 +#define BNX2_EMAC_TX_STAT_ETHERSTATSPKTS64OCTETS 0x00001638 +#define BNX2_EMAC_TX_STAT_ETHERSTATSPKTS65OCTETSTO127OCTETS 0x0000163c +#define BNX2_EMAC_TX_STAT_ETHERSTATSPKTS128OCTETSTO255OCTETS 0x00001640 +#define BNX2_EMAC_TX_STAT_ETHERSTATSPKTS256OCTETSTO511OCTETS 0x00001644 +#define BNX2_EMAC_TX_STAT_ETHERSTATSPKTS512OCTETSTO1023OCTETS 0x00001648 +#define BNX2_EMAC_TX_STAT_ETHERSTATSPKTS1024OCTETSTO1522OCTETS 0x0000164c +#define BNX2_EMAC_TX_STAT_ETHERSTATSPKTS1523OCTETSTO9022OCTETS 0x00001650 +#define BNX2_EMAC_TX_STAT_DOT3STATSINTERNALMACTRANSMITERRORS 0x00001654 +#define BNX2_EMAC_TXMAC_DEBUG0 0x00001658 +#define BNX2_EMAC_TXMAC_DEBUG1 0x0000165c +#define BNX2_EMAC_TXMAC_DEBUG1_ODI_STATE (0xfL<<0) +#define BNX2_EMAC_TXMAC_DEBUG1_ODI_STATE_IDLE (0x0L<<0) +#define BNX2_EMAC_TXMAC_DEBUG1_ODI_STATE_START0 (0x1L<<0) +#define BNX2_EMAC_TXMAC_DEBUG1_ODI_STATE_DATA0 (0x4L<<0) +#define BNX2_EMAC_TXMAC_DEBUG1_ODI_STATE_DATA1 (0x5L<<0) +#define BNX2_EMAC_TXMAC_DEBUG1_ODI_STATE_DATA2 (0x6L<<0) +#define BNX2_EMAC_TXMAC_DEBUG1_ODI_STATE_DATA3 (0x7L<<0) +#define BNX2_EMAC_TXMAC_DEBUG1_ODI_STATE_WAIT0 (0x8L<<0) +#define BNX2_EMAC_TXMAC_DEBUG1_ODI_STATE_WAIT1 (0x9L<<0) +#define BNX2_EMAC_TXMAC_DEBUG1_CRS_ENABLE (1L<<4) +#define BNX2_EMAC_TXMAC_DEBUG1_BAD_CRC (1L<<5) +#define BNX2_EMAC_TXMAC_DEBUG1_SE_COUNTER (0xfL<<6) +#define BNX2_EMAC_TXMAC_DEBUG1_SEND_PAUSE (1L<<10) +#define BNX2_EMAC_TXMAC_DEBUG1_LATE_COLLISION (1L<<11) +#define BNX2_EMAC_TXMAC_DEBUG1_MAX_DEFER (1L<<12) +#define BNX2_EMAC_TXMAC_DEBUG1_DEFERRED (1L<<13) +#define BNX2_EMAC_TXMAC_DEBUG1_ONE_BYTE (1L<<14) +#define BNX2_EMAC_TXMAC_DEBUG1_IPG_TIME (0xfL<<15) +#define BNX2_EMAC_TXMAC_DEBUG1_SLOT_TIME (0xffL<<19) + +#define BNX2_EMAC_TXMAC_DEBUG2 0x00001660 +#define BNX2_EMAC_TXMAC_DEBUG2_BACK_OFF (0x3ffL<<0) +#define BNX2_EMAC_TXMAC_DEBUG2_BYTE_COUNT (0xffffL<<10) +#define BNX2_EMAC_TXMAC_DEBUG2_COL_COUNT (0x1fL<<26) +#define BNX2_EMAC_TXMAC_DEBUG2_COL_BIT (1L<<31) + +#define BNX2_EMAC_TXMAC_DEBUG3 0x00001664 +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE (0xfL<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_IDLE (0x0L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_PRE1 (0x1L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_PRE2 (0x2L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_SFD (0x3L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_DATA (0x4L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_CRC1 (0x5L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_CRC2 (0x6L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_EXT (0x7L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_STATB (0x8L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_STATG (0x9L<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_JAM (0xaL<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_EJAM (0xbL<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_BJAM (0xcL<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_SWAIT (0xdL<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_SM_STATE_BACKOFF (0xeL<<0) +#define BNX2_EMAC_TXMAC_DEBUG3_FILT_STATE (0x7L<<4) +#define BNX2_EMAC_TXMAC_DEBUG3_FILT_STATE_IDLE (0x0L<<4) +#define BNX2_EMAC_TXMAC_DEBUG3_FILT_STATE_WAIT (0x1L<<4) +#define BNX2_EMAC_TXMAC_DEBUG3_FILT_STATE_UNI (0x2L<<4) +#define BNX2_EMAC_TXMAC_DEBUG3_FILT_STATE_MC (0x3L<<4) +#define BNX2_EMAC_TXMAC_DEBUG3_FILT_STATE_BC2 (0x4L<<4) +#define BNX2_EMAC_TXMAC_DEBUG3_FILT_STATE_BC3 (0x5L<<4) +#define BNX2_EMAC_TXMAC_DEBUG3_FILT_STATE_BC (0x6L<<4) +#define BNX2_EMAC_TXMAC_DEBUG3_CRS_DONE (1L<<7) +#define BNX2_EMAC_TXMAC_DEBUG3_XOFF (1L<<8) +#define BNX2_EMAC_TXMAC_DEBUG3_SE_COUNTER (0xfL<<9) +#define BNX2_EMAC_TXMAC_DEBUG3_QUANTA_COUNTER (0x1fL<<13) + +#define BNX2_EMAC_TXMAC_DEBUG4 0x00001668 +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_COUNTER (0xffffL<<0) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE (0xfL<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_IDLE (0x0L<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_MCA1 (0x2L<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_MCA2 (0x3L<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_MCA3 (0x6L<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_SRC1 (0x7L<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_SRC2 (0x5L<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_SRC3 (0x4L<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_TYPE (0xcL<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_CMD (0xeL<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_TIME (0xaL<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_CRC1 (0x8L<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_CRC2 (0x9L<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_PAUSE_STATE_WAIT (0xdL<<16) +#define BNX2_EMAC_TXMAC_DEBUG4_STATS0_VALID (1L<<20) +#define BNX2_EMAC_TXMAC_DEBUG4_APPEND_CRC (1L<<21) +#define BNX2_EMAC_TXMAC_DEBUG4_SLOT_FILLED (1L<<22) +#define BNX2_EMAC_TXMAC_DEBUG4_MAX_DEFER (1L<<23) +#define BNX2_EMAC_TXMAC_DEBUG4_SEND_EXTEND (1L<<24) +#define BNX2_EMAC_TXMAC_DEBUG4_SEND_PADDING (1L<<25) +#define BNX2_EMAC_TXMAC_DEBUG4_EOF_LOC (1L<<26) +#define BNX2_EMAC_TXMAC_DEBUG4_COLLIDING (1L<<27) +#define BNX2_EMAC_TXMAC_DEBUG4_COL_IN (1L<<28) +#define BNX2_EMAC_TXMAC_DEBUG4_BURSTING (1L<<29) +#define BNX2_EMAC_TXMAC_DEBUG4_ADVANCE (1L<<30) +#define BNX2_EMAC_TXMAC_DEBUG4_GO (1L<<31) + +#define BNX2_EMAC_TX_STAT_AC0 0x00001680 +#define BNX2_EMAC_TX_STAT_AC1 0x00001684 +#define BNX2_EMAC_TX_STAT_AC2 0x00001688 +#define BNX2_EMAC_TX_STAT_AC3 0x0000168c +#define BNX2_EMAC_TX_STAT_AC4 0x00001690 +#define BNX2_EMAC_TX_STAT_AC5 0x00001694 +#define BNX2_EMAC_TX_STAT_AC6 0x00001698 +#define BNX2_EMAC_TX_STAT_AC7 0x0000169c +#define BNX2_EMAC_TX_STAT_AC8 0x000016a0 +#define BNX2_EMAC_TX_STAT_AC9 0x000016a4 +#define BNX2_EMAC_TX_STAT_AC10 0x000016a8 +#define BNX2_EMAC_TX_STAT_AC11 0x000016ac +#define BNX2_EMAC_TX_STAT_AC12 0x000016b0 +#define BNX2_EMAC_TX_STAT_AC13 0x000016b4 +#define BNX2_EMAC_TX_STAT_AC14 0x000016b8 +#define BNX2_EMAC_TX_STAT_AC15 0x000016bc +#define BNX2_EMAC_TX_STAT_AC16 0x000016c0 +#define BNX2_EMAC_TX_STAT_AC17 0x000016c4 +#define BNX2_EMAC_TX_STAT_AC18 0x000016c8 +#define BNX2_EMAC_TX_STAT_AC19 0x000016cc +#define BNX2_EMAC_TX_STAT_AC20 0x000016d0 +#define BNX2_EMAC_TX_STAT_AC21 0x000016d4 +#define BNX2_EMAC_TXMAC_SUC_DBG_OVERRUNVEC 0x000016d8 + + +/* + * rpm_reg definition + * offset: 0x1800 + */ +#define BNX2_RPM_COMMAND 0x00001800 +#define BNX2_RPM_COMMAND_ENABLED (1L<<0) +#define BNX2_RPM_COMMAND_OVERRUN_ABORT (1L<<4) + +#define BNX2_RPM_STATUS 0x00001804 +#define BNX2_RPM_STATUS_MBUF_WAIT (1L<<0) +#define BNX2_RPM_STATUS_FREE_WAIT (1L<<1) + +#define BNX2_RPM_CONFIG 0x00001808 +#define BNX2_RPM_CONFIG_NO_PSD_HDR_CKSUM (1L<<0) +#define BNX2_RPM_CONFIG_ACPI_ENA (1L<<1) +#define BNX2_RPM_CONFIG_ACPI_KEEP (1L<<2) +#define BNX2_RPM_CONFIG_MP_KEEP (1L<<3) +#define BNX2_RPM_CONFIG_SORT_VECT_VAL (0xfL<<4) +#define BNX2_RPM_CONFIG_IGNORE_VLAN (1L<<31) + +#define BNX2_RPM_VLAN_MATCH0 0x00001810 +#define BNX2_RPM_VLAN_MATCH0_RPM_VLAN_MTCH0_VALUE (0xfffL<<0) + +#define BNX2_RPM_VLAN_MATCH1 0x00001814 +#define BNX2_RPM_VLAN_MATCH1_RPM_VLAN_MTCH1_VALUE (0xfffL<<0) + +#define BNX2_RPM_VLAN_MATCH2 0x00001818 +#define BNX2_RPM_VLAN_MATCH2_RPM_VLAN_MTCH2_VALUE (0xfffL<<0) + +#define BNX2_RPM_VLAN_MATCH3 0x0000181c +#define BNX2_RPM_VLAN_MATCH3_RPM_VLAN_MTCH3_VALUE (0xfffL<<0) + +#define BNX2_RPM_SORT_USER0 0x00001820 +#define BNX2_RPM_SORT_USER0_PM_EN (0xffffL<<0) +#define BNX2_RPM_SORT_USER0_BC_EN (1L<<16) +#define BNX2_RPM_SORT_USER0_MC_EN (1L<<17) +#define BNX2_RPM_SORT_USER0_MC_HSH_EN (1L<<18) +#define BNX2_RPM_SORT_USER0_PROM_EN (1L<<19) +#define BNX2_RPM_SORT_USER0_VLAN_EN (0xfL<<20) +#define BNX2_RPM_SORT_USER0_PROM_VLAN (1L<<24) +#define BNX2_RPM_SORT_USER0_ENA (1L<<31) + +#define BNX2_RPM_SORT_USER1 0x00001824 +#define BNX2_RPM_SORT_USER1_PM_EN (0xffffL<<0) +#define BNX2_RPM_SORT_USER1_BC_EN (1L<<16) +#define BNX2_RPM_SORT_USER1_MC_EN (1L<<17) +#define BNX2_RPM_SORT_USER1_MC_HSH_EN (1L<<18) +#define BNX2_RPM_SORT_USER1_PROM_EN (1L<<19) +#define BNX2_RPM_SORT_USER1_VLAN_EN (0xfL<<20) +#define BNX2_RPM_SORT_USER1_PROM_VLAN (1L<<24) +#define BNX2_RPM_SORT_USER1_ENA (1L<<31) + +#define BNX2_RPM_SORT_USER2 0x00001828 +#define BNX2_RPM_SORT_USER2_PM_EN (0xffffL<<0) +#define BNX2_RPM_SORT_USER2_BC_EN (1L<<16) +#define BNX2_RPM_SORT_USER2_MC_EN (1L<<17) +#define BNX2_RPM_SORT_USER2_MC_HSH_EN (1L<<18) +#define BNX2_RPM_SORT_USER2_PROM_EN (1L<<19) +#define BNX2_RPM_SORT_USER2_VLAN_EN (0xfL<<20) +#define BNX2_RPM_SORT_USER2_PROM_VLAN (1L<<24) +#define BNX2_RPM_SORT_USER2_ENA (1L<<31) + +#define BNX2_RPM_SORT_USER3 0x0000182c +#define BNX2_RPM_SORT_USER3_PM_EN (0xffffL<<0) +#define BNX2_RPM_SORT_USER3_BC_EN (1L<<16) +#define BNX2_RPM_SORT_USER3_MC_EN (1L<<17) +#define BNX2_RPM_SORT_USER3_MC_HSH_EN (1L<<18) +#define BNX2_RPM_SORT_USER3_PROM_EN (1L<<19) +#define BNX2_RPM_SORT_USER3_VLAN_EN (0xfL<<20) +#define BNX2_RPM_SORT_USER3_PROM_VLAN (1L<<24) +#define BNX2_RPM_SORT_USER3_ENA (1L<<31) + +#define BNX2_RPM_STAT_L2_FILTER_DISCARDS 0x00001840 +#define BNX2_RPM_STAT_RULE_CHECKER_DISCARDS 0x00001844 +#define BNX2_RPM_STAT_IFINFTQDISCARDS 0x00001848 +#define BNX2_RPM_STAT_IFINMBUFDISCARD 0x0000184c +#define BNX2_RPM_STAT_RULE_CHECKER_P4_HIT 0x00001850 +#define BNX2_RPM_STAT_AC0 0x00001880 +#define BNX2_RPM_STAT_AC1 0x00001884 +#define BNX2_RPM_STAT_AC2 0x00001888 +#define BNX2_RPM_STAT_AC3 0x0000188c +#define BNX2_RPM_STAT_AC4 0x00001890 +#define BNX2_RPM_RC_CNTL_0 0x00001900 +#define BNX2_RPM_RC_CNTL_0_OFFSET (0xffL<<0) +#define BNX2_RPM_RC_CNTL_0_CLASS (0x7L<<8) +#define BNX2_RPM_RC_CNTL_0_PRIORITY (1L<<11) +#define BNX2_RPM_RC_CNTL_0_P4 (1L<<12) +#define BNX2_RPM_RC_CNTL_0_HDR_TYPE (0x7L<<13) +#define BNX2_RPM_RC_CNTL_0_HDR_TYPE_START (0L<<13) +#define BNX2_RPM_RC_CNTL_0_HDR_TYPE_IP (1L<<13) +#define BNX2_RPM_RC_CNTL_0_HDR_TYPE_TCP (2L<<13) +#define BNX2_RPM_RC_CNTL_0_HDR_TYPE_UDP (3L<<13) +#define BNX2_RPM_RC_CNTL_0_HDR_TYPE_DATA (4L<<13) +#define BNX2_RPM_RC_CNTL_0_COMP (0x3L<<16) +#define BNX2_RPM_RC_CNTL_0_COMP_EQUAL (0L<<16) +#define BNX2_RPM_RC_CNTL_0_COMP_NEQUAL (1L<<16) +#define BNX2_RPM_RC_CNTL_0_COMP_GREATER (2L<<16) +#define BNX2_RPM_RC_CNTL_0_COMP_LESS (3L<<16) +#define BNX2_RPM_RC_CNTL_0_SBIT (1L<<19) +#define BNX2_RPM_RC_CNTL_0_CMDSEL (0xfL<<20) +#define BNX2_RPM_RC_CNTL_0_MAP (1L<<24) +#define BNX2_RPM_RC_CNTL_0_DISCARD (1L<<25) +#define BNX2_RPM_RC_CNTL_0_MASK (1L<<26) +#define BNX2_RPM_RC_CNTL_0_P1 (1L<<27) +#define BNX2_RPM_RC_CNTL_0_P2 (1L<<28) +#define BNX2_RPM_RC_CNTL_0_P3 (1L<<29) +#define BNX2_RPM_RC_CNTL_0_NBIT (1L<<30) + +#define BNX2_RPM_RC_VALUE_MASK_0 0x00001904 +#define BNX2_RPM_RC_VALUE_MASK_0_VALUE (0xffffL<<0) +#define BNX2_RPM_RC_VALUE_MASK_0_MASK (0xffffL<<16) + +#define BNX2_RPM_RC_CNTL_1 0x00001908 +#define BNX2_RPM_RC_CNTL_1_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_1_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_1 0x0000190c +#define BNX2_RPM_RC_CNTL_2 0x00001910 +#define BNX2_RPM_RC_CNTL_2_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_2_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_2 0x00001914 +#define BNX2_RPM_RC_CNTL_3 0x00001918 +#define BNX2_RPM_RC_CNTL_3_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_3_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_3 0x0000191c +#define BNX2_RPM_RC_CNTL_4 0x00001920 +#define BNX2_RPM_RC_CNTL_4_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_4_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_4 0x00001924 +#define BNX2_RPM_RC_CNTL_5 0x00001928 +#define BNX2_RPM_RC_CNTL_5_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_5_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_5 0x0000192c +#define BNX2_RPM_RC_CNTL_6 0x00001930 +#define BNX2_RPM_RC_CNTL_6_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_6_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_6 0x00001934 +#define BNX2_RPM_RC_CNTL_7 0x00001938 +#define BNX2_RPM_RC_CNTL_7_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_7_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_7 0x0000193c +#define BNX2_RPM_RC_CNTL_8 0x00001940 +#define BNX2_RPM_RC_CNTL_8_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_8_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_8 0x00001944 +#define BNX2_RPM_RC_CNTL_9 0x00001948 +#define BNX2_RPM_RC_CNTL_9_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_9_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_9 0x0000194c +#define BNX2_RPM_RC_CNTL_10 0x00001950 +#define BNX2_RPM_RC_CNTL_10_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_10_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_10 0x00001954 +#define BNX2_RPM_RC_CNTL_11 0x00001958 +#define BNX2_RPM_RC_CNTL_11_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_11_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_11 0x0000195c +#define BNX2_RPM_RC_CNTL_12 0x00001960 +#define BNX2_RPM_RC_CNTL_12_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_12_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_12 0x00001964 +#define BNX2_RPM_RC_CNTL_13 0x00001968 +#define BNX2_RPM_RC_CNTL_13_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_13_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_13 0x0000196c +#define BNX2_RPM_RC_CNTL_14 0x00001970 +#define BNX2_RPM_RC_CNTL_14_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_14_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_14 0x00001974 +#define BNX2_RPM_RC_CNTL_15 0x00001978 +#define BNX2_RPM_RC_CNTL_15_A (0x3ffffL<<0) +#define BNX2_RPM_RC_CNTL_15_B (0xfffL<<19) + +#define BNX2_RPM_RC_VALUE_MASK_15 0x0000197c +#define BNX2_RPM_RC_CONFIG 0x00001980 +#define BNX2_RPM_RC_CONFIG_RULE_ENABLE (0xffffL<<0) +#define BNX2_RPM_RC_CONFIG_DEF_CLASS (0x7L<<24) + +#define BNX2_RPM_DEBUG0 0x00001984 +#define BNX2_RPM_DEBUG0_FM_BCNT (0xffffL<<0) +#define BNX2_RPM_DEBUG0_T_DATA_OFST_VLD (1L<<16) +#define BNX2_RPM_DEBUG0_T_UDP_OFST_VLD (1L<<17) +#define BNX2_RPM_DEBUG0_T_TCP_OFST_VLD (1L<<18) +#define BNX2_RPM_DEBUG0_T_IP_OFST_VLD (1L<<19) +#define BNX2_RPM_DEBUG0_IP_MORE_FRGMT (1L<<20) +#define BNX2_RPM_DEBUG0_T_IP_NO_TCP_UDP_HDR (1L<<21) +#define BNX2_RPM_DEBUG0_LLC_SNAP (1L<<22) +#define BNX2_RPM_DEBUG0_FM_STARTED (1L<<23) +#define BNX2_RPM_DEBUG0_DONE (1L<<24) +#define BNX2_RPM_DEBUG0_WAIT_4_DONE (1L<<25) +#define BNX2_RPM_DEBUG0_USE_TPBUF_CKSUM (1L<<26) +#define BNX2_RPM_DEBUG0_RX_NO_PSD_HDR_CKSUM (1L<<27) +#define BNX2_RPM_DEBUG0_IGNORE_VLAN (1L<<28) +#define BNX2_RPM_DEBUG0_RP_ENA_ACTIVE (1L<<31) + +#define BNX2_RPM_DEBUG1 0x00001988 +#define BNX2_RPM_DEBUG1_FSM_CUR_ST (0xffffL<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_IDLE (0L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_ETYPE_B6_ALL (1L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_ETYPE_B2_IPLLC (2L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_ETYPE_B6_IP (4L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_ETYPE_B2_IP (8L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_IP_START (16L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_IP (32L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_TCP (64L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_UDP (128L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_AH (256L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_ESP (512L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_ESP_PAYLOAD (1024L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_DATA (2048L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_ADD_CARRY (0x2000L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_ADD_CARRYOUT (0x4000L<<0) +#define BNX2_RPM_DEBUG1_FSM_CUR_ST_LATCH_RESULT (0x8000L<<0) +#define BNX2_RPM_DEBUG1_HDR_BCNT (0x7ffL<<16) +#define BNX2_RPM_DEBUG1_UNKNOWN_ETYPE_D (1L<<28) +#define BNX2_RPM_DEBUG1_VLAN_REMOVED_D2 (1L<<29) +#define BNX2_RPM_DEBUG1_VLAN_REMOVED_D1 (1L<<30) +#define BNX2_RPM_DEBUG1_EOF_0XTRA_WD (1L<<31) + +#define BNX2_RPM_DEBUG2 0x0000198c +#define BNX2_RPM_DEBUG2_CMD_HIT_VEC (0xffffL<<0) +#define BNX2_RPM_DEBUG2_IP_BCNT (0xffL<<16) +#define BNX2_RPM_DEBUG2_THIS_CMD_M4 (1L<<24) +#define BNX2_RPM_DEBUG2_THIS_CMD_M3 (1L<<25) +#define BNX2_RPM_DEBUG2_THIS_CMD_M2 (1L<<26) +#define BNX2_RPM_DEBUG2_THIS_CMD_M1 (1L<<27) +#define BNX2_RPM_DEBUG2_IPIPE_EMPTY (1L<<28) +#define BNX2_RPM_DEBUG2_FM_DISCARD (1L<<29) +#define BNX2_RPM_DEBUG2_LAST_RULE_IN_FM_D2 (1L<<30) +#define BNX2_RPM_DEBUG2_LAST_RULE_IN_FM_D1 (1L<<31) + +#define BNX2_RPM_DEBUG3 0x00001990 +#define BNX2_RPM_DEBUG3_AVAIL_MBUF_PTR (0x1ffL<<0) +#define BNX2_RPM_DEBUG3_RDE_RLUPQ_WR_REQ_INT (1L<<9) +#define BNX2_RPM_DEBUG3_RDE_RBUF_WR_LAST_INT (1L<<10) +#define BNX2_RPM_DEBUG3_RDE_RBUF_WR_REQ_INT (1L<<11) +#define BNX2_RPM_DEBUG3_RDE_RBUF_FREE_REQ (1L<<12) +#define BNX2_RPM_DEBUG3_RDE_RBUF_ALLOC_REQ (1L<<13) +#define BNX2_RPM_DEBUG3_DFSM_MBUF_NOTAVAIL (1L<<14) +#define BNX2_RPM_DEBUG3_RBUF_RDE_SOF_DROP (1L<<15) +#define BNX2_RPM_DEBUG3_DFIFO_VLD_ENTRY_CT (0xfL<<16) +#define BNX2_RPM_DEBUG3_RDE_SRC_FIFO_ALMFULL (1L<<21) +#define BNX2_RPM_DEBUG3_DROP_NXT_VLD (1L<<22) +#define BNX2_RPM_DEBUG3_DROP_NXT (1L<<23) +#define BNX2_RPM_DEBUG3_FTQ_FSM (0x3L<<24) +#define BNX2_RPM_DEBUG3_FTQ_FSM_IDLE (0x0L<<24) +#define BNX2_RPM_DEBUG3_FTQ_FSM_WAIT_ACK (0x1L<<24) +#define BNX2_RPM_DEBUG3_FTQ_FSM_WAIT_FREE (0x2L<<24) +#define BNX2_RPM_DEBUG3_MBWRITE_FSM (0x3L<<26) +#define BNX2_RPM_DEBUG3_MBWRITE_FSM_WAIT_SOF (0x0L<<26) +#define BNX2_RPM_DEBUG3_MBWRITE_FSM_GET_MBUF (0x1L<<26) +#define BNX2_RPM_DEBUG3_MBWRITE_FSM_DMA_DATA (0x2L<<26) +#define BNX2_RPM_DEBUG3_MBWRITE_FSM_WAIT_DATA (0x3L<<26) +#define BNX2_RPM_DEBUG3_MBWRITE_FSM_WAIT_EOF (0x4L<<26) +#define BNX2_RPM_DEBUG3_MBWRITE_FSM_WAIT_MF_ACK (0x5L<<26) +#define BNX2_RPM_DEBUG3_MBWRITE_FSM_WAIT_DROP_NXT_VLD (0x6L<<26) +#define BNX2_RPM_DEBUG3_MBWRITE_FSM_DONE (0x7L<<26) +#define BNX2_RPM_DEBUG3_MBFREE_FSM (1L<<29) +#define BNX2_RPM_DEBUG3_MBFREE_FSM_IDLE (0L<<29) +#define BNX2_RPM_DEBUG3_MBFREE_FSM_WAIT_ACK (1L<<29) +#define BNX2_RPM_DEBUG3_MBALLOC_FSM (1L<<30) +#define BNX2_RPM_DEBUG3_MBALLOC_FSM_ET_MBUF (0x0L<<30) +#define BNX2_RPM_DEBUG3_MBALLOC_FSM_IVE_MBUF (0x1L<<30) +#define BNX2_RPM_DEBUG3_CCODE_EOF_ERROR (1L<<31) + +#define BNX2_RPM_DEBUG4 0x00001994 +#define BNX2_RPM_DEBUG4_DFSM_MBUF_CLUSTER (0x1ffffffL<<0) +#define BNX2_RPM_DEBUG4_DFIFO_CUR_CCODE (0x7L<<25) +#define BNX2_RPM_DEBUG4_MBWRITE_FSM (0x7L<<28) +#define BNX2_RPM_DEBUG4_DFIFO_EMPTY (1L<<31) + +#define BNX2_RPM_DEBUG5 0x00001998 +#define BNX2_RPM_DEBUG5_RDROP_WPTR (0x1fL<<0) +#define BNX2_RPM_DEBUG5_RDROP_ACPI_RPTR (0x1fL<<5) +#define BNX2_RPM_DEBUG5_RDROP_MC_RPTR (0x1fL<<10) +#define BNX2_RPM_DEBUG5_RDROP_RC_RPTR (0x1fL<<15) +#define BNX2_RPM_DEBUG5_RDROP_ACPI_EMPTY (1L<<20) +#define BNX2_RPM_DEBUG5_RDROP_MC_EMPTY (1L<<21) +#define BNX2_RPM_DEBUG5_RDROP_AEOF_VEC_AT_RDROP_MC_RPTR (1L<<22) +#define BNX2_RPM_DEBUG5_HOLDREG_WOL_DROP_INT (1L<<23) +#define BNX2_RPM_DEBUG5_HOLDREG_DISCARD (1L<<24) +#define BNX2_RPM_DEBUG5_HOLDREG_MBUF_NOTAVAIL (1L<<25) +#define BNX2_RPM_DEBUG5_HOLDREG_MC_EMPTY (1L<<26) +#define BNX2_RPM_DEBUG5_HOLDREG_RC_EMPTY (1L<<27) +#define BNX2_RPM_DEBUG5_HOLDREG_FC_EMPTY (1L<<28) +#define BNX2_RPM_DEBUG5_HOLDREG_ACPI_EMPTY (1L<<29) +#define BNX2_RPM_DEBUG5_HOLDREG_FULL_T (1L<<30) +#define BNX2_RPM_DEBUG5_HOLDREG_RD (1L<<31) + +#define BNX2_RPM_DEBUG6 0x0000199c +#define BNX2_RPM_DEBUG6_ACPI_VEC (0xffffL<<0) +#define BNX2_RPM_DEBUG6_VEC (0xffffL<<16) + +#define BNX2_RPM_DEBUG7 0x000019a0 +#define BNX2_RPM_DEBUG7_RPM_DBG7_LAST_CRC (0xffffffffL<<0) + +#define BNX2_RPM_DEBUG8 0x000019a4 +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM (0xfL<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_IDLE (0L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_SOF_W1_ADDR (1L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_SOF_W2_ADDR (2L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_SOF_W3_ADDR (3L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_SOF_WAIT_THBUF (4L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_W3_DATA (5L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_W0_ADDR (6L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_W1_ADDR (7L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_W2_ADDR (8L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_W3_ADDR (9L<<0) +#define BNX2_RPM_DEBUG8_PS_ACPI_FSM_WAIT_THBUF (10L<<0) +#define BNX2_RPM_DEBUG8_COMPARE_AT_W0 (1L<<4) +#define BNX2_RPM_DEBUG8_COMPARE_AT_W3_DATA (1L<<5) +#define BNX2_RPM_DEBUG8_COMPARE_AT_SOF_WAIT (1L<<6) +#define BNX2_RPM_DEBUG8_COMPARE_AT_SOF_W3 (1L<<7) +#define BNX2_RPM_DEBUG8_COMPARE_AT_SOF_W2 (1L<<8) +#define BNX2_RPM_DEBUG8_EOF_W_LTEQ6_VLDBYTES (1L<<9) +#define BNX2_RPM_DEBUG8_EOF_W_LTEQ4_VLDBYTES (1L<<10) +#define BNX2_RPM_DEBUG8_NXT_EOF_W_12_VLDBYTES (1L<<11) +#define BNX2_RPM_DEBUG8_EOF_DET (1L<<12) +#define BNX2_RPM_DEBUG8_SOF_DET (1L<<13) +#define BNX2_RPM_DEBUG8_WAIT_4_SOF (1L<<14) +#define BNX2_RPM_DEBUG8_ALL_DONE (1L<<15) +#define BNX2_RPM_DEBUG8_THBUF_ADDR (0x7fL<<16) +#define BNX2_RPM_DEBUG8_BYTE_CTR (0xffL<<24) + +#define BNX2_RPM_DEBUG9 0x000019a8 +#define BNX2_RPM_DEBUG9_OUTFIFO_COUNT (0x7L<<0) +#define BNX2_RPM_DEBUG9_RDE_ACPI_RDY (1L<<3) +#define BNX2_RPM_DEBUG9_VLD_RD_ENTRY_CT (0x7L<<4) +#define BNX2_RPM_DEBUG9_OUTFIFO_OVERRUN_OCCURRED (1L<<28) +#define BNX2_RPM_DEBUG9_INFIFO_OVERRUN_OCCURRED (1L<<29) +#define BNX2_RPM_DEBUG9_ACPI_MATCH_INT (1L<<30) +#define BNX2_RPM_DEBUG9_ACPI_ENABLE_SYN (1L<<31) + +#define BNX2_RPM_ACPI_DBG_BUF_W00 0x000019c0 +#define BNX2_RPM_ACPI_DBG_BUF_W01 0x000019c4 +#define BNX2_RPM_ACPI_DBG_BUF_W02 0x000019c8 +#define BNX2_RPM_ACPI_DBG_BUF_W03 0x000019cc +#define BNX2_RPM_ACPI_DBG_BUF_W10 0x000019d0 +#define BNX2_RPM_ACPI_DBG_BUF_W11 0x000019d4 +#define BNX2_RPM_ACPI_DBG_BUF_W12 0x000019d8 +#define BNX2_RPM_ACPI_DBG_BUF_W13 0x000019dc +#define BNX2_RPM_ACPI_DBG_BUF_W20 0x000019e0 +#define BNX2_RPM_ACPI_DBG_BUF_W21 0x000019e4 +#define BNX2_RPM_ACPI_DBG_BUF_W22 0x000019e8 +#define BNX2_RPM_ACPI_DBG_BUF_W23 0x000019ec +#define BNX2_RPM_ACPI_DBG_BUF_W30 0x000019f0 +#define BNX2_RPM_ACPI_DBG_BUF_W31 0x000019f4 +#define BNX2_RPM_ACPI_DBG_BUF_W32 0x000019f8 +#define BNX2_RPM_ACPI_DBG_BUF_W33 0x000019fc + + +/* + * rbuf_reg definition + * offset: 0x200000 + */ +#define BNX2_RBUF_COMMAND 0x00200000 +#define BNX2_RBUF_COMMAND_ENABLED (1L<<0) +#define BNX2_RBUF_COMMAND_FREE_INIT (1L<<1) +#define BNX2_RBUF_COMMAND_RAM_INIT (1L<<2) +#define BNX2_RBUF_COMMAND_OVER_FREE (1L<<4) +#define BNX2_RBUF_COMMAND_ALLOC_REQ (1L<<5) + +#define BNX2_RBUF_STATUS1 0x00200004 +#define BNX2_RBUF_STATUS1_FREE_COUNT (0x3ffL<<0) + +#define BNX2_RBUF_STATUS2 0x00200008 +#define BNX2_RBUF_STATUS2_FREE_TAIL (0x3ffL<<0) +#define BNX2_RBUF_STATUS2_FREE_HEAD (0x3ffL<<16) + +#define BNX2_RBUF_CONFIG 0x0020000c +#define BNX2_RBUF_CONFIG_XOFF_TRIP (0x3ffL<<0) +#define BNX2_RBUF_CONFIG_XON_TRIP (0x3ffL<<16) + +#define BNX2_RBUF_FW_BUF_ALLOC 0x00200010 +#define BNX2_RBUF_FW_BUF_ALLOC_VALUE (0x1ffL<<7) + +#define BNX2_RBUF_FW_BUF_FREE 0x00200014 +#define BNX2_RBUF_FW_BUF_FREE_COUNT (0x7fL<<0) +#define BNX2_RBUF_FW_BUF_FREE_TAIL (0x1ffL<<7) +#define BNX2_RBUF_FW_BUF_FREE_HEAD (0x1ffL<<16) + +#define BNX2_RBUF_FW_BUF_SEL 0x00200018 +#define BNX2_RBUF_FW_BUF_SEL_COUNT (0x7fL<<0) +#define BNX2_RBUF_FW_BUF_SEL_TAIL (0x1ffL<<7) +#define BNX2_RBUF_FW_BUF_SEL_HEAD (0x1ffL<<16) + +#define BNX2_RBUF_CONFIG2 0x0020001c +#define BNX2_RBUF_CONFIG2_MAC_DROP_TRIP (0x3ffL<<0) +#define BNX2_RBUF_CONFIG2_MAC_KEEP_TRIP (0x3ffL<<16) + +#define BNX2_RBUF_CONFIG3 0x00200020 +#define BNX2_RBUF_CONFIG3_CU_DROP_TRIP (0x3ffL<<0) +#define BNX2_RBUF_CONFIG3_CU_KEEP_TRIP (0x3ffL<<16) + +#define BNX2_RBUF_PKT_DATA 0x00208000 +#define BNX2_RBUF_CLIST_DATA 0x00210000 +#define BNX2_RBUF_BUF_DATA 0x00220000 + + +/* + * rv2p_reg definition + * offset: 0x2800 + */ +#define BNX2_RV2P_COMMAND 0x00002800 +#define BNX2_RV2P_COMMAND_ENABLED (1L<<0) +#define BNX2_RV2P_COMMAND_PROC1_INTRPT (1L<<1) +#define BNX2_RV2P_COMMAND_PROC2_INTRPT (1L<<2) +#define BNX2_RV2P_COMMAND_ABORT0 (1L<<4) +#define BNX2_RV2P_COMMAND_ABORT1 (1L<<5) +#define BNX2_RV2P_COMMAND_ABORT2 (1L<<6) +#define BNX2_RV2P_COMMAND_ABORT3 (1L<<7) +#define BNX2_RV2P_COMMAND_ABORT4 (1L<<8) +#define BNX2_RV2P_COMMAND_ABORT5 (1L<<9) +#define BNX2_RV2P_COMMAND_PROC1_RESET (1L<<16) +#define BNX2_RV2P_COMMAND_PROC2_RESET (1L<<17) +#define BNX2_RV2P_COMMAND_CTXIF_RESET (1L<<18) + +#define BNX2_RV2P_STATUS 0x00002804 +#define BNX2_RV2P_STATUS_ALWAYS_0 (1L<<0) +#define BNX2_RV2P_STATUS_RV2P_GEN_STAT0_CNT (1L<<8) +#define BNX2_RV2P_STATUS_RV2P_GEN_STAT1_CNT (1L<<9) +#define BNX2_RV2P_STATUS_RV2P_GEN_STAT2_CNT (1L<<10) +#define BNX2_RV2P_STATUS_RV2P_GEN_STAT3_CNT (1L<<11) +#define BNX2_RV2P_STATUS_RV2P_GEN_STAT4_CNT (1L<<12) +#define BNX2_RV2P_STATUS_RV2P_GEN_STAT5_CNT (1L<<13) + +#define BNX2_RV2P_CONFIG 0x00002808 +#define BNX2_RV2P_CONFIG_STALL_PROC1 (1L<<0) +#define BNX2_RV2P_CONFIG_STALL_PROC2 (1L<<1) +#define BNX2_RV2P_CONFIG_PROC1_STALL_ON_ABORT0 (1L<<8) +#define BNX2_RV2P_CONFIG_PROC1_STALL_ON_ABORT1 (1L<<9) +#define BNX2_RV2P_CONFIG_PROC1_STALL_ON_ABORT2 (1L<<10) +#define BNX2_RV2P_CONFIG_PROC1_STALL_ON_ABORT3 (1L<<11) +#define BNX2_RV2P_CONFIG_PROC1_STALL_ON_ABORT4 (1L<<12) +#define BNX2_RV2P_CONFIG_PROC1_STALL_ON_ABORT5 (1L<<13) +#define BNX2_RV2P_CONFIG_PROC2_STALL_ON_ABORT0 (1L<<16) +#define BNX2_RV2P_CONFIG_PROC2_STALL_ON_ABORT1 (1L<<17) +#define BNX2_RV2P_CONFIG_PROC2_STALL_ON_ABORT2 (1L<<18) +#define BNX2_RV2P_CONFIG_PROC2_STALL_ON_ABORT3 (1L<<19) +#define BNX2_RV2P_CONFIG_PROC2_STALL_ON_ABORT4 (1L<<20) +#define BNX2_RV2P_CONFIG_PROC2_STALL_ON_ABORT5 (1L<<21) +#define BNX2_RV2P_CONFIG_PAGE_SIZE (0xfL<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_256 (0L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_512 (1L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_1K (2L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_2K (3L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_4K (4L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_8K (5L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_16K (6L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_32K (7L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_64K (8L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_128K (9L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_256K (10L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_512K (11L<<24) +#define BNX2_RV2P_CONFIG_PAGE_SIZE_1M (12L<<24) + +#define BNX2_RV2P_GEN_BFR_ADDR_0 0x00002810 +#define BNX2_RV2P_GEN_BFR_ADDR_0_VALUE (0xffffL<<16) + +#define BNX2_RV2P_GEN_BFR_ADDR_1 0x00002814 +#define BNX2_RV2P_GEN_BFR_ADDR_1_VALUE (0xffffL<<16) + +#define BNX2_RV2P_GEN_BFR_ADDR_2 0x00002818 +#define BNX2_RV2P_GEN_BFR_ADDR_2_VALUE (0xffffL<<16) + +#define BNX2_RV2P_GEN_BFR_ADDR_3 0x0000281c +#define BNX2_RV2P_GEN_BFR_ADDR_3_VALUE (0xffffL<<16) + +#define BNX2_RV2P_INSTR_HIGH 0x00002830 +#define BNX2_RV2P_INSTR_HIGH_HIGH (0x1fL<<0) + +#define BNX2_RV2P_INSTR_LOW 0x00002834 +#define BNX2_RV2P_PROC1_ADDR_CMD 0x00002838 +#define BNX2_RV2P_PROC1_ADDR_CMD_ADD (0x3ffL<<0) +#define BNX2_RV2P_PROC1_ADDR_CMD_RDWR (1L<<31) + +#define BNX2_RV2P_PROC2_ADDR_CMD 0x0000283c +#define BNX2_RV2P_PROC2_ADDR_CMD_ADD (0x3ffL<<0) +#define BNX2_RV2P_PROC2_ADDR_CMD_RDWR (1L<<31) + +#define BNX2_RV2P_PROC1_GRC_DEBUG 0x00002840 +#define BNX2_RV2P_PROC2_GRC_DEBUG 0x00002844 +#define BNX2_RV2P_GRC_PROC_DEBUG 0x00002848 +#define BNX2_RV2P_DEBUG_VECT_PEEK 0x0000284c +#define BNX2_RV2P_DEBUG_VECT_PEEK_1_VALUE (0x7ffL<<0) +#define BNX2_RV2P_DEBUG_VECT_PEEK_1_PEEK_EN (1L<<11) +#define BNX2_RV2P_DEBUG_VECT_PEEK_1_SEL (0xfL<<12) +#define BNX2_RV2P_DEBUG_VECT_PEEK_2_VALUE (0x7ffL<<16) +#define BNX2_RV2P_DEBUG_VECT_PEEK_2_PEEK_EN (1L<<27) +#define BNX2_RV2P_DEBUG_VECT_PEEK_2_SEL (0xfL<<28) + +#define BNX2_RV2P_PFTQ_DATA 0x00002b40 +#define BNX2_RV2P_PFTQ_CMD 0x00002b78 +#define BNX2_RV2P_PFTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_RV2P_PFTQ_CMD_WR_TOP (1L<<10) +#define BNX2_RV2P_PFTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_RV2P_PFTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_RV2P_PFTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_RV2P_PFTQ_CMD_RD_DATA (1L<<26) +#define BNX2_RV2P_PFTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_RV2P_PFTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_RV2P_PFTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_RV2P_PFTQ_CMD_POP (1L<<30) +#define BNX2_RV2P_PFTQ_CMD_BUSY (1L<<31) + +#define BNX2_RV2P_PFTQ_CTL 0x00002b7c +#define BNX2_RV2P_PFTQ_CTL_INTERVENE (1L<<0) +#define BNX2_RV2P_PFTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_RV2P_PFTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_RV2P_PFTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_RV2P_PFTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_RV2P_TFTQ_DATA 0x00002b80 +#define BNX2_RV2P_TFTQ_CMD 0x00002bb8 +#define BNX2_RV2P_TFTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_RV2P_TFTQ_CMD_WR_TOP (1L<<10) +#define BNX2_RV2P_TFTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_RV2P_TFTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_RV2P_TFTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_RV2P_TFTQ_CMD_RD_DATA (1L<<26) +#define BNX2_RV2P_TFTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_RV2P_TFTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_RV2P_TFTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_RV2P_TFTQ_CMD_POP (1L<<30) +#define BNX2_RV2P_TFTQ_CMD_BUSY (1L<<31) + +#define BNX2_RV2P_TFTQ_CTL 0x00002bbc +#define BNX2_RV2P_TFTQ_CTL_INTERVENE (1L<<0) +#define BNX2_RV2P_TFTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_RV2P_TFTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_RV2P_TFTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_RV2P_TFTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_RV2P_MFTQ_DATA 0x00002bc0 +#define BNX2_RV2P_MFTQ_CMD 0x00002bf8 +#define BNX2_RV2P_MFTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_RV2P_MFTQ_CMD_WR_TOP (1L<<10) +#define BNX2_RV2P_MFTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_RV2P_MFTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_RV2P_MFTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_RV2P_MFTQ_CMD_RD_DATA (1L<<26) +#define BNX2_RV2P_MFTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_RV2P_MFTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_RV2P_MFTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_RV2P_MFTQ_CMD_POP (1L<<30) +#define BNX2_RV2P_MFTQ_CMD_BUSY (1L<<31) + +#define BNX2_RV2P_MFTQ_CTL 0x00002bfc +#define BNX2_RV2P_MFTQ_CTL_INTERVENE (1L<<0) +#define BNX2_RV2P_MFTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_RV2P_MFTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_RV2P_MFTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_RV2P_MFTQ_CTL_CUR_DEPTH (0x3ffL<<22) + + + +/* + * mq_reg definition + * offset: 0x3c00 + */ +#define BNX2_MQ_COMMAND 0x00003c00 +#define BNX2_MQ_COMMAND_ENABLED (1L<<0) +#define BNX2_MQ_COMMAND_OVERFLOW (1L<<4) +#define BNX2_MQ_COMMAND_WR_ERROR (1L<<5) +#define BNX2_MQ_COMMAND_RD_ERROR (1L<<6) + +#define BNX2_MQ_STATUS 0x00003c04 +#define BNX2_MQ_STATUS_CTX_ACCESS_STAT (1L<<16) +#define BNX2_MQ_STATUS_CTX_ACCESS64_STAT (1L<<17) +#define BNX2_MQ_STATUS_PCI_STALL_STAT (1L<<18) + +#define BNX2_MQ_CONFIG 0x00003c08 +#define BNX2_MQ_CONFIG_TX_HIGH_PRI (1L<<0) +#define BNX2_MQ_CONFIG_HALT_DIS (1L<<1) +#define BNX2_MQ_CONFIG_KNL_BYP_BLK_SIZE (0x7L<<4) +#define BNX2_MQ_CONFIG_KNL_BYP_BLK_SIZE_256 (0L<<4) +#define BNX2_MQ_CONFIG_KNL_BYP_BLK_SIZE_512 (1L<<4) +#define BNX2_MQ_CONFIG_KNL_BYP_BLK_SIZE_1K (2L<<4) +#define BNX2_MQ_CONFIG_KNL_BYP_BLK_SIZE_2K (3L<<4) +#define BNX2_MQ_CONFIG_KNL_BYP_BLK_SIZE_4K (4L<<4) +#define BNX2_MQ_CONFIG_MAX_DEPTH (0x7fL<<8) +#define BNX2_MQ_CONFIG_CUR_DEPTH (0x7fL<<20) + +#define BNX2_MQ_ENQUEUE1 0x00003c0c +#define BNX2_MQ_ENQUEUE1_OFFSET (0x3fL<<2) +#define BNX2_MQ_ENQUEUE1_CID (0x3fffL<<8) +#define BNX2_MQ_ENQUEUE1_BYTE_MASK (0xfL<<24) +#define BNX2_MQ_ENQUEUE1_KNL_MODE (1L<<28) + +#define BNX2_MQ_ENQUEUE2 0x00003c10 +#define BNX2_MQ_BAD_WR_ADDR 0x00003c14 +#define BNX2_MQ_BAD_RD_ADDR 0x00003c18 +#define BNX2_MQ_KNL_BYP_WIND_START 0x00003c1c +#define BNX2_MQ_KNL_BYP_WIND_START_VALUE (0xfffffL<<12) + +#define BNX2_MQ_KNL_WIND_END 0x00003c20 +#define BNX2_MQ_KNL_WIND_END_VALUE (0xffffffL<<8) + +#define BNX2_MQ_KNL_WRITE_MASK1 0x00003c24 +#define BNX2_MQ_KNL_TX_MASK1 0x00003c28 +#define BNX2_MQ_KNL_CMD_MASK1 0x00003c2c +#define BNX2_MQ_KNL_COND_ENQUEUE_MASK1 0x00003c30 +#define BNX2_MQ_KNL_RX_V2P_MASK1 0x00003c34 +#define BNX2_MQ_KNL_WRITE_MASK2 0x00003c38 +#define BNX2_MQ_KNL_TX_MASK2 0x00003c3c +#define BNX2_MQ_KNL_CMD_MASK2 0x00003c40 +#define BNX2_MQ_KNL_COND_ENQUEUE_MASK2 0x00003c44 +#define BNX2_MQ_KNL_RX_V2P_MASK2 0x00003c48 +#define BNX2_MQ_KNL_BYP_WRITE_MASK1 0x00003c4c +#define BNX2_MQ_KNL_BYP_TX_MASK1 0x00003c50 +#define BNX2_MQ_KNL_BYP_CMD_MASK1 0x00003c54 +#define BNX2_MQ_KNL_BYP_COND_ENQUEUE_MASK1 0x00003c58 +#define BNX2_MQ_KNL_BYP_RX_V2P_MASK1 0x00003c5c +#define BNX2_MQ_KNL_BYP_WRITE_MASK2 0x00003c60 +#define BNX2_MQ_KNL_BYP_TX_MASK2 0x00003c64 +#define BNX2_MQ_KNL_BYP_CMD_MASK2 0x00003c68 +#define BNX2_MQ_KNL_BYP_COND_ENQUEUE_MASK2 0x00003c6c +#define BNX2_MQ_KNL_BYP_RX_V2P_MASK2 0x00003c70 +#define BNX2_MQ_MEM_WR_ADDR 0x00003c74 +#define BNX2_MQ_MEM_WR_ADDR_VALUE (0x3fL<<0) + +#define BNX2_MQ_MEM_WR_DATA0 0x00003c78 +#define BNX2_MQ_MEM_WR_DATA0_VALUE (0xffffffffL<<0) + +#define BNX2_MQ_MEM_WR_DATA1 0x00003c7c +#define BNX2_MQ_MEM_WR_DATA1_VALUE (0xffffffffL<<0) + +#define BNX2_MQ_MEM_WR_DATA2 0x00003c80 +#define BNX2_MQ_MEM_WR_DATA2_VALUE (0x3fffffffL<<0) + +#define BNX2_MQ_MEM_RD_ADDR 0x00003c84 +#define BNX2_MQ_MEM_RD_ADDR_VALUE (0x3fL<<0) + +#define BNX2_MQ_MEM_RD_DATA0 0x00003c88 +#define BNX2_MQ_MEM_RD_DATA0_VALUE (0xffffffffL<<0) + +#define BNX2_MQ_MEM_RD_DATA1 0x00003c8c +#define BNX2_MQ_MEM_RD_DATA1_VALUE (0xffffffffL<<0) + +#define BNX2_MQ_MEM_RD_DATA2 0x00003c90 +#define BNX2_MQ_MEM_RD_DATA2_VALUE (0x3fffffffL<<0) + + + +/* + * tbdr_reg definition + * offset: 0x5000 + */ +#define BNX2_TBDR_COMMAND 0x00005000 +#define BNX2_TBDR_COMMAND_ENABLE (1L<<0) +#define BNX2_TBDR_COMMAND_SOFT_RST (1L<<1) +#define BNX2_TBDR_COMMAND_MSTR_ABORT (1L<<4) + +#define BNX2_TBDR_STATUS 0x00005004 +#define BNX2_TBDR_STATUS_DMA_WAIT (1L<<0) +#define BNX2_TBDR_STATUS_FTQ_WAIT (1L<<1) +#define BNX2_TBDR_STATUS_FIFO_OVERFLOW (1L<<2) +#define BNX2_TBDR_STATUS_FIFO_UNDERFLOW (1L<<3) +#define BNX2_TBDR_STATUS_SEARCHMISS_ERROR (1L<<4) +#define BNX2_TBDR_STATUS_FTQ_ENTRY_CNT (1L<<5) +#define BNX2_TBDR_STATUS_BURST_CNT (1L<<6) + +#define BNX2_TBDR_CONFIG 0x00005008 +#define BNX2_TBDR_CONFIG_MAX_BDS (0xffL<<0) +#define BNX2_TBDR_CONFIG_SWAP_MODE (1L<<8) +#define BNX2_TBDR_CONFIG_PRIORITY (1L<<9) +#define BNX2_TBDR_CONFIG_CACHE_NEXT_PAGE_PTRS (1L<<10) +#define BNX2_TBDR_CONFIG_PAGE_SIZE (0xfL<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_256 (0L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_512 (1L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_1K (2L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_2K (3L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_4K (4L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_8K (5L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_16K (6L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_32K (7L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_64K (8L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_128K (9L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_256K (10L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_512K (11L<<24) +#define BNX2_TBDR_CONFIG_PAGE_SIZE_1M (12L<<24) + +#define BNX2_TBDR_DEBUG_VECT_PEEK 0x0000500c +#define BNX2_TBDR_DEBUG_VECT_PEEK_1_VALUE (0x7ffL<<0) +#define BNX2_TBDR_DEBUG_VECT_PEEK_1_PEEK_EN (1L<<11) +#define BNX2_TBDR_DEBUG_VECT_PEEK_1_SEL (0xfL<<12) +#define BNX2_TBDR_DEBUG_VECT_PEEK_2_VALUE (0x7ffL<<16) +#define BNX2_TBDR_DEBUG_VECT_PEEK_2_PEEK_EN (1L<<27) +#define BNX2_TBDR_DEBUG_VECT_PEEK_2_SEL (0xfL<<28) + +#define BNX2_TBDR_FTQ_DATA 0x000053c0 +#define BNX2_TBDR_FTQ_CMD 0x000053f8 +#define BNX2_TBDR_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_TBDR_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_TBDR_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_TBDR_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_TBDR_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_TBDR_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_TBDR_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_TBDR_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_TBDR_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_TBDR_FTQ_CMD_POP (1L<<30) +#define BNX2_TBDR_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_TBDR_FTQ_CTL 0x000053fc +#define BNX2_TBDR_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_TBDR_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_TBDR_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_TBDR_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_TBDR_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + + + +/* + * tdma_reg definition + * offset: 0x5c00 + */ +#define BNX2_TDMA_COMMAND 0x00005c00 +#define BNX2_TDMA_COMMAND_ENABLED (1L<<0) +#define BNX2_TDMA_COMMAND_MASTER_ABORT (1L<<4) +#define BNX2_TDMA_COMMAND_BAD_L2_LENGTH_ABORT (1L<<7) + +#define BNX2_TDMA_STATUS 0x00005c04 +#define BNX2_TDMA_STATUS_DMA_WAIT (1L<<0) +#define BNX2_TDMA_STATUS_PAYLOAD_WAIT (1L<<1) +#define BNX2_TDMA_STATUS_PATCH_FTQ_WAIT (1L<<2) +#define BNX2_TDMA_STATUS_LOCK_WAIT (1L<<3) +#define BNX2_TDMA_STATUS_FTQ_ENTRY_CNT (1L<<16) +#define BNX2_TDMA_STATUS_BURST_CNT (1L<<17) + +#define BNX2_TDMA_CONFIG 0x00005c08 +#define BNX2_TDMA_CONFIG_ONE_DMA (1L<<0) +#define BNX2_TDMA_CONFIG_ONE_RECORD (1L<<1) +#define BNX2_TDMA_CONFIG_LIMIT_SZ (0xfL<<4) +#define BNX2_TDMA_CONFIG_LIMIT_SZ_64 (0L<<4) +#define BNX2_TDMA_CONFIG_LIMIT_SZ_128 (0x4L<<4) +#define BNX2_TDMA_CONFIG_LIMIT_SZ_256 (0x6L<<4) +#define BNX2_TDMA_CONFIG_LIMIT_SZ_512 (0x8L<<4) +#define BNX2_TDMA_CONFIG_LINE_SZ (0xfL<<8) +#define BNX2_TDMA_CONFIG_LINE_SZ_64 (0L<<8) +#define BNX2_TDMA_CONFIG_LINE_SZ_128 (4L<<8) +#define BNX2_TDMA_CONFIG_LINE_SZ_256 (6L<<8) +#define BNX2_TDMA_CONFIG_LINE_SZ_512 (8L<<8) +#define BNX2_TDMA_CONFIG_ALIGN_ENA (1L<<15) +#define BNX2_TDMA_CONFIG_CHK_L2_BD (1L<<16) +#define BNX2_TDMA_CONFIG_FIFO_CMP (0xfL<<20) + +#define BNX2_TDMA_PAYLOAD_PROD 0x00005c0c +#define BNX2_TDMA_PAYLOAD_PROD_VALUE (0x1fffL<<3) + +#define BNX2_TDMA_DBG_WATCHDOG 0x00005c10 +#define BNX2_TDMA_DBG_TRIGGER 0x00005c14 +#define BNX2_TDMA_DMAD_FSM 0x00005c80 +#define BNX2_TDMA_DMAD_FSM_BD_INVLD (1L<<0) +#define BNX2_TDMA_DMAD_FSM_PUSH (0xfL<<4) +#define BNX2_TDMA_DMAD_FSM_ARB_TBDC (0x3L<<8) +#define BNX2_TDMA_DMAD_FSM_ARB_CTX (1L<<12) +#define BNX2_TDMA_DMAD_FSM_DR_INTF (1L<<16) +#define BNX2_TDMA_DMAD_FSM_DMAD (0x7L<<20) +#define BNX2_TDMA_DMAD_FSM_BD (0xfL<<24) + +#define BNX2_TDMA_DMAD_STATUS 0x00005c84 +#define BNX2_TDMA_DMAD_STATUS_RHOLD_PUSH_ENTRY (0x3L<<0) +#define BNX2_TDMA_DMAD_STATUS_RHOLD_DMAD_ENTRY (0x3L<<4) +#define BNX2_TDMA_DMAD_STATUS_RHOLD_BD_ENTRY (0x3L<<8) +#define BNX2_TDMA_DMAD_STATUS_IFTQ_ENUM (0xfL<<12) + +#define BNX2_TDMA_DR_INTF_FSM 0x00005c88 +#define BNX2_TDMA_DR_INTF_FSM_L2_COMP (0x3L<<0) +#define BNX2_TDMA_DR_INTF_FSM_TPATQ (0x7L<<4) +#define BNX2_TDMA_DR_INTF_FSM_TPBUF (0x3L<<8) +#define BNX2_TDMA_DR_INTF_FSM_DR_BUF (0x7L<<12) +#define BNX2_TDMA_DR_INTF_FSM_DMAD (0x7L<<16) + +#define BNX2_TDMA_DR_INTF_STATUS 0x00005c8c +#define BNX2_TDMA_DR_INTF_STATUS_HOLE_PHASE (0x7L<<0) +#define BNX2_TDMA_DR_INTF_STATUS_DATA_AVAIL (0x3L<<4) +#define BNX2_TDMA_DR_INTF_STATUS_SHIFT_ADDR (0x7L<<8) +#define BNX2_TDMA_DR_INTF_STATUS_NXT_PNTR (0xfL<<12) +#define BNX2_TDMA_DR_INTF_STATUS_BYTE_COUNT (0x7L<<16) + +#define BNX2_TDMA_FTQ_DATA 0x00005fc0 +#define BNX2_TDMA_FTQ_CMD 0x00005ff8 +#define BNX2_TDMA_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_TDMA_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_TDMA_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_TDMA_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_TDMA_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_TDMA_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_TDMA_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_TDMA_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_TDMA_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_TDMA_FTQ_CMD_POP (1L<<30) +#define BNX2_TDMA_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_TDMA_FTQ_CTL 0x00005ffc +#define BNX2_TDMA_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_TDMA_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_TDMA_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_TDMA_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_TDMA_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + + + +/* + * hc_reg definition + * offset: 0x6800 + */ +#define BNX2_HC_COMMAND 0x00006800 +#define BNX2_HC_COMMAND_ENABLE (1L<<0) +#define BNX2_HC_COMMAND_SKIP_ABORT (1L<<4) +#define BNX2_HC_COMMAND_COAL_NOW (1L<<16) +#define BNX2_HC_COMMAND_COAL_NOW_WO_INT (1L<<17) +#define BNX2_HC_COMMAND_STATS_NOW (1L<<18) +#define BNX2_HC_COMMAND_FORCE_INT (0x3L<<19) +#define BNX2_HC_COMMAND_FORCE_INT_NULL (0L<<19) +#define BNX2_HC_COMMAND_FORCE_INT_HIGH (1L<<19) +#define BNX2_HC_COMMAND_FORCE_INT_LOW (2L<<19) +#define BNX2_HC_COMMAND_FORCE_INT_FREE (3L<<19) +#define BNX2_HC_COMMAND_CLR_STAT_NOW (1L<<21) + +#define BNX2_HC_STATUS 0x00006804 +#define BNX2_HC_STATUS_MASTER_ABORT (1L<<0) +#define BNX2_HC_STATUS_PARITY_ERROR_STATE (1L<<1) +#define BNX2_HC_STATUS_PCI_CLK_CNT_STAT (1L<<16) +#define BNX2_HC_STATUS_CORE_CLK_CNT_STAT (1L<<17) +#define BNX2_HC_STATUS_NUM_STATUS_BLOCKS_STAT (1L<<18) +#define BNX2_HC_STATUS_NUM_INT_GEN_STAT (1L<<19) +#define BNX2_HC_STATUS_NUM_INT_MBOX_WR_STAT (1L<<20) +#define BNX2_HC_STATUS_CORE_CLKS_TO_HW_INTACK_STAT (1L<<23) +#define BNX2_HC_STATUS_CORE_CLKS_TO_SW_INTACK_STAT (1L<<24) +#define BNX2_HC_STATUS_CORE_CLKS_DURING_SW_INTACK_STAT (1L<<25) + +#define BNX2_HC_CONFIG 0x00006808 +#define BNX2_HC_CONFIG_COLLECT_STATS (1L<<0) +#define BNX2_HC_CONFIG_RX_TMR_MODE (1L<<1) +#define BNX2_HC_CONFIG_TX_TMR_MODE (1L<<2) +#define BNX2_HC_CONFIG_COM_TMR_MODE (1L<<3) +#define BNX2_HC_CONFIG_CMD_TMR_MODE (1L<<4) +#define BNX2_HC_CONFIG_STATISTIC_PRIORITY (1L<<5) +#define BNX2_HC_CONFIG_STATUS_PRIORITY (1L<<6) +#define BNX2_HC_CONFIG_STAT_MEM_ADDR (0xffL<<8) + +#define BNX2_HC_ATTN_BITS_ENABLE 0x0000680c +#define BNX2_HC_STATUS_ADDR_L 0x00006810 +#define BNX2_HC_STATUS_ADDR_H 0x00006814 +#define BNX2_HC_STATISTICS_ADDR_L 0x00006818 +#define BNX2_HC_STATISTICS_ADDR_H 0x0000681c +#define BNX2_HC_TX_QUICK_CONS_TRIP 0x00006820 +#define BNX2_HC_TX_QUICK_CONS_TRIP_VALUE (0xffL<<0) +#define BNX2_HC_TX_QUICK_CONS_TRIP_INT (0xffL<<16) + +#define BNX2_HC_COMP_PROD_TRIP 0x00006824 +#define BNX2_HC_COMP_PROD_TRIP_VALUE (0xffL<<0) +#define BNX2_HC_COMP_PROD_TRIP_INT (0xffL<<16) + +#define BNX2_HC_RX_QUICK_CONS_TRIP 0x00006828 +#define BNX2_HC_RX_QUICK_CONS_TRIP_VALUE (0xffL<<0) +#define BNX2_HC_RX_QUICK_CONS_TRIP_INT (0xffL<<16) + +#define BNX2_HC_RX_TICKS 0x0000682c +#define BNX2_HC_RX_TICKS_VALUE (0x3ffL<<0) +#define BNX2_HC_RX_TICKS_INT (0x3ffL<<16) + +#define BNX2_HC_TX_TICKS 0x00006830 +#define BNX2_HC_TX_TICKS_VALUE (0x3ffL<<0) +#define BNX2_HC_TX_TICKS_INT (0x3ffL<<16) + +#define BNX2_HC_COM_TICKS 0x00006834 +#define BNX2_HC_COM_TICKS_VALUE (0x3ffL<<0) +#define BNX2_HC_COM_TICKS_INT (0x3ffL<<16) + +#define BNX2_HC_CMD_TICKS 0x00006838 +#define BNX2_HC_CMD_TICKS_VALUE (0x3ffL<<0) +#define BNX2_HC_CMD_TICKS_INT (0x3ffL<<16) + +#define BNX2_HC_PERIODIC_TICKS 0x0000683c +#define BNX2_HC_PERIODIC_TICKS_HC_PERIODIC_TICKS (0xffffL<<0) + +#define BNX2_HC_STAT_COLLECT_TICKS 0x00006840 +#define BNX2_HC_STAT_COLLECT_TICKS_HC_STAT_COLL_TICKS (0xffL<<4) + +#define BNX2_HC_STATS_TICKS 0x00006844 +#define BNX2_HC_STATS_TICKS_HC_STAT_TICKS (0xffffL<<8) + +#define BNX2_HC_STAT_MEM_DATA 0x0000684c +#define BNX2_HC_STAT_GEN_SEL_0 0x00006850 +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0 (0x7fL<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT0 (0L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT1 (1L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT2 (2L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT3 (3L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT4 (4L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT5 (5L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT6 (6L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT7 (7L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT8 (8L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT9 (9L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT10 (10L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXP_STAT11 (11L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TXP_STAT0 (12L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TXP_STAT1 (13L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TXP_STAT2 (14L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TXP_STAT3 (15L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TXP_STAT4 (16L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TXP_STAT5 (17L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TXP_STAT6 (18L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TXP_STAT7 (19L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT0 (20L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT1 (21L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT2 (22L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT3 (23L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT4 (24L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT5 (25L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT6 (26L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT7 (27L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT8 (28L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT9 (29L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT10 (30L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COM_STAT11 (31L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TPAT_STAT0 (32L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TPAT_STAT1 (33L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TPAT_STAT2 (34L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TPAT_STAT3 (35L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CP_STAT0 (36L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CP_STAT1 (37L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CP_STAT2 (38L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CP_STAT3 (39L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CP_STAT4 (40L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CP_STAT5 (41L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CP_STAT6 (42L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CP_STAT7 (43L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MCP_STAT0 (44L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MCP_STAT1 (45L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MCP_STAT2 (46L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MCP_STAT3 (47L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MCP_STAT4 (48L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MCP_STAT5 (49L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MCP_STAT6 (50L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MCP_STAT7 (51L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_PCI_CLK_CNT (52L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CORE_CLK_CNT (53L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_HC_NUM_STATUS_BLOCKS (54L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_HC_NUM_INT_GEN (55L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_HC_NUM_INT_MBOX_WR (56L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_HC_CORE_CLKS_TO_HW_INTACK (59L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_HC_CORE_CLKS_TO_SW_INTACK (60L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_HC_CORE_CLKS_DURING_SW_INTACK (61L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TSCH_CMD_CNT (62L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TSCH_SLOT_CNT (63L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CSCH_CMD_CNT (64L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CSCH_SLOT_CNT (65L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RLUPQ_VALID_CNT (66L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXPQ_VALID_CNT (67L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RXPCQ_VALID_CNT (68L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RV2PPQ_VALID_CNT (69L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RV2PMQ_VALID_CNT (70L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RV2PTQ_VALID_CNT (71L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RDMAQ_VALID_CNT (72L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TSCHQ_VALID_CNT (73L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TBDRQ_VALID_CNT (74L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TXPQ_VALID_CNT (75L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TDMAQ_VALID_CNT (76L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TPATQ_VALID_CNT (77L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TASQ_VALID_CNT (78L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CSQ_VALID_CNT (79L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CPQ_VALID_CNT (80L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COMXQ_VALID_CNT (81L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COMTQ_VALID_CNT (82L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_COMQ_VALID_CNT (83L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MGMQ_VALID_CNT (84L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_READ_TRANSFERS_CNT (85L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_READ_DELAY_PCI_CLKS_CNT (86L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_BIG_READ_TRANSFERS_CNT (87L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_BIG_READ_DELAY_PCI_CLKS_CNT (88L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_BIG_READ_RETRY_AFTER_DATA_CNT (89L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_WRITE_TRANSFERS_CNT (90L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_WRITE_DELAY_PCI_CLKS_CNT (91L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_BIG_WRITE_TRANSFERS_CNT (92L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_BIG_WRITE_DELAY_PCI_CLKS_CNT (93L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_DMAE_BIG_WRITE_RETRY_AFTER_DATA_CNT (94L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CTX_WR_CNT64 (95L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CTX_RD_CNT64 (96L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CTX_ACC_STALL_CLKS (97L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_CTX_LOCK_STALL_CLKS (98L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MBQ_CTX_ACCESS_STAT (99L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MBQ_CTX_ACCESS64_STAT (100L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_MBQ_PCI_STALL_STAT (101L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TBDR_FTQ_ENTRY_CNT (102L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TBDR_BURST_CNT (103L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TDMA_FTQ_ENTRY_CNT (104L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TDMA_BURST_CNT (105L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RDMA_FTQ_ENTRY_CNT (106L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RDMA_BURST_CNT (107L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RLUP_MATCH_CNT (108L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TMR_POLL_PASS_CNT (109L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TMR_TMR1_CNT (110L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TMR_TMR2_CNT (111L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TMR_TMR3_CNT (112L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TMR_TMR4_CNT (113L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_TMR_TMR5_CNT (114L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RV2P_STAT0 (115L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RV2P_STAT1 (116L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RV2P_STAT2 (117L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RV2P_STAT3 (118L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RV2P_STAT4 (119L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RV2P_STAT5 (120L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RBDC_PROC1_MISS (121L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RBDC_PROC2_MISS (122L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_0_RBDC_BURST_CNT (127L<<0) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_1 (0x7fL<<8) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_2 (0x7fL<<16) +#define BNX2_HC_STAT_GEN_SEL_0_GEN_SEL_3 (0x7fL<<24) + +#define BNX2_HC_STAT_GEN_SEL_1 0x00006854 +#define BNX2_HC_STAT_GEN_SEL_1_GEN_SEL_4 (0x7fL<<0) +#define BNX2_HC_STAT_GEN_SEL_1_GEN_SEL_5 (0x7fL<<8) +#define BNX2_HC_STAT_GEN_SEL_1_GEN_SEL_6 (0x7fL<<16) +#define BNX2_HC_STAT_GEN_SEL_1_GEN_SEL_7 (0x7fL<<24) + +#define BNX2_HC_STAT_GEN_SEL_2 0x00006858 +#define BNX2_HC_STAT_GEN_SEL_2_GEN_SEL_8 (0x7fL<<0) +#define BNX2_HC_STAT_GEN_SEL_2_GEN_SEL_9 (0x7fL<<8) +#define BNX2_HC_STAT_GEN_SEL_2_GEN_SEL_10 (0x7fL<<16) +#define BNX2_HC_STAT_GEN_SEL_2_GEN_SEL_11 (0x7fL<<24) + +#define BNX2_HC_STAT_GEN_SEL_3 0x0000685c +#define BNX2_HC_STAT_GEN_SEL_3_GEN_SEL_12 (0x7fL<<0) +#define BNX2_HC_STAT_GEN_SEL_3_GEN_SEL_13 (0x7fL<<8) +#define BNX2_HC_STAT_GEN_SEL_3_GEN_SEL_14 (0x7fL<<16) +#define BNX2_HC_STAT_GEN_SEL_3_GEN_SEL_15 (0x7fL<<24) + +#define BNX2_HC_STAT_GEN_STAT0 0x00006888 +#define BNX2_HC_STAT_GEN_STAT1 0x0000688c +#define BNX2_HC_STAT_GEN_STAT2 0x00006890 +#define BNX2_HC_STAT_GEN_STAT3 0x00006894 +#define BNX2_HC_STAT_GEN_STAT4 0x00006898 +#define BNX2_HC_STAT_GEN_STAT5 0x0000689c +#define BNX2_HC_STAT_GEN_STAT6 0x000068a0 +#define BNX2_HC_STAT_GEN_STAT7 0x000068a4 +#define BNX2_HC_STAT_GEN_STAT8 0x000068a8 +#define BNX2_HC_STAT_GEN_STAT9 0x000068ac +#define BNX2_HC_STAT_GEN_STAT10 0x000068b0 +#define BNX2_HC_STAT_GEN_STAT11 0x000068b4 +#define BNX2_HC_STAT_GEN_STAT12 0x000068b8 +#define BNX2_HC_STAT_GEN_STAT13 0x000068bc +#define BNX2_HC_STAT_GEN_STAT14 0x000068c0 +#define BNX2_HC_STAT_GEN_STAT15 0x000068c4 +#define BNX2_HC_STAT_GEN_STAT_AC0 0x000068c8 +#define BNX2_HC_STAT_GEN_STAT_AC1 0x000068cc +#define BNX2_HC_STAT_GEN_STAT_AC2 0x000068d0 +#define BNX2_HC_STAT_GEN_STAT_AC3 0x000068d4 +#define BNX2_HC_STAT_GEN_STAT_AC4 0x000068d8 +#define BNX2_HC_STAT_GEN_STAT_AC5 0x000068dc +#define BNX2_HC_STAT_GEN_STAT_AC6 0x000068e0 +#define BNX2_HC_STAT_GEN_STAT_AC7 0x000068e4 +#define BNX2_HC_STAT_GEN_STAT_AC8 0x000068e8 +#define BNX2_HC_STAT_GEN_STAT_AC9 0x000068ec +#define BNX2_HC_STAT_GEN_STAT_AC10 0x000068f0 +#define BNX2_HC_STAT_GEN_STAT_AC11 0x000068f4 +#define BNX2_HC_STAT_GEN_STAT_AC12 0x000068f8 +#define BNX2_HC_STAT_GEN_STAT_AC13 0x000068fc +#define BNX2_HC_STAT_GEN_STAT_AC14 0x00006900 +#define BNX2_HC_STAT_GEN_STAT_AC15 0x00006904 +#define BNX2_HC_VIS 0x00006908 +#define BNX2_HC_VIS_STAT_BUILD_STATE (0xfL<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_IDLE (0L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_START (1L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_REQUEST (2L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_UPDATE64 (3L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_UPDATE32 (4L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_UPDATE_DONE (5L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_DMA (6L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_MSI_CONTROL (7L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_MSI_LOW (8L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_MSI_HIGH (9L<<0) +#define BNX2_HC_VIS_STAT_BUILD_STATE_MSI_DATA (10L<<0) +#define BNX2_HC_VIS_DMA_STAT_STATE (0xfL<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_IDLE (0L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_STATUS_PARAM (1L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_STATUS_DMA (2L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_WRITE_COMP (3L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_COMP (4L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_STATISTIC_PARAM (5L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_STATISTIC_DMA (6L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_WRITE_COMP_1 (7L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_WRITE_COMP_2 (8L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_WAIT (9L<<8) +#define BNX2_HC_VIS_DMA_STAT_STATE_ABORT (15L<<8) +#define BNX2_HC_VIS_DMA_MSI_STATE (0x7L<<12) +#define BNX2_HC_VIS_STATISTIC_DMA_EN_STATE (0x3L<<15) +#define BNX2_HC_VIS_STATISTIC_DMA_EN_STATE_IDLE (0L<<15) +#define BNX2_HC_VIS_STATISTIC_DMA_EN_STATE_COUNT (1L<<15) +#define BNX2_HC_VIS_STATISTIC_DMA_EN_STATE_START (2L<<15) + +#define BNX2_HC_VIS_1 0x0000690c +#define BNX2_HC_VIS_1_HW_INTACK_STATE (1L<<4) +#define BNX2_HC_VIS_1_HW_INTACK_STATE_IDLE (0L<<4) +#define BNX2_HC_VIS_1_HW_INTACK_STATE_COUNT (1L<<4) +#define BNX2_HC_VIS_1_SW_INTACK_STATE (1L<<5) +#define BNX2_HC_VIS_1_SW_INTACK_STATE_IDLE (0L<<5) +#define BNX2_HC_VIS_1_SW_INTACK_STATE_COUNT (1L<<5) +#define BNX2_HC_VIS_1_DURING_SW_INTACK_STATE (1L<<6) +#define BNX2_HC_VIS_1_DURING_SW_INTACK_STATE_IDLE (0L<<6) +#define BNX2_HC_VIS_1_DURING_SW_INTACK_STATE_COUNT (1L<<6) +#define BNX2_HC_VIS_1_MAILBOX_COUNT_STATE (1L<<7) +#define BNX2_HC_VIS_1_MAILBOX_COUNT_STATE_IDLE (0L<<7) +#define BNX2_HC_VIS_1_MAILBOX_COUNT_STATE_COUNT (1L<<7) +#define BNX2_HC_VIS_1_RAM_RD_ARB_STATE (0xfL<<17) +#define BNX2_HC_VIS_1_RAM_RD_ARB_STATE_IDLE (0L<<17) +#define BNX2_HC_VIS_1_RAM_RD_ARB_STATE_DMA (1L<<17) +#define BNX2_HC_VIS_1_RAM_RD_ARB_STATE_UPDATE (2L<<17) +#define BNX2_HC_VIS_1_RAM_RD_ARB_STATE_ASSIGN (3L<<17) +#define BNX2_HC_VIS_1_RAM_RD_ARB_STATE_WAIT (4L<<17) +#define BNX2_HC_VIS_1_RAM_RD_ARB_STATE_REG_UPDATE (5L<<17) +#define BNX2_HC_VIS_1_RAM_RD_ARB_STATE_REG_ASSIGN (6L<<17) +#define BNX2_HC_VIS_1_RAM_RD_ARB_STATE_REG_WAIT (7L<<17) +#define BNX2_HC_VIS_1_RAM_WR_ARB_STATE (0x3L<<21) +#define BNX2_HC_VIS_1_RAM_WR_ARB_STATE_NORMAL (0L<<21) +#define BNX2_HC_VIS_1_RAM_WR_ARB_STATE_CLEAR (1L<<21) +#define BNX2_HC_VIS_1_INT_GEN_STATE (1L<<23) +#define BNX2_HC_VIS_1_INT_GEN_STATE_DLE (0L<<23) +#define BNX2_HC_VIS_1_INT_GEN_STATE_NTERRUPT (1L<<23) +#define BNX2_HC_VIS_1_STAT_CHAN_ID (0x7L<<24) +#define BNX2_HC_VIS_1_INT_B (1L<<27) + +#define BNX2_HC_DEBUG_VECT_PEEK 0x00006910 +#define BNX2_HC_DEBUG_VECT_PEEK_1_VALUE (0x7ffL<<0) +#define BNX2_HC_DEBUG_VECT_PEEK_1_PEEK_EN (1L<<11) +#define BNX2_HC_DEBUG_VECT_PEEK_1_SEL (0xfL<<12) +#define BNX2_HC_DEBUG_VECT_PEEK_2_VALUE (0x7ffL<<16) +#define BNX2_HC_DEBUG_VECT_PEEK_2_PEEK_EN (1L<<27) +#define BNX2_HC_DEBUG_VECT_PEEK_2_SEL (0xfL<<28) + + + +/* + * txp_reg definition + * offset: 0x40000 + */ +#define BNX2_TXP_CPU_MODE 0x00045000 +#define BNX2_TXP_CPU_MODE_LOCAL_RST (1L<<0) +#define BNX2_TXP_CPU_MODE_STEP_ENA (1L<<1) +#define BNX2_TXP_CPU_MODE_PAGE_0_DATA_ENA (1L<<2) +#define BNX2_TXP_CPU_MODE_PAGE_0_INST_ENA (1L<<3) +#define BNX2_TXP_CPU_MODE_MSG_BIT1 (1L<<6) +#define BNX2_TXP_CPU_MODE_INTERRUPT_ENA (1L<<7) +#define BNX2_TXP_CPU_MODE_SOFT_HALT (1L<<10) +#define BNX2_TXP_CPU_MODE_BAD_DATA_HALT_ENA (1L<<11) +#define BNX2_TXP_CPU_MODE_BAD_INST_HALT_ENA (1L<<12) +#define BNX2_TXP_CPU_MODE_FIO_ABORT_HALT_ENA (1L<<13) +#define BNX2_TXP_CPU_MODE_SPAD_UNDERFLOW_HALT_ENA (1L<<15) + +#define BNX2_TXP_CPU_STATE 0x00045004 +#define BNX2_TXP_CPU_STATE_BREAKPOINT (1L<<0) +#define BNX2_TXP_CPU_STATE_BAD_INST_HALTED (1L<<2) +#define BNX2_TXP_CPU_STATE_PAGE_0_DATA_HALTED (1L<<3) +#define BNX2_TXP_CPU_STATE_PAGE_0_INST_HALTED (1L<<4) +#define BNX2_TXP_CPU_STATE_BAD_DATA_ADDR_HALTED (1L<<5) +#define BNX2_TXP_CPU_STATE_BAD_pc_HALTED (1L<<6) +#define BNX2_TXP_CPU_STATE_ALIGN_HALTED (1L<<7) +#define BNX2_TXP_CPU_STATE_FIO_ABORT_HALTED (1L<<8) +#define BNX2_TXP_CPU_STATE_SOFT_HALTED (1L<<10) +#define BNX2_TXP_CPU_STATE_SPAD_UNDERFLOW (1L<<11) +#define BNX2_TXP_CPU_STATE_INTERRRUPT (1L<<12) +#define BNX2_TXP_CPU_STATE_DATA_ACCESS_STALL (1L<<14) +#define BNX2_TXP_CPU_STATE_INST_FETCH_STALL (1L<<15) +#define BNX2_TXP_CPU_STATE_BLOCKED_READ (1L<<31) + +#define BNX2_TXP_CPU_EVENT_MASK 0x00045008 +#define BNX2_TXP_CPU_EVENT_MASK_BREAKPOINT_MASK (1L<<0) +#define BNX2_TXP_CPU_EVENT_MASK_BAD_INST_HALTED_MASK (1L<<2) +#define BNX2_TXP_CPU_EVENT_MASK_PAGE_0_DATA_HALTED_MASK (1L<<3) +#define BNX2_TXP_CPU_EVENT_MASK_PAGE_0_INST_HALTED_MASK (1L<<4) +#define BNX2_TXP_CPU_EVENT_MASK_BAD_DATA_ADDR_HALTED_MASK (1L<<5) +#define BNX2_TXP_CPU_EVENT_MASK_BAD_PC_HALTED_MASK (1L<<6) +#define BNX2_TXP_CPU_EVENT_MASK_ALIGN_HALTED_MASK (1L<<7) +#define BNX2_TXP_CPU_EVENT_MASK_FIO_ABORT_MASK (1L<<8) +#define BNX2_TXP_CPU_EVENT_MASK_SOFT_HALTED_MASK (1L<<10) +#define BNX2_TXP_CPU_EVENT_MASK_SPAD_UNDERFLOW_MASK (1L<<11) +#define BNX2_TXP_CPU_EVENT_MASK_INTERRUPT_MASK (1L<<12) + +#define BNX2_TXP_CPU_PROGRAM_COUNTER 0x0004501c +#define BNX2_TXP_CPU_INSTRUCTION 0x00045020 +#define BNX2_TXP_CPU_DATA_ACCESS 0x00045024 +#define BNX2_TXP_CPU_INTERRUPT_ENABLE 0x00045028 +#define BNX2_TXP_CPU_INTERRUPT_VECTOR 0x0004502c +#define BNX2_TXP_CPU_INTERRUPT_SAVED_PC 0x00045030 +#define BNX2_TXP_CPU_HW_BREAKPOINT 0x00045034 +#define BNX2_TXP_CPU_HW_BREAKPOINT_DISABLE (1L<<0) +#define BNX2_TXP_CPU_HW_BREAKPOINT_ADDRESS (0x3fffffffL<<2) + +#define BNX2_TXP_CPU_DEBUG_VECT_PEEK 0x00045038 +#define BNX2_TXP_CPU_DEBUG_VECT_PEEK_1_VALUE (0x7ffL<<0) +#define BNX2_TXP_CPU_DEBUG_VECT_PEEK_1_PEEK_EN (1L<<11) +#define BNX2_TXP_CPU_DEBUG_VECT_PEEK_1_SEL (0xfL<<12) +#define BNX2_TXP_CPU_DEBUG_VECT_PEEK_2_VALUE (0x7ffL<<16) +#define BNX2_TXP_CPU_DEBUG_VECT_PEEK_2_PEEK_EN (1L<<27) +#define BNX2_TXP_CPU_DEBUG_VECT_PEEK_2_SEL (0xfL<<28) + +#define BNX2_TXP_CPU_LAST_BRANCH_ADDR 0x00045048 +#define BNX2_TXP_CPU_LAST_BRANCH_ADDR_TYPE (1L<<1) +#define BNX2_TXP_CPU_LAST_BRANCH_ADDR_TYPE_JUMP (0L<<1) +#define BNX2_TXP_CPU_LAST_BRANCH_ADDR_TYPE_BRANCH (1L<<1) +#define BNX2_TXP_CPU_LAST_BRANCH_ADDR_LBA (0x3fffffffL<<2) + +#define BNX2_TXP_CPU_REG_FILE 0x00045200 +#define BNX2_TXP_FTQ_DATA 0x000453c0 +#define BNX2_TXP_FTQ_CMD 0x000453f8 +#define BNX2_TXP_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_TXP_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_TXP_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_TXP_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_TXP_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_TXP_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_TXP_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_TXP_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_TXP_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_TXP_FTQ_CMD_POP (1L<<30) +#define BNX2_TXP_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_TXP_FTQ_CTL 0x000453fc +#define BNX2_TXP_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_TXP_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_TXP_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_TXP_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_TXP_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_TXP_SCRATCH 0x00060000 + + +/* + * tpat_reg definition + * offset: 0x80000 + */ +#define BNX2_TPAT_CPU_MODE 0x00085000 +#define BNX2_TPAT_CPU_MODE_LOCAL_RST (1L<<0) +#define BNX2_TPAT_CPU_MODE_STEP_ENA (1L<<1) +#define BNX2_TPAT_CPU_MODE_PAGE_0_DATA_ENA (1L<<2) +#define BNX2_TPAT_CPU_MODE_PAGE_0_INST_ENA (1L<<3) +#define BNX2_TPAT_CPU_MODE_MSG_BIT1 (1L<<6) +#define BNX2_TPAT_CPU_MODE_INTERRUPT_ENA (1L<<7) +#define BNX2_TPAT_CPU_MODE_SOFT_HALT (1L<<10) +#define BNX2_TPAT_CPU_MODE_BAD_DATA_HALT_ENA (1L<<11) +#define BNX2_TPAT_CPU_MODE_BAD_INST_HALT_ENA (1L<<12) +#define BNX2_TPAT_CPU_MODE_FIO_ABORT_HALT_ENA (1L<<13) +#define BNX2_TPAT_CPU_MODE_SPAD_UNDERFLOW_HALT_ENA (1L<<15) + +#define BNX2_TPAT_CPU_STATE 0x00085004 +#define BNX2_TPAT_CPU_STATE_BREAKPOINT (1L<<0) +#define BNX2_TPAT_CPU_STATE_BAD_INST_HALTED (1L<<2) +#define BNX2_TPAT_CPU_STATE_PAGE_0_DATA_HALTED (1L<<3) +#define BNX2_TPAT_CPU_STATE_PAGE_0_INST_HALTED (1L<<4) +#define BNX2_TPAT_CPU_STATE_BAD_DATA_ADDR_HALTED (1L<<5) +#define BNX2_TPAT_CPU_STATE_BAD_pc_HALTED (1L<<6) +#define BNX2_TPAT_CPU_STATE_ALIGN_HALTED (1L<<7) +#define BNX2_TPAT_CPU_STATE_FIO_ABORT_HALTED (1L<<8) +#define BNX2_TPAT_CPU_STATE_SOFT_HALTED (1L<<10) +#define BNX2_TPAT_CPU_STATE_SPAD_UNDERFLOW (1L<<11) +#define BNX2_TPAT_CPU_STATE_INTERRRUPT (1L<<12) +#define BNX2_TPAT_CPU_STATE_DATA_ACCESS_STALL (1L<<14) +#define BNX2_TPAT_CPU_STATE_INST_FETCH_STALL (1L<<15) +#define BNX2_TPAT_CPU_STATE_BLOCKED_READ (1L<<31) + +#define BNX2_TPAT_CPU_EVENT_MASK 0x00085008 +#define BNX2_TPAT_CPU_EVENT_MASK_BREAKPOINT_MASK (1L<<0) +#define BNX2_TPAT_CPU_EVENT_MASK_BAD_INST_HALTED_MASK (1L<<2) +#define BNX2_TPAT_CPU_EVENT_MASK_PAGE_0_DATA_HALTED_MASK (1L<<3) +#define BNX2_TPAT_CPU_EVENT_MASK_PAGE_0_INST_HALTED_MASK (1L<<4) +#define BNX2_TPAT_CPU_EVENT_MASK_BAD_DATA_ADDR_HALTED_MASK (1L<<5) +#define BNX2_TPAT_CPU_EVENT_MASK_BAD_PC_HALTED_MASK (1L<<6) +#define BNX2_TPAT_CPU_EVENT_MASK_ALIGN_HALTED_MASK (1L<<7) +#define BNX2_TPAT_CPU_EVENT_MASK_FIO_ABORT_MASK (1L<<8) +#define BNX2_TPAT_CPU_EVENT_MASK_SOFT_HALTED_MASK (1L<<10) +#define BNX2_TPAT_CPU_EVENT_MASK_SPAD_UNDERFLOW_MASK (1L<<11) +#define BNX2_TPAT_CPU_EVENT_MASK_INTERRUPT_MASK (1L<<12) + +#define BNX2_TPAT_CPU_PROGRAM_COUNTER 0x0008501c +#define BNX2_TPAT_CPU_INSTRUCTION 0x00085020 +#define BNX2_TPAT_CPU_DATA_ACCESS 0x00085024 +#define BNX2_TPAT_CPU_INTERRUPT_ENABLE 0x00085028 +#define BNX2_TPAT_CPU_INTERRUPT_VECTOR 0x0008502c +#define BNX2_TPAT_CPU_INTERRUPT_SAVED_PC 0x00085030 +#define BNX2_TPAT_CPU_HW_BREAKPOINT 0x00085034 +#define BNX2_TPAT_CPU_HW_BREAKPOINT_DISABLE (1L<<0) +#define BNX2_TPAT_CPU_HW_BREAKPOINT_ADDRESS (0x3fffffffL<<2) + +#define BNX2_TPAT_CPU_DEBUG_VECT_PEEK 0x00085038 +#define BNX2_TPAT_CPU_DEBUG_VECT_PEEK_1_VALUE (0x7ffL<<0) +#define BNX2_TPAT_CPU_DEBUG_VECT_PEEK_1_PEEK_EN (1L<<11) +#define BNX2_TPAT_CPU_DEBUG_VECT_PEEK_1_SEL (0xfL<<12) +#define BNX2_TPAT_CPU_DEBUG_VECT_PEEK_2_VALUE (0x7ffL<<16) +#define BNX2_TPAT_CPU_DEBUG_VECT_PEEK_2_PEEK_EN (1L<<27) +#define BNX2_TPAT_CPU_DEBUG_VECT_PEEK_2_SEL (0xfL<<28) + +#define BNX2_TPAT_CPU_LAST_BRANCH_ADDR 0x00085048 +#define BNX2_TPAT_CPU_LAST_BRANCH_ADDR_TYPE (1L<<1) +#define BNX2_TPAT_CPU_LAST_BRANCH_ADDR_TYPE_JUMP (0L<<1) +#define BNX2_TPAT_CPU_LAST_BRANCH_ADDR_TYPE_BRANCH (1L<<1) +#define BNX2_TPAT_CPU_LAST_BRANCH_ADDR_LBA (0x3fffffffL<<2) + +#define BNX2_TPAT_CPU_REG_FILE 0x00085200 +#define BNX2_TPAT_FTQ_DATA 0x000853c0 +#define BNX2_TPAT_FTQ_CMD 0x000853f8 +#define BNX2_TPAT_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_TPAT_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_TPAT_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_TPAT_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_TPAT_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_TPAT_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_TPAT_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_TPAT_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_TPAT_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_TPAT_FTQ_CMD_POP (1L<<30) +#define BNX2_TPAT_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_TPAT_FTQ_CTL 0x000853fc +#define BNX2_TPAT_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_TPAT_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_TPAT_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_TPAT_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_TPAT_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_TPAT_SCRATCH 0x000a0000 + + +/* + * rxp_reg definition + * offset: 0xc0000 + */ +#define BNX2_RXP_CPU_MODE 0x000c5000 +#define BNX2_RXP_CPU_MODE_LOCAL_RST (1L<<0) +#define BNX2_RXP_CPU_MODE_STEP_ENA (1L<<1) +#define BNX2_RXP_CPU_MODE_PAGE_0_DATA_ENA (1L<<2) +#define BNX2_RXP_CPU_MODE_PAGE_0_INST_ENA (1L<<3) +#define BNX2_RXP_CPU_MODE_MSG_BIT1 (1L<<6) +#define BNX2_RXP_CPU_MODE_INTERRUPT_ENA (1L<<7) +#define BNX2_RXP_CPU_MODE_SOFT_HALT (1L<<10) +#define BNX2_RXP_CPU_MODE_BAD_DATA_HALT_ENA (1L<<11) +#define BNX2_RXP_CPU_MODE_BAD_INST_HALT_ENA (1L<<12) +#define BNX2_RXP_CPU_MODE_FIO_ABORT_HALT_ENA (1L<<13) +#define BNX2_RXP_CPU_MODE_SPAD_UNDERFLOW_HALT_ENA (1L<<15) + +#define BNX2_RXP_CPU_STATE 0x000c5004 +#define BNX2_RXP_CPU_STATE_BREAKPOINT (1L<<0) +#define BNX2_RXP_CPU_STATE_BAD_INST_HALTED (1L<<2) +#define BNX2_RXP_CPU_STATE_PAGE_0_DATA_HALTED (1L<<3) +#define BNX2_RXP_CPU_STATE_PAGE_0_INST_HALTED (1L<<4) +#define BNX2_RXP_CPU_STATE_BAD_DATA_ADDR_HALTED (1L<<5) +#define BNX2_RXP_CPU_STATE_BAD_pc_HALTED (1L<<6) +#define BNX2_RXP_CPU_STATE_ALIGN_HALTED (1L<<7) +#define BNX2_RXP_CPU_STATE_FIO_ABORT_HALTED (1L<<8) +#define BNX2_RXP_CPU_STATE_SOFT_HALTED (1L<<10) +#define BNX2_RXP_CPU_STATE_SPAD_UNDERFLOW (1L<<11) +#define BNX2_RXP_CPU_STATE_INTERRRUPT (1L<<12) +#define BNX2_RXP_CPU_STATE_DATA_ACCESS_STALL (1L<<14) +#define BNX2_RXP_CPU_STATE_INST_FETCH_STALL (1L<<15) +#define BNX2_RXP_CPU_STATE_BLOCKED_READ (1L<<31) + +#define BNX2_RXP_CPU_EVENT_MASK 0x000c5008 +#define BNX2_RXP_CPU_EVENT_MASK_BREAKPOINT_MASK (1L<<0) +#define BNX2_RXP_CPU_EVENT_MASK_BAD_INST_HALTED_MASK (1L<<2) +#define BNX2_RXP_CPU_EVENT_MASK_PAGE_0_DATA_HALTED_MASK (1L<<3) +#define BNX2_RXP_CPU_EVENT_MASK_PAGE_0_INST_HALTED_MASK (1L<<4) +#define BNX2_RXP_CPU_EVENT_MASK_BAD_DATA_ADDR_HALTED_MASK (1L<<5) +#define BNX2_RXP_CPU_EVENT_MASK_BAD_PC_HALTED_MASK (1L<<6) +#define BNX2_RXP_CPU_EVENT_MASK_ALIGN_HALTED_MASK (1L<<7) +#define BNX2_RXP_CPU_EVENT_MASK_FIO_ABORT_MASK (1L<<8) +#define BNX2_RXP_CPU_EVENT_MASK_SOFT_HALTED_MASK (1L<<10) +#define BNX2_RXP_CPU_EVENT_MASK_SPAD_UNDERFLOW_MASK (1L<<11) +#define BNX2_RXP_CPU_EVENT_MASK_INTERRUPT_MASK (1L<<12) + +#define BNX2_RXP_CPU_PROGRAM_COUNTER 0x000c501c +#define BNX2_RXP_CPU_INSTRUCTION 0x000c5020 +#define BNX2_RXP_CPU_DATA_ACCESS 0x000c5024 +#define BNX2_RXP_CPU_INTERRUPT_ENABLE 0x000c5028 +#define BNX2_RXP_CPU_INTERRUPT_VECTOR 0x000c502c +#define BNX2_RXP_CPU_INTERRUPT_SAVED_PC 0x000c5030 +#define BNX2_RXP_CPU_HW_BREAKPOINT 0x000c5034 +#define BNX2_RXP_CPU_HW_BREAKPOINT_DISABLE (1L<<0) +#define BNX2_RXP_CPU_HW_BREAKPOINT_ADDRESS (0x3fffffffL<<2) + +#define BNX2_RXP_CPU_DEBUG_VECT_PEEK 0x000c5038 +#define BNX2_RXP_CPU_DEBUG_VECT_PEEK_1_VALUE (0x7ffL<<0) +#define BNX2_RXP_CPU_DEBUG_VECT_PEEK_1_PEEK_EN (1L<<11) +#define BNX2_RXP_CPU_DEBUG_VECT_PEEK_1_SEL (0xfL<<12) +#define BNX2_RXP_CPU_DEBUG_VECT_PEEK_2_VALUE (0x7ffL<<16) +#define BNX2_RXP_CPU_DEBUG_VECT_PEEK_2_PEEK_EN (1L<<27) +#define BNX2_RXP_CPU_DEBUG_VECT_PEEK_2_SEL (0xfL<<28) + +#define BNX2_RXP_CPU_LAST_BRANCH_ADDR 0x000c5048 +#define BNX2_RXP_CPU_LAST_BRANCH_ADDR_TYPE (1L<<1) +#define BNX2_RXP_CPU_LAST_BRANCH_ADDR_TYPE_JUMP (0L<<1) +#define BNX2_RXP_CPU_LAST_BRANCH_ADDR_TYPE_BRANCH (1L<<1) +#define BNX2_RXP_CPU_LAST_BRANCH_ADDR_LBA (0x3fffffffL<<2) + +#define BNX2_RXP_CPU_REG_FILE 0x000c5200 +#define BNX2_RXP_CFTQ_DATA 0x000c5380 +#define BNX2_RXP_CFTQ_CMD 0x000c53b8 +#define BNX2_RXP_CFTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_RXP_CFTQ_CMD_WR_TOP (1L<<10) +#define BNX2_RXP_CFTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_RXP_CFTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_RXP_CFTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_RXP_CFTQ_CMD_RD_DATA (1L<<26) +#define BNX2_RXP_CFTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_RXP_CFTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_RXP_CFTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_RXP_CFTQ_CMD_POP (1L<<30) +#define BNX2_RXP_CFTQ_CMD_BUSY (1L<<31) + +#define BNX2_RXP_CFTQ_CTL 0x000c53bc +#define BNX2_RXP_CFTQ_CTL_INTERVENE (1L<<0) +#define BNX2_RXP_CFTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_RXP_CFTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_RXP_CFTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_RXP_CFTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_RXP_FTQ_DATA 0x000c53c0 +#define BNX2_RXP_FTQ_CMD 0x000c53f8 +#define BNX2_RXP_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_RXP_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_RXP_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_RXP_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_RXP_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_RXP_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_RXP_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_RXP_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_RXP_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_RXP_FTQ_CMD_POP (1L<<30) +#define BNX2_RXP_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_RXP_FTQ_CTL 0x000c53fc +#define BNX2_RXP_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_RXP_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_RXP_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_RXP_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_RXP_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_RXP_SCRATCH 0x000e0000 + + +/* + * com_reg definition + * offset: 0x100000 + */ +#define BNX2_COM_CPU_MODE 0x00105000 +#define BNX2_COM_CPU_MODE_LOCAL_RST (1L<<0) +#define BNX2_COM_CPU_MODE_STEP_ENA (1L<<1) +#define BNX2_COM_CPU_MODE_PAGE_0_DATA_ENA (1L<<2) +#define BNX2_COM_CPU_MODE_PAGE_0_INST_ENA (1L<<3) +#define BNX2_COM_CPU_MODE_MSG_BIT1 (1L<<6) +#define BNX2_COM_CPU_MODE_INTERRUPT_ENA (1L<<7) +#define BNX2_COM_CPU_MODE_SOFT_HALT (1L<<10) +#define BNX2_COM_CPU_MODE_BAD_DATA_HALT_ENA (1L<<11) +#define BNX2_COM_CPU_MODE_BAD_INST_HALT_ENA (1L<<12) +#define BNX2_COM_CPU_MODE_FIO_ABORT_HALT_ENA (1L<<13) +#define BNX2_COM_CPU_MODE_SPAD_UNDERFLOW_HALT_ENA (1L<<15) + +#define BNX2_COM_CPU_STATE 0x00105004 +#define BNX2_COM_CPU_STATE_BREAKPOINT (1L<<0) +#define BNX2_COM_CPU_STATE_BAD_INST_HALTED (1L<<2) +#define BNX2_COM_CPU_STATE_PAGE_0_DATA_HALTED (1L<<3) +#define BNX2_COM_CPU_STATE_PAGE_0_INST_HALTED (1L<<4) +#define BNX2_COM_CPU_STATE_BAD_DATA_ADDR_HALTED (1L<<5) +#define BNX2_COM_CPU_STATE_BAD_pc_HALTED (1L<<6) +#define BNX2_COM_CPU_STATE_ALIGN_HALTED (1L<<7) +#define BNX2_COM_CPU_STATE_FIO_ABORT_HALTED (1L<<8) +#define BNX2_COM_CPU_STATE_SOFT_HALTED (1L<<10) +#define BNX2_COM_CPU_STATE_SPAD_UNDERFLOW (1L<<11) +#define BNX2_COM_CPU_STATE_INTERRRUPT (1L<<12) +#define BNX2_COM_CPU_STATE_DATA_ACCESS_STALL (1L<<14) +#define BNX2_COM_CPU_STATE_INST_FETCH_STALL (1L<<15) +#define BNX2_COM_CPU_STATE_BLOCKED_READ (1L<<31) + +#define BNX2_COM_CPU_EVENT_MASK 0x00105008 +#define BNX2_COM_CPU_EVENT_MASK_BREAKPOINT_MASK (1L<<0) +#define BNX2_COM_CPU_EVENT_MASK_BAD_INST_HALTED_MASK (1L<<2) +#define BNX2_COM_CPU_EVENT_MASK_PAGE_0_DATA_HALTED_MASK (1L<<3) +#define BNX2_COM_CPU_EVENT_MASK_PAGE_0_INST_HALTED_MASK (1L<<4) +#define BNX2_COM_CPU_EVENT_MASK_BAD_DATA_ADDR_HALTED_MASK (1L<<5) +#define BNX2_COM_CPU_EVENT_MASK_BAD_PC_HALTED_MASK (1L<<6) +#define BNX2_COM_CPU_EVENT_MASK_ALIGN_HALTED_MASK (1L<<7) +#define BNX2_COM_CPU_EVENT_MASK_FIO_ABORT_MASK (1L<<8) +#define BNX2_COM_CPU_EVENT_MASK_SOFT_HALTED_MASK (1L<<10) +#define BNX2_COM_CPU_EVENT_MASK_SPAD_UNDERFLOW_MASK (1L<<11) +#define BNX2_COM_CPU_EVENT_MASK_INTERRUPT_MASK (1L<<12) + +#define BNX2_COM_CPU_PROGRAM_COUNTER 0x0010501c +#define BNX2_COM_CPU_INSTRUCTION 0x00105020 +#define BNX2_COM_CPU_DATA_ACCESS 0x00105024 +#define BNX2_COM_CPU_INTERRUPT_ENABLE 0x00105028 +#define BNX2_COM_CPU_INTERRUPT_VECTOR 0x0010502c +#define BNX2_COM_CPU_INTERRUPT_SAVED_PC 0x00105030 +#define BNX2_COM_CPU_HW_BREAKPOINT 0x00105034 +#define BNX2_COM_CPU_HW_BREAKPOINT_DISABLE (1L<<0) +#define BNX2_COM_CPU_HW_BREAKPOINT_ADDRESS (0x3fffffffL<<2) + +#define BNX2_COM_CPU_DEBUG_VECT_PEEK 0x00105038 +#define BNX2_COM_CPU_DEBUG_VECT_PEEK_1_VALUE (0x7ffL<<0) +#define BNX2_COM_CPU_DEBUG_VECT_PEEK_1_PEEK_EN (1L<<11) +#define BNX2_COM_CPU_DEBUG_VECT_PEEK_1_SEL (0xfL<<12) +#define BNX2_COM_CPU_DEBUG_VECT_PEEK_2_VALUE (0x7ffL<<16) +#define BNX2_COM_CPU_DEBUG_VECT_PEEK_2_PEEK_EN (1L<<27) +#define BNX2_COM_CPU_DEBUG_VECT_PEEK_2_SEL (0xfL<<28) + +#define BNX2_COM_CPU_LAST_BRANCH_ADDR 0x00105048 +#define BNX2_COM_CPU_LAST_BRANCH_ADDR_TYPE (1L<<1) +#define BNX2_COM_CPU_LAST_BRANCH_ADDR_TYPE_JUMP (0L<<1) +#define BNX2_COM_CPU_LAST_BRANCH_ADDR_TYPE_BRANCH (1L<<1) +#define BNX2_COM_CPU_LAST_BRANCH_ADDR_LBA (0x3fffffffL<<2) + +#define BNX2_COM_CPU_REG_FILE 0x00105200 +#define BNX2_COM_COMXQ_FTQ_DATA 0x00105340 +#define BNX2_COM_COMXQ_FTQ_CMD 0x00105378 +#define BNX2_COM_COMXQ_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_COM_COMXQ_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_COM_COMXQ_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_COM_COMXQ_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_COM_COMXQ_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_COM_COMXQ_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_COM_COMXQ_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_COM_COMXQ_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_COM_COMXQ_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_COM_COMXQ_FTQ_CMD_POP (1L<<30) +#define BNX2_COM_COMXQ_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_COM_COMXQ_FTQ_CTL 0x0010537c +#define BNX2_COM_COMXQ_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_COM_COMXQ_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_COM_COMXQ_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_COM_COMXQ_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_COM_COMXQ_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_COM_COMTQ_FTQ_DATA 0x00105380 +#define BNX2_COM_COMTQ_FTQ_CMD 0x001053b8 +#define BNX2_COM_COMTQ_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_COM_COMTQ_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_COM_COMTQ_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_COM_COMTQ_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_COM_COMTQ_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_COM_COMTQ_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_COM_COMTQ_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_COM_COMTQ_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_COM_COMTQ_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_COM_COMTQ_FTQ_CMD_POP (1L<<30) +#define BNX2_COM_COMTQ_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_COM_COMTQ_FTQ_CTL 0x001053bc +#define BNX2_COM_COMTQ_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_COM_COMTQ_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_COM_COMTQ_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_COM_COMTQ_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_COM_COMTQ_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_COM_COMQ_FTQ_DATA 0x001053c0 +#define BNX2_COM_COMQ_FTQ_CMD 0x001053f8 +#define BNX2_COM_COMQ_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_COM_COMQ_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_COM_COMQ_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_COM_COMQ_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_COM_COMQ_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_COM_COMQ_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_COM_COMQ_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_COM_COMQ_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_COM_COMQ_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_COM_COMQ_FTQ_CMD_POP (1L<<30) +#define BNX2_COM_COMQ_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_COM_COMQ_FTQ_CTL 0x001053fc +#define BNX2_COM_COMQ_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_COM_COMQ_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_COM_COMQ_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_COM_COMQ_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_COM_COMQ_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_COM_SCRATCH 0x00120000 + + +/* + * cp_reg definition + * offset: 0x180000 + */ +#define BNX2_CP_CPU_MODE 0x00185000 +#define BNX2_CP_CPU_MODE_LOCAL_RST (1L<<0) +#define BNX2_CP_CPU_MODE_STEP_ENA (1L<<1) +#define BNX2_CP_CPU_MODE_PAGE_0_DATA_ENA (1L<<2) +#define BNX2_CP_CPU_MODE_PAGE_0_INST_ENA (1L<<3) +#define BNX2_CP_CPU_MODE_MSG_BIT1 (1L<<6) +#define BNX2_CP_CPU_MODE_INTERRUPT_ENA (1L<<7) +#define BNX2_CP_CPU_MODE_SOFT_HALT (1L<<10) +#define BNX2_CP_CPU_MODE_BAD_DATA_HALT_ENA (1L<<11) +#define BNX2_CP_CPU_MODE_BAD_INST_HALT_ENA (1L<<12) +#define BNX2_CP_CPU_MODE_FIO_ABORT_HALT_ENA (1L<<13) +#define BNX2_CP_CPU_MODE_SPAD_UNDERFLOW_HALT_ENA (1L<<15) + +#define BNX2_CP_CPU_STATE 0x00185004 +#define BNX2_CP_CPU_STATE_BREAKPOINT (1L<<0) +#define BNX2_CP_CPU_STATE_BAD_INST_HALTED (1L<<2) +#define BNX2_CP_CPU_STATE_PAGE_0_DATA_HALTED (1L<<3) +#define BNX2_CP_CPU_STATE_PAGE_0_INST_HALTED (1L<<4) +#define BNX2_CP_CPU_STATE_BAD_DATA_ADDR_HALTED (1L<<5) +#define BNX2_CP_CPU_STATE_BAD_pc_HALTED (1L<<6) +#define BNX2_CP_CPU_STATE_ALIGN_HALTED (1L<<7) +#define BNX2_CP_CPU_STATE_FIO_ABORT_HALTED (1L<<8) +#define BNX2_CP_CPU_STATE_SOFT_HALTED (1L<<10) +#define BNX2_CP_CPU_STATE_SPAD_UNDERFLOW (1L<<11) +#define BNX2_CP_CPU_STATE_INTERRRUPT (1L<<12) +#define BNX2_CP_CPU_STATE_DATA_ACCESS_STALL (1L<<14) +#define BNX2_CP_CPU_STATE_INST_FETCH_STALL (1L<<15) +#define BNX2_CP_CPU_STATE_BLOCKED_READ (1L<<31) + +#define BNX2_CP_CPU_EVENT_MASK 0x00185008 +#define BNX2_CP_CPU_EVENT_MASK_BREAKPOINT_MASK (1L<<0) +#define BNX2_CP_CPU_EVENT_MASK_BAD_INST_HALTED_MASK (1L<<2) +#define BNX2_CP_CPU_EVENT_MASK_PAGE_0_DATA_HALTED_MASK (1L<<3) +#define BNX2_CP_CPU_EVENT_MASK_PAGE_0_INST_HALTED_MASK (1L<<4) +#define BNX2_CP_CPU_EVENT_MASK_BAD_DATA_ADDR_HALTED_MASK (1L<<5) +#define BNX2_CP_CPU_EVENT_MASK_BAD_PC_HALTED_MASK (1L<<6) +#define BNX2_CP_CPU_EVENT_MASK_ALIGN_HALTED_MASK (1L<<7) +#define BNX2_CP_CPU_EVENT_MASK_FIO_ABORT_MASK (1L<<8) +#define BNX2_CP_CPU_EVENT_MASK_SOFT_HALTED_MASK (1L<<10) +#define BNX2_CP_CPU_EVENT_MASK_SPAD_UNDERFLOW_MASK (1L<<11) +#define BNX2_CP_CPU_EVENT_MASK_INTERRUPT_MASK (1L<<12) + +#define BNX2_CP_CPU_PROGRAM_COUNTER 0x0018501c +#define BNX2_CP_CPU_INSTRUCTION 0x00185020 +#define BNX2_CP_CPU_DATA_ACCESS 0x00185024 +#define BNX2_CP_CPU_INTERRUPT_ENABLE 0x00185028 +#define BNX2_CP_CPU_INTERRUPT_VECTOR 0x0018502c +#define BNX2_CP_CPU_INTERRUPT_SAVED_PC 0x00185030 +#define BNX2_CP_CPU_HW_BREAKPOINT 0x00185034 +#define BNX2_CP_CPU_HW_BREAKPOINT_DISABLE (1L<<0) +#define BNX2_CP_CPU_HW_BREAKPOINT_ADDRESS (0x3fffffffL<<2) + +#define BNX2_CP_CPU_DEBUG_VECT_PEEK 0x00185038 +#define BNX2_CP_CPU_DEBUG_VECT_PEEK_1_VALUE (0x7ffL<<0) +#define BNX2_CP_CPU_DEBUG_VECT_PEEK_1_PEEK_EN (1L<<11) +#define BNX2_CP_CPU_DEBUG_VECT_PEEK_1_SEL (0xfL<<12) +#define BNX2_CP_CPU_DEBUG_VECT_PEEK_2_VALUE (0x7ffL<<16) +#define BNX2_CP_CPU_DEBUG_VECT_PEEK_2_PEEK_EN (1L<<27) +#define BNX2_CP_CPU_DEBUG_VECT_PEEK_2_SEL (0xfL<<28) + +#define BNX2_CP_CPU_LAST_BRANCH_ADDR 0x00185048 +#define BNX2_CP_CPU_LAST_BRANCH_ADDR_TYPE (1L<<1) +#define BNX2_CP_CPU_LAST_BRANCH_ADDR_TYPE_JUMP (0L<<1) +#define BNX2_CP_CPU_LAST_BRANCH_ADDR_TYPE_BRANCH (1L<<1) +#define BNX2_CP_CPU_LAST_BRANCH_ADDR_LBA (0x3fffffffL<<2) + +#define BNX2_CP_CPU_REG_FILE 0x00185200 +#define BNX2_CP_CPQ_FTQ_DATA 0x001853c0 +#define BNX2_CP_CPQ_FTQ_CMD 0x001853f8 +#define BNX2_CP_CPQ_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_CP_CPQ_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_CP_CPQ_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_CP_CPQ_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_CP_CPQ_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_CP_CPQ_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_CP_CPQ_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_CP_CPQ_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_CP_CPQ_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_CP_CPQ_FTQ_CMD_POP (1L<<30) +#define BNX2_CP_CPQ_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_CP_CPQ_FTQ_CTL 0x001853fc +#define BNX2_CP_CPQ_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_CP_CPQ_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_CP_CPQ_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_CP_CPQ_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_CP_CPQ_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_CP_SCRATCH 0x001a0000 + + +/* + * mcp_reg definition + * offset: 0x140000 + */ +#define BNX2_MCP_CPU_MODE 0x00145000 +#define BNX2_MCP_CPU_MODE_LOCAL_RST (1L<<0) +#define BNX2_MCP_CPU_MODE_STEP_ENA (1L<<1) +#define BNX2_MCP_CPU_MODE_PAGE_0_DATA_ENA (1L<<2) +#define BNX2_MCP_CPU_MODE_PAGE_0_INST_ENA (1L<<3) +#define BNX2_MCP_CPU_MODE_MSG_BIT1 (1L<<6) +#define BNX2_MCP_CPU_MODE_INTERRUPT_ENA (1L<<7) +#define BNX2_MCP_CPU_MODE_SOFT_HALT (1L<<10) +#define BNX2_MCP_CPU_MODE_BAD_DATA_HALT_ENA (1L<<11) +#define BNX2_MCP_CPU_MODE_BAD_INST_HALT_ENA (1L<<12) +#define BNX2_MCP_CPU_MODE_FIO_ABORT_HALT_ENA (1L<<13) +#define BNX2_MCP_CPU_MODE_SPAD_UNDERFLOW_HALT_ENA (1L<<15) + +#define BNX2_MCP_CPU_STATE 0x00145004 +#define BNX2_MCP_CPU_STATE_BREAKPOINT (1L<<0) +#define BNX2_MCP_CPU_STATE_BAD_INST_HALTED (1L<<2) +#define BNX2_MCP_CPU_STATE_PAGE_0_DATA_HALTED (1L<<3) +#define BNX2_MCP_CPU_STATE_PAGE_0_INST_HALTED (1L<<4) +#define BNX2_MCP_CPU_STATE_BAD_DATA_ADDR_HALTED (1L<<5) +#define BNX2_MCP_CPU_STATE_BAD_pc_HALTED (1L<<6) +#define BNX2_MCP_CPU_STATE_ALIGN_HALTED (1L<<7) +#define BNX2_MCP_CPU_STATE_FIO_ABORT_HALTED (1L<<8) +#define BNX2_MCP_CPU_STATE_SOFT_HALTED (1L<<10) +#define BNX2_MCP_CPU_STATE_SPAD_UNDERFLOW (1L<<11) +#define BNX2_MCP_CPU_STATE_INTERRRUPT (1L<<12) +#define BNX2_MCP_CPU_STATE_DATA_ACCESS_STALL (1L<<14) +#define BNX2_MCP_CPU_STATE_INST_FETCH_STALL (1L<<15) +#define BNX2_MCP_CPU_STATE_BLOCKED_READ (1L<<31) + +#define BNX2_MCP_CPU_EVENT_MASK 0x00145008 +#define BNX2_MCP_CPU_EVENT_MASK_BREAKPOINT_MASK (1L<<0) +#define BNX2_MCP_CPU_EVENT_MASK_BAD_INST_HALTED_MASK (1L<<2) +#define BNX2_MCP_CPU_EVENT_MASK_PAGE_0_DATA_HALTED_MASK (1L<<3) +#define BNX2_MCP_CPU_EVENT_MASK_PAGE_0_INST_HALTED_MASK (1L<<4) +#define BNX2_MCP_CPU_EVENT_MASK_BAD_DATA_ADDR_HALTED_MASK (1L<<5) +#define BNX2_MCP_CPU_EVENT_MASK_BAD_PC_HALTED_MASK (1L<<6) +#define BNX2_MCP_CPU_EVENT_MASK_ALIGN_HALTED_MASK (1L<<7) +#define BNX2_MCP_CPU_EVENT_MASK_FIO_ABORT_MASK (1L<<8) +#define BNX2_MCP_CPU_EVENT_MASK_SOFT_HALTED_MASK (1L<<10) +#define BNX2_MCP_CPU_EVENT_MASK_SPAD_UNDERFLOW_MASK (1L<<11) +#define BNX2_MCP_CPU_EVENT_MASK_INTERRUPT_MASK (1L<<12) + +#define BNX2_MCP_CPU_PROGRAM_COUNTER 0x0014501c +#define BNX2_MCP_CPU_INSTRUCTION 0x00145020 +#define BNX2_MCP_CPU_DATA_ACCESS 0x00145024 +#define BNX2_MCP_CPU_INTERRUPT_ENABLE 0x00145028 +#define BNX2_MCP_CPU_INTERRUPT_VECTOR 0x0014502c +#define BNX2_MCP_CPU_INTERRUPT_SAVED_PC 0x00145030 +#define BNX2_MCP_CPU_HW_BREAKPOINT 0x00145034 +#define BNX2_MCP_CPU_HW_BREAKPOINT_DISABLE (1L<<0) +#define BNX2_MCP_CPU_HW_BREAKPOINT_ADDRESS (0x3fffffffL<<2) + +#define BNX2_MCP_CPU_DEBUG_VECT_PEEK 0x00145038 +#define BNX2_MCP_CPU_DEBUG_VECT_PEEK_1_VALUE (0x7ffL<<0) +#define BNX2_MCP_CPU_DEBUG_VECT_PEEK_1_PEEK_EN (1L<<11) +#define BNX2_MCP_CPU_DEBUG_VECT_PEEK_1_SEL (0xfL<<12) +#define BNX2_MCP_CPU_DEBUG_VECT_PEEK_2_VALUE (0x7ffL<<16) +#define BNX2_MCP_CPU_DEBUG_VECT_PEEK_2_PEEK_EN (1L<<27) +#define BNX2_MCP_CPU_DEBUG_VECT_PEEK_2_SEL (0xfL<<28) + +#define BNX2_MCP_CPU_LAST_BRANCH_ADDR 0x00145048 +#define BNX2_MCP_CPU_LAST_BRANCH_ADDR_TYPE (1L<<1) +#define BNX2_MCP_CPU_LAST_BRANCH_ADDR_TYPE_JUMP (0L<<1) +#define BNX2_MCP_CPU_LAST_BRANCH_ADDR_TYPE_BRANCH (1L<<1) +#define BNX2_MCP_CPU_LAST_BRANCH_ADDR_LBA (0x3fffffffL<<2) + +#define BNX2_MCP_CPU_REG_FILE 0x00145200 +#define BNX2_MCP_MCPQ_FTQ_DATA 0x001453c0 +#define BNX2_MCP_MCPQ_FTQ_CMD 0x001453f8 +#define BNX2_MCP_MCPQ_FTQ_CMD_OFFSET (0x3ffL<<0) +#define BNX2_MCP_MCPQ_FTQ_CMD_WR_TOP (1L<<10) +#define BNX2_MCP_MCPQ_FTQ_CMD_WR_TOP_0 (0L<<10) +#define BNX2_MCP_MCPQ_FTQ_CMD_WR_TOP_1 (1L<<10) +#define BNX2_MCP_MCPQ_FTQ_CMD_SFT_RESET (1L<<25) +#define BNX2_MCP_MCPQ_FTQ_CMD_RD_DATA (1L<<26) +#define BNX2_MCP_MCPQ_FTQ_CMD_ADD_INTERVEN (1L<<27) +#define BNX2_MCP_MCPQ_FTQ_CMD_ADD_DATA (1L<<28) +#define BNX2_MCP_MCPQ_FTQ_CMD_INTERVENE_CLR (1L<<29) +#define BNX2_MCP_MCPQ_FTQ_CMD_POP (1L<<30) +#define BNX2_MCP_MCPQ_FTQ_CMD_BUSY (1L<<31) + +#define BNX2_MCP_MCPQ_FTQ_CTL 0x001453fc +#define BNX2_MCP_MCPQ_FTQ_CTL_INTERVENE (1L<<0) +#define BNX2_MCP_MCPQ_FTQ_CTL_OVERFLOW (1L<<1) +#define BNX2_MCP_MCPQ_FTQ_CTL_FORCE_INTERVENE (1L<<2) +#define BNX2_MCP_MCPQ_FTQ_CTL_MAX_DEPTH (0x3ffL<<12) +#define BNX2_MCP_MCPQ_FTQ_CTL_CUR_DEPTH (0x3ffL<<22) + +#define BNX2_MCP_ROM 0x00150000 +#define BNX2_MCP_SCRATCH 0x00160000 + + +#define NUM_MC_HASH_REGISTERS 8 + + +/* PHY_ID1: bits 31-16; PHY_ID2: bits 15-0. */ +#define PHY_BCM5706_PHY_ID 0x00206160 + +#define PHY_ID(id) ((id) & 0xfffffff0) +#define PHY_REV_ID(id) ((id) & 0xf) + +#define MIN_ETHERNET_PACKET_SIZE 60 +#define MAX_ETHERNET_PACKET_SIZE 1514 +#define MAX_ETHERNET_JUMBO_PACKET_SIZE 9014 + +#define RX_COPY_THRESH 92 + +#define DMA_READ_CHANS 5 +#define DMA_WRITE_CHANS 3 + +#define BCM_PAGE_BITS 12 +#define BCM_PAGE_SIZE (1 << BCM_PAGE_BITS) + +#define TX_DESC_CNT (BCM_PAGE_SIZE / sizeof(struct tx_bd)) +#define MAX_TX_DESC_CNT (TX_DESC_CNT - 1) + +#define RX_DESC_CNT (BCM_PAGE_SIZE / sizeof(struct rx_bd)) +#define MAX_RX_DESC_CNT (RX_DESC_CNT - 1) + +#define NEXT_TX_BD(x) (((x) & (MAX_TX_DESC_CNT - 1)) == \ + (MAX_TX_DESC_CNT - 1)) ? \ + (x) + 2 : (x) + 1 + +#define TX_RING_IDX(x) ((x) & MAX_TX_DESC_CNT) + +#define NEXT_RX_BD(x) (((x) & (MAX_RX_DESC_CNT - 1)) == \ + (MAX_RX_DESC_CNT - 1)) ? \ + (x) + 2 : (x) + 1 + +#define RX_RING_IDX(x) ((x) & MAX_RX_DESC_CNT) + + +/* Context size. */ +#define CTX_SHIFT 7 +#define CTX_SIZE (1 << CTX_SHIFT) +#define CTX_MASK (CTX_SIZE - 1) +#define GET_CID_ADDR(_cid) ((_cid) << CTX_SHIFT) +#define GET_CID(_cid_addr) ((_cid_addr) >> CTX_SHIFT) + +#define PHY_CTX_SHIFT 6 +#define PHY_CTX_SIZE (1 << PHY_CTX_SHIFT) +#define PHY_CTX_MASK (PHY_CTX_SIZE - 1) +#define GET_PCID_ADDR(_pcid) ((_pcid) << PHY_CTX_SHIFT) +#define GET_PCID(_pcid_addr) ((_pcid_addr) >> PHY_CTX_SHIFT) + +#define MB_KERNEL_CTX_SHIFT 8 +#define MB_KERNEL_CTX_SIZE (1 << MB_KERNEL_CTX_SHIFT) +#define MB_KERNEL_CTX_MASK (MB_KERNEL_CTX_SIZE - 1) +#define MB_GET_CID_ADDR(_cid) (0x10000 + ((_cid) << MB_KERNEL_CTX_SHIFT)) + +#define MAX_CID_CNT 0x4000 +#define MAX_CID_ADDR (GET_CID_ADDR(MAX_CID_CNT)) +#define INVALID_CID_ADDR 0xffffffff + +#define TX_CID 16 +#define RX_CID 0 + +#define MB_TX_CID_ADDR MB_GET_CID_ADDR(TX_CID) +#define MB_RX_CID_ADDR MB_GET_CID_ADDR(RX_CID) + +struct sw_bd { + struct sk_buff *skb; + DECLARE_PCI_UNMAP_ADDR(mapping) +}; + +/* Buffered flash (Atmel: AT45DB011B) specific information */ +#define SEEPROM_PAGE_BITS 2 +#define SEEPROM_PHY_PAGE_SIZE (1 << SEEPROM_PAGE_BITS) +#define SEEPROM_BYTE_ADDR_MASK (SEEPROM_PHY_PAGE_SIZE-1) +#define SEEPROM_PAGE_SIZE 4 +#define SEEPROM_TOTAL_SIZE 65536 + +#define BUFFERED_FLASH_PAGE_BITS 9 +#define BUFFERED_FLASH_PHY_PAGE_SIZE (1 << BUFFERED_FLASH_PAGE_BITS) +#define BUFFERED_FLASH_BYTE_ADDR_MASK (BUFFERED_FLASH_PHY_PAGE_SIZE-1) +#define BUFFERED_FLASH_PAGE_SIZE 264 +#define BUFFERED_FLASH_TOTAL_SIZE 131072 + +#define SAIFUN_FLASH_PAGE_BITS 8 +#define SAIFUN_FLASH_PHY_PAGE_SIZE (1 << SAIFUN_FLASH_PAGE_BITS) +#define SAIFUN_FLASH_BYTE_ADDR_MASK (SAIFUN_FLASH_PHY_PAGE_SIZE-1) +#define SAIFUN_FLASH_PAGE_SIZE 256 +#define SAIFUN_FLASH_BASE_TOTAL_SIZE 65536 + +#define NVRAM_TIMEOUT_COUNT 30000 + + +#define FLASH_STRAP_MASK (BNX2_NVM_CFG1_FLASH_MODE | \ + BNX2_NVM_CFG1_BUFFER_MODE | \ + BNX2_NVM_CFG1_PROTECT_MODE | \ + BNX2_NVM_CFG1_FLASH_SIZE) + +struct flash_spec { + u32 strapping; + u32 config1; + u32 config2; + u32 config3; + u32 write1; + u32 buffered; + u32 page_bits; + u32 page_size; + u32 addr_mask; + u32 total_size; + u8 *name; +}; + +struct bnx2 { + /* Fields used in the tx and intr/napi performance paths are grouped */ + /* together in the beginning of the structure. */ + void __iomem *regview; + + struct net_device *dev; + struct pci_dev *pdev; + + atomic_t intr_sem; + + struct status_block *status_blk; + u32 last_status_idx; + + atomic_t tx_avail_bd; + struct tx_bd *tx_desc_ring; + struct sw_bd *tx_buf_ring; + u32 tx_prod_bseq; + u16 tx_prod; + u16 tx_cons; + +#ifdef BCM_VLAN + struct vlan_group *vlgrp; +#endif + + u32 rx_offset; + u32 rx_buf_use_size; /* useable size */ + u32 rx_buf_size; /* with alignment */ + struct rx_bd *rx_desc_ring; + struct sw_bd *rx_buf_ring; + u32 rx_prod_bseq; + u16 rx_prod; + u16 rx_cons; + + u32 rx_csum; + + /* Only used to synchronize netif_stop_queue/wake_queue when tx */ + /* ring is full */ + spinlock_t tx_lock; + + /* End of fileds used in the performance code paths. */ + + char *name; + + int timer_interval; + struct timer_list timer; + struct work_struct reset_task; + + /* Used to synchronize phy accesses. */ + spinlock_t phy_lock; + + u32 flags; +#define PCIX_FLAG 1 +#define PCI_32BIT_FLAG 2 +#define ONE_TDMA_FLAG 4 /* no longer used */ +#define NO_WOL_FLAG 8 +#define USING_DAC_FLAG 0x10 +#define USING_MSI_FLAG 0x20 + + u32 phy_flags; +#define PHY_SERDES_FLAG 1 +#define PHY_CRC_FIX_FLAG 2 +#define PHY_PARALLEL_DETECT_FLAG 4 +#define PHY_INT_MODE_MASK_FLAG 0x300 +#define PHY_INT_MODE_AUTO_POLLING_FLAG 0x100 +#define PHY_INT_MODE_LINK_READY_FLAG 0x200 + + u32 chip_id; + /* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */ +#define CHIP_NUM(bp) (((bp)->chip_id) & 0xffff0000) +#define CHIP_NUM_5706 0x57060000 + +#define CHIP_REV(bp) (((bp)->chip_id) & 0x0000f000) +#define CHIP_REV_Ax 0x00000000 +#define CHIP_REV_Bx 0x00001000 +#define CHIP_REV_Cx 0x00002000 + +#define CHIP_METAL(bp) (((bp)->chip_id) & 0x00000ff0) +#define CHIP_BONDING(bp) (((bp)->chip_id) & 0x0000000f) + +#define CHIP_ID(bp) (((bp)->chip_id) & 0xfffffff0) +#define CHIP_ID_5706_A0 0x57060000 +#define CHIP_ID_5706_A1 0x57060010 + +#define CHIP_BOND_ID(bp) (((bp)->chip_id) & 0xf) + +/* A serdes chip will have the first bit of the bond id set. */ +#define CHIP_BOND_ID_SERDES_BIT 0x01 + + u32 phy_addr; + u32 phy_id; + + u16 bus_speed_mhz; + u8 wol; + + u8 fw_timed_out; + + u16 fw_wr_seq; + u16 fw_drv_pulse_wr_seq; + + int tx_ring_size; + dma_addr_t tx_desc_mapping; + + + int rx_ring_size; + dma_addr_t rx_desc_mapping; + + u16 tx_quick_cons_trip; + u16 tx_quick_cons_trip_int; + u16 rx_quick_cons_trip; + u16 rx_quick_cons_trip_int; + u16 comp_prod_trip; + u16 comp_prod_trip_int; + u16 tx_ticks; + u16 tx_ticks_int; + u16 com_ticks; + u16 com_ticks_int; + u16 cmd_ticks; + u16 cmd_ticks_int; + u16 rx_ticks; + u16 rx_ticks_int; + + u32 stats_ticks; + + dma_addr_t status_blk_mapping; + + struct statistics_block *stats_blk; + dma_addr_t stats_blk_mapping; + + u32 rx_mode; + + u16 req_line_speed; + u8 req_duplex; + + u8 link_up; + + u16 line_speed; + u8 duplex; + u8 flow_ctrl; /* actual flow ctrl settings */ + /* may be different from */ + /* req_flow_ctrl if autoneg */ +#define FLOW_CTRL_TX 1 +#define FLOW_CTRL_RX 2 + + u32 advertising; + + u8 req_flow_ctrl; /* flow ctrl advertisement */ + /* settings or forced */ + /* settings */ + u8 autoneg; +#define AUTONEG_SPEED 1 +#define AUTONEG_FLOW_CTRL 2 + + u8 loopback; +#define MAC_LOOPBACK 1 +#define PHY_LOOPBACK 2 + + u8 serdes_an_pending; +#define SERDES_AN_TIMEOUT (2 * HZ) + + u8 mac_addr[8]; + + u32 fw_ver; + + int pm_cap; + int pcix_cap; + + struct net_device_stats net_stats; + + struct flash_spec *flash_info; +}; + +static u32 bnx2_reg_rd_ind(struct bnx2 *bp, u32 offset); +static void bnx2_reg_wr_ind(struct bnx2 *bp, u32 offset, u32 val); + +#define REG_RD(bp, offset) \ + readl(bp->regview + offset) + +#define REG_WR(bp, offset, val) \ + writel(val, bp->regview + offset) + +#define REG_WR16(bp, offset, val) \ + writew(val, bp->regview + offset) + +#define REG_RD_IND(bp, offset) \ + bnx2_reg_rd_ind(bp, offset) + +#define REG_WR_IND(bp, offset, val) \ + bnx2_reg_wr_ind(bp, offset, val) + +/* Indirect context access. Unlike the MBQ_WR, these macros will not + * trigger a chip event. */ +static void bnx2_ctx_wr(struct bnx2 *bp, u32 cid_addr, u32 offset, u32 val); + +#define CTX_WR(bp, cid_addr, offset, val) \ + bnx2_ctx_wr(bp, cid_addr, offset, val) + +struct cpu_reg { + u32 mode; + u32 mode_value_halt; + u32 mode_value_sstep; + + u32 state; + u32 state_value_clear; + + u32 gpr0; + u32 evmask; + u32 pc; + u32 inst; + u32 bp; + + u32 spad_base; + + u32 mips_view_base; +}; + +struct fw_info { + u32 ver_major; + u32 ver_minor; + u32 ver_fix; + + u32 start_addr; + + /* Text section. */ + u32 text_addr; + u32 text_len; + u32 text_index; + u32 *text; + + /* Data section. */ + u32 data_addr; + u32 data_len; + u32 data_index; + u32 *data; + + /* SBSS section. */ + u32 sbss_addr; + u32 sbss_len; + u32 sbss_index; + u32 *sbss; + + /* BSS section. */ + u32 bss_addr; + u32 bss_len; + u32 bss_index; + u32 *bss; + + /* Read-only section. */ + u32 rodata_addr; + u32 rodata_len; + u32 rodata_index; + u32 *rodata; +}; + +#define RV2P_PROC1 0 +#define RV2P_PROC2 1 + + +/* This value (in milliseconds) determines the frequency of the driver + * issuing the PULSE message code. The firmware monitors this periodic + * pulse to determine when to switch to an OS-absent mode. */ +#define DRV_PULSE_PERIOD_MS 250 + +/* This value (in milliseconds) determines how long the driver should + * wait for an acknowledgement from the firmware before timing out. Once + * the firmware has timed out, the driver will assume there is no firmware + * running and there won't be any firmware-driver synchronization during a + * driver reset. */ +#define FW_ACK_TIME_OUT_MS 50 + + +#define BNX2_DRV_RESET_SIGNATURE 0x00000000 +#define BNX2_DRV_RESET_SIGNATURE_MAGIC 0x4841564b /* HAVK */ +//#define DRV_RESET_SIGNATURE_MAGIC 0x47495352 /* RSIG */ + +#define BNX2_DRV_MB 0x00000004 +#define BNX2_DRV_MSG_CODE 0xff000000 +#define BNX2_DRV_MSG_CODE_RESET 0x01000000 +#define BNX2_DRV_MSG_CODE_UNLOAD 0x02000000 +#define BNX2_DRV_MSG_CODE_SHUTDOWN 0x03000000 +#define BNX2_DRV_MSG_CODE_SUSPEND_WOL 0x04000000 +#define BNX2_DRV_MSG_CODE_FW_TIMEOUT 0x05000000 +#define BNX2_DRV_MSG_CODE_PULSE 0x06000000 +#define BNX2_DRV_MSG_CODE_DIAG 0x07000000 +#define BNX2_DRV_MSG_CODE_SUSPEND_NO_WOL 0x09000000 + +#define BNX2_DRV_MSG_DATA 0x00ff0000 +#define BNX2_DRV_MSG_DATA_WAIT0 0x00010000 +#define BNX2_DRV_MSG_DATA_WAIT1 0x00020000 +#define BNX2_DRV_MSG_DATA_WAIT2 0x00030000 +#define BNX2_DRV_MSG_DATA_WAIT3 0x00040000 + +#define BNX2_DRV_MSG_SEQ 0x0000ffff + +#define BNX2_FW_MB 0x00000008 +#define BNX2_FW_MSG_ACK 0x0000ffff +#define BNX2_FW_MSG_STATUS_MASK 0x00ff0000 +#define BNX2_FW_MSG_STATUS_OK 0x00000000 +#define BNX2_FW_MSG_STATUS_FAILURE 0x00ff0000 + +#define BNX2_LINK_STATUS 0x0000000c + +#define BNX2_DRV_PULSE_MB 0x00000010 +#define BNX2_DRV_PULSE_SEQ_MASK 0x0000ffff + +/* Indicate to the firmware not to go into the + * OS absent when it is not getting driver pulse. + * This is used for debugging. */ +#define BNX2_DRV_MSG_DATA_PULSE_CODE_ALWAYS_ALIVE 0x00010000 + +#define BNX2_DEV_INFO_SIGNATURE 0x00000020 +#define BNX2_DEV_INFO_SIGNATURE_MAGIC 0x44564900 +#define BNX2_DEV_INFO_SIGNATURE_MAGIC_MASK 0xffffff00 +#define BNX2_DEV_INFO_FEATURE_CFG_VALID 0x01 +#define BNX2_DEV_INFO_SECONDARY_PORT 0x80 +#define BNX2_DEV_INFO_DRV_ALWAYS_ALIVE 0x40 + +#define BNX2_SHARED_HW_CFG_PART_NUM 0x00000024 + +#define BNX2_SHARED_HW_CFG_POWER_DISSIPATED 0x00000034 +#define BNX2_SHARED_HW_CFG_POWER_STATE_D3_MASK 0xff000000 +#define BNX2_SHARED_HW_CFG_POWER_STATE_D2_MASK 0xff0000 +#define BNX2_SHARED_HW_CFG_POWER_STATE_D1_MASK 0xff00 +#define BNX2_SHARED_HW_CFG_POWER_STATE_D0_MASK 0xff + +#define BNX2_SHARED_HW_CFG POWER_CONSUMED 0x00000038 +#define BNX2_SHARED_HW_CFG_CONFIG 0x0000003c +#define BNX2_SHARED_HW_CFG_DESIGN_NIC 0 +#define BNX2_SHARED_HW_CFG_DESIGN_LOM 0x1 +#define BNX2_SHARED_HW_CFG_PHY_COPPER 0 +#define BNX2_SHARED_HW_CFG_PHY_FIBER 0x2 +#define BNX2_SHARED_HW_CFG_LED_MODE_SHIFT_BITS 8 +#define BNX2_SHARED_HW_CFG_LED_MODE_MASK 0x300 +#define BNX2_SHARED_HW_CFG_LED_MODE_MAC 0 +#define BNX2_SHARED_HW_CFG_LED_MODE_GPHY1 0x100 +#define BNX2_SHARED_HW_CFG_LED_MODE_GPHY2 0x200 + +#define BNX2_DEV_INFO_BC_REV 0x0000004c + +#define BNX2_PORT_HW_CFG_MAC_UPPER 0x00000050 +#define BNX2_PORT_HW_CFG_UPPERMAC_MASK 0xffff + +#define BNX2_PORT_HW_CFG_MAC_LOWER 0x00000054 +#define BNX2_PORT_HW_CFG_CONFIG 0x00000058 + +#define BNX2_PORT_HW_CFG_IMD_MAC_A_UPPER 0x00000068 +#define BNX2_PORT_HW_CFG_IMD_MAC_A_LOWER 0x0000006c +#define BNX2_PORT_HW_CFG_IMD_MAC_B_UPPER 0x00000070 +#define BNX2_PORT_HW_CFG_IMD_MAC_B_LOWER 0x00000074 +#define BNX2_PORT_HW_CFG_ISCSI_MAC_UPPER 0x00000078 +#define BNX2_PORT_HW_CFG_ISCSI_MAC_LOWER 0x0000007c + +#define BNX2_DEV_INFO_PER_PORT_HW_CONFIG2 0x000000b4 + +#define BNX2_DEV_INFO_FORMAT_REV 0x000000c4 +#define BNX2_DEV_INFO_FORMAT_REV_MASK 0xff000000 +#define BNX2_DEV_INFO_FORMAT_REV_ID ('A' << 24) + +#define BNX2_SHARED_FEATURE 0x000000c8 +#define BNX2_SHARED_FEATURE_MASK 0xffffffff + +#define BNX2_PORT_FEATURE 0x000000d8 +#define BNX2_PORT2_FEATURE 0x00000014c +#define BNX2_PORT_FEATURE_WOL_ENABLED 0x01000000 +#define BNX2_PORT_FEATURE_MBA_ENABLED 0x02000000 +#define BNX2_PORT_FEATURE_ASF_ENABLED 0x04000000 +#define BNX2_PORT_FEATURE_IMD_ENABLED 0x08000000 +#define BNX2_PORT_FEATURE_BAR1_SIZE_MASK 0xf +#define BNX2_PORT_FEATURE_BAR1_SIZE_DISABLED 0x0 +#define BNX2_PORT_FEATURE_BAR1_SIZE_64K 0x1 +#define BNX2_PORT_FEATURE_BAR1_SIZE_128K 0x2 +#define BNX2_PORT_FEATURE_BAR1_SIZE_256K 0x3 +#define BNX2_PORT_FEATURE_BAR1_SIZE_512K 0x4 +#define BNX2_PORT_FEATURE_BAR1_SIZE_1M 0x5 +#define BNX2_PORT_FEATURE_BAR1_SIZE_2M 0x6 +#define BNX2_PORT_FEATURE_BAR1_SIZE_4M 0x7 +#define BNX2_PORT_FEATURE_BAR1_SIZE_8M 0x8 +#define BNX2_PORT_FEATURE_BAR1_SIZE_16M 0x9 +#define BNX2_PORT_FEATURE_BAR1_SIZE_32M 0xa +#define BNX2_PORT_FEATURE_BAR1_SIZE_64M 0xb +#define BNX2_PORT_FEATURE_BAR1_SIZE_128M 0xc +#define BNX2_PORT_FEATURE_BAR1_SIZE_256M 0xd +#define BNX2_PORT_FEATURE_BAR1_SIZE_512M 0xe +#define BNX2_PORT_FEATURE_BAR1_SIZE_1G 0xf + +#define BNX2_PORT_FEATURE_WOL 0xdc +#define BNX2_PORT2_FEATURE_WOL 0x150 +#define BNX2_PORT_FEATURE_WOL_DEFAULT_SHIFT_BITS 4 +#define BNX2_PORT_FEATURE_WOL_DEFAULT_MASK 0x30 +#define BNX2_PORT_FEATURE_WOL_DEFAULT_DISABLE 0 +#define BNX2_PORT_FEATURE_WOL_DEFAULT_MAGIC 0x10 +#define BNX2_PORT_FEATURE_WOL_DEFAULT_ACPI 0x20 +#define BNX2_PORT_FEATURE_WOL_DEFAULT_MAGIC_AND_ACPI 0x30 +#define BNX2_PORT_FEATURE_WOL_LINK_SPEED_MASK 0xf +#define BNX2_PORT_FEATURE_WOL_LINK_SPEED_AUTONEG 0 +#define BNX2_PORT_FEATURE_WOL_LINK_SPEED_10HALF 1 +#define BNX2_PORT_FEATURE_WOL_LINK_SPEED_10FULL 2 +#define BNX2_PORT_FEATURE_WOL_LINK_SPEED_100HALF 3 +#define BNX2_PORT_FEATURE_WOL_LINK_SPEED_100FULL 4 +#define BNX2_PORT_FEATURE_WOL_LINK_SPEED_1000HALF 5 +#define BNX2_PORT_FEATURE_WOL_LINK_SPEED_1000FULL 6 +#define BNX2_PORT_FEATURE_WOL_AUTONEG_ADVERTISE_1000 0x40 +#define BNX2_PORT_FEATURE_WOL_RESERVED_PAUSE_CAP 0x400 +#define BNX2_PORT_FEATURE_WOL_RESERVED_ASYM_PAUSE_CAP 0x800 + +#define BNX2_PORT_FEATURE_MBA 0xe0 +#define BNX2_PORT2_FEATURE_MBA 0x154 +#define BNX2_PORT_FEATURE_MBA_BOOT_AGENT_TYPE_SHIFT_BITS 0 +#define BNX2_PORT_FEATURE_MBA_BOOT_AGENT_TYPE_MASK 0x3 +#define BNX2_PORT_FEATURE_MBA_BOOT_AGENT_TYPE_PXE 0 +#define BNX2_PORT_FEATURE_MBA_BOOT_AGENT_TYPE_RPL 1 +#define BNX2_PORT_FEATURE_MBA_BOOT_AGENT_TYPE_BOOTP 2 +#define BNX2_PORT_FEATURE_MBA_LINK_SPEED_SHIFT_BITS 2 +#define BNX2_PORT_FEATURE_MBA_LINK_SPEED_MASK 0x3c +#define BNX2_PORT_FEATURE_MBA_LINK_SPEED_AUTONEG 0 +#define BNX2_PORT_FEATURE_MBA_LINK_SPEED_10HALF 0x4 +#define BNX2_PORT_FEATURE_MBA_LINK_SPEED_10FULL 0x8 +#define BNX2_PORT_FEATURE_MBA_LINK_SPEED_100HALF 0xc +#define BNX2_PORT_FEATURE_MBA_LINK_SPEED_100FULL 0x10 +#define BNX2_PORT_FEATURE_MBA_LINK_SPEED_1000HALF 0x14 +#define BNX2_PORT_FEATURE_MBA_LINK_SPEED_1000FULL 0x18 +#define BNX2_PORT_FEATURE_MBA_SETUP_PROMPT_ENABLE 0x40 +#define BNX2_PORT_FEATURE_MBA_HOTKEY_CTRL_S 0 +#define BNX2_PORT_FEATURE_MBA_HOTKEY_CTRL_B 0x80 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_SHIFT_BITS 8 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_MASK 0xff00 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_DISABLED 0 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_1K 0x100 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_2K 0x200 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_4K 0x300 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_8K 0x400 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_16K 0x500 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_32K 0x600 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_64K 0x700 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_128K 0x800 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_256K 0x900 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_512K 0xa00 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_1M 0xb00 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_2M 0xc00 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_4M 0xd00 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_8M 0xe00 +#define BNX2_PORT_FEATURE_MBA_EXP_ROM_SIZE_16M 0xf00 +#define BNX2_PORT_FEATURE_MBA_MSG_TIMEOUT_SHIFT_BITS 16 +#define BNX2_PORT_FEATURE_MBA_MSG_TIMEOUT_MASK 0xf0000 +#define BNX2_PORT_FEATURE_MBA_BIOS_BOOTSTRAP_SHIFT_BITS 20 +#define BNX2_PORT_FEATURE_MBA_BIOS_BOOTSTRAP_MASK 0x300000 +#define BNX2_PORT_FEATURE_MBA_BIOS_BOOTSTRAP_AUTO 0 +#define BNX2_PORT_FEATURE_MBA_BIOS_BOOTSTRAP_BBS 0x100000 +#define BNX2_PORT_FEATURE_MBA_BIOS_BOOTSTRAP_INT18H 0x200000 +#define BNX2_PORT_FEATURE_MBA_BIOS_BOOTSTRAP_INT19H 0x300000 + +#define BNX2_PORT_FEATURE_IMD 0xe4 +#define BNX2_PORT2_FEATURE_IMD 0x158 +#define BNX2_PORT_FEATURE_IMD_LINK_OVERRIDE_DEFAULT 0 +#define BNX2_PORT_FEATURE_IMD_LINK_OVERRIDE_ENABLE 1 + +#define BNX2_PORT_FEATURE_VLAN 0xe8 +#define BNX2_PORT2_FEATURE_VLAN 0x15c +#define BNX2_PORT_FEATURE_MBA_VLAN_TAG_MASK 0xffff +#define BNX2_PORT_FEATURE_MBA_VLAN_ENABLE 0x10000 + +#define BNX2_BC_STATE_RESET_TYPE 0x000001c0 +#define BNX2_BC_STATE_RESET_TYPE_SIG 0x00005254 +#define BNX2_BC_STATE_RESET_TYPE_SIG_MASK 0x0000ffff +#define BNX2_BC_STATE_RESET_TYPE_NONE (BNX2_BC_STATE_RESET_TYPE_SIG | \ + 0x00010000) +#define BNX2_BC_STATE_RESET_TYPE_PCI (BNX2_BC_STATE_RESET_TYPE_SIG | \ + 0x00020000) +#define BNX2_BC_STATE_RESET_TYPE_VAUX (BNX2_BC_STATE_RESET_TYPE_SIG | \ + 0x00030000) +#define BNX2_BC_STATE_RESET_TYPE_DRV_MASK DRV_MSG_CODE +#define BNX2_BC_STATE_RESET_TYPE_DRV_RESET (BNX2_BC_STATE_RESET_TYPE_SIG | \ + DRV_MSG_CODE_RESET) +#define BNX2_BC_STATE_RESET_TYPE_DRV_UNLOAD (BNX2_BC_STATE_RESET_TYPE_SIG | \ + DRV_MSG_CODE_UNLOAD) +#define BNX2_BC_STATE_RESET_TYPE_DRV_SHUTDOWN (BNX2_BC_STATE_RESET_TYPE_SIG | \ + DRV_MSG_CODE_SHUTDOWN) +#define BNX2_BC_STATE_RESET_TYPE_DRV_WOL (BNX2_BC_STATE_RESET_TYPE_SIG | \ + DRV_MSG_CODE_WOL) +#define BNX2_BC_STATE_RESET_TYPE_DRV_DIAG (BNX2_BC_STATE_RESET_TYPE_SIG | \ + DRV_MSG_CODE_DIAG) +#define BNX2_BC_STATE_RESET_TYPE_VALUE(msg) (BNX2_BC_STATE_RESET_TYPE_SIG | \ + (msg)) + +#define BNX2_BC_STATE 0x000001c4 +#define BNX2_BC_STATE_ERR_MASK 0x0000ff00 +#define BNX2_BC_STATE_SIGN 0x42530000 +#define BNX2_BC_STATE_SIGN_MASK 0xffff0000 +#define BNX2_BC_STATE_BC1_START (BNX2_BC_STATE_SIGN | 0x1) +#define BNX2_BC_STATE_GET_NVM_CFG1 (BNX2_BC_STATE_SIGN | 0x2) +#define BNX2_BC_STATE_PROG_BAR (BNX2_BC_STATE_SIGN | 0x3) +#define BNX2_BC_STATE_INIT_VID (BNX2_BC_STATE_SIGN | 0x4) +#define BNX2_BC_STATE_GET_NVM_CFG2 (BNX2_BC_STATE_SIGN | 0x5) +#define BNX2_BC_STATE_APPLY_WKARND (BNX2_BC_STATE_SIGN | 0x6) +#define BNX2_BC_STATE_LOAD_BC2 (BNX2_BC_STATE_SIGN | 0x7) +#define BNX2_BC_STATE_GOING_BC2 (BNX2_BC_STATE_SIGN | 0x8) +#define BNX2_BC_STATE_GOING_DIAG (BNX2_BC_STATE_SIGN | 0x9) +#define BNX2_BC_STATE_RT_FINAL_INIT (BNX2_BC_STATE_SIGN | 0x81) +#define BNX2_BC_STATE_RT_WKARND (BNX2_BC_STATE_SIGN | 0x82) +#define BNX2_BC_STATE_RT_DRV_PULSE (BNX2_BC_STATE_SIGN | 0x83) +#define BNX2_BC_STATE_RT_FIOEVTS (BNX2_BC_STATE_SIGN | 0x84) +#define BNX2_BC_STATE_RT_DRV_CMD (BNX2_BC_STATE_SIGN | 0x85) +#define BNX2_BC_STATE_RT_LOW_POWER (BNX2_BC_STATE_SIGN | 0x86) +#define BNX2_BC_STATE_RT_SET_WOL (BNX2_BC_STATE_SIGN | 0x87) +#define BNX2_BC_STATE_RT_OTHER_FW (BNX2_BC_STATE_SIGN | 0x88) +#define BNX2_BC_STATE_RT_GOING_D3 (BNX2_BC_STATE_SIGN | 0x89) +#define BNX2_BC_STATE_ERR_BAD_VERSION (BNX2_BC_STATE_SIGN | 0x0100) +#define BNX2_BC_STATE_ERR_BAD_BC2_CRC (BNX2_BC_STATE_SIGN | 0x0200) +#define BNX2_BC_STATE_ERR_BC1_LOOP (BNX2_BC_STATE_SIGN | 0x0300) +#define BNX2_BC_STATE_ERR_UNKNOWN_CMD (BNX2_BC_STATE_SIGN | 0x0400) +#define BNX2_BC_STATE_ERR_DRV_DEAD (BNX2_BC_STATE_SIGN | 0x0500) +#define BNX2_BC_STATE_ERR_NO_RXP (BNX2_BC_STATE_SIGN | 0x0600) +#define BNX2_BC_STATE_ERR_TOO_MANY_RBUF (BNX2_BC_STATE_SIGN | 0x0700) + +#define BNX2_BC_STATE_DEBUG_CMD 0x1dc +#define BNX2_BC_STATE_BC_DBG_CMD_SIGNATURE 0x42440000 +#define BNX2_BC_STATE_BC_DBG_CMD_SIGNATURE_MASK 0xffff0000 +#define BNX2_BC_STATE_BC_DBG_CMD_LOOP_CNT_MASK 0xffff +#define BNX2_BC_STATE_BC_DBG_CMD_LOOP_INFINITE 0xffff + +#define HOST_VIEW_SHMEM_BASE 0x167c00 + +#endif diff --git a/drivers/net/bnx2_fw.h b/drivers/net/bnx2_fw.h new file mode 100644 index 000000000000..35f3a2ae5ef1 --- /dev/null +++ b/drivers/net/bnx2_fw.h @@ -0,0 +1,2468 @@ +/* bnx2_fw.h: Broadcom NX2 network driver. + * + * Copyright (c) 2004, 2005 Broadcom Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, except as noted below. + * + * This file contains firmware data derived from proprietary unpublished + * source code, Copyright (c) 2004, 2005 Broadcom Corporation. + * + * Permission is hereby granted for the distribution of this firmware data + * in hexadecimal or equivalent format, provided this copyright notice is + * accompanying it. + */ + + +static int bnx2_COM_b06FwReleaseMajor = 0x0; +static int bnx2_COM_b06FwReleaseMinor = 0x0; +static int bnx2_COM_b06FwReleaseFix = 0x0; +static u32 bnx2_COM_b06FwStartAddr = 0x080004a0; +static u32 bnx2_COM_b06FwTextAddr = 0x08000000; +static int bnx2_COM_b06FwTextLen = 0x4594; +static u32 bnx2_COM_b06FwDataAddr = 0x080045e0; +static int bnx2_COM_b06FwDataLen = 0x0; +static u32 bnx2_COM_b06FwRodataAddr = 0x08004598; +static int bnx2_COM_b06FwRodataLen = 0x18; +static u32 bnx2_COM_b06FwBssAddr = 0x08004600; +static int bnx2_COM_b06FwBssLen = 0x88; +static u32 bnx2_COM_b06FwSbssAddr = 0x080045e0; +static int bnx2_COM_b06FwSbssLen = 0x1c; +static u32 bnx2_COM_b06FwText[(0x4594/4) + 1] = { + 0x0a000128, 0x00000000, 0x00000000, 0x0000000d, 0x636f6d20, 0x302e362e, + 0x39000000, 0x00060902, 0x00000000, 0x00000003, 0x00000014, 0x00000032, + 0x00000003, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000010, 0x000003e8, 0x0000ea60, 0x00000001, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000002, 0x00000020, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x10000003, 0x00000000, 0x0000000d, + 0x0000000d, 0x3c020800, 0x244245e0, 0x3c030800, 0x24634688, 0xac400000, + 0x0043202b, 0x1480fffd, 0x24420004, 0x3c1d0800, 0x37bd7ffc, 0x03a0f021, + 0x3c100800, 0x261004a0, 0x3c1c0800, 0x279c45e0, 0x0e0001f2, 0x00000000, + 0x0000000d, 0x27bdffe8, 0x3c1a8000, 0x3c020008, 0x0342d825, 0x3c036010, + 0xafbf0010, 0x8c655000, 0x3c020800, 0x24470ac8, 0x3c040800, 0x24864600, + 0x2402ff7f, 0x00a22824, 0x34a5380c, 0xac655000, 0x00002821, 0x24020037, + 0x24030c80, 0xaf420008, 0xaf430024, 0xacc70000, 0x24a50001, 0x2ca20016, + 0x1440fffc, 0x24c60004, 0x24844600, 0x3c020800, 0x24420ad4, 0x3c030800, + 0x246309d4, 0xac820004, 0x3c020800, 0x24420618, 0x3c050800, 0x24a50ca0, + 0xac82000c, 0x3c020800, 0x24423100, 0xac830008, 0x3c030800, 0x246325c8, + 0xac820014, 0x3c020800, 0x24422b0c, 0xac830018, 0xac83001c, 0x3c030800, + 0x24630adc, 0xac820024, 0x3c020800, 0x24423040, 0xac83002c, 0x3c030800, + 0x24633060, 0xac820030, 0x3c020800, 0x24422f6c, 0xac830034, 0x3c030800, + 0x24632c60, 0xac82003c, 0x3c020800, 0x24420b6c, 0xac850010, 0xac850020, + 0xac830040, 0x0e000bd6, 0xac820050, 0x8fbf0010, 0x03e00008, 0x27bd0018, + 0x27bdffe0, 0xafb00010, 0x27500100, 0xafbf0018, 0xafb10014, 0x9203000b, + 0x24020003, 0x1462005b, 0x96110008, 0x32220001, 0x10400009, 0x27430080, + 0x8e020000, 0x96040014, 0x000211c2, 0x00021040, 0x00621821, 0xa4640000, + 0x0a0001cb, 0x3c020800, 0x3c020800, 0x8c430020, 0x1060002a, 0x3c030800, + 0x0e001006, 0x00000000, 0x97420108, 0x8f850018, 0x9743010c, 0x3042003e, + 0x00021400, 0x00621825, 0xaca30000, 0x8f840018, 0x8f420100, 0xac820004, + 0x97430116, 0x9742010e, 0x8f840018, 0x00031c00, 0x00431025, 0xac820008, + 0x97430110, 0x97440112, 0x8f850018, 0x00031c00, 0x00832025, 0xaca4000c, + 0x97420114, 0x8f840018, 0x3042ffff, 0xac820010, 0x8f830018, 0xac600014, + 0x8f820018, 0x3c030800, 0xac400018, 0x9462466e, 0x8f840018, 0x3c032000, + 0x00431025, 0xac82001c, 0x0e001044, 0x24040001, 0x3c030800, 0x8c620040, + 0x24420001, 0xac620040, 0x3c020800, 0x8c430044, 0x32240004, 0x24630001, + 0x10800017, 0xac430044, 0x8f4202b8, 0x04430007, 0x8e020020, 0x3c040800, + 0x8c830060, 0x24020001, 0x24630001, 0x0a0001ed, 0xac830060, 0x3c060800, + 0x8cc4005c, 0xaf420280, 0x96030016, 0x00001021, 0xa7430284, 0x8e050004, + 0x24840001, 0x3c031000, 0xaf450288, 0xaf4302b8, 0x0a0001ed, 0xacc4005c, + 0x32220002, 0x0a0001ed, 0x0002102b, 0x3c026000, 0xac400808, 0x0000000d, + 0x00001021, 0x8fbf0018, 0x8fb10014, 0x8fb00010, 0x03e00008, 0x27bd0020, + 0x27bdffc8, 0xafbf0034, 0xafbe0030, 0xafb7002c, 0xafb60028, 0xafb50024, + 0xafb40020, 0xafb3001c, 0xafb20018, 0xafb10014, 0x0e00013f, 0xafb00010, + 0x24110020, 0x24150030, 0x2794000c, 0x27930008, 0x3c124000, 0x3c1e0800, + 0x3c170800, 0x3c160800, 0x8f820004, 0x3c040800, 0x8c830020, 0x10430004, + 0x00000000, 0xaf830004, 0x0e00110b, 0x00000000, 0x8f500000, 0x32020007, + 0x1040fff5, 0x32020001, 0x1040002b, 0x32020002, 0x8f420100, 0xaf420020, + 0x8f430104, 0xaf4300a8, 0x9342010b, 0x93630000, 0x306300ff, 0x10710005, + 0x304400ff, 0x10750006, 0x2c820016, 0x0a000227, 0x00000000, 0xaf940000, + 0x0a000228, 0x2c820016, 0xaf930000, 0x0a000228, 0x00000000, 0xaf800000, + 0x14400005, 0x00041880, 0x0e0002b2, 0x00000000, 0x0a000234, 0x00000000, + 0x3c020800, 0x24424600, 0x00621821, 0x8c620000, 0x0040f809, 0x00000000, + 0x10400005, 0x8fc20034, 0x8f420104, 0x3c016020, 0xac220014, 0x8fc20034, + 0xaf520138, 0x24420001, 0xafc20034, 0x32020002, 0x10400019, 0x32020004, + 0x8f420140, 0xaf420020, 0x93630000, 0x306300ff, 0x10710005, 0x00000000, + 0x10750006, 0x00000000, 0x0a000250, 0x00000000, 0xaf940000, 0x0a000251, + 0x00000000, 0xaf930000, 0x0a000251, 0x00000000, 0xaf800000, 0x0e0008b9, + 0x00000000, 0x8ee20038, 0xaf520178, 0x24420001, 0xaee20038, 0x32020004, + 0x1040ffad, 0x00000000, 0x8f420180, 0xaf420020, 0x93630000, 0x306300ff, + 0x10710005, 0x00000000, 0x10750006, 0x00000000, 0x0a00026a, 0x00000000, + 0xaf940000, 0x0a00026b, 0x00000000, 0xaf930000, 0x0a00026b, 0x00000000, + 0xaf800000, 0x93620000, 0x14510004, 0x8ec2003c, 0x0e000835, 0x00000000, + 0x8ec2003c, 0xaf5201b8, 0x24420001, 0x0a000206, 0xaec2003c, 0x27bdffe8, + 0xafbf0010, 0x97420108, 0x24033000, 0x30447000, 0x10830012, 0x28823001, + 0x10400007, 0x24024000, 0x1080000b, 0x24022000, 0x1082001a, 0x24020001, + 0x0a000299, 0x00000000, 0x1082000c, 0x24025000, 0x1082000e, 0x00000000, + 0x0a000299, 0x00000000, 0x0000000d, 0x0a00029b, 0x00001021, 0x0e000300, + 0x00000000, 0x0a00029b, 0x00001021, 0x0e00048f, 0x00000000, 0x0a00029b, + 0x00001021, 0x0e000fdf, 0x00000000, 0x0a00029b, 0x00001021, 0x0000000d, + 0x00001021, 0x8fbf0010, 0x03e00008, 0x27bd0018, 0x93620000, 0x24030020, + 0x304400ff, 0x10830005, 0x24020030, 0x10820007, 0x00000000, 0x0a0002af, + 0x00000000, 0x2782000c, 0xaf820000, 0x03e00008, 0x00000000, 0x27820008, + 0xaf820000, 0x03e00008, 0x00000000, 0xaf800000, 0x03e00008, 0x00000000, + 0x0000000d, 0x03e00008, 0x00001021, 0x03e00008, 0x00001021, 0x27440100, + 0x94830008, 0x30620004, 0x10400017, 0x30620002, 0x8f4202b8, 0x04430007, + 0x8c820020, 0x3c040800, 0x8c830060, 0x24020001, 0x24630001, 0x03e00008, + 0xac830060, 0xaf420280, 0x94830016, 0x3c060800, 0xa7430284, 0x8c850004, + 0x8cc4005c, 0x00001021, 0x3c031000, 0x24840001, 0xaf450288, 0xaf4302b8, + 0x03e00008, 0xacc4005c, 0x14400003, 0x3c040800, 0x03e00008, 0x00001021, + 0x8c830084, 0x24020001, 0x24630001, 0x03e00008, 0xac830084, 0x27450100, + 0x3c040800, 0x8c820088, 0x94a3000c, 0x24420001, 0x007a1821, 0xac820088, + 0x8ca40018, 0x90664000, 0xaf440038, 0x8ca2001c, 0x2403fff8, 0x00063600, + 0x00431024, 0x34420004, 0x3c030005, 0xaf42003c, 0xaf430030, 0x00000000, + 0x00000000, 0x00000000, 0xaf460404, 0x00000000, 0x00000000, 0x00000000, + 0x3c020006, 0x34420001, 0xaf420030, 0x00000000, 0x00000000, 0x00000000, + 0x8f420000, 0x30420010, 0x1040fffd, 0x00001021, 0x03e00008, 0x00000000, + 0x3c020800, 0x8c430020, 0x27bdffe8, 0xafb00010, 0x27500100, 0x1060001e, + 0xafbf0014, 0x0e001006, 0x00000000, 0x8f830018, 0x8e020018, 0xac620000, + 0x8f840018, 0x9602000c, 0xac820004, 0x8f830018, 0xac600008, 0x8f820018, + 0xac40000c, 0x8f830018, 0xac600010, 0x8f820018, 0xac400014, 0x8f840018, + 0x3c026000, 0x8c434448, 0xac830018, 0x96020008, 0x3c030800, 0x9464466e, + 0x8f850018, 0x00021400, 0x00441025, 0x24040001, 0x0e001044, 0xaca2001c, + 0x8fbf0014, 0x8fb00010, 0x03e00008, 0x27bd0018, 0x27bdffc8, 0xafb3001c, + 0x00009821, 0xafb7002c, 0x0000b821, 0xafbe0030, 0x0000f021, 0xafb50024, + 0x27550100, 0xafbf0034, 0xafb60028, 0xafb40020, 0xafb20018, 0xafb10014, + 0xafb00010, 0x96a20008, 0x8f540100, 0x8eb20018, 0x30420001, 0x10400037, + 0x02a0b021, 0x8f630054, 0x2642ffff, 0x00431023, 0x18400006, 0x00000000, + 0x0000000d, 0x00000000, 0x24000128, 0x0a000372, 0x00002021, 0x8f62004c, + 0x02421023, 0x18400028, 0x00002021, 0x93650120, 0x93640121, 0x3c030800, + 0x8c62008c, 0x308400ff, 0x24420001, 0x30a500ff, 0x00803821, 0x1485000b, + 0xac62008c, 0x3c040800, 0x8c830090, 0x24630001, 0xac830090, 0x93620122, + 0x30420001, 0x00021023, 0x30420005, 0x0a000372, 0x34440004, 0x27660100, + 0x00041080, 0x00c21021, 0x8c430000, 0x02431823, 0x04600004, 0x24820001, + 0x30440007, 0x1485fff9, 0x00041080, 0x10870007, 0x3c030800, 0xa3640121, + 0x8c620094, 0x24040005, 0x24420001, 0x0a000372, 0xac620094, 0x24040004, + 0x00809821, 0x9362003f, 0x304400ff, 0x38830016, 0x2c630001, 0x38820010, + 0x2c420001, 0x00621825, 0x1460000c, 0x24020001, 0x38830008, 0x2c630001, + 0x38820014, 0x2c420001, 0x00621825, 0x14600005, 0x24020001, 0x24020012, + 0x14820002, 0x00001021, 0x24020001, 0x50400007, 0x8eb10020, 0x8ea20020, + 0x8f630040, 0x00408821, 0x00431023, 0x5c400001, 0x8f710040, 0x9343010b, + 0x24020004, 0x54620005, 0x36730080, 0x96a20008, 0x36730002, 0x24170001, + 0x305e0020, 0x2402fffb, 0x02628024, 0x1200002a, 0x3c030800, 0x8c620030, + 0x02021024, 0x10400026, 0x3c020800, 0x8c430020, 0x10600024, 0x32620004, + 0x0e001006, 0x00000000, 0x8f830018, 0x8f420100, 0xac620000, 0x8f840018, + 0x02201821, 0x32620002, 0xac900004, 0x8f840018, 0x50400001, 0x8ec30014, + 0xac830008, 0x8f830018, 0x8ec20020, 0xac62000c, 0x8f840018, 0x8f620040, + 0xac820010, 0x8f830018, 0x8ec20018, 0xac620014, 0x8f840018, 0x3c026000, + 0x8c434448, 0x3c020800, 0xac830018, 0x9443466e, 0x8f840018, 0x3c024010, + 0x00621825, 0xac83001c, 0x0e001044, 0x24040001, 0x32620004, 0x10400076, + 0x00003821, 0x3c029000, 0x34420001, 0x3c038000, 0x02821025, 0xa360007c, + 0xaf420020, 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x93620023, + 0x30420080, 0x10400011, 0x00000000, 0x8f65005c, 0x8f63004c, 0x9764003c, + 0x8f620064, 0x00a32823, 0x00852821, 0x00a2102b, 0x54400006, 0x3c023fff, + 0x93620023, 0x3042007f, 0xa3620023, 0xaf720064, 0x3c023fff, 0x0a0003f1, + 0x3442ffff, 0x8f62005c, 0x02421023, 0x04400011, 0x00000000, 0x8f65005c, + 0x8f630064, 0x9764003c, 0x3c023fff, 0x3442ffff, 0xaf720064, 0x00a32823, + 0x00852821, 0x0045102b, 0x10400004, 0x02451021, 0x3c053fff, 0x34a5ffff, + 0x02451021, 0xaf62005c, 0x24070001, 0xaf72004c, 0x8f620054, 0x16420005, + 0x00000000, 0x93620023, 0x30420040, 0x10400017, 0x24020001, 0x9762006a, + 0x00022880, 0x50a00001, 0x24050001, 0x97630068, 0x93640081, 0x3c020800, + 0x8c46004c, 0x00652821, 0x00852804, 0x00c5102b, 0x54400001, 0x00a03021, + 0x3c020800, 0x8c440050, 0x00c4182b, 0x54600001, 0x00c02021, 0x8f420074, + 0x2403fffe, 0x00832824, 0x00a21021, 0xaf62000c, 0x3c028000, 0x34420001, + 0x02821025, 0xa3600081, 0xaf420020, 0x9363007e, 0x9362007a, 0x10620004, + 0x00000000, 0x0e000f2a, 0x00000000, 0x00403821, 0x10e00017, 0x3c029000, + 0x34420001, 0x02821025, 0xaf420020, 0x3c038000, 0x8f420020, 0x00431024, + 0x1440fffd, 0x3c028000, 0x9363007d, 0x34420001, 0x3c048000, 0x02821025, + 0xa363007d, 0xaf420020, 0x8f4201f8, 0x00441024, 0x1440fffd, 0x24020002, + 0x3c031000, 0xaf5401c0, 0xa34201c4, 0xaf4301f8, 0x8ea30014, 0x8f620040, + 0x14430003, 0x00431023, 0x0a000443, 0x00001021, 0x28420001, 0x10400034, + 0x00000000, 0x8f620040, 0xaf630040, 0x9362003e, 0x30420001, 0x1440000b, + 0x3c029000, 0x93620022, 0x24420001, 0xa3620022, 0x93630022, 0x3c020800, + 0x8c440098, 0x0064182b, 0x1460001e, 0x3c020800, 0x3c029000, 0x34420001, + 0x02821025, 0xaf420020, 0x3c038000, 0x8f420020, 0x00431024, 0x1440fffd, + 0x00000000, 0x3c038000, 0x9362007d, 0x34630001, 0x3c048000, 0x02831825, + 0x34420001, 0xa362007d, 0xaf430020, 0x8f4201f8, 0x00441024, 0x1440fffd, + 0x24020002, 0x3c031000, 0xaf5401c0, 0xa34201c4, 0x24020001, 0xaf4301f8, + 0xa7620012, 0x0a000476, 0xa3600022, 0x9743007a, 0x9444002a, 0x00641821, + 0x3063fffe, 0xa7630012, 0x0e000b68, 0x00000000, 0x12e00003, 0x00000000, + 0x0e000f27, 0x00000000, 0x53c00004, 0x96a20008, 0x0e000c10, 0x00000000, + 0x96a20008, 0x8fbf0034, 0x8fbe0030, 0x8fb7002c, 0x8fb60028, 0x8fb50024, + 0x8fb40020, 0x8fb3001c, 0x8fb20018, 0x8fb10014, 0x8fb00010, 0x00021042, + 0x30420001, 0x03e00008, 0x27bd0038, 0x27bdffe8, 0xafbf0010, 0x97420108, + 0x2403000b, 0x304400ff, 0x1083004e, 0x2882000c, 0x10400011, 0x24020006, + 0x1082003e, 0x28820007, 0x10400007, 0x28820008, 0x1080002b, 0x24020001, + 0x1082002e, 0x3c026000, 0x0a000504, 0x00000000, 0x14400061, 0x2882000a, + 0x1440002b, 0x00000000, 0x0a0004ec, 0x00000000, 0x2402001c, 0x1082004e, + 0x2882001d, 0x1040000e, 0x24020019, 0x10820041, 0x2882001a, 0x10400005, + 0x2402000e, 0x10820036, 0x00000000, 0x0a000504, 0x00000000, 0x2402001b, + 0x1082003c, 0x00000000, 0x0a000504, 0x00000000, 0x240200c1, 0x10820040, + 0x288200c2, 0x10400005, 0x24020080, 0x1082001f, 0x00000000, 0x0a000504, + 0x00000000, 0x240200c2, 0x1082003b, 0x00000000, 0x0a000504, 0x00000000, + 0x3c026000, 0x0e000c7d, 0xac400808, 0x0a000506, 0x8fbf0010, 0x8c444448, + 0x3c030800, 0xac640064, 0x0e000c7d, 0x00000000, 0x3c026000, 0x8c444448, + 0x3c030800, 0x0a000505, 0xac640068, 0x8f440100, 0x0e000508, 0x00000000, + 0x3c026000, 0x8c444448, 0x3c030800, 0x0a000505, 0xac64006c, 0x0e000cab, + 0x00000000, 0x0a000506, 0x8fbf0010, 0x8f440100, 0x0e000cd5, 0x00000000, + 0x0a000506, 0x8fbf0010, 0x0e000d1c, 0x00000000, 0x0a000506, 0x8fbf0010, + 0x0000000d, 0x0a000506, 0x8fbf0010, 0x0e0005d7, 0x00000000, 0x0a000506, + 0x8fbf0010, 0x8f440100, 0x0e000d7e, 0x00000000, 0x0a000506, 0x8fbf0010, + 0x0e000e95, 0x00000000, 0x0a000506, 0x8fbf0010, 0x0e000626, 0x00000000, + 0x0a000506, 0x8fbf0010, 0x0e000b68, 0x00000000, 0x0a000506, 0x8fbf0010, + 0x0000000d, 0x8fbf0010, 0x03e00008, 0x27bd0018, 0x27bdffe8, 0x3c029000, + 0x34420001, 0xafb00010, 0x00808021, 0x02021025, 0x3c038000, 0xafbf0014, + 0xaf420020, 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x93620005, + 0x34420001, 0xa3620005, 0x8f63004c, 0x8f620054, 0x10620019, 0x3c028000, + 0x9762006a, 0x00022880, 0x50a00001, 0x24050001, 0x97630068, 0x93640081, + 0x3c020800, 0x8c46004c, 0x00652821, 0x00852804, 0x00c5102b, 0x54400001, + 0x00a03021, 0x3c020800, 0x8c440050, 0x00c4182b, 0x54600001, 0x00c02021, + 0x8f420074, 0x2403fffe, 0x00832824, 0x00a21021, 0xaf62000c, 0x3c028000, + 0x34420001, 0x02021025, 0x0e000c7d, 0xaf420020, 0x3c029000, 0x34420001, + 0x3c038000, 0x02021025, 0xaf420020, 0x8f420020, 0x00431024, 0x1440fffd, + 0x3c028000, 0x9363007d, 0x34420001, 0x3c048000, 0x02021025, 0xa363007d, + 0xaf420020, 0x8f4201f8, 0x00441024, 0x1440fffd, 0x8fbf0014, 0xaf5001c0, + 0x8fb00010, 0x24020002, 0x3c031000, 0xa34201c4, 0xaf4301f8, 0x03e00008, + 0x27bd0018, 0x27bdffd8, 0xafbf0020, 0xafb3001c, 0xafb20018, 0xafb10014, + 0xafb00010, 0x93630005, 0x00809021, 0x24020030, 0x30630030, 0x14620072, + 0x00a09821, 0x3c020800, 0x8c430020, 0x1060006c, 0x00000000, 0x0e001006, + 0x00000000, 0x8f820018, 0xac520000, 0x9363003e, 0x9362003f, 0x8f840018, + 0x00031a00, 0x00431025, 0xac820004, 0x93630081, 0x93620082, 0x8f850018, + 0x00031e00, 0x00021400, 0x00621825, 0xaca30008, 0x8f840018, 0x8f620040, + 0xac82000c, 0x8f830018, 0x8f620048, 0xac620010, 0x8f840018, 0x8f62004c, + 0x3c110800, 0xac820014, 0x8f830018, 0x8f620050, 0x26304660, 0x00002021, + 0xac620018, 0x9602000e, 0x8f850018, 0x3c03c00b, 0x00431025, 0x0e001044, + 0xaca2001c, 0x8f830018, 0x8f620054, 0xac620000, 0x8f840018, 0x8f620058, + 0xac820004, 0x8f830018, 0x8f62005c, 0xac620008, 0x8f840018, 0x8f620060, + 0xac82000c, 0x8f850018, 0x8f620064, 0xaca20010, 0x97630068, 0x9762006a, + 0x8f840018, 0x00031c00, 0x00431025, 0xac820014, 0x8f830018, 0x00002021, + 0xac600018, 0x9602000e, 0x8f850018, 0x3c03c00c, 0x00431025, 0x0e001044, + 0xaca2001c, 0x8f840018, 0x8f630018, 0xac830000, 0x936200c4, 0x30420002, + 0x10400006, 0x00000000, 0x976200c8, 0x8f830018, 0x3042ffff, 0x0a0005b5, + 0xac620004, 0x8f820018, 0xac400004, 0x8f830018, 0x8f62006c, 0xac620008, + 0x8f840018, 0x8f6200dc, 0xac82000c, 0x8f830018, 0xac600010, 0x93620005, + 0x8f830018, 0x00021600, 0x00531025, 0xac620014, 0x8f850018, 0x3c026000, + 0x8c434448, 0x24040001, 0x26224660, 0xaca30018, 0x9443000e, 0x8f850018, + 0x3c02400d, 0x00621825, 0x0e001044, 0xaca3001c, 0x0e000d48, 0x02402021, + 0x8fbf0020, 0x8fb3001c, 0x8fb20018, 0x8fb10014, 0x8fb00010, 0x03e00008, + 0x27bd0028, 0x27bdffe0, 0xafb00010, 0x27500100, 0xafbf0018, 0xafb10014, + 0x9603000c, 0x240200c1, 0x5462001d, 0x8e040000, 0x3c029000, 0x8f440100, + 0x34420001, 0x3c038000, 0x00821025, 0xaf420020, 0x8f420020, 0x00431024, + 0x1440fffd, 0x00000000, 0x3c038000, 0x9362007d, 0x34630001, 0x3c058000, + 0x00831825, 0x34420004, 0xa362007d, 0xaf430020, 0x8f4201f8, 0x00451024, + 0x1440fffd, 0x24020002, 0x3c031000, 0xaf4401c0, 0xa34201c4, 0xaf4301f8, + 0x0a000622, 0x8fbf0018, 0x8f65004c, 0x24060001, 0x0e000db5, 0x2407049f, + 0x3c020800, 0x8c430020, 0x9611000c, 0x1060001d, 0x8e100000, 0x0e001006, + 0x00000000, 0x8f820018, 0xac500000, 0x8f840018, 0x00111400, 0xac820004, + 0x8f830018, 0xac600008, 0x8f820018, 0xac40000c, 0x8f830018, 0xac600010, + 0x8f840018, 0x240204a2, 0xac820014, 0x8f850018, 0x3c026000, 0x8c434448, + 0x24040001, 0x3c020800, 0xaca30018, 0x9443466e, 0x8f850018, 0x3c024019, + 0x00621825, 0x0e001044, 0xaca3001c, 0x8fbf0018, 0x8fb10014, 0x8fb00010, + 0x03e00008, 0x27bd0020, 0x27bdffb0, 0xafb1002c, 0x27510100, 0xafbf004c, + 0xafbe0048, 0xafb70044, 0xafb60040, 0xafb5003c, 0xafb40038, 0xafb30034, + 0xafb20030, 0xafb00028, 0x8e350000, 0x9634000c, 0x3c026000, 0x8c434448, + 0x0000f021, 0xaf630170, 0x8f620040, 0x8e230014, 0x0000b821, 0x00431023, + 0x044001ec, 0x0000b021, 0x32820010, 0x1040002e, 0x3c026000, 0x9363003f, + 0x9222000e, 0x10430006, 0x2402000c, 0x9223000f, 0x10620003, 0x24020014, + 0x14620025, 0x3c026000, 0x32820004, 0x10400007, 0x241e0001, 0x8f620050, + 0x24420001, 0xaf620050, 0x8f630054, 0x24630001, 0xaf630054, 0x32830102, + 0x24020002, 0x5462000d, 0x9222000f, 0x8f620040, 0x24420001, 0xaf620040, + 0x8f630048, 0x8f620040, 0x24630001, 0x54620005, 0x9222000f, 0x8f620048, + 0x24420001, 0xaf620048, 0x9222000f, 0xa362003f, 0x9223000f, 0x24020012, + 0x14620007, 0x3c026000, 0x3c030800, 0x8c620074, 0x24420001, 0x0e000f6e, + 0xac620074, 0x3c026000, 0x8c434448, 0x32820040, 0xaf630174, 0x32830020, + 0xafa30010, 0x32830080, 0xafa30014, 0x32830001, 0xafa3001c, 0x32830008, + 0xafa30020, 0x32830100, 0x104000bb, 0xafa30018, 0x8e260010, 0x8f630054, + 0x24c2ffff, 0x00431023, 0x18400006, 0x00000000, 0x0000000d, 0x00000000, + 0x24000128, 0x0a0006b2, 0x00009021, 0x8f62004c, 0x00c21023, 0x18400028, + 0x00009021, 0x93650120, 0x93640121, 0x3c030800, 0x8c62008c, 0x308400ff, + 0x24420001, 0x30a500ff, 0x00804021, 0x1485000b, 0xac62008c, 0x3c040800, + 0x8c830090, 0x24630001, 0xac830090, 0x93620122, 0x30420001, 0x00021023, + 0x30420005, 0x0a0006b2, 0x34520004, 0x27670100, 0x00041080, 0x00e21021, + 0x8c430000, 0x00c31823, 0x04600004, 0x24820001, 0x30440007, 0x1485fff9, + 0x00041080, 0x10880007, 0x3c030800, 0xa3640121, 0x8c620094, 0x24120005, + 0x24420001, 0x0a0006b2, 0xac620094, 0x24120004, 0x32420001, 0x10400020, + 0x3c020800, 0x8c430020, 0x8e300000, 0x1060001c, 0x8e330010, 0x0e001006, + 0x00000000, 0x8f820018, 0xac500000, 0x8f840018, 0x24020001, 0xac820004, + 0x8f830018, 0xac600008, 0x8f820018, 0xac40000c, 0x8f830018, 0xac600010, + 0x8f820018, 0xac530014, 0x8f850018, 0x3c026000, 0x8c434448, 0x24040001, + 0x3c020800, 0xaca30018, 0x9443466e, 0x8f850018, 0x3c024010, 0x00621825, + 0x0e001044, 0xaca3001c, 0x32420004, 0x10400060, 0x00003821, 0x3c029000, + 0x8e260010, 0x34420001, 0x3c038000, 0x02a21025, 0xa360007c, 0xaf420020, + 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x93620023, 0x30420080, + 0x10400011, 0x00000000, 0x8f65005c, 0x8f63004c, 0x9764003c, 0x8f620064, + 0x00a32823, 0x00852821, 0x00a2102b, 0x54400006, 0x3c023fff, 0x93620023, + 0x3042007f, 0xa3620023, 0xaf660064, 0x3c023fff, 0x0a000702, 0x3442ffff, + 0x8f62005c, 0x00c21023, 0x04400011, 0x00000000, 0x8f65005c, 0x8f630064, + 0x9764003c, 0x3c023fff, 0x3442ffff, 0xaf660064, 0x00a32823, 0x00852821, + 0x0045102b, 0x10400004, 0x00c51021, 0x3c053fff, 0x34a5ffff, 0x00c51021, + 0xaf62005c, 0x24070001, 0xaf66004c, 0x8f620054, 0x14c20005, 0x00000000, + 0x93620023, 0x30420040, 0x10400017, 0x24020001, 0x9762006a, 0x00022880, + 0x50a00001, 0x24050001, 0x97630068, 0x93640081, 0x3c020800, 0x8c46004c, + 0x00652821, 0x00852804, 0x00c5102b, 0x54400001, 0x00a03021, 0x3c020800, + 0x8c440050, 0x00c4182b, 0x54600001, 0x00c02021, 0x8f420074, 0x2403fffe, + 0x00832824, 0x00a21021, 0xaf62000c, 0x3c028000, 0x34420001, 0x02a21025, + 0xa3600081, 0xaf420020, 0x9363007e, 0x9362007a, 0x10620005, 0x00e0b021, + 0x0e000f2a, 0x00000000, 0x00403821, 0x00e0b021, 0x8fa20010, 0x10400008, + 0x00000000, 0x8e220018, 0xaf620018, 0x8e23001c, 0xaf63001c, 0x8e220020, + 0x24160001, 0xaf620058, 0x13c00036, 0x32820004, 0x10400035, 0x8fa30014, + 0x93620023, 0x30420040, 0x10400031, 0x3c020800, 0x8c430020, 0x1060001c, + 0x8e300000, 0x0e001006, 0x00000000, 0x8f820018, 0xac500000, 0x8f830018, + 0xac600004, 0x8f820018, 0xac400008, 0x8f830018, 0xac60000c, 0x8f820018, + 0xac400010, 0x8f830018, 0x24020587, 0xac620014, 0x8f850018, 0x3c026000, + 0x8c434448, 0x24040001, 0x3c020800, 0xaca30018, 0x9443466e, 0x8f850018, + 0x3c024019, 0x00621825, 0x0e001044, 0xaca3001c, 0x3c029000, 0x34420001, + 0x02a21025, 0xaf420020, 0x3c038000, 0x8f420020, 0x00431024, 0x1440fffd, + 0x24020001, 0xaf62000c, 0x93630023, 0x3c028000, 0x34420001, 0x02a21025, + 0x306300bf, 0xa3630023, 0xaf420020, 0x8fa30014, 0x10600012, 0x8fa3001c, + 0x9362007c, 0x24420001, 0xa362007c, 0x9363007e, 0x9362007a, 0x1462000b, + 0x8fa3001c, 0x9362007c, 0x3c030800, 0x8c640024, 0x0044102b, 0x14400005, + 0x8fa3001c, 0x0e000f2a, 0x00000000, 0x02c2b025, 0x8fa3001c, 0x3062ffff, + 0x10400003, 0x32820200, 0x0a000793, 0x24170004, 0x10400003, 0x00000000, + 0x24170040, 0x24160001, 0x13c0005d, 0x32820002, 0x1040005c, 0x8fa20020, + 0x9222000a, 0x30420020, 0x10400033, 0x3c100800, 0x93620023, 0x30420040, + 0x1040002f, 0x8e020020, 0x1040001e, 0x3c029000, 0x0e001006, 0x00000000, + 0x8f820018, 0xac550000, 0x8f840018, 0x3c02008d, 0xac820004, 0x8f830018, + 0xac600008, 0x8f820018, 0xac40000c, 0x8f830018, 0xac600010, 0x8f840018, + 0x240205bf, 0xac820014, 0x8f850018, 0x3c026000, 0x8c434448, 0x24040001, + 0x3c020800, 0xaca30018, 0x9443466e, 0x8f850018, 0x3c024019, 0x00621825, + 0x0e001044, 0xaca3001c, 0x3c029000, 0x34420001, 0x02a21025, 0xaf420020, + 0x3c038000, 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x93630023, + 0x3c028000, 0x34420001, 0x02a21025, 0x306300bf, 0xa3630023, 0xaf420020, + 0x8e020020, 0x10400023, 0x8fa20020, 0x0e001006, 0x00000000, 0x8f840018, + 0x8e230000, 0xac830000, 0x9222000a, 0x8f830018, 0x00021600, 0xac620004, + 0x8f840018, 0x8f620040, 0xac820008, 0x8f850018, 0x8f63004c, 0xaca3000c, + 0x9362003f, 0x8f840018, 0x304200ff, 0xac820010, 0x8f830018, 0x3c026000, + 0xac600014, 0x8f850018, 0x8c434448, 0x24040001, 0x3c020800, 0xaca30018, + 0x9443466e, 0x8f850018, 0x3c02401a, 0x00621825, 0x0e001044, 0xaca3001c, + 0x8fa20020, 0x1040000e, 0x8fa20018, 0x9222000a, 0xa3620082, 0x56e00005, + 0x36f70008, 0x8fa30018, 0x10600004, 0x00000000, 0x36f70008, 0x0a000801, + 0x24160001, 0x0e000de1, 0x02a02021, 0x8fa20018, 0x10400003, 0x00000000, + 0x36f70010, 0x24160001, 0x12c00019, 0x3c029000, 0x34420001, 0x02a21025, + 0xaf420020, 0x3c038000, 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, + 0x3c038000, 0x9362007d, 0x34630001, 0x3c048000, 0x02a31825, 0x02e21025, + 0xa362007d, 0xaf430020, 0x8f4201f8, 0x00441024, 0x1440fffd, 0x24020002, + 0x3c031000, 0xaf5501c0, 0xa34201c4, 0xaf4301f8, 0x9363003f, 0x24020012, + 0x14620004, 0x3c026000, 0x0e000f6e, 0x00000000, 0x3c026000, 0x8c434448, + 0xaf630178, 0x8fbf004c, 0x8fbe0048, 0x8fb70044, 0x8fb60040, 0x8fb5003c, + 0x8fb40038, 0x8fb30034, 0x8fb20030, 0x8fb1002c, 0x8fb00028, 0x03e00008, + 0x27bd0050, 0x27bdffe8, 0xafbf0014, 0xafb00010, 0x8f500180, 0x97420184, + 0x30420200, 0x14400015, 0x00000000, 0x8f430188, 0x3c02ff00, 0x00621824, + 0x3c020200, 0x10620031, 0x0043102b, 0x14400007, 0x3c020300, 0x1060000b, + 0x3c020100, 0x1062000d, 0x00000000, 0x0a0008b4, 0x00000000, 0x10620027, + 0x3c020400, 0x1062003e, 0x02002021, 0x0a0008b4, 0x00000000, 0x0e000e1e, + 0x02002021, 0x0a0008b6, 0x8fbf0014, 0x93620005, 0x30420020, 0x1440005e, + 0x8fbf0014, 0x3c029000, 0x34420001, 0x02021025, 0xaf420020, 0x3c038000, + 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x93620005, 0x3c038000, + 0x34630001, 0x02031825, 0x34420020, 0xa3620005, 0xaf430020, 0x93620005, + 0x30420020, 0x14400003, 0x02002021, 0x0000000d, 0x02002021, 0x0e000553, + 0x24055854, 0x0a0008b6, 0x8fbf0014, 0x93620005, 0x30420001, 0x1040003f, + 0x3c029000, 0x34420001, 0x02021025, 0xaf420020, 0x3c038000, 0x8f420020, + 0x00431024, 0x1440fffd, 0x00000000, 0x93620005, 0x3c048000, 0x3c030800, + 0x304200fe, 0xa3620005, 0x8c620020, 0x34840001, 0x02042025, 0xaf440020, + 0x1040002d, 0x8fbf0014, 0x0a000894, 0x00000000, 0x00002821, 0x00003021, + 0x0e000f78, 0x240706a4, 0x3c020800, 0x8c430020, 0x10600023, 0x8fbf0014, + 0x0e001006, 0x00000000, 0x8f820018, 0xac500000, 0x93630082, 0x9362003f, + 0x8f840018, 0x00031a00, 0x00431025, 0xac820004, 0x8f830018, 0xac600008, + 0x8f820018, 0xac40000c, 0x8f830018, 0xac600010, 0x8f820018, 0xac400014, + 0x8f850018, 0x3c026000, 0x8c434448, 0x24040001, 0x3c020800, 0xaca30018, + 0x9443466e, 0x8f850018, 0x3c02400a, 0x00621825, 0x0e001044, 0xaca3001c, + 0x0a0008b6, 0x8fbf0014, 0x0000000d, 0x8fbf0014, 0x8fb00010, 0x03e00008, + 0x27bd0018, 0x27bdffe8, 0xafbf0010, 0x93420148, 0x2444ffff, 0x2c830005, + 0x10600047, 0x3c020800, 0x24424598, 0x00041880, 0x00621821, 0x8c640000, + 0x00800008, 0x00000000, 0x8f430144, 0x8f62000c, 0x14620006, 0x24020001, + 0xaf62000c, 0x0e000909, 0x00000000, 0x0a000907, 0x8fbf0010, 0x8f62000c, + 0x0a000900, 0x00000000, 0x97630010, 0x8f420144, 0x14430006, 0x24020001, + 0xa7620010, 0x0e000eeb, 0x00000000, 0x0a000907, 0x8fbf0010, 0x97620010, + 0x0a000900, 0x00000000, 0x97630012, 0x8f420144, 0x14430006, 0x24020001, + 0xa7620012, 0x0e000f06, 0x00000000, 0x0a000907, 0x8fbf0010, 0x97620012, + 0x0a000900, 0x00000000, 0x97630014, 0x8f420144, 0x14430006, 0x24020001, + 0xa7620014, 0x0e000f21, 0x00000000, 0x0a000907, 0x8fbf0010, 0x97620014, + 0x0a000900, 0x00000000, 0x97630016, 0x8f420144, 0x14430006, 0x24020001, + 0xa7620016, 0x0e000f24, 0x00000000, 0x0a000907, 0x8fbf0010, 0x97620016, + 0x14400006, 0x8fbf0010, 0x3c030800, 0x8c620070, 0x24420001, 0xac620070, + 0x8fbf0010, 0x03e00008, 0x27bd0018, 0x27bdffe8, 0xafbf0010, 0x93620081, + 0x3c030800, 0x8c640048, 0x0044102b, 0x14400028, 0x3c029000, 0x8f460140, + 0x34420001, 0x3c038000, 0x00c21025, 0xaf420020, 0x8f420020, 0x00431024, + 0x1440fffd, 0x3c048000, 0x34840001, 0x3c059000, 0x34a50001, 0x3c078000, + 0x24020012, 0x24030080, 0x00c42025, 0x00c52825, 0xa362003f, 0xa3630082, + 0xaf440020, 0xaf450020, 0x8f420020, 0x00471024, 0x1440fffd, 0x3c038000, + 0x9362007d, 0x34630001, 0x3c048000, 0x00c31825, 0x34420020, 0xa362007d, + 0xaf430020, 0x8f4201f8, 0x00441024, 0x1440fffd, 0x24020002, 0x3c031000, + 0x0a00096d, 0xaf4601c0, 0x93620081, 0x24420001, 0x0e000f2a, 0xa3620081, + 0x9763006a, 0x00032880, 0x14a00002, 0x00403821, 0x24050001, 0x97630068, + 0x93640081, 0x3c020800, 0x8c46004c, 0x00652821, 0x00852804, 0x00c5102b, + 0x54400001, 0x00a03021, 0x3c020800, 0x8c440050, 0x00c4182b, 0x54600001, + 0x00c02021, 0x8f420074, 0x2403fffe, 0x00832824, 0x00a21021, 0xaf62000c, + 0x10e0001a, 0x3c029000, 0x8f440140, 0x34420001, 0x3c038000, 0x00821025, + 0xaf420020, 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x3c038000, + 0x9362007d, 0x34630001, 0x3c058000, 0x00831825, 0x34420004, 0xa362007d, + 0xaf430020, 0x8f4201f8, 0x00451024, 0x1440fffd, 0x24020002, 0x3c031000, + 0xaf4401c0, 0xa34201c4, 0xaf4301f8, 0x8fbf0010, 0x03e00008, 0x27bd0018, + 0x27bdffd8, 0xafb3001c, 0x27530100, 0xafbf0024, 0xafb40020, 0xafb20018, + 0xafb10014, 0xafb00010, 0x96620008, 0x3c140800, 0x8f520100, 0x30420001, + 0x104000cf, 0x00000000, 0x8e700018, 0x8f630054, 0x2602ffff, 0x00431023, + 0x18400006, 0x00000000, 0x0000000d, 0x00000000, 0x24000128, 0x0a0009b6, + 0x00008821, 0x8f62004c, 0x02021023, 0x18400028, 0x00008821, 0x93650120, + 0x93640121, 0x3c030800, 0x8c62008c, 0x308400ff, 0x24420001, 0x30a500ff, + 0x00803821, 0x1485000b, 0xac62008c, 0x3c040800, 0x8c830090, 0x24630001, + 0xac830090, 0x93620122, 0x30420001, 0x00021023, 0x30420005, 0x0a0009b6, + 0x34510004, 0x27660100, 0x00041080, 0x00c21021, 0x8c430000, 0x02031823, + 0x04600004, 0x24820001, 0x30440007, 0x1485fff9, 0x00041080, 0x10870007, + 0x3c030800, 0xa3640121, 0x8c620094, 0x24110005, 0x24420001, 0x0a0009b6, + 0xac620094, 0x24110004, 0x32220001, 0x1040001e, 0x8e820020, 0x1040001d, + 0x32220004, 0x0e001006, 0x00000000, 0x8f820018, 0xac520000, 0x8f840018, + 0x24020001, 0xac820004, 0x8f830018, 0xac600008, 0x8f820018, 0xac40000c, + 0x8f830018, 0xac600010, 0x8f820018, 0xac500014, 0x8f850018, 0x3c026000, + 0x8c434448, 0x24040001, 0x3c020800, 0xaca30018, 0x9443466e, 0x8f850018, + 0x3c024010, 0x00621825, 0x0e001044, 0xaca3001c, 0x32220004, 0x10400076, + 0x00003821, 0x3c029000, 0x34420001, 0x3c038000, 0x02421025, 0xa360007c, + 0xaf420020, 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x93620023, + 0x30420080, 0x10400011, 0x00000000, 0x8f65005c, 0x8f63004c, 0x9764003c, + 0x8f620064, 0x00a32823, 0x00852821, 0x00a2102b, 0x54400006, 0x3c023fff, + 0x93620023, 0x3042007f, 0xa3620023, 0xaf700064, 0x3c023fff, 0x0a000a03, + 0x3442ffff, 0x8f62005c, 0x02021023, 0x04400011, 0x00000000, 0x8f65005c, + 0x8f630064, 0x9764003c, 0x3c023fff, 0x3442ffff, 0xaf700064, 0x00a32823, + 0x00852821, 0x0045102b, 0x10400004, 0x02051021, 0x3c053fff, 0x34a5ffff, + 0x02051021, 0xaf62005c, 0x24070001, 0xaf70004c, 0x8f620054, 0x16020005, + 0x00000000, 0x93620023, 0x30420040, 0x10400017, 0x24020001, 0x9762006a, + 0x00022880, 0x50a00001, 0x24050001, 0x97630068, 0x93640081, 0x3c020800, + 0x8c46004c, 0x00652821, 0x00852804, 0x00c5102b, 0x54400001, 0x00a03021, + 0x3c020800, 0x8c440050, 0x00c4182b, 0x54600001, 0x00c02021, 0x8f420074, + 0x2403fffe, 0x00832824, 0x00a21021, 0xaf62000c, 0x3c028000, 0x34420001, + 0x02421025, 0xa3600081, 0xaf420020, 0x9363007e, 0x9362007a, 0x10620004, + 0x00000000, 0x0e000f2a, 0x00000000, 0x00403821, 0x10e00017, 0x3c029000, + 0x34420001, 0x02421025, 0xaf420020, 0x3c038000, 0x8f420020, 0x00431024, + 0x1440fffd, 0x3c028000, 0x9363007d, 0x34420001, 0x3c048000, 0x02421025, + 0xa363007d, 0xaf420020, 0x8f4201f8, 0x00441024, 0x1440fffd, 0x24020002, + 0x3c031000, 0xaf5201c0, 0xa34201c4, 0xaf4301f8, 0x9342010b, 0x8e830020, + 0x27500100, 0x38420006, 0x10600029, 0x2c510001, 0x0e001006, 0x00000000, + 0x8f830018, 0x8e020000, 0xac620000, 0x8f840018, 0x96020008, 0xac820004, + 0x8f830018, 0x8e020014, 0xac620008, 0x8f850018, 0x3c026000, 0x8c434448, + 0xaca3000c, 0x8f840018, 0x96020012, 0xac820010, 0x8f850018, 0x8e030020, + 0xaca30014, 0x9602000c, 0x9603000e, 0x8f840018, 0x00021400, 0x00431025, + 0xac820018, 0x12200005, 0x3c020800, 0x9443466e, 0x8f840018, 0x0a000a78, + 0x3c024013, 0x9443466e, 0x8f840018, 0x3c024014, 0x00621825, 0xac83001c, + 0x0e001044, 0x24040001, 0x8e630014, 0x8f620040, 0x14430003, 0x00431023, + 0x0a000a83, 0x00001021, 0x28420001, 0x10400034, 0x00000000, 0x8f620040, + 0xaf630040, 0x9362003e, 0x30420001, 0x1440000b, 0x3c029000, 0x93620022, + 0x24420001, 0xa3620022, 0x93630022, 0x3c020800, 0x8c440098, 0x0064182b, + 0x1460001e, 0x3c020800, 0x3c029000, 0x34420001, 0x02421025, 0xaf420020, + 0x3c038000, 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x3c038000, + 0x9362007d, 0x34630001, 0x3c048000, 0x02431825, 0x34420001, 0xa362007d, + 0xaf430020, 0x8f4201f8, 0x00441024, 0x1440fffd, 0x24020002, 0x3c031000, + 0xaf5201c0, 0xa34201c4, 0x24020001, 0xaf4301f8, 0xa7620012, 0x0a000ab6, + 0xa3600022, 0x9743007a, 0x9444002a, 0x00641821, 0x3063fffe, 0xa7630012, + 0x0e000b68, 0x00000000, 0x97420108, 0x8fbf0024, 0x8fb40020, 0x8fb3001c, + 0x8fb20018, 0x8fb10014, 0x8fb00010, 0x00021042, 0x30420001, 0x03e00008, + 0x27bd0028, 0x27bdffe0, 0xafb20018, 0x3c120800, 0x8e420020, 0xafb00010, + 0x27500100, 0xafbf001c, 0x10400046, 0xafb10014, 0x0e001006, 0x00000000, + 0x8f840018, 0x8e020000, 0xac820000, 0x936300b1, 0x936200c5, 0x8f850018, + 0x00031e00, 0x00021400, 0x34420100, 0x00621825, 0xaca30004, 0x8f840018, + 0x8e02001c, 0xac820008, 0x8f830018, 0x8f620048, 0xac62000c, 0x8f840018, + 0x96020012, 0xac820010, 0x8f830018, 0x8f620040, 0x24040001, 0xac620014, + 0x8f850018, 0x3c026000, 0x8c434448, 0x3c020800, 0x24514660, 0xaca30018, + 0x9623000e, 0x8f850018, 0x3c024016, 0x00621825, 0x0e001044, 0xaca3001c, + 0x96030008, 0x30630010, 0x1060001c, 0x8e420020, 0x1040001a, 0x8e100000, + 0x0e001006, 0x00000000, 0x8f820018, 0xac500000, 0x8f830018, 0xac600004, + 0x8f820018, 0xac400008, 0x8f830018, 0xac60000c, 0x8f820018, 0xac400010, + 0x8f830018, 0xac600014, 0x8f850018, 0x3c036000, 0x8c634448, 0x24040001, + 0xaca30018, 0x9622000e, 0x8f850018, 0x3c034015, 0x00431025, 0x0e001044, + 0xaca2001c, 0x00001021, 0x8fbf001c, 0x8fb20018, 0x8fb10014, 0x8fb00010, + 0x03e00008, 0x27bd0020, 0x27bdffe0, 0xafb20018, 0x3c120800, 0x8e420020, + 0xafb00010, 0x27500100, 0xafbf001c, 0x10400041, 0xafb10014, 0x0e001006, + 0x00000000, 0x8f830018, 0x8e020000, 0xac620000, 0x8f840018, 0x24020100, + 0xac820004, 0x8f830018, 0x8e02001c, 0xac620008, 0x8f840018, 0x8e020018, + 0xac82000c, 0x8f830018, 0x96020012, 0xac620010, 0x8f840018, 0x96020008, + 0xac820014, 0x8f850018, 0x3c026000, 0x8c434448, 0x24040001, 0x3c020800, + 0x24514660, 0xaca30018, 0x9623000e, 0x8f850018, 0x3c024017, 0x00621825, + 0x0e001044, 0xaca3001c, 0x96030008, 0x30630010, 0x1060001c, 0x8e420020, + 0x1040001a, 0x8e100000, 0x0e001006, 0x00000000, 0x8f820018, 0xac500000, + 0x8f830018, 0xac600004, 0x8f820018, 0xac400008, 0x8f830018, 0xac60000c, + 0x8f820018, 0xac400010, 0x8f830018, 0xac600014, 0x8f850018, 0x3c036000, + 0x8c634448, 0x24040001, 0xaca30018, 0x9622000e, 0x8f850018, 0x3c034015, + 0x00431025, 0x0e001044, 0xaca2001c, 0x00001021, 0x8fbf001c, 0x8fb20018, + 0x8fb10014, 0x8fb00010, 0x03e00008, 0x27bd0020, 0x27bdffe8, 0xafbf0010, + 0x936200c4, 0x30420002, 0x10400019, 0x00000000, 0x936200c5, 0x936300b1, + 0x00431023, 0x304400ff, 0x30830080, 0x10600004, 0x00000000, 0x0000000d, + 0x00000000, 0x24000a6a, 0x93620004, 0x00441023, 0x304400ff, 0x30830080, + 0x10600004, 0x2482ffff, 0x8f650024, 0x0a000b82, 0x00000000, 0x00022b00, + 0x8f620024, 0x0045102b, 0x10400002, 0x00000000, 0x8f650024, 0x8f620048, + 0x8f630040, 0x00431823, 0x0065202b, 0x10800004, 0x00000000, 0x8f620040, + 0x00451021, 0xaf620048, 0x9762003c, 0x0062102b, 0x10400041, 0x8fbf0010, + 0x10a0003f, 0x3c029000, 0x34420001, 0x3c040800, 0x8c830080, 0x8f450100, + 0x3c068000, 0x24630001, 0x00a21025, 0xac830080, 0xaf420020, 0x8f420020, + 0x00461024, 0x1440fffd, 0x3c038000, 0x9362007d, 0x34630001, 0x3c048000, + 0x00a31825, 0x34420004, 0xa362007d, 0xaf430020, 0x8f4201f8, 0x00441024, + 0x1440fffd, 0x24020002, 0x3c030800, 0xaf4501c0, 0xa34201c4, 0x8c640020, + 0x3c021000, 0xaf4201f8, 0x1080001f, 0x8fbf0010, 0x0e001006, 0x00000000, + 0x8f830018, 0x8f420100, 0xac620000, 0x8f840018, 0x8f620040, 0xac820004, + 0x8f850018, 0x8f620048, 0xaca20008, 0x8f830018, 0xac60000c, 0x8f820018, + 0xac400010, 0x8f830018, 0x3c026000, 0xac600014, 0x8f840018, 0x8c434448, + 0x3c020800, 0xac830018, 0x9443466e, 0x8f840018, 0x3c0240c2, 0x00621825, + 0xac83001c, 0x0e001044, 0x24040001, 0x8fbf0010, 0x03e00008, 0x27bd0018, + 0x3c020800, 0x24423958, 0xaf82000c, 0x03e00008, 0x00000000, 0x27bdffe8, + 0xafb00010, 0x27500100, 0xafbf0014, 0x8e02001c, 0x14400003, 0x3c020800, + 0x0000000d, 0x3c020800, 0x8c430020, 0x10600026, 0x00001021, 0x0e001006, + 0x00000000, 0x8f830018, 0x8e020000, 0xac620000, 0x8f840018, 0x8e02001c, + 0xac820004, 0x8f830018, 0xac600008, 0x8f840018, 0x8e020018, 0xac82000c, + 0x8f850018, 0x96020012, 0xaca20010, 0x8f830018, 0x3c106000, 0xac600014, + 0x8f840018, 0x8e024448, 0x3c030800, 0xac820018, 0x9462466e, 0x8f840018, + 0x3c034012, 0x00431025, 0xac82001c, 0x0e001044, 0x24040001, 0x8e036800, + 0x00001021, 0x3c040001, 0x00641825, 0xae036800, 0x0a000c0d, 0x8fbf0014, + 0x8fbf0014, 0x8fb00010, 0x03e00008, 0x27bd0018, 0x3c020800, 0x97430078, + 0x9444002e, 0x00001021, 0x00641821, 0x3063fffe, 0x03e00008, 0xa7630010, + 0x27450100, 0x8f640048, 0x8ca30018, 0x00641023, 0x18400021, 0x00000000, + 0xaf630048, 0x8f620040, 0x9763003c, 0x00821023, 0x0043102a, 0x1040001a, + 0x3c029000, 0x8ca40000, 0x34420001, 0x3c038000, 0x00821025, 0xaf420020, + 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x3c038000, 0x9362007d, + 0x34630001, 0x3c058000, 0x00831825, 0x34420004, 0xa362007d, 0xaf430020, + 0x8f4201f8, 0x00451024, 0x1440fffd, 0x24020002, 0x3c031000, 0xaf4401c0, + 0xa34201c4, 0xaf4301f8, 0x03e00008, 0x00001021, 0x8f420100, 0x34420001, + 0xaf4200a4, 0x03e00008, 0x00001021, 0x27bdffe0, 0xafbf0018, 0xafb10014, + 0xafb00010, 0x9362007e, 0x30d000ff, 0x16020029, 0x00808821, 0x93620080, + 0x16020026, 0x00000000, 0x9362007f, 0x16020023, 0x00000000, 0x9362007a, + 0x16020004, 0x00000000, 0x0000000d, 0x00000000, 0x24000771, 0x0e000f49, + 0x00000000, 0x3c039000, 0x34630001, 0x3c048000, 0x02231825, 0xa370007a, + 0xaf430020, 0x8f420020, 0x00441024, 0x1440fffd, 0x3c028000, 0x9363007d, + 0x34420001, 0x3c048000, 0x02221025, 0xa363007d, 0xaf420020, 0x8f4201f8, + 0x00441024, 0x1440fffd, 0x24020002, 0x3c031000, 0xaf5101c0, 0xa34201c4, + 0xaf4301f8, 0x0a000c79, 0x8fbf0018, 0x0000000d, 0x00000000, 0x24000781, + 0x8fbf0018, 0x8fb10014, 0x8fb00010, 0x03e00008, 0x27bd0020, 0x3c020800, + 0x8c430020, 0x27bdffe8, 0xafb00010, 0x27500100, 0x10600024, 0xafbf0014, + 0x0e001006, 0x00000000, 0x8f830018, 0x8e020000, 0xac620000, 0x8f840018, + 0x8e020004, 0xac820004, 0x8f830018, 0x8e020018, 0xac620008, 0x8f840018, + 0x8e03001c, 0xac83000c, 0x9602000c, 0x9203000a, 0x8f840018, 0x00021400, + 0x00431025, 0xac820010, 0x8f830018, 0x3c026000, 0xac600014, 0x8f840018, + 0x8c434448, 0xac830018, 0x96020008, 0x3c030800, 0x9464466e, 0x8f850018, + 0x00021400, 0x00441025, 0x24040001, 0x0e001044, 0xaca2001c, 0x8fbf0014, + 0x8fb00010, 0x03e00008, 0x27bd0018, 0x3c020800, 0x8c430020, 0x27bdffe8, + 0xafb00010, 0x27500100, 0x10600020, 0xafbf0014, 0x0e001006, 0x00000000, + 0x8f820018, 0xac400000, 0x8f830018, 0xac600004, 0x8f820018, 0xac400008, + 0x8f830018, 0xac60000c, 0x9602000c, 0x9603000e, 0x8f840018, 0x00021400, + 0x00431025, 0xac820010, 0x8f830018, 0x3c026000, 0xac600014, 0x8f840018, + 0x8c434448, 0xac830018, 0x96020008, 0x3c030800, 0x9464466e, 0x8f850018, + 0x00021400, 0x00441025, 0x24040001, 0x0e001044, 0xaca2001c, 0x8fbf0014, + 0x8fb00010, 0x03e00008, 0x27bd0018, 0x27bdffe8, 0xafb00010, 0x27500100, + 0xafbf0014, 0x9602000c, 0x10400024, 0x00802821, 0x3c020800, 0x8c430020, + 0x1060003a, 0x8fbf0014, 0x0e001006, 0x00000000, 0x8f840018, 0x8e030000, + 0xac830000, 0x9602000c, 0x8f840018, 0x00021400, 0xac820004, 0x8f830018, + 0xac600008, 0x8f820018, 0xac40000c, 0x8f830018, 0xac600010, 0x8f820018, + 0xac400014, 0x8f850018, 0x3c026000, 0x8c434448, 0x24040001, 0x3c020800, + 0xaca30018, 0x9443466e, 0x8f850018, 0x3c02400b, 0x00621825, 0x0e001044, + 0xaca3001c, 0x0a000d19, 0x8fbf0014, 0x93620005, 0x30420010, 0x14400015, + 0x3c029000, 0x34420001, 0x00a21025, 0xaf420020, 0x3c038000, 0x8f420020, + 0x00431024, 0x1440fffd, 0x00000000, 0x3c038000, 0x93620005, 0x34630001, + 0x00a02021, 0x00a31825, 0x24055852, 0x34420010, 0xa3620005, 0x0e000553, + 0xaf430020, 0x0a000d19, 0x8fbf0014, 0x0000000d, 0x8fbf0014, 0x8fb00010, + 0x03e00008, 0x27bd0018, 0x3c020800, 0x8c430020, 0x27bdffe8, 0xafb00010, + 0x27500100, 0x10600022, 0xafbf0014, 0x0e001006, 0x00000000, 0x8f840018, + 0x8e020004, 0xac820000, 0x9603000c, 0x9762002c, 0x8f840018, 0x00031c00, + 0x00431025, 0xac820004, 0x8f830018, 0xac600008, 0x8f820018, 0xac40000c, + 0x8f830018, 0xac600010, 0x8f820018, 0xac400014, 0x8f850018, 0x3c026000, + 0x8c434448, 0x24040001, 0x3c020800, 0xaca30018, 0x9443466e, 0x8f850018, + 0x3c02400e, 0x00621825, 0x0e001044, 0xaca3001c, 0x0e000d48, 0x8e040000, + 0x8fbf0014, 0x8fb00010, 0x03e00008, 0x27bd0018, 0x3c038000, 0x8f420278, + 0x00431024, 0x1440fffd, 0x24020002, 0x3c031000, 0xaf440240, 0xa3420244, + 0x03e00008, 0xaf430278, 0x3c020800, 0x8c430020, 0x27bdffe0, 0xafb10014, + 0x00808821, 0xafb20018, 0x00c09021, 0xafb00010, 0x30b0ffff, 0x1060001c, + 0xafbf001c, 0x0e001006, 0x00000000, 0x8f820018, 0xac510000, 0x8f840018, + 0x00101400, 0xac820004, 0x8f830018, 0xac600008, 0x8f820018, 0xac40000c, + 0x8f830018, 0xac600010, 0x8f820018, 0xac520014, 0x8f840018, 0x3c026000, + 0x8c434448, 0x3c020800, 0xac830018, 0x9443466e, 0x8f840018, 0x3c024019, + 0x00621825, 0xac83001c, 0x0e001044, 0x24040001, 0x8fbf001c, 0x8fb20018, + 0x8fb10014, 0x8fb00010, 0x03e00008, 0x27bd0020, 0x27bdffe8, 0x27450100, + 0xafbf0010, 0x94a3000c, 0x240200c1, 0x14620029, 0x00803021, 0x3c029000, + 0x34420001, 0x00c21025, 0xaf420020, 0x3c038000, 0x8f420020, 0x00431024, + 0x1440fffd, 0x3c028000, 0x34420001, 0x3c049000, 0x34840001, 0x3c058000, + 0x24030012, 0x00c21025, 0x00c42025, 0xa363003f, 0xaf420020, 0xaf440020, + 0x8f420020, 0x00451024, 0x1440fffd, 0x3c038000, 0x9362007d, 0x34630001, + 0x3c048000, 0x00c31825, 0x34420020, 0xa362007d, 0xaf430020, 0x8f4201f8, + 0x00441024, 0x1440fffd, 0x24020002, 0x3c031000, 0xaf4601c0, 0xa34201c4, + 0xaf4301f8, 0x0a000db3, 0x8fbf0010, 0x00c02021, 0x94a5000c, 0x24060001, + 0x0e000f78, 0x240706d8, 0x8fbf0010, 0x03e00008, 0x27bd0018, 0x3c020800, + 0x8c430020, 0x27bdffe0, 0xafb00010, 0x00808021, 0xafb20018, 0x00a09021, + 0xafb10014, 0x30d100ff, 0x1060001c, 0xafbf001c, 0x0e001006, 0x00000000, + 0x8f820018, 0xac500000, 0x8f840018, 0x24020001, 0xac820004, 0x8f830018, + 0xac600008, 0x8f820018, 0xac40000c, 0x8f830018, 0xac600010, 0x8f820018, + 0xac520014, 0x8f840018, 0x3c026000, 0x8c434448, 0x3c020800, 0xac830018, + 0x9443466e, 0x8f840018, 0x3c024010, 0x00621825, 0xac83001c, 0x0e001044, + 0x02202021, 0x8fbf001c, 0x8fb20018, 0x8fb10014, 0x8fb00010, 0x03e00008, + 0x27bd0020, 0x27bdffe8, 0xafbf0014, 0xafb00010, 0x93620005, 0x30420001, + 0x10400033, 0x00808021, 0x3c029000, 0x34420001, 0x02021025, 0xaf420020, + 0x3c038000, 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x93620005, + 0x3c048000, 0x3c030800, 0x304200fe, 0xa3620005, 0x8c620020, 0x34840001, + 0x02042025, 0xaf440020, 0x10400020, 0x8fbf0014, 0x0e001006, 0x00000000, + 0x8f820018, 0xac500000, 0x93630082, 0x9362003f, 0x8f840018, 0x00031a00, + 0x00431025, 0xac820004, 0x8f830018, 0xac600008, 0x8f820018, 0xac40000c, + 0x8f830018, 0xac600010, 0x8f820018, 0xac400014, 0x8f840018, 0x3c026000, + 0x8c434448, 0x3c020800, 0xac830018, 0x9443466e, 0x8f840018, 0x3c02400a, + 0x00621825, 0xac83001c, 0x0e001044, 0x24040001, 0x8fbf0014, 0x8fb00010, + 0x03e00008, 0x27bd0018, 0x27bdffe8, 0xafbf0010, 0x8f420188, 0x00803021, + 0x9364003f, 0x24030012, 0x00021402, 0x1483001c, 0x304500ff, 0x3c029000, + 0x34420001, 0x3c038000, 0x00c21025, 0xa3650080, 0xa365007a, 0xaf420020, + 0x8f420020, 0x00431024, 0x1440fffd, 0x3c028000, 0x9363007d, 0x34420001, + 0x3c048000, 0x00c21025, 0xa363007d, 0xaf420020, 0x8f4201f8, 0x00441024, + 0x1440fffd, 0x24020002, 0x3c031000, 0xaf4601c0, 0xa34201c4, 0xaf4301f8, + 0x0a000e54, 0x8fbf0010, 0x9362007e, 0x1445000e, 0x00000000, 0x93620080, + 0x1045000b, 0x00000000, 0xa3650080, 0x8f820000, 0x93660080, 0x8f440180, + 0x8f65004c, 0x8c430000, 0x0060f809, 0x00000000, 0x0a000e54, 0x8fbf0010, + 0xa3650080, 0x8fbf0010, 0x03e00008, 0x27bd0018, 0x3c020800, 0x8c430020, + 0x27bdffe0, 0xafb10014, 0x00808821, 0xafb20018, 0x00a09021, 0xafb00010, + 0x30d000ff, 0x1060002f, 0xafbf001c, 0x0e001006, 0x00000000, 0x8f820018, + 0xac510000, 0x8f830018, 0xac700004, 0x8f820018, 0xac520008, 0x8f830018, + 0xac60000c, 0x8f820018, 0xac400010, 0x9763006a, 0x00032880, 0x50a00001, + 0x24050001, 0x97630068, 0x93640081, 0x3c020800, 0x8c46004c, 0x00652821, + 0x00852804, 0x00c5102b, 0x54400001, 0x00a03021, 0x3c020800, 0x8c440050, + 0x00c4182b, 0x54600001, 0x00c02021, 0x8f830018, 0x2402fffe, 0x00822824, + 0x3c026000, 0xac650014, 0x8f840018, 0x8c434448, 0x3c020800, 0xac830018, + 0x9443466e, 0x8f840018, 0x3c024011, 0x00621825, 0xac83001c, 0x0e001044, + 0x24040001, 0x8fbf001c, 0x8fb20018, 0x8fb10014, 0x8fb00010, 0x03e00008, + 0x27bd0020, 0x27bdffe8, 0xafbf0014, 0xafb00010, 0x8f440100, 0x27500100, + 0x8f650050, 0x0e000c45, 0x9206001b, 0x3c020800, 0x8c430020, 0x1060001d, + 0x8e100018, 0x0e001006, 0x00000000, 0x8f840018, 0x8f420100, 0xac820000, + 0x8f830018, 0xac700004, 0x8f840018, 0x8f620050, 0xac820008, 0x8f830018, + 0xac60000c, 0x8f820018, 0xac400010, 0x8f830018, 0x3c026000, 0xac600014, + 0x8f850018, 0x8c434448, 0x24040001, 0x3c020800, 0xaca30018, 0x9443466e, + 0x8f850018, 0x3c02401c, 0x00621825, 0x0e001044, 0xaca3001c, 0x8fbf0014, + 0x8fb00010, 0x03e00008, 0x27bd0018, 0x3c029000, 0x8f460140, 0x34420001, + 0x3c038000, 0x00c21025, 0xaf420020, 0x8f420020, 0x00431024, 0x1440fffd, + 0x3c048000, 0x34840001, 0x3c059000, 0x34a50001, 0x3c078000, 0x24020012, + 0x24030080, 0x00c42025, 0x00c52825, 0xa362003f, 0xa3630082, 0xaf440020, + 0xaf450020, 0x8f420020, 0x00471024, 0x1440fffd, 0x3c038000, 0x9362007d, + 0x34630001, 0x3c048000, 0x00c31825, 0x34420020, 0xa362007d, 0xaf430020, + 0x8f4201f8, 0x00441024, 0x1440fffd, 0x24020002, 0x3c031000, 0xaf4601c0, + 0xa34201c4, 0x03e00008, 0xaf4301f8, 0x8f430238, 0x3c020800, 0x04610013, + 0x8c44009c, 0x2406fffe, 0x3c050800, 0x3c038000, 0x2484ffff, 0x14800009, + 0x00000000, 0x97420078, 0x8ca3007c, 0x24420001, 0x00461024, 0x24630001, + 0xa7620010, 0x03e00008, 0xaca3007c, 0x8f420238, 0x00431024, 0x1440fff3, + 0x2484ffff, 0x8f420140, 0x3c031000, 0xaf420200, 0x03e00008, 0xaf430238, + 0x3c029000, 0x8f440140, 0x34420001, 0x3c038000, 0x00821025, 0xaf420020, + 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x3c038000, 0x9362007d, + 0x34630001, 0x3c058000, 0x00831825, 0x34420001, 0xa362007d, 0xaf430020, + 0x8f4201f8, 0x00451024, 0x1440fffd, 0x24020002, 0x3c031000, 0xaf4401c0, + 0xa34201c4, 0x03e00008, 0xaf4301f8, 0x0000000d, 0x03e00008, 0x00000000, + 0x0000000d, 0x03e00008, 0x00000000, 0x24020001, 0x03e00008, 0xa7620010, + 0x9362003f, 0x304400ff, 0x3883000e, 0x2c630001, 0x38820010, 0x2c420001, + 0x00621825, 0x14600003, 0x24020012, 0x14820003, 0x00000000, 0x03e00008, + 0x00001021, 0x9363007e, 0x9362007a, 0x14620006, 0x00000000, 0x9363007e, + 0x24020001, 0x24630001, 0x03e00008, 0xa363007e, 0x9363007e, 0x93620080, + 0x14620004, 0x24020001, 0xa362000b, 0x03e00008, 0x24020001, 0x03e00008, + 0x00001021, 0x9362000b, 0x10400021, 0x00001021, 0xa360000b, 0x9362003f, + 0x304400ff, 0x3883000e, 0x2c630001, 0x38820010, 0x2c420001, 0x00621825, + 0x14600015, 0x00001821, 0x24020012, 0x10820012, 0x00000000, 0x9363007e, + 0x9362007a, 0x14620007, 0x00000000, 0x9362007e, 0x24030001, 0x24420001, + 0xa362007e, 0x03e00008, 0x00601021, 0x9363007e, 0x93620080, 0x14620004, + 0x00001821, 0x24020001, 0xa362000b, 0x24030001, 0x03e00008, 0x00601021, + 0x03e00008, 0x00000000, 0x24040001, 0xaf64000c, 0x8f6300dc, 0x8f6200cc, + 0x50620001, 0xa7640010, 0xa7640012, 0xa7640014, 0x03e00008, 0xa7640016, + 0x27bdffd8, 0xafb00010, 0x00808021, 0xafb3001c, 0x00c09821, 0xafbf0020, + 0xafb20018, 0xafb10014, 0x93620023, 0x00e09021, 0x30420040, 0x10400020, + 0x30b1ffff, 0x3c020800, 0x8c430020, 0x1060001c, 0x00000000, 0x0e001006, + 0x00000000, 0x8f820018, 0xac500000, 0x8f840018, 0x3c02008d, 0xac820004, + 0x8f830018, 0xac600008, 0x8f820018, 0xac40000c, 0x8f830018, 0xac600010, + 0x8f820018, 0xac520014, 0x8f850018, 0x3c026000, 0x8c434448, 0x24040001, + 0x3c020800, 0xaca30018, 0x9443466e, 0x8f850018, 0x3c024019, 0x00621825, + 0x0e001044, 0xaca3001c, 0x93620023, 0x30420020, 0x14400003, 0x3c020800, + 0x52600020, 0x3c029000, 0x8c430020, 0x1060001d, 0x3c029000, 0x0e001006, + 0x00000000, 0x8f820018, 0xac500000, 0x8f840018, 0x00111400, 0xac820004, + 0x8f830018, 0xac720008, 0x8f820018, 0xac40000c, 0x8f830018, 0xac600010, + 0x8f820018, 0xac400014, 0x8f850018, 0x3c026000, 0x8c434448, 0x24040001, + 0x3c020800, 0xaca30018, 0x9443466e, 0x8f850018, 0x3c02401b, 0x00621825, + 0x0e001044, 0xaca3001c, 0x3c029000, 0x34420001, 0x02021025, 0xaf420020, + 0x3c038000, 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x93630023, + 0x3c028000, 0x34420001, 0x02021025, 0x8fbf0020, 0x8fb3001c, 0x8fb20018, + 0x8fb10014, 0x8fb00010, 0x3063009f, 0xa3630023, 0xaf420020, 0x03e00008, + 0x27bd0028, 0x3c020800, 0x8c430020, 0x27bdffe8, 0xafb00010, 0x27500100, + 0x1060001d, 0xafbf0014, 0x0e001006, 0x00000000, 0x8f830018, 0x8e020004, + 0xac620000, 0x8f840018, 0x8e020018, 0xac820004, 0x8f850018, 0x8e020000, + 0xaca20008, 0x8f830018, 0xac60000c, 0x8f820018, 0xac400010, 0x8f830018, + 0xac600014, 0x8f820018, 0xac400018, 0x96030008, 0x3c020800, 0x9444466e, + 0x8f850018, 0x00031c00, 0x00641825, 0x24040001, 0x0e001044, 0xaca3001c, + 0x8fbf0014, 0x8fb00010, 0x03e00008, 0x27bd0018, 0x3c060800, 0x24c54660, + 0x3c02000a, 0x03421821, 0x94640006, 0x94a2000a, 0x00441023, 0x00021400, + 0x00021c03, 0x04610006, 0xa4a40006, 0x0000000d, 0x00000000, 0x2400005a, + 0x0a00101b, 0x24020001, 0x8f820014, 0x0062102b, 0x14400002, 0x00001021, + 0x24020001, 0x304200ff, 0x1040001c, 0x274a0400, 0x3c07000a, 0x3c020800, + 0x24454660, 0x94a9000a, 0x8f880014, 0x03471021, 0x94430006, 0x00402021, + 0xa4a30006, 0x94820006, 0xa4a20006, 0x01221023, 0x00021400, 0x00021403, + 0x04410006, 0x0048102b, 0x0000000d, 0x00000000, 0x2400005a, 0x0a001036, + 0x24020001, 0x14400002, 0x00001021, 0x24020001, 0x304200ff, 0x1440ffec, + 0x03471021, 0x24c44660, 0x8c820010, 0xaf420038, 0x8c830014, 0x3c020005, + 0xaf43003c, 0xaf420030, 0xaf800010, 0xaf8a0018, 0x03e00008, 0x00000000, + 0x27bdffe0, 0x8f820010, 0x8f850018, 0x3c070800, 0x24e84660, 0xafbf001c, + 0xafb20018, 0xafb10014, 0xafb00010, 0x9503000a, 0x8d060014, 0x00009021, + 0x309000ff, 0x00e08821, 0x24420001, 0x24a50020, 0x24630001, 0xaf820010, + 0xaf850018, 0xa503000a, 0x24c30020, 0x3c028000, 0x04c10007, 0xad030014, + 0x00621024, 0x14400005, 0x26224660, 0x8d020010, 0x24420001, 0xad020010, + 0x26224660, 0x9444000a, 0x94450018, 0x0010102b, 0x00a41826, 0x2c630001, + 0x00621825, 0x1060001c, 0x3c030006, 0x8f820010, 0x24120001, 0x00021140, + 0x00431025, 0xaf420030, 0x00000000, 0x00000000, 0x00000000, 0x27450400, + 0x8f420000, 0x30420010, 0x1040fffd, 0x26224660, 0x9444000a, 0x94430018, + 0xaf800010, 0xaf850018, 0x14830012, 0x26274660, 0x0e0010d2, 0x00000000, + 0x1600000e, 0x26274660, 0x0e001006, 0x00000000, 0x0a00108f, 0x26274660, + 0x00041c00, 0x00031c03, 0x00051400, 0x00021403, 0x00621823, 0x18600002, + 0x3c026000, 0xac400808, 0x26274660, 0x94e2000e, 0x94e3000c, 0x24420001, + 0xa4e2000e, 0x3042ffff, 0x50430001, 0xa4e0000e, 0x12000005, 0x3c02000a, + 0x94e2000a, 0xa74200a2, 0x0a0010cc, 0x02401021, 0x03421821, 0x94640006, + 0x94e2000a, 0x00441023, 0x00021400, 0x00021c03, 0x04610006, 0xa4e40006, + 0x0000000d, 0x00000000, 0x2400005a, 0x0a0010ae, 0x24020001, 0x8f820014, + 0x0062102b, 0x14400002, 0x00001021, 0x24020001, 0x304200ff, 0x1040001b, + 0x3c020800, 0x3c06000a, 0x24454660, 0x94a8000a, 0x8f870014, 0x03461021, + 0x94430006, 0x00402021, 0xa4a30006, 0x94820006, 0xa4a20006, 0x01021023, + 0x00021400, 0x00021403, 0x04410006, 0x0047102b, 0x0000000d, 0x00000000, + 0x2400005a, 0x0a0010c8, 0x24020001, 0x14400002, 0x00001021, 0x24020001, + 0x304200ff, 0x1440ffec, 0x03461021, 0x02401021, 0x8fbf001c, 0x8fb20018, + 0x8fb10014, 0x8fb00010, 0x03e00008, 0x27bd0020, 0x3c020800, 0x24454660, + 0x94a3001a, 0x8ca40024, 0x00403021, 0x000318c0, 0x00832021, 0xaf44003c, + 0x8ca20020, 0xaf420038, 0x3c020050, 0x34420008, 0xaf420030, 0x00000000, + 0x00000000, 0x00000000, 0x8f420000, 0x30420020, 0x1040fffd, 0x00000000, + 0x8f430400, 0x24c64660, 0xacc30010, 0x8f420404, 0x3c030020, 0xacc20014, + 0xaf430030, 0x94c40018, 0x94c3001c, 0x94c2001a, 0x94c5001e, 0x00832021, + 0x24420001, 0xa4c2001a, 0x3042ffff, 0x14450002, 0xa4c40018, 0xa4c0001a, + 0x03e00008, 0x00000000, 0x8f820010, 0x3c030006, 0x00021140, 0x00431025, + 0xaf420030, 0x00000000, 0x00000000, 0x00000000, 0x27430400, 0x8f420000, + 0x30420010, 0x1040fffd, 0x00000000, 0xaf800010, 0xaf830018, 0x03e00008, + 0x00000000, 0x27bdffe8, 0xafb00010, 0x3c100800, 0x26104660, 0x3c05000a, + 0x02002021, 0x03452821, 0xafbf0014, 0x0e001128, 0x2406000a, 0x96020002, + 0x9603001e, 0x3042000f, 0x24420003, 0x00431804, 0x24027fff, 0x0043102b, + 0xaf830014, 0x10400004, 0x00000000, 0x0000000d, 0x00000000, 0x24000043, + 0x0e0010d2, 0x00000000, 0x8fbf0014, 0x8fb00010, 0x03e00008, 0x27bd0018, + 0x10c00007, 0x00000000, 0x8ca20000, 0x24c6ffff, 0x24a50004, 0xac820000, + 0x14c0fffb, 0x24840004, 0x03e00008, 0x00000000, 0x0a001137, 0x00a01021, + 0xac860000, 0x24840004, 0x00a01021, 0x1440fffc, 0x24a5ffff, 0x03e00008, + 0x00000000, 0x3c036000, 0x8c642b7c, 0x3c036010, 0x8c6553fc, 0x00041582, + 0x00042302, 0x308403ff, 0x00052d82, 0x00441026, 0x0002102b, 0x0005282b, + 0x00451025, 0x1440000d, 0x3c020050, 0x34420004, 0xaf400038, 0xaf40003c, + 0xaf420030, 0x00000000, 0x00000000, 0x8f420000, 0x30420020, 0x1040fffd, + 0x3c020020, 0xaf420030, 0x0000000d, 0x03e00008, 0x00000000, 0x3c020050, + 0x34420004, 0xaf440038, 0xaf45003c, 0xaf420030, 0x00000000, 0x00000000, + 0x8f420000, 0x30420020, 0x1040fffd, 0x3c020020, 0xaf420030, 0x03e00008, + 0x00000000, 0x00000000 }; + +static u32 bnx2_COM_b06FwData[(0x0/4) + 1] = { 0x00000000 }; +static u32 bnx2_COM_b06FwRodata[(0x18/4) + 1] = { + 0x08002318, 0x08002348, 0x08002378, 0x080023a8, 0x080023d8, 0x00000000, + 0x00000000 }; + +static u32 bnx2_COM_b06FwBss[(0x88/4) + 1] = { 0x00000000 }; +static u32 bnx2_COM_b06FwSbss[(0x1c/4) + 1] = { 0x00000000 }; + +static int bnx2_RXP_b06FwReleaseMajor = 0x0; +static int bnx2_RXP_b06FwReleaseMinor = 0x0; +static int bnx2_RXP_b06FwReleaseFix = 0x0; +static u32 bnx2_RXP_b06FwStartAddr = 0x08000060; +static u32 bnx2_RXP_b06FwTextAddr = 0x08000000; +static int bnx2_RXP_b06FwTextLen = 0x20b8; +static u32 bnx2_RXP_b06FwDataAddr = 0x080020e0; +static int bnx2_RXP_b06FwDataLen = 0x0; +static u32 bnx2_RXP_b06FwRodataAddr = 0x00000000; +static int bnx2_RXP_b06FwRodataLen = 0x0; +static u32 bnx2_RXP_b06FwBssAddr = 0x08002100; +static int bnx2_RXP_b06FwBssLen = 0x239c; +static u32 bnx2_RXP_b06FwSbssAddr = 0x080020e0; +static int bnx2_RXP_b06FwSbssLen = 0x14; + +static u32 bnx2_RXP_b06FwText[(0x20b8/4) + 1] = { + 0x0a000018, 0x00000000, 0x00000000, 0x0000000d, 0x72787020, 0x302e362e, + 0x39000000, 0x00060903, 0x00000000, 0x0000000d, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x10000003, 0x00000000, 0x0000000d, 0x0000000d, 0x3c020800, + 0x244220e0, 0x3c030800, 0x2463449c, 0xac400000, 0x0043202b, 0x1480fffd, + 0x24420004, 0x3c1d0800, 0x37bd3ffc, 0x03a0f021, 0x3c100800, 0x26100060, + 0x3c1c0800, 0x279c20e0, 0x0e000329, 0x00000000, 0x0000000d, 0x8f870008, + 0x2ce20080, 0x10400018, 0x3c030800, 0x24633490, 0x8f460100, 0x00072140, + 0x00831021, 0xac460000, 0x8f450104, 0x00641021, 0xac450004, 0x8f460108, + 0xac460008, 0x8f45010c, 0xac45000c, 0x8f460114, 0xac460010, 0x8f450118, + 0xac450014, 0x8f460124, 0xac460018, 0x8f450128, 0x00641821, 0x24e20001, + 0xaf820008, 0xac65001c, 0x03e00008, 0x00000000, 0x00804021, 0x8f830000, + 0x24070001, 0x3c020001, 0x00621024, 0x10400037, 0x00603021, 0x9742010e, + 0x3c038000, 0x3045ffff, 0x8f4201b8, 0x00431024, 0x1440fffd, 0x24020003, + 0xa342018b, 0x8f840004, 0x24020080, 0x24030002, 0xaf420180, 0xa743018c, + 0x10800005, 0xa745018e, 0x9743011c, 0x9742011e, 0x0a000069, 0x00021400, + 0x9743011e, 0x9742011c, 0x00021400, 0x00621825, 0xaf4301a8, 0x8f84000c, + 0x24020003, 0x30838000, 0x1060000d, 0xa7420188, 0x93420116, 0x304200fc, + 0x005a1021, 0x24424004, 0x8c430000, 0x3063ffff, 0x14600005, 0x00000000, + 0x3c02ffff, 0x34427fff, 0x00821024, 0xaf82000c, 0x9782000e, 0x9743010c, + 0x8f440104, 0x3042bfff, 0x00031c00, 0x3084ffff, 0x00641825, 0xa74201a6, + 0xaf4301ac, 0x3c021000, 0xaf4201b8, 0x03e00008, 0x00001021, 0x30c21000, + 0x1040000f, 0x00000000, 0x9742010c, 0x3042fc00, 0x5440000b, 0x24070005, + 0x3c021000, 0x00c21024, 0x10400007, 0x3c030dff, 0x3463ffff, 0x3c020e00, + 0x00c21024, 0x0062182b, 0x54600001, 0x24070005, 0x8f82000c, 0x30434000, + 0x10600016, 0x00404821, 0x3c020f00, 0x00c21024, 0x14400012, 0x00000000, + 0x93420116, 0x34424000, 0x03421821, 0x94650002, 0x2ca21389, 0x1040000b, + 0x3c020800, 0x24422100, 0x00051942, 0x00031880, 0x00621821, 0x30a5001f, + 0x8c640000, 0x24020001, 0x00a21004, 0x00822024, 0x01044025, 0x11000037, + 0x3c021000, 0x9742010e, 0x34e60002, 0x3c038000, 0x24420004, 0x3045ffff, + 0x8f4201b8, 0x00431024, 0x1440fffd, 0x24020003, 0xa342018b, 0x8f840004, + 0x24020180, 0x24030002, 0xaf420180, 0xa743018c, 0x10800005, 0xa745018e, + 0x9743011c, 0x9742011e, 0x0a0000cd, 0x00021400, 0x9743011e, 0x9742011c, + 0x00021400, 0x00621825, 0xaf4301a8, 0x8f84000c, 0x30828000, 0x1040000c, + 0xa7460188, 0x93420116, 0x304200fc, 0x005a1021, 0x24424004, 0x8c430000, + 0x3063ffff, 0x14600004, 0x3c02ffff, 0x34427fff, 0x00821024, 0xaf82000c, + 0x9782000e, 0x9743010c, 0x8f440104, 0x3042bfff, 0x00031c00, 0x3084ffff, + 0x00641825, 0xa74201a6, 0xaf4301ac, 0x3c021000, 0xaf4201b8, 0x03e00008, + 0x00001021, 0x00c21024, 0x104000ba, 0x3c020800, 0x8c430030, 0x1060003e, + 0x31224000, 0x1040003c, 0x3c030f00, 0x00c31824, 0x3c020100, 0x0043102b, + 0x14400038, 0x3c030800, 0x9742010e, 0x34e60002, 0x3c038000, 0x24420004, + 0x3045ffff, 0x8f4201b8, 0x00431024, 0x1440fffd, 0x24020003, 0xa342018b, + 0x8f840004, 0x24020080, 0x24030002, 0xaf420180, 0xa743018c, 0x10800005, + 0xa745018e, 0x9743011c, 0x9742011e, 0x0a000110, 0x00021400, 0x9743011e, + 0x9742011c, 0x00021400, 0x00621825, 0xaf4301a8, 0x8f84000c, 0x30828000, + 0x1040000c, 0xa7460188, 0x93420116, 0x304200fc, 0x005a1021, 0x24424004, + 0x8c430000, 0x3063ffff, 0x14600004, 0x3c02ffff, 0x34427fff, 0x00821024, + 0xaf82000c, 0x9782000e, 0x9743010c, 0x8f440104, 0x3042bfff, 0x00031c00, + 0x3084ffff, 0x00641825, 0xa74201a6, 0xaf4301ac, 0x3c021000, 0xaf4201b8, + 0x03e00008, 0x00001021, 0x3c030800, 0x8c620024, 0x30420008, 0x1040003d, + 0x34e80002, 0x3c020f00, 0x00c21024, 0x5440003a, 0x3107ffff, 0x9742010c, + 0x30420200, 0x50400036, 0x3107ffff, 0x9742010e, 0x30e6fffb, 0x3c038000, + 0x24420004, 0x3045ffff, 0x8f4201b8, 0x00431024, 0x1440fffd, 0x24020003, + 0xa342018b, 0x8f840004, 0x24020180, 0x24030002, 0xaf420180, 0xa743018c, + 0x10800005, 0xa745018e, 0x9743011c, 0x9742011e, 0x0a000153, 0x00021400, + 0x9743011e, 0x9742011c, 0x00021400, 0x00621825, 0xaf4301a8, 0x8f84000c, + 0x30828000, 0x1040000c, 0xa7460188, 0x93420116, 0x304200fc, 0x005a1021, + 0x24424004, 0x8c430000, 0x3063ffff, 0x14600004, 0x3c02ffff, 0x34427fff, + 0x00821024, 0xaf82000c, 0x9782000e, 0x9743010c, 0x8f440104, 0x3042bfff, + 0x00031c00, 0x3084ffff, 0x00641825, 0xa74201a6, 0xaf4301ac, 0x3c021000, + 0xaf4201b8, 0x3107ffff, 0x8f820000, 0x3c068000, 0x9743010e, 0x00021442, + 0x30440780, 0x24630004, 0x3065ffff, 0x8f4201b8, 0x00461024, 0x1440fffd, + 0x24020003, 0xa342018b, 0x8f830004, 0x24020002, 0xaf440180, 0xa742018c, + 0x10600005, 0xa745018e, 0x9743011c, 0x9742011e, 0x0a000189, 0x00021400, + 0x9743011e, 0x9742011c, 0x00021400, 0x00621825, 0xaf4301a8, 0x8f84000c, + 0x30828000, 0x1040000c, 0xa7470188, 0x93420116, 0x304200fc, 0x005a1021, + 0x24424004, 0x8c430000, 0x3063ffff, 0x14600004, 0x3c02ffff, 0x34427fff, + 0x00821024, 0xaf82000c, 0x9782000e, 0x9743010c, 0x8f440104, 0x3042bfff, + 0x00031c00, 0x3084ffff, 0x00641825, 0xa74201a6, 0xaf4301ac, 0x3c021000, + 0xaf4201b8, 0x03e00008, 0x00001021, 0x8f424000, 0x30420100, 0x104000ef, + 0x3c020800, 0x8c440024, 0x24030001, 0x14830036, 0x00404021, 0x9742010e, + 0x34e50002, 0x3c038000, 0x24420004, 0x3044ffff, 0x8f4201b8, 0x00431024, + 0x1440fffd, 0x24020003, 0xa342018b, 0x8f830004, 0x24020002, 0xaf400180, + 0xa742018c, 0x10600005, 0xa744018e, 0x9743011c, 0x9742011e, 0x0a0001c6, + 0x00021400, 0x9743011e, 0x9742011c, 0x00021400, 0x00621825, 0xaf4301a8, + 0x8f84000c, 0x30828000, 0x1040000c, 0xa7450188, 0x93420116, 0x304200fc, + 0x005a1021, 0x24424004, 0x8c430000, 0x3063ffff, 0x14600004, 0x3c02ffff, + 0x34427fff, 0x00821024, 0xaf82000c, 0x9782000e, 0x9743010c, 0x8f440104, + 0x3042bfff, 0x00031c00, 0x3084ffff, 0x00641825, 0xa74201a6, 0xaf4301ac, + 0x3c021000, 0xaf4201b8, 0x03e00008, 0x00001021, 0x30820001, 0x10400035, + 0x30e90004, 0x9742010e, 0x30e6fffb, 0x3c038000, 0x24420004, 0x3044ffff, + 0x8f4201b8, 0x00431024, 0x1440fffd, 0x24020003, 0xa342018b, 0x8f830004, + 0x24020002, 0xaf400180, 0xa742018c, 0x10600005, 0xa744018e, 0x9743011c, + 0x9742011e, 0x0a0001fe, 0x00021400, 0x9743011e, 0x9742011c, 0x00021400, + 0x00621825, 0xaf4301a8, 0x8f84000c, 0x30828000, 0x1040000c, 0xa7470188, + 0x93420116, 0x304200fc, 0x005a1021, 0x24424004, 0x8c430000, 0x3063ffff, + 0x14600004, 0x3c02ffff, 0x34427fff, 0x00821024, 0xaf82000c, 0x9782000e, + 0x9743010c, 0x8f440104, 0x3042bfff, 0x00031c00, 0x3084ffff, 0x00641825, + 0xa74201a6, 0xaf4301ac, 0x3c021000, 0xaf4201b8, 0x30c7ffff, 0x8d020024, + 0x30420004, 0x10400037, 0x8d020024, 0x9742010e, 0x30e6fffb, 0x3c038000, + 0x24420004, 0x3045ffff, 0x8f4201b8, 0x00431024, 0x1440fffd, 0x24020003, + 0xa342018b, 0x8f840004, 0x24020100, 0x24030002, 0xaf420180, 0xa743018c, + 0x10800005, 0xa745018e, 0x9743011c, 0x9742011e, 0x0a000237, 0x00021400, + 0x9743011e, 0x9742011c, 0x00021400, 0x00621825, 0xaf4301a8, 0x8f84000c, + 0x30828000, 0x1040000c, 0xa7470188, 0x93420116, 0x304200fc, 0x005a1021, + 0x24424004, 0x8c430000, 0x3063ffff, 0x14600004, 0x3c02ffff, 0x34427fff, + 0x00821024, 0xaf82000c, 0x9782000e, 0x9743010c, 0x8f440104, 0x3042bfff, + 0x00031c00, 0x3084ffff, 0x00641825, 0xa74201a6, 0xaf4301ac, 0x3c021000, + 0xaf4201b8, 0x30c7ffff, 0x8d020024, 0x30420008, 0x10400034, 0x00000000, + 0x9742010e, 0x3c038000, 0x24420004, 0x3045ffff, 0x8f4201b8, 0x00431024, + 0x1440fffd, 0x24020003, 0xa342018b, 0x8f840004, 0x24020180, 0x24030002, + 0xaf420180, 0xa743018c, 0x10800005, 0xa745018e, 0x9743011c, 0x9742011e, + 0x0a00026f, 0x00021400, 0x9743011e, 0x9742011c, 0x00021400, 0x00621825, + 0xaf4301a8, 0x8f84000c, 0x30828000, 0x1040000c, 0xa7470188, 0x93420116, + 0x304200fc, 0x005a1021, 0x24424004, 0x8c430000, 0x3063ffff, 0x14600004, + 0x3c02ffff, 0x34427fff, 0x00821024, 0xaf82000c, 0x9782000e, 0x9743010c, + 0x8f440104, 0x3042bfff, 0x00031c00, 0x3084ffff, 0x00641825, 0xa74201a6, + 0xaf4301ac, 0x3c021000, 0xaf4201b8, 0x15200046, 0x00001021, 0x3c038000, + 0x8f4201b8, 0x00431024, 0x1440fffd, 0x24020002, 0x24032000, 0xa342018b, + 0xa7430188, 0x3c021000, 0xaf4201b8, 0x03e00008, 0x00001021, 0x3c030800, + 0x8c620024, 0x30420001, 0x10400035, 0x00001021, 0x9742010e, 0x34e50002, + 0x3c038000, 0x24420004, 0x3044ffff, 0x8f4201b8, 0x00431024, 0x1440fffd, + 0x24020003, 0xa342018b, 0x8f830004, 0x24020002, 0xaf400180, 0xa742018c, + 0x10600005, 0xa744018e, 0x9743011c, 0x9742011e, 0x0a0002b5, 0x00021400, + 0x9743011e, 0x9742011c, 0x00021400, 0x00621825, 0xaf4301a8, 0x8f84000c, + 0x30828000, 0x1040000c, 0xa7450188, 0x93420116, 0x304200fc, 0x005a1021, + 0x24424004, 0x8c430000, 0x3063ffff, 0x14600004, 0x3c02ffff, 0x34427fff, + 0x00821024, 0xaf82000c, 0x9782000e, 0x9743010c, 0x8f440104, 0x3042bfff, + 0x00031c00, 0x3084ffff, 0x00641825, 0xa74201a6, 0xaf4301ac, 0x3c021000, + 0xaf4201b8, 0x00001021, 0x03e00008, 0x00000000, 0x27bdffe0, 0xafbf0018, + 0xafb10014, 0xafb00010, 0x8f420140, 0xaf420020, 0x8f430148, 0x3c027000, + 0x00621824, 0x3c024000, 0x1062000c, 0x0043102b, 0x14400006, 0x3c025000, + 0x3c023000, 0x1062000b, 0x3c024000, 0x0a00031f, 0x00000000, 0x10620034, + 0x3c024000, 0x0a00031f, 0x00000000, 0x0e00067c, 0x00000000, 0x0a00031f, + 0x3c024000, 0x8f420148, 0x24030002, 0x3044ffff, 0x00021402, 0x305000ff, + 0x1203000c, 0x27510180, 0x2a020003, 0x10400005, 0x24020003, 0x0600001d, + 0x36053000, 0x0a00030a, 0x3c038000, 0x12020007, 0x00000000, 0x0a000317, + 0x00000000, 0x0e000423, 0x00000000, 0x0a000308, 0x00402021, 0x0e000435, + 0x00000000, 0x00402021, 0x36053000, 0x3c038000, 0x8f4201b8, 0x00431024, + 0x1440fffd, 0x24020002, 0xa6250008, 0xa222000b, 0xa6240010, 0x8f420144, + 0x3c031000, 0xae220024, 0xaf4301b8, 0x0a00031f, 0x3c024000, 0x0000000d, + 0x00000000, 0x240001c3, 0x0a00031f, 0x3c024000, 0x0e0007f7, 0x00000000, + 0x3c024000, 0xaf420178, 0x00000000, 0x8fbf0018, 0x8fb10014, 0x8fb00010, + 0x03e00008, 0x27bd0020, 0x24020800, 0x03e00008, 0xaf4201b8, 0x27bdffe8, + 0x3c04600c, 0xafbf0014, 0xafb00010, 0x8c825000, 0x3c1a8000, 0x2403ff7f, + 0x3c106000, 0x00431024, 0x3442380c, 0x24030003, 0xac825000, 0x3c040008, + 0xaf430008, 0x8e020808, 0x3c030800, 0xac600020, 0x3042fff0, 0x2c420001, + 0xaf820004, 0x0e000819, 0x0344d825, 0x0e000781, 0x00000000, 0x3c020400, + 0x3442000c, 0x3c03ffff, 0x34630806, 0xae021948, 0xae03194c, 0x8e021980, + 0x34420200, 0xae021980, 0x8f500000, 0x32020003, 0x1040fffd, 0x32020001, + 0x10400004, 0x32020002, 0x0e0003bd, 0x00000000, 0x32020002, 0x1040fff6, + 0x00000000, 0x0e0002d4, 0x00000000, 0x0a00034a, 0x00000000, 0x27bdffe8, + 0x3c04600c, 0xafbf0014, 0xafb00010, 0x8c825000, 0x3c1a8000, 0x2403ff7f, + 0x3c106000, 0x00431024, 0x3442380c, 0x24030003, 0xac825000, 0x3c040008, + 0xaf430008, 0x8e020808, 0x3c030800, 0xac600020, 0x3042fff0, 0x2c420001, + 0xaf820004, 0x0e000819, 0x0344d825, 0x0e000781, 0x00000000, 0x3c020400, + 0x3442000c, 0x3c03ffff, 0x34630806, 0xae021948, 0xae03194c, 0x8e021980, + 0x8fbf0014, 0x34420200, 0xae021980, 0x8fb00010, 0x03e00008, 0x27bd0018, + 0x30a5ffff, 0x30c6ffff, 0x30e7ffff, 0x3c038000, 0x8f4201b8, 0x00431024, + 0x1440fffd, 0x24020003, 0xa342018b, 0x8f830004, 0xaf440180, 0xa745018c, + 0x10600005, 0xa746018e, 0x9743011c, 0x9742011e, 0x0a000393, 0x00021400, + 0x9743011e, 0x9742011c, 0x00021400, 0x00621825, 0xaf4301a8, 0x8f84000c, + 0x30828000, 0x1040000c, 0xa7470188, 0x93420116, 0x304200fc, 0x005a1021, + 0x24424004, 0x8c430000, 0x3063ffff, 0x14600004, 0x3c02ffff, 0x34427fff, + 0x00821024, 0xaf82000c, 0x9782000e, 0x9743010c, 0x8f440104, 0x3042bfff, + 0x00031c00, 0x3084ffff, 0x00641825, 0xa74201a6, 0xaf4301ac, 0x3c021000, + 0xaf4201b8, 0x03e00008, 0x00000000, 0x3c038000, 0x8f4201b8, 0x00431024, + 0x1440fffd, 0x24020002, 0x24032000, 0xa342018b, 0xa7430188, 0x3c021000, + 0xaf4201b8, 0x03e00008, 0x00000000, 0x27bdffe8, 0xafbf0010, 0x8f460128, + 0xaf460020, 0x8f420104, 0x8f450100, 0x24030800, 0x3c040010, 0xaf820000, + 0x00441024, 0xaf85000c, 0xaf4301b8, 0x14400005, 0x3c02001f, 0x3c030800, + 0x8c620020, 0x0a0003d5, 0x00002021, 0x3442ff00, 0x14c20009, 0x2402bfff, + 0x3c030800, 0x8c620020, 0x24040001, 0x24420001, 0x0e00004c, 0xac620020, + 0x0a0003e4, 0x00000000, 0x00a21024, 0x14400006, 0x00000000, 0xaf400048, + 0x0e000448, 0xaf400040, 0x0a0003e4, 0x00000000, 0x0e000783, 0x00000000, + 0x10400005, 0x3c024000, 0x8f430124, 0x3c026020, 0xac430014, 0x3c024000, + 0xaf420138, 0x00000000, 0x8fbf0010, 0x03e00008, 0x27bd0018, 0x27bdffe0, + 0xafbf0018, 0xafb10014, 0xafb00010, 0x8f420148, 0x24030002, 0x3044ffff, + 0x00021402, 0x305000ff, 0x1203000c, 0x27510180, 0x2a020003, 0x10400005, + 0x24020003, 0x0600001d, 0x36053000, 0x0a00040e, 0x3c038000, 0x12020007, + 0x00000000, 0x0a00041b, 0x00000000, 0x0e000423, 0x00000000, 0x0a00040c, + 0x00402021, 0x0e000435, 0x00000000, 0x00402021, 0x36053000, 0x3c038000, + 0x8f4201b8, 0x00431024, 0x1440fffd, 0x24020002, 0xa6250008, 0xa222000b, + 0xa6240010, 0x8f420144, 0x3c031000, 0xae220024, 0xaf4301b8, 0x0a00041f, + 0x8fbf0018, 0x0000000d, 0x00000000, 0x240001c3, 0x8fbf0018, 0x8fb10014, + 0x8fb00010, 0x03e00008, 0x27bd0020, 0x3084ffff, 0x2c821389, 0x1040000d, + 0x00001021, 0x3c030800, 0x24632100, 0x00042942, 0x00052880, 0x00a32821, + 0x3086001f, 0x8ca40000, 0x24030001, 0x00c31804, 0x00832025, 0x03e00008, + 0xaca40000, 0x03e00008, 0x24020091, 0x3084ffff, 0x2c821389, 0x1040000e, + 0x00001021, 0x3c030800, 0x24632100, 0x00042942, 0x00052880, 0x00a32821, + 0x3086001f, 0x24030001, 0x8ca40000, 0x00c31804, 0x00031827, 0x00832024, + 0x03e00008, 0xaca40000, 0x03e00008, 0x24020091, 0x27bdffb0, 0x3c026000, + 0xafbf0048, 0x8c434448, 0xaf630140, 0x93620005, 0x30420001, 0x14400005, + 0x00000000, 0x0e0007ed, 0x00000000, 0x0a00067a, 0x8fbf0048, 0x93420116, + 0x93430112, 0x8f430104, 0x3c040020, 0x34424000, 0x00641824, 0x1060000d, + 0x03426021, 0x8f430128, 0x27420180, 0xac430000, 0x8f650040, 0x24040008, + 0x240340c1, 0xa4430008, 0x24030002, 0xa043000b, 0x3c031000, 0x0a000563, + 0xa044000a, 0x8f420104, 0x3c030040, 0x00431024, 0x10400007, 0x00000000, + 0x8f430128, 0x27420180, 0xac430000, 0x8f650040, 0x0a00055c, 0x24040010, + 0xaf400048, 0xaf400054, 0xaf400040, 0x8f630048, 0x8f620040, 0x00624823, + 0x05210004, 0x00000000, 0x0000000d, 0x00000000, 0x24000132, 0x9742011a, + 0x3046ffff, 0x10c00004, 0x8d880004, 0x01061021, 0x0a000487, 0x2445ffff, + 0x01002821, 0x918a000d, 0xa7a00020, 0xafa00028, 0x9364003f, 0x3c026000, + 0x8c434448, 0x308700ff, 0x31420004, 0x10400033, 0xaf630144, 0x24090012, + 0x14e90006, 0x3c040800, 0x8c830028, 0x24020001, 0x24630001, 0x0a00054e, + 0xac830028, 0x8f620044, 0x15020012, 0x97a20020, 0x27a60010, 0x27450180, + 0x3442001a, 0xa7a20020, 0x8f630040, 0x3c048000, 0x24020020, 0xa3a70022, + 0xa3a90023, 0xa3a2001a, 0xafa30028, 0x8f4201b8, 0x00441024, 0x1440fffd, + 0x00000000, 0x0a000533, 0x00000000, 0x8f620044, 0x01021023, 0x0440009e, + 0x24020001, 0x8f620048, 0x01021023, 0x0441009a, 0x24020001, 0x97a20020, + 0x27a60010, 0x34420001, 0xa7a20020, 0x8f630040, 0x27450180, 0x3c048000, + 0xafa30028, 0x8f4201b8, 0x00441024, 0x1440fffd, 0x00000000, 0x0a000533, + 0x00000000, 0x3c026000, 0x8c424448, 0xaf620148, 0x8f630040, 0x00685823, + 0x19600013, 0x00cb102a, 0x54400007, 0x314a00fe, 0x5566000c, 0x010b4021, + 0x31420001, 0x54400009, 0x010b4021, 0x314a00fe, 0x24020001, 0xa7a20020, + 0x8f630040, 0x00c05821, 0x00003021, 0x0a0004dd, 0xafa30028, 0x00cb1023, + 0x0a0004dd, 0x3046ffff, 0x00005821, 0x8f620048, 0x2442ffff, 0x00a21823, + 0x18600019, 0x0066102a, 0x14400013, 0x24020001, 0xa7a20020, 0x8f630040, + 0xafa30028, 0x8f620040, 0x55020005, 0x27a60010, 0x55200003, 0x27a60010, + 0x0a0004f6, 0x00c01821, 0x27450180, 0x3c038000, 0x8f4201b8, 0x00431024, + 0x1440fffd, 0x00000000, 0x0a000533, 0x00000000, 0x8f650048, 0x00c31023, + 0x3046ffff, 0x314a00f6, 0x3c046000, 0x8c824448, 0x31430002, 0x1060001e, + 0xaf62014c, 0x8f620044, 0x1502000e, 0x97a20020, 0x27a60010, 0x34420200, + 0xa7a20020, 0x8f630040, 0x27450180, 0x3c048000, 0xafa30028, 0x8f4201b8, + 0x00441024, 0x1440fffd, 0x00000000, 0x0a000533, 0x00000000, 0x27a60010, + 0x34420001, 0xa7a20020, 0x8f630040, 0x27450180, 0x3c048000, 0xafa30028, + 0x8f4201b8, 0x00441024, 0x1440fffd, 0x00000000, 0x0a000533, 0x00000000, + 0x3c026000, 0x8c424448, 0x31430010, 0xaf620150, 0x54600003, 0x8d890008, + 0x0a00054e, 0x24020001, 0x8f630054, 0x2522ffff, 0x00431023, 0x1840002a, + 0x24020001, 0x27a60010, 0xa7a20020, 0x8f630040, 0x27450180, 0x3c048000, + 0xafa30028, 0x8f4201b8, 0x00441024, 0x1440fffd, 0x00000000, 0x8f420128, + 0xaca20000, 0x8cc30018, 0x240240c1, 0xa4a20008, 0xaca30018, 0x90c4000a, + 0x24020002, 0xa0a2000b, 0xa0a4000a, 0x94c20010, 0xa4a20010, 0x90c30012, + 0xa0a30012, 0x90c20013, 0xa0a20013, 0x8cc30014, 0xaca30014, 0x8cc20024, + 0xaca20024, 0x8cc30028, 0xaca30028, 0x8cc4002c, 0x24020001, 0x3c031000, + 0xaca4002c, 0xaf4301b8, 0xaf400044, 0xaf400050, 0x0a00067a, 0x8fbf0048, + 0x3c026000, 0x8c424448, 0x31430020, 0x10600019, 0xaf620154, 0x8f430128, + 0x27420180, 0xac430000, 0x8f650040, 0x24040004, 0x240340c1, 0xa4430008, + 0x24030002, 0xa044000a, 0x24040008, 0xa043000b, 0x3c031000, 0xa4440010, + 0xa0400012, 0xa0400013, 0xac400014, 0xac400024, 0xac400028, 0xac40002c, + 0xac450018, 0x0e0007ed, 0xaf4301b8, 0x0a00067a, 0x8fbf0048, 0x8f430104, + 0x8c824448, 0x38e3000a, 0x2c630001, 0xaf620158, 0x38e2000c, 0x2c420001, + 0x00621825, 0x14600003, 0x2402000e, 0x14e2002a, 0x00000000, 0x50c00008, + 0x9584000e, 0x10c00004, 0xa7a60040, 0x01061021, 0x0a000583, 0x2445ffff, + 0x01002821, 0x9584000e, 0x93630035, 0x8f62004c, 0x00642004, 0x00892021, + 0x00821023, 0x1840001f, 0x3c026000, 0x8f620018, 0x01021023, 0x1c40000f, + 0x97a20020, 0x8f620018, 0x15020018, 0x3c026000, 0x8f62001c, 0x01221023, + 0x1c400008, 0x97a20020, 0x8f62001c, 0x15220011, 0x3c026000, 0x8f620058, + 0x00821023, 0x1840000c, 0x97a20020, 0xafa50028, 0xafa80034, 0xafa90038, + 0xafa4003c, 0x34420020, 0x0a0005a8, 0xa7a20020, 0x8f680040, 0x00003021, + 0x8f640058, 0x01002821, 0x3c026000, 0x8c434448, 0xaf63015c, 0x8f62004c, + 0x01221023, 0x18400009, 0x00000000, 0x8f620054, 0x01221023, 0x1c400005, + 0x97a20020, 0xafa50028, 0xafa90024, 0x0a0005c3, 0x34420040, 0x9742011a, + 0x1440000c, 0x24020014, 0x8f620058, 0x14820009, 0x24020014, 0x8f63004c, + 0x8f620054, 0x10620004, 0x97a20020, 0xafa50028, 0x34420080, 0xa7a20020, + 0x24020014, 0x10e2000a, 0x28e20015, 0x10400005, 0x2402000c, 0x10e20006, + 0x3c026000, 0x0a000600, 0x00000000, 0x24020016, 0x14e20031, 0x3c026000, + 0x8f620054, 0x24420001, 0x1522002d, 0x3c026000, 0x24020014, 0x10e2001e, + 0x28e20015, 0x10400005, 0x2402000c, 0x10e20008, 0x3c026000, 0x0a000600, + 0x00000000, 0x24020016, 0x10e2000c, 0x97a20020, 0x0a000600, 0x3c026000, + 0x97a30020, 0x2402000e, 0xafa50028, 0xa3a70022, 0xa3a20023, 0xafa90024, + 0x34630054, 0x0a0005ff, 0xa7a30020, 0x24030010, 0x24040002, 0xafa50028, + 0xa3a70022, 0xa3a30023, 0xa3a4001a, 0xafa90024, 0x0a0005fe, 0x3442005d, + 0x97a20020, 0x24030012, 0x24040002, 0xafa50028, 0xa3a70022, 0xa3a30023, + 0xa3a4001a, 0xafa90024, 0x3042fffe, 0x3442005c, 0xa7a20020, 0x3c026000, + 0x8c434448, 0x31420001, 0xaf630160, 0x1040002c, 0x2402000c, 0x10e20014, + 0x28e2000d, 0x10400005, 0x2402000a, 0x10e20008, 0x97a20020, 0x0a000631, + 0x3c026000, 0x2402000e, 0x10e20018, 0x3c026000, 0x0a000631, 0x00000000, + 0x24030008, 0x24040002, 0xafa50028, 0xa3a70022, 0xa3a30023, 0xa3a4001a, + 0x0a00062f, 0x34420013, 0x97a30020, 0x30620004, 0x1440000b, 0x97a20020, + 0x3462001b, 0xa7a20020, 0x24020016, 0x24030002, 0xafa50028, 0xa3a70022, + 0xa3a20023, 0x0a000630, 0xa3a3001a, 0x97a20020, 0x24030010, 0x24040002, + 0xafa50028, 0xa3a70022, 0xa3a30023, 0xa3a4001a, 0x3442001b, 0xa7a20020, + 0x3c026000, 0x8c434448, 0x31420009, 0x0002102b, 0x00021023, 0x30420007, + 0x34440003, 0xaf630164, 0x10c00016, 0x24030800, 0x8f820010, 0x27450180, + 0x24420001, 0xaf820010, 0x24020004, 0xaf4301b8, 0xa4a40008, 0xa0a2000b, + 0x93440120, 0x3c031000, 0xa4a6000e, 0xaca90024, 0xaca80028, 0x008b2021, + 0xa4a4000c, 0xaf4301b8, 0x97a20020, 0x00003021, 0x3042ffbf, 0x0a000650, + 0xa7a20020, 0x24060001, 0x3c026000, 0x8c434448, 0xaf630168, 0x97a20020, + 0x10400020, 0x27450180, 0x3c038000, 0x8f4201b8, 0x00431024, 0x1440fffd, + 0x00000000, 0x8f420128, 0xaca20000, 0x8fa30028, 0x240240c1, 0xa4a20008, + 0xaca30018, 0x93a4001a, 0x24020002, 0xa0a2000b, 0xa0a4000a, 0x97a20020, + 0xa4a20010, 0x93a30022, 0xa0a30012, 0x93a20023, 0xa0a20013, 0x8fa30024, + 0xaca30014, 0x8fa20034, 0xaca20024, 0x8fa30038, 0xaca30028, 0x8fa2003c, + 0x3c031000, 0xaca2002c, 0xaf4301b8, 0x3c026000, 0x8c434448, 0x00c01021, + 0xaf63016c, 0x8fbf0048, 0x03e00008, 0x27bd0050, 0x8f460140, 0x8f470148, + 0x3c028000, 0x00e24024, 0x00072c02, 0x30a300ff, 0x2402000b, 0x1062008f, + 0x27440180, 0x2862000c, 0x10400011, 0x24020006, 0x1062005a, 0x28620007, + 0x10400007, 0x24020008, 0x10600024, 0x24020001, 0x10620037, 0x00000000, + 0x0a00077e, 0x00000000, 0x106200a9, 0x24020009, 0x106200bb, 0x00071c02, + 0x0a00077e, 0x00000000, 0x2402001b, 0x106200c7, 0x2862001c, 0x10400007, + 0x2402000e, 0x106200b1, 0x24020019, 0x106200c2, 0x00071c02, 0x0a00077e, + 0x00000000, 0x24020080, 0x10620060, 0x28620081, 0x10400005, 0x2402001c, + 0x10620094, 0x00071c02, 0x0a00077e, 0x00000000, 0x240200c2, 0x106200c5, + 0x00a01821, 0x0a00077e, 0x00000000, 0x00a01821, 0x3c058000, 0x8f4201b8, + 0x00451024, 0x1440fffd, 0x24020001, 0xa4830008, 0x24030002, 0xac860000, + 0xac800004, 0xa082000a, 0xa083000b, 0xa4870010, 0x8f430144, 0x3c021000, + 0xac800028, 0xac830024, 0x3c036000, 0xaf4201b8, 0x03e00008, 0xac600808, + 0x11000009, 0x00a01821, 0x3c020800, 0x24030002, 0xa0434490, 0x24424490, + 0xac460008, 0x8f430144, 0x03e00008, 0xac430004, 0x3c058000, 0x8f4201b8, + 0x00451024, 0x1440fffd, 0x24020002, 0xac800000, 0xac860004, 0xa4830008, + 0xa082000a, 0xa082000b, 0xa4870010, 0xac800024, 0x8f420144, 0x3c031000, + 0xac820028, 0x3c026000, 0xaf4301b8, 0x03e00008, 0xac400808, 0x00a01821, + 0x3c080800, 0x3c058000, 0x8f4201b8, 0x00451024, 0x1440fffd, 0x00000000, + 0xac860000, 0x91024490, 0x00002821, 0x10400002, 0x25064490, 0x8cc50008, + 0xac850004, 0xa4830008, 0x91034490, 0x24020002, 0xa082000b, 0xa4870010, + 0x34630001, 0xa083000a, 0x8f420144, 0xac820024, 0x91034490, 0x10600002, + 0x00001021, 0x8cc20004, 0xac820028, 0x3c021000, 0xaf4201b8, 0x3c026000, + 0xa1004490, 0x03e00008, 0xac400808, 0x00a01821, 0x3c058000, 0x8f4201b8, + 0x00451024, 0x1440fffd, 0x24020002, 0xa082000b, 0xa4830008, 0xa4870010, + 0x8f420144, 0x3c031000, 0xa4820012, 0x03e00008, 0xaf4301b8, 0x30e2ffff, + 0x14400028, 0x00071c02, 0x93620005, 0x30420004, 0x14400020, 0x3c029000, + 0x34420001, 0x00c21025, 0xaf420020, 0x3c038000, 0x8f420020, 0x00431024, + 0x1440fffd, 0x00000000, 0x93620005, 0x3c038000, 0x34630001, 0x00c31825, + 0x34420004, 0xa3620005, 0xaf430020, 0x93620005, 0x30420004, 0x14400003, + 0x3c038000, 0x0000000d, 0x3c038000, 0x8f4201b8, 0x00431024, 0x1440fffd, + 0x24020005, 0x3c031000, 0xac860000, 0xa082000b, 0xaf4301b8, 0x0a00073d, + 0x00071c02, 0x0000000d, 0x03e00008, 0x00000000, 0x00071c02, 0x3c058000, + 0x8f4201b8, 0x00451024, 0x1440fffd, 0x24020001, 0xa4830008, 0x24030002, + 0xac860000, 0xac800004, 0xa082000a, 0xa083000b, 0xa4870010, 0x8f430144, + 0x3c021000, 0xac800028, 0xac830024, 0x03e00008, 0xaf4201b8, 0x00071c02, + 0x3c058000, 0x8f4201b8, 0x00451024, 0x1440fffd, 0x24020002, 0xac800000, + 0xac860004, 0xa4830008, 0xa082000a, 0xa082000b, 0xa4870010, 0xac800024, + 0x8f420144, 0x3c031000, 0xac820028, 0x03e00008, 0xaf4301b8, 0x00071c02, + 0x3c058000, 0x8f4201b8, 0x00451024, 0x1440fffd, 0x24020001, 0xa4830008, + 0x24030002, 0xa082000a, 0x3c021000, 0xac860000, 0xac800004, 0xa083000b, + 0xa4870010, 0xac800024, 0xac800028, 0x03e00008, 0xaf4201b8, 0x3c058000, + 0x8f4201b8, 0x00451024, 0x1440fffd, 0x24020002, 0xac860000, 0xac800004, + 0xa4830008, 0xa080000a, 0x0a000748, 0xa082000b, 0x0000000d, 0x03e00008, + 0x00000000, 0x03e00008, 0x00000000, 0x8f420100, 0x3042003e, 0x14400011, + 0x24020001, 0xaf400048, 0x8f420100, 0x304207c0, 0x10400005, 0x00000000, + 0xaf40004c, 0xaf400050, 0x03e00008, 0x24020001, 0xaf400054, 0xaf400040, + 0x8f420100, 0x30423800, 0x54400001, 0xaf400044, 0x24020001, 0x03e00008, + 0x00000000, 0x3c029000, 0x34420001, 0x00822025, 0xaf440020, 0x3c038000, + 0x8f420020, 0x00431024, 0x1440fffd, 0x00000000, 0x03e00008, 0x00000000, + 0x3c028000, 0x34420001, 0x00822025, 0x03e00008, 0xaf440020, 0x8f430128, + 0x27420180, 0xac430000, 0x8f650040, 0x240340c1, 0xa4430008, 0x24030002, + 0xa044000a, 0x24040008, 0xa043000b, 0x3c031000, 0xa4440010, 0xa0400012, + 0xa0400013, 0xac400014, 0xac400024, 0xac400028, 0xac40002c, 0xac450018, + 0x03e00008, 0xaf4301b8, 0x24020001, 0xacc40000, 0x03e00008, 0xa4e50000, + 0x03e00008, 0x24020001, 0x24020001, 0xaf400044, 0x03e00008, 0xaf400050, + 0x00803021, 0x27450180, 0x3c038000, 0x8f4201b8, 0x00431024, 0x1440fffd, + 0x00000000, 0x8f420128, 0xaca20000, 0x8cc30018, 0x240240c1, 0xa4a20008, + 0xaca30018, 0x90c4000a, 0x24020002, 0xa0a2000b, 0xa0a4000a, 0x94c20010, + 0xa4a20010, 0x90c30012, 0xa0a30012, 0x90c20013, 0xa0a20013, 0x8cc30014, + 0xaca30014, 0x8cc20024, 0xaca20024, 0x8cc30028, 0xaca30028, 0x8cc2002c, + 0x3c031000, 0xaca2002c, 0x24020001, 0xaf4301b8, 0xaf400044, 0x03e00008, + 0xaf400050, 0x27bdffe8, 0xafbf0010, 0x0e000326, 0x00000000, 0x00002021, + 0x0e00004c, 0xaf400180, 0x8fbf0010, 0x03e00008, 0x27bd0018, 0x8f460148, + 0x27450180, 0x3c038000, 0x00061402, 0x304700ff, 0x8f4201b8, 0x00431024, + 0x1440fffd, 0x00000000, 0x8f440140, 0x00061202, 0x304200ff, 0x00061c02, + 0xaca20004, 0x24020002, 0xa4a30008, 0x30c300ff, 0xa0a2000b, 0xaca30024, + 0x10e0000a, 0xaca40000, 0x28e20004, 0x14400005, 0x24020001, 0x24020005, + 0x54e20005, 0xa0a0000a, 0x24020001, 0x0a000816, 0xa0a2000a, 0xa0a0000a, + 0x3c021000, 0x03e00008, 0xaf4201b8, 0x03e00008, 0x00001021, 0x10c00007, + 0x00000000, 0x8ca20000, 0x24c6ffff, 0x24a50004, 0xac820000, 0x14c0fffb, + 0x24840004, 0x03e00008, 0x00000000, 0x0a00082a, 0x00a01021, 0xac860000, + 0x24840004, 0x00a01021, 0x1440fffc, 0x24a5ffff, 0x03e00008, 0x00000000, + 0x00000000 }; + +static u32 bnx2_RXP_b06FwData[(0x0/4) + 1] = { 0x00000000 }; +static u32 bnx2_RXP_b06FwRodata[(0x0/4) + 1] = { 0x00000000 }; +static u32 bnx2_RXP_b06FwBss[(0x239c/4) + 1] = { 0x00000000 }; +static u32 bnx2_RXP_b06FwSbss[(0x14/4) + 1] = { 0x00000000 }; + +static u32 bnx2_rv2p_proc1[] = { + 0x00000008, 0xac000001, 0x0000000c, 0x2f800001, 0x00000010, 0x213f0004, + 0x00000010, 0x20bf002c, 0x00000010, 0x203f0143, 0x00000018, 0x8000fffd, + 0x00000010, 0xb1b8b017, 0x0000000b, 0x2fdf0002, 0x00000000, 0x03d80000, + 0x00000000, 0x2c380000, 0x00000008, 0x2c800000, 0x00000008, 0x2d000000, + 0x00000010, 0x91d40000, 0x00000008, 0x2d800108, 0x00000008, 0x02000002, + 0x00000010, 0x91de0000, 0x0000000f, 0x42e0001c, 0x00000010, 0x91840a08, + 0x00000008, 0x2c8000b0, 0x00000008, 0x2d000008, 0x00000008, 0x2d800150, + 0x00000000, 0x00000000, 0x00000010, 0x91de0000, 0x00000010, 0x2c620002, + 0x00000018, 0x80000012, 0x0000000b, 0x2fdf0002, 0x0000000c, 0x1f800002, + 0x00000000, 0x2c070000, 0x00000018, 0x8000ffe6, 0x00000008, 0x02000002, + 0x0000000f, 0x42e0001c, 0x00000010, 0x91840a08, 0x00000008, 0x2c8000b0, + 0x00000008, 0x2d000008, 0x00000010, 0x91d40000, 0x00000008, 0x2d800108, + 0x00000000, 0x00000000, 0x00000010, 0x91de0000, 0x00000018, 0x80000004, + 0x0000000c, 0x1f800002, 0x00000000, 0x00000000, 0x00000018, 0x8000ffd9, + 0x0000000c, 0x29800002, 0x0000000c, 0x1f800002, 0x00000000, 0x2adf0000, + 0x00000008, 0x2a000005, 0x00000018, 0x8000ffd4, 0x00000008, 0x02240030, + 0x00000018, 0x00040000, 0x00000018, 0x80000015, 0x00000018, 0x80000017, + 0x00000018, 0x8000001b, 0x00000018, 0x8000004c, 0x00000018, 0x8000008c, + 0x00000018, 0x8000000f, 0x00000018, 0x8000000e, 0x00000018, 0x8000000d, + 0x00000018, 0x8000000c, 0x00000018, 0x800000c2, 0x00000018, 0x8000000a, + 0x00000018, 0x80000009, 0x00000018, 0x80000008, 0x00000018, 0x800000fd, + 0x00000018, 0x80000006, 0x00000018, 0x80000005, 0x00000018, 0x800000ff, + 0x00000018, 0x80000104, 0x00000018, 0x80000002, 0x00000018, 0x80000098, + 0x00000018, 0x80000000, 0x0000000c, 0x1f800001, 0x00000000, 0x00000000, + 0x00000018, 0x8000ffba, 0x00000010, 0x91d40000, 0x0000000c, 0x29800001, + 0x0000000c, 0x1f800001, 0x00000008, 0x2a000002, 0x00000018, 0x8000ffb5, + 0x00000010, 0xb1a0b012, 0x0000000b, 0x2fdf0002, 0x00000000, 0x2c200000, + 0x00000008, 0x2c800000, 0x00000008, 0x2d000000, 0x00000010, 0x91d40000, + 0x00000008, 0x2d80011c, 0x00000000, 0x00000000, 0x00000010, 0x91de0000, + 0x0000000f, 0x47600008, 0x0000000f, 0x060e0001, 0x00000010, 0x001f0000, + 0x00000000, 0x0f580000, 0x00000000, 0x0a640000, 0x00000000, 0x0ae50000, + 0x00000000, 0x0b660000, 0x00000000, 0x0d610000, 0x00000018, 0x80000013, + 0x0000000f, 0x47600008, 0x0000000b, 0x2fdf0002, 0x00000008, 0x2c800000, + 0x00000008, 0x2d000000, 0x00000010, 0x91d40000, 0x00000008, 0x2d80011c, + 0x0000000f, 0x060e0001, 0x00000010, 0x001f0000, 0x00000000, 0x0f580000, + 0x00000010, 0x91de0000, 0x00000000, 0x0a640000, 0x00000000, 0x0ae50000, + 0x00000000, 0x0b660000, 0x00000000, 0x0d610000, 0x00000000, 0x02620000, + 0x0000000b, 0x2fdf0002, 0x00000000, 0x309a0000, 0x00000000, 0x31040000, + 0x00000000, 0x0c961800, 0x00000009, 0x0c99ffff, 0x00000004, 0xcc993400, + 0x00000010, 0xb1963202, 0x00000008, 0x0f800000, 0x0000000c, 0x29800001, + 0x00000010, 0x00220002, 0x0000000c, 0x29520001, 0x0000000c, 0x29520000, + 0x00000008, 0x22000001, 0x0000000c, 0x1f800001, 0x00000000, 0x2adf0000, + 0x00000008, 0x2a000003, 0x00000018, 0x8000ff83, 0x00000010, 0xb1a0b01d, + 0x0000000b, 0x2fdf0002, 0x00000000, 0x2c200000, 0x00000008, 0x2c8000b0, + 0x00000008, 0x2d000008, 0x00000010, 0x91d40000, 0x00000008, 0x2d800150, + 0x00000000, 0x00000000, 0x00000010, 0x205f0000, 0x00000008, 0x2c800000, + 0x00000008, 0x2d000000, 0x00000008, 0x2d800108, 0x00000000, 0x00000000, + 0x00000010, 0x91de0000, 0x0000000f, 0x47600008, 0x00000000, 0x060e0000, + 0x00000010, 0x001f0000, 0x00000000, 0x0f580000, 0x00000010, 0x91de0000, + 0x00000000, 0x0a640000, 0x00000000, 0x0ae50000, 0x00000000, 0x0b670000, + 0x00000000, 0x0d620000, 0x00000000, 0x0ce71800, 0x00000009, 0x0c99ffff, + 0x00000004, 0xcc993400, 0x00000010, 0xb1963220, 0x00000008, 0x0f800000, + 0x00000018, 0x8000001e, 0x0000000f, 0x47600008, 0x0000000b, 0x2fdf0002, + 0x00000008, 0x2c8000b0, 0x00000008, 0x2d000008, 0x00000010, 0x91d40000, + 0x00000008, 0x2d80012c, 0x0000000f, 0x060e0001, 0x00000010, 0x001f0000, + 0x00000000, 0x0f580000, 0x00000010, 0x91de0000, 0x00000000, 0x0a640000, + 0x00000000, 0x0ae50000, 0x00000000, 0x0b670000, 0x00000000, 0x0d620000, + 0x00000000, 0x02630000, 0x0000000f, 0x47620010, 0x00000000, 0x0ce71800, + 0x0000000b, 0x2fdf0002, 0x00000000, 0x311a0000, 0x00000000, 0x31840000, + 0x0000000b, 0xc20000ff, 0x00000002, 0x42040000, 0x00000001, 0x31620800, + 0x0000000f, 0x020e0010, 0x00000002, 0x31620800, 0x00000009, 0x0c99ffff, + 0x00000004, 0xcc993400, 0x00000010, 0xb1963202, 0x00000008, 0x0f800000, + 0x0000000c, 0x29800001, 0x0000000c, 0x1f800001, 0x0000000c, 0x61420006, + 0x00000008, 0x22000008, 0x00000000, 0x2adf0000, 0x00000008, 0x2a000004, + 0x00000018, 0x8000ff42, 0x00000008, 0x2c8000b0, 0x00000008, 0x2d000008, + 0x00000010, 0x91a0b008, 0x00000010, 0x91d40000, 0x0000000c, 0x31620018, + 0x00000008, 0x2d800001, 0x00000000, 0x00000000, 0x00000010, 0x91de0000, + 0x00000008, 0xac000001, 0x00000018, 0x8000000e, 0x00000000, 0x0380b000, + 0x0000000b, 0x2fdf0002, 0x00000000, 0x2c004000, 0x00000010, 0x91d40000, + 0x00000008, 0x2d800101, 0x00000000, 0x00000000, 0x00000010, 0x91de0000, + 0x0000000c, 0x31620018, 0x00000008, 0x2d800001, 0x00000000, 0x00000000, + 0x00000010, 0x91de0000, 0x0000000b, 0x2fdf0002, 0x00000000, 0x2c000e00, + 0x0000000c, 0x29800001, 0x0000000c, 0x1f800001, 0x00000008, 0x2a000007, + 0x00000018, 0x8000ff27, 0x00000010, 0xb1a0b016, 0x0000000b, 0x2fdf0002, + 0x00000000, 0x03d80000, 0x00000000, 0x2c200000, 0x00000008, 0x2c8000b0, + 0x00000008, 0x2d000008, 0x00000010, 0x91d40000, 0x00000008, 0x2d800150, + 0x00000000, 0x00000000, 0x00000010, 0x205f0000, 0x00000008, 0x2c800000, + 0x00000008, 0x2d000000, 0x00000008, 0x2d800108, 0x00000008, 0x07000001, + 0x00000010, 0xb5de1c00, 0x00000010, 0x2c620002, 0x00000018, 0x8000000a, + 0x0000000b, 0x2fdf0002, 0x00000000, 0x2c070000, 0x0000000c, 0x1f800001, + 0x00000010, 0x91de0000, 0x00000018, 0x8000ff11, 0x00000008, 0x2c8000b0, + 0x00000008, 0x2d000008, 0x00000010, 0x91d40000, 0x00000008, 0x2d800108, + 0x0000000c, 0x29800001, 0x0000000c, 0x1f800001, 0x00000010, 0x91de0000, + 0x00000000, 0x2adf0000, 0x00000008, 0x2a00000a, 0x00000018, 0x8000ff07, + 0x00000000, 0x82265600, 0x0000000f, 0x47220008, 0x00000009, 0x070e000f, + 0x00000008, 0x070e0008, 0x00000008, 0x02800001, 0x00000007, 0x02851c00, + 0x00000008, 0x82850001, 0x00000000, 0x02840a00, 0x00000007, 0x42851c00, + 0x00000003, 0xc3aa5200, 0x00000000, 0x03b10e00, 0x00000010, 0x001f0000, + 0x0000000f, 0x0f280007, 0x00000007, 0x4b071c00, 0x00000000, 0x00000000, + 0x0000000f, 0x0a960003, 0x00000000, 0x0a955c00, 0x00000000, 0x4a005a00, + 0x00000000, 0x0c960a00, 0x00000009, 0x0c99ffff, 0x00000008, 0x0d00ffff, + 0x00000010, 0xb1963202, 0x00000008, 0x0f800005, 0x00000010, 0x00220020, + 0x00000000, 0x02a70000, 0x00000010, 0xb1850002, 0x00000008, 0x82850200, + 0x00000000, 0x02000000, 0x00000000, 0x03a60000, 0x00000018, 0x8000004e, + 0x00000000, 0x072b0000, 0x00000001, 0x878c1c00, 0x00000000, 0x870e1e00, + 0x00000000, 0x860c1e00, 0x00000000, 0x03061e00, 0x00000010, 0xb18e0003, + 0x00000018, 0x80000047, 0x00000018, 0x8000fffa, 0x00000010, 0x918c0003, + 0x00000010, 0xb1870002, 0x00000018, 0x80000043, 0x00000010, 0x91d40000, + 0x0000000c, 0x29800001, 0x00000000, 0x2a860000, 0x00000000, 0x230c0000, + 0x00000000, 0x2b070000, 0x00000010, 0xb187000e, 0x00000008, 0x2a000008, + 0x00000018, 0x8000003b, 0x00000010, 0x91d40000, 0x00000000, 0x28d18c00, + 0x00000000, 0x2a860000, 0x00000000, 0x230c0000, 0x00000000, 0x2b070000, + 0x00000018, 0x8000fff8, 0x00000010, 0x91d40000, 0x0000000c, 0x29800001, + 0x00000000, 0x2aab0000, 0x00000000, 0xa3265600, 0x00000000, 0x2b000000, + 0x0000000c, 0x1f800001, 0x00000008, 0x2a000008, 0x00000018, 0x8000fec8, + 0x00000010, 0x91d40000, 0x0000000c, 0x29800001, 0x0000000c, 0x1f800001, + 0x00000008, 0x2a000009, 0x00000018, 0x8000fec3, 0x00000010, 0x91d40000, + 0x0000000c, 0x29800001, 0x0000000c, 0x1f800001, 0x00000000, 0x29420000, + 0x00000008, 0x2a000002, 0x00000018, 0x8000febd, 0x00000018, 0x8000febc, + 0x00000010, 0xb1bcb016, 0x0000000b, 0x2fdf0002, 0x00000000, 0x03d80000, + 0x00000000, 0x2c3c0000, 0x00000008, 0x2c8000b0, 0x00000008, 0x2d000008, + 0x00000010, 0x91d40000, 0x00000008, 0x2d800150, 0x00000000, 0x00000000, + 0x00000010, 0x205f0000, 0x00000008, 0x2c800000, 0x00000008, 0x2d000000, + 0x00000008, 0x2d800108, 0x00000008, 0x07000001, 0x00000010, 0xb5de1c00, + 0x00000010, 0x2c620002, 0x00000018, 0x8000000a, 0x0000000b, 0x2fdf0002, + 0x00000000, 0x2c070000, 0x0000000c, 0x1f800000, 0x00000010, 0x91de0000, + 0x00000018, 0x8000fea6, 0x00000008, 0x2c8000b0, 0x00000008, 0x2d000008, + 0x00000010, 0x91d40000, 0x00000008, 0x2d800108, 0x0000000c, 0x29800000, + 0x0000000c, 0x1f800000, 0x00000010, 0x91de0000, 0x00000000, 0x2adf0000, + 0x00000008, 0x2a000006, 0x00000018, 0x8000fe9c, 0x00000008, 0x03050004, + 0x00000006, 0x83040c00, 0x00000008, 0x02850200, 0x00000000, 0x86050c00, + 0x00000001, 0x860c0e00, 0x00000008, 0x02040004, 0x00000000, 0x02041800, + 0x00000000, 0x83871800, 0x00000018, 0x00020000 }; + +static u32 bnx2_rv2p_proc2[] = { + 0x00000000, 0x2a000000, 0x00000010, 0xb1d40000, 0x00000008, 0x02540003, + 0x00000018, 0x00040000, 0x00000018, 0x8000000a, 0x00000018, 0x8000000a, + 0x00000018, 0x8000000e, 0x00000018, 0x80000056, 0x00000018, 0x800001b9, + 0x00000018, 0x800001e1, 0x00000018, 0x8000019b, 0x00000018, 0x800001f9, + 0x00000018, 0x8000019f, 0x00000018, 0x800001a6, 0x00000018, 0x80000000, + 0x0000000c, 0x29800001, 0x00000000, 0x2a000000, 0x0000000c, 0x29800000, + 0x00000010, 0x20530000, 0x00000018, 0x8000ffee, 0x0000000c, 0x29800001, + 0x00000010, 0x91de0000, 0x00000010, 0x001f0000, 0x00000000, 0x2f80aa00, + 0x00000000, 0x2a000000, 0x00000000, 0x0d610000, 0x00000000, 0x03620000, + 0x00000000, 0x2c400000, 0x00000000, 0x02638c00, 0x00000000, 0x26460000, + 0x00000010, 0x00420002, 0x00000008, 0x02040012, 0x00000010, 0xb9060836, + 0x00000000, 0x0f580000, 0x00000000, 0x0a640000, 0x00000000, 0x0ae50000, + 0x00000000, 0x0b660000, 0x00000000, 0x0c000000, 0x00000000, 0x0b800000, + 0x00000010, 0x00420009, 0x00000008, 0x0cc60012, 0x00000008, 0x0f800003, + 0x00000000, 0x00000000, 0x00000010, 0x009f0000, 0x00000008, 0x27110012, + 0x00000000, 0x66900000, 0x00000008, 0xa31b0012, 0x00000018, 0x80000008, + 0x00000000, 0x0cc60000, 0x00000008, 0x0f800003, 0x00000000, 0x00000000, + 0x00000010, 0x009f0000, 0x00000000, 0x27110000, 0x00000000, 0x66900000, + 0x00000000, 0x231b0000, 0x00000010, 0xb197320e, 0x00000000, 0x25960000, + 0x00000000, 0x021b0000, 0x00000010, 0x001f0000, 0x00000008, 0x0f800003, + 0x0000000c, 0x29800000, 0x00000010, 0x20530000, 0x00000000, 0x22c50800, + 0x00000010, 0x009f0000, 0x00000000, 0x27002200, 0x00000000, 0x26802000, + 0x00000000, 0x231b0000, 0x0000000c, 0x69520001, 0x00000018, 0x8000fff3, + 0x00000010, 0x01130002, 0x00000010, 0xb1980003, 0x00000010, 0x001f0000, + 0x00000008, 0x0f800004, 0x00000008, 0x22000003, 0x00000008, 0x2c80000c, + 0x00000008, 0x2d00000c, 0x00000010, 0x009f0000, 0x00000000, 0x25960000, + 0x0000000c, 0x29800000, 0x00000000, 0x32140000, 0x00000000, 0x32950000, + 0x00000000, 0x33160000, 0x00000000, 0x31e32e00, 0x00000008, 0x2d800010, + 0x00000010, 0x20530000, 0x00000018, 0x8000ffac, 0x00000000, 0x23000000, + 0x00000000, 0x25e60000, 0x00000008, 0x2200000b, 0x0000000c, 0x69520000, + 0x0000000c, 0x29800000, 0x00000010, 0x20530000, 0x00000018, 0x8000ffa5, + 0x0000000c, 0x29800001, 0x00000010, 0x91de0000, 0x00000000, 0x2fd50000, + 0x00000010, 0x001f0000, 0x00000000, 0x02700000, 0x00000000, 0x0d620000, + 0x00000000, 0xbb630800, 0x00000000, 0x2a000000, 0x00000009, 0x076000ff, + 0x0000000f, 0x2c0e0007, 0x00000008, 0x2c800000, 0x00000008, 0x2d000064, + 0x00000008, 0x2d80011c, 0x00000009, 0x06420002, 0x0000000c, 0x61420001, + 0x00000000, 0x0f400000, 0x00000000, 0x02d08c00, 0x00000000, 0x23000000, + 0x00000004, 0x826da000, 0x00000000, 0x8304a000, 0x00000000, 0x22c50c00, + 0x00000000, 0x03760000, 0x00000004, 0x83860a00, 0x00000000, 0x83870c00, + 0x00000010, 0x91de0000, 0x00000000, 0x037c0000, 0x00000000, 0x837b0c00, + 0x00000001, 0x83060e00, 0x00000000, 0x83870c00, 0x00000000, 0x82850e00, + 0x00000010, 0xb1860016, 0x0000000f, 0x47610018, 0x00000000, 0x068e0000, + 0x0000000f, 0x47670010, 0x0000000f, 0x47e20010, 0x00000000, 0x870e1e00, + 0x00000010, 0xb70e1a10, 0x00000010, 0x0ce7000e, 0x00000008, 0x22000009, + 0x00000000, 0x286d0000, 0x0000000f, 0x65680010, 0x00000003, 0xf66c9400, + 0x00000010, 0xb972a003, 0x0000000c, 0x73e70019, 0x0000000c, 0x21420004, + 0x00000018, 0x8000023f, 0x00000000, 0x37ed0000, 0x0000000c, 0x73e7001a, + 0x00000010, 0x20530000, 0x00000008, 0x22000008, 0x0000000c, 0x61420004, + 0x00000000, 0x02f60000, 0x00000004, 0x82840a00, 0x00000010, 0xb1840a2b, + 0x00000010, 0x2d67000a, 0x00000010, 0xb96d0804, 0x00000004, 0xb6ed0a00, + 0x00000000, 0x37ed0000, 0x00000018, 0x80000029, 0x0000000c, 0x61420000, + 0x00000000, 0x37040000, 0x00000000, 0x37850000, 0x0000000c, 0x33e7001a, + 0x00000018, 0x80000024, 0x00000010, 0xb96d0809, 0x00000004, 0xb6ed0a00, + 0x00000000, 0x036d0000, 0x00000004, 0xb76e0c00, 0x00000010, 0x91ee0c1f, + 0x0000000c, 0x73e7001a, 0x00000004, 0xb6ef0c00, 0x00000000, 0x37ed0000, + 0x00000018, 0x8000001b, 0x0000000c, 0x61420000, 0x00000010, 0xb7ee0a05, + 0x00000010, 0xb96f0815, 0x00000003, 0xb76e0800, 0x00000004, 0xb7ef0a00, + 0x00000018, 0x80000015, 0x00000010, 0x0ce7000c, 0x00000008, 0x22000009, + 0x00000000, 0x286d0000, 0x0000000f, 0x65680010, 0x00000003, 0xf66c9400, + 0x00000010, 0xb972a003, 0x0000000c, 0x73e70019, 0x0000000c, 0x21420004, + 0x00000018, 0x80000215, 0x00000010, 0x20530000, 0x00000008, 0x22000008, + 0x0000000c, 0x61420004, 0x00000000, 0x37040000, 0x00000000, 0x37850000, + 0x00000000, 0x036d0000, 0x00000003, 0xb8f10c00, 0x00000018, 0x80000004, + 0x00000000, 0x02840000, 0x00000002, 0x21421800, 0x0000000c, 0x61420000, + 0x00000000, 0x286d0000, 0x0000000f, 0x65ed0010, 0x00000009, 0x266dffff, + 0x00000000, 0x23000000, 0x00000010, 0xb1840a3d, 0x00000010, 0x01420002, + 0x00000004, 0xb8f10a00, 0x00000003, 0x83760a00, 0x00000010, 0xb8040c39, + 0x00000010, 0xb7e6080a, 0x00000000, 0x0a640000, 0x00000000, 0x0ae50000, + 0x00000009, 0x0c68ffff, 0x00000009, 0x0b67ffff, 0x00000000, 0x0be60000, + 0x00000000, 0x0c840000, 0x00000010, 0xb197320c, 0x00000008, 0x0f800002, + 0x00000018, 0x8000000a, 0x00000000, 0x0a6a0000, 0x00000000, 0x0aeb0000, + 0x00000000, 0x0c000000, 0x00000009, 0x0b6cffff, 0x00000000, 0x0be90000, + 0x00000000, 0x0c840000, 0x00000010, 0xb1973203, 0x00000008, 0x0f800002, + 0x00000018, 0x80000001, 0x00000010, 0x001f0000, 0x00000000, 0x0c860000, + 0x00000000, 0x06980000, 0x00000008, 0x0f800003, 0x00000000, 0x00000000, + 0x00000010, 0x009f0000, 0x00000010, 0xb1973210, 0x00000000, 0x231b0000, + 0x00000000, 0x02043600, 0x00000003, 0x8384a000, 0x0000000f, 0x65870010, + 0x00000009, 0x2607ffff, 0x00000000, 0x27111a00, 0x00000000, 0x66900000, + 0x0000000c, 0x29000000, 0x00000018, 0x800001de, 0x00000000, 0x06980000, + 0x00000010, 0x20530000, 0x00000000, 0x22c58c00, 0x00000010, 0x001f0000, + 0x00000008, 0x0f800003, 0x00000018, 0x8000fff0, 0x00000000, 0x02043600, + 0x00000000, 0x231b0000, 0x00000003, 0x8384a000, 0x0000000f, 0x65870010, + 0x00000009, 0x2607ffff, 0x00000000, 0x27111a00, 0x00000000, 0x66900000, + 0x0000000c, 0x29000000, 0x00000010, 0x91840a02, 0x00000002, 0x21421800, + 0x00000000, 0x32140000, 0x00000000, 0x32950000, 0x00000005, 0x73e72c00, + 0x00000005, 0x74683000, 0x00000000, 0x33170000, 0x00000018, 0x80000138, + 0x00000010, 0x91c60004, 0x00000008, 0x07000004, 0x00000010, 0xb1c41c02, + 0x00000010, 0x91840a04, 0x00000018, 0x800001c3, 0x00000010, 0x20530000, + 0x00000000, 0x22c58c00, 0x00000010, 0xb1840a8e, 0x0000000c, 0x21420006, + 0x00000010, 0x0ce7001a, 0x0000000f, 0x43680010, 0x00000000, 0x03f30c00, + 0x00000010, 0x91870850, 0x0000000f, 0x46ec0010, 0x00000010, 0xb68d0c4e, + 0x00000000, 0x838d0c00, 0x00000000, 0xa3050800, 0x00000001, 0xa3460e00, + 0x00000000, 0x02048c00, 0x00000010, 0x91840a02, 0x00000002, 0x21421800, + 0x00000010, 0x001f0000, 0x00000008, 0x22000008, 0x00000003, 0x8384a000, + 0x0000000f, 0x65870010, 0x00000009, 0x2607ffff, 0x00000000, 0x27750c00, + 0x00000000, 0x66f40000, 0x0000000c, 0x29000000, 0x00000018, 0x800001aa, + 0x00000000, 0x03068c00, 0x00000003, 0xf4680c00, 0x00000010, 0x20530000, + 0x00000000, 0x22c58c00, 0x00000018, 0x8000ffe5, 0x00000000, 0x39760000, + 0x00000000, 0x39840000, 0x0000000c, 0x33e70019, 0x00000010, 0x001f0000, + 0x00000000, 0x031e0000, 0x00000000, 0x0760fe00, 0x0000000f, 0x0f0e0007, + 0x00000000, 0x83850800, 0x00000000, 0x0a7d0000, 0x00000000, 0x0afe0000, + 0x00000000, 0x0b7f0000, 0x00000000, 0x0d7a0000, 0x00000000, 0x0c000000, + 0x00000000, 0x0bfc0000, 0x00000000, 0x0c970e00, 0x00000008, 0x0f800003, + 0x0000000f, 0x47670010, 0x00000008, 0x070e0001, 0x0000000b, 0xc38000ff, + 0x00000002, 0x43870000, 0x00000001, 0x33e70e00, 0x0000000f, 0x038e0010, + 0x00000002, 0x33e70e00, 0x00000000, 0x28f30000, 0x00000010, 0x009f0000, + 0x00000000, 0x02043600, 0x00000010, 0x91840a02, 0x00000002, 0x21421800, + 0x00000008, 0x22000006, 0x00000000, 0x231b0000, 0x00000000, 0x23ff0000, + 0x00000000, 0x241b0000, 0x00000003, 0x8384a000, 0x0000000f, 0x65870010, + 0x00000009, 0x2607ffff, 0x00000000, 0x27110000, 0x00000000, 0x26900000, + 0x0000000c, 0x29000000, 0x00000018, 0x8000017e, 0x00000003, 0xf4683600, + 0x00000000, 0x3a100000, 0x00000000, 0x3a910000, 0x00000003, 0xf66c2400, + 0x00000010, 0x001f0000, 0x00000010, 0xb1923604, 0x00000008, 0x0f800004, + 0x00000000, 0x00000000, 0x00000010, 0x009f0000, 0x00000000, 0x3e170000, + 0x00000000, 0x3e940000, 0x00000000, 0x3f150000, 0x00000000, 0x3f960000, + 0x00000010, 0x001f0000, 0x00000000, 0x0f060000, 0x00000010, 0x20530000, + 0x00000000, 0x22c53600, 0x00000018, 0x8000ffac, 0x00000010, 0x001f0000, + 0x00000000, 0x031e0000, 0x00000000, 0x83850800, 0x00000009, 0x076000ff, + 0x0000000f, 0x0f0e0007, 0x00000000, 0x0c000000, 0x00000000, 0x0a7d0000, + 0x00000000, 0x0afe0000, 0x00000000, 0x0b7f0000, 0x00000000, 0x0d7a0000, + 0x00000000, 0x0bfc0000, 0x00000000, 0x0c970e00, 0x00000008, 0x0f800003, + 0x0000000f, 0x47670010, 0x00000008, 0x070e0001, 0x0000000b, 0xc38000ff, + 0x00000002, 0x43870000, 0x00000001, 0x33e70e00, 0x0000000f, 0x038e0010, + 0x00000002, 0x33e70e00, 0x00000000, 0x39840000, 0x00000003, 0xb9720800, + 0x00000000, 0x28f30000, 0x0000000f, 0x65680010, 0x00000010, 0x009f0000, + 0x00000000, 0x02043600, 0x00000010, 0x91840a02, 0x00000002, 0x21421800, + 0x00000008, 0x22000007, 0x00000000, 0x231b0000, 0x00000000, 0x23ff0000, + 0x00000000, 0x241b0000, 0x00000003, 0x8384a000, 0x0000000f, 0x65870010, + 0x00000009, 0x2607ffff, 0x00000000, 0x27110000, 0x00000000, 0x26900000, + 0x0000000c, 0x29000000, 0x00000018, 0x80000145, 0x00000003, 0xf4683600, + 0x00000000, 0x3a100000, 0x00000000, 0x3a910000, 0x00000003, 0xf66c2400, + 0x00000010, 0x001f0000, 0x00000010, 0xb1923604, 0x00000008, 0x0f800004, + 0x00000000, 0x00000000, 0x00000010, 0x009f0000, 0x00000000, 0x3e170000, + 0x00000000, 0x3e940000, 0x00000000, 0x3f150000, 0x00000000, 0x3f960000, + 0x00000010, 0x001f0000, 0x00000000, 0x0f060000, 0x00000010, 0x20530000, + 0x00000000, 0x22c53600, 0x00000018, 0x8000ff73, 0x00000010, 0x0ce70005, + 0x00000008, 0x2c80000c, 0x00000008, 0x2d000070, 0x00000008, 0x2d800010, + 0x00000000, 0x00000000, 0x00000010, 0x205f0000, 0x00000018, 0x8000011d, + 0x00000000, 0x2c1e0000, 0x00000008, 0x2c8000b8, 0x00000008, 0x2d000010, + 0x00000008, 0x2d800048, 0x00000000, 0x00000000, 0x00000010, 0x91de0000, + 0x00000018, 0x8000fe5d, 0x0000000c, 0x29800001, 0x00000000, 0x2a000000, + 0x00000010, 0x001f0000, 0x00000000, 0x0f008000, 0x00000008, 0x0f800007, + 0x00000018, 0x80000006, 0x0000000c, 0x29800001, 0x00000000, 0x2a000000, + 0x00000010, 0x001f0000, 0x0000000f, 0x0f470007, 0x00000008, 0x0f800008, + 0x00000018, 0x80000119, 0x00000010, 0x20530000, 0x00000018, 0x8000fe4f, + 0x0000000c, 0x29800001, 0x00000010, 0x91de0000, 0x00000000, 0x2fd50000, + 0x00000000, 0x2a000000, 0x00000009, 0x0261ffff, 0x0000000d, 0x70e10001, + 0x00000018, 0x80000101, 0x00000000, 0x2c400000, 0x00000008, 0x2c8000c4, + 0x00000008, 0x2d00001c, 0x00000008, 0x2d800001, 0x00000005, 0x70e10800, + 0x00000010, 0x91de0000, 0x00000018, 0x8000fe41, 0x0000000c, 0x29800001, + 0x00000010, 0x91de0000, 0x00000000, 0x2fd50000, 0x00000010, 0x001f0000, + 0x00000000, 0x02700000, 0x00000000, 0x0d620000, 0x00000000, 0xbb630800, + 0x00000000, 0x2a000000, 0x00000000, 0x0f400000, 0x00000000, 0x2c400000, + 0x0000000c, 0x73e7001b, 0x00000010, 0x0ce7000e, 0x00000000, 0x286d0000, + 0x0000000f, 0x65ed0010, 0x00000009, 0x266dffff, 0x00000018, 0x80000069, + 0x00000008, 0x02000004, 0x00000010, 0x91c40803, 0x00000018, 0x800000f6, + 0x00000010, 0x20530000, 0x00000018, 0x800000e5, 0x00000008, 0x2c8000b8, + 0x00000008, 0x2d000010, 0x00000008, 0x2d800048, 0x00000018, 0x80000005, + 0x00000008, 0x2c8000c4, 0x00000008, 0x2d00001c, 0x00000008, 0x2d800001, + 0x00000000, 0x00000000, 0x00000010, 0x205f0000, 0x00000008, 0x2c800048, + 0x00000008, 0x2d000068, 0x00000008, 0x2d800104, 0x00000000, 0x00000000, + 0x00000010, 0x91de0000, 0x00000000, 0x27f60000, 0x00000010, 0xb87a9e04, + 0x00000008, 0x2200000d, 0x00000018, 0x800000e2, 0x00000010, 0x20530000, + 0x00000018, 0x8000fe18, 0x0000000c, 0x29800001, 0x00000010, 0x91de0000, + 0x00000000, 0x2fd50000, 0x00000010, 0x001f0000, 0x00000000, 0x02700000, + 0x00000000, 0x0d620000, 0x00000000, 0xbb630800, 0x00000000, 0x2a000000, + 0x00000010, 0x0e670011, 0x00000000, 0x286d0000, 0x0000000f, 0x65ed0010, + 0x00000009, 0x266dffff, 0x00000004, 0xb8f1a000, 0x00000000, 0x0f400000, + 0x0000000c, 0x73e7001c, 0x00000018, 0x80000040, 0x00000008, 0x02000004, + 0x00000010, 0x91c40802, 0x00000018, 0x800000cd, 0x00000000, 0x2c1e0000, + 0x00000008, 0x2c8000b8, 0x00000008, 0x2d000010, 0x00000008, 0x2d800048, + 0x00000010, 0x20530000, 0x00000010, 0x91de0000, 0x00000018, 0x8000fdfe, + 0x0000000c, 0x29800001, 0x00000000, 0x03550000, 0x00000000, 0x06460000, + 0x00000000, 0x03d60000, 0x00000000, 0x2a000000, 0x0000000f, 0x0f480007, + 0x00000010, 0xb18c0027, 0x0000000f, 0x47420008, 0x00000009, 0x070e000f, + 0x00000008, 0x070e0008, 0x00000010, 0x001f0000, 0x00000008, 0x09000001, + 0x00000007, 0x09121c00, 0x00000003, 0xcbca9200, 0x00000000, 0x0b97a200, + 0x00000007, 0x4b171c00, 0x0000000f, 0x0a960003, 0x00000000, 0x0a959c00, + 0x00000000, 0x4a009a00, 0x00000008, 0x82120001, 0x00000001, 0x0c170800, + 0x00000000, 0x02180000, 0x00000000, 0x0c971800, 0x00000008, 0x0d00ffff, + 0x00000008, 0x0f800006, 0x0000000c, 0x29000000, 0x00000008, 0x22000001, + 0x00000000, 0x22c50c00, 0x00000010, 0x009f0000, 0x00000010, 0xb197320b, + 0x00000000, 0x231b0000, 0x00000000, 0x27110800, 0x00000000, 0x66900000, + 0x00000018, 0x800000a4, 0x00000000, 0x02180000, 0x00000010, 0x20530000, + 0x00000000, 0x22c53600, 0x00000010, 0x001f0000, 0x00000008, 0x0f800006, + 0x00000018, 0x8000fff5, 0x00000010, 0x91870002, 0x00000008, 0x2200000a, + 0x00000000, 0x231b0000, 0x00000000, 0x27110800, 0x00000000, 0x66900000, + 0x00000018, 0x80000098, 0x00000008, 0x0200000a, 0x00000010, 0x91c40804, + 0x00000010, 0x02c20003, 0x00000010, 0x001f0000, 0x00000008, 0x0f800008, + 0x00000010, 0x20530000, 0x00000018, 0x8000fdc9, 0x00000000, 0x06820000, + 0x00000010, 0x001f0000, 0x00000010, 0x0ce70028, 0x00000000, 0x03720000, + 0x00000000, 0xa8760c00, 0x00000000, 0x0cf60000, 0x00000010, 0xb8723224, + 0x00000000, 0x03440000, 0x00000008, 0x22000010, 0x00000000, 0x03ca0000, + 0x0000000f, 0x65680010, 0x00000000, 0x0bcf0000, 0x00000000, 0x27f20000, + 0x00000010, 0xb7ef3203, 0x0000000c, 0x21420004, 0x0000000c, 0x73e70019, + 0x00000000, 0x07520000, 0x00000000, 0x29000000, 0x00000018, 0x8000007e, + 0x00000004, 0xb9723200, 0x00000010, 0x20530000, 0x00000000, 0x22060000, + 0x0000000c, 0x61420004, 0x00000000, 0x25070000, 0x00000000, 0x27970000, + 0x00000000, 0x290e0000, 0x00000010, 0x0ce70010, 0x00000010, 0xb873320f, + 0x0000000f, 0x436c0010, 0x00000000, 0x03f30c00, 0x00000000, 0x03f30000, + 0x00000000, 0x83990e00, 0x00000001, 0x83860e00, 0x00000000, 0x83060e00, + 0x00000003, 0xf66c0c00, 0x00000000, 0x39f30e00, 0x00000000, 0x3af50e00, + 0x00000000, 0x7a740000, 0x0000000f, 0x43680010, 0x00000001, 0x83860e00, + 0x00000000, 0x83060e00, 0x00000003, 0xf4680c00, 0x00000000, 0x286d0000, + 0x00000000, 0x03690000, 0x00000010, 0xb1f60c54, 0x00000000, 0x0a6a0000, + 0x00000000, 0x0aeb0000, 0x00000009, 0x0b6cffff, 0x00000000, 0x0c000000, + 0x00000000, 0x0be90000, 0x00000003, 0x8cf6a000, 0x0000000c, 0x09800002, + 0x00000010, 0x009f0000, 0x00000010, 0xb8173209, 0x00000000, 0x35140000, + 0x00000000, 0x35950000, 0x00000005, 0x766c2c00, 0x00000000, 0x34970000, + 0x00000004, 0xb8f12e00, 0x00000010, 0x001f0000, 0x00000008, 0x0f800004, + 0x00000018, 0x8000fff7, 0x00000000, 0x03e90000, 0x00000010, 0xb8f6a01a, + 0x00000010, 0x20130019, 0x00000010, 0xb1f10e18, 0x00000000, 0x83973200, + 0x00000000, 0x38700e00, 0x00000000, 0xbb760e00, 0x00000000, 0x37d00000, + 0x0000000c, 0x73e7001a, 0x00000003, 0xb8f1a000, 0x00000000, 0x32140000, + 0x00000000, 0x32950000, 0x00000005, 0x73e72c00, 0x00000000, 0x33190000, + 0x00000005, 0x74680000, 0x00000010, 0x0ce7000d, 0x00000008, 0x22000009, + 0x00000000, 0x07520000, 0x00000000, 0x29000000, 0x0000000c, 0x73e70019, + 0x0000000f, 0x65680010, 0x0000000c, 0x21420004, 0x00000018, 0x8000003c, + 0x00000010, 0x20530000, 0x0000000c, 0x61420004, 0x00000000, 0x290e0000, + 0x00000018, 0x80000002, 0x00000010, 0x91973206, 0x00000000, 0x35140000, + 0x00000000, 0x35950000, 0x00000005, 0x766c2c00, 0x00000000, 0x34990000, + 0x00000004, 0xb8f13200, 0x00000000, 0x83690c00, 0x00000010, 0xb1860013, + 0x00000000, 0x28e90000, 0x00000008, 0x22000004, 0x00000000, 0x23ec0000, + 0x00000000, 0x03690000, 0x00000010, 0xb8660c07, 0x00000009, 0x036cffff, + 0x00000000, 0x326a0000, 0x00000000, 0x32eb0000, 0x00000005, 0x73e70c00, + 0x00000000, 0x33690000, 0x00000005, 0x74680000, 0x0000000c, 0x73e7001c, + 0x00000000, 0x03690000, 0x00000010, 0xb1f60c12, 0x00000010, 0xb1d00c11, + 0x0000000c, 0x21420005, 0x0000000c, 0x33e7001c, 0x00000018, 0x8000000e, + 0x00000010, 0x2e67000d, 0x00000000, 0x03690000, 0x00000010, 0xb1f60c0b, + 0x00000010, 0xb1d00c0a, 0x00000000, 0x03440000, 0x00000008, 0x2200000c, + 0x00000000, 0x07520000, 0x00000000, 0x29000000, 0x00000018, 0x80000015, + 0x0000000c, 0x33e7001c, 0x00000010, 0x20530000, 0x00000000, 0x22060000, + 0x00000000, 0x290e0000, 0x00000018, 0x000d0000, 0x00000000, 0x06820000, + 0x00000010, 0x2de7000d, 0x00000010, 0x0ce7000c, 0x00000000, 0x27f20000, + 0x00000010, 0xb96d9e0a, 0x00000000, 0xa86d9e00, 0x00000009, 0x0361ffff, + 0x00000010, 0xb7500c07, 0x00000008, 0x2200000f, 0x0000000f, 0x65680010, + 0x00000000, 0x29000000, 0x00000018, 0x80000004, 0x0000000c, 0x33e7001b, + 0x00000010, 0x20530000, 0x00000018, 0x000d0000, 0x00000000, 0x2b820000, + 0x00000010, 0x20d2002f, 0x00000010, 0x0052002e, 0x00000009, 0x054e0007, + 0x00000010, 0xb18a002c, 0x00000000, 0x050a8c00, 0x00000008, 0x850a0008, + 0x00000010, 0x918a0029, 0x00000003, 0xc5008800, 0x00000008, 0xa3460001, + 0x00000010, 0xb1c60007, 0x00000008, 0x22000001, 0x0000000c, 0x29800000, + 0x00000010, 0x20530000, 0x00000000, 0x274e8c00, 0x00000000, 0x66cd0000, + 0x00000000, 0x22c58c00, 0x00000008, 0x22000014, 0x00000003, 0x22c58e00, + 0x00000003, 0x23c58e00, 0x00000003, 0x22c58e00, 0x00000003, 0x26cd9e00, + 0x00000003, 0x27cd9e00, 0x00000003, 0x26cd9e00, 0x00000003, 0x274ea000, + 0x00000003, 0x284ea000, 0x00000003, 0x274ea000, 0x0000000c, 0x69520000, + 0x0000000c, 0x29800000, 0x00000010, 0x20530000, 0x00000003, 0x22c58e00, + 0x00000003, 0x23c58e00, 0x00000003, 0x22c58e00, 0x00000003, 0x26cd9e00, + 0x00000003, 0x27cd9e00, 0x00000003, 0x26cd9e00, 0x00000003, 0x274ea000, + 0x00000003, 0x284ea000, 0x00000003, 0x274ea000, 0x00000000, 0xa2c58c00, + 0x00000000, 0xa74e8c00, 0x00000000, 0xe6cd0000, 0x0000000f, 0x620a0010, + 0x00000008, 0x23460001, 0x0000000c, 0x29800000, 0x00000010, 0x20530000, + 0x0000000c, 0x29520000, 0x00000018, 0x80000002, 0x0000000c, 0x29800000, + 0x00000018, 0x00570000 }; + +static int bnx2_TPAT_b06FwReleaseMajor = 0x0; +static int bnx2_TPAT_b06FwReleaseMinor = 0x0; +static int bnx2_TPAT_b06FwReleaseFix = 0x0; +static u32 bnx2_TPAT_b06FwStartAddr = 0x08000858; +static u32 bnx2_TPAT_b06FwTextAddr = 0x08000800; +static int bnx2_TPAT_b06FwTextLen = 0x1314; +static u32 bnx2_TPAT_b06FwDataAddr = 0x08001b40; +static int bnx2_TPAT_b06FwDataLen = 0x0; +static u32 bnx2_TPAT_b06FwRodataAddr = 0x00000000; +static int bnx2_TPAT_b06FwRodataLen = 0x0; +static u32 bnx2_TPAT_b06FwBssAddr = 0x08001b90; +static int bnx2_TPAT_b06FwBssLen = 0x80; +static u32 bnx2_TPAT_b06FwSbssAddr = 0x08001b40; +static int bnx2_TPAT_b06FwSbssLen = 0x48; + +static u32 bnx2_TPAT_b06FwText[(0x1314/4) + 1] = { + 0x0a000216, 0x00000000, 0x00000000, 0x0000000d, 0x74706174, 0x20302e36, + 0x2e390000, 0x00060901, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x10000003, + 0x00000000, 0x0000000d, 0x0000000d, 0x3c020800, 0x24421b40, 0x3c030800, + 0x24631c10, 0xac400000, 0x0043202b, 0x1480fffd, 0x24420004, 0x3c1d0800, + 0x37bd3ffc, 0x03a0f021, 0x3c100800, 0x26100858, 0x3c1c0800, 0x279c1b40, + 0x0e00051f, 0x00000000, 0x0000000d, 0x8f820024, 0x27bdffe8, 0xafbf0014, + 0x10400004, 0xafb00010, 0x0000000d, 0x00000000, 0x2400015f, 0x8f82001c, + 0x8c450008, 0x24030800, 0xaf430178, 0x97430104, 0x3c020008, 0xaf420140, + 0x8f820034, 0x30420001, 0x10400006, 0x3070ffff, 0x24020002, 0x2603fffe, + 0xa7420146, 0x0a000246, 0xa7430148, 0xa7400146, 0x8f850034, 0x30a20020, + 0x0002102b, 0x00021023, 0x30460009, 0x30a30c00, 0x24020400, 0x14620002, + 0x34c40001, 0x34c40005, 0xa744014a, 0x3c020800, 0x8c440820, 0x3c030048, + 0x24020002, 0x00832025, 0x30a30006, 0x1062000d, 0x2c620003, 0x50400005, + 0x24020004, 0x10600012, 0x3c020001, 0x0a000271, 0x00000000, 0x10620007, + 0x24020006, 0x1462000f, 0x3c020111, 0x0a000269, 0x00821025, 0x0a000268, + 0x3c020101, 0x3c020011, 0x00821025, 0x24030001, 0xaf421000, 0xaf830030, + 0x0a000271, 0x00000000, 0x00821025, 0xaf421000, 0xaf800030, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x8f830030, 0x1060003f, 0x3c048000, + 0x8f421000, 0x00441024, 0x1040fffd, 0x00000000, 0x10600039, 0x00000000, + 0x8f421000, 0x3c030020, 0x00431024, 0x10400034, 0x00000000, 0x97421014, + 0x14400031, 0x00000000, 0x97421008, 0x8f84001c, 0x24420006, 0x00024082, + 0x00081880, 0x00643821, 0x8ce50000, 0x30430003, 0x30420001, 0x10400004, + 0x00000000, 0x0000000d, 0x0a0002b0, 0x00081080, 0x5460000f, 0x30a5ffff, + 0x3c06ffff, 0x00a62824, 0x0005182b, 0x00a61026, 0x0002102b, 0x00621824, + 0x10600004, 0x00000000, 0x0000000d, 0x00000000, 0x240001fc, 0x8ce20000, + 0x0a0002af, 0x00462825, 0x0005182b, 0x38a2ffff, 0x0002102b, 0x00621824, + 0x10600004, 0x00000000, 0x0000000d, 0x00000000, 0x24000206, 0x8ce20000, + 0x3445ffff, 0x00081080, 0x00441021, 0x3c030800, 0xac450000, 0x8c620840, + 0x24420001, 0xac620840, 0x8f820008, 0x10400003, 0x00000000, 0x0e000660, + 0x00000000, 0x8f840028, 0x02002821, 0x24820008, 0x30421fff, 0x24434000, + 0x0343d821, 0x30a30007, 0xaf840018, 0xaf820028, 0xaf420084, 0x10600002, + 0x24a20007, 0x3045fff8, 0x8f820044, 0x8f840004, 0x00451821, 0xaf82002c, + 0x0064102b, 0xaf830044, 0x14400002, 0x00641023, 0xaf820044, 0x8f840044, + 0x34028000, 0x8fbf0014, 0x8fb00010, 0x00821021, 0x03421821, 0x3c021000, + 0xaf83001c, 0xaf440080, 0xaf420178, 0x03e00008, 0x27bd0018, 0x8f820024, + 0x27bdffe8, 0xafbf0014, 0x10400004, 0xafb00010, 0x0000000d, 0x00000000, + 0x24000249, 0x8f85001c, 0x24020001, 0xaf820024, 0x8ca70008, 0xa3800023, + 0x8f620004, 0x3c100800, 0x26041b90, 0x00021402, 0xa3820010, 0x304600ff, + 0x24c60005, 0x0e00064a, 0x00063082, 0x8f640004, 0x8f430108, 0x3c021000, + 0x00621824, 0xa7840020, 0x10600008, 0x00000000, 0x97420104, 0x93830023, + 0x2442ffec, 0x34630002, 0xa3830023, 0x0a000304, 0x3045ffff, 0x97420104, + 0x2442fff0, 0x3045ffff, 0x8f620004, 0x3042ffff, 0x2c420013, 0x14400004, + 0x00000000, 0x93820023, 0x34420001, 0xa3820023, 0x93830023, 0x24020001, + 0x10620009, 0x28620002, 0x14400014, 0x24020002, 0x10620012, 0x24020003, + 0x1062000a, 0x00000000, 0x0a000325, 0x00000000, 0x8f82001c, 0x8c43000c, + 0x3c04ffff, 0x00641824, 0x00651825, 0x0a000325, 0xac43000c, 0x8f82001c, + 0x8c430010, 0x3c04ffff, 0x00641824, 0x00651825, 0xac430010, 0x8f620004, + 0x3042ffff, 0x24420002, 0x00021083, 0xa3820038, 0x304500ff, 0x8f82001c, + 0x3c04ffff, 0x00052880, 0x00a22821, 0x8ca70000, 0x97820020, 0x97430104, + 0x00e42024, 0x24420002, 0x00621823, 0x00833825, 0xaca70000, 0x93840038, + 0x26061b90, 0x00041080, 0x00461021, 0x90430000, 0x3063000f, 0x00832021, + 0xa3840022, 0x308200ff, 0x3c04fff6, 0x24420003, 0x00021080, 0x00461021, + 0x8c450000, 0x93830022, 0x8f82001c, 0x3484ffff, 0x00a43824, 0x00031880, + 0x00621821, 0xaf850000, 0xac67000c, 0x93820022, 0x93830022, 0x8f84001c, + 0x24420003, 0x00021080, 0x00461021, 0x24630004, 0x00031880, 0xac470000, + 0x93820022, 0x00661821, 0x94670002, 0x00021080, 0x00441021, 0xac670000, + 0x24030010, 0xac470010, 0xa7430140, 0x24030002, 0xa7400142, 0xa7400144, + 0xa7430146, 0x97420104, 0x8f840034, 0x24030001, 0x2442fffe, 0x30840006, + 0xa7420148, 0x24020002, 0xa743014a, 0x1082000d, 0x2c820003, 0x10400005, + 0x24020004, 0x10800011, 0x3c020009, 0x0a000383, 0x00000000, 0x10820007, + 0x24020006, 0x1482000d, 0x3c020119, 0x0a00037d, 0x24030001, 0x0a00037c, + 0x3c020109, 0x3c020019, 0x24030001, 0xaf421000, 0xaf830030, 0x0a000383, + 0x00000000, 0xaf421000, 0xaf800030, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x93820010, 0x24030008, 0x8f840030, 0x24420002, 0x30420007, + 0x00621823, 0x30630007, 0xaf83000c, 0x10800005, 0x3c038000, 0x8f421000, + 0x00431024, 0x1040fffd, 0x00000000, 0x8f820028, 0xaf820018, 0x24420010, + 0x30421fff, 0xaf820028, 0xaf420084, 0x97430104, 0x24424000, 0x0342d821, + 0x3063ffff, 0x30620007, 0x10400002, 0x24620007, 0x3043fff8, 0x8f820044, + 0x8f840004, 0x00431821, 0xaf82002c, 0x0064102b, 0xaf830044, 0x14400002, + 0x00641023, 0xaf820044, 0x8f840044, 0x34028000, 0x8fbf0014, 0x8fb00010, + 0x00821021, 0x03421821, 0x3c021000, 0xaf83001c, 0xaf440080, 0xaf420178, + 0x03e00008, 0x27bd0018, 0x8f820024, 0x27bdffe8, 0xafbf0014, 0x14400004, + 0xafb00010, 0x0000000d, 0x00000000, 0x240002db, 0x8f620004, 0x04410009, + 0x3c050800, 0x93820022, 0x8f830000, 0x24a41b90, 0xaf800024, 0x24420003, + 0x00021080, 0x00441021, 0xac430000, 0x93820038, 0x24a51b90, 0x93860010, + 0x3c040001, 0x27700008, 0x24420001, 0x00021080, 0x00451021, 0x8c430000, + 0x24c60005, 0x00063082, 0x00641821, 0x02002021, 0x0e00064a, 0xac430000, + 0x93840022, 0x3c057fff, 0x8f620004, 0x00042080, 0x00902021, 0x8c830004, + 0x34a5ffff, 0x00451024, 0x00621821, 0xac830004, 0x93850038, 0x3c07ffff, + 0x93840010, 0x00052880, 0x00b02821, 0x8ca30000, 0x97420104, 0x97860020, + 0x00671824, 0x00441021, 0x00461023, 0x3042ffff, 0x00621825, 0xaca30000, + 0x93830023, 0x24020001, 0x10620009, 0x28620002, 0x1440001a, 0x24020002, + 0x10620018, 0x24020003, 0x1062000d, 0x00000000, 0x0a000411, 0x00000000, + 0x93820010, 0x97430104, 0x8e04000c, 0x00621821, 0x2463fff2, 0x3063ffff, + 0x00872024, 0x00832025, 0x0a000411, 0xae04000c, 0x93820010, 0x97430104, + 0x8e040010, 0x00621821, 0x2463ffee, 0x3063ffff, 0x00872024, 0x00832025, + 0xae040010, 0x9783000e, 0x8f840034, 0x2402000a, 0xa7420140, 0xa7430142, + 0x93820010, 0xa7420144, 0xa7400146, 0x97430104, 0x30840006, 0x24020001, + 0xa7430148, 0xa742014a, 0x24020002, 0x1082000d, 0x2c820003, 0x10400005, + 0x24020004, 0x10800011, 0x3c020041, 0x0a000437, 0x00000000, 0x10820007, + 0x24020006, 0x1482000d, 0x3c020151, 0x0a000431, 0x24030001, 0x0a000430, + 0x3c020141, 0x3c020051, 0x24030001, 0xaf421000, 0xaf830030, 0x0a000437, + 0x00000000, 0xaf421000, 0xaf800030, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x8f820030, 0x93840010, 0x8f850028, 0x10400005, 0x3c038000, + 0x8f421000, 0x00431024, 0x1040fffd, 0x00000000, 0x2483000a, 0x30620007, + 0x10400002, 0x24620007, 0x304303f8, 0x00a31021, 0x30421fff, 0xaf850018, + 0xaf820028, 0xaf420084, 0x97430104, 0x24424000, 0x0342d821, 0x3063ffff, + 0x30620007, 0x10400002, 0x24620007, 0x3043fff8, 0x8f820044, 0x8f840004, + 0x00431821, 0xaf82002c, 0x0064102b, 0xaf830044, 0x14400002, 0x00641023, + 0xaf820044, 0x8f840044, 0x34028000, 0x8fbf0014, 0x8fb00010, 0x00821021, + 0x03421821, 0x3c021000, 0xaf83001c, 0xaf440080, 0xaf420178, 0x03e00008, + 0x27bd0018, 0x3c026000, 0x8c444448, 0x3c030800, 0xac64082c, 0x8f620000, + 0x97430104, 0x3c048000, 0x3046ffff, 0x3067ffff, 0x8f420178, 0x00441024, + 0x1440fffd, 0x2402000a, 0x30c30007, 0xa7420140, 0x24020008, 0x00431023, + 0x30420007, 0x24c3fffe, 0xa7420142, 0xa7430144, 0xa7400146, 0xa7470148, + 0x8f420108, 0x3c036000, 0x8f850034, 0x30420020, 0x0002102b, 0x00021023, + 0x30420009, 0x34420001, 0xa742014a, 0x8c644448, 0x3c020800, 0x30a50006, + 0xac440830, 0x24020002, 0x10a2000d, 0x2ca20003, 0x10400005, 0x24020004, + 0x10a00011, 0x3c020041, 0x0a0004a8, 0x00000000, 0x10a20007, 0x24020006, + 0x14a2000d, 0x3c020151, 0x0a0004a2, 0x24030001, 0x0a0004a1, 0x3c020141, + 0x3c020051, 0x24030001, 0xaf421000, 0xaf830030, 0x0a0004a8, 0x00000000, + 0xaf421000, 0xaf800030, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x8f820030, 0x24c30008, 0x10400006, 0x30e6ffff, 0x3c048000, 0x8f421000, + 0x00441024, 0x1040fffd, 0x00000000, 0x3c026000, 0x8c444448, 0x3065ffff, + 0x3c020800, 0x30a30007, 0x10600003, 0xac440834, 0x24a20007, 0x3045fff8, + 0x8f840028, 0x00851021, 0x30421fff, 0x24434000, 0x0343d821, 0x30c30007, + 0xaf840018, 0xaf820028, 0xaf420084, 0x10600002, 0x24c20007, 0x3046fff8, + 0x8f820044, 0x8f840004, 0x00461821, 0xaf82002c, 0x0064102b, 0xaf830044, + 0x14400002, 0x00641023, 0xaf820044, 0x8f840044, 0x34028000, 0x3c030800, + 0x8c650844, 0x00821021, 0x03421821, 0xaf83001c, 0xaf440080, 0x10a00006, + 0x2402000e, 0x93830043, 0x14620004, 0x3c021000, 0x2402043f, 0xa7420148, + 0x3c021000, 0x3c036000, 0xaf420178, 0x8c644448, 0x3c020800, 0x03e00008, + 0xac440838, 0x8f820034, 0x30424000, 0x10400005, 0x24020800, 0x0000000d, + 0x00000000, 0x24000405, 0x24020800, 0xaf420178, 0x97440104, 0x3c030008, + 0xaf430140, 0x8f820034, 0x30420001, 0x10400006, 0x3085ffff, 0x24020002, + 0x24a3fffe, 0xa7420146, 0x0a0004ff, 0xa7430148, 0xa7400146, 0x8f840028, + 0x2402000d, 0xa742014a, 0x24830008, 0x30631fff, 0x24624000, 0x0342d821, + 0x30a20007, 0xaf840018, 0xaf830028, 0xaf430084, 0x10400002, 0x24a20007, + 0x3045fff8, 0x8f820044, 0x8f840004, 0x00451821, 0xaf82002c, 0x0064102b, + 0xaf830044, 0x14400002, 0x00641023, 0xaf820044, 0x8f840044, 0x34028000, + 0x00821021, 0x03421821, 0x3c021000, 0xaf83001c, 0xaf440080, 0x03e00008, + 0xaf420178, 0x27bdffe8, 0x3c046008, 0xafbf0014, 0xafb00010, 0x8c825000, + 0x3c1a8000, 0x2403ff7f, 0x375b4000, 0x00431024, 0x3442380c, 0xac825000, + 0x8f430008, 0x3c100800, 0x37428000, 0x34630001, 0xaf430008, 0xaf82001c, + 0x3c02601c, 0xaf800028, 0xaf400080, 0xaf400084, 0x8c450008, 0x3c036000, + 0x8c620808, 0x3c040800, 0x3c030080, 0xac830820, 0x3042fff0, 0x38420010, + 0x2c420001, 0xaf850004, 0xaf820008, 0x0e00062f, 0x00000000, 0x8f420000, + 0x30420001, 0x1040fffb, 0x00000000, 0x8f440108, 0x30822000, 0xaf840034, + 0x10400004, 0x8e02083c, 0x24420001, 0x0a00059d, 0xae02083c, 0x30820200, + 0x10400027, 0x00000000, 0x97420104, 0x1040001c, 0x30824000, 0x14400005, + 0x00000000, 0x0e00022d, 0x00000000, 0x0a000592, 0x00000000, 0x8f620008, + 0x8f630000, 0x24020030, 0x00031e02, 0x306300f0, 0x10620007, 0x28620031, + 0x14400031, 0x24020040, 0x10620007, 0x00000000, 0x0a000592, 0x00000000, + 0x0e0002dd, 0x00000000, 0x0a000592, 0x00000000, 0x0e0003b8, 0x00000000, + 0x0a000592, 0x00000000, 0x30820040, 0x1440002d, 0x00000000, 0x0000000d, + 0x00000000, 0x240004a6, 0x0a00059d, 0x00000000, 0x8f430100, 0x24020d00, + 0x1462000f, 0x30820006, 0x97420104, 0x10400005, 0x30820040, 0x0e0004e9, + 0x00000000, 0x0a000592, 0x00000000, 0x1440001b, 0x00000000, 0x0000000d, + 0x00000000, 0x240004b8, 0x0a00059d, 0x00000000, 0x1040000e, 0x30821000, + 0x10400005, 0x00000000, 0x0e00065d, 0x00000000, 0x0a000592, 0x00000000, + 0x0e00046b, 0x00000000, 0x8f820040, 0x24420001, 0xaf820040, 0x0a00059d, + 0x00000000, 0x30820040, 0x14400004, 0x00000000, 0x0000000d, 0x00000000, + 0x240004cf, 0x8f420138, 0x3c034000, 0x00431025, 0xaf420138, 0x0a00053f, + 0x00000000, 0x3c046008, 0x8c835000, 0x3c1a8000, 0x2402ff7f, 0x375b4000, + 0x00621824, 0x3463380c, 0xac835000, 0x8f420008, 0x3c056000, 0x3c03601c, + 0x34420001, 0xaf420008, 0x37428000, 0xaf800028, 0xaf82001c, 0xaf400080, + 0xaf400084, 0x8c660008, 0x8ca20808, 0x3c040800, 0x3c030080, 0xac830820, + 0x3042fff0, 0x38420010, 0x2c420001, 0xaf860004, 0xaf820008, 0x03e00008, + 0x00000000, 0x3084ffff, 0x30820007, 0x10400002, 0x24820007, 0x3044fff8, + 0x8f820028, 0x00441821, 0x30631fff, 0x24644000, 0x0344d821, 0xaf820018, + 0xaf830028, 0x03e00008, 0xaf430084, 0x3084ffff, 0x30820007, 0x10400002, + 0x24820007, 0x3044fff8, 0x8f820044, 0x8f830004, 0x00442021, 0xaf82002c, + 0x0083102b, 0xaf840044, 0x14400002, 0x00831023, 0xaf820044, 0x8f820044, + 0x34038000, 0x00431821, 0x03432021, 0xaf84001c, 0x03e00008, 0xaf420080, + 0x8f830034, 0x24020002, 0x30630006, 0x1062000d, 0x2c620003, 0x50400005, + 0x24020004, 0x10600012, 0x3c020001, 0x0a000601, 0x00000000, 0x10620007, + 0x24020006, 0x1462000f, 0x3c020111, 0x0a0005f9, 0x00821025, 0x0a0005f8, + 0x3c020101, 0x3c020011, 0x00821025, 0x24030001, 0xaf421000, 0xaf830030, + 0x0a000601, 0x00000000, 0x00821025, 0xaf421000, 0xaf800030, 0x00000000, + 0x00000000, 0x00000000, 0x03e00008, 0x00000000, 0x8f820030, 0x10400005, + 0x3c038000, 0x8f421000, 0x00431024, 0x1040fffd, 0x00000000, 0x03e00008, + 0x00000000, 0x8f820034, 0x27bdffe8, 0x30424000, 0x14400005, 0xafbf0010, + 0x0e00022d, 0x00000000, 0x0a00062d, 0x8fbf0010, 0x8f620008, 0x8f630000, + 0x24020030, 0x00031e02, 0x306300f0, 0x10620008, 0x28620031, 0x1440000d, + 0x8fbf0010, 0x24020040, 0x10620007, 0x00000000, 0x0a00062d, 0x00000000, + 0x0e0002dd, 0x00000000, 0x0a00062d, 0x8fbf0010, 0x0e0003b8, 0x00000000, + 0x8fbf0010, 0x03e00008, 0x27bd0018, 0x8f84003c, 0x1080000f, 0x3c026000, + 0x8c430c3c, 0x30630fff, 0xaf830014, 0x14600011, 0x3082000f, 0x10400005, + 0x308200f0, 0x10400003, 0x30820f00, 0x14400006, 0x00000000, 0x0000000d, + 0x00000000, 0x2400050e, 0x03e00008, 0x00000000, 0x0000000d, 0x00000000, + 0x24000513, 0x03e00008, 0x00000000, 0xaf83003c, 0x03e00008, 0x00000000, + 0x10c00007, 0x00000000, 0x8ca20000, 0x24c6ffff, 0x24a50004, 0xac820000, + 0x14c0fffb, 0x24840004, 0x03e00008, 0x00000000, 0x0a000659, 0x00a01021, + 0xac860000, 0x24840004, 0x00a01021, 0x1440fffc, 0x24a5ffff, 0x03e00008, + 0x00000000, 0x0000000d, 0x03e00008, 0x00000000, 0x3c040800, 0x8c82084c, + 0x54400007, 0xac80084c, 0x8f820034, 0x24030400, 0x30420c00, 0x1443005b, + 0x00000000, 0xac80084c, 0x0000000d, 0x00000000, 0x2400003c, 0x3c026000, + 0x8c444448, 0x3c030800, 0xac640850, 0x24000043, 0x97420104, 0x3045ffff, + 0x000530c2, 0x24a2007f, 0x000239c2, 0x2400004e, 0x3c046020, 0x24030020, + 0xac830000, 0x8c820000, 0x30420020, 0x10400005, 0x3c036020, 0x8c620000, + 0x30420020, 0x1440fffd, 0x00000000, 0x3c026020, 0x8c430010, 0x24040001, + 0x0087102b, 0x30ea007f, 0x24abfffe, 0x10400010, 0x00034240, 0x3c056020, + 0x24090020, 0xaca90000, 0x8ca20000, 0x30420020, 0x10400006, 0x24840001, + 0x3c036020, 0x8c620000, 0x30420020, 0x1440fffd, 0x00000000, 0x0087102b, + 0x1440fff4, 0x00000000, 0x8f85001c, 0x3c026020, 0x8c430010, 0x3c046020, + 0x34848000, 0x006a1825, 0x01034025, 0x2400006b, 0x10c0000b, 0x00000000, + 0x8ca30000, 0x24a50004, 0x8ca20000, 0x24a50004, 0x24c6ffff, 0xac820000, + 0x24840004, 0xac830000, 0x14c0fff7, 0x24840004, 0x24000077, 0x3c020007, + 0x34427700, 0x3c036000, 0xac6223c8, 0xac6b23cc, 0xac6823e4, 0x24000086, + 0x3c046000, 0x3c038000, 0x8c8223f8, 0x00431024, 0x1440fffd, 0x3c021000, + 0x3c056000, 0x24030019, 0xaca223f8, 0xa743014a, 0x8ca44448, 0x3c020800, + 0xac440854, 0x03e00008, 0x00000000, 0x00000000 }; + +static u32 bnx2_TPAT_b06FwData[(0x0/4) + 1] = { 0x00000000 }; +static u32 bnx2_TPAT_b06FwRodata[(0x0/4) + 1] = { 0x00000000 }; +static u32 bnx2_TPAT_b06FwBss[(0x80/4) + 1] = { 0x00000000 }; +static u32 bnx2_TPAT_b06FwSbss[(0x48/4) + 1] = { 0x00000000 }; + +static int bnx2_TXP_b06FwReleaseMajor = 0x0; +static int bnx2_TXP_b06FwReleaseMinor = 0x0; +static int bnx2_TXP_b06FwReleaseFix = 0x0; +static u32 bnx2_TXP_b06FwStartAddr = 0x08002090; +static u32 bnx2_TXP_b06FwTextAddr = 0x08000000; +static int bnx2_TXP_b06FwTextLen = 0x3ffc; +static u32 bnx2_TXP_b06FwDataAddr = 0x08004020; +static int bnx2_TXP_b06FwDataLen = 0x0; +static u32 bnx2_TXP_b06FwRodataAddr = 0x00000000; +static int bnx2_TXP_b06FwRodataLen = 0x0; +static u32 bnx2_TXP_b06FwBssAddr = 0x08004060; +static int bnx2_TXP_b06FwBssLen = 0x194; +static u32 bnx2_TXP_b06FwSbssAddr = 0x08004020; +static int bnx2_TXP_b06FwSbssLen = 0x34; +static u32 bnx2_TXP_b06FwText[(0x3ffc/4) + 1] = { + 0x0a000824, 0x00000000, 0x00000000, 0x0000000d, 0x74787020, 0x302e362e, + 0x39000000, 0x00060900, 0x0000000a, 0x000003e8, 0x0000ea60, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x10000003, 0x00000000, 0x0000000d, + 0x0000000d, 0x3c020800, 0x24424020, 0x3c030800, 0x246341f4, 0xac400000, + 0x0043202b, 0x1480fffd, 0x24420004, 0x3c1d0800, 0x37bd7ffc, 0x03a0f021, + 0x3c100800, 0x26102090, 0x3c1c0800, 0x279c4020, 0x0e000a0e, 0x00000000, + 0x0000000d, 0x8f840014, 0x27bdffe8, 0xafb00010, 0x8f460104, 0x8f830008, + 0x8c8500ac, 0xaf430080, 0x948200a8, 0xa7420e10, 0x948300aa, 0xa7430e12, + 0x8c8200ac, 0xaf420e18, 0x97430e10, 0xa7430e14, 0x97420e12, 0xa7420e16, + 0x8f430e18, 0x00005021, 0x00c53023, 0x10c001a3, 0xaf430e1c, 0x240f0800, + 0x3c0e1000, 0x2419fff8, 0x24100010, 0x3c188100, 0x93620008, 0x10400009, + 0x00000000, 0x97620010, 0x00c2102b, 0x14400005, 0x00000000, 0x97620010, + 0x3042ffff, 0x0a000862, 0xaf420e00, 0xaf460e00, 0x8f420000, 0x30420008, + 0x1040fffd, 0x00000000, 0x97420e08, 0x8f450e04, 0x3044ffff, 0x30820001, + 0x14400005, 0x00000000, 0x14a00005, 0x3083a040, 0x0a0009e6, 0x00000000, + 0x0000000d, 0x3083a040, 0x24020040, 0x14620049, 0x3082a000, 0x8f87000c, + 0x30880036, 0x30890008, 0xaf4f0178, 0x00e01821, 0x9742008a, 0x00431023, + 0x2442ffff, 0x30421fff, 0x2c420008, 0x1440fffa, 0x00000000, 0x8f830018, + 0x00a05021, 0x00c53023, 0x24e24000, 0x03422821, 0x306b00ff, 0x24630001, + 0xaf830018, 0x93840012, 0x000b1400, 0x3c030100, 0x00431025, 0xaca20000, + 0x8f820018, 0x30840007, 0x00042240, 0x34870001, 0x00e83825, 0x1120000f, + 0xaca20004, 0x97430e0a, 0x8f84000c, 0x00ee3825, 0x2402000e, 0x00781825, + 0xaf430160, 0x25430006, 0x24840008, 0x30841fff, 0xa742015a, 0xa7430158, + 0xaf84000c, 0x0a0008a9, 0x00000000, 0x8f83000c, 0x25420002, 0xa7420158, + 0x24630008, 0x30631fff, 0xaf83000c, 0x54c0000c, 0x8f420e14, 0x97420e10, + 0x97430e12, 0x8f840014, 0x00021400, 0x00621825, 0xac8300a8, 0x8f850014, + 0x8f420e18, 0x34e70040, 0xaca200ac, 0x8f420e14, 0x8f430e1c, 0xaf420144, + 0xaf430148, 0xa34b0152, 0xaf470154, 0x0a0009f1, 0xaf4e0178, 0x10400128, + 0x00000000, 0x97620010, 0x00a2102b, 0x10400003, 0x30820040, 0x10400122, + 0x00000000, 0xafa60008, 0xa7840010, 0xaf850004, 0x93620008, 0x1440005e, + 0x27ac0008, 0xaf60000c, 0x97820010, 0x30424000, 0x10400002, 0x2403000e, + 0x24030016, 0xa363000a, 0x24034007, 0xaf630014, 0x93820012, 0x8f630014, + 0x30420007, 0x00021240, 0x00621825, 0xaf630014, 0x97820010, 0x8f630014, + 0x30420010, 0x00621825, 0xaf630014, 0x97820010, 0x30420008, 0x5040000e, + 0x00002821, 0x8f620014, 0x004e1025, 0xaf620014, 0x97430e0a, 0x2402000e, + 0x00781825, 0xaf630004, 0xa3620002, 0x9363000a, 0x3405fffc, 0x24630004, + 0x0a0008f2, 0xa363000a, 0xaf600004, 0xa3600002, 0x97820010, 0x9363000a, + 0x30421f00, 0x00021182, 0x24420028, 0x00621821, 0xa3630009, 0x97420e0c, + 0xa7620010, 0x93630009, 0x24020008, 0x24630002, 0x30630007, 0x00431023, + 0x30420007, 0xa362000b, 0x93640009, 0x97620010, 0x8f890004, 0x97830010, + 0x00441021, 0x00a21021, 0x30630040, 0x10600006, 0x3045ffff, 0x15250005, + 0x0125102b, 0x3c068000, 0x0a000925, 0x00005821, 0x0125102b, 0x144000c8, + 0x00005021, 0x97420e14, 0xa7420e10, 0x97430e16, 0xa7430e12, 0x8f420e1c, + 0xaf420e18, 0xaf450e00, 0x8f420000, 0x30420008, 0x1040fffd, 0x00000000, + 0x97420e08, 0x00a04821, 0xa7820010, 0x8f430e04, 0x00003021, 0x240b0001, + 0xaf830004, 0x97620010, 0x0a000936, 0x304dffff, 0x8f890004, 0x97820010, + 0x30420040, 0x10400004, 0x01206821, 0x3c068000, 0x0a000936, 0x00005821, + 0x97630010, 0x8f820004, 0x144300a7, 0x00005021, 0x00003021, 0x240b0001, + 0x8d820000, 0x00491023, 0x1440000d, 0xad820000, 0x8f620014, 0x34420040, + 0xaf620014, 0x97430e10, 0x97420e12, 0x8f840014, 0x00031c00, 0x00431025, + 0xac8200a8, 0x8f830014, 0x8f420e18, 0xac6200ac, 0x93620008, 0x1440003f, + 0x00000000, 0x25260002, 0x8f84000c, 0x9743008a, 0x3063ffff, 0xafa30000, + 0x8fa20000, 0x00441023, 0x2442ffff, 0x30421fff, 0x2c420010, 0x1440fff7, + 0x00000000, 0x8f82000c, 0x8f830018, 0x00021082, 0x00021080, 0x24424000, + 0x03422821, 0x00605021, 0x24630001, 0x314200ff, 0x00021400, 0xaf830018, + 0x3c033200, 0x00431025, 0xaca20000, 0x93630009, 0x9362000a, 0x00031c00, + 0x00431025, 0xaca20004, 0x8f830018, 0xaca30008, 0x97820010, 0x30420008, + 0x10400002, 0x00c04021, 0x25280006, 0x97430e14, 0x93640002, 0x8f450e1c, + 0x8f660004, 0x8f670014, 0xaf4f0178, 0x3063ffff, 0xa7430144, 0x97420e16, + 0xa7420146, 0xaf450148, 0xa34a0152, 0x8f82000c, 0x308400ff, 0xa744015a, + 0xaf460160, 0xa7480158, 0xaf470154, 0xaf4e0178, 0x00501021, 0x30421fff, + 0xaf82000c, 0x0a0009c5, 0x8d820000, 0x93620009, 0x9363000b, 0x8f85000c, + 0x2463000a, 0x00435021, 0x25440007, 0x00992024, 0x9743008a, 0x3063ffff, + 0xafa30000, 0x8fa20000, 0x00451023, 0x2442ffff, 0x30421fff, 0x0044102b, + 0x1440fff7, 0x00000000, 0x8f82000c, 0x8f840018, 0x00021082, 0x00021080, + 0x24424000, 0x03422821, 0x00804021, 0x24840001, 0xaf840018, 0x93630009, + 0x310200ff, 0x00022400, 0x3c024100, 0x24630002, 0x00621825, 0x00832025, + 0xaca40000, 0x8f62000c, 0x00461025, 0xaca20004, 0x97430e14, 0x93640002, + 0x8f450e1c, 0x8f660004, 0x8f670014, 0xaf4f0178, 0x3063ffff, 0xa7430144, + 0x97420e16, 0x308400ff, 0xa7420146, 0xaf450148, 0xa3480152, 0x8f83000c, + 0x25420007, 0x00591024, 0xa744015a, 0xaf460160, 0xa7490158, 0xaf470154, + 0xaf4e0178, 0x00621821, 0x30631fff, 0xaf83000c, 0x8d820000, 0x14400005, + 0x00000000, 0x8f620014, 0x2403ffbf, 0x00431024, 0xaf620014, 0x8f62000c, + 0x004d1021, 0xaf62000c, 0x93630008, 0x14600008, 0x00000000, 0x11600006, + 0x00000000, 0x8f630014, 0x3c02efff, 0x3442fffe, 0x00621824, 0xaf630014, + 0xa36b0008, 0x01205021, 0x15400016, 0x8fa60008, 0x97420e14, 0x97430e16, + 0x8f850014, 0x00021400, 0x00621825, 0xaca300a8, 0x8f840014, 0x8f420e1c, + 0x0a0009f3, 0xac8200ac, 0x97420e14, 0x97430e16, 0x8f840014, 0x00021400, + 0x00621825, 0xac8300a8, 0x8f850014, 0x8f420e1c, 0x00005021, 0x0a0009f3, + 0xaca200ac, 0x14c0fe64, 0x00000000, 0x55400018, 0x8fb00010, 0x3c038000, + 0x8f420178, 0x00431024, 0x1440fffd, 0x00000000, 0x97430e14, 0x8f440e1c, + 0x24020800, 0xaf420178, 0x3063ffff, 0xa7430144, 0x97420e16, 0x3c031000, + 0xa7420146, 0x24020240, 0xaf440148, 0xa3400152, 0xa740015a, 0xaf400160, + 0xa7400158, 0xaf420154, 0xaf430178, 0x8fb00010, 0x03e00008, 0x27bd0018, + 0x27bdffd8, 0x3c1a8000, 0x3c0420ff, 0x3484fffd, 0x3c020008, 0x03421821, + 0xafbf0020, 0xafb3001c, 0xafb20018, 0xafb10014, 0xafb00010, 0xaf830014, + 0xaf440e00, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x3c0200ff, 0x3442fffd, 0x3c046004, 0xaf420e00, 0x8c835000, 0x24130d00, + 0x3c120800, 0x3c114000, 0x2402ff7f, 0x00621824, 0x3463380c, 0x24020009, + 0xac835000, 0xaf420008, 0xaf800018, 0xaf80000c, 0x0e000fa1, 0x00000000, + 0x0e000a96, 0x00000000, 0x3c020800, 0x24504080, 0x8f420000, 0x30420001, + 0x1040fffd, 0x00000000, 0x8f440100, 0xaf840008, 0xaf440020, 0x93430108, + 0xa3830012, 0x93820012, 0x30420001, 0x10400008, 0x00000000, 0x93820012, + 0x30420006, 0x00021100, 0x0e00083b, 0x0050d821, 0x0a000a52, 0x00000000, + 0x14930005, 0x00000000, 0x0e00083b, 0x265b4100, 0x0a000a52, 0x00000000, + 0x0e000ba3, 0x00000000, 0xaf510138, 0x0a000a36, 0x00000000, 0x27bdfff8, + 0x3084ffff, 0x24820007, 0x3044fff8, 0x8f85000c, 0x9743008a, 0x3063ffff, + 0xafa30000, 0x8fa20000, 0x00451023, 0x2442ffff, 0x30421fff, 0x0044102b, + 0x1440fff7, 0x00000000, 0x8f82000c, 0x00021082, 0x00021080, 0x24424000, + 0x03421021, 0x03e00008, 0x27bd0008, 0x3084ffff, 0x8f82000c, 0x24840007, + 0x3084fff8, 0x00441021, 0x30421fff, 0xaf82000c, 0x03e00008, 0x00000000, + 0x27bdffe8, 0x3c1a8000, 0x3c0420ff, 0x3484fffd, 0x3c020008, 0x03421821, + 0xafbf0010, 0xaf830014, 0xaf440e00, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x3c0200ff, 0x3442fffd, 0x3c046004, 0xaf420e00, + 0x8c825000, 0x2403ff7f, 0x00431024, 0x3442380c, 0x24030009, 0xac825000, + 0xaf430008, 0xaf800018, 0xaf80000c, 0x0e000fa1, 0x00000000, 0x0e000a96, + 0x00000000, 0x8fbf0010, 0x03e00008, 0x27bd0018, 0x27bdffe8, 0x3c02000a, + 0x03421821, 0x3c040800, 0x24844120, 0x24050018, 0xafbf0010, 0xaf830024, + 0x0e000fad, 0x00003021, 0x3c050800, 0x3c020800, 0x24423d60, 0xaca24180, + 0x24a54180, 0x3c020800, 0x24423e18, 0x3c030800, 0x24633e2c, 0x3c040800, + 0xaca20004, 0x3c020800, 0x24423d68, 0xaca30008, 0xac824190, 0x24844190, + 0x3c020800, 0x24423da4, 0x3c070800, 0x24e73de4, 0x3c060800, 0x24c63e40, + 0x3c050800, 0x24a52b28, 0x3c030800, 0xac820004, 0x3c020800, 0x24423e48, + 0xac870008, 0xac86000c, 0xac850010, 0xac6241b0, 0x246341b0, 0x8fbf0010, + 0x3c020800, 0x24423e60, 0xac620004, 0xac670008, 0xac66000c, 0xac650010, + 0x03e00008, 0x27bd0018, 0x27bdffc8, 0x3c020800, 0x24424120, 0xafbf0030, + 0xafb3002c, 0xafb20028, 0xafb10024, 0xafb00020, 0x90470021, 0x8c510008, + 0x8c45001c, 0x8f900020, 0x3c060800, 0x3c038000, 0x8f420178, 0x00431024, + 0x1440fffd, 0x8cc2414c, 0x24c3414c, 0x2473ffd4, 0xaf420144, 0x8e620030, + 0x30b22000, 0xaf420148, 0x3c021000, 0xaf50014c, 0xa3470152, 0xa7510158, + 0xaf450154, 0xaf420178, 0x12400004, 0x3c030800, 0x8c620030, 0x24420001, + 0xac620030, 0x93420109, 0x9344010a, 0x00111c00, 0xafa30018, 0x00071a00, + 0xafa50014, 0x8cc5414c, 0x00021600, 0x00042400, 0x00441025, 0x00431025, + 0xafa20010, 0x8f440100, 0x8e660030, 0x0e000fe1, 0x02003821, 0x1640000e, + 0x8fbf0030, 0x8f820000, 0x8e630030, 0x8c44017c, 0x02031823, 0x00711823, + 0x00641823, 0x2c630002, 0x14600006, 0x8fb3002c, 0x0000000d, 0x00000000, + 0x240000ca, 0x8fbf0030, 0x8fb3002c, 0x8fb20028, 0x8fb10024, 0x8fb00020, + 0x03e00008, 0x27bd0038, 0x974309da, 0x00804021, 0xad030000, 0x8f4209dc, + 0xad020004, 0x8f4309e0, 0xad030008, 0x934409d9, 0x24020001, 0x30840003, + 0x1082001f, 0x30a900ff, 0x28820002, 0x10400005, 0x24020002, 0x10800009, + 0x3c0a0800, 0x0a000b64, 0x93420934, 0x1082000b, 0x24020003, 0x10820026, + 0x3c0a0800, 0x0a000b64, 0x93420934, 0x974209e4, 0x00021400, 0x34420800, + 0xad02000c, 0x0a000b63, 0x25080010, 0x974209e4, 0x00021400, 0x34428100, + 0xad02000c, 0x974309e8, 0x3c0a0800, 0x00031c00, 0x34630800, 0xad030010, + 0x0a000b63, 0x25080014, 0x974409e4, 0x3c050800, 0x24a24120, 0x94430018, + 0x94460010, 0x9447000c, 0x00a05021, 0x24020800, 0xad000010, 0xad020014, + 0x00042400, 0x00661821, 0x00671823, 0x2463fff2, 0x00832025, 0xad04000c, + 0x0a000b63, 0x25080018, 0x974209e4, 0x3c050800, 0x00021400, 0x34428100, + 0xad02000c, 0x974409e8, 0x24a24120, 0x94430018, 0x94460010, 0x9447000c, + 0x00a05021, 0x24020800, 0xad000014, 0xad020018, 0x00042400, 0x00661821, + 0x00671823, 0x2463ffee, 0x00832025, 0xad040010, 0x2508001c, 0x93420934, + 0x93450921, 0x3c074000, 0x25444120, 0x94830014, 0x94860010, 0x00021082, + 0x00021600, 0x00052c00, 0x00a72825, 0x00451025, 0x00661821, 0x00431025, + 0xad020000, 0x97830028, 0x974209ea, 0x00621821, 0x00031c00, 0xad030004, + 0x97820028, 0x24420001, 0x30427fff, 0xa7820028, 0x93430920, 0x3c020006, + 0x00031e00, 0x00621825, 0xad030008, 0x8f42092c, 0xad02000c, 0x8f430930, + 0xad030010, 0x8f440938, 0x25080014, 0xad040000, 0x8f820020, 0x11200004, + 0xad020004, 0x8f420940, 0x0a000b8d, 0x2442ffff, 0x8f420940, 0xad020008, + 0x8f440948, 0x8f420940, 0x93430936, 0x00822823, 0x00652806, 0x3402ffff, + 0x0045102b, 0x54400001, 0x3405ffff, 0x93420937, 0x25444120, 0x90830020, + 0xad000010, 0x00021700, 0x34630010, 0x00031c00, 0x00431025, 0x00451025, + 0xad02000c, 0x03e00008, 0x25020014, 0x27bdffb0, 0x3c020008, 0x03421821, + 0xafbf004c, 0xafbe0048, 0xafb70044, 0xafb60040, 0xafb5003c, 0xafb40038, + 0xafb30034, 0xafb20030, 0xafb1002c, 0xafb00028, 0xaf830000, 0x24020040, + 0xaf420814, 0xaf400810, 0x8f420944, 0x8f430950, 0x8f440954, 0x8f45095c, + 0xaf820030, 0xaf830020, 0xaf84001c, 0xaf85002c, 0x93430900, 0x24020020, + 0x10620005, 0x24020030, 0x10620022, 0x3c030800, 0x0a000bf1, 0x8c62002c, + 0x24020088, 0xaf420818, 0x3c020800, 0x24424180, 0xafa20020, 0x93430109, + 0x3c020800, 0x10600009, 0x24574190, 0x3c026000, 0x24030100, 0xac43081c, + 0x3c030001, 0xac43081c, 0x0000000d, 0x00000000, 0x2400031d, 0x9342010a, + 0x30420080, 0x1440001c, 0x00000000, 0x3c026000, 0x24030100, 0xac43081c, + 0x3c030001, 0xac43081c, 0x0000000d, 0x00000000, 0x24000324, 0x0a000bf4, + 0x00000000, 0x93430109, 0x3063007f, 0x00031140, 0x000318c0, 0x00431021, + 0x24430088, 0xaf430818, 0x0000000d, 0x3c020800, 0x244241d0, 0x3c030800, + 0x247741e0, 0x0a000bf4, 0xafa20020, 0x24420001, 0x0a000f4c, 0xac62002c, + 0x8f840000, 0x8f850020, 0x24020800, 0xaf420178, 0x8f4209a4, 0x8c83017c, + 0x00a21023, 0x00431023, 0x2c420002, 0x14400004, 0x00000000, 0x0000000d, + 0x00000000, 0x24000349, 0x8f420104, 0x8f430988, 0x00431023, 0x58400005, + 0x8f4209a0, 0x0000000d, 0x00000000, 0x2400034d, 0x8f4209a0, 0x3c100800, + 0xae02414c, 0x8f4309a4, 0x2604414c, 0x2491ffd4, 0xae230030, 0x8f420104, + 0xae250024, 0x00431023, 0xac82ffd4, 0x8fa30020, 0x8c620000, 0x0040f809, + 0x0200b021, 0x00409021, 0x32440010, 0x32420002, 0x10400007, 0xafa40024, + 0x8e22001c, 0x32500040, 0x2403ffbf, 0x00431024, 0x0a000f13, 0xae22001c, + 0x32420020, 0x10400002, 0x3c020800, 0x245741b0, 0x32420001, 0x14400007, + 0x00000000, 0x8f820008, 0xaf420080, 0x8ec3414c, 0xaf430e10, 0x8e220030, + 0xaf420e18, 0x9343010b, 0x93420905, 0x30420008, 0x1040003c, 0x307400ff, + 0x8f820000, 0x8c430074, 0x0460000a, 0x00000000, 0x3c026000, 0x24030100, + 0xac43081c, 0x3c030001, 0xac43081c, 0x0000000d, 0x00000000, 0x24000384, + 0x8f820000, 0x9044007b, 0x9343010a, 0x14830027, 0x32500040, 0x24072000, + 0x3c090800, 0x3c038000, 0x8f420178, 0x00431024, 0x1440fffd, 0x8ec2414c, + 0x26c4414c, 0x2484ffd4, 0xaf420144, 0x8c820030, 0x3c030100, 0xaf420148, + 0x24020047, 0xaf43014c, 0x00001821, 0xa3420152, 0x3c021000, 0xa7430158, + 0xaf470154, 0xaf420178, 0x8ec5414c, 0x8d230030, 0x8c860030, 0x24630001, + 0xad230030, 0x93420109, 0x9343010a, 0xafa70014, 0xafa00018, 0x00021600, + 0x00031c00, 0x00431025, 0x34424700, 0xafa20010, 0x8f440100, 0x0e000fe1, + 0x3c070100, 0x3c030800, 0x24624120, 0x0a000d01, 0x8c43001c, 0x32820002, + 0x10400047, 0x3c039000, 0x34630001, 0x8f820008, 0x32500040, 0x3c048000, + 0x00431025, 0xaf420020, 0x8f420020, 0x00441024, 0x1440fffd, 0x00000000, + 0x8f830000, 0x90620005, 0x3c058000, 0x34420008, 0xa0620005, 0x8f860000, + 0x34a50001, 0x8f840008, 0x8cc20074, 0x3c038000, 0x00852025, 0x00431025, + 0xacc20074, 0xaf440020, 0x90c3007b, 0x9342010a, 0x14620028, 0x3c040800, + 0x24072000, 0x3c090800, 0x3c038000, 0x8f420178, 0x00431024, 0x1440fffd, + 0x8ec2414c, 0x26c4414c, 0x2484ffd4, 0xaf420144, 0x8c820030, 0x3c030100, + 0xaf420148, 0x24020046, 0xaf43014c, 0x00001821, 0xa3420152, 0x3c021000, + 0xa7430158, 0xaf470154, 0xaf420178, 0x8ec5414c, 0x8d230030, 0x8c860030, + 0x24630001, 0xad230030, 0x93420109, 0x9343010a, 0xafa70014, 0xafa00018, + 0x00021600, 0x00031c00, 0x00431025, 0x34424600, 0xafa20010, 0x8f440100, + 0x0e000fe1, 0x3c070100, 0x3c040800, 0x24824120, 0x0a000d01, 0x8c43001c, + 0x93420108, 0x30420010, 0x50400050, 0x9343093f, 0x8f860000, 0x90c3007f, + 0x90c2007e, 0x90c40080, 0x306800ff, 0x00021600, 0x00081c00, 0x00431025, + 0x00042200, 0x90c3007a, 0x90c5000a, 0x00441025, 0x11050028, 0x00623825, + 0xa0c8000a, 0x24086000, 0x3c090800, 0x3c038000, 0x8f420178, 0x00431024, + 0x1440fffd, 0x8ec2414c, 0x26c4414c, 0x2484ffd4, 0xaf420144, 0x8c820030, + 0x00001821, 0xaf420148, 0x24020052, 0xaf47014c, 0xa3420152, 0x3c021000, + 0xa7430158, 0xaf480154, 0xaf420178, 0x8ec5414c, 0x8d230030, 0x8c860030, + 0x24630001, 0xad230030, 0x93420109, 0x9343010a, 0xafa80014, 0xafa00018, + 0x00021600, 0x00031c00, 0x00431025, 0x34425200, 0xafa20010, 0x0e000fe1, + 0x8f440100, 0x0a000cfb, 0x00000000, 0x3c026000, 0x24030100, 0xac43081c, + 0x3c030001, 0xac43081c, 0x0000000d, 0x00000000, 0x240003cd, 0x16800009, + 0x3c040800, 0x3c030800, 0x24624120, 0x8c43001c, 0x32500040, 0x2404ffbf, + 0x00641824, 0x0a000f13, 0xac43001c, 0x8c824120, 0x10400005, 0x3c030800, + 0x8c620034, 0xac804120, 0x24420001, 0xac620034, 0x9343093f, 0x24020012, + 0x1462000f, 0x329e0038, 0x17c0000c, 0x3c030800, 0x8f830000, 0x8c62004c, + 0xac62005c, 0x3c020800, 0x24444120, 0x8c82001c, 0x32500040, 0x2403ffbf, + 0x00431024, 0x0a000f13, 0xac82001c, 0xac604120, 0x97420908, 0x000211c0, + 0xaf420024, 0x97420908, 0x3c030080, 0x34630003, 0x000211c0, 0xaf42080c, + 0xaf43081c, 0x974209ec, 0x8f4309a4, 0xa7820028, 0x3c020800, 0x24444120, + 0xac830028, 0x93420937, 0x93430934, 0x00021080, 0x00621821, 0xa4830014, + 0x934209d8, 0x00621821, 0xa4830016, 0x934209d8, 0x93430934, 0x00809821, + 0x00431021, 0x24420010, 0xa4820012, 0x0000a821, 0x24020006, 0x13c00003, + 0xae62001c, 0x0a000d82, 0x24120008, 0x8f420958, 0x8f830020, 0x8f84002c, + 0x00431023, 0x00832023, 0x04800003, 0xae620004, 0x04410003, 0x0082102b, + 0x0a000d4e, 0xae600004, 0x54400001, 0xae640004, 0x8ee20000, 0x0040f809, + 0x00000000, 0x00409021, 0x32420001, 0x5440001e, 0x8ee20004, 0x8e630008, + 0x1060002b, 0x3c02c000, 0x00621025, 0xaf420e00, 0x8f420000, 0x30420008, + 0x1040fffd, 0x00000000, 0x97420e08, 0xa7820010, 0x8f430e04, 0x8e620008, + 0xaf830004, 0x8f840004, 0x0044102b, 0x1040000b, 0x24150001, 0x24020100, + 0x3c016000, 0xac22081c, 0x3c020001, 0x3c016000, 0xac22081c, 0x0000000d, + 0x00000000, 0x24000449, 0x24150001, 0x8ee20004, 0x0040f809, 0x00000000, + 0x02429025, 0x32420002, 0x5040001d, 0x8f470940, 0x12a00006, 0x8ec2414c, + 0x8f830000, 0xac6200a8, 0x8f840000, 0x8e620030, 0xac8200ac, 0x32420004, + 0x50400013, 0x8f470940, 0x3c020800, 0x3283007d, 0x106000fe, 0x245741b0, + 0x32820001, 0x50400006, 0x36520002, 0x8f830030, 0x8f420940, 0x106200f7, + 0x00000000, 0x36520002, 0x24020008, 0xa660000c, 0xa662000e, 0xae600008, + 0xa2600020, 0x8f470940, 0x3c030800, 0x24684120, 0x8d020028, 0x8d050008, + 0x9504000c, 0x9506000a, 0x95030022, 0x00451021, 0x00862021, 0x00641821, + 0xaf870030, 0xad020028, 0x32820030, 0x10400006, 0xa5030010, 0x91020020, + 0x32910040, 0x34420004, 0x0a000dd4, 0xa1020020, 0x93420923, 0x30420040, + 0x10400029, 0x32910040, 0x8f830000, 0x8f840020, 0x8c620084, 0x00441023, + 0x0442000a, 0x3c039000, 0x95020010, 0x8c630084, 0x00821021, 0x00621823, + 0x1c600004, 0x3c039000, 0x91020020, 0x34420001, 0xa1020020, 0x34630001, + 0x8f820008, 0x32910040, 0x3c048000, 0x00431025, 0xaf420020, 0x8f420020, + 0x00441024, 0x1440fffd, 0x00000000, 0x8f840000, 0x9083003f, 0x2402000a, + 0x10620005, 0x2402000c, 0x9083003f, 0x24020008, 0x14620002, 0x24020014, + 0xa082003f, 0x8f830008, 0x3c028000, 0x34420001, 0x00621825, 0xaf430020, + 0x3c040800, 0x24904120, 0x9602000c, 0x96030016, 0x9604000e, 0x00431021, + 0x00442021, 0x24840002, 0x3084ffff, 0x0e000a55, 0xa6020018, 0x8f850018, + 0x00a01821, 0xa2030021, 0x8ee60008, 0x00402021, 0x24a50001, 0xaf850018, + 0x00c0f809, 0x00000000, 0x00402021, 0x0e000b12, 0x02202821, 0x8ee3000c, + 0x0060f809, 0x00402021, 0x96040018, 0x9602000e, 0x00822021, 0x24840002, + 0x0e000a6b, 0x3084ffff, 0x3c030800, 0x8c624120, 0x8e030008, 0x3c040800, + 0x00431023, 0x14400012, 0xac824120, 0x54600006, 0x8e02001c, 0x3243004a, + 0x24020002, 0x14620005, 0x00000000, 0x8e02001c, 0x34420040, 0x0a000e0b, + 0xae02001c, 0x52a00006, 0x36520002, 0x8e02002c, 0xaf420e10, 0x8e030030, + 0xaf430e18, 0x36520002, 0x52a00008, 0x96670010, 0x8f830000, 0x8f420e10, + 0xac6200a8, 0x8f840000, 0x8f420e18, 0xac8200ac, 0x96670010, 0x92680020, + 0x24020040, 0xaf420814, 0x8f830020, 0x8f82001c, 0x00671821, 0x00621023, + 0xaf830020, 0x58400005, 0x8f42095c, 0x8f820000, 0xaf83001c, 0xac430054, + 0x8f42095c, 0x31030008, 0xaf82002c, 0x1060001a, 0x00000000, 0x8f840000, + 0x90820120, 0x90830121, 0x304600ff, 0x00c31823, 0x30630007, 0x24020007, + 0x1062000e, 0x00000000, 0x90820122, 0x304200fe, 0xa0820122, 0x8f850000, + 0x00061880, 0x8f840020, 0x24a20100, 0x00431021, 0x24c30001, 0x30630007, + 0xac440000, 0x0a000e40, 0xa0a30120, 0x90820122, 0x34420001, 0xa0820122, + 0x14e00003, 0x31020001, 0x10400031, 0x32510002, 0x8f820000, 0x8c43000c, + 0x30630001, 0x1060002c, 0x32510002, 0x3c029000, 0x8f830008, 0x34420001, + 0x3c048000, 0x00621825, 0xaf430020, 0x8f420020, 0x00441024, 0x1440fffd, + 0x00000000, 0x8f870000, 0x8ce2000c, 0x30420001, 0x10400018, 0x00000000, + 0x94e2006a, 0x00022880, 0x50a00001, 0x24050001, 0x94e30068, 0x90e40081, + 0x3c020800, 0x8c460024, 0x00652821, 0x00852804, 0x00c5102b, 0x54400001, + 0x00a03021, 0x3c020800, 0x8c440028, 0x00c4182b, 0x54600001, 0x00c02021, + 0x8f430074, 0x2402fffe, 0x00822824, 0x00a31821, 0xace3000c, 0x8f830008, + 0x3c028000, 0x34420001, 0x00621825, 0xaf430020, 0x8f830020, 0x3c020800, + 0x24504120, 0xae030024, 0x8ee20010, 0x0040f809, 0x00000000, 0x12a00005, + 0x00000000, 0x8f420e10, 0xae02002c, 0x8f430e18, 0xae030030, 0x1220feba, + 0x0000a821, 0x8f870024, 0x97860028, 0x8f830000, 0x8f820030, 0x8f840020, + 0x8f85001c, 0x32500040, 0xa4e6002c, 0xac620044, 0x32420008, 0xac640050, + 0xac650054, 0x1040007a, 0x32820020, 0x10400027, 0x32910010, 0x24072000, + 0x3c090800, 0x3c038000, 0x8f420178, 0x00431024, 0x1440fffd, 0x8ec2414c, + 0x26c4414c, 0x2484ffd4, 0xaf420144, 0x8c820030, 0x3c030400, 0xaf420148, + 0x24020041, 0xaf43014c, 0x00001821, 0xa3420152, 0x3c021000, 0xa7430158, + 0xaf470154, 0xaf420178, 0x8ec5414c, 0x8d230030, 0x8c860030, 0x24630001, + 0xad230030, 0x93420109, 0x9343010a, 0xafa70014, 0xafa00018, 0x00021600, + 0x00031c00, 0x00431025, 0x34424100, 0xafa20010, 0x8f440100, 0x0e000fe1, + 0x3c070400, 0x12200028, 0x24072000, 0x3c090800, 0x3c038000, 0x8f420178, + 0x00431024, 0x1440fffd, 0x8ec2414c, 0x26c4414c, 0x2484ffd4, 0xaf420144, + 0x8c820030, 0x3c030300, 0xaf420148, 0x2402004e, 0xaf43014c, 0x00001821, + 0xa3420152, 0x3c021000, 0xa7430158, 0xaf470154, 0xaf420178, 0x8ec5414c, + 0x8d230030, 0x8c860030, 0x24630001, 0xad230030, 0x93420109, 0x9343010a, + 0xafa70014, 0xafa00018, 0x00021600, 0x00031c00, 0x00431025, 0x34424e00, + 0xafa20010, 0x8f440100, 0x0e000fe1, 0x3c070300, 0x0a000f0b, 0x8fa30024, + 0x32820008, 0x10400026, 0x3c090800, 0x24072000, 0x3c038000, 0x8f420178, + 0x00431024, 0x1440fffd, 0x8ec2414c, 0x26c4414c, 0x2484ffd4, 0xaf420144, + 0x8c820030, 0x3c030200, 0xaf420148, 0x2402004b, 0xaf43014c, 0x00001821, + 0xa3420152, 0x3c021000, 0xa7430158, 0xaf470154, 0xaf420178, 0x8ec5414c, + 0x8d230030, 0x8c860030, 0x24630001, 0xad230030, 0x93420109, 0x9343010a, + 0xafa70014, 0xafa00018, 0x00021600, 0x00031c00, 0x00431025, 0x34424b00, + 0xafa20010, 0x8f440100, 0x0e000fe1, 0x3c070200, 0x8fa30024, 0x14600004, + 0x8fa40020, 0x32420010, 0x10400004, 0x00000000, 0x8c820004, 0x0040f809, + 0x00000000, 0x12000006, 0x8fa30020, 0x8c620008, 0x0040f809, 0x00000000, + 0x0a000f4d, 0x8fbf004c, 0x3c030800, 0x8c62413c, 0x30420040, 0x1440002f, + 0x8fbf004c, 0x24040040, 0x8f910020, 0x3c038000, 0x8f420178, 0x00431024, + 0x1440fffd, 0x8ec2414c, 0x26d0414c, 0x2610ffd4, 0xaf420144, 0x8e020030, + 0x00001821, 0xaf420148, 0x24020049, 0xaf51014c, 0xa3420152, 0x3c021000, + 0xa7430158, 0xaf440154, 0xaf420178, 0x8ec5414c, 0x8e060030, 0x93420109, + 0x9343010a, 0xafa40014, 0xafa00018, 0x00021600, 0x00031c00, 0x00431025, + 0x34424900, 0xafa20010, 0x8f440100, 0x0e000fe1, 0x02203821, 0x8f830000, + 0x8e020030, 0x8c64017c, 0x02221023, 0x00441023, 0x2c420002, 0x14400005, + 0x8fbf004c, 0x0000000d, 0x00000000, 0x240000ca, 0x8fbf004c, 0x8fbe0048, + 0x8fb70044, 0x8fb60040, 0x8fb5003c, 0x8fb40038, 0x8fb30034, 0x8fb20030, + 0x8fb1002c, 0x8fb00028, 0x03e00008, 0x27bd0050, 0x03e00008, 0x00001021, + 0x3c030800, 0x24654120, 0x8ca40004, 0x8c634120, 0x0064102b, 0x54400001, + 0x00602021, 0x9743093c, 0x0083102b, 0x54400001, 0x00801821, 0x00001021, + 0xaca30008, 0x03e00008, 0xa4a00022, 0x8f850004, 0x97840010, 0x3c030800, + 0x24634120, 0x24020008, 0xa462000e, 0x8f820004, 0xa460000c, 0x000420c2, + 0x30840008, 0x2c420001, 0x00021023, 0x30420006, 0xac650008, 0x03e00008, + 0xa0640020, 0x3c020800, 0x24424120, 0x90450021, 0x94430018, 0x3c021100, + 0xac800004, 0x00052c00, 0x24630002, 0x00621825, 0x00a32825, 0x24820008, + 0x03e00008, 0xac850000, 0x0000000d, 0x00000000, 0x2400016f, 0x03e00008, + 0x00000000, 0x0000000d, 0x00000000, 0x2400017b, 0x03e00008, 0x00000000, + 0x03e00008, 0x00000000, 0x3c020800, 0x24424120, 0xac400008, 0xa4400022, + 0x03e00008, 0x24020001, 0x3c020800, 0x24424120, 0x24030008, 0xac400008, + 0xa440000c, 0xa443000e, 0xa0400020, 0x03e00008, 0x24020004, 0x03e00008, + 0x00001021, 0x10c00007, 0x00000000, 0x8ca20000, 0x24c6ffff, 0x24a50004, + 0xac820000, 0x14c0fffb, 0x24840004, 0x03e00008, 0x00000000, 0x0a000fb2, + 0x00a01021, 0xac860000, 0x24840004, 0x00a01021, 0x1440fffc, 0x24a5ffff, + 0x03e00008, 0x00000000, 0x3c0a0800, 0x8d490068, 0x3c050800, 0x24a51090, + 0x00093140, 0x00c51021, 0xac440000, 0x8f440e04, 0x00a61021, 0xac440004, + 0x97430e08, 0x97420e0c, 0x00a62021, 0x00031c00, 0x00431025, 0xac820008, + 0x8f430e10, 0x00801021, 0xac43000c, 0x8f440e14, 0xac440010, 0x8f430e18, + 0x3c0800ff, 0xac430014, 0x8f470e1c, 0x3508ffff, 0x25290001, 0xac470018, + 0x3c070800, 0x8ce3006c, 0x9344010a, 0x3c026000, 0x24630001, 0xace3006c, + 0x8c434448, 0x3129007f, 0x00a62821, 0xad490068, 0x00042600, 0x00681824, + 0x00832025, 0x03e00008, 0xaca4001c, 0x8fac0010, 0x8fad0014, 0x8fae0018, + 0x3c0b0800, 0x8d6a0060, 0x3c080800, 0x25080078, 0x000a4940, 0x01281021, + 0x01091821, 0xac440000, 0x00601021, 0xac650004, 0xac460008, 0xac67000c, + 0xac4c0010, 0xac6d0014, 0x3c036000, 0xac4e0018, 0x8c654448, 0x3c040800, + 0x8c820064, 0x254a0001, 0x314a007f, 0x01094021, 0xad6a0060, 0x24420001, + 0xac820064, 0x03e00008, 0xad05001c, 0x00000000 }; + +static u32 bnx2_TXP_b06FwData[(0x0/4) + 1] = { 0x00000000 }; +static u32 bnx2_TXP_b06FwRodata[(0x0/4) + 1] = { 0x00000000 }; +static u32 bnx2_TXP_b06FwBss[(0x194/4) + 1] = { 0x00000000 }; +static u32 bnx2_TXP_b06FwSbss[(0x34/4) + 1] = { 0x00000000 }; + diff --git a/include/linux/mii.h b/include/linux/mii.h index 20971fe78a8d..374b615ea9ea 100644 --- a/include/linux/mii.h +++ b/include/linux/mii.h @@ -65,9 +65,13 @@ #define ADVERTISE_SLCT 0x001f /* Selector bits */ #define ADVERTISE_CSMA 0x0001 /* Only selector supported */ #define ADVERTISE_10HALF 0x0020 /* Try for 10mbps half-duplex */ +#define ADVERTISE_1000XFULL 0x0020 /* Try for 1000BASE-X full-duplex */ #define ADVERTISE_10FULL 0x0040 /* Try for 10mbps full-duplex */ +#define ADVERTISE_1000XHALF 0x0040 /* Try for 1000BASE-X half-duplex */ #define ADVERTISE_100HALF 0x0080 /* Try for 100mbps half-duplex */ +#define ADVERTISE_1000XPAUSE 0x0080 /* Try for 1000BASE-X pause */ #define ADVERTISE_100FULL 0x0100 /* Try for 100mbps full-duplex */ +#define ADVERTISE_1000XPSE_ASYM 0x0100 /* Try for 1000BASE-X asym pause */ #define ADVERTISE_100BASE4 0x0200 /* Try for 100mbps 4k packets */ #define ADVERTISE_PAUSE_CAP 0x0400 /* Try for pause */ #define ADVERTISE_PAUSE_ASYM 0x0800 /* Try for asymetric pause */ @@ -84,9 +88,13 @@ /* Link partner ability register. */ #define LPA_SLCT 0x001f /* Same as advertise selector */ #define LPA_10HALF 0x0020 /* Can do 10mbps half-duplex */ +#define LPA_1000XFULL 0x0020 /* Can do 1000BASE-X full-duplex */ #define LPA_10FULL 0x0040 /* Can do 10mbps full-duplex */ +#define LPA_1000XHALF 0x0040 /* Can do 1000BASE-X half-duplex */ #define LPA_100HALF 0x0080 /* Can do 100mbps half-duplex */ +#define LPA_1000XPAUSE 0x0080 /* Can do 1000BASE-X pause */ #define LPA_100FULL 0x0100 /* Can do 100mbps full-duplex */ +#define LPA_1000XPAUSE_ASYM 0x0100 /* Can do 1000BASE-X pause asym*/ #define LPA_100BASE4 0x0200 /* Can do 100mbps 4k packets */ #define LPA_PAUSE_CAP 0x0400 /* Can pause */ #define LPA_PAUSE_ASYM 0x0800 /* Can pause asymetrically */ diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 7b9720e35361..7ccbc2e4272c 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2071,6 +2071,7 @@ #define PCI_DEVICE_ID_TIGON3_5703 0x1647 #define PCI_DEVICE_ID_TIGON3_5704 0x1648 #define PCI_DEVICE_ID_TIGON3_5704S_2 0x1649 +#define PCI_DEVICE_ID_NX2_5706 0x164a #define PCI_DEVICE_ID_TIGON3_5702FE 0x164d #define PCI_DEVICE_ID_TIGON3_5705 0x1653 #define PCI_DEVICE_ID_TIGON3_5705_2 0x1654 @@ -2090,6 +2091,7 @@ #define PCI_DEVICE_ID_TIGON3_5702X 0x16a6 #define PCI_DEVICE_ID_TIGON3_5703X 0x16a7 #define PCI_DEVICE_ID_TIGON3_5704S 0x16a8 +#define PCI_DEVICE_ID_NX2_5706S 0x16aa #define PCI_DEVICE_ID_TIGON3_5702A3 0x16c6 #define PCI_DEVICE_ID_TIGON3_5703A3 0x16c7 #define PCI_DEVICE_ID_TIGON3_5781 0x16dd -- cgit v1.2.3 From aa8f0dc6c3dbf1cf3ff58f3e945c981be134814d Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Thu, 26 May 2005 21:54:27 -0400 Subject: libata: Fix use-after-iounmap Jens Axboe pointed out that the iounmap() call in libata was occurring too early, and some drivers (ahci, probably others) were using ioremap'd memory after it had been unmapped. The patch should address that problem by way of improving the libata driver API: * move ->host_stop() call after all ->port_stop() calls have occurred. * create default helper function ata_host_stop(), and move iounmap() call there. * add ->host_stop_prewalk() hook, use it in sata_qstor.c (hi Mark). sata_qstor appears to require the host-stop-before-port-stop ordering that existed prior to applying the attached patch. --- drivers/scsi/ahci.c | 2 ++ drivers/scsi/ata_piix.c | 2 ++ drivers/scsi/libata-core.c | 15 +++++++++++---- drivers/scsi/sata_nv.c | 2 ++ drivers/scsi/sata_promise.c | 1 + drivers/scsi/sata_qstor.c | 2 ++ drivers/scsi/sata_sil.c | 1 + drivers/scsi/sata_sis.c | 1 + drivers/scsi/sata_svw.c | 1 + drivers/scsi/sata_sx4.c | 2 ++ drivers/scsi/sata_uli.c | 1 + drivers/scsi/sata_via.c | 1 + drivers/scsi/sata_vsc.c | 1 + include/linux/libata.h | 1 + 14 files changed, 29 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/scsi/ahci.c b/drivers/scsi/ahci.c index da5bd33d982d..8b468a628700 100644 --- a/drivers/scsi/ahci.c +++ b/drivers/scsi/ahci.c @@ -289,6 +289,8 @@ static void ahci_host_stop(struct ata_host_set *host_set) { struct ahci_host_priv *hpriv = host_set->private_data; kfree(hpriv); + + ata_host_stop(host_set); } static int ahci_port_start(struct ata_port *ap) diff --git a/drivers/scsi/ata_piix.c b/drivers/scsi/ata_piix.c index 3867f91ef8c7..54c52349adc5 100644 --- a/drivers/scsi/ata_piix.c +++ b/drivers/scsi/ata_piix.c @@ -153,6 +153,7 @@ static struct ata_port_operations piix_pata_ops = { .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_operations piix_sata_ops = { @@ -180,6 +181,7 @@ static struct ata_port_operations piix_sata_ops = { .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info piix_port_info[] = { diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c index ee9b96da841e..2b41cd3a8ec6 100644 --- a/drivers/scsi/libata-core.c +++ b/drivers/scsi/libata-core.c @@ -3321,6 +3321,13 @@ void ata_port_stop (struct ata_port *ap) dma_free_coherent(dev, ATA_PRD_TBL_SZ, ap->prd, ap->prd_dma); } +void ata_host_stop (struct ata_host_set *host_set) +{ + if (host_set->mmio_base) + iounmap(host_set->mmio_base); +} + + /** * ata_host_remove - Unregister SCSI host structure with upper layers * @ap: Port to unregister @@ -3877,10 +3884,6 @@ void ata_pci_remove_one (struct pci_dev *pdev) } free_irq(host_set->irq, host_set); - if (host_set->ops->host_stop) - host_set->ops->host_stop(host_set); - if (host_set->mmio_base) - iounmap(host_set->mmio_base); for (i = 0; i < host_set->n_ports; i++) { ap = host_set->ports[i]; @@ -3899,6 +3902,9 @@ void ata_pci_remove_one (struct pci_dev *pdev) scsi_host_put(ap->host); } + if (host_set->ops->host_stop) + host_set->ops->host_stop(host_set); + kfree(host_set); pci_release_regions(pdev); @@ -3996,6 +4002,7 @@ EXPORT_SYMBOL_GPL(ata_chk_err); EXPORT_SYMBOL_GPL(ata_exec_command); EXPORT_SYMBOL_GPL(ata_port_start); EXPORT_SYMBOL_GPL(ata_port_stop); +EXPORT_SYMBOL_GPL(ata_host_stop); EXPORT_SYMBOL_GPL(ata_interrupt); EXPORT_SYMBOL_GPL(ata_qc_prep); EXPORT_SYMBOL_GPL(ata_bmdma_setup); diff --git a/drivers/scsi/sata_nv.c b/drivers/scsi/sata_nv.c index 69009f853a49..b0403ccd8a25 100644 --- a/drivers/scsi/sata_nv.c +++ b/drivers/scsi/sata_nv.c @@ -329,6 +329,8 @@ static void nv_host_stop (struct ata_host_set *host_set) host->host_desc->disable_hotplug(host_set); kfree(host); + + ata_host_stop(host_set); } static int nv_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) diff --git a/drivers/scsi/sata_promise.c b/drivers/scsi/sata_promise.c index c4e9e0298122..b18c90582e67 100644 --- a/drivers/scsi/sata_promise.c +++ b/drivers/scsi/sata_promise.c @@ -122,6 +122,7 @@ static struct ata_port_operations pdc_ata_ops = { .scr_write = pdc_sata_scr_write, .port_start = pdc_port_start, .port_stop = pdc_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info pdc_port_info[] = { diff --git a/drivers/scsi/sata_qstor.c b/drivers/scsi/sata_qstor.c index dfd362104717..1383e8a28d72 100644 --- a/drivers/scsi/sata_qstor.c +++ b/drivers/scsi/sata_qstor.c @@ -536,6 +536,8 @@ static void qs_host_stop(struct ata_host_set *host_set) writeb(0, mmio_base + QS_HCT_CTRL); /* disable host interrupts */ writeb(QS_CNFG3_GSRST, mmio_base + QS_HCF_CNFG3); /* global reset */ + + ata_host_stop(host_set); } static void qs_host_init(unsigned int chip_id, struct ata_probe_ent *pe) diff --git a/drivers/scsi/sata_sil.c b/drivers/scsi/sata_sil.c index 2b2ff48be396..238580d244e6 100644 --- a/drivers/scsi/sata_sil.c +++ b/drivers/scsi/sata_sil.c @@ -161,6 +161,7 @@ static struct ata_port_operations sil_ops = { .scr_write = sil_scr_write, .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info sil_port_info[] = { diff --git a/drivers/scsi/sata_sis.c b/drivers/scsi/sata_sis.c index 5105ddd08447..e418b89c6b9d 100644 --- a/drivers/scsi/sata_sis.c +++ b/drivers/scsi/sata_sis.c @@ -114,6 +114,7 @@ static struct ata_port_operations sis_ops = { .scr_write = sis_scr_write, .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info sis_port_info = { diff --git a/drivers/scsi/sata_svw.c b/drivers/scsi/sata_svw.c index 05075bd3a893..edef1fa969fc 100644 --- a/drivers/scsi/sata_svw.c +++ b/drivers/scsi/sata_svw.c @@ -313,6 +313,7 @@ static struct ata_port_operations k2_sata_ops = { .scr_write = k2_sata_scr_write, .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static void k2_sata_setup_port(struct ata_ioports *port, unsigned long base) diff --git a/drivers/scsi/sata_sx4.c b/drivers/scsi/sata_sx4.c index 70118650c461..140cea05de3f 100644 --- a/drivers/scsi/sata_sx4.c +++ b/drivers/scsi/sata_sx4.c @@ -245,6 +245,8 @@ static void pdc20621_host_stop(struct ata_host_set *host_set) iounmap(dimm_mmio); kfree(hpriv); + + ata_host_stop(host_set); } static int pdc_port_start(struct ata_port *ap) diff --git a/drivers/scsi/sata_uli.c b/drivers/scsi/sata_uli.c index 0bff4f475f26..a71fb54eebd3 100644 --- a/drivers/scsi/sata_uli.c +++ b/drivers/scsi/sata_uli.c @@ -113,6 +113,7 @@ static struct ata_port_operations uli_ops = { .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info uli_port_info = { diff --git a/drivers/scsi/sata_via.c b/drivers/scsi/sata_via.c index 3a7830667277..f43183c19a12 100644 --- a/drivers/scsi/sata_via.c +++ b/drivers/scsi/sata_via.c @@ -134,6 +134,7 @@ static struct ata_port_operations svia_sata_ops = { .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info svia_port_info = { diff --git a/drivers/scsi/sata_vsc.c b/drivers/scsi/sata_vsc.c index 2c28f0ad73c2..f67c34330ae9 100644 --- a/drivers/scsi/sata_vsc.c +++ b/drivers/scsi/sata_vsc.c @@ -230,6 +230,7 @@ static struct ata_port_operations vsc_sata_ops = { .scr_write = vsc_sata_scr_write, .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static void __devinit vsc_sata_setup_port(struct ata_ioports *port, unsigned long base) diff --git a/include/linux/libata.h b/include/linux/libata.h index 1f7e2039a04e..e74f301e9bae 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -410,6 +410,7 @@ extern u8 ata_chk_err(struct ata_port *ap); extern void ata_exec_command(struct ata_port *ap, struct ata_taskfile *tf); extern int ata_port_start (struct ata_port *ap); extern void ata_port_stop (struct ata_port *ap); +extern void ata_host_stop (struct ata_host_set *host_set); extern irqreturn_t ata_interrupt (int irq, void *dev_instance, struct pt_regs *regs); extern void ata_qc_prep(struct ata_queued_cmd *qc); extern int ata_qc_issue_prot(struct ata_queued_cmd *qc); -- cgit v1.2.3 From 8f37d47c9bf74cb48692691086b482e315d07f40 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 27 May 2005 12:17:28 +0100 Subject: AUDIT: Record working directory when syscall arguments are pathnames Signed-off-by: David Woodhouse --- include/linux/audit.h | 3 ++- kernel/auditsc.c | 23 +++++++++++++++++++++++ 2 files changed, 25 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index 3278ddf41ce6..bf2ad3ba72eb 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -61,11 +61,12 @@ #define AUDIT_SYSCALL 1300 /* Syscall event */ #define AUDIT_FS_WATCH 1301 /* Filesystem watch event */ -#define AUDIT_PATH 1302 /* Filname path information */ +#define AUDIT_PATH 1302 /* Filename path information */ #define AUDIT_IPC 1303 /* IPC record */ #define AUDIT_SOCKETCALL 1304 /* sys_socketcall arguments */ #define AUDIT_CONFIG_CHANGE 1305 /* Audit system configuration change */ #define AUDIT_SOCKADDR 1306 /* sockaddr copied as syscall arg */ +#define AUDIT_CWD 1307 /* Current working directory */ #define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ #define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 7556c479d5af..e75f84e1a1a0 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -145,6 +145,8 @@ struct audit_context { int auditable; /* 1 if record should be written */ int name_count; struct audit_names names[AUDIT_NAMES]; + struct dentry * pwd; + struct vfsmount * pwdmnt; struct audit_context *previous; /* For nested syscalls */ struct audit_aux_data *aux; @@ -552,6 +554,12 @@ static inline void audit_free_names(struct audit_context *context) if (context->names[i].name) __putname(context->names[i].name); context->name_count = 0; + if (context->pwd) + dput(context->pwd); + if (context->pwdmnt) + mntput(context->pwdmnt); + context->pwd = NULL; + context->pwdmnt = NULL; } static inline void audit_free_aux(struct audit_context *context) @@ -745,10 +753,18 @@ static void audit_log_exit(struct audit_context *context) audit_log_end(ab); } + if (context->pwd && context->pwdmnt) { + ab = audit_log_start(context, AUDIT_CWD); + if (ab) { + audit_log_d_path(ab, "cwd=", context->pwd, context->pwdmnt); + audit_log_end(ab); + } + } for (i = 0; i < context->name_count; i++) { ab = audit_log_start(context, AUDIT_PATH); if (!ab) continue; /* audit_panic has been called */ + audit_log_format(ab, "item=%d", i); if (context->names[i].name) { audit_log_format(ab, " name="); @@ -929,6 +945,13 @@ void audit_getname(const char *name) context->names[context->name_count].name = name; context->names[context->name_count].ino = (unsigned long)-1; ++context->name_count; + if (!context->pwd) { + read_lock(¤t->fs->lock); + context->pwd = dget(current->fs->pwd); + context->pwdmnt = mntget(current->fs->pwdmnt); + read_unlock(¤t->fs->lock); + } + } /* Intercept a putname request. Called from -- cgit v1.2.3 From 25be5e6ccc5c9cab77012fabbb72520f9b6e4618 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 27 May 2005 04:21:50 -0400 Subject: [PATCH] VIA IRQ quirk Delete quirk_via_bridge(), restore quirk_via_irqpic() -- but now improved to be invoked upon device ENABLE, and now only for VIA devices -- not all devices behind VIA bridges. Signed-off-by: Bjorn Helgaas Signed-off-by: Len Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/pci/irq.c | 5 ----- drivers/acpi/pci_irq.c | 4 ---- drivers/pci/quirks.c | 40 ++++++++++++++++++++++++---------------- include/linux/acpi.h | 5 ++++- 4 files changed, 28 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c index d6598da4b67b..da21b1d07c15 100644 --- a/arch/i386/pci/irq.c +++ b/arch/i386/pci/irq.c @@ -1029,7 +1029,6 @@ void pcibios_penalize_isa_irq(int irq) static int pirq_enable_irq(struct pci_dev *dev) { u8 pin; - extern int via_interrupt_line_quirk; struct pci_dev *temp_dev; pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin); @@ -1084,10 +1083,6 @@ static int pirq_enable_irq(struct pci_dev *dev) printk(KERN_WARNING "PCI: No IRQ known for interrupt pin %c of device %s.%s\n", 'A' + pin, pci_name(dev), msg); } - /* VIA bridges use interrupt line for apic/pci steering across - the V-Link */ - else if (via_interrupt_line_quirk) - pci_write_config_byte(dev, PCI_INTERRUPT_LINE, dev->irq & 15); return 0; } diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c index 12b0eea63407..8093f2e00321 100644 --- a/drivers/acpi/pci_irq.c +++ b/drivers/acpi/pci_irq.c @@ -391,7 +391,6 @@ acpi_pci_irq_enable ( u8 pin = 0; int edge_level = ACPI_LEVEL_SENSITIVE; int active_high_low = ACPI_ACTIVE_LOW; - extern int via_interrupt_line_quirk; char *link = NULL; ACPI_FUNCTION_TRACE("acpi_pci_irq_enable"); @@ -444,9 +443,6 @@ acpi_pci_irq_enable ( } } - if (via_interrupt_line_quirk) - pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq & 15); - dev->irq = acpi_register_gsi(irq, edge_level, active_high_low); printk(KERN_INFO PREFIX "PCI Interrupt %s[%c] -> ", diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 026aa04669a2..afc9d29acd8a 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -18,6 +18,7 @@ #include #include #include +#include #include "pci.h" /* Deal with broken BIOS'es that neglect to enable passive release, @@ -467,9 +468,6 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_APIC, * non-x86 architectures (yes Via exists on PPC among other places), * we must mask the PCI_INTERRUPT_LINE value versus 0xf to get * interrupts delivered properly. - * - * TODO: When we have device-specific interrupt routers, - * quirk_via_irqpic will go away from quirks. */ /* @@ -494,6 +492,29 @@ static void __devinit quirk_via_acpi(struct pci_dev *d) DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_3, quirk_via_acpi ); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4, quirk_via_acpi ); +static void __devinit quirk_via_irqpic(struct pci_dev *dev) +{ + u8 irq, new_irq; + +#ifdef CONFIG_X86_IO_APIC + if (nr_ioapics && !skip_ioapic_setup) + return; +#endif +#ifdef CONFIG_ACPI + if (acpi_irq_model != ACPI_IRQ_MODEL_PIC) + return; +#endif + new_irq = dev->irq & 0xf; + pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &irq); + if (new_irq != irq) { + printk(KERN_INFO "PCI: Via PIC IRQ fixup for %s, from %d to %d\n", + pci_name(dev), irq, new_irq); + udelay(15); /* unknown if delay really needed */ + pci_write_config_byte(dev, PCI_INTERRUPT_LINE, new_irq); + } +} +DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_ANY_ID, quirk_via_irqpic); + /* * PIIX3 USB: We have to disable USB interrupts that are * hardwired to PIRQD# and may be shared with an @@ -683,19 +704,6 @@ static void __init quirk_disable_pxb(struct pci_dev *pdev) } DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82454NX, quirk_disable_pxb ); -/* - * VIA northbridges care about PCI_INTERRUPT_LINE - */ -int via_interrupt_line_quirk; - -static void __devinit quirk_via_bridge(struct pci_dev *pdev) -{ - if(pdev->devfn == 0) { - printk(KERN_INFO "PCI: Via IRQ fixup\n"); - via_interrupt_line_quirk = 1; - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_ANY_ID, quirk_via_bridge ); /* * Serverworks CSB5 IDE does not fully support native mode diff --git a/include/linux/acpi.h b/include/linux/acpi.h index aefe6d051ace..d5a55bdb9c3c 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -25,6 +25,8 @@ #ifndef _LINUX_ACPI_H #define _LINUX_ACPI_H +#ifdef CONFIG_ACPI + #ifndef _LINUX #define _LINUX #endif @@ -533,4 +535,5 @@ static inline int acpi_get_pxm(acpi_handle handle) extern int pnpacpi_disabled; -#endif /*_LINUX_ACPI_H*/ +#endif /* CONFIG_ACPI */ +#endif /*_LINUX_ACPI_H*/ -- cgit v1.2.3 From 668d1e6093110f7534e661e2ff43d54c74659b6d Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sat, 28 May 2005 02:11:12 -0500 Subject: Input: This patch adds dummy gameport_register_port, gameport_unregister_port and gameport_set_phys functions to gameport.h for the case when a driver can't use gameport. This fixes the compilation of some OSS drivers with GAMEPORT=n without the need to #if inside every single driver. This patch also removes the non-working and now obsolete SOUND_GAMEPORT. This patch is also an alternative solution for ALSA drivers with similar problems (but #if's inside the drivers might have the advantage of saving some more bytes of gameport is not available). The only user-visible change is that for GAMEPORT=m the affected OSS drivers are now allowed to be built statically (but they won't have gameport support). Signed-off-by: Adrian Bunk Signed-off-by: Vojtech Pavlik Signed-off-by: Dmitry Torokhov --- drivers/input/gameport/Kconfig | 20 -------------------- include/linux/gameport.h | 28 +++++++++++++++++++++++++--- sound/oss/Kconfig | 12 ++++++------ 3 files changed, 31 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/drivers/input/gameport/Kconfig b/drivers/input/gameport/Kconfig index 6282f460aba0..1d93f5092904 100644 --- a/drivers/input/gameport/Kconfig +++ b/drivers/input/gameport/Kconfig @@ -68,23 +68,3 @@ config GAMEPORT_CS461X depends on PCI endif - -# Yes, SOUND_GAMEPORT looks a bit odd. Yes, it ends up being turned on -# in every .config. Please don't touch it. It is here to handle an -# unusual dependency between GAMEPORT and sound drivers. -# -# Some sound drivers call gameport functions. If GAMEPORT is -# not selected, empty stubs are provided for the functions and all is -# well. -# If GAMEPORT is built in, everything is fine. -# If GAMEPORT is a module, however, it would need to be loaded for the -# sound driver to be able to link properly. Therefore, the sound -# driver must be a module as well in that case. Since there's no way -# to express that directly in Kconfig, we use SOUND_GAMEPORT to -# express it. SOUND_GAMEPORT boils down to "if GAMEPORT is 'm', -# anything that depends on SOUND_GAMEPORT must be 'm' as well. if -# GAMEPORT is 'y' or 'n', it can be anything". -config SOUND_GAMEPORT - tristate - default m if GAMEPORT=m - default y diff --git a/include/linux/gameport.h b/include/linux/gameport.h index b1272f822cfa..cd623eccdbea 100644 --- a/include/linux/gameport.h +++ b/include/linux/gameport.h @@ -67,6 +67,8 @@ int gameport_open(struct gameport *gameport, struct gameport_driver *drv, int mo void gameport_close(struct gameport *gameport); void gameport_rescan(struct gameport *gameport); +#if defined(CONFIG_GAMEPORT) || (defined(MODULE) && defined(CONFIG_GAMEPORT_MODULE)) + void __gameport_register_port(struct gameport *gameport, struct module *owner); static inline void gameport_register_port(struct gameport *gameport) { @@ -75,6 +77,29 @@ static inline void gameport_register_port(struct gameport *gameport) void gameport_unregister_port(struct gameport *gameport); +void gameport_set_phys(struct gameport *gameport, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); + +#else + +static inline void gameport_register_port(struct gameport *gameport) +{ + return; +} + +static inline void gameport_unregister_port(struct gameport *gameport) +{ + return; +} + +static inline void gameport_set_phys(struct gameport *gameport, + const char *fmt, ...) +{ + return; +} + +#endif + static inline struct gameport *gameport_allocate_port(void) { struct gameport *gameport = kcalloc(1, sizeof(struct gameport), GFP_KERNEL); @@ -92,9 +117,6 @@ static inline void gameport_set_name(struct gameport *gameport, const char *name strlcpy(gameport->name, name, sizeof(gameport->name)); } -void gameport_set_phys(struct gameport *gameport, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); - /* * Use the following fucntions to manipulate gameport's per-port * driver-specific data. diff --git a/sound/oss/Kconfig b/sound/oss/Kconfig index a9602f89d6b1..e537bd66a707 100644 --- a/sound/oss/Kconfig +++ b/sound/oss/Kconfig @@ -112,7 +112,7 @@ config SOUND_BCM_CS4297A config SOUND_ES1370 tristate "Ensoniq AudioPCI (ES1370)" - depends on SOUND_PRIME!=n && SOUND && PCI && SOUND_GAMEPORT + depends on SOUND_PRIME!=n && SOUND && PCI help Say Y or M if you have a PCI sound card utilizing the Ensoniq ES1370 chipset, such as Ensoniq's AudioPCI (non-97). To find @@ -125,7 +125,7 @@ config SOUND_ES1370 config SOUND_ES1371 tristate "Creative Ensoniq AudioPCI 97 (ES1371)" - depends on SOUND_PRIME!=n && SOUND && PCI && SOUND_GAMEPORT + depends on SOUND_PRIME!=n && SOUND && PCI help Say Y or M if you have a PCI sound card utilizing the Ensoniq ES1371 chipset, such as Ensoniq's AudioPCI97. To find out if @@ -138,7 +138,7 @@ config SOUND_ES1371 config SOUND_ESSSOLO1 tristate "ESS Technology Solo1" - depends on SOUND_PRIME!=n && SOUND && SOUND_GAMEPORT && PCI + depends on SOUND_PRIME!=n && SOUND && PCI help Say Y or M if you have a PCI sound card utilizing the ESS Technology Solo1 chip. To find out if your sound card uses a @@ -179,7 +179,7 @@ config SOUND_HARMONY config SOUND_SONICVIBES tristate "S3 SonicVibes" - depends on SOUND_PRIME!=n && SOUND && SOUND_GAMEPORT + depends on SOUND_PRIME!=n && SOUND help Say Y or M if you have a PCI sound card utilizing the S3 SonicVibes chipset. To find out if your sound card uses a @@ -226,7 +226,7 @@ config SOUND_AU1550_AC97 config SOUND_TRIDENT tristate "Trident 4DWave DX/NX, SiS 7018 or ALi 5451 PCI Audio Core" - depends on SOUND_PRIME!=n && SOUND && SOUND_GAMEPORT + depends on SOUND_PRIME!=n && SOUND ---help--- Say Y or M if you have a PCI sound card utilizing the Trident 4DWave-DX/NX chipset or your mother board chipset has SiS 7018 @@ -739,7 +739,7 @@ config SOUND_NM256 config SOUND_MAD16 tristate "OPTi MAD16 and/or Mozart based cards" - depends on SOUND_OSS && SOUND_GAMEPORT + depends on SOUND_OSS ---help--- Answer Y if your card has a Mozart (OAK OTI-601) or MAD16 (OPTi 82C928 or 82C929 or 82C931) audio interface chip. These chips are -- cgit v1.2.3 From 8f28e8fa46625310102aea06fac61ba04c8b5b88 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sat, 28 May 2005 15:52:02 -0700 Subject: [PATCH] irq code: Add coherence test for PREEMPT_ACTIVE After porting this fixlet to UML: http://linux.bkbits.net:8080/linux-2.5/cset@41791ab52lfMuF2i3V-eTIGRBbDYKQ , I've also added a warning which should refuse compilation with insane values for PREEMPT_ACTIVE... maybe we should simply move PREEMPT_ACTIVE out of architectures using GENERIC_IRQS. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/hardirq.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index ebc712e91066..8336dba18971 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -43,13 +43,17 @@ #define __IRQ_MASK(x) ((1UL << (x))-1) #define PREEMPT_MASK (__IRQ_MASK(PREEMPT_BITS) << PREEMPT_SHIFT) -#define HARDIRQ_MASK (__IRQ_MASK(HARDIRQ_BITS) << HARDIRQ_SHIFT) #define SOFTIRQ_MASK (__IRQ_MASK(SOFTIRQ_BITS) << SOFTIRQ_SHIFT) +#define HARDIRQ_MASK (__IRQ_MASK(HARDIRQ_BITS) << HARDIRQ_SHIFT) #define PREEMPT_OFFSET (1UL << PREEMPT_SHIFT) #define SOFTIRQ_OFFSET (1UL << SOFTIRQ_SHIFT) #define HARDIRQ_OFFSET (1UL << HARDIRQ_SHIFT) +#if PREEMPT_ACTIVE < (1 << (HARDIRQ_SHIFT + HARDIRQ_BITS)) +#error PREEMPT_ACTIVE is too low! +#endif + #define hardirq_count() (preempt_count() & HARDIRQ_MASK) #define softirq_count() (preempt_count() & SOFTIRQ_MASK) #define irq_count() (preempt_count() & (HARDIRQ_MASK | SOFTIRQ_MASK)) -- cgit v1.2.3 From d8a33ac435c43a1a404b2ec560ef1d1536710c36 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Sun, 29 May 2005 14:13:47 -0700 Subject: [BRIDGE]: features change notification Resend of earlier patch (no changes) from Catalin used to provide device feature change notification. Signed-off-by: Catalin BOIE Acked-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 + include/linux/notifier.h | 1 + net/core/dev.c | 12 ++++++++++++ net/core/ethtool.c | 8 +++++++- 4 files changed, 21 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index b25bd02720d3..d8c65ecef9d9 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -913,6 +913,7 @@ extern void dev_mc_discard(struct net_device *dev); extern void dev_set_promiscuity(struct net_device *dev, int inc); extern void dev_set_allmulti(struct net_device *dev, int inc); extern void netdev_state_change(struct net_device *dev); +extern void netdev_features_change(struct net_device *dev); /* Load a device via the kmod */ extern void dev_load(const char *name); extern void dev_mcast_init(void); diff --git a/include/linux/notifier.h b/include/linux/notifier.h index 9303a003e9ab..5937dd6053c3 100644 --- a/include/linux/notifier.h +++ b/include/linux/notifier.h @@ -56,6 +56,7 @@ extern int notifier_call_chain(struct notifier_block **n, unsigned long val, voi #define NETDEV_CHANGEADDR 0x0008 #define NETDEV_GOING_DOWN 0x0009 #define NETDEV_CHANGENAME 0x000A +#define NETDEV_FEAT_CHANGE 0x000B #define SYS_DOWN 0x0001 /* Notify of system down */ #define SYS_RESTART SYS_DOWN diff --git a/net/core/dev.c b/net/core/dev.c index d4d9e2680adb..f15a3ffff635 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -760,6 +760,18 @@ int dev_change_name(struct net_device *dev, char *newname) return err; } +/** + * netdev_features_change - device changes fatures + * @dev: device to cause notification + * + * Called to indicate a device has changed features. + */ +void netdev_features_change(struct net_device *dev) +{ + notifier_call_chain(&netdev_chain, NETDEV_FEAT_CHANGE, dev); +} +EXPORT_SYMBOL(netdev_features_change); + /** * netdev_state_change - device changes state * @dev: device to cause notification diff --git a/net/core/ethtool.c b/net/core/ethtool.c index f05fde97c43d..252bfc6f03f4 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -682,6 +682,7 @@ int dev_ethtool(struct ifreq *ifr) void __user *useraddr = ifr->ifr_data; u32 ethcmd; int rc; + int old_features; /* * XXX: This can be pushed down into the ethtool_* handlers that @@ -703,6 +704,8 @@ int dev_ethtool(struct ifreq *ifr) if ((rc = dev->ethtool_ops->begin(dev)) < 0) return rc; + old_features = dev->features; + switch (ethcmd) { case ETHTOOL_GSET: rc = ethtool_get_settings(dev, useraddr); @@ -712,7 +715,6 @@ int dev_ethtool(struct ifreq *ifr) break; case ETHTOOL_GDRVINFO: rc = ethtool_get_drvinfo(dev, useraddr); - break; case ETHTOOL_GREGS: rc = ethtool_get_regs(dev, useraddr); @@ -801,6 +803,10 @@ int dev_ethtool(struct ifreq *ifr) if(dev->ethtool_ops->complete) dev->ethtool_ops->complete(dev); + + if (old_features != dev->features) + netdev_features_change(dev); + return rc; ioctl: -- cgit v1.2.3 From 81e8157583c559c27aac75c708d40a35f563d734 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Sun, 29 May 2005 14:14:35 -0700 Subject: [BRIDGE]: make dev->features unsigned The features field in netdevice is really a bitmask, and bitmask's should be unsigned. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- net/core/ethtool.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d8c65ecef9d9..470af8c1a4a0 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -401,7 +401,7 @@ struct net_device } reg_state; /* Net device features */ - int features; + unsigned long features; #define NETIF_F_SG 1 /* Scatter/gather IO. */ #define NETIF_F_IP_CSUM 2 /* Can checksum only TCP/UDP over IPv4. */ #define NETIF_F_NO_CSUM 4 /* Does not require checksum. F.e. loopack. */ diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 252bfc6f03f4..2a56a5218363 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -682,7 +682,7 @@ int dev_ethtool(struct ifreq *ifr) void __user *useraddr = ifr->ifr_data; u32 ethcmd; int rc; - int old_features; + unsigned long old_features; /* * XXX: This can be pushed down into the ethtool_* handlers that -- cgit v1.2.3 From 8f937c6099858eee15fae14009dcbd05177fa91d Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Sun, 29 May 2005 20:23:46 -0700 Subject: [IPV4]: Primary and secondary addresses Add an option to make secondary IP addresses get promoted when primary IP addresses are removed from the device. It defaults to off to preserve existing behavior. Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- include/linux/inetdevice.h | 2 ++ include/linux/sysctl.h | 1 + net/ipv4/devinet.c | 34 +++++++++++++++++++++++++++++----- 3 files changed, 32 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index 6fafb27877a7..7e1e15f934f3 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -29,6 +29,7 @@ struct ipv4_devconf int no_xfrm; int no_policy; int force_igmp_version; + int promote_secondaries; void *sysctl; }; @@ -71,6 +72,7 @@ struct in_device #define IN_DEV_SEC_REDIRECTS(in_dev) (ipv4_devconf.secure_redirects || (in_dev)->cnf.secure_redirects) #define IN_DEV_IDTAG(in_dev) ((in_dev)->cnf.tag) #define IN_DEV_MEDIUM_ID(in_dev) ((in_dev)->cnf.medium_id) +#define IN_DEV_PROMOTE_SECONDARIES(in_dev) (ipv4_devconf.promote_secondaries || (in_dev)->cnf.promote_secondaries) #define IN_DEV_RX_REDIRECTS(in_dev) \ ((IN_DEV_FORWARD(in_dev) && \ diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 772998147e3e..23032d9d6071 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -399,6 +399,7 @@ enum NET_IPV4_CONF_FORCE_IGMP_VERSION=17, NET_IPV4_CONF_ARP_ANNOUNCE=18, NET_IPV4_CONF_ARP_IGNORE=19, + NET_IPV4_CONF_PROMOTE_SECONDARIES=20, __NET_IPV4_CONF_MAX }; diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 3cc96730c4ed..478a30179a52 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -233,11 +233,14 @@ int inet_addr_onlink(struct in_device *in_dev, u32 a, u32 b) static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, int destroy) { + struct in_ifaddr *promote = NULL; struct in_ifaddr *ifa1 = *ifap; ASSERT_RTNL(); - /* 1. Deleting primary ifaddr forces deletion all secondaries */ + /* 1. Deleting primary ifaddr forces deletion all secondaries + * unless alias promotion is set + **/ if (!(ifa1->ifa_flags & IFA_F_SECONDARY)) { struct in_ifaddr *ifa; @@ -251,11 +254,16 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, continue; } - *ifap1 = ifa->ifa_next; + if (!IN_DEV_PROMOTE_SECONDARIES(in_dev)) { + *ifap1 = ifa->ifa_next; - rtmsg_ifa(RTM_DELADDR, ifa); - notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa); - inet_free_ifa(ifa); + rtmsg_ifa(RTM_DELADDR, ifa); + notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa); + inet_free_ifa(ifa); + } else { + promote = ifa; + break; + } } } @@ -281,6 +289,13 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, if (!in_dev->ifa_list) inetdev_destroy(in_dev); } + + if (promote && IN_DEV_PROMOTE_SECONDARIES(in_dev)) { + /* not sure if we should send a delete notify first? */ + promote->ifa_flags &= ~IFA_F_SECONDARY; + rtmsg_ifa(RTM_NEWADDR, promote); + notifier_call_chain(&inetaddr_chain, NETDEV_UP, promote); + } } static int inet_insert_ifa(struct in_ifaddr *ifa) @@ -1384,6 +1399,15 @@ static struct devinet_sysctl_table { .proc_handler = &ipv4_doint_and_flush, .strategy = &ipv4_doint_and_flush_strategy, }, + { + .ctl_name = NET_IPV4_CONF_PROMOTE_SECONDARIES, + .procname = "promote_secondaries", + .data = &ipv4_devconf.promote_secondaries, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &ipv4_doint_and_flush, + .strategy = &ipv4_doint_and_flush_strategy, + }, }, .devinet_dev = { { -- cgit v1.2.3 From 79165121bc09c209451487d977df910c4ff6fc94 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Sun, 29 May 2005 20:24:30 -0700 Subject: [NET]: Add is_multicast_ether_addr() in include/linux/etherdevice.h This patch adds is_multicast_ether_addr() to go along with is_valid_ether_addr() and friends. It then changes is_valid_ether_addr() to use the new macro, and fixes up the comment on that function to move implementation details out of the API doco. Signed-off-by: Michael Ellerman Signed-off-by: David S. Miller --- include/linux/etherdevice.h | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index 220748b7abea..a1478258d002 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -55,19 +55,33 @@ static inline int is_zero_ether_addr(const u8 *addr) return !(addr[0] | addr[1] | addr[2] | addr[3] | addr[4] | addr[5]); } +/** + * is_multicast_ether_addr - Determine if the given Ethernet address is a + * multicast address. + * + * @addr: Pointer to a six-byte array containing the Ethernet address + * + * Return true if the address is a multicast address. + */ +static inline int is_multicast_ether_addr(const u8 *addr) +{ + return addr[0] & 0x01; +} + /** * is_valid_ether_addr - Determine if the given Ethernet address is valid * @addr: Pointer to a six-byte array containing the Ethernet address * * Check that the Ethernet address (MAC) is not 00:00:00:00:00:00, is not - * a multicast address, and is not FF:FF:FF:FF:FF:FF. The multicast - * and FF:FF:... tests are combined into the single test "!(addr[0]&1)". + * a multicast address, and is not FF:FF:FF:FF:FF:FF. * * Return true if the address is valid. */ static inline int is_valid_ether_addr(const u8 *addr) { - return !(addr[0]&1) && !is_zero_ether_addr(addr); + /* FF:FF:FF:FF:FF:FF is a multicast address so we don't need to + * explicitly check for it here. */ + return !is_multicast_ether_addr(addr) && !is_zero_ether_addr(addr); } /** @@ -83,6 +97,6 @@ static inline void random_ether_addr(u8 *addr) addr [0] &= 0xfe; /* clear multicast bit */ addr [0] |= 0x02; /* set local assignment bit (IEEE802) */ } -#endif +#endif /* __KERNEL__ */ #endif /* _LINUX_ETHERDEVICE_H */ -- cgit v1.2.3 From 69f6a0fafcdf0bfe85af182695d6d38f80f9d549 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Sun, 29 May 2005 20:27:24 -0700 Subject: [NET]: Add ethtool support for NETIF_F_HW_CSUM. Signed-off-by: Jon Mason Signed-off-by: David S. Miller --- include/linux/ethtool.h | 1 + net/core/ethtool.c | 12 +++++++++++- 2 files changed, 12 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index c85b210490ea..a0ab26aab450 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -256,6 +256,7 @@ struct net_device; u32 ethtool_op_get_link(struct net_device *dev); u32 ethtool_op_get_tx_csum(struct net_device *dev); int ethtool_op_set_tx_csum(struct net_device *dev, u32 data); +int ethtool_op_set_tx_hw_csum(struct net_device *dev, u32 data); u32 ethtool_op_get_sg(struct net_device *dev); int ethtool_op_set_sg(struct net_device *dev, u32 data); u32 ethtool_op_get_tso(struct net_device *dev); diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 2a56a5218363..8ec484894d68 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -29,7 +29,7 @@ u32 ethtool_op_get_link(struct net_device *dev) u32 ethtool_op_get_tx_csum(struct net_device *dev) { - return (dev->features & NETIF_F_IP_CSUM) != 0; + return (dev->features & (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM)) != 0; } int ethtool_op_set_tx_csum(struct net_device *dev, u32 data) @@ -42,6 +42,15 @@ int ethtool_op_set_tx_csum(struct net_device *dev, u32 data) return 0; } +int ethtool_op_set_tx_hw_csum(struct net_device *dev, u32 data) +{ + if (data) + dev->features |= NETIF_F_HW_CSUM; + else + dev->features &= ~NETIF_F_HW_CSUM; + + return 0; +} u32 ethtool_op_get_sg(struct net_device *dev) { return (dev->features & NETIF_F_SG) != 0; @@ -823,3 +832,4 @@ EXPORT_SYMBOL(ethtool_op_get_tx_csum); EXPORT_SYMBOL(ethtool_op_set_sg); EXPORT_SYMBOL(ethtool_op_set_tso); EXPORT_SYMBOL(ethtool_op_set_tx_csum); +EXPORT_SYMBOL(ethtool_op_set_tx_hw_csum); -- cgit v1.2.3 From af00f9811e0ccbd3db84ddc4cffb0da942653393 Mon Sep 17 00:00:00 2001 From: Andy Currid Date: Mon, 23 May 2005 08:55:45 -0700 Subject: [PATCH] PCI: amd74xx patch for new NVIDIA device IDs Here's the 2.6 amd74xx patch for NVIDIA MCP51. Signed-off-by: Andy Currid Signed-off-by: Greg Kroah-Hartman --- drivers/ide/pci/amd74xx.c | 3 +++ include/linux/pci_ids.h | 6 ++++++ 2 files changed, 9 insertions(+) (limited to 'include/linux') diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c index 47225e324356..4e0f13d1d060 100644 --- a/drivers/ide/pci/amd74xx.c +++ b/drivers/ide/pci/amd74xx.c @@ -72,6 +72,7 @@ static struct amd_ide_chip { { PCI_DEVICE_ID_NVIDIA_NFORCE3S_SATA2, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_CK804_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP04_IDE, 0x50, AMD_UDMA_133 }, + { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE, 0x50, AMD_UDMA_133 }, { 0 } }; @@ -487,6 +488,7 @@ static ide_pci_device_t amd74xx_chipsets[] __devinitdata = { /* 12 */ DECLARE_NV_DEV("NFORCE3-250-SATA2"), /* 13 */ DECLARE_NV_DEV("NFORCE-CK804"), /* 14 */ DECLARE_NV_DEV("NFORCE-MCP04"), + /* 15 */ DECLARE_NV_DEV("NFORCE-MCP51"), }; static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_id *id) @@ -521,6 +523,7 @@ static struct pci_device_id amd74xx_pci_tbl[] = { #endif { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_CK804_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 13 }, { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP04_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 14 }, + { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 15 }, { 0, }, }; MODULE_DEVICE_TABLE(pci, amd74xx_pci_tbl); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 7ccbc2e4272c..b0d6134e1ee6 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1230,6 +1230,12 @@ #define PCI_DEVICE_ID_NVIDIA_QUADRO4_900XGL 0x0258 #define PCI_DEVICE_ID_NVIDIA_QUADRO4_750XGL 0x0259 #define PCI_DEVICE_ID_NVIDIA_QUADRO4_700XGL 0x025B +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE 0x0265 +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_SATA 0x0266 +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_SATA2 0x0267 +#define PCI_DEVICE_ID_NVIDIA_NVENET_12 0x0268 +#define PCI_DEVICE_ID_NVIDIA_NVENET_13 0x0269 +#define PCI_DEVICE_ID_NVIDIA_MCP51_AUDIO 0x026B #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_TI_4800 0x0280 #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_TI_4800_8X 0x0281 #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_TI_4800SE 0x0282 -- cgit v1.2.3 From b53cc6ead046093477ec7a3354d620337101ea5b Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Tue, 31 May 2005 19:03:47 -0700 Subject: [CPUFREQ] fix up comment in cpufreq.h Fix up comment in cpufreq.h stating transition latency should be passed in microseconds -- it was decided long ago to switch to nanoseconds. Signed-off-by: Dominik Brodowski Signed-off-by: Dave Jones --- include/linux/cpufreq.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index f21af067d015..927daa86c9b3 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -49,7 +49,7 @@ int cpufreq_unregister_notifier(struct notifier_block *nb, unsigned int list); /* Frequency values here are CPU kHz so that hardware which doesn't run * with some frequencies can complain without having to guess what per * cent / per mille means. - * Maximum transition latency is in microseconds - if it's unknown, + * Maximum transition latency is in nanoseconds - if it's unknown, * CPUFREQ_ETERNAL shall be used. */ -- cgit v1.2.3 From 0baab86b00cdf9785ac2bb2ce1ab63995b3866ca Mon Sep 17 00:00:00 2001 From: Edward Falk Date: Thu, 2 Jun 2005 18:17:13 -0400 Subject: libata: update inline source docs --- drivers/scsi/libata-core.c | 283 +++++++++++++++++++++++++++++++++++++++++++-- include/linux/libata.h | 58 ++++++++++ 2 files changed, 334 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c index 31b2984f379a..86e84ed87bee 100644 --- a/drivers/scsi/libata-core.c +++ b/drivers/scsi/libata-core.c @@ -186,6 +186,28 @@ static void ata_tf_load_mmio(struct ata_port *ap, struct ata_taskfile *tf) ata_wait_idle(ap); } + +/** + * ata_tf_load - send taskfile registers to host controller + * @ap: Port to which output is sent + * @tf: ATA taskfile register set + * + * Outputs ATA taskfile to standard ATA host controller using MMIO + * or PIO as indicated by the ATA_FLAG_MMIO flag. + * Writes the control, feature, nsect, lbal, lbam, and lbah registers. + * Optionally (ATA_TFLAG_LBA48) writes hob_feature, hob_nsect, + * hob_lbal, hob_lbam, and hob_lbah. + * + * This function waits for idle (!BUSY and !DRQ) after writing + * registers. If the control register has a new value, this + * function also waits for idle after writing control and before + * writing the remaining registers. + * + * May be used as the tf_load() entry in ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ void ata_tf_load(struct ata_port *ap, struct ata_taskfile *tf) { if (ap->flags & ATA_FLAG_MMIO) @@ -195,11 +217,11 @@ void ata_tf_load(struct ata_port *ap, struct ata_taskfile *tf) } /** - * ata_exec_command - issue ATA command to host controller + * ata_exec_command_pio - issue ATA command to host controller * @ap: port to which command is being issued * @tf: ATA taskfile register set * - * Issues PIO/MMIO write to ATA command register, with proper + * Issues PIO write to ATA command register, with proper * synchronization with interrupt handler / other threads. * * LOCKING: @@ -235,6 +257,18 @@ static void ata_exec_command_mmio(struct ata_port *ap, struct ata_taskfile *tf) ata_pause(ap); } + +/** + * ata_exec_command - issue ATA command to host controller + * @ap: port to which command is being issued + * @tf: ATA taskfile register set + * + * Issues PIO/MMIO write to ATA command register, with proper + * synchronization with interrupt handler / other threads. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ void ata_exec_command(struct ata_port *ap, struct ata_taskfile *tf) { if (ap->flags & ATA_FLAG_MMIO) @@ -305,7 +339,7 @@ void ata_tf_to_host_nolock(struct ata_port *ap, struct ata_taskfile *tf) } /** - * ata_tf_read - input device's ATA taskfile shadow registers + * ata_tf_read_pio - input device's ATA taskfile shadow registers * @ap: Port from which input is read * @tf: ATA taskfile register set for storing input * @@ -368,6 +402,23 @@ static void ata_tf_read_mmio(struct ata_port *ap, struct ata_taskfile *tf) } } + +/** + * ata_tf_read - input device's ATA taskfile shadow registers + * @ap: Port from which input is read + * @tf: ATA taskfile register set for storing input + * + * Reads ATA taskfile registers for currently-selected device + * into @tf. + * + * Reads nsect, lbal, lbam, lbah, and device. If ATA_TFLAG_LBA48 + * is set, also reads the hob registers. + * + * May be used as the tf_read() entry in ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf) { if (ap->flags & ATA_FLAG_MMIO) @@ -381,7 +432,7 @@ void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf) * @ap: port where the device is * * Reads ATA taskfile status register for currently-selected device - * and return it's value. This also clears pending interrupts + * and return its value. This also clears pending interrupts * from this device * * LOCKING: @@ -397,7 +448,7 @@ static u8 ata_check_status_pio(struct ata_port *ap) * @ap: port where the device is * * Reads ATA taskfile status register for currently-selected device - * via MMIO and return it's value. This also clears pending interrupts + * via MMIO and return its value. This also clears pending interrupts * from this device * * LOCKING: @@ -408,6 +459,20 @@ static u8 ata_check_status_mmio(struct ata_port *ap) return readb((void __iomem *) ap->ioaddr.status_addr); } + +/** + * ata_check_status - Read device status reg & clear interrupt + * @ap: port where the device is + * + * Reads ATA taskfile status register for currently-selected device + * and return its value. This also clears pending interrupts + * from this device + * + * May be used as the check_status() entry in ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ u8 ata_check_status(struct ata_port *ap) { if (ap->flags & ATA_FLAG_MMIO) @@ -415,6 +480,20 @@ u8 ata_check_status(struct ata_port *ap) return ata_check_status_pio(ap); } + +/** + * ata_altstatus - Read device alternate status reg + * @ap: port where the device is + * + * Reads ATA taskfile alternate status register for + * currently-selected device and return its value. + * + * Note: may NOT be used as the check_altstatus() entry in + * ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ u8 ata_altstatus(struct ata_port *ap) { if (ap->ops->check_altstatus) @@ -425,6 +504,20 @@ u8 ata_altstatus(struct ata_port *ap) return inb(ap->ioaddr.altstatus_addr); } + +/** + * ata_chk_err - Read device error reg + * @ap: port where the device is + * + * Reads ATA taskfile error register for + * currently-selected device and return its value. + * + * Note: may NOT be used as the check_err() entry in + * ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ u8 ata_chk_err(struct ata_port *ap) { if (ap->ops->check_err) @@ -873,10 +966,24 @@ void ata_dev_id_string(u16 *id, unsigned char *s, } } + +/** + * ata_noop_dev_select - Select device 0/1 on ATA bus + * @ap: ATA channel to manipulate + * @device: ATA device (numbered from zero) to select + * + * This function performs no actual function. + * + * May be used as the dev_select() entry in ata_port_operations. + * + * LOCKING: + * caller. + */ void ata_noop_dev_select (struct ata_port *ap, unsigned int device) { } + /** * ata_std_dev_select - Select device 0/1 on ATA bus * @ap: ATA channel to manipulate @@ -884,7 +991,9 @@ void ata_noop_dev_select (struct ata_port *ap, unsigned int device) * * Use the method defined in the ATA specification to * make either device 0, or device 1, active on the - * ATA channel. + * ATA channel. Works with both PIO and MMIO. + * + * May be used as the dev_select() entry in ata_port_operations. * * LOCKING: * caller. @@ -2127,6 +2236,19 @@ void ata_qc_prep(struct ata_queued_cmd *qc) * spin_lock_irqsave(host_set lock) */ + + +/** + * ata_sg_init_one - Prepare a one-entry scatter-gather list. + * @qc: Queued command + * @buf: transfer buffer + * @buflen: length of buf + * + * Builds a single-entry scatter-gather list to initiate a + * transfer utilizing the specified buffer. + * + * LOCKING: + */ void ata_sg_init_one(struct ata_queued_cmd *qc, void *buf, unsigned int buflen) { struct scatterlist *sg; @@ -2158,6 +2280,18 @@ void ata_sg_init_one(struct ata_queued_cmd *qc, void *buf, unsigned int buflen) * spin_lock_irqsave(host_set lock) */ + +/** + * ata_sg_init - Assign a scatter gather list to a queued command + * @qc: Queued command + * @sg: Scatter-gather list + * @n_elem: length of sg list + * + * Attaches a scatter-gather list to a queued command. + * + * LOCKING: + */ + void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg, unsigned int n_elem) { @@ -2331,6 +2465,18 @@ static void ata_pio_complete (struct ata_port *ap) ata_qc_complete(qc, drv_stat); } + +/** + * swap_buf_le16 - + * @buf: Buffer to swap + * @buf_words: Number of 16-bit words in buffer. + * + * Swap halves of 16-bit words if needed to convert from + * little-endian byte order to native cpu byte order, or + * vice-versa. + * + * LOCKING: + */ void swap_buf_le16(u16 *buf, unsigned int buf_words) { #ifdef __BIG_ENDIAN @@ -2992,6 +3138,7 @@ err_out: return -1; } + /** * ata_qc_issue_prot - issue taskfile to device in proto-dependent manner * @qc: command to issue to device @@ -3001,6 +3148,8 @@ err_out: * classes called "protocols", and issuing each type of protocol * is slightly different. * + * May be used as the qc_issue() entry in ata_port_operations. + * * LOCKING: * spin_lock_irqsave(host_set lock) * @@ -3058,7 +3207,7 @@ int ata_qc_issue_prot(struct ata_queued_cmd *qc) } /** - * ata_bmdma_setup - Set up PCI IDE BMDMA transaction + * ata_bmdma_setup_mmio - Set up PCI IDE BMDMA transaction * @qc: Info associated with this ATA transaction. * * LOCKING: @@ -3165,6 +3314,18 @@ static void ata_bmdma_start_pio (struct ata_queued_cmd *qc) ap->ioaddr.bmdma_addr + ATA_DMA_CMD); } + +/** + * ata_bmdma_start - Start a PCI IDE BMDMA transaction + * @qc: Info associated with this ATA transaction. + * + * Writes the ATA_DMA_START flag to the DMA command register. + * + * May be used as the bmdma_start() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ void ata_bmdma_start(struct ata_queued_cmd *qc) { if (qc->ap->flags & ATA_FLAG_MMIO) @@ -3173,6 +3334,20 @@ void ata_bmdma_start(struct ata_queued_cmd *qc) ata_bmdma_start_pio(qc); } + +/** + * ata_bmdma_setup - Set up PCI IDE BMDMA transaction + * @qc: Info associated with this ATA transaction. + * + * Writes address of PRD table to device's PRD Table Address + * register, sets the DMA control register, and calls + * ops->exec_command() to start the transfer. + * + * May be used as the bmdma_setup() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ void ata_bmdma_setup(struct ata_queued_cmd *qc) { if (qc->ap->flags & ATA_FLAG_MMIO) @@ -3181,6 +3356,19 @@ void ata_bmdma_setup(struct ata_queued_cmd *qc) ata_bmdma_setup_pio(qc); } + +/** + * ata_bmdma_irq_clear - Clear PCI IDE BMDMA interrupt. + * @qc: Info associated with this ATA transaction. + * + * Clear interrupt and error flags in DMA status register. + * + * May be used as the irq_clear() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ + void ata_bmdma_irq_clear(struct ata_port *ap) { if (ap->flags & ATA_FLAG_MMIO) { @@ -3193,6 +3381,19 @@ void ata_bmdma_irq_clear(struct ata_port *ap) } + +/** + * ata_bmdma_status - Read PCI IDE BMDMA status + * @qc: Info associated with this ATA transaction. + * + * Read and return BMDMA status register. + * + * May be used as the bmdma_status() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ + u8 ata_bmdma_status(struct ata_port *ap) { u8 host_stat; @@ -3204,6 +3405,19 @@ u8 ata_bmdma_status(struct ata_port *ap) return host_stat; } + +/** + * ata_bmdma_stop - Stop PCI IDE BMDMA transfer + * @qc: Info associated with this ATA transaction. + * + * Clears the ATA_DMA_START flag in the dma control register + * + * May be used as the bmdma_stop() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ + void ata_bmdma_stop(struct ata_port *ap) { if (ap->flags & ATA_FLAG_MMIO) { @@ -3407,6 +3621,19 @@ err_out: ata_qc_complete(qc, ATA_ERR); } + +/** + * ata_port_start - Set port up for dma. + * @ap: Port to initialize + * + * Called just after data structures for each port are + * initialized. Allocates space for PRD table. + * + * May be used as the port_start() entry in ata_port_operations. + * + * LOCKING: + */ + int ata_port_start (struct ata_port *ap) { struct device *dev = ap->host_set->dev; @@ -3420,6 +3647,18 @@ int ata_port_start (struct ata_port *ap) return 0; } + +/** + * ata_port_stop - Undo ata_port_start() + * @ap: Port to shut down + * + * Frees the PRD table. + * + * May be used as the port_stop() entry in ata_port_operations. + * + * LOCKING: + */ + void ata_port_stop (struct ata_port *ap) { struct device *dev = ap->host_set->dev; @@ -3720,7 +3959,15 @@ int ata_scsi_release(struct Scsi_Host *host) /** * ata_std_ports - initialize ioaddr with standard port offsets. * @ioaddr: IO address structure to be initialized + * + * Utility function which initializes data_addr, error_addr, + * feature_addr, nsect_addr, lbal_addr, lbam_addr, lbah_addr, + * device_addr, status_addr, and command_addr to standard offsets + * relative to cmd_addr. + * + * Does not set ctl_addr, altstatus_addr, bmdma_addr, or scr_addr. */ + void ata_std_ports(struct ata_ioports *ioaddr) { ioaddr->data_addr = ioaddr->cmd_addr + ATA_REG_DATA; @@ -3762,6 +4009,20 @@ ata_probe_ent_alloc(struct device *dev, struct ata_port_info *port) return probe_ent; } + + +/** + * ata_pci_init_native_mode - Initialize native-mode driver + * @pdev: pci device to be initialized + * @port: array[2] of pointers to port info structures. + * + * Utility function which allocates and initializes an + * ata_probe_ent structure for a standard dual-port + * PIO-based IDE controller. The returned ata_probe_ent + * structure can be passed to ata_device_add(). The returned + * ata_probe_ent structure should then be freed with kfree(). + */ + #ifdef CONFIG_PCI struct ata_probe_ent * ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port) @@ -3843,6 +4104,14 @@ ata_pci_init_legacy_mode(struct pci_dev *pdev, struct ata_port_info **port, * @port_info: Information from low-level host driver * @n_ports: Number of ports attached to host controller * + * This is a helper function which can be called from a driver's + * xxx_init_one() probe function if the hardware uses traditional + * IDE taskfile registers. + * + * This function calls pci_enable_device(), reserves its register + * regions, sets the dma mask, enables bus master mode, and calls + * ata_device_add() + * * LOCKING: * Inherited from PCI layer (may sleep). * diff --git a/include/linux/libata.h b/include/linux/libata.h index 1f7e2039a04e..ad410590664f 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -466,12 +466,34 @@ static inline u8 ata_chk_status(struct ata_port *ap) return ap->ops->check_status(ap); } + +/** + * ata_pause - Flush writes and pause 400 nanoseconds. + * @ap: Port to wait for. + * + * LOCKING: + * Inherited from caller. + */ + static inline void ata_pause(struct ata_port *ap) { ata_altstatus(ap); ndelay(400); } + +/** + * ata_busy_wait - Wait for a port status register + * @ap: Port to wait for. + * + * Waits up to max*10 microseconds for the selected bits in the port's + * status register to be cleared. + * Returns final value of status register. + * + * LOCKING: + * Inherited from caller. + */ + static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits, unsigned int max) { @@ -486,6 +508,18 @@ static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits, return status; } + +/** + * ata_wait_idle - Wait for a port to be idle. + * @ap: Port to wait for. + * + * Waits up to 10ms for port's BUSY and DRQ signals to clear. + * Returns final value of status register. + * + * LOCKING: + * Inherited from caller. + */ + static inline u8 ata_wait_idle(struct ata_port *ap) { u8 status = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000); @@ -524,6 +558,18 @@ static inline void ata_tf_init(struct ata_port *ap, struct ata_taskfile *tf, uns tf->device = ATA_DEVICE_OBS | ATA_DEV1; } + +/** + * ata_irq_on - Enable interrupts on a port. + * @ap: Port on which interrupts are enabled. + * + * Enable interrupts on a legacy IDE device using MMIO or PIO, + * wait for idle, clear any pending interrupts. + * + * LOCKING: + * Inherited from caller. + */ + static inline u8 ata_irq_on(struct ata_port *ap) { struct ata_ioports *ioaddr = &ap->ioaddr; @@ -543,6 +589,18 @@ static inline u8 ata_irq_on(struct ata_port *ap) return tmp; } + +/** + * ata_irq_ack - Acknowledge a device interrupt. + * @ap: Port on which interrupts are enabled. + * + * Wait up to 10 ms for legacy IDE device to become idle (BUSY + * or BUSY+DRQ clear). Obtain dma status and port status from + * device. Clear the interrupt. Return port status. + * + * LOCKING: + */ + static inline u8 ata_irq_ack(struct ata_port *ap, unsigned int chk_drq) { unsigned int bits = chk_drq ? ATA_BUSY | ATA_DRQ : ATA_BUSY; -- cgit v1.2.3 From b597ef4712c05c962640a655386a7f06cc1a1fbc Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 2 Jun 2005 16:36:00 -0700 Subject: [NET]: Fix locking in shaper driver. o use a semaphore instead of an opencoded and racy lock o move locking out of shaper_kick and into the callers - most just released the lock before calling shaper_kick o remove in_interrupt() tests. from ->close we can always block, from ->hard_start_xmit and timer context never Signed-off-by: Christoph Hellwig Signed-off-by: David S. Miller --- drivers/net/shaper.c | 86 +++++++++++------------------------------------ include/linux/if_shaper.h | 3 +- 2 files changed, 20 insertions(+), 69 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/shaper.c b/drivers/net/shaper.c index e68cf5fb4920..20edeb345792 100644 --- a/drivers/net/shaper.c +++ b/drivers/net/shaper.c @@ -100,35 +100,8 @@ static int sh_debug; /* Debug flag */ #define SHAPER_BANNER "CymruNet Traffic Shaper BETA 0.04 for Linux 2.1\n" -/* - * Locking - */ - -static int shaper_lock(struct shaper *sh) -{ - /* - * Lock in an interrupt must fail - */ - while (test_and_set_bit(0, &sh->locked)) - { - if (!in_interrupt()) - sleep_on(&sh->wait_queue); - else - return 0; - - } - return 1; -} - static void shaper_kick(struct shaper *sh); -static void shaper_unlock(struct shaper *sh) -{ - clear_bit(0, &sh->locked); - wake_up(&sh->wait_queue); - shaper_kick(sh); -} - /* * Compute clocks on a buffer */ @@ -157,17 +130,15 @@ static void shaper_setspeed(struct shaper *shaper, int bitspersec) * Throw a frame at a shaper. */ -static int shaper_qframe(struct shaper *shaper, struct sk_buff *skb) + +static int shaper_start_xmit(struct sk_buff *skb, struct net_device *dev) { + struct shaper *shaper = dev->priv; struct sk_buff *ptr; - /* - * Get ready to work on this shaper. Lock may fail if its - * an interrupt and locked. - */ - - if(!shaper_lock(shaper)) - return -1; + if (down_trylock(&shaper->sem)) + return -1; + ptr=shaper->sendq.prev; /* @@ -260,7 +231,8 @@ static int shaper_qframe(struct shaper *shaper, struct sk_buff *skb) dev_kfree_skb(ptr); shaper->stats.collisions++; } - shaper_unlock(shaper); + shaper_kick(shaper); + up(&shaper->sem); return 0; } @@ -297,8 +269,13 @@ static void shaper_queue_xmit(struct shaper *shaper, struct sk_buff *skb) static void shaper_timer(unsigned long data) { - struct shaper *sh=(struct shaper *)data; - shaper_kick(sh); + struct shaper *shaper = (struct shaper *)data; + + if (!down_trylock(&shaper->sem)) { + shaper_kick(shaper); + up(&shaper->sem); + } else + mod_timer(&shaper->timer, jiffies); } /* @@ -310,19 +287,6 @@ static void shaper_kick(struct shaper *shaper) { struct sk_buff *skb; - /* - * Shaper unlock will kick - */ - - if (test_and_set_bit(0, &shaper->locked)) - { - if(sh_debug) - printk("Shaper locked.\n"); - mod_timer(&shaper->timer, jiffies); - return; - } - - /* * Walk the list (may be empty) */ @@ -364,8 +328,6 @@ static void shaper_kick(struct shaper *shaper) if(skb!=NULL) mod_timer(&shaper->timer, SHAPERCB(skb)->shapeclock); - - clear_bit(0, &shaper->locked); } @@ -376,14 +338,12 @@ static void shaper_kick(struct shaper *shaper) static void shaper_flush(struct shaper *shaper) { struct sk_buff *skb; - if(!shaper_lock(shaper)) - { - printk(KERN_ERR "shaper: shaper_flush() called by an irq!\n"); - return; - } + + down(&shaper->sem); while((skb=skb_dequeue(&shaper->sendq))!=NULL) dev_kfree_skb(skb); - shaper_unlock(shaper); + shaper_kick(shaper); + up(&shaper->sem); } /* @@ -426,13 +386,6 @@ static int shaper_close(struct net_device *dev) * ARP and other resolutions and not before. */ - -static int shaper_start_xmit(struct sk_buff *skb, struct net_device *dev) -{ - struct shaper *sh=dev->priv; - return shaper_qframe(sh, skb); -} - static struct net_device_stats *shaper_get_stats(struct net_device *dev) { struct shaper *sh=dev->priv; @@ -623,7 +576,6 @@ static void shaper_init_priv(struct net_device *dev) init_timer(&sh->timer); sh->timer.function=shaper_timer; sh->timer.data=(unsigned long)sh; - init_waitqueue_head(&sh->wait_queue); } /* diff --git a/include/linux/if_shaper.h b/include/linux/if_shaper.h index 0485b256d043..004e6f09a6e2 100644 --- a/include/linux/if_shaper.h +++ b/include/linux/if_shaper.h @@ -23,7 +23,7 @@ struct shaper __u32 shapeclock; unsigned long recovery; /* Time we can next clock a packet out on an empty queue */ - unsigned long locked; + struct semaphore sem; struct net_device_stats stats; struct net_device *dev; int (*hard_start_xmit) (struct sk_buff *skb, @@ -38,7 +38,6 @@ struct shaper int (*hard_header_cache)(struct neighbour *neigh, struct hh_cache *hh); void (*header_cache_update)(struct hh_cache *hh, struct net_device *dev, unsigned char * haddr); struct net_device_stats* (*get_stats)(struct net_device *dev); - wait_queue_head_t wait_queue; struct timer_list timer; }; -- cgit v1.2.3 From 5ba0eac6e0b7e2889649a1105d97c600595e2bb1 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Thu, 2 Jun 2005 16:48:05 -0700 Subject: [NET]: Fix HH_DATA_OFF. When the hardware header size is a multiple of HH_DATA_MOD, HH_DATA_OFF() incorrectly returns HH_DATA_MOD (instead of 0). This affects ieee80211 layer as 802.11 header is 32 bytes long. Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 470af8c1a4a0..ba5d1236aa17 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -204,7 +204,7 @@ struct hh_cache /* cached hardware header; allow for machine alignment needs. */ #define HH_DATA_MOD 16 #define HH_DATA_OFF(__len) \ - (HH_DATA_MOD - ((__len) & (HH_DATA_MOD - 1))) + (HH_DATA_MOD - (((__len - 1) & (HH_DATA_MOD - 1)) + 1)) #define HH_DATA_ALIGN(__len) \ (((__len)+(HH_DATA_MOD-1))&~(HH_DATA_MOD - 1)) unsigned long hh_data[HH_DATA_ALIGN(LL_MAX_HEADER) / sizeof(long)]; -- cgit v1.2.3 From 719df469cb51199316ae2a11c75a8046be34b899 Mon Sep 17 00:00:00 2001 From: Roman Kagan Date: Fri, 6 May 2005 00:55:56 +0400 Subject: [PATCH] USB: update urb documentation On Wed, May 04, 2005 at 01:37:30PM -0700, David Brownell wrote: > On Wednesday 04 May 2005 12:19 pm, Roman Kagan wrote: > > struct urb { > > /* private, usb core and host controller only fields in the urb */ > > ... > > struct list_head urb_list; /* list pointer to all active urbs */ > > ... > > }; > > > > Is it safe to use it for driver's purposes when the driver owns the urb, > > that is, starting from the completion routine until the urb is submitted > > with usb_submit_urb()? > > Right now, it should be. Great! FWIW I've briefly tested a modified version of usbatm using the list head in struct urb instead of creating a wrapper struct, and I haven't seen any failures yet. So I tend to believe that your "should be" actually means "is" :) > > If it is, can it be guaranteed in future, e.g. > > by moving the list head into the public section of struct urb? > > In fact I'm not sure why it ever got called "private" to usbcore/hcds. > I thought the idea was that it should be like urb->status, reserved for > whoever controls the URB. OK then how about the following (essentially documentation) patch? Signed-off-by: Roman Kagan Acked-by: David Brownell Signed-off-by: Greg Kroah-Hartman --- include/linux/usb.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/usb.h b/include/linux/usb.h index 41d1a644c9d4..2d1ac5058534 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -796,6 +796,10 @@ typedef void (*usb_complete_t)(struct urb *, struct pt_regs *); * of the iso_frame_desc array, and the number of errors is reported in * error_count. Completion callbacks for ISO transfers will normally * (re)submit URBs to ensure a constant transfer rate. + * + * Note that even fields marked "public" should not be touched by the driver + * when the urb is owned by the hcd, that is, since the call to + * usb_submit_urb() till the entry into the completion routine. */ struct urb { @@ -803,12 +807,12 @@ struct urb struct kref kref; /* reference count of the URB */ spinlock_t lock; /* lock for the URB */ void *hcpriv; /* private data for host controller */ - struct list_head urb_list; /* list pointer to all active urbs */ int bandwidth; /* bandwidth for INT/ISO request */ atomic_t use_count; /* concurrent submissions counter */ u8 reject; /* submissions will fail */ /* public, documented fields in the urb that can be used by drivers */ + struct list_head urb_list; /* list head for use by the urb owner */ struct usb_device *dev; /* (in) pointer to associated device */ unsigned int pipe; /* (in) pipe information */ int status; /* (return) non-ISO status */ -- cgit v1.2.3 From 3f5948fa2cbbda1261eec9a39ef3004b3caf73fb Mon Sep 17 00:00:00 2001 From: David Mosberger Date: Mon, 6 Jun 2005 15:50:09 -0700 Subject: [PATCH] Include before testing CONFIG_ACPI I'm not sure why this issue is suddenly showing, but without this patchlet, the zx1 config won't compile anymore (e.g., to see the compilation-error, look for "***" in [1]). [1] http://www.gelato.unsw.edu.au/kerncomp/results//2005-06-06-17-00/zx1_defconfig-log.html Signed-off-by: David Mosberger-Tang Cc: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/acpi.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/acpi.h b/include/linux/acpi.h index d5a55bdb9c3c..b123cc08773d 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -25,6 +25,8 @@ #ifndef _LINUX_ACPI_H #define _LINUX_ACPI_H +#include + #ifdef CONFIG_ACPI #ifndef _LINUX -- cgit v1.2.3 From d0de98fa16169562bd74913c6c9b3857f9065c79 Mon Sep 17 00:00:00 2001 From: Alan Hourihane Date: Tue, 31 May 2005 19:50:49 +0100 Subject: [PATCH] i945G patch for agpgart Attached is a small patch for i945G support against 2.6.11.11. From: Alan Hourihane Signed-off-by: Dave Jones --- drivers/char/agp/intel-agp.c | 15 +++++++++++++-- include/linux/pci_ids.h | 2 ++ 2 files changed, 15 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c index 8c7d727432bb..6a5047e0d333 100644 --- a/drivers/char/agp/intel-agp.c +++ b/drivers/char/agp/intel-agp.c @@ -418,7 +418,8 @@ static void intel_i830_init_gtt_entries(void) case I915_GMCH_GMS_STOLEN_48M: /* Check it's really I915G */ if (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915G_HB || - agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915GM_HB) + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915GM_HB || + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82945G_HB) gtt_entries = MB(48) - KB(size); else gtt_entries = 0; @@ -426,7 +427,8 @@ static void intel_i830_init_gtt_entries(void) case I915_GMCH_GMS_STOLEN_64M: /* Check it's really I915G */ if (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915G_HB || - agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915GM_HB) + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915GM_HB || + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82945G_HB) gtt_entries = MB(64) - KB(size); else gtt_entries = 0; @@ -1662,6 +1664,14 @@ static int __devinit agp_intel_probe(struct pci_dev *pdev, } name = "915GM"; break; + case PCI_DEVICE_ID_INTEL_82945G_HB: + if (find_i830(PCI_DEVICE_ID_INTEL_82945G_IG)) { + bridge->driver = &intel_915_driver; + } else { + bridge->driver = &intel_845_driver; + } + name = "945G"; + break; case PCI_DEVICE_ID_INTEL_7505_0: bridge->driver = &intel_7505_driver; name = "E7505"; @@ -1801,6 +1811,7 @@ static struct pci_device_id agp_intel_pci_table[] = { ID(PCI_DEVICE_ID_INTEL_7205_0), ID(PCI_DEVICE_ID_INTEL_82915G_HB), ID(PCI_DEVICE_ID_INTEL_82915GM_HB), + ID(PCI_DEVICE_ID_INTEL_82945G_HB), { } }; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index b0d6134e1ee6..18f734ec9181 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2382,6 +2382,8 @@ #define PCI_DEVICE_ID_INTEL_82915G_IG 0x2582 #define PCI_DEVICE_ID_INTEL_82915GM_HB 0x2590 #define PCI_DEVICE_ID_INTEL_82915GM_IG 0x2592 +#define PCI_DEVICE_ID_INTEL_82945G_HB 0x2770 +#define PCI_DEVICE_ID_INTEL_82945G_IG 0x2772 #define PCI_DEVICE_ID_INTEL_ICH6_0 0x2640 #define PCI_DEVICE_ID_INTEL_ICH6_1 0x2641 #define PCI_DEVICE_ID_INTEL_ICH6_2 0x2642 -- cgit v1.2.3 From 6d1cfbab4de64f2d0c5b0f81177ade0d75b69288 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 8 Jun 2005 14:13:14 -0700 Subject: [TG3]: Fix 5700/5701 DMA corruption on Apple G4. Fix 5700/5701 DMA write corruption on Apple G4 by detecting the Apple UniNorth PCI 1.5 chipset and adjusting the DMA write boundary to 16. DMA test fails to detect the problem with this chipset. Thanks to Manuel Perez Ayala for reporting the problem and helping to debug it. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/tg3.c | 18 ++++++++++++++++-- include/linux/pci_ids.h | 1 + 2 files changed, 17 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index e944aac258e3..77337c3b4285 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -9695,10 +9695,24 @@ static int __devinit tg3_test_dma(struct tg3 *tp) } if ((tp->dma_rwctrl & DMA_RWCTRL_WRITE_BNDRY_MASK) != DMA_RWCTRL_WRITE_BNDRY_16) { + static struct pci_device_id dma_wait_state_chipsets[] = { + { PCI_DEVICE(PCI_VENDOR_ID_APPLE, + PCI_DEVICE_ID_APPLE_UNI_N_PCI15) }, + { }, + }; + /* DMA test passed without adjusting DMA boundary, - * just restore the calculated DMA boundary + * now look for chipsets that are known to expose the + * DMA bug without failing the test. */ - tp->dma_rwctrl = saved_dma_rwctrl; + if (pci_dev_present(dma_wait_state_chipsets)) { + tp->dma_rwctrl &= ~DMA_RWCTRL_WRITE_BNDRY_MASK; + tp->dma_rwctrl |= DMA_RWCTRL_WRITE_BNDRY_16; + } + else + /* Safe to use the calculated DMA boundary. */ + tp->dma_rwctrl = saved_dma_rwctrl; + tw32(TG3PCI_DMA_RW_CTRL, tp->dma_rwctrl); } diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 18f734ec9181..b8b4ebf9abf1 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -874,6 +874,7 @@ #define PCI_DEVICE_ID_APPLE_KL_USB_P 0x0026 #define PCI_DEVICE_ID_APPLE_UNI_N_AGP_P 0x0027 #define PCI_DEVICE_ID_APPLE_UNI_N_AGP15 0x002d +#define PCI_DEVICE_ID_APPLE_UNI_N_PCI15 0x002e #define PCI_DEVICE_ID_APPLE_UNI_N_FW2 0x0030 #define PCI_DEVICE_ID_APPLE_UNI_N_GMAC2 0x0032 #define PCI_DEVIEC_ID_APPLE_UNI_N_ATA 0x0033 -- cgit v1.2.3 From 4890062960cbc4d3cebdbd8261a68bc85efcf5d4 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Wed, 8 Jun 2005 15:10:48 -0700 Subject: [PKT_SCHED]: Allow socket attributes to be matched on via meta ematch Adds meta collectors for all socket attributes that make sense to be filtered upon. Some of them are only useful for debugging but having them doesn't hurt. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/tc_ematch/tc_em_meta.h | 30 ++++ net/sched/em_meta.c | 291 ++++++++++++++++++++++++++++++++--- 2 files changed, 297 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h index aa6b48bb4dcd..a6b2cc530af5 100644 --- a/include/linux/tc_ematch/tc_em_meta.h +++ b/include/linux/tc_ematch/tc_em_meta.h @@ -56,6 +56,36 @@ enum TCF_META_ID_TCCLASSID, TCF_META_ID_RTCLASSID, TCF_META_ID_RTIIF, + TCF_META_ID_SK_FAMILY, + TCF_META_ID_SK_STATE, + TCF_META_ID_SK_REUSE, + TCF_META_ID_SK_BOUND_IF, + TCF_META_ID_SK_REFCNT, + TCF_META_ID_SK_SHUTDOWN, + TCF_META_ID_SK_PROTO, + TCF_META_ID_SK_TYPE, + TCF_META_ID_SK_RCVBUF, + TCF_META_ID_SK_RMEM_ALLOC, + TCF_META_ID_SK_WMEM_ALLOC, + TCF_META_ID_SK_OMEM_ALLOC, + TCF_META_ID_SK_WMEM_QUEUED, + TCF_META_ID_SK_RCV_QLEN, + TCF_META_ID_SK_SND_QLEN, + TCF_META_ID_SK_ERR_QLEN, + TCF_META_ID_SK_FORWARD_ALLOCS, + TCF_META_ID_SK_SNDBUF, + TCF_META_ID_SK_ALLOCS, + TCF_META_ID_SK_ROUTE_CAPS, + TCF_META_ID_SK_HASHENT, + TCF_META_ID_SK_LINGERTIME, + TCF_META_ID_SK_ACK_BACKLOG, + TCF_META_ID_SK_MAX_ACK_BACKLOG, + TCF_META_ID_SK_PRIO, + TCF_META_ID_SK_RCVLOWAT, + TCF_META_ID_SK_RCVTIMEO, + TCF_META_ID_SK_SNDTIMEO, + TCF_META_ID_SK_SENDMSG_OFF, + TCF_META_ID_SK_WRITE_PENDING, __TCF_META_ID_MAX }; #define TCF_META_ID_MAX (__TCF_META_ID_MAX - 1) diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index f1eeaf65cee5..ed2a46cbb67f 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -32,7 +32,7 @@ * +-----------+ +-----------+ * | | * ---> meta_ops[INT][INDEV](...) | - * | | + * | | * ----------- | * V V * +-----------+ +-----------+ @@ -70,6 +70,7 @@ #include #include #include +#include struct meta_obj { @@ -283,6 +284,214 @@ META_COLLECTOR(int_rtiif) dst->value = ((struct rtable*) skb->dst)->fl.iif; } +/************************************************************************** + * Socket Attributes + **************************************************************************/ + +#define SKIP_NONLOCAL(skb) \ + if (unlikely(skb->sk == NULL)) { \ + *err = -1; \ + return; \ + } + +META_COLLECTOR(int_sk_family) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_family; +} + +META_COLLECTOR(int_sk_state) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_state; +} + +META_COLLECTOR(int_sk_reuse) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_reuse; +} + +META_COLLECTOR(int_sk_bound_if) +{ + SKIP_NONLOCAL(skb); + /* No error if bound_dev_if is 0, legal userspace check */ + dst->value = skb->sk->sk_bound_dev_if; +} + +META_COLLECTOR(var_sk_bound_if) +{ + SKIP_NONLOCAL(skb); + + if (skb->sk->sk_bound_dev_if == 0) { + dst->value = (unsigned long) "any"; + dst->len = 3; + } else { + struct net_device *dev; + + dev = dev_get_by_index(skb->sk->sk_bound_dev_if); + *err = var_dev(dev, dst); + if (dev) + dev_put(dev); + } +} + +META_COLLECTOR(int_sk_refcnt) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_refcnt); +} + +META_COLLECTOR(int_sk_rcvbuf) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_rcvbuf; +} + +META_COLLECTOR(int_sk_shutdown) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_shutdown; +} + +META_COLLECTOR(int_sk_proto) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_protocol; +} + +META_COLLECTOR(int_sk_type) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_type; +} + +META_COLLECTOR(int_sk_rmem_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_rmem_alloc); +} + +META_COLLECTOR(int_sk_wmem_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_wmem_alloc); +} + +META_COLLECTOR(int_sk_omem_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_omem_alloc); +} + +META_COLLECTOR(int_sk_rcv_qlen) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_receive_queue.qlen; +} + +META_COLLECTOR(int_sk_snd_qlen) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_write_queue.qlen; +} + +META_COLLECTOR(int_sk_wmem_queued) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_wmem_queued; +} + +META_COLLECTOR(int_sk_fwd_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_forward_alloc; +} + +META_COLLECTOR(int_sk_sndbuf) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_sndbuf; +} + +META_COLLECTOR(int_sk_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_allocation; +} + +META_COLLECTOR(int_sk_route_caps) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_route_caps; +} + +META_COLLECTOR(int_sk_hashent) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_hashent; +} + +META_COLLECTOR(int_sk_lingertime) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_lingertime / HZ; +} + +META_COLLECTOR(int_sk_err_qlen) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_error_queue.qlen; +} + +META_COLLECTOR(int_sk_ack_bl) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_ack_backlog; +} + +META_COLLECTOR(int_sk_max_ack_bl) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_max_ack_backlog; +} + +META_COLLECTOR(int_sk_prio) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_priority; +} + +META_COLLECTOR(int_sk_rcvlowat) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_rcvlowat; +} + +META_COLLECTOR(int_sk_rcvtimeo) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_rcvtimeo / HZ; +} + +META_COLLECTOR(int_sk_sndtimeo) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_sndtimeo / HZ; +} + +META_COLLECTOR(int_sk_sendmsg_off) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_sndmsg_off; +} + +META_COLLECTOR(int_sk_write_pend) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_write_pending; +} + /************************************************************************** * Meta value collectors assignment table **************************************************************************/ @@ -293,41 +502,75 @@ struct meta_ops struct meta_value *, struct meta_obj *, int *); }; +#define META_ID(name) TCF_META_ID_##name +#define META_FUNC(name) { .get = meta_##name } + /* Meta value operations table listing all meta value collectors and * assigns them to a type and meta id. */ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [TCF_META_TYPE_VAR] = { - [TCF_META_ID_DEV] = { .get = meta_var_dev }, - [TCF_META_ID_INDEV] = { .get = meta_var_indev }, - [TCF_META_ID_REALDEV] = { .get = meta_var_realdev } + [META_ID(DEV)] = META_FUNC(var_dev), + [META_ID(INDEV)] = META_FUNC(var_indev), + [META_ID(REALDEV)] = META_FUNC(var_realdev), + [META_ID(SK_BOUND_IF)] = META_FUNC(var_sk_bound_if), }, [TCF_META_TYPE_INT] = { - [TCF_META_ID_RANDOM] = { .get = meta_int_random }, - [TCF_META_ID_LOADAVG_0] = { .get = meta_int_loadavg_0 }, - [TCF_META_ID_LOADAVG_1] = { .get = meta_int_loadavg_1 }, - [TCF_META_ID_LOADAVG_2] = { .get = meta_int_loadavg_2 }, - [TCF_META_ID_DEV] = { .get = meta_int_dev }, - [TCF_META_ID_INDEV] = { .get = meta_int_indev }, - [TCF_META_ID_REALDEV] = { .get = meta_int_realdev }, - [TCF_META_ID_PRIORITY] = { .get = meta_int_priority }, - [TCF_META_ID_PROTOCOL] = { .get = meta_int_protocol }, - [TCF_META_ID_SECURITY] = { .get = meta_int_security }, - [TCF_META_ID_PKTTYPE] = { .get = meta_int_pkttype }, - [TCF_META_ID_PKTLEN] = { .get = meta_int_pktlen }, - [TCF_META_ID_DATALEN] = { .get = meta_int_datalen }, - [TCF_META_ID_MACLEN] = { .get = meta_int_maclen }, + [META_ID(RANDOM)] = META_FUNC(int_random), + [META_ID(LOADAVG_0)] = META_FUNC(int_loadavg_0), + [META_ID(LOADAVG_1)] = META_FUNC(int_loadavg_1), + [META_ID(LOADAVG_2)] = META_FUNC(int_loadavg_2), + [META_ID(DEV)] = META_FUNC(int_dev), + [META_ID(INDEV)] = META_FUNC(int_indev), + [META_ID(REALDEV)] = META_FUNC(int_realdev), + [META_ID(PRIORITY)] = META_FUNC(int_priority), + [META_ID(PROTOCOL)] = META_FUNC(int_protocol), + [META_ID(SECURITY)] = META_FUNC(int_security), + [META_ID(PKTTYPE)] = META_FUNC(int_pkttype), + [META_ID(PKTLEN)] = META_FUNC(int_pktlen), + [META_ID(DATALEN)] = META_FUNC(int_datalen), + [META_ID(MACLEN)] = META_FUNC(int_maclen), #ifdef CONFIG_NETFILTER - [TCF_META_ID_NFMARK] = { .get = meta_int_nfmark }, + [META_ID(NFMARK)] = META_FUNC(int_nfmark), #endif - [TCF_META_ID_TCINDEX] = { .get = meta_int_tcindex }, + [META_ID(TCINDEX)] = META_FUNC(int_tcindex), #ifdef CONFIG_NET_CLS_ACT - [TCF_META_ID_TCVERDICT] = { .get = meta_int_tcverd }, - [TCF_META_ID_TCCLASSID] = { .get = meta_int_tcclassid }, + [META_ID(TCVERDICT)] = META_FUNC(int_tcverd), + [META_ID(TCCLASSID)] = META_FUNC(int_tcclassid), #endif #ifdef CONFIG_NET_CLS_ROUTE - [TCF_META_ID_RTCLASSID] = { .get = meta_int_rtclassid }, + [META_ID(RTCLASSID)] = META_FUNC(int_rtclassid), #endif - [TCF_META_ID_RTIIF] = { .get = meta_int_rtiif } + [META_ID(RTIIF)] = META_FUNC(int_rtiif), + [META_ID(SK_FAMILY)] = META_FUNC(int_sk_family), + [META_ID(SK_STATE)] = META_FUNC(int_sk_state), + [META_ID(SK_REUSE)] = META_FUNC(int_sk_reuse), + [META_ID(SK_BOUND_IF)] = META_FUNC(int_sk_bound_if), + [META_ID(SK_REFCNT)] = META_FUNC(int_sk_refcnt), + [META_ID(SK_RCVBUF)] = META_FUNC(int_sk_rcvbuf), + [META_ID(SK_SNDBUF)] = META_FUNC(int_sk_sndbuf), + [META_ID(SK_SHUTDOWN)] = META_FUNC(int_sk_shutdown), + [META_ID(SK_PROTO)] = META_FUNC(int_sk_proto), + [META_ID(SK_TYPE)] = META_FUNC(int_sk_type), + [META_ID(SK_RMEM_ALLOC)] = META_FUNC(int_sk_rmem_alloc), + [META_ID(SK_WMEM_ALLOC)] = META_FUNC(int_sk_wmem_alloc), + [META_ID(SK_OMEM_ALLOC)] = META_FUNC(int_sk_omem_alloc), + [META_ID(SK_WMEM_QUEUED)] = META_FUNC(int_sk_wmem_queued), + [META_ID(SK_RCV_QLEN)] = META_FUNC(int_sk_rcv_qlen), + [META_ID(SK_SND_QLEN)] = META_FUNC(int_sk_snd_qlen), + [META_ID(SK_ERR_QLEN)] = META_FUNC(int_sk_err_qlen), + [META_ID(SK_FORWARD_ALLOCS)] = META_FUNC(int_sk_fwd_alloc), + [META_ID(SK_ALLOCS)] = META_FUNC(int_sk_alloc), + [META_ID(SK_ROUTE_CAPS)] = META_FUNC(int_sk_route_caps), + [META_ID(SK_HASHENT)] = META_FUNC(int_sk_hashent), + [META_ID(SK_LINGERTIME)] = META_FUNC(int_sk_lingertime), + [META_ID(SK_ACK_BACKLOG)] = META_FUNC(int_sk_ack_bl), + [META_ID(SK_MAX_ACK_BACKLOG)] = META_FUNC(int_sk_max_ack_bl), + [META_ID(SK_PRIO)] = META_FUNC(int_sk_prio), + [META_ID(SK_RCVLOWAT)] = META_FUNC(int_sk_rcvlowat), + [META_ID(SK_RCVTIMEO)] = META_FUNC(int_sk_rcvtimeo), + [META_ID(SK_SNDTIMEO)] = META_FUNC(int_sk_sndtimeo), + [META_ID(SK_SENDMSG_OFF)] = META_FUNC(int_sk_sendmsg_off), + [META_ID(SK_WRITE_PENDING)] = META_FUNC(int_sk_write_pend), } }; -- cgit v1.2.3 From a58e76f25432dc5e3e84d04c27bec03347ca365b Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 12 Jun 2005 10:56:26 +0200 Subject: [PATCH] Remove obsolete HAVE_ARCH_GET_SIGNAL_TO_DELIVER? Now m68k no longer sets HAVE_ARCH_GET_SIGNAL_TO_DELIVER, can it be removed completely? Or may ARM26 still need it? Note that its usage was removed from kernel/signal.c about 2 months ago. Signed-off-by: Geert Uytterhoeven Signed-off-by: Linus Torvalds --- include/asm-arm26/signal.h | 3 --- include/linux/signal.h | 2 -- 2 files changed, 5 deletions(-) (limited to 'include/linux') diff --git a/include/asm-arm26/signal.h b/include/asm-arm26/signal.h index dedb29280303..37ad25355591 100644 --- a/include/asm-arm26/signal.h +++ b/include/asm-arm26/signal.h @@ -166,9 +166,6 @@ typedef struct sigaltstack { #include #define sigmask(sig) (1UL << ((sig) - 1)) -//FIXME!!! -//#define HAVE_ARCH_GET_SIGNAL_TO_DELIVER - #endif diff --git a/include/linux/signal.h b/include/linux/signal.h index 0a98f5ec5cae..7be18b5e2fb4 100644 --- a/include/linux/signal.h +++ b/include/linux/signal.h @@ -231,10 +231,8 @@ extern int __group_send_sig_info(int, struct siginfo *, struct task_struct *); extern long do_sigpending(void __user *, unsigned long); extern int sigprocmask(int, sigset_t *, sigset_t *); -#ifndef HAVE_ARCH_GET_SIGNAL_TO_DELIVER struct pt_regs; extern int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka, struct pt_regs *regs, void *cookie); -#endif #endif /* __KERNEL__ */ -- cgit v1.2.3 From 03722adce90a248d0bea77d390decbd05991e2d2 Mon Sep 17 00:00:00 2001 From: Tom Rini Date: Mon, 13 Jun 2005 13:57:10 -0700 Subject: [NET]: linux/if_tr.h needs asm/byteorder.h uses __be16, but does not directly include . Add this in, so that dhcp/net-tools token ring code can compile again. Signed-off-by: Tom Rini Signed-off-by: David S. Miller --- include/linux/if_tr.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/if_tr.h b/include/linux/if_tr.h index 6688b414c529..3fba9e2f5427 100644 --- a/include/linux/if_tr.h +++ b/include/linux/if_tr.h @@ -19,6 +19,8 @@ #ifndef _LINUX_IF_TR_H #define _LINUX_IF_TR_H +#include /* For __be16 */ + /* IEEE 802.5 Token-Ring magic constants. The frame sizes omit the preamble and FCS/CRC (frame check sequence). */ #define TR_ALEN 6 /* Octets in one token-ring addr */ -- cgit v1.2.3 From 1c2fb7f93cb20621772bf304f3dba0849942e5db Mon Sep 17 00:00:00 2001 From: "J. Simonetti" Date: Mon, 13 Jun 2005 15:19:03 -0700 Subject: [IPV4]: Sysctl configurable icmp error source address. This patch alows you to change the source address of icmp error messages. It applies cleanly to 2.6.11.11 and retains the default behaviour. In the old (default) behaviour icmp error messages are sent with the ip of the exiting interface. The new behaviour (when the sysctl variable is toggled on), it will send the message with the ip of the interface that received the packet that caused the icmp error. This is the behaviour network administrators will expect from a router. It makes debugging complicated network layouts much easier. Also, all 'vendor routers' I know of have the later behaviour. Signed-off-by: David S. Miller --- include/linux/sysctl.h | 1 + net/ipv4/icmp.c | 9 +++++++-- net/ipv4/sysctl_net_ipv4.c | 9 +++++++++ 3 files changed, 17 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 23032d9d6071..a17745c80a91 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -346,6 +346,7 @@ enum NET_TCP_MODERATE_RCVBUF=106, NET_TCP_TSO_WIN_DIVISOR=107, NET_TCP_BIC_BETA=108, + NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR=109, }; enum { diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 85bf0d3e294b..cb759484979d 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -207,6 +207,7 @@ int sysctl_icmp_ignore_bogus_error_responses; int sysctl_icmp_ratelimit = 1 * HZ; int sysctl_icmp_ratemask = 0x1818; +int sysctl_icmp_errors_use_inbound_ifaddr; /* * ICMP control array. This specifies what to do with each ICMP. @@ -511,8 +512,12 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, u32 info) */ saddr = iph->daddr; - if (!(rt->rt_flags & RTCF_LOCAL)) - saddr = 0; + if (!(rt->rt_flags & RTCF_LOCAL)) { + if (sysctl_icmp_errors_use_inbound_ifaddr) + saddr = inet_select_addr(skb_in->dev, 0, RT_SCOPE_LINK); + else + saddr = 0; + } tos = icmp_pointers[type].error ? ((iph->tos & IPTOS_TOS_MASK) | IPTOS_PREC_INTERNETCONTROL) : diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 3aafb298c1c1..23068bddbf0b 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -23,6 +23,7 @@ extern int sysctl_ip_nonlocal_bind; extern int sysctl_icmp_echo_ignore_all; extern int sysctl_icmp_echo_ignore_broadcasts; extern int sysctl_icmp_ignore_bogus_error_responses; +extern int sysctl_icmp_errors_use_inbound_ifaddr; /* From ip_fragment.c */ extern int sysctl_ipfrag_low_thresh; @@ -395,6 +396,14 @@ ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = &proc_dointvec }, + { + .ctl_name = NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR, + .procname = "icmp_errors_use_inbound_ifaddr", + .data = &sysctl_icmp_errors_use_inbound_ifaddr, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, { .ctl_name = NET_IPV4_ROUTE, .procname = "route", -- cgit v1.2.3 From 986a80d5c154808cc78170584670324a22fd8219 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Thu, 16 Jun 2005 15:14:00 -0700 Subject: [PATCH] avoid signed vs unsigned comparison in efi_range_is_wc() warning when building with gcc -W : include/linux/efi.h: In function `efi_range_is_wc': include/linux/efi.h:320: warning: comparison between signed and unsigned It looks to me like a significantly large 'len' passed in could cause the loop to never end. Isn't it safer to make 'i' an unsigned long as well? Like this little patch below (which of course also kills the warning) : Signed-off-by: Jesper Juhl Signed-off-by: Tony Luck --- include/linux/efi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/efi.h b/include/linux/efi.h index 047e7222df7a..73781ec165b4 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -315,7 +315,7 @@ extern struct efi_memory_map memmap; */ static inline int efi_range_is_wc(unsigned long start, unsigned long len) { - int i; + unsigned long i; for (i = 0; i < len; i += (1UL << EFI_PAGE_SHIFT)) { unsigned long paddr = __pa(start + i); -- cgit v1.2.3 From b8112df71cae7d6a86158caeb19d215f56c4f9ab Mon Sep 17 00:00:00 2001 From: Lee Revell Date: Wed, 15 Jun 2005 14:19:03 -0400 Subject: [SCSI] Add DMA mask constants other than 32 and 64 bit Signed-Off-By: Lee Revell Signed-off-by: James Bottomley --- include/linux/dma-mapping.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index 806c305332c1..2d80cc761a15 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -14,7 +14,12 @@ enum dma_data_direction { }; #define DMA_64BIT_MASK 0xffffffffffffffffULL +#define DMA_40BIT_MASK 0x000000ffffffffffULL +#define DMA_39BIT_MASK 0x0000007fffffffffULL #define DMA_32BIT_MASK 0x00000000ffffffffULL +#define DMA_31BIT_MASK 0x000000007fffffffULL +#define DMA_30BIT_MASK 0x000000003fffffffULL +#define DMA_29BIT_MASK 0x000000001fffffffULL #include -- cgit v1.2.3 From 26b15dad9f1c19d6d4f7b999b07eaa6d98e4b375 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sat, 18 Jun 2005 22:42:13 -0700 Subject: [IPSEC] Add complete xfrm event notification Heres the final patch. What this patch provides - netlink xfrm events - ability to have events generated by netlink propagated to pfkey and vice versa. - fixes the acquire lets-be-happy-with-one-success issue Signed-off-by: Jamal Hadi Salim Signed-off-by: Herbert Xu --- include/linux/xfrm.h | 2 + include/net/xfrm.h | 29 +++- net/key/af_key.c | 357 +++++++++++++++++++++++++++++++++++++------------- net/xfrm/xfrm_state.c | 74 ++++++++--- net/xfrm/xfrm_user.c | 272 +++++++++++++++++++++++++++++++++++++- 5 files changed, 616 insertions(+), 118 deletions(-) (limited to 'include/linux') diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index fd2ef742a9fd..03bc600516ea 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -257,5 +257,7 @@ struct xfrm_usersa_flush { #define XFRMGRP_ACQUIRE 1 #define XFRMGRP_EXPIRE 2 +#define XFRMGRP_SA 4 +#define XFRMGRP_POLICY 8 #endif /* _LINUX_XFRM_H */ diff --git a/include/net/xfrm.h b/include/net/xfrm.h index d675836ba6c3..a159655ebede 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -158,6 +158,27 @@ enum { XFRM_STATE_DEAD }; +/* events that could be sent by kernel */ +enum { + XFRM_SAP_INVALID, + XFRM_SAP_EXPIRED, + XFRM_SAP_ADDED, + XFRM_SAP_UPDATED, + XFRM_SAP_DELETED, + XFRM_SAP_FLUSHED, + __XFRM_SAP_MAX +}; +#define XFRM_SAP_MAX (__XFRM_SAP_MAX - 1) + +/* callback structure passed from either netlink or pfkey */ +struct km_event +{ + u32 data; + u32 seq; + u32 pid; + u32 event; +}; + struct xfrm_type; struct xfrm_dst; struct xfrm_policy_afinfo { @@ -179,6 +200,8 @@ struct xfrm_policy_afinfo { extern int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo); extern int xfrm_policy_unregister_afinfo(struct xfrm_policy_afinfo *afinfo); +extern void km_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c); +extern void km_state_notify(struct xfrm_state *x, struct km_event *c); #define XFRM_ACQ_EXPIRES 30 @@ -290,11 +313,11 @@ struct xfrm_mgr { struct list_head list; char *id; - int (*notify)(struct xfrm_state *x, int event); + int (*notify)(struct xfrm_state *x, struct km_event *c); int (*acquire)(struct xfrm_state *x, struct xfrm_tmpl *, struct xfrm_policy *xp, int dir); struct xfrm_policy *(*compile_policy)(u16 family, int opt, u8 *data, int len, int *dir); int (*new_mapping)(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport); - int (*notify_policy)(struct xfrm_policy *x, int dir, int event); + int (*notify_policy)(struct xfrm_policy *x, int dir, struct km_event *c); }; extern int xfrm_register_km(struct xfrm_mgr *km); @@ -817,7 +840,7 @@ extern int xfrm_state_add(struct xfrm_state *x); extern int xfrm_state_update(struct xfrm_state *x); extern struct xfrm_state *xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto, unsigned short family); extern struct xfrm_state *xfrm_find_acq_byseq(u32 seq); -extern void xfrm_state_delete(struct xfrm_state *x); +extern int xfrm_state_delete(struct xfrm_state *x); extern void xfrm_state_flush(u8 proto); extern int xfrm_replay_check(struct xfrm_state *x, u32 seq); extern void xfrm_replay_advance(struct xfrm_state *x, u32 seq); diff --git a/net/key/af_key.c b/net/key/af_key.c index ce980aa94ed8..d086c117f5f0 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -1240,13 +1240,85 @@ static int pfkey_acquire(struct sock *sk, struct sk_buff *skb, struct sadb_msg * return 0; } +static inline int event2poltype(int event) +{ + switch (event) { + case XFRM_SAP_DELETED: + return SADB_X_SPDDELETE; + case XFRM_SAP_ADDED: + return SADB_X_SPDADD; + case XFRM_SAP_UPDATED: + return SADB_X_SPDUPDATE; + case XFRM_SAP_EXPIRED: + // return SADB_X_SPDEXPIRE; + default: + printk("pfkey: Unknown policy event %d\n", event); + break; + } + + return 0; +} + +static inline int event2keytype(int event) +{ + switch (event) { + case XFRM_SAP_DELETED: + return SADB_DELETE; + case XFRM_SAP_ADDED: + return SADB_ADD; + case XFRM_SAP_UPDATED: + return SADB_UPDATE; + case XFRM_SAP_EXPIRED: + return SADB_EXPIRE; + default: + printk("pfkey: Unknown SA event %d\n", event); + break; + } + + return 0; +} + +/* ADD/UPD/DEL */ +static int key_notify_sa(struct xfrm_state *x, struct km_event *c) +{ + struct sk_buff *skb; + struct sadb_msg *hdr; + int hsc = 3; + + if (c->event == XFRM_SAP_DELETED) + hsc = 0; + + if (c->event == XFRM_SAP_EXPIRED) { + if (c->data) + hsc = 2; + else + hsc = 1; + } + + skb = pfkey_xfrm_state2msg(x, 0, hsc); + + if (IS_ERR(skb)) + return PTR_ERR(skb); + + hdr = (struct sadb_msg *) skb->data; + hdr->sadb_msg_version = PF_KEY_V2; + hdr->sadb_msg_type = event2keytype(c->event); + hdr->sadb_msg_satype = pfkey_proto2satype(x->id.proto); + hdr->sadb_msg_errno = 0; + hdr->sadb_msg_reserved = 0; + hdr->sadb_msg_seq = c->seq; + hdr->sadb_msg_pid = c->pid; + + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL); + + return 0; +} static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; struct xfrm_state *x; int err; + struct km_event c; xfrm_probe_algs(); @@ -1254,6 +1326,7 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, if (IS_ERR(x)) return PTR_ERR(x); + xfrm_state_hold(x); if (hdr->sadb_msg_type == SADB_ADD) err = xfrm_state_add(x); else @@ -1265,27 +1338,23 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, return err; } - out_skb = pfkey_xfrm_state2msg(x, 0, 3); - if (IS_ERR(out_skb)) - return PTR_ERR(out_skb); /* XXX Should we return 0 here ? */ - - out_hdr = (struct sadb_msg *) out_skb->data; - out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = hdr->sadb_msg_type; - out_hdr->sadb_msg_satype = pfkey_proto2satype(x->id.proto); - out_hdr->sadb_msg_errno = 0; - out_hdr->sadb_msg_reserved = 0; - out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; - out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); + if (hdr->sadb_msg_type == SADB_ADD) + c.event = XFRM_SAP_ADDED; + else + c.event = XFRM_SAP_UPDATED; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + km_state_notify(x, &c); + xfrm_state_put(x); - return 0; + return err; } static int pfkey_delete(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { struct xfrm_state *x; + struct km_event c; + int err; if (!ext_hdrs[SADB_EXT_SA-1] || !present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], @@ -1301,13 +1370,19 @@ static int pfkey_delete(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h return -EPERM; } - xfrm_state_delete(x); - xfrm_state_put(x); + err = xfrm_state_delete(x); + if (err < 0) { + xfrm_state_put(x); + return err; + } - pfkey_broadcast(skb_clone(skb, GFP_KERNEL), GFP_KERNEL, - BROADCAST_ALL, sk); + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + c.event = XFRM_SAP_DELETED; + km_state_notify(x, &c); + xfrm_state_put(x); - return 0; + return err; } static int pfkey_get(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) @@ -1445,28 +1520,42 @@ static int pfkey_register(struct sock *sk, struct sk_buff *skb, struct sadb_msg return 0; } +static int key_notify_sa_flush(struct km_event *c) +{ + struct sk_buff *skb; + struct sadb_msg *hdr; + + skb = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_ATOMIC); + if (!skb) + return -ENOBUFS; + hdr = (struct sadb_msg *) skb_put(skb, sizeof(struct sadb_msg)); + hdr->sadb_msg_satype = pfkey_proto2satype(c->data); + hdr->sadb_msg_seq = c->seq; + hdr->sadb_msg_pid = c->pid; + hdr->sadb_msg_version = PF_KEY_V2; + hdr->sadb_msg_errno = (uint8_t) 0; + hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); + + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL); + + return 0; +} + static int pfkey_flush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { unsigned proto; - struct sk_buff *skb_out; - struct sadb_msg *hdr_out; + struct km_event c; proto = pfkey_satype2proto(hdr->sadb_msg_satype); if (proto == 0) return -EINVAL; - skb_out = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_KERNEL); - if (!skb_out) - return -ENOBUFS; - xfrm_state_flush(proto); - - hdr_out = (struct sadb_msg *) skb_put(skb_out, sizeof(struct sadb_msg)); - pfkey_hdr_dup(hdr_out, hdr); - hdr_out->sadb_msg_errno = (uint8_t) 0; - hdr_out->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); - - pfkey_broadcast(skb_out, GFP_KERNEL, BROADCAST_ALL, NULL); + c.data = proto; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + c.event = XFRM_SAP_FLUSHED; + km_state_notify(NULL, &c); return 0; } @@ -1859,6 +1948,35 @@ static void pfkey_xfrm_policy2msg(struct sk_buff *skb, struct xfrm_policy *xp, i hdr->sadb_msg_reserved = atomic_read(&xp->refcnt); } +static int key_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + struct sk_buff *out_skb; + struct sadb_msg *out_hdr; + int err; + + out_skb = pfkey_xfrm_policy2msg_prep(xp); + if (IS_ERR(out_skb)) { + err = PTR_ERR(out_skb); + goto out; + } + pfkey_xfrm_policy2msg(out_skb, xp, dir); + + out_hdr = (struct sadb_msg *) out_skb->data; + out_hdr->sadb_msg_version = PF_KEY_V2; + + if (c->data && c->event == XFRM_SAP_DELETED) + out_hdr->sadb_msg_type = SADB_X_SPDDELETE2; + else + out_hdr->sadb_msg_type = event2poltype(c->event); + out_hdr->sadb_msg_errno = 0; + out_hdr->sadb_msg_seq = c->seq; + out_hdr->sadb_msg_pid = c->pid; + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, NULL); +out: + return 0; + +} + static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { int err; @@ -1866,8 +1984,7 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h struct sadb_address *sa; struct sadb_x_policy *pol; struct xfrm_policy *xp; - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; + struct km_event c; if (!present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], ext_hdrs[SADB_EXT_ADDRESS_DST-1]) || @@ -1935,31 +2052,23 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h (err = parse_ipsecrequests(xp, pol)) < 0) goto out; - out_skb = pfkey_xfrm_policy2msg_prep(xp); - if (IS_ERR(out_skb)) { - err = PTR_ERR(out_skb); - goto out; - } - err = xfrm_policy_insert(pol->sadb_x_policy_dir-1, xp, hdr->sadb_msg_type != SADB_X_SPDUPDATE); if (err) { - kfree_skb(out_skb); - goto out; + kfree(xp); + return err; } - pfkey_xfrm_policy2msg(out_skb, xp, pol->sadb_x_policy_dir-1); + if (hdr->sadb_msg_type == SADB_X_SPDUPDATE) + c.event = XFRM_SAP_UPDATED; + else + c.event = XFRM_SAP_ADDED; - xfrm_pol_put(xp); + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; - out_hdr = (struct sadb_msg *) out_skb->data; - out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = hdr->sadb_msg_type; - out_hdr->sadb_msg_satype = 0; - out_hdr->sadb_msg_errno = 0; - out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; - out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); + km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); + xfrm_pol_put(xp); return 0; out: @@ -1973,9 +2082,8 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg struct sadb_address *sa; struct sadb_x_policy *pol; struct xfrm_policy *xp; - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; struct xfrm_selector sel; + struct km_event c; if (!present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], ext_hdrs[SADB_EXT_ADDRESS_DST-1]) || @@ -2010,25 +2118,40 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg err = 0; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + c.event = XFRM_SAP_DELETED; + km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); + + xfrm_pol_put(xp); + return err; +} + +static int key_pol_get_resp(struct sock *sk, struct xfrm_policy *xp, struct sadb_msg *hdr, int dir) +{ + int err; + struct sk_buff *out_skb; + struct sadb_msg *out_hdr; + err = 0; + out_skb = pfkey_xfrm_policy2msg_prep(xp); if (IS_ERR(out_skb)) { err = PTR_ERR(out_skb); goto out; } - pfkey_xfrm_policy2msg(out_skb, xp, pol->sadb_x_policy_dir-1); + pfkey_xfrm_policy2msg(out_skb, xp, dir); out_hdr = (struct sadb_msg *) out_skb->data; out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = SADB_X_SPDDELETE; + out_hdr->sadb_msg_type = hdr->sadb_msg_type; out_hdr->sadb_msg_satype = 0; out_hdr->sadb_msg_errno = 0; out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ONE, sk); err = 0; out: - xfrm_pol_put(xp); return err; } @@ -2037,8 +2160,7 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h int err; struct sadb_x_policy *pol; struct xfrm_policy *xp; - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; + struct km_event c; if ((pol = ext_hdrs[SADB_X_EXT_POLICY-1]) == NULL) return -EINVAL; @@ -2050,24 +2172,16 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h err = 0; - out_skb = pfkey_xfrm_policy2msg_prep(xp); - if (IS_ERR(out_skb)) { - err = PTR_ERR(out_skb); - goto out; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + if (hdr->sadb_msg_type == SADB_X_SPDDELETE2) { + c.data = 1; // to signal pfkey of SADB_X_SPDDELETE2 + c.event = XFRM_SAP_DELETED; + km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); + } else { + err = key_pol_get_resp(sk, xp, hdr, pol->sadb_x_policy_dir-1); } - pfkey_xfrm_policy2msg(out_skb, xp, pol->sadb_x_policy_dir-1); - out_hdr = (struct sadb_msg *) out_skb->data; - out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = hdr->sadb_msg_type; - out_hdr->sadb_msg_satype = 0; - out_hdr->sadb_msg_errno = 0; - out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; - out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); - err = 0; - -out: xfrm_pol_put(xp); return err; } @@ -2102,22 +2216,34 @@ static int pfkey_spddump(struct sock *sk, struct sk_buff *skb, struct sadb_msg * return xfrm_policy_walk(dump_sp, &data); } -static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) +static int key_notify_policy_flush(struct km_event *c) { struct sk_buff *skb_out; - struct sadb_msg *hdr_out; + struct sadb_msg *hdr; - skb_out = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_KERNEL); + skb_out = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_ATOMIC); if (!skb_out) return -ENOBUFS; + hdr = (struct sadb_msg *) skb_put(skb_out, sizeof(struct sadb_msg)); + hdr->sadb_msg_seq = c->seq; + hdr->sadb_msg_pid = c->pid; + hdr->sadb_msg_version = PF_KEY_V2; + hdr->sadb_msg_errno = (uint8_t) 0; + hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); + pfkey_broadcast(skb_out, GFP_ATOMIC, BROADCAST_ALL, NULL); + return 0; - xfrm_policy_flush(); +} + +static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) +{ + struct km_event c; - hdr_out = (struct sadb_msg *) skb_put(skb_out, sizeof(struct sadb_msg)); - pfkey_hdr_dup(hdr_out, hdr); - hdr_out->sadb_msg_errno = (uint8_t) 0; - hdr_out->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); - pfkey_broadcast(skb_out, GFP_KERNEL, BROADCAST_ALL, NULL); + xfrm_policy_flush(); + c.event = XFRM_SAP_FLUSHED; + c.pid = hdr->sadb_msg_pid; + c.seq = hdr->sadb_msg_seq; + km_policy_notify(NULL, 0, &c); return 0; } @@ -2317,11 +2443,23 @@ static void dump_esp_combs(struct sk_buff *skb, struct xfrm_tmpl *t) } } -static int pfkey_send_notify(struct xfrm_state *x, int hard) +static int key_notify_policy_expire(struct xfrm_policy *xp, struct km_event *c) +{ + return 0; +} + +static int key_notify_sa_expire(struct xfrm_state *x, struct km_event *c) { struct sk_buff *out_skb; struct sadb_msg *out_hdr; - int hsc = (hard ? 2 : 1); + int hard; + int hsc; + + hard = c->data; + if (hard) + hsc = 2; + else + hsc = 1; out_skb = pfkey_xfrm_state2msg(x, 0, hsc); if (IS_ERR(out_skb)) @@ -2340,6 +2478,44 @@ static int pfkey_send_notify(struct xfrm_state *x, int hard) return 0; } +static int pfkey_send_notify(struct xfrm_state *x, struct km_event *c) +{ + switch (c->event) { + case XFRM_SAP_EXPIRED: + return key_notify_sa_expire(x, c); + case XFRM_SAP_DELETED: + case XFRM_SAP_ADDED: + case XFRM_SAP_UPDATED: + return key_notify_sa(x, c); + case XFRM_SAP_FLUSHED: + return key_notify_sa_flush(c); + default: + printk("pfkey: Unknown SA event %d\n", c->event); + break; + } + + return 0; +} + +static int pfkey_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + switch (c->event) { + case XFRM_SAP_EXPIRED: + return key_notify_policy_expire(xp, c); + case XFRM_SAP_DELETED: + case XFRM_SAP_ADDED: + case XFRM_SAP_UPDATED: + return key_notify_policy(xp, dir, c); + case XFRM_SAP_FLUSHED: + return key_notify_policy_flush(c); + default: + printk("pfkey: Unknown policy event %d\n", c->event); + break; + } + + return 0; +} + static u32 get_acqseq(void) { u32 res; @@ -2856,6 +3032,7 @@ static struct xfrm_mgr pfkeyv2_mgr = .acquire = pfkey_send_acquire, .compile_policy = pfkey_compile_policy, .new_mapping = pfkey_send_new_mapping, + .notify_policy = pfkey_send_policy_notify, }; static void __exit ipsec_pfkey_exit(void) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index d11747c2a763..918a94c552a5 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -50,7 +50,7 @@ static DEFINE_SPINLOCK(xfrm_state_gc_lock); static int xfrm_state_gc_flush_bundles; -static void __xfrm_state_delete(struct xfrm_state *x); +static int __xfrm_state_delete(struct xfrm_state *x); static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned short family); static void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo); @@ -215,8 +215,10 @@ void __xfrm_state_destroy(struct xfrm_state *x) } EXPORT_SYMBOL(__xfrm_state_destroy); -static void __xfrm_state_delete(struct xfrm_state *x) +static int __xfrm_state_delete(struct xfrm_state *x) { + int err = -ESRCH; + if (x->km.state != XFRM_STATE_DEAD) { x->km.state = XFRM_STATE_DEAD; spin_lock(&xfrm_state_lock); @@ -245,14 +247,21 @@ static void __xfrm_state_delete(struct xfrm_state *x) * is what we are dropping here. */ atomic_dec(&x->refcnt); + err = 0; } + + return err; } -void xfrm_state_delete(struct xfrm_state *x) +int xfrm_state_delete(struct xfrm_state *x) { + int err; + spin_lock_bh(&x->lock); - __xfrm_state_delete(x); + err = __xfrm_state_delete(x); spin_unlock_bh(&x->lock); + + return err; } EXPORT_SYMBOL(xfrm_state_delete); @@ -796,34 +805,60 @@ EXPORT_SYMBOL(xfrm_replay_advance); static struct list_head xfrm_km_list = LIST_HEAD_INIT(xfrm_km_list); static DEFINE_RWLOCK(xfrm_km_lock); -static void km_state_expired(struct xfrm_state *x, int hard) +void km_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) { struct xfrm_mgr *km; - if (hard) - x->km.state = XFRM_STATE_EXPIRED; - else - x->km.dying = 1; + read_lock(&xfrm_km_lock); + list_for_each_entry(km, &xfrm_km_list, list) + if (km->notify_policy) + km->notify_policy(xp, dir, c); + read_unlock(&xfrm_km_lock); +} +void km_state_notify(struct xfrm_state *x, struct km_event *c) +{ + struct xfrm_mgr *km; read_lock(&xfrm_km_lock); list_for_each_entry(km, &xfrm_km_list, list) - km->notify(x, hard); + if (km->notify) + km->notify(x, c); read_unlock(&xfrm_km_lock); +} + +EXPORT_SYMBOL(km_policy_notify); +EXPORT_SYMBOL(km_state_notify); + +static void km_state_expired(struct xfrm_state *x, int hard) +{ + struct km_event c; + + if (hard) + x->km.state = XFRM_STATE_EXPIRED; + else + x->km.dying = 1; + c.data = hard; + c.event = XFRM_SAP_EXPIRED; + km_state_notify(x, &c); if (hard) wake_up(&km_waitq); } +/* + * We send to all registered managers regardless of failure + * We are happy with one success +*/ static int km_query(struct xfrm_state *x, struct xfrm_tmpl *t, struct xfrm_policy *pol) { - int err = -EINVAL; + int err = -EINVAL, acqret; struct xfrm_mgr *km; read_lock(&xfrm_km_lock); list_for_each_entry(km, &xfrm_km_list, list) { - err = km->acquire(x, t, pol, XFRM_POLICY_OUT); - if (!err) - break; + acqret = km->acquire(x, t, pol, XFRM_POLICY_OUT); + if (!acqret) + err = acqret; } read_unlock(&xfrm_km_lock); return err; @@ -848,13 +883,12 @@ EXPORT_SYMBOL(km_new_mapping); void km_policy_expired(struct xfrm_policy *pol, int dir, int hard) { - struct xfrm_mgr *km; + struct km_event c; - read_lock(&xfrm_km_lock); - list_for_each_entry(km, &xfrm_km_list, list) - if (km->notify_policy) - km->notify_policy(pol, dir, hard); - read_unlock(&xfrm_km_lock); + c.data = hard; + c.data = hard; + c.event = XFRM_SAP_EXPIRED; + km_policy_notify(pol, dir, &c); if (hard) wake_up(&km_waitq); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 97509011c274..bd8e6882c083 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -277,6 +277,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) struct xfrm_usersa_info *p = NLMSG_DATA(nlh); struct xfrm_state *x; int err; + struct km_event c; err = verify_newsa_info(p, (struct rtattr **) xfrma); if (err) @@ -286,6 +287,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) if (!x) return err; + xfrm_state_hold(x); if (nlh->nlmsg_type == XFRM_MSG_NEWSA) err = xfrm_state_add(x); else @@ -294,14 +296,27 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) if (err < 0) { x->km.state = XFRM_STATE_DEAD; xfrm_state_put(x); + return err; } + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + if (nlh->nlmsg_type == XFRM_MSG_NEWSA) + c.event = XFRM_SAP_ADDED; + else + c.event = XFRM_SAP_UPDATED; + + km_state_notify(x, &c); + xfrm_state_put(x); + return err; } static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) { struct xfrm_state *x; + int err; + struct km_event c; struct xfrm_usersa_id *p = NLMSG_DATA(nlh); x = xfrm_state_lookup(&p->daddr, p->spi, p->proto, p->family); @@ -313,10 +328,19 @@ static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) return -EPERM; } - xfrm_state_delete(x); + err = xfrm_state_delete(x); + if (err < 0) { + xfrm_state_put(x); + return err; + } + + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + c.event = XFRM_SAP_DELETED; + km_state_notify(x, &c); xfrm_state_put(x); - return 0; + return err; } static void copy_to_user_state(struct xfrm_state *x, struct xfrm_usersa_info *p) @@ -681,6 +705,7 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr { struct xfrm_userpolicy_info *p = NLMSG_DATA(nlh); struct xfrm_policy *xp; + struct km_event c; int err; int excl; @@ -692,6 +717,10 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr if (!xp) return err; + /* shouldnt excl be based on nlh flags?? + * Aha! this is anti-netlink really i.e more pfkey derived + * in netlink excl is a flag and you wouldnt need + * a type XFRM_MSG_UPDPOLICY - JHS */ excl = nlh->nlmsg_type == XFRM_MSG_NEWPOLICY; err = xfrm_policy_insert(p->dir, xp, excl); if (err) { @@ -699,6 +728,15 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr return err; } + if (!excl) + c.event = XFRM_SAP_UPDATED; + else + c.event = XFRM_SAP_ADDED; + + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_policy_notify(xp, p->dir, &c); + xfrm_pol_put(xp); return 0; @@ -816,6 +854,7 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr struct xfrm_policy *xp; struct xfrm_userpolicy_id *p; int err; + struct km_event c; int delete; p = NLMSG_DATA(nlh); @@ -843,6 +882,11 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr NETLINK_CB(skb).pid, MSG_DONTWAIT); } + } else { + c.event = XFRM_SAP_DELETED; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_policy_notify(xp, p->dir, &c); } xfrm_pol_put(xp); @@ -852,15 +896,28 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr static int xfrm_flush_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) { + struct km_event c; struct xfrm_usersa_flush *p = NLMSG_DATA(nlh); xfrm_state_flush(p->proto); + c.data = p->proto; + c.event = XFRM_SAP_FLUSHED; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_state_notify(NULL, &c); + return 0; } static int xfrm_flush_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) { + struct km_event c; + xfrm_policy_flush(); + c.event = XFRM_SAP_FLUSHED; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_policy_notify(NULL, 0, &c); return 0; } @@ -1069,10 +1126,12 @@ nlmsg_failure: return -1; } -static int xfrm_send_state_notify(struct xfrm_state *x, int hard) +static int xfrm_exp_state_notify(struct xfrm_state *x, struct km_event *c) { struct sk_buff *skb; + int hard = c ->data; + /* fix to do alloc using NLM macros */ skb = alloc_skb(sizeof(struct xfrm_user_expire) + 16, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; @@ -1085,6 +1144,122 @@ static int xfrm_send_state_notify(struct xfrm_state *x, int hard) return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_EXPIRE, GFP_ATOMIC); } +static int xfrm_notify_sa_flush(struct km_event *c) +{ + struct xfrm_usersa_flush *p; + struct nlmsghdr *nlh; + struct sk_buff *skb; + unsigned char *b; + int len = NLMSG_LENGTH(sizeof(struct xfrm_usersa_flush)); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + nlh = NLMSG_PUT(skb, c->pid, c->seq, + XFRM_MSG_FLUSHSA, sizeof(*p)); + nlh->nlmsg_flags = 0; + + p = NLMSG_DATA(nlh); + p->proto = c->data; + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_SA, GFP_ATOMIC); + +nlmsg_failure: + kfree_skb(skb); + return -1; +} + +static int inline xfrm_sa_len(struct xfrm_state *x) +{ + int l = NLMSG_LENGTH(sizeof(struct xfrm_usersa_info)); + if (x->aalg) + l += RTA_SPACE(sizeof(*x->aalg) + (x->aalg->alg_key_len+7)/8); + if (x->ealg) + l += RTA_SPACE(sizeof(*x->ealg) + (x->ealg->alg_key_len+7)/8); + if (x->calg) + l += RTA_SPACE(sizeof(*x->calg)); + if (x->encap) + l += RTA_SPACE(sizeof(*x->encap)); + + return l; +} + +static int xfrm_notify_sa(struct xfrm_state *x, struct km_event *c) +{ + struct xfrm_usersa_info *p; + struct nlmsghdr *nlh; + struct sk_buff *skb; + u32 nlt; + unsigned char *b; + int len = xfrm_sa_len(x); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + if (c->event == XFRM_SAP_ADDED) + nlt = XFRM_MSG_NEWSA; + else if (c->event == XFRM_SAP_UPDATED) + nlt = XFRM_MSG_UPDSA; + else if (c->event == XFRM_SAP_DELETED) + nlt = XFRM_MSG_DELSA; + else + goto nlmsg_failure; + + nlh = NLMSG_PUT(skb, c->pid, c->seq, nlt, sizeof(*p)); + nlh->nlmsg_flags = 0; + + p = NLMSG_DATA(nlh); + copy_to_user_state(x, p); + + if (x->aalg) + RTA_PUT(skb, XFRMA_ALG_AUTH, + sizeof(*(x->aalg))+(x->aalg->alg_key_len+7)/8, x->aalg); + if (x->ealg) + RTA_PUT(skb, XFRMA_ALG_CRYPT, + sizeof(*(x->ealg))+(x->ealg->alg_key_len+7)/8, x->ealg); + if (x->calg) + RTA_PUT(skb, XFRMA_ALG_COMP, sizeof(*(x->calg)), x->calg); + + if (x->encap) + RTA_PUT(skb, XFRMA_ENCAP, sizeof(*x->encap), x->encap); + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_SA, GFP_ATOMIC); + +nlmsg_failure: +rtattr_failure: + kfree_skb(skb); + return -1; +} + +static int xfrm_send_state_notify(struct xfrm_state *x, struct km_event *c) +{ + + switch (c->event) { + case XFRM_SAP_EXPIRED: + return xfrm_exp_state_notify(x, c); + case XFRM_SAP_DELETED: + case XFRM_SAP_UPDATED: + case XFRM_SAP_ADDED: + return xfrm_notify_sa(x, c); + case XFRM_SAP_FLUSHED: + return xfrm_notify_sa_flush(c); + default: + printk("xfrm_user: Unknown SA event %d\n", c->event); + break; + } + + return 0; + +} + static int build_acquire(struct sk_buff *skb, struct xfrm_state *x, struct xfrm_tmpl *xt, struct xfrm_policy *xp, int dir) @@ -1218,7 +1393,7 @@ nlmsg_failure: return -1; } -static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, int hard) +static int xfrm_exp_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) { struct sk_buff *skb; size_t len; @@ -1229,7 +1404,7 @@ static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, int hard) if (skb == NULL) return -ENOMEM; - if (build_polexpire(skb, xp, dir, hard) < 0) + if (build_polexpire(skb, xp, dir, c->data) < 0) BUG(); NETLINK_CB(skb).dst_groups = XFRMGRP_EXPIRE; @@ -1237,6 +1412,93 @@ static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, int hard) return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_EXPIRE, GFP_ATOMIC); } +static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + struct xfrm_userpolicy_info *p; + struct nlmsghdr *nlh; + struct sk_buff *skb; + u32 nlt = 0 ; + unsigned char *b; + int len = RTA_SPACE(sizeof(struct xfrm_user_tmpl) * xp->xfrm_nr); + len += NLMSG_SPACE(sizeof(struct xfrm_userpolicy_info)); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + if (c->event == XFRM_SAP_ADDED) + nlt = XFRM_MSG_NEWPOLICY; + else if (c->event == XFRM_SAP_UPDATED) + nlt = XFRM_MSG_UPDPOLICY; + else if (c->event == XFRM_SAP_DELETED) + nlt = XFRM_MSG_DELPOLICY; + else + goto nlmsg_failure; + + nlh = NLMSG_PUT(skb, c->pid, c->seq, nlt, sizeof(*p)); + + p = NLMSG_DATA(nlh); + + nlh->nlmsg_flags = 0; + + copy_to_user_policy(xp, p, dir); + if (copy_to_user_tmpl(xp, skb) < 0) + goto nlmsg_failure; + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_POLICY, GFP_ATOMIC); + +nlmsg_failure: + kfree_skb(skb); + return -1; +} + +static int xfrm_notify_policy_flush(struct km_event *c) +{ + struct nlmsghdr *nlh; + struct sk_buff *skb; + unsigned char *b; + int len = NLMSG_LENGTH(0); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + + nlh = NLMSG_PUT(skb, c->pid, c->seq, XFRM_MSG_FLUSHPOLICY, 0); + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_POLICY, GFP_ATOMIC); + +nlmsg_failure: + kfree_skb(skb); + return -1; +} + +static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + + switch (c->event) { + case XFRM_SAP_ADDED: + case XFRM_SAP_UPDATED: + case XFRM_SAP_DELETED: + return xfrm_notify_policy(xp, dir, c); + case XFRM_SAP_FLUSHED: + return xfrm_notify_policy_flush(c); + case XFRM_SAP_EXPIRED: + return xfrm_exp_policy_notify(xp, dir, c); + default: + printk("xfrm_user: Unknown Policy event %d\n", c->event); + } + + return 0; + +} + static struct xfrm_mgr netlink_mgr = { .id = "netlink", .notify = xfrm_send_state_notify, -- cgit v1.2.3 From 1944972d3bb651474a5021c9da8d0166ae19f1eb Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 18 Jun 2005 22:46:19 -0700 Subject: [SLAB] Introduce kmem_cache_name This is for use with slab users that pass a dynamically allocated slab name in kmem_cache_create, so that before destroying the slab one can retrieve the name and free its memory. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/slab.h | 1 + mm/slab.c | 6 ++++++ 2 files changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/slab.h b/include/linux/slab.h index 7d66385ae750..76cf7e60216c 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -64,6 +64,7 @@ extern int kmem_cache_shrink(kmem_cache_t *); extern void *kmem_cache_alloc(kmem_cache_t *, unsigned int __nocast); extern void kmem_cache_free(kmem_cache_t *, void *); extern unsigned int kmem_cache_size(kmem_cache_t *); +extern const char *kmem_cache_name(kmem_cache_t *); extern kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags); /* Size description struct for general caches. */ diff --git a/mm/slab.c b/mm/slab.c index 840742641152..c78d343b3c5f 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2620,6 +2620,12 @@ unsigned int kmem_cache_size(kmem_cache_t *cachep) } EXPORT_SYMBOL(kmem_cache_size); +const char *kmem_cache_name(kmem_cache_t *cachep) +{ + return cachep->name; +} +EXPORT_SYMBOL_GPL(kmem_cache_name); + struct ccupdate_struct { kmem_cache_t *cachep; struct array_cache *new[NR_CPUS]; -- cgit v1.2.3 From 2e6599cb899ba4b133f42cbf9d2b1883d2dc583a Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 18 Jun 2005 22:46:52 -0700 Subject: [NET] Generalise TCP's struct open_request minisock infrastructure Kept this first changeset minimal, without changing existing names to ease peer review. Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn has two new members: ->slab, that replaces tcp_openreq_cachep ->obj_size, to inform the size of the openreq descendant for a specific protocol The protocol specific fields in struct open_request were moved to a class hierarchy, with the things that are common to all connection oriented PF_INET protocols in struct inet_request_sock, the TCP ones in tcp_request_sock, that is an inet_request_sock, that is an open_request. I.e. this uses the same approach used for the struct sock class hierarchy, with sk_prot indicating if the protocol wants to use the open_request infrastructure by filling in sk_prot->rsk_prot with an or_calltable. Results? Performance is improved and TCP v4 now uses only 64 bytes per open request minisock, down from 96 without this patch :-) Next changeset will rename some of the structs, fields and functions mentioned above, struct or_calltable is way unclear, better name it struct request_sock_ops, s/struct open_request/struct request_sock/g, etc. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/ip.h | 21 ++++++++++ include/linux/ipv6.h | 13 +++++++ include/linux/tcp.h | 11 ++++++ include/net/request_sock.h | 77 +++++++++++++++++++++++++++++++++++++ include/net/sock.h | 4 ++ include/net/tcp.h | 87 ++++++----------------------------------- include/net/tcp_ecn.h | 7 ++-- net/core/sock.c | 35 +++++++++++++++++ net/ipv4/syncookies.c | 39 ++++++++++--------- net/ipv4/tcp.c | 9 ----- net/ipv4/tcp_diag.c | 25 ++++++------ net/ipv4/tcp_ipv4.c | 72 +++++++++++++++++++--------------- net/ipv4/tcp_minisocks.c | 48 ++++++++++++----------- net/ipv4/tcp_output.c | 25 ++++++------ net/ipv4/tcp_timer.c | 2 +- net/ipv6/tcp_ipv6.c | 96 ++++++++++++++++++++++++---------------------- 16 files changed, 341 insertions(+), 230 deletions(-) create mode 100644 include/net/request_sock.h (limited to 'include/linux') diff --git a/include/linux/ip.h b/include/linux/ip.h index 8438c68591f9..d5b7c907204e 100644 --- a/include/linux/ip.h +++ b/include/linux/ip.h @@ -81,6 +81,7 @@ #ifdef __KERNEL__ #include #include +#include #include #include #include @@ -107,6 +108,26 @@ struct ip_options { #define optlength(opt) (sizeof(struct ip_options) + opt->optlen) +struct inet_request_sock { + struct open_request req; + u32 loc_addr; + u32 rmt_addr; + u16 rmt_port; + u16 snd_wscale : 4, + rcv_wscale : 4, + tstamp_ok : 1, + sack_ok : 1, + wscale_ok : 1, + ecn_ok : 1, + acked : 1; + struct ip_options *opt; +}; + +static inline struct inet_request_sock *inet_rsk(const struct open_request *sk) +{ + return (struct inet_request_sock *)sk; +} + struct ipv6_pinfo; struct inet_sock { diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index ab0d0efbf240..98acdbf3d446 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -193,6 +193,19 @@ struct inet6_skb_parm { #define IP6CB(skb) ((struct inet6_skb_parm*)((skb)->cb)) +struct tcp6_request_sock { + struct tcp_request_sock req; + struct in6_addr loc_addr; + struct in6_addr rmt_addr; + struct sk_buff *pktopts; + int iif; +}; + +static inline struct tcp6_request_sock *tcp6_rsk(const struct open_request *sk) +{ + return (struct tcp6_request_sock *)sk; +} + /** * struct ipv6_pinfo - ipv6 private area * diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 14a55e3e3a50..86771b37b80d 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -230,6 +230,17 @@ struct tcp_options_received { __u16 mss_clamp; /* Maximal mss, negotiated at connection setup */ }; +struct tcp_request_sock { + struct inet_request_sock req; + __u32 rcv_isn; + __u32 snt_isn; +}; + +static inline struct tcp_request_sock *tcp_rsk(const struct open_request *req) +{ + return (struct tcp_request_sock *)req; +} + struct tcp_sock { /* inet_sock has to be the first member of tcp_sock */ struct inet_sock inet; diff --git a/include/net/request_sock.h b/include/net/request_sock.h new file mode 100644 index 000000000000..9502f5587931 --- /dev/null +++ b/include/net/request_sock.h @@ -0,0 +1,77 @@ +/* + * NET Generic infrastructure for Network protocols. + * + * Definitions for request_sock + * + * Authors: Arnaldo Carvalho de Melo + * + * From code originally in include/net/tcp.h + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#ifndef _REQUEST_SOCK_H +#define _REQUEST_SOCK_H + +#include +#include +#include + +struct open_request; +struct sk_buff; +struct dst_entry; +struct proto; + +struct or_calltable { + int family; + kmem_cache_t *slab; + int obj_size; + int (*rtx_syn_ack)(struct sock *sk, + struct open_request *req, + struct dst_entry *dst); + void (*send_ack)(struct sk_buff *skb, + struct open_request *req); + void (*send_reset)(struct sk_buff *skb); + void (*destructor)(struct open_request *req); +}; + +/* struct open_request - mini sock to represent a connection request + */ +struct open_request { + struct open_request *dl_next; /* Must be first member! */ + u16 mss; + u8 retrans; + u8 __pad; + /* The following two fields can be easily recomputed I think -AK */ + u32 window_clamp; /* window clamp at creation time */ + u32 rcv_wnd; /* rcv_wnd offered first time */ + u32 ts_recent; + unsigned long expires; + struct or_calltable *class; + struct sock *sk; +}; + +static inline struct open_request *tcp_openreq_alloc(struct or_calltable *class) +{ + struct open_request *req = kmem_cache_alloc(class->slab, SLAB_ATOMIC); + + if (req != NULL) + req->class = class; + + return req; +} + +static inline void tcp_openreq_fastfree(struct open_request *req) +{ + kmem_cache_free(req->class->slab, req); +} + +static inline void tcp_openreq_free(struct open_request *req) +{ + req->class->destructor(req); + tcp_openreq_fastfree(req); +} + +#endif /* _REQUEST_SOCK_H */ diff --git a/include/net/sock.h b/include/net/sock.h index a9ef3a6a13f3..6919276af8af 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -484,6 +484,8 @@ extern void sk_stream_kill_queues(struct sock *sk); extern int sk_wait_data(struct sock *sk, long *timeo); +struct or_calltable; + /* Networking protocol blocks we attach to sockets. * socket layer -> transport layer interface * transport -> network interface is defined by struct inet_proto @@ -547,6 +549,8 @@ struct proto { kmem_cache_t *slab; unsigned int obj_size; + struct or_calltable *rsk_prot; + struct module *owner; char name[32]; diff --git a/include/net/tcp.h b/include/net/tcp.h index e71f8ba3e101..d438ba566b89 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include @@ -613,74 +614,6 @@ extern atomic_t tcp_memory_allocated; extern atomic_t tcp_sockets_allocated; extern int tcp_memory_pressure; -struct open_request; - -struct or_calltable { - int family; - int (*rtx_syn_ack) (struct sock *sk, struct open_request *req, struct dst_entry*); - void (*send_ack) (struct sk_buff *skb, struct open_request *req); - void (*destructor) (struct open_request *req); - void (*send_reset) (struct sk_buff *skb); -}; - -struct tcp_v4_open_req { - __u32 loc_addr; - __u32 rmt_addr; - struct ip_options *opt; -}; - -#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) -struct tcp_v6_open_req { - struct in6_addr loc_addr; - struct in6_addr rmt_addr; - struct sk_buff *pktopts; - int iif; -}; -#endif - -/* this structure is too big */ -struct open_request { - struct open_request *dl_next; /* Must be first member! */ - __u32 rcv_isn; - __u32 snt_isn; - __u16 rmt_port; - __u16 mss; - __u8 retrans; - __u8 __pad; - __u16 snd_wscale : 4, - rcv_wscale : 4, - tstamp_ok : 1, - sack_ok : 1, - wscale_ok : 1, - ecn_ok : 1, - acked : 1; - /* The following two fields can be easily recomputed I think -AK */ - __u32 window_clamp; /* window clamp at creation time */ - __u32 rcv_wnd; /* rcv_wnd offered first time */ - __u32 ts_recent; - unsigned long expires; - struct or_calltable *class; - struct sock *sk; - union { - struct tcp_v4_open_req v4_req; -#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) - struct tcp_v6_open_req v6_req; -#endif - } af; -}; - -/* SLAB cache for open requests. */ -extern kmem_cache_t *tcp_openreq_cachep; - -#define tcp_openreq_alloc() kmem_cache_alloc(tcp_openreq_cachep, SLAB_ATOMIC) -#define tcp_openreq_fastfree(req) kmem_cache_free(tcp_openreq_cachep, req) - -static inline void tcp_openreq_free(struct open_request *req) -{ - req->class->destructor(req); - tcp_openreq_fastfree(req); -} - #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) #define TCP_INET_FAMILY(fam) ((fam) == AF_INET) #else @@ -1832,17 +1765,19 @@ static __inline__ void tcp_openreq_init(struct open_request *req, struct tcp_options_received *rx_opt, struct sk_buff *skb) { + struct inet_request_sock *ireq = inet_rsk(req); + req->rcv_wnd = 0; /* So that tcp_send_synack() knows! */ - req->rcv_isn = TCP_SKB_CB(skb)->seq; + tcp_rsk(req)->rcv_isn = TCP_SKB_CB(skb)->seq; req->mss = rx_opt->mss_clamp; req->ts_recent = rx_opt->saw_tstamp ? rx_opt->rcv_tsval : 0; - req->tstamp_ok = rx_opt->tstamp_ok; - req->sack_ok = rx_opt->sack_ok; - req->snd_wscale = rx_opt->snd_wscale; - req->wscale_ok = rx_opt->wscale_ok; - req->acked = 0; - req->ecn_ok = 0; - req->rmt_port = skb->h.th->source; + ireq->tstamp_ok = rx_opt->tstamp_ok; + ireq->sack_ok = rx_opt->sack_ok; + ireq->snd_wscale = rx_opt->snd_wscale; + ireq->wscale_ok = rx_opt->wscale_ok; + ireq->acked = 0; + ireq->ecn_ok = 0; + ireq->rmt_port = skb->h.th->source; } extern void tcp_enter_memory_pressure(void); diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h index dc1456389a97..94ad970e844a 100644 --- a/include/net/tcp_ecn.h +++ b/include/net/tcp_ecn.h @@ -2,6 +2,7 @@ #define _NET_TCP_ECN_H_ 1 #include +#include #define TCP_HP_BITS (~(TCP_RESERVED_BITS|TCP_FLAG_PSH)) @@ -40,7 +41,7 @@ static inline void TCP_ECN_send_syn(struct sock *sk, struct tcp_sock *tp, static __inline__ void TCP_ECN_make_synack(struct open_request *req, struct tcphdr *th) { - if (req->ecn_ok) + if (inet_rsk(req)->ecn_ok) th->ece = 1; } @@ -113,14 +114,14 @@ static inline int TCP_ECN_rcv_ecn_echo(struct tcp_sock *tp, struct tcphdr *th) static inline void TCP_ECN_openreq_child(struct tcp_sock *tp, struct open_request *req) { - tp->ecn_flags = req->ecn_ok ? TCP_ECN_OK : 0; + tp->ecn_flags = inet_rsk(req)->ecn_ok ? TCP_ECN_OK : 0; } static __inline__ void TCP_ECN_create_request(struct open_request *req, struct tcphdr *th) { if (sysctl_tcp_ecn && th->ece && th->cwr) - req->ecn_ok = 1; + inet_rsk(req)->ecn_ok = 1; } #endif diff --git a/net/core/sock.c b/net/core/sock.c index 96e00b08698f..a6ec3ada7f9e 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -118,6 +118,7 @@ #include #include #include +#include #include #include #include @@ -1363,6 +1364,7 @@ static LIST_HEAD(proto_list); int proto_register(struct proto *prot, int alloc_slab) { + char *request_sock_slab_name; int rc = -ENOBUFS; if (alloc_slab) { @@ -1374,6 +1376,25 @@ int proto_register(struct proto *prot, int alloc_slab) prot->name); goto out; } + + if (prot->rsk_prot != NULL) { + static const char mask[] = "request_sock_%s"; + + request_sock_slab_name = kmalloc(strlen(prot->name) + sizeof(mask) - 1, GFP_KERNEL); + if (request_sock_slab_name == NULL) + goto out_free_sock_slab; + + sprintf(request_sock_slab_name, mask, prot->name); + prot->rsk_prot->slab = kmem_cache_create(request_sock_slab_name, + prot->rsk_prot->obj_size, 0, + SLAB_HWCACHE_ALIGN, NULL, NULL); + + if (prot->rsk_prot->slab == NULL) { + printk(KERN_CRIT "%s: Can't create request sock SLAB cache!\n", + prot->name); + goto out_free_request_sock_slab_name; + } + } } write_lock(&proto_list_lock); @@ -1382,6 +1403,12 @@ int proto_register(struct proto *prot, int alloc_slab) rc = 0; out: return rc; +out_free_request_sock_slab_name: + kfree(request_sock_slab_name); +out_free_sock_slab: + kmem_cache_destroy(prot->slab); + prot->slab = NULL; + goto out; } EXPORT_SYMBOL(proto_register); @@ -1395,6 +1422,14 @@ void proto_unregister(struct proto *prot) prot->slab = NULL; } + if (prot->rsk_prot != NULL && prot->rsk_prot->slab != NULL) { + const char *name = kmem_cache_name(prot->rsk_prot->slab); + + kmem_cache_destroy(prot->rsk_prot->slab); + kfree(name); + prot->rsk_prot->slab = NULL; + } + list_del(&prot->node); write_unlock(&proto_list_lock); } diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index e923d2f021aa..dd47e6da6fb3 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -190,6 +190,8 @@ static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct ip_options *opt) { + struct inet_request_sock *ireq; + struct tcp_request_sock *treq; struct tcp_sock *tp = tcp_sk(sk); __u32 cookie = ntohl(skb->h.th->ack_seq) - 1; struct sock *ret = sk; @@ -209,19 +211,20 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV); - req = tcp_openreq_alloc(); ret = NULL; + req = tcp_openreq_alloc(&or_ipv4); /* for safety */ if (!req) goto out; - req->rcv_isn = htonl(skb->h.th->seq) - 1; - req->snt_isn = cookie; + ireq = inet_rsk(req); + treq = tcp_rsk(req); + treq->rcv_isn = htonl(skb->h.th->seq) - 1; + treq->snt_isn = cookie; req->mss = mss; - req->rmt_port = skb->h.th->source; - req->af.v4_req.loc_addr = skb->nh.iph->daddr; - req->af.v4_req.rmt_addr = skb->nh.iph->saddr; - req->class = &or_ipv4; /* for savety */ - req->af.v4_req.opt = NULL; + ireq->rmt_port = skb->h.th->source; + ireq->loc_addr = skb->nh.iph->daddr; + ireq->rmt_addr = skb->nh.iph->saddr; + ireq->opt = NULL; /* We throwed the options of the initial SYN away, so we hope * the ACK carries the same options again (see RFC1122 4.2.3.8) @@ -229,17 +232,15 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, if (opt && opt->optlen) { int opt_size = sizeof(struct ip_options) + opt->optlen; - req->af.v4_req.opt = kmalloc(opt_size, GFP_ATOMIC); - if (req->af.v4_req.opt) { - if (ip_options_echo(req->af.v4_req.opt, skb)) { - kfree(req->af.v4_req.opt); - req->af.v4_req.opt = NULL; - } + ireq->opt = kmalloc(opt_size, GFP_ATOMIC); + if (ireq->opt != NULL && ip_options_echo(ireq->opt, skb)) { + kfree(ireq->opt); + ireq->opt = NULL; } } - req->snd_wscale = req->rcv_wscale = req->tstamp_ok = 0; - req->wscale_ok = req->sack_ok = 0; + ireq->snd_wscale = ireq->rcv_wscale = ireq->tstamp_ok = 0; + ireq->wscale_ok = ireq->sack_ok = 0; req->expires = 0UL; req->retrans = 0; @@ -253,8 +254,8 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct flowi fl = { .nl_u = { .ip4_u = { .daddr = ((opt && opt->srr) ? opt->faddr : - req->af.v4_req.rmt_addr), - .saddr = req->af.v4_req.loc_addr, + ireq->rmt_addr), + .saddr = ireq->loc_addr, .tos = RT_CONN_FLAGS(sk) } }, .proto = IPPROTO_TCP, .uli_u = { .ports = @@ -272,7 +273,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, &req->rcv_wnd, &req->window_clamp, 0, &rcv_wscale); /* BTW win scale with syncookies is 0 by definition */ - req->rcv_wscale = rcv_wscale; + ireq->rcv_wscale = rcv_wscale; ret = get_cookie_sock(sk, skb, req, &rt->u.dst); out: return ret; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 0d9a4fd5f1a4..a3cabfa2022a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -271,7 +271,6 @@ int sysctl_tcp_fin_timeout = TCP_FIN_TIMEOUT; DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics); -kmem_cache_t *tcp_openreq_cachep; kmem_cache_t *tcp_bucket_cachep; kmem_cache_t *tcp_timewait_cachep; @@ -2271,13 +2270,6 @@ void __init tcp_init(void) __skb_cb_too_small_for_tcp(sizeof(struct tcp_skb_cb), sizeof(skb->cb)); - tcp_openreq_cachep = kmem_cache_create("tcp_open_request", - sizeof(struct open_request), - 0, SLAB_HWCACHE_ALIGN, - NULL, NULL); - if (!tcp_openreq_cachep) - panic("tcp_init: Cannot alloc open_request cache."); - tcp_bucket_cachep = kmem_cache_create("tcp_bind_bucket", sizeof(struct tcp_bind_bucket), 0, SLAB_HWCACHE_ALIGN, @@ -2374,7 +2366,6 @@ EXPORT_SYMBOL(tcp_destroy_sock); EXPORT_SYMBOL(tcp_disconnect); EXPORT_SYMBOL(tcp_getsockopt); EXPORT_SYMBOL(tcp_ioctl); -EXPORT_SYMBOL(tcp_openreq_cachep); EXPORT_SYMBOL(tcp_poll); EXPORT_SYMBOL(tcp_read_sock); EXPORT_SYMBOL(tcp_recvmsg); diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 8faa8948f75c..700ff2413588 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -458,6 +458,7 @@ static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, struct open_request *req, u32 pid, u32 seq) { + const struct inet_request_sock *ireq = inet_rsk(req); struct inet_sock *inet = inet_sk(sk); unsigned char *b = skb->tail; struct tcpdiagmsg *r; @@ -482,9 +483,9 @@ static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, tmo = 0; r->id.tcpdiag_sport = inet->sport; - r->id.tcpdiag_dport = req->rmt_port; - r->id.tcpdiag_src[0] = req->af.v4_req.loc_addr; - r->id.tcpdiag_dst[0] = req->af.v4_req.rmt_addr; + r->id.tcpdiag_dport = ireq->rmt_port; + r->id.tcpdiag_src[0] = ireq->loc_addr; + r->id.tcpdiag_dst[0] = ireq->rmt_addr; r->tcpdiag_expires = jiffies_to_msecs(tmo), r->tcpdiag_rqueue = 0; r->tcpdiag_wqueue = 0; @@ -493,9 +494,9 @@ static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, #ifdef CONFIG_IP_TCPDIAG_IPV6 if (r->tcpdiag_family == AF_INET6) { ipv6_addr_copy((struct in6_addr *)r->id.tcpdiag_src, - &req->af.v6_req.loc_addr); + &tcp6_rsk(req)->loc_addr); ipv6_addr_copy((struct in6_addr *)r->id.tcpdiag_dst, - &req->af.v6_req.rmt_addr); + &tcp6_rsk(req)->rmt_addr); } #endif nlh->nlmsg_len = skb->tail - b; @@ -545,9 +546,11 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, reqnum = 0; for (req = head; req; reqnum++, req = req->dl_next) { + struct inet_request_sock *ireq = inet_rsk(req); + if (reqnum < s_reqnum) continue; - if (r->id.tcpdiag_dport != req->rmt_port && + if (r->id.tcpdiag_dport != ireq->rmt_port && r->id.tcpdiag_dport) continue; @@ -555,16 +558,16 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, entry.saddr = #ifdef CONFIG_IP_TCPDIAG_IPV6 (entry.family == AF_INET6) ? - req->af.v6_req.loc_addr.s6_addr32 : + tcp6_rsk(req)->loc_addr.s6_addr32 : #endif - &req->af.v4_req.loc_addr; + &ireq->loc_addr; entry.daddr = #ifdef CONFIG_IP_TCPDIAG_IPV6 (entry.family == AF_INET6) ? - req->af.v6_req.rmt_addr.s6_addr32 : + tcp6_rsk(req)->rmt_addr.s6_addr32 : #endif - &req->af.v4_req.rmt_addr; - entry.dport = ntohs(req->rmt_port); + &ireq->rmt_addr; + entry.dport = ntohs(ireq->rmt_port); if (!tcpdiag_bc_run(RTA_DATA(bc), RTA_PAYLOAD(bc), &entry)) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index dad98e4a5043..e156be90df14 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -880,9 +880,11 @@ static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, for (prev = &lopt->syn_table[tcp_v4_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; prev = &req->dl_next) { - if (req->rmt_port == rport && - req->af.v4_req.rmt_addr == raddr && - req->af.v4_req.loc_addr == laddr && + const struct inet_request_sock *ireq = inet_rsk(req); + + if (ireq->rmt_port == rport && + ireq->rmt_addr == raddr && + ireq->loc_addr == laddr && TCP_INET_FAMILY(req->class->family)) { BUG_TRAP(!req->sk); *prevp = prev; @@ -897,7 +899,7 @@ static void tcp_v4_synq_add(struct sock *sk, struct open_request *req) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; - u32 h = tcp_v4_synq_hash(req->af.v4_req.rmt_addr, req->rmt_port, lopt->hash_rnd); + u32 h = tcp_v4_synq_hash(inet_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); req->expires = jiffies + TCP_TIMEOUT_INIT; req->retrans = 0; @@ -1065,7 +1067,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info) */ BUG_TRAP(!req->sk); - if (seq != req->snt_isn) { + if (seq != tcp_rsk(req)->snt_isn) { NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS); goto out; } @@ -1256,7 +1258,7 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) static void tcp_v4_or_send_ack(struct sk_buff *skb, struct open_request *req) { - tcp_v4_send_ack(skb, req->snt_isn + 1, req->rcv_isn + 1, req->rcv_wnd, + tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -1264,18 +1266,19 @@ static struct dst_entry* tcp_v4_route_req(struct sock *sk, struct open_request *req) { struct rtable *rt; - struct ip_options *opt = req->af.v4_req.opt; + const struct inet_request_sock *ireq = inet_rsk(req); + struct ip_options *opt = inet_rsk(req)->opt; struct flowi fl = { .oif = sk->sk_bound_dev_if, .nl_u = { .ip4_u = { .daddr = ((opt && opt->srr) ? opt->faddr : - req->af.v4_req.rmt_addr), - .saddr = req->af.v4_req.loc_addr, + ireq->rmt_addr), + .saddr = ireq->loc_addr, .tos = RT_CONN_FLAGS(sk) } }, .proto = IPPROTO_TCP, .uli_u = { .ports = { .sport = inet_sk(sk)->sport, - .dport = req->rmt_port } } }; + .dport = ireq->rmt_port } } }; if (ip_route_output_flow(&rt, &fl, sk, 0)) { IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); @@ -1297,6 +1300,7 @@ static struct dst_entry* tcp_v4_route_req(struct sock *sk, static int tcp_v4_send_synack(struct sock *sk, struct open_request *req, struct dst_entry *dst) { + const struct inet_request_sock *ireq = inet_rsk(req); int err = -1; struct sk_buff * skb; @@ -1310,14 +1314,14 @@ static int tcp_v4_send_synack(struct sock *sk, struct open_request *req, struct tcphdr *th = skb->h.th; th->check = tcp_v4_check(th, skb->len, - req->af.v4_req.loc_addr, - req->af.v4_req.rmt_addr, + ireq->loc_addr, + ireq->rmt_addr, csum_partial((char *)th, skb->len, skb->csum)); - err = ip_build_and_send_pkt(skb, sk, req->af.v4_req.loc_addr, - req->af.v4_req.rmt_addr, - req->af.v4_req.opt); + err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr, + ireq->rmt_addr, + ireq->opt); if (err == NET_XMIT_CN) err = 0; } @@ -1332,8 +1336,8 @@ out: */ static void tcp_v4_or_free(struct open_request *req) { - if (req->af.v4_req.opt) - kfree(req->af.v4_req.opt); + if (inet_rsk(req)->opt) + kfree(inet_rsk(req)->opt); } static inline void syn_flood_warning(struct sk_buff *skb) @@ -1387,6 +1391,7 @@ int sysctl_max_syn_backlog = 256; struct or_calltable or_ipv4 = { .family = PF_INET, + .obj_size = sizeof(struct tcp_request_sock), .rtx_syn_ack = tcp_v4_send_synack, .send_ack = tcp_v4_or_send_ack, .destructor = tcp_v4_or_free, @@ -1395,6 +1400,7 @@ struct or_calltable or_ipv4 = { int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) { + struct inet_request_sock *ireq; struct tcp_options_received tmp_opt; struct open_request *req; __u32 saddr = skb->nh.iph->saddr; @@ -1433,7 +1439,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(); + req = tcp_openreq_alloc(&or_ipv4); if (!req) goto drop; @@ -1461,10 +1467,10 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) tcp_openreq_init(req, &tmp_opt, skb); - req->af.v4_req.loc_addr = daddr; - req->af.v4_req.rmt_addr = saddr; - req->af.v4_req.opt = tcp_v4_save_options(sk, skb); - req->class = &or_ipv4; + ireq = inet_rsk(req); + ireq->loc_addr = daddr; + ireq->rmt_addr = saddr; + ireq->opt = tcp_v4_save_options(sk, skb); if (!want_cookie) TCP_ECN_create_request(req, skb->h.th); @@ -1523,7 +1529,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) isn = tcp_v4_init_sequence(sk, skb); } - req->snt_isn = isn; + tcp_rsk(req)->snt_isn = isn; if (tcp_v4_send_synack(sk, req, dst)) goto drop_and_free; @@ -1551,6 +1557,7 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, struct open_request *req, struct dst_entry *dst) { + struct inet_request_sock *ireq; struct inet_sock *newinet; struct tcp_sock *newtp; struct sock *newsk; @@ -1570,11 +1577,12 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, newtp = tcp_sk(newsk); newinet = inet_sk(newsk); - newinet->daddr = req->af.v4_req.rmt_addr; - newinet->rcv_saddr = req->af.v4_req.loc_addr; - newinet->saddr = req->af.v4_req.loc_addr; - newinet->opt = req->af.v4_req.opt; - req->af.v4_req.opt = NULL; + ireq = inet_rsk(req); + newinet->daddr = ireq->rmt_addr; + newinet->rcv_saddr = ireq->loc_addr; + newinet->saddr = ireq->loc_addr; + newinet->opt = ireq->opt; + ireq->opt = NULL; newinet->mc_index = tcp_v4_iif(skb); newinet->mc_ttl = skb->nh.iph->ttl; newtp->ext_header_len = 0; @@ -2454,15 +2462,16 @@ void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo) static void get_openreq4(struct sock *sk, struct open_request *req, char *tmpbuf, int i, int uid) { + const struct inet_request_sock *ireq = inet_rsk(req); int ttd = req->expires - jiffies; sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X" " %02X %08X:%08X %02X:%08lX %08X %5d %8d %u %d %p", i, - req->af.v4_req.loc_addr, + ireq->loc_addr, ntohs(inet_sk(sk)->sport), - req->af.v4_req.rmt_addr, - ntohs(req->rmt_port), + ireq->rmt_addr, + ntohs(ireq->rmt_port), TCP_SYN_RECV, 0, 0, /* could print option size, but that is af dependent. */ 1, /* timers active (only the expire timer) */ @@ -2618,6 +2627,7 @@ struct proto tcp_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp_sock), + .rsk_prot = &or_ipv4, }; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index eea1a17a9ac2..1037401c7cc8 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -692,6 +692,8 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, struct sock *newsk = sk_alloc(PF_INET, GFP_ATOMIC, sk->sk_prot, 0); if(newsk != NULL) { + struct inet_request_sock *ireq = inet_rsk(req); + struct tcp_request_sock *treq = tcp_rsk(req); struct tcp_sock *newtp; struct sk_filter *filter; @@ -703,7 +705,7 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, tcp_sk(newsk)->bind_hash = NULL; /* Clone the TCP header template */ - inet_sk(newsk)->dport = req->rmt_port; + inet_sk(newsk)->dport = ireq->rmt_port; sock_lock_init(newsk); bh_lock_sock(newsk); @@ -739,14 +741,14 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, /* Now setup tcp_sock */ newtp = tcp_sk(newsk); newtp->pred_flags = 0; - newtp->rcv_nxt = req->rcv_isn + 1; - newtp->snd_nxt = req->snt_isn + 1; - newtp->snd_una = req->snt_isn + 1; - newtp->snd_sml = req->snt_isn + 1; + newtp->rcv_nxt = treq->rcv_isn + 1; + newtp->snd_nxt = treq->snt_isn + 1; + newtp->snd_una = treq->snt_isn + 1; + newtp->snd_sml = treq->snt_isn + 1; tcp_prequeue_init(newtp); - tcp_init_wl(newtp, req->snt_isn, req->rcv_isn); + tcp_init_wl(newtp, treq->snt_isn, treq->rcv_isn); newtp->retransmits = 0; newtp->backoff = 0; @@ -775,10 +777,10 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, tcp_set_ca_state(newtp, TCP_CA_Open); tcp_init_xmit_timers(newsk); skb_queue_head_init(&newtp->out_of_order_queue); - newtp->rcv_wup = req->rcv_isn + 1; - newtp->write_seq = req->snt_isn + 1; + newtp->rcv_wup = treq->rcv_isn + 1; + newtp->write_seq = treq->snt_isn + 1; newtp->pushed_seq = newtp->write_seq; - newtp->copied_seq = req->rcv_isn + 1; + newtp->copied_seq = treq->rcv_isn + 1; newtp->rx_opt.saw_tstamp = 0; @@ -808,18 +810,18 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, newsk->sk_socket = NULL; newsk->sk_sleep = NULL; - newtp->rx_opt.tstamp_ok = req->tstamp_ok; - if((newtp->rx_opt.sack_ok = req->sack_ok) != 0) { + newtp->rx_opt.tstamp_ok = ireq->tstamp_ok; + if((newtp->rx_opt.sack_ok = ireq->sack_ok) != 0) { if (sysctl_tcp_fack) newtp->rx_opt.sack_ok |= 2; } newtp->window_clamp = req->window_clamp; newtp->rcv_ssthresh = req->rcv_wnd; newtp->rcv_wnd = req->rcv_wnd; - newtp->rx_opt.wscale_ok = req->wscale_ok; + newtp->rx_opt.wscale_ok = ireq->wscale_ok; if (newtp->rx_opt.wscale_ok) { - newtp->rx_opt.snd_wscale = req->snd_wscale; - newtp->rx_opt.rcv_wscale = req->rcv_wscale; + newtp->rx_opt.snd_wscale = ireq->snd_wscale; + newtp->rx_opt.rcv_wscale = ireq->rcv_wscale; } else { newtp->rx_opt.snd_wscale = newtp->rx_opt.rcv_wscale = 0; newtp->window_clamp = min(newtp->window_clamp, 65535U); @@ -881,7 +883,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, } /* Check for pure retransmitted SYN. */ - if (TCP_SKB_CB(skb)->seq == req->rcv_isn && + if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn && flg == TCP_FLAG_SYN && !paws_reject) { /* @@ -959,7 +961,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, * Invalid ACK: reset will be sent by listening socket */ if ((flg & TCP_FLAG_ACK) && - (TCP_SKB_CB(skb)->ack_seq != req->snt_isn+1)) + (TCP_SKB_CB(skb)->ack_seq != tcp_rsk(req)->snt_isn + 1)) return sk; /* Also, it would be not so bad idea to check rcv_tsecr, which @@ -970,7 +972,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, /* RFC793: "first check sequence number". */ if (paws_reject || !tcp_in_window(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq, - req->rcv_isn+1, req->rcv_isn+1+req->rcv_wnd)) { + tcp_rsk(req)->rcv_isn + 1, tcp_rsk(req)->rcv_isn + 1 + req->rcv_wnd)) { /* Out of window: send ACK and drop. */ if (!(flg & TCP_FLAG_RST)) req->class->send_ack(skb, req); @@ -981,12 +983,12 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, /* In sequence, PAWS is OK. */ - if (tmp_opt.saw_tstamp && !after(TCP_SKB_CB(skb)->seq, req->rcv_isn+1)) + if (tmp_opt.saw_tstamp && !after(TCP_SKB_CB(skb)->seq, tcp_rsk(req)->rcv_isn + 1)) req->ts_recent = tmp_opt.rcv_tsval; - if (TCP_SKB_CB(skb)->seq == req->rcv_isn) { + if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn) { /* Truncate SYN, it is out of window starting - at req->rcv_isn+1. */ + at tcp_rsk(req)->rcv_isn + 1. */ flg &= ~TCP_FLAG_SYN; } @@ -1003,8 +1005,8 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, return NULL; /* If TCP_DEFER_ACCEPT is set, drop bare ACK. */ - if (tp->defer_accept && TCP_SKB_CB(skb)->end_seq == req->rcv_isn+1) { - req->acked = 1; + if (tp->defer_accept && TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) { + inet_rsk(req)->acked = 1; return NULL; } @@ -1026,7 +1028,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, listen_overflow: if (!sysctl_tcp_abort_on_overflow) { - req->acked = 1; + inet_rsk(req)->acked = 1; return NULL; } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index fa24e7ae1f40..f3c8747caf91 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1358,6 +1358,7 @@ int tcp_send_synack(struct sock *sk) struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, struct open_request *req) { + struct inet_request_sock *ireq = inet_rsk(req); struct tcp_sock *tp = tcp_sk(sk); struct tcphdr *th; int tcp_header_size; @@ -1373,47 +1374,47 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, skb->dst = dst_clone(dst); tcp_header_size = (sizeof(struct tcphdr) + TCPOLEN_MSS + - (req->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0) + - (req->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) + + (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0) + + (ireq->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) + /* SACK_PERM is in the place of NOP NOP of TS */ - ((req->sack_ok && !req->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0)); + ((ireq->sack_ok && !ireq->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0)); skb->h.th = th = (struct tcphdr *) skb_push(skb, tcp_header_size); memset(th, 0, sizeof(struct tcphdr)); th->syn = 1; th->ack = 1; if (dst->dev->features&NETIF_F_TSO) - req->ecn_ok = 0; + ireq->ecn_ok = 0; TCP_ECN_make_synack(req, th); th->source = inet_sk(sk)->sport; - th->dest = req->rmt_port; - TCP_SKB_CB(skb)->seq = req->snt_isn; + th->dest = ireq->rmt_port; + TCP_SKB_CB(skb)->seq = tcp_rsk(req)->snt_isn; TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + 1; TCP_SKB_CB(skb)->sacked = 0; skb_shinfo(skb)->tso_segs = 1; skb_shinfo(skb)->tso_size = 0; th->seq = htonl(TCP_SKB_CB(skb)->seq); - th->ack_seq = htonl(req->rcv_isn + 1); + th->ack_seq = htonl(tcp_rsk(req)->rcv_isn + 1); if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */ __u8 rcv_wscale; /* Set this up on the first call only */ req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW); /* tcp_full_space because it is guaranteed to be the first packet */ tcp_select_initial_window(tcp_full_space(sk), - dst_metric(dst, RTAX_ADVMSS) - (req->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), + dst_metric(dst, RTAX_ADVMSS) - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), &req->rcv_wnd, &req->window_clamp, - req->wscale_ok, + ireq->wscale_ok, &rcv_wscale); - req->rcv_wscale = rcv_wscale; + ireq->rcv_wscale = rcv_wscale; } /* RFC1323: The window in SYN & SYN/ACK segments is never scaled. */ th->window = htons(req->rcv_wnd); TCP_SKB_CB(skb)->when = tcp_time_stamp; - tcp_syn_build_options((__u32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), req->tstamp_ok, - req->sack_ok, req->wscale_ok, req->rcv_wscale, + tcp_syn_build_options((__u32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok, + ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale, TCP_SKB_CB(skb)->when, req->ts_recent); diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 799ebe061e2c..ba30ca0aa6a3 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -513,7 +513,7 @@ static void tcp_synack_timer(struct sock *sk) while ((req = *reqp) != NULL) { if (time_after_eq(now, req->expires)) { if ((req->retrans < thresh || - (req->acked && req->retrans < max_retries)) + (inet_rsk(req)->acked && req->retrans < max_retries)) && !req->class->rtx_syn_ack(sk, req, NULL)) { unsigned long timeo; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 0f69e800a0ad..9199ad2fde0d 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -407,11 +407,13 @@ static struct open_request *tcp_v6_search_req(struct tcp_sock *tp, for (prev = &lopt->syn_table[tcp_v6_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; prev = &req->dl_next) { - if (req->rmt_port == rport && + const struct tcp6_request_sock *treq = tcp6_rsk(req); + + if (inet_rsk(req)->rmt_port == rport && req->class->family == AF_INET6 && - ipv6_addr_equal(&req->af.v6_req.rmt_addr, raddr) && - ipv6_addr_equal(&req->af.v6_req.loc_addr, laddr) && - (!req->af.v6_req.iif || req->af.v6_req.iif == iif)) { + ipv6_addr_equal(&treq->rmt_addr, raddr) && + ipv6_addr_equal(&treq->loc_addr, laddr) && + (!treq->iif || treq->iif == iif)) { BUG_TRAP(req->sk == NULL); *prevp = prev; return req; @@ -923,7 +925,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, */ BUG_TRAP(req->sk == NULL); - if (seq != req->snt_isn) { + if (seq != tcp_rsk(req)->snt_isn) { NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS); goto out; } @@ -960,6 +962,7 @@ out: static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, struct dst_entry *dst) { + struct tcp6_request_sock *treq = tcp6_rsk(req); struct ipv6_pinfo *np = inet6_sk(sk); struct sk_buff * skb; struct ipv6_txoptions *opt = NULL; @@ -969,19 +972,19 @@ static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, memset(&fl, 0, sizeof(fl)); fl.proto = IPPROTO_TCP; - ipv6_addr_copy(&fl.fl6_dst, &req->af.v6_req.rmt_addr); - ipv6_addr_copy(&fl.fl6_src, &req->af.v6_req.loc_addr); + ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); + ipv6_addr_copy(&fl.fl6_src, &treq->loc_addr); fl.fl6_flowlabel = 0; - fl.oif = req->af.v6_req.iif; - fl.fl_ip_dport = req->rmt_port; + fl.oif = treq->iif; + fl.fl_ip_dport = inet_rsk(req)->rmt_port; fl.fl_ip_sport = inet_sk(sk)->sport; if (dst == NULL) { opt = np->opt; if (opt == NULL && np->rxopt.bits.srcrt == 2 && - req->af.v6_req.pktopts) { - struct sk_buff *pktopts = req->af.v6_req.pktopts; + treq->pktopts) { + struct sk_buff *pktopts = treq->pktopts; struct inet6_skb_parm *rxopt = IP6CB(pktopts); if (rxopt->srcrt) opt = ipv6_invert_rthdr(sk, (struct ipv6_rt_hdr*)(pktopts->nh.raw + rxopt->srcrt)); @@ -1008,10 +1011,10 @@ static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, struct tcphdr *th = skb->h.th; th->check = tcp_v6_check(th, skb->len, - &req->af.v6_req.loc_addr, &req->af.v6_req.rmt_addr, + &treq->loc_addr, &treq->rmt_addr, csum_partial((char *)th, skb->len, skb->csum)); - ipv6_addr_copy(&fl.fl6_dst, &req->af.v6_req.rmt_addr); + ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); err = ip6_xmit(sk, skb, &fl, opt, 0); if (err == NET_XMIT_CN) err = 0; @@ -1026,12 +1029,13 @@ done: static void tcp_v6_or_free(struct open_request *req) { - if (req->af.v6_req.pktopts) - kfree_skb(req->af.v6_req.pktopts); + if (tcp6_rsk(req)->pktopts) + kfree_skb(tcp6_rsk(req)->pktopts); } static struct or_calltable or_ipv6 = { .family = AF_INET6, + .obj_size = sizeof(struct tcp6_request_sock), .rtx_syn_ack = tcp_v6_send_synack, .send_ack = tcp_v6_or_send_ack, .destructor = tcp_v6_or_free, @@ -1221,7 +1225,7 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req) { - tcp_v6_send_ack(skb, req->snt_isn+1, req->rcv_isn+1, req->rcv_wnd, req->ts_recent); + tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -1264,7 +1268,7 @@ static void tcp_v6_synq_add(struct sock *sk, struct open_request *req) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; - u32 h = tcp_v6_synq_hash(&req->af.v6_req.rmt_addr, req->rmt_port, lopt->hash_rnd); + u32 h = tcp_v6_synq_hash(&tcp6_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); req->sk = NULL; req->expires = jiffies + TCP_TIMEOUT_INIT; @@ -1284,6 +1288,7 @@ static void tcp_v6_synq_add(struct sock *sk, struct open_request *req) */ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) { + struct tcp6_request_sock *treq; struct ipv6_pinfo *np = inet6_sk(sk); struct tcp_options_received tmp_opt; struct tcp_sock *tp = tcp_sk(sk); @@ -1308,7 +1313,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(); + req = tcp_openreq_alloc(&or_ipv6); if (req == NULL) goto drop; @@ -1321,28 +1326,28 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) tmp_opt.tstamp_ok = tmp_opt.saw_tstamp; tcp_openreq_init(req, &tmp_opt, skb); - req->class = &or_ipv6; - ipv6_addr_copy(&req->af.v6_req.rmt_addr, &skb->nh.ipv6h->saddr); - ipv6_addr_copy(&req->af.v6_req.loc_addr, &skb->nh.ipv6h->daddr); + treq = tcp6_rsk(req); + ipv6_addr_copy(&treq->rmt_addr, &skb->nh.ipv6h->saddr); + ipv6_addr_copy(&treq->loc_addr, &skb->nh.ipv6h->daddr); TCP_ECN_create_request(req, skb->h.th); - req->af.v6_req.pktopts = NULL; + treq->pktopts = NULL; if (ipv6_opt_accepted(sk, skb) || np->rxopt.bits.rxinfo || np->rxopt.bits.rxhlim) { atomic_inc(&skb->users); - req->af.v6_req.pktopts = skb; + treq->pktopts = skb; } - req->af.v6_req.iif = sk->sk_bound_dev_if; + treq->iif = sk->sk_bound_dev_if; /* So that link locals have meaning */ if (!sk->sk_bound_dev_if && - ipv6_addr_type(&req->af.v6_req.rmt_addr) & IPV6_ADDR_LINKLOCAL) - req->af.v6_req.iif = tcp_v6_iif(skb); + ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL) + treq->iif = tcp_v6_iif(skb); if (isn == 0) isn = tcp_v6_init_sequence(sk,skb); - req->snt_isn = isn; + tcp_rsk(req)->snt_isn = isn; if (tcp_v6_send_synack(sk, req, NULL)) goto drop; @@ -1363,6 +1368,7 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, struct open_request *req, struct dst_entry *dst) { + struct tcp6_request_sock *treq = tcp6_rsk(req); struct ipv6_pinfo *newnp, *np = inet6_sk(sk); struct tcp6_sock *newtcp6sk; struct inet_sock *newinet; @@ -1426,10 +1432,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, goto out_overflow; if (np->rxopt.bits.srcrt == 2 && - opt == NULL && req->af.v6_req.pktopts) { - struct inet6_skb_parm *rxopt = IP6CB(req->af.v6_req.pktopts); + opt == NULL && treq->pktopts) { + struct inet6_skb_parm *rxopt = IP6CB(treq->pktopts); if (rxopt->srcrt) - opt = ipv6_invert_rthdr(sk, (struct ipv6_rt_hdr*)(req->af.v6_req.pktopts->nh.raw+rxopt->srcrt)); + opt = ipv6_invert_rthdr(sk, (struct ipv6_rt_hdr *)(treq->pktopts->nh.raw + rxopt->srcrt)); } if (dst == NULL) { @@ -1438,16 +1444,16 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, memset(&fl, 0, sizeof(fl)); fl.proto = IPPROTO_TCP; - ipv6_addr_copy(&fl.fl6_dst, &req->af.v6_req.rmt_addr); + ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); if (opt && opt->srcrt) { struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt; ipv6_addr_copy(&final, &fl.fl6_dst); ipv6_addr_copy(&fl.fl6_dst, rt0->addr); final_p = &final; } - ipv6_addr_copy(&fl.fl6_src, &req->af.v6_req.loc_addr); + ipv6_addr_copy(&fl.fl6_src, &treq->loc_addr); fl.oif = sk->sk_bound_dev_if; - fl.fl_ip_dport = req->rmt_port; + fl.fl_ip_dport = inet_rsk(req)->rmt_port; fl.fl_ip_sport = inet_sk(sk)->sport; if (ip6_dst_lookup(sk, &dst, &fl)) @@ -1482,10 +1488,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, memcpy(newnp, np, sizeof(struct ipv6_pinfo)); - ipv6_addr_copy(&newnp->daddr, &req->af.v6_req.rmt_addr); - ipv6_addr_copy(&newnp->saddr, &req->af.v6_req.loc_addr); - ipv6_addr_copy(&newnp->rcv_saddr, &req->af.v6_req.loc_addr); - newsk->sk_bound_dev_if = req->af.v6_req.iif; + ipv6_addr_copy(&newnp->daddr, &treq->rmt_addr); + ipv6_addr_copy(&newnp->saddr, &treq->loc_addr); + ipv6_addr_copy(&newnp->rcv_saddr, &treq->loc_addr); + newsk->sk_bound_dev_if = treq->iif; /* Now IPv6 options... @@ -1498,11 +1504,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, /* Clone pktoptions received with SYN */ newnp->pktoptions = NULL; - if (req->af.v6_req.pktopts) { - newnp->pktoptions = skb_clone(req->af.v6_req.pktopts, - GFP_ATOMIC); - kfree_skb(req->af.v6_req.pktopts); - req->af.v6_req.pktopts = NULL; + if (treq->pktopts != NULL) { + newnp->pktoptions = skb_clone(treq->pktopts, GFP_ATOMIC); + kfree_skb(treq->pktopts); + treq->pktopts = NULL; if (newnp->pktoptions) skb_set_owner_r(newnp->pktoptions, newsk); } @@ -2058,8 +2063,8 @@ static void get_openreq6(struct seq_file *seq, if (ttd < 0) ttd = 0; - src = &req->af.v6_req.loc_addr; - dest = &req->af.v6_req.rmt_addr; + src = &tcp6_rsk(req)->loc_addr; + dest = &tcp6_rsk(req)->rmt_addr; seq_printf(seq, "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X " "%02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %p\n", @@ -2069,7 +2074,7 @@ static void get_openreq6(struct seq_file *seq, ntohs(inet_sk(sk)->sport), dest->s6_addr32[0], dest->s6_addr32[1], dest->s6_addr32[2], dest->s6_addr32[3], - ntohs(req->rmt_port), + ntohs(inet_rsk(req)->rmt_port), TCP_SYN_RECV, 0,0, /* could print option size, but that is af dependent. */ 1, /* timers active (only the expire timer) */ @@ -2239,6 +2244,7 @@ struct proto tcpv6_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp6_sock), + .rsk_prot = &or_ipv6, }; static struct inet6_protocol tcpv6_protocol = { -- cgit v1.2.3 From 60236fdd08b2169045a3bbfc5ffe1576e6c3c17b Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 18 Jun 2005 22:47:21 -0700 Subject: [NET] Rename open_request to request_sock Ok, this one just renames some stuff to have a better namespace and to dissassociate it from TCP: struct open_request -> struct request_sock tcp_openreq_alloc -> reqsk_alloc tcp_openreq_free -> reqsk_free tcp_openreq_fastfree -> __reqsk_free With this most of the infrastructure closely resembles a struct sock methods subset. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/ip.h | 4 ++-- include/linux/ipv6.h | 2 +- include/linux/tcp.h | 6 ++--- include/net/request_sock.h | 34 ++++++++++++++-------------- include/net/sock.h | 4 ++-- include/net/tcp.h | 30 ++++++++++++------------- include/net/tcp_ecn.h | 6 ++--- net/ipv4/syncookies.c | 12 +++++----- net/ipv4/tcp.c | 12 +++++----- net/ipv4/tcp_diag.c | 4 ++-- net/ipv4/tcp_ipv4.c | 56 +++++++++++++++++++++++----------------------- net/ipv4/tcp_minisocks.c | 14 ++++++------ net/ipv4/tcp_output.c | 2 +- net/ipv4/tcp_timer.c | 6 ++--- net/ipv6/tcp_ipv6.c | 42 +++++++++++++++++----------------- 15 files changed, 117 insertions(+), 117 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ip.h b/include/linux/ip.h index d5b7c907204e..31e7cedd9f84 100644 --- a/include/linux/ip.h +++ b/include/linux/ip.h @@ -109,7 +109,7 @@ struct ip_options { #define optlength(opt) (sizeof(struct ip_options) + opt->optlen) struct inet_request_sock { - struct open_request req; + struct request_sock req; u32 loc_addr; u32 rmt_addr; u16 rmt_port; @@ -123,7 +123,7 @@ struct inet_request_sock { struct ip_options *opt; }; -static inline struct inet_request_sock *inet_rsk(const struct open_request *sk) +static inline struct inet_request_sock *inet_rsk(const struct request_sock *sk) { return (struct inet_request_sock *)sk; } diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 98acdbf3d446..6fcd6a0ade24 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -201,7 +201,7 @@ struct tcp6_request_sock { int iif; }; -static inline struct tcp6_request_sock *tcp6_rsk(const struct open_request *sk) +static inline struct tcp6_request_sock *tcp6_rsk(const struct request_sock *sk) { return (struct tcp6_request_sock *)sk; } diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 86771b37b80d..fb54292a15aa 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -236,7 +236,7 @@ struct tcp_request_sock { __u32 snt_isn; }; -static inline struct tcp_request_sock *tcp_rsk(const struct open_request *req) +static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req) { return (struct tcp_request_sock *)req; } @@ -393,8 +393,8 @@ struct tcp_sock { struct tcp_listen_opt *listen_opt; /* FIFO of established children */ - struct open_request *accept_queue; - struct open_request *accept_queue_tail; + struct request_sock *accept_queue; + struct request_sock *accept_queue_tail; unsigned int keepalive_time; /* time before keep alive takes place */ unsigned int keepalive_intvl; /* time interval between keep alive probes */ diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 9502f5587931..08a8fd1d1610 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -19,28 +19,28 @@ #include #include -struct open_request; +struct request_sock; struct sk_buff; struct dst_entry; struct proto; -struct or_calltable { +struct request_sock_ops { int family; kmem_cache_t *slab; int obj_size; int (*rtx_syn_ack)(struct sock *sk, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst); void (*send_ack)(struct sk_buff *skb, - struct open_request *req); + struct request_sock *req); void (*send_reset)(struct sk_buff *skb); - void (*destructor)(struct open_request *req); + void (*destructor)(struct request_sock *req); }; -/* struct open_request - mini sock to represent a connection request +/* struct request_sock - mini sock to represent a connection request */ -struct open_request { - struct open_request *dl_next; /* Must be first member! */ +struct request_sock { + struct request_sock *dl_next; /* Must be first member! */ u16 mss; u8 retrans; u8 __pad; @@ -49,29 +49,29 @@ struct open_request { u32 rcv_wnd; /* rcv_wnd offered first time */ u32 ts_recent; unsigned long expires; - struct or_calltable *class; + struct request_sock_ops *rsk_ops; struct sock *sk; }; -static inline struct open_request *tcp_openreq_alloc(struct or_calltable *class) +static inline struct request_sock *reqsk_alloc(struct request_sock_ops *ops) { - struct open_request *req = kmem_cache_alloc(class->slab, SLAB_ATOMIC); + struct request_sock *req = kmem_cache_alloc(ops->slab, SLAB_ATOMIC); if (req != NULL) - req->class = class; + req->rsk_ops = ops; return req; } -static inline void tcp_openreq_fastfree(struct open_request *req) +static inline void __reqsk_free(struct request_sock *req) { - kmem_cache_free(req->class->slab, req); + kmem_cache_free(req->rsk_ops->slab, req); } -static inline void tcp_openreq_free(struct open_request *req) +static inline void reqsk_free(struct request_sock *req) { - req->class->destructor(req); - tcp_openreq_fastfree(req); + req->rsk_ops->destructor(req); + __reqsk_free(req); } #endif /* _REQUEST_SOCK_H */ diff --git a/include/net/sock.h b/include/net/sock.h index 6919276af8af..e593af5b1ecc 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -484,7 +484,7 @@ extern void sk_stream_kill_queues(struct sock *sk); extern int sk_wait_data(struct sock *sk, long *timeo); -struct or_calltable; +struct request_sock_ops; /* Networking protocol blocks we attach to sockets. * socket layer -> transport layer interface @@ -549,7 +549,7 @@ struct proto { kmem_cache_t *slab; unsigned int obj_size; - struct or_calltable *rsk_prot; + struct request_sock_ops *rsk_prot; struct module *owner; diff --git a/include/net/tcp.h b/include/net/tcp.h index d438ba566b89..6663086a5e35 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -641,7 +641,7 @@ struct tcp_func { struct sock * (*syn_recv_sock) (struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst); int (*remember_stamp) (struct sock *sk); @@ -785,8 +785,8 @@ extern enum tcp_tw_status tcp_timewait_state_process(struct tcp_tw_bucket *tw, unsigned len); extern struct sock * tcp_check_req(struct sock *sk,struct sk_buff *skb, - struct open_request *req, - struct open_request **prev); + struct request_sock *req, + struct request_sock **prev); extern int tcp_child_process(struct sock *parent, struct sock *child, struct sk_buff *skb); @@ -836,12 +836,12 @@ extern int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb); extern struct sock * tcp_create_openreq_child(struct sock *sk, - struct open_request *req, + struct request_sock *req, struct sk_buff *skb); extern struct sock * tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst); extern int tcp_v4_do_rcv(struct sock *sk, @@ -855,7 +855,7 @@ extern int tcp_connect(struct sock *sk); extern struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, - struct open_request *req); + struct request_sock *req); extern int tcp_disconnect(struct sock *sk, int flags); @@ -1683,7 +1683,7 @@ static inline int tcp_full_space(const struct sock *sk) return tcp_win_from_space(sk->sk_rcvbuf); } -static inline void tcp_acceptq_queue(struct sock *sk, struct open_request *req, +static inline void tcp_acceptq_queue(struct sock *sk, struct request_sock *req, struct sock *child) { struct tcp_sock *tp = tcp_sk(sk); @@ -1707,11 +1707,11 @@ struct tcp_listen_opt int qlen_young; int clock_hand; u32 hash_rnd; - struct open_request *syn_table[TCP_SYNQ_HSIZE]; + struct request_sock *syn_table[TCP_SYNQ_HSIZE]; }; static inline void -tcp_synq_removed(struct sock *sk, struct open_request *req) +tcp_synq_removed(struct sock *sk, struct request_sock *req) { struct tcp_listen_opt *lopt = tcp_sk(sk)->listen_opt; @@ -1745,23 +1745,23 @@ static inline int tcp_synq_is_full(struct sock *sk) return tcp_synq_len(sk) >> tcp_sk(sk)->listen_opt->max_qlen_log; } -static inline void tcp_synq_unlink(struct tcp_sock *tp, struct open_request *req, - struct open_request **prev) +static inline void tcp_synq_unlink(struct tcp_sock *tp, struct request_sock *req, + struct request_sock **prev) { write_lock(&tp->syn_wait_lock); *prev = req->dl_next; write_unlock(&tp->syn_wait_lock); } -static inline void tcp_synq_drop(struct sock *sk, struct open_request *req, - struct open_request **prev) +static inline void tcp_synq_drop(struct sock *sk, struct request_sock *req, + struct request_sock **prev) { tcp_synq_unlink(tcp_sk(sk), req, prev); tcp_synq_removed(sk, req); - tcp_openreq_free(req); + reqsk_free(req); } -static __inline__ void tcp_openreq_init(struct open_request *req, +static __inline__ void tcp_openreq_init(struct request_sock *req, struct tcp_options_received *rx_opt, struct sk_buff *skb) { diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h index 94ad970e844a..64980ee8c92a 100644 --- a/include/net/tcp_ecn.h +++ b/include/net/tcp_ecn.h @@ -39,7 +39,7 @@ static inline void TCP_ECN_send_syn(struct sock *sk, struct tcp_sock *tp, } static __inline__ void -TCP_ECN_make_synack(struct open_request *req, struct tcphdr *th) +TCP_ECN_make_synack(struct request_sock *req, struct tcphdr *th) { if (inet_rsk(req)->ecn_ok) th->ece = 1; @@ -112,13 +112,13 @@ static inline int TCP_ECN_rcv_ecn_echo(struct tcp_sock *tp, struct tcphdr *th) } static inline void TCP_ECN_openreq_child(struct tcp_sock *tp, - struct open_request *req) + struct request_sock *req) { tp->ecn_flags = inet_rsk(req)->ecn_ok ? TCP_ECN_OK : 0; } static __inline__ void -TCP_ECN_create_request(struct open_request *req, struct tcphdr *th) +TCP_ECN_create_request(struct request_sock *req, struct tcphdr *th) { if (sysctl_tcp_ecn && th->ece && th->cwr) inet_rsk(req)->ecn_ok = 1; diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index dd47e6da6fb3..72d014442185 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -169,10 +169,10 @@ static inline int cookie_check(struct sk_buff *skb, __u32 cookie) return mssind < NUM_MSS ? msstab[mssind] + 1 : 0; } -extern struct or_calltable or_ipv4; +extern struct request_sock_ops tcp_request_sock_ops; static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst) { struct tcp_sock *tp = tcp_sk(sk); @@ -182,7 +182,7 @@ static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, if (child) tcp_acceptq_queue(sk, req, child); else - tcp_openreq_free(req); + reqsk_free(req); return child; } @@ -195,7 +195,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct tcp_sock *tp = tcp_sk(sk); __u32 cookie = ntohl(skb->h.th->ack_seq) - 1; struct sock *ret = sk; - struct open_request *req; + struct request_sock *req; int mss; struct rtable *rt; __u8 rcv_wscale; @@ -212,7 +212,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV); ret = NULL; - req = tcp_openreq_alloc(&or_ipv4); /* for safety */ + req = reqsk_alloc(&tcp_request_sock_ops); /* for safety */ if (!req) goto out; @@ -262,7 +262,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, { .sport = skb->h.th->dest, .dport = skb->h.th->source } } }; if (ip_route_output_key(&rt, &fl)) { - tcp_openreq_free(req); + reqsk_free(req); goto out; } } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index a3cabfa2022a..1c29feb6b35f 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -516,8 +516,8 @@ static void tcp_listen_stop (struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; - struct open_request *acc_req = tp->accept_queue; - struct open_request *req; + struct request_sock *acc_req = tp->accept_queue; + struct request_sock *req; int i; tcp_delete_keepalive_timer(sk); @@ -533,7 +533,7 @@ static void tcp_listen_stop (struct sock *sk) while ((req = lopt->syn_table[i]) != NULL) { lopt->syn_table[i] = req->dl_next; lopt->qlen--; - tcp_openreq_free(req); + reqsk_free(req); /* Following specs, it would be better either to send FIN * (and enter FIN-WAIT-1, it is normal close) @@ -573,7 +573,7 @@ static void tcp_listen_stop (struct sock *sk) sock_put(child); sk_acceptq_removed(sk); - tcp_openreq_fastfree(req); + __reqsk_free(req); } BUG_TRAP(!sk->sk_ack_backlog); } @@ -1894,7 +1894,7 @@ static int wait_for_connect(struct sock *sk, long timeo) struct sock *tcp_accept(struct sock *sk, int flags, int *err) { struct tcp_sock *tp = tcp_sk(sk); - struct open_request *req; + struct request_sock *req; struct sock *newsk; int error; @@ -1927,7 +1927,7 @@ struct sock *tcp_accept(struct sock *sk, int flags, int *err) newsk = req->sk; sk_acceptq_removed(sk); - tcp_openreq_fastfree(req); + __reqsk_free(req); BUG_TRAP(newsk->sk_state != TCP_SYN_RECV); release_sock(sk); return newsk; diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 700ff2413588..67277800d0c1 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -455,7 +455,7 @@ static int tcpdiag_dump_sock(struct sk_buff *skb, struct sock *sk, } static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, - struct open_request *req, + struct request_sock *req, u32 pid, u32 seq) { const struct inet_request_sock *ireq = inet_rsk(req); @@ -542,7 +542,7 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, } for (j = s_j; j < TCP_SYNQ_HSIZE; j++) { - struct open_request *req, *head = lopt->syn_table[j]; + struct request_sock *req, *head = lopt->syn_table[j]; reqnum = 0; for (req = head; req; reqnum++, req = req->dl_next) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index e156be90df14..95528a75a63d 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -36,7 +36,7 @@ * ACK bit. * Andi Kleen : Implemented fast path mtu discovery. * Fixed many serious bugs in the - * open_request handling and moved + * request_sock handling and moved * most of it into the af independent code. * Added tail drop and some other bugfixes. * Added new listen sematics. @@ -869,13 +869,13 @@ static __inline__ u32 tcp_v4_synq_hash(u32 raddr, u16 rport, u32 rnd) return (jhash_2words(raddr, (u32) rport, rnd) & (TCP_SYNQ_HSIZE - 1)); } -static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, - struct open_request ***prevp, +static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp, + struct request_sock ***prevp, __u16 rport, __u32 raddr, __u32 laddr) { struct tcp_listen_opt *lopt = tp->listen_opt; - struct open_request *req, **prev; + struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v4_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; @@ -885,7 +885,7 @@ static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, if (ireq->rmt_port == rport && ireq->rmt_addr == raddr && ireq->loc_addr == laddr && - TCP_INET_FAMILY(req->class->family)) { + TCP_INET_FAMILY(req->rsk_ops->family)) { BUG_TRAP(!req->sk); *prevp = prev; break; @@ -895,7 +895,7 @@ static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, return req; } -static void tcp_v4_synq_add(struct sock *sk, struct open_request *req) +static void tcp_v4_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; @@ -1052,7 +1052,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info) } switch (sk->sk_state) { - struct open_request *req, **prev; + struct request_sock *req, **prev; case TCP_LISTEN: if (sock_owned_by_user(sk)) goto out; @@ -1256,14 +1256,14 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) tcp_tw_put(tw); } -static void tcp_v4_or_send_ack(struct sk_buff *skb, struct open_request *req) +static void tcp_v4_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } static struct dst_entry* tcp_v4_route_req(struct sock *sk, - struct open_request *req) + struct request_sock *req) { struct rtable *rt; const struct inet_request_sock *ireq = inet_rsk(req); @@ -1294,10 +1294,10 @@ static struct dst_entry* tcp_v4_route_req(struct sock *sk, /* * Send a SYN-ACK after having received an ACK. - * This still operates on a open_request only, not on a big + * This still operates on a request_sock only, not on a big * socket. */ -static int tcp_v4_send_synack(struct sock *sk, struct open_request *req, +static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req, struct dst_entry *dst) { const struct inet_request_sock *ireq = inet_rsk(req); @@ -1332,9 +1332,9 @@ out: } /* - * IPv4 open_request destructor. + * IPv4 request_sock destructor. */ -static void tcp_v4_or_free(struct open_request *req) +static void tcp_v4_reqsk_destructor(struct request_sock *req) { if (inet_rsk(req)->opt) kfree(inet_rsk(req)->opt); @@ -1353,7 +1353,7 @@ static inline void syn_flood_warning(struct sk_buff *skb) } /* - * Save and compile IPv4 options into the open_request if needed. + * Save and compile IPv4 options into the request_sock if needed. */ static inline struct ip_options *tcp_v4_save_options(struct sock *sk, struct sk_buff *skb) @@ -1389,12 +1389,12 @@ static inline struct ip_options *tcp_v4_save_options(struct sock *sk, */ int sysctl_max_syn_backlog = 256; -struct or_calltable or_ipv4 = { +struct request_sock_ops tcp_request_sock_ops = { .family = PF_INET, .obj_size = sizeof(struct tcp_request_sock), .rtx_syn_ack = tcp_v4_send_synack, - .send_ack = tcp_v4_or_send_ack, - .destructor = tcp_v4_or_free, + .send_ack = tcp_v4_reqsk_send_ack, + .destructor = tcp_v4_reqsk_destructor, .send_reset = tcp_v4_send_reset, }; @@ -1402,7 +1402,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) { struct inet_request_sock *ireq; struct tcp_options_received tmp_opt; - struct open_request *req; + struct request_sock *req; __u32 saddr = skb->nh.iph->saddr; __u32 daddr = skb->nh.iph->daddr; __u32 isn = TCP_SKB_CB(skb)->when; @@ -1439,7 +1439,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(&or_ipv4); + req = reqsk_alloc(&tcp_request_sock_ops); if (!req) goto drop; @@ -1535,14 +1535,14 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) goto drop_and_free; if (want_cookie) { - tcp_openreq_free(req); + reqsk_free(req); } else { tcp_v4_synq_add(sk, req); } return 0; drop_and_free: - tcp_openreq_free(req); + reqsk_free(req); drop: TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS); return 0; @@ -1554,7 +1554,7 @@ drop: * now create the new socket. */ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst) { struct inet_request_sock *ireq; @@ -1613,9 +1613,9 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) struct iphdr *iph = skb->nh.iph; struct tcp_sock *tp = tcp_sk(sk); struct sock *nsk; - struct open_request **prev; + struct request_sock **prev; /* Find possible connection requests. */ - struct open_request *req = tcp_v4_search_req(tp, &prev, th->source, + struct request_sock *req = tcp_v4_search_req(tp, &prev, th->source, iph->saddr, iph->daddr); if (req) return tcp_check_req(sk, skb, req, prev); @@ -2152,13 +2152,13 @@ static void *listening_get_next(struct seq_file *seq, void *cur) ++st->num; if (st->state == TCP_SEQ_STATE_OPENREQ) { - struct open_request *req = cur; + struct request_sock *req = cur; tp = tcp_sk(st->syn_wait_sk); req = req->dl_next; while (1) { while (req) { - if (req->class->family == st->family) { + if (req->rsk_ops->family == st->family) { cur = req; goto out; } @@ -2459,7 +2459,7 @@ void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo) memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops)); } -static void get_openreq4(struct sock *sk, struct open_request *req, +static void get_openreq4(struct sock *sk, struct request_sock *req, char *tmpbuf, int i, int uid) { const struct inet_request_sock *ireq = inet_rsk(req); @@ -2627,7 +2627,7 @@ struct proto tcp_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp_sock), - .rsk_prot = &or_ipv4, + .rsk_prot = &tcp_request_sock_ops, }; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 1037401c7cc8..0e6d525a8341 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -684,7 +684,7 @@ out: * Actually, we could lots of memory writes here. tp of listening * socket contains all necessary default parameters. */ -struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, struct sk_buff *skb) +struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, struct sk_buff *skb) { /* allocate the newsk from the same slab of the master sock, * if not, at sk_free time we'll try to free it from the wrong @@ -853,12 +853,12 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, /* * Process an incoming packet for SYN_RECV sockets represented - * as an open_request. + * as a request_sock. */ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, - struct open_request *req, - struct open_request **prev) + struct request_sock *req, + struct request_sock **prev) { struct tcphdr *th = skb->h.th; struct tcp_sock *tp = tcp_sk(sk); @@ -903,7 +903,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, * Enforce "SYN-ACK" according to figure 8, figure 6 * of RFC793, fixed by RFC1122. */ - req->class->rtx_syn_ack(sk, req, NULL); + req->rsk_ops->rtx_syn_ack(sk, req, NULL); return NULL; } @@ -975,7 +975,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, tcp_rsk(req)->rcv_isn + 1, tcp_rsk(req)->rcv_isn + 1 + req->rcv_wnd)) { /* Out of window: send ACK and drop. */ if (!(flg & TCP_FLAG_RST)) - req->class->send_ack(skb, req); + req->rsk_ops->send_ack(skb, req); if (paws_reject) NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED); return NULL; @@ -1035,7 +1035,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, embryonic_reset: NET_INC_STATS_BH(LINUX_MIB_EMBRYONICRSTS); if (!(flg & TCP_FLAG_RST)) - req->class->send_reset(skb); + req->rsk_ops->send_reset(skb); tcp_synq_drop(sk, req, prev); return NULL; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f3c8747caf91..f17c6577e337 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1356,7 +1356,7 @@ int tcp_send_synack(struct sock *sk) * Prepare a SYN-ACK. */ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, - struct open_request *req) + struct request_sock *req) { struct inet_request_sock *ireq = inet_rsk(req); struct tcp_sock *tp = tcp_sk(sk); diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index ba30ca0aa6a3..f03efe5fb76a 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -468,7 +468,7 @@ static void tcp_synack_timer(struct sock *sk) int max_retries = tp->syn_retries ? : sysctl_tcp_synack_retries; int thresh = max_retries; unsigned long now = jiffies; - struct open_request **reqp, *req; + struct request_sock **reqp, *req; int i, budget; if (lopt == NULL || lopt->qlen == 0) @@ -514,7 +514,7 @@ static void tcp_synack_timer(struct sock *sk) if (time_after_eq(now, req->expires)) { if ((req->retrans < thresh || (inet_rsk(req)->acked && req->retrans < max_retries)) - && !req->class->rtx_syn_ack(sk, req, NULL)) { + && !req->rsk_ops->rtx_syn_ack(sk, req, NULL)) { unsigned long timeo; if (req->retrans++ == 0) @@ -533,7 +533,7 @@ static void tcp_synack_timer(struct sock *sk) lopt->qlen--; if (req->retrans == 0) lopt->qlen_young--; - tcp_openreq_free(req); + reqsk_free(req); continue; } reqp = &req->dl_next; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 9199ad2fde0d..068cd4a8c292 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -65,7 +65,7 @@ #include static void tcp_v6_send_reset(struct sk_buff *skb); -static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req); +static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req); static void tcp_v6_send_check(struct sock *sk, struct tcphdr *th, int len, struct sk_buff *skb); @@ -394,15 +394,15 @@ static u32 tcp_v6_synq_hash(struct in6_addr *raddr, u16 rport, u32 rnd) return c & (TCP_SYNQ_HSIZE - 1); } -static struct open_request *tcp_v6_search_req(struct tcp_sock *tp, - struct open_request ***prevp, +static struct request_sock *tcp_v6_search_req(struct tcp_sock *tp, + struct request_sock ***prevp, __u16 rport, struct in6_addr *raddr, struct in6_addr *laddr, int iif) { struct tcp_listen_opt *lopt = tp->listen_opt; - struct open_request *req, **prev; + struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v6_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; @@ -410,7 +410,7 @@ static struct open_request *tcp_v6_search_req(struct tcp_sock *tp, const struct tcp6_request_sock *treq = tcp6_rsk(req); if (inet_rsk(req)->rmt_port == rport && - req->class->family == AF_INET6 && + req->rsk_ops->family == AF_INET6 && ipv6_addr_equal(&treq->rmt_addr, raddr) && ipv6_addr_equal(&treq->loc_addr, laddr) && (!treq->iif || treq->iif == iif)) { @@ -908,9 +908,9 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, icmpv6_err_convert(type, code, &err); - /* Might be for an open_request */ + /* Might be for an request_sock */ switch (sk->sk_state) { - struct open_request *req, **prev; + struct request_sock *req, **prev; case TCP_LISTEN: if (sock_owned_by_user(sk)) goto out; @@ -959,7 +959,7 @@ out: } -static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, +static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req, struct dst_entry *dst) { struct tcp6_request_sock *treq = tcp6_rsk(req); @@ -1027,18 +1027,18 @@ done: return err; } -static void tcp_v6_or_free(struct open_request *req) +static void tcp_v6_reqsk_destructor(struct request_sock *req) { if (tcp6_rsk(req)->pktopts) kfree_skb(tcp6_rsk(req)->pktopts); } -static struct or_calltable or_ipv6 = { +static struct request_sock_ops tcp6_request_sock_ops = { .family = AF_INET6, .obj_size = sizeof(struct tcp6_request_sock), .rtx_syn_ack = tcp_v6_send_synack, - .send_ack = tcp_v6_or_send_ack, - .destructor = tcp_v6_or_free, + .send_ack = tcp_v6_reqsk_send_ack, + .destructor = tcp_v6_reqsk_destructor, .send_reset = tcp_v6_send_reset }; @@ -1223,7 +1223,7 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) tcp_tw_put(tw); } -static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req) +static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -1231,7 +1231,7 @@ static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req) static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) { - struct open_request *req, **prev; + struct request_sock *req, **prev; struct tcphdr *th = skb->h.th; struct tcp_sock *tp = tcp_sk(sk); struct sock *nsk; @@ -1264,7 +1264,7 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) return sk; } -static void tcp_v6_synq_add(struct sock *sk, struct open_request *req) +static void tcp_v6_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; @@ -1292,7 +1292,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) struct ipv6_pinfo *np = inet6_sk(sk); struct tcp_options_received tmp_opt; struct tcp_sock *tp = tcp_sk(sk); - struct open_request *req = NULL; + struct request_sock *req = NULL; __u32 isn = TCP_SKB_CB(skb)->when; if (skb->protocol == htons(ETH_P_IP)) @@ -1313,7 +1313,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(&or_ipv6); + req = reqsk_alloc(&tcp6_request_sock_ops); if (req == NULL) goto drop; @@ -1358,14 +1358,14 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) drop: if (req) - tcp_openreq_free(req); + reqsk_free(req); TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS); return 0; /* don't send reset */ } static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst) { struct tcp6_request_sock *treq = tcp6_rsk(req); @@ -2055,7 +2055,7 @@ static int tcp_v6_destroy_sock(struct sock *sk) /* Proc filesystem TCPv6 sock list dumping. */ static void get_openreq6(struct seq_file *seq, - struct sock *sk, struct open_request *req, int i, int uid) + struct sock *sk, struct request_sock *req, int i, int uid) { struct in6_addr *dest, *src; int ttd = req->expires - jiffies; @@ -2244,7 +2244,7 @@ struct proto tcpv6_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp6_sock), - .rsk_prot = &or_ipv6, + .rsk_prot = &tcp6_request_sock_ops, }; static struct inet6_protocol tcpv6_protocol = { -- cgit v1.2.3 From 0e87506fcc734647c7b2497eee4eb81e785c857a Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 18 Jun 2005 22:47:59 -0700 Subject: [NET] Generalise tcp_listen_opt This chunks out the accept_queue and tcp_listen_opt code and moves them to net/core/request_sock.c and include/net/request_sock.h, to make it useful for other transport protocols, DCCP being the first one to use it. Next patches will rename tcp_listen_opt to accept_sock and remove the inline tcp functions that just call a reqsk_queue_ function. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/tcp.h | 17 +---- include/net/request_sock.h | 178 +++++++++++++++++++++++++++++++++++++++++++++ include/net/tcp.h | 46 ++---------- net/core/Makefile | 3 +- net/core/request_sock.c | 48 ++++++++++++ net/ipv4/tcp.c | 67 ++++++----------- net/ipv4/tcp_diag.c | 6 +- net/ipv4/tcp_ipv4.c | 32 +++----- net/ipv4/tcp_minisocks.c | 6 +- net/ipv4/tcp_timer.c | 10 +-- net/ipv6/tcp_ipv6.c | 14 +--- 11 files changed, 281 insertions(+), 146 deletions(-) create mode 100644 net/core/request_sock.c (limited to 'include/linux') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index fb54292a15aa..97a7c9e03df5 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -379,22 +379,7 @@ struct tcp_sock { __u32 total_retrans; /* Total retransmits for entire connection */ - /* The syn_wait_lock is necessary only to avoid proc interface having - * to grab the main lock sock while browsing the listening hash - * (otherwise it's deadlock prone). - * This lock is acquired in read mode only from listening_get_next() - * and it's acquired in write mode _only_ from code that is actively - * changing the syn_wait_queue. All readers that are holding - * the master sock lock don't need to grab this lock in read mode - * too as the syn_wait_queue writes are always protected from - * the main sock lock. - */ - rwlock_t syn_wait_lock; - struct tcp_listen_opt *listen_opt; - - /* FIFO of established children */ - struct request_sock *accept_queue; - struct request_sock *accept_queue_tail; + struct request_sock_queue accept_queue; /* FIFO of established children */ unsigned int keepalive_time; /* time before keep alive takes place */ unsigned int keepalive_intvl; /* time interval between keep alive probes */ diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 08a8fd1d1610..38943ed04e73 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -16,7 +16,9 @@ #define _REQUEST_SOCK_H #include +#include #include + #include struct request_sock; @@ -74,4 +76,180 @@ static inline void reqsk_free(struct request_sock *req) __reqsk_free(req); } +extern int sysctl_max_syn_backlog; + +/** struct tcp_listen_opt - listen state + * + * @max_qlen_log - log_2 of maximal queued SYNs/REQUESTs + */ +struct tcp_listen_opt { + u8 max_qlen_log; + /* 3 bytes hole, try to use */ + int qlen; + int qlen_young; + int clock_hand; + u32 hash_rnd; + struct request_sock *syn_table[0]; +}; + +/** struct request_sock_queue - queue of request_socks + * + * @rskq_accept_head - FIFO head of established children + * @rskq_accept_tail - FIFO tail of established children + * @syn_wait_lock - serializer + * + * %syn_wait_lock is necessary only to avoid proc interface having to grab the main + * lock sock while browsing the listening hash (otherwise it's deadlock prone). + * + * This lock is acquired in read mode only from listening_get_next() seq_file + * op and it's acquired in write mode _only_ from code that is actively + * changing rskq_accept_head. All readers that are holding the master sock lock + * don't need to grab this lock in read mode too as rskq_accept_head. writes + * are always protected from the main sock lock. + */ +struct request_sock_queue { + struct request_sock *rskq_accept_head; + struct request_sock *rskq_accept_tail; + rwlock_t syn_wait_lock; + struct tcp_listen_opt *listen_opt; +}; + +extern int reqsk_queue_alloc(struct request_sock_queue *queue, + const int nr_table_entries); + +static inline struct tcp_listen_opt *reqsk_queue_yank_listen_sk(struct request_sock_queue *queue) +{ + struct tcp_listen_opt *lopt; + + write_lock_bh(&queue->syn_wait_lock); + lopt = queue->listen_opt; + queue->listen_opt = NULL; + write_unlock_bh(&queue->syn_wait_lock); + + return lopt; +} + +static inline void reqsk_queue_destroy(struct request_sock_queue *queue) +{ + kfree(reqsk_queue_yank_listen_sk(queue)); +} + +static inline struct request_sock * + reqsk_queue_yank_acceptq(struct request_sock_queue *queue) +{ + struct request_sock *req = queue->rskq_accept_head; + + queue->rskq_accept_head = queue->rskq_accept_head = NULL; + return req; +} + +static inline int reqsk_queue_empty(struct request_sock_queue *queue) +{ + return queue->rskq_accept_head == NULL; +} + +static inline void reqsk_queue_unlink(struct request_sock_queue *queue, + struct request_sock *req, + struct request_sock **prev_req) +{ + write_lock(&queue->syn_wait_lock); + *prev_req = req->dl_next; + write_unlock(&queue->syn_wait_lock); +} + +static inline void reqsk_queue_add(struct request_sock_queue *queue, + struct request_sock *req, + struct sock *parent, + struct sock *child) +{ + req->sk = child; + sk_acceptq_added(parent); + + if (queue->rskq_accept_head == NULL) + queue->rskq_accept_head = req; + else + queue->rskq_accept_tail->dl_next = req; + + queue->rskq_accept_tail = req; + req->dl_next = NULL; +} + +static inline struct request_sock *reqsk_queue_remove(struct request_sock_queue *queue) +{ + struct request_sock *req = queue->rskq_accept_head; + + BUG_TRAP(req != NULL); + + queue->rskq_accept_head = req->dl_next; + if (queue->rskq_accept_head == NULL) + queue->rskq_accept_tail = NULL; + + return req; +} + +static inline struct sock *reqsk_queue_get_child(struct request_sock_queue *queue, + struct sock *parent) +{ + struct request_sock *req = reqsk_queue_remove(queue); + struct sock *child = req->sk; + + BUG_TRAP(child != NULL); + + sk_acceptq_removed(parent); + __reqsk_free(req); + return child; +} + +static inline int reqsk_queue_removed(struct request_sock_queue *queue, + struct request_sock *req) +{ + struct tcp_listen_opt *lopt = queue->listen_opt; + + if (req->retrans == 0) + --lopt->qlen_young; + + return --lopt->qlen; +} + +static inline int reqsk_queue_added(struct request_sock_queue *queue) +{ + struct tcp_listen_opt *lopt = queue->listen_opt; + const int prev_qlen = lopt->qlen; + + lopt->qlen_young++; + lopt->qlen++; + return prev_qlen; +} + +static inline int reqsk_queue_len(struct request_sock_queue *queue) +{ + return queue->listen_opt != NULL ? queue->listen_opt->qlen : 0; +} + +static inline int reqsk_queue_len_young(struct request_sock_queue *queue) +{ + return queue->listen_opt->qlen_young; +} + +static inline int reqsk_queue_is_full(struct request_sock_queue *queue) +{ + return queue->listen_opt->qlen >> queue->listen_opt->max_qlen_log; +} + +static inline void reqsk_queue_hash_req(struct request_sock_queue *queue, + u32 hash, struct request_sock *req, + unsigned timeout) +{ + struct tcp_listen_opt *lopt = queue->listen_opt; + + req->expires = jiffies + timeout; + req->retrans = 0; + req->sk = NULL; + req->dl_next = lopt->syn_table[hash]; + + write_lock(&queue->syn_wait_lock); + lopt->syn_table[hash] = req; + write_unlock(&queue->syn_wait_lock); +} + #endif /* _REQUEST_SOCK_H */ diff --git a/include/net/tcp.h b/include/net/tcp.h index 6663086a5e35..a2e323c54457 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1686,71 +1686,41 @@ static inline int tcp_full_space(const struct sock *sk) static inline void tcp_acceptq_queue(struct sock *sk, struct request_sock *req, struct sock *child) { - struct tcp_sock *tp = tcp_sk(sk); - - req->sk = child; - sk_acceptq_added(sk); - - if (!tp->accept_queue_tail) { - tp->accept_queue = req; - } else { - tp->accept_queue_tail->dl_next = req; - } - tp->accept_queue_tail = req; - req->dl_next = NULL; + reqsk_queue_add(&tcp_sk(sk)->accept_queue, req, sk, child); } -struct tcp_listen_opt -{ - u8 max_qlen_log; /* log_2 of maximal queued SYNs */ - int qlen; - int qlen_young; - int clock_hand; - u32 hash_rnd; - struct request_sock *syn_table[TCP_SYNQ_HSIZE]; -}; - static inline void tcp_synq_removed(struct sock *sk, struct request_sock *req) { - struct tcp_listen_opt *lopt = tcp_sk(sk)->listen_opt; - - if (--lopt->qlen == 0) + if (reqsk_queue_removed(&tcp_sk(sk)->accept_queue, req) == 0) tcp_delete_keepalive_timer(sk); - if (req->retrans == 0) - lopt->qlen_young--; } static inline void tcp_synq_added(struct sock *sk) { - struct tcp_listen_opt *lopt = tcp_sk(sk)->listen_opt; - - if (lopt->qlen++ == 0) + if (reqsk_queue_added(&tcp_sk(sk)->accept_queue) == 0) tcp_reset_keepalive_timer(sk, TCP_TIMEOUT_INIT); - lopt->qlen_young++; } static inline int tcp_synq_len(struct sock *sk) { - return tcp_sk(sk)->listen_opt->qlen; + return reqsk_queue_len(&tcp_sk(sk)->accept_queue); } static inline int tcp_synq_young(struct sock *sk) { - return tcp_sk(sk)->listen_opt->qlen_young; + return reqsk_queue_len_young(&tcp_sk(sk)->accept_queue); } static inline int tcp_synq_is_full(struct sock *sk) { - return tcp_synq_len(sk) >> tcp_sk(sk)->listen_opt->max_qlen_log; + return reqsk_queue_is_full(&tcp_sk(sk)->accept_queue); } static inline void tcp_synq_unlink(struct tcp_sock *tp, struct request_sock *req, - struct request_sock **prev) + struct request_sock **prev) { - write_lock(&tp->syn_wait_lock); - *prev = req->dl_next; - write_unlock(&tp->syn_wait_lock); + reqsk_queue_unlink(&tp->accept_queue, req, prev); } static inline void tcp_synq_drop(struct sock *sk, struct request_sock *req, diff --git a/net/core/Makefile b/net/core/Makefile index 81f03243fe2f..5e0c56b7f607 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -2,7 +2,8 @@ # Makefile for the Linux networking core. # -obj-y := sock.o skbuff.o iovec.o datagram.o stream.o scm.o gen_stats.o gen_estimator.o +obj-y := sock.o request_sock.o skbuff.o iovec.o datagram.o stream.o scm.o \ + gen_stats.o gen_estimator.o obj-$(CONFIG_SYSCTL) += sysctl_net_core.o diff --git a/net/core/request_sock.c b/net/core/request_sock.c new file mode 100644 index 000000000000..1258333ca007 --- /dev/null +++ b/net/core/request_sock.c @@ -0,0 +1,48 @@ +/* + * NET Generic infrastructure for Network protocols. + * + * Authors: Arnaldo Carvalho de Melo + * + * From code originally in include/net/tcp.h + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include + +#include + +int reqsk_queue_alloc(struct request_sock_queue *queue, + const int nr_table_entries) +{ + const int lopt_size = sizeof(struct tcp_listen_opt) + + nr_table_entries * sizeof(struct request_sock *); + struct tcp_listen_opt *lopt = kmalloc(lopt_size, GFP_KERNEL); + + if (lopt == NULL) + return -ENOMEM; + + memset(lopt, 0, lopt_size); + + for (lopt->max_qlen_log = 6; + (1 << lopt->max_qlen_log) < sysctl_max_syn_backlog; + lopt->max_qlen_log++); + + get_random_bytes(&lopt->hash_rnd, sizeof(lopt->hash_rnd)); + rwlock_init(&queue->syn_wait_lock); + queue->rskq_accept_head = queue->rskq_accept_head = NULL; + + write_lock_bh(&queue->syn_wait_lock); + queue->listen_opt = lopt; + write_unlock_bh(&queue->syn_wait_lock); + + return 0; +} + +EXPORT_SYMBOL(reqsk_queue_alloc); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 1c29feb6b35f..b85a46dd40a0 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -316,7 +316,7 @@ EXPORT_SYMBOL(tcp_enter_memory_pressure); static __inline__ unsigned int tcp_listen_poll(struct sock *sk, poll_table *wait) { - return tcp_sk(sk)->accept_queue ? (POLLIN | POLLRDNORM) : 0; + return !reqsk_queue_empty(&tcp_sk(sk)->accept_queue) ? (POLLIN | POLLRDNORM) : 0; } /* @@ -462,28 +462,15 @@ int tcp_listen_start(struct sock *sk) { struct inet_sock *inet = inet_sk(sk); struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt; + int rc = reqsk_queue_alloc(&tp->accept_queue, TCP_SYNQ_HSIZE); + + if (rc != 0) + return rc; sk->sk_max_ack_backlog = 0; sk->sk_ack_backlog = 0; - tp->accept_queue = tp->accept_queue_tail = NULL; - rwlock_init(&tp->syn_wait_lock); tcp_delack_init(tp); - lopt = kmalloc(sizeof(struct tcp_listen_opt), GFP_KERNEL); - if (!lopt) - return -ENOMEM; - - memset(lopt, 0, sizeof(struct tcp_listen_opt)); - for (lopt->max_qlen_log = 6; ; lopt->max_qlen_log++) - if ((1 << lopt->max_qlen_log) >= sysctl_max_syn_backlog) - break; - get_random_bytes(&lopt->hash_rnd, 4); - - write_lock_bh(&tp->syn_wait_lock); - tp->listen_opt = lopt; - write_unlock_bh(&tp->syn_wait_lock); - /* There is race window here: we announce ourselves listening, * but this transition is still not validated by get_port(). * It is OK, because this socket enters to hash table only @@ -500,10 +487,7 @@ int tcp_listen_start(struct sock *sk) } sk->sk_state = TCP_CLOSE; - write_lock_bh(&tp->syn_wait_lock); - tp->listen_opt = NULL; - write_unlock_bh(&tp->syn_wait_lock); - kfree(lopt); + reqsk_queue_destroy(&tp->accept_queue); return -EADDRINUSE; } @@ -515,18 +499,16 @@ int tcp_listen_start(struct sock *sk) static void tcp_listen_stop (struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; - struct request_sock *acc_req = tp->accept_queue; + struct tcp_listen_opt *lopt; + struct request_sock *acc_req; struct request_sock *req; int i; tcp_delete_keepalive_timer(sk); /* make all the listen_opt local to us */ - write_lock_bh(&tp->syn_wait_lock); - tp->listen_opt = NULL; - write_unlock_bh(&tp->syn_wait_lock); - tp->accept_queue = tp->accept_queue_tail = NULL; + lopt = reqsk_queue_yank_listen_sk(&tp->accept_queue); + acc_req = reqsk_queue_yank_acceptq(&tp->accept_queue); if (lopt->qlen) { for (i = 0; i < TCP_SYNQ_HSIZE; i++) { @@ -1867,11 +1849,11 @@ static int wait_for_connect(struct sock *sk, long timeo) prepare_to_wait_exclusive(sk->sk_sleep, &wait, TASK_INTERRUPTIBLE); release_sock(sk); - if (!tp->accept_queue) + if (reqsk_queue_empty(&tp->accept_queue)) timeo = schedule_timeout(timeo); lock_sock(sk); err = 0; - if (tp->accept_queue) + if (!reqsk_queue_empty(&tp->accept_queue)) break; err = -EINVAL; if (sk->sk_state != TCP_LISTEN) @@ -1894,7 +1876,6 @@ static int wait_for_connect(struct sock *sk, long timeo) struct sock *tcp_accept(struct sock *sk, int flags, int *err) { struct tcp_sock *tp = tcp_sk(sk); - struct request_sock *req; struct sock *newsk; int error; @@ -1905,37 +1886,31 @@ struct sock *tcp_accept(struct sock *sk, int flags, int *err) */ error = -EINVAL; if (sk->sk_state != TCP_LISTEN) - goto out; + goto out_err; /* Find already established connection */ - if (!tp->accept_queue) { + if (reqsk_queue_empty(&tp->accept_queue)) { long timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); /* If this is a non blocking socket don't sleep */ error = -EAGAIN; if (!timeo) - goto out; + goto out_err; error = wait_for_connect(sk, timeo); if (error) - goto out; + goto out_err; } - req = tp->accept_queue; - if ((tp->accept_queue = req->dl_next) == NULL) - tp->accept_queue_tail = NULL; - - newsk = req->sk; - sk_acceptq_removed(sk); - __reqsk_free(req); + newsk = reqsk_queue_get_child(&tp->accept_queue, sk); BUG_TRAP(newsk->sk_state != TCP_SYN_RECV); - release_sock(sk); - return newsk; - out: release_sock(sk); + return newsk; +out_err: + newsk = NULL; *err = error; - return NULL; + goto out; } /* diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 67277800d0c1..c3328fa48837 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -529,9 +529,9 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, entry.family = sk->sk_family; - read_lock_bh(&tp->syn_wait_lock); + read_lock_bh(&tp->accept_queue.syn_wait_lock); - lopt = tp->listen_opt; + lopt = tp->accept_queue.listen_opt; if (!lopt || !lopt->qlen) goto out; @@ -588,7 +588,7 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, } out: - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); return err; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 95528a75a63d..1745dc8d25e6 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -874,7 +874,7 @@ static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp, __u16 rport, __u32 raddr, __u32 laddr) { - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v4_synq_hash(raddr, rport, lopt->hash_rnd)]; @@ -898,18 +898,10 @@ static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp, static void tcp_v4_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; u32 h = tcp_v4_synq_hash(inet_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); - req->expires = jiffies + TCP_TIMEOUT_INIT; - req->retrans = 0; - req->sk = NULL; - req->dl_next = lopt->syn_table[h]; - - write_lock(&tp->syn_wait_lock); - lopt->syn_table[h] = req; - write_unlock(&tp->syn_wait_lock); - + reqsk_queue_hash_req(&tp->accept_queue, h, req, TCP_TIMEOUT_INIT); tcp_synq_added(sk); } @@ -2167,17 +2159,17 @@ static void *listening_get_next(struct seq_file *seq, void *cur) if (++st->sbucket >= TCP_SYNQ_HSIZE) break; get_req: - req = tp->listen_opt->syn_table[st->sbucket]; + req = tp->accept_queue.listen_opt->syn_table[st->sbucket]; } sk = sk_next(st->syn_wait_sk); st->state = TCP_SEQ_STATE_LISTENING; - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); } else { tp = tcp_sk(sk); - read_lock_bh(&tp->syn_wait_lock); - if (tp->listen_opt && tp->listen_opt->qlen) + read_lock_bh(&tp->accept_queue.syn_wait_lock); + if (reqsk_queue_len(&tp->accept_queue)) goto start_req; - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); sk = sk_next(sk); } get_sk: @@ -2187,8 +2179,8 @@ get_sk: goto out; } tp = tcp_sk(sk); - read_lock_bh(&tp->syn_wait_lock); - if (tp->listen_opt && tp->listen_opt->qlen) { + read_lock_bh(&tp->accept_queue.syn_wait_lock); + if (reqsk_queue_len(&tp->accept_queue)) { start_req: st->uid = sock_i_uid(sk); st->syn_wait_sk = sk; @@ -2196,7 +2188,7 @@ start_req: st->sbucket = 0; goto get_req; } - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); } if (++st->bucket < TCP_LHTABLE_SIZE) { sk = sk_head(&tcp_listening_hash[st->bucket]); @@ -2383,7 +2375,7 @@ static void tcp_seq_stop(struct seq_file *seq, void *v) case TCP_SEQ_STATE_OPENREQ: if (v) { struct tcp_sock *tp = tcp_sk(st->syn_wait_sk); - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); } case TCP_SEQ_STATE_LISTENING: if (v != SEQ_START_TOKEN) diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 0e6d525a8341..b3943e7562f3 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -790,10 +790,8 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, newtp->probes_out = 0; newtp->rx_opt.num_sacks = 0; newtp->urg_data = 0; - newtp->listen_opt = NULL; - newtp->accept_queue = newtp->accept_queue_tail = NULL; - /* Deinitialize syn_wait_lock to trap illegal accesses. */ - memset(&newtp->syn_wait_lock, 0, sizeof(newtp->syn_wait_lock)); + /* Deinitialize accept_queue to trap illegal accesses. */ + memset(&newtp->accept_queue, 0, sizeof(newtp->accept_queue)); /* Back to base struct sock members. */ newsk->sk_err = 0; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index f03efe5fb76a..d97d191149c1 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -464,7 +464,7 @@ out_unlock: static void tcp_synack_timer(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; int max_retries = tp->syn_retries ? : sysctl_tcp_synack_retries; int thresh = max_retries; unsigned long now = jiffies; @@ -527,12 +527,8 @@ static void tcp_synack_timer(struct sock *sk) } /* Drop this request */ - write_lock(&tp->syn_wait_lock); - *reqp = req->dl_next; - write_unlock(&tp->syn_wait_lock); - lopt->qlen--; - if (req->retrans == 0) - lopt->qlen_young--; + tcp_synq_unlink(tp, req, reqp); + reqsk_queue_removed(&tp->accept_queue, req); reqsk_free(req); continue; } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 068cd4a8c292..84091daad6b5 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -401,7 +401,7 @@ static struct request_sock *tcp_v6_search_req(struct tcp_sock *tp, struct in6_addr *laddr, int iif) { - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v6_synq_hash(raddr, rport, lopt->hash_rnd)]; @@ -1267,18 +1267,10 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) static void tcp_v6_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; u32 h = tcp_v6_synq_hash(&tcp6_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); - req->sk = NULL; - req->expires = jiffies + TCP_TIMEOUT_INIT; - req->retrans = 0; - req->dl_next = lopt->syn_table[h]; - - write_lock(&tp->syn_wait_lock); - lopt->syn_table[h] = req; - write_unlock(&tp->syn_wait_lock); - + reqsk_queue_hash_req(&tp->accept_queue, h, req, TCP_TIMEOUT_INIT); tcp_synq_added(sk); } -- cgit v1.2.3 From f88a10d65620d97b6d0a7e352a3493c1b7e7409b Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:50:12 -0700 Subject: [NETLINK]: New message building macros NLMSG_PUT_ANSWER(skb, nlcb, type, length) Start a new netlink message as answer to a request, returns the message header. NLMSG_END(skb, nlh) End a netlink message, fixes total message length, returns skb->len. NLMSG_CANCEL(skb, nlh) Cancel the building process and trim whole message from skb again, returns -1. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/netlink.h | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index b2738ac8bc99..8d1cb419a930 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -171,8 +171,21 @@ __nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len) } #define NLMSG_PUT(skb, pid, seq, type, len) \ -({ if (skb_tailroom(skb) < (int)NLMSG_SPACE(len)) goto nlmsg_failure; \ - __nlmsg_put(skb, pid, seq, type, len); }) +({ if (skb_tailroom(skb) < (int)NLMSG_SPACE(len)) \ + goto nlmsg_failure; \ + __nlmsg_put(skb, pid, seq, type, len); }) + +#define NLMSG_PUT_ANSWER(skb, cb, type, len) \ + NLMSG_PUT(skb, NETLINK_CB((cb)->skb).pid, \ + (cb)->nlh->nlmsg_seq, type, len) + +#define NLMSG_END(skb, nlh) \ +({ (nlh)->nlmsg_len = (skb)->tail - (unsigned char *) (nlh); \ + (skb)->len; }) + +#define NLMSG_CANCEL(skb, nlh) \ +({ skb_trim(skb, (unsigned char *) (nlh) - (skb)->data); \ + -1; }) extern int netlink_dump_start(struct sock *ssk, struct sk_buff *skb, struct nlmsghdr *nlh, -- cgit v1.2.3 From 00768244923f66801958a8d2d103f7b65608c9b6 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:50:38 -0700 Subject: [NETLINK] Routing attribute related shortcuts RTA_GET_U(32|64)(tlv) Assumes TLV is a u32/u64 field and returns its value. RTA_GET_[M]SECS(tlv) Assumes TLV is a u64 and transports jiffies converted to seconds or milliseconds and returns its value. RTA_PUT_U(32|64)(skb, type, value) Appends %value as fixed u32/u64 to %skb as TLV %type. RTA_PUT_[M]SECS(skb, type, jiffies) Converts %jiffies to secs/msecs and appends it as u64 to %skb as TLV %type. RTA_PUT_STRING(skb, type, string) Appends %NUL terminated %string to %skb as TLV %type. RTA_NEST(skb, type) Starts a nested TLV %type and returns the nesting handle. RTA_NEST_END(skb, nesting_handle) Finishes the nested TLV %nesting_handle, must be called symmetric to RTA_NEST(). Returns skb->len RTA_NEST_CANCEL(skb, nesting_handle) Cancel the nested TLV %nesting_handle and trim nested TLV from skb again, returns -1. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 45 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 45 insertions(+) (limited to 'include/linux') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 91ac97c20777..a09b5d42babf 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -789,6 +789,51 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi ({ if (unlikely(skb_tailroom(skb) < (int)(attrlen))) \ goto rtattr_failure; \ memcpy(skb_put(skb, RTA_ALIGN(attrlen)), data, attrlen); }) + +#define RTA_PUT_U32(skb, attrtype, value) \ +({ u32 _tmp = (value); \ + RTA_PUT(skb, attrtype, sizeof(u32), &_tmp); }) + +#define RTA_PUT_U64(skb, attrtype, value) \ +({ u64 _tmp = (value); \ + RTA_PUT(skb, attrtype, sizeof(u64), &_tmp); }) + +#define RTA_PUT_SECS(skb, attrtype, value) \ + RTA_PUT_U64(skb, attrtype, (value) / HZ) + +#define RTA_PUT_MSECS(skb, attrtype, value) \ + RTA_PUT_U64(skb, attrtype, jiffies_to_msecs(value)) + +#define RTA_PUT_STRING(skb, attrtype, value) \ + RTA_PUT(skb, attrtype, strlen(value) + 1, value) + +#define RTA_NEST(skb, type) \ +({ struct rtattr *__start = (struct rtattr *) (skb)->tail; \ + RTA_PUT(skb, type, 0, NULL); \ + __start; }) + +#define RTA_NEST_END(skb, start) \ +({ (start)->rta_len = ((skb)->tail - (unsigned char *) (start)); \ + (skb)->len; }) + +#define RTA_NEST_CANCEL(skb, start) \ +({ skb_trim(skb, (unsigned char *) (start) - (skb)->data); \ + -1; }) + +#define RTA_GET_U32(rta) \ +({ if (!rta || RTA_PAYLOAD(rta) < sizeof(u32)) \ + goto rtattr_failure; \ + *(u32 *) RTA_DATA(rta); }) + +#define RTA_GET_U64(rta) \ +({ u64 _tmp; \ + if (!rta || RTA_PAYLOAD(rta) < sizeof(u64)) \ + goto rtattr_failure; \ + memcpy(&_tmp, RTA_DATA(rta), sizeof(_tmp)); \ + _tmp; }) + +#define RTA_GET_SECS(rta) ((unsigned long) RTA_GET_U64(rta) * HZ) +#define RTA_GET_MSECS(rta) (msecs_to_jiffies((unsigned long) RTA_GET_U64(rta))) static inline struct rtattr * __rta_reserve(struct sk_buff *skb, int attrtype, int attrlen) -- cgit v1.2.3 From c7fb64db001f83ece669c76a02d8ec2fdb1dd307 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:50:55 -0700 Subject: [NETLINK]: Neighbour table configuration and statistics via rtnetlink To retrieve the neighbour tables send RTM_GETNEIGHTBL with the NLM_F_DUMP flag set. Every neighbour table configuration is spread over multiple messages to avoid running into message size limits on systems with many interfaces. The first message in the sequence transports all not device specific data such as statistics, configuration, and the default parameter set. This message is followed by 0..n messages carrying device specific parameter sets. Although the ordering should be sufficient, NDTA_NAME can be used to identify sequences. The initial message can be identified by checking for NDTA_CONFIG. The device specific messages do not contain this TLV but have NDTPA_IFINDEX set to the corresponding interface index. To change neighbour table attributes, send RTM_SETNEIGHTBL with NDTA_NAME set. Changeable attribute include NDTA_THRESH[1-3], NDTA_GC_INTERVAL, and all TLVs in NDTA_PARMS unless marked otherwise. Device specific parameter sets can be changed by setting NDTPA_IFINDEX to the interface index of the corresponding device. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 107 +++++++++++++++ include/net/neighbour.h | 4 + net/core/neighbour.c | 317 +++++++++++++++++++++++++++++++++++++++++++- net/core/rtnetlink.c | 20 +-- security/selinux/nlmsgtab.c | 2 + 5 files changed, 439 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index a09b5d42babf..5a5cda160267 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -89,6 +89,13 @@ enum { RTM_GETANYCAST = 62, #define RTM_GETANYCAST RTM_GETANYCAST + RTM_NEWNEIGHTBL = 64, +#define RTM_NEWNEIGHTBL RTM_NEWNEIGHTBL + RTM_GETNEIGHTBL = 66, +#define RTM_GETNEIGHTBL RTM_GETNEIGHTBL + RTM_SETNEIGHTBL, +#define RTM_SETNEIGHTBL RTM_SETNEIGHTBL + __RTM_MAX, #define RTM_MAX (((__RTM_MAX + 3) & ~3) - 1) }; @@ -493,6 +500,106 @@ struct nda_cacheinfo __u32 ndm_refcnt; }; + +/***************************************************************** + * Neighbour tables specific messages. + * + * To retrieve the neighbour tables send RTM_GETNEIGHTBL with the + * NLM_F_DUMP flag set. Every neighbour table configuration is + * spread over multiple messages to avoid running into message + * size limits on systems with many interfaces. The first message + * in the sequence transports all not device specific data such as + * statistics, configuration, and the default parameter set. + * This message is followed by 0..n messages carrying device + * specific parameter sets. + * Although the ordering should be sufficient, NDTA_NAME can be + * used to identify sequences. The initial message can be identified + * by checking for NDTA_CONFIG. The device specific messages do + * not contain this TLV but have NDTPA_IFINDEX set to the + * corresponding interface index. + * + * To change neighbour table attributes, send RTM_SETNEIGHTBL + * with NDTA_NAME set. Changeable attribute include NDTA_THRESH[1-3], + * NDTA_GC_INTERVAL, and all TLVs in NDTA_PARMS unless marked + * otherwise. Device specific parameter sets can be changed by + * setting NDTPA_IFINDEX to the interface index of the corresponding + * device. + ****/ + +struct ndt_stats +{ + __u64 ndts_allocs; + __u64 ndts_destroys; + __u64 ndts_hash_grows; + __u64 ndts_res_failed; + __u64 ndts_lookups; + __u64 ndts_hits; + __u64 ndts_rcv_probes_mcast; + __u64 ndts_rcv_probes_ucast; + __u64 ndts_periodic_gc_runs; + __u64 ndts_forced_gc_runs; +}; + +enum { + NDTPA_UNSPEC, + NDTPA_IFINDEX, /* u32, unchangeable */ + NDTPA_REFCNT, /* u32, read-only */ + NDTPA_REACHABLE_TIME, /* u64, read-only, msecs */ + NDTPA_BASE_REACHABLE_TIME, /* u64, msecs */ + NDTPA_RETRANS_TIME, /* u64, msecs */ + NDTPA_GC_STALETIME, /* u64, msecs */ + NDTPA_DELAY_PROBE_TIME, /* u64, msecs */ + NDTPA_QUEUE_LEN, /* u32 */ + NDTPA_APP_PROBES, /* u32 */ + NDTPA_UCAST_PROBES, /* u32 */ + NDTPA_MCAST_PROBES, /* u32 */ + NDTPA_ANYCAST_DELAY, /* u64, msecs */ + NDTPA_PROXY_DELAY, /* u64, msecs */ + NDTPA_PROXY_QLEN, /* u32 */ + NDTPA_LOCKTIME, /* u64, msecs */ + __NDTPA_MAX +}; +#define NDTPA_MAX (__NDTPA_MAX - 1) + +struct ndtmsg +{ + __u8 ndtm_family; + __u8 ndtm_pad1; + __u16 ndtm_pad2; +}; + +struct ndt_config +{ + __u16 ndtc_key_len; + __u16 ndtc_entry_size; + __u32 ndtc_entries; + __u32 ndtc_last_flush; /* delta to now in msecs */ + __u32 ndtc_last_rand; /* delta to now in msecs */ + __u32 ndtc_hash_rnd; + __u32 ndtc_hash_mask; + __u32 ndtc_hash_chain_gc; + __u32 ndtc_proxy_qlen; +}; + +enum { + NDTA_UNSPEC, + NDTA_NAME, /* char *, unchangeable */ + NDTA_THRESH1, /* u32 */ + NDTA_THRESH2, /* u32 */ + NDTA_THRESH3, /* u32 */ + NDTA_CONFIG, /* struct ndt_config, read-only */ + NDTA_PARMS, /* nested TLV NDTPA_* */ + NDTA_STATS, /* struct ndt_stats, read-only */ + NDTA_GC_INTERVAL, /* u64, msecs */ + __NDTA_MAX +}; +#define NDTA_MAX (__NDTA_MAX - 1) + +#define NDTA_RTA(r) ((struct rtattr*)(((char*)(r)) + \ + NLMSG_ALIGN(sizeof(struct ndtmsg)))) +#define NDTA_PAYLOAD(n) NLMSG_PAYLOAD(n,sizeof(struct ndtmsg)) + + /**** * General form of address family dependent message. ****/ diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 4f33bbc21e7f..17191ac9be70 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -65,6 +65,7 @@ struct neighbour; struct neigh_parms { + struct net_device *dev; struct neigh_parms *next; int (*neigh_setup)(struct neighbour *); struct neigh_table *tbl; @@ -252,6 +253,9 @@ extern int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg); extern int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg); extern void neigh_app_ns(struct neighbour *n); +extern int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb); +extern int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg); + extern void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void *), void *cookie); extern void __neigh_for_each_release(struct neigh_table *tbl, int (*cb)(struct neighbour *)); extern void pneigh_for_each(struct neigh_table *tbl, void (*cb)(struct pneigh_entry *)); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 43bdc521e20d..0841ac78c67d 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1276,9 +1276,14 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev, INIT_RCU_HEAD(&p->rcu_head); p->reachable_time = neigh_rand_reach_time(p->base_reachable_time); - if (dev && dev->neigh_setup && dev->neigh_setup(dev, p)) { - kfree(p); - return NULL; + if (dev) { + if (dev->neigh_setup && dev->neigh_setup(dev, p)) { + kfree(p); + return NULL; + } + + dev_hold(dev); + p->dev = dev; } p->sysctl_table = NULL; write_lock_bh(&tbl->lock); @@ -1309,6 +1314,8 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms) *p = parms->next; parms->dead = 1; write_unlock_bh(&tbl->lock); + if (parms->dev) + dev_put(parms->dev); call_rcu(&parms->rcu_head, neigh_rcu_free_parms); return; } @@ -1546,6 +1553,308 @@ out: return err; } +static int neightbl_fill_parms(struct sk_buff *skb, struct neigh_parms *parms) +{ + struct rtattr *nest = RTA_NEST(skb, NDTA_PARMS); + + if (parms->dev) + RTA_PUT_U32(skb, NDTPA_IFINDEX, parms->dev->ifindex); + + RTA_PUT_U32(skb, NDTPA_REFCNT, atomic_read(&parms->refcnt)); + RTA_PUT_U32(skb, NDTPA_QUEUE_LEN, parms->queue_len); + RTA_PUT_U32(skb, NDTPA_PROXY_QLEN, parms->proxy_qlen); + RTA_PUT_U32(skb, NDTPA_APP_PROBES, parms->app_probes); + RTA_PUT_U32(skb, NDTPA_UCAST_PROBES, parms->ucast_probes); + RTA_PUT_U32(skb, NDTPA_MCAST_PROBES, parms->mcast_probes); + RTA_PUT_MSECS(skb, NDTPA_REACHABLE_TIME, parms->reachable_time); + RTA_PUT_MSECS(skb, NDTPA_BASE_REACHABLE_TIME, + parms->base_reachable_time); + RTA_PUT_MSECS(skb, NDTPA_GC_STALETIME, parms->gc_staletime); + RTA_PUT_MSECS(skb, NDTPA_DELAY_PROBE_TIME, parms->delay_probe_time); + RTA_PUT_MSECS(skb, NDTPA_RETRANS_TIME, parms->retrans_time); + RTA_PUT_MSECS(skb, NDTPA_ANYCAST_DELAY, parms->anycast_delay); + RTA_PUT_MSECS(skb, NDTPA_PROXY_DELAY, parms->proxy_delay); + RTA_PUT_MSECS(skb, NDTPA_LOCKTIME, parms->locktime); + + return RTA_NEST_END(skb, nest); + +rtattr_failure: + return RTA_NEST_CANCEL(skb, nest); +} + +static int neightbl_fill_info(struct neigh_table *tbl, struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct nlmsghdr *nlh; + struct ndtmsg *ndtmsg; + + nlh = NLMSG_PUT_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg)); + ndtmsg = NLMSG_DATA(nlh); + + NLMSG_SET_MULTIPART(nlh); + + read_lock_bh(&tbl->lock); + ndtmsg->ndtm_family = tbl->family; + + RTA_PUT_STRING(skb, NDTA_NAME, tbl->id); + RTA_PUT_MSECS(skb, NDTA_GC_INTERVAL, tbl->gc_interval); + RTA_PUT_U32(skb, NDTA_THRESH1, tbl->gc_thresh1); + RTA_PUT_U32(skb, NDTA_THRESH2, tbl->gc_thresh2); + RTA_PUT_U32(skb, NDTA_THRESH3, tbl->gc_thresh3); + + { + unsigned long now = jiffies; + unsigned int flush_delta = now - tbl->last_flush; + unsigned int rand_delta = now - tbl->last_rand; + + struct ndt_config ndc = { + .ndtc_key_len = tbl->key_len, + .ndtc_entry_size = tbl->entry_size, + .ndtc_entries = atomic_read(&tbl->entries), + .ndtc_last_flush = jiffies_to_msecs(flush_delta), + .ndtc_last_rand = jiffies_to_msecs(rand_delta), + .ndtc_hash_rnd = tbl->hash_rnd, + .ndtc_hash_mask = tbl->hash_mask, + .ndtc_hash_chain_gc = tbl->hash_chain_gc, + .ndtc_proxy_qlen = tbl->proxy_queue.qlen, + }; + + RTA_PUT(skb, NDTA_CONFIG, sizeof(ndc), &ndc); + } + + { + int cpu; + struct ndt_stats ndst; + + memset(&ndst, 0, sizeof(ndst)); + + for (cpu = 0; cpu < NR_CPUS; cpu++) { + struct neigh_statistics *st; + + if (!cpu_possible(cpu)) + continue; + + st = per_cpu_ptr(tbl->stats, cpu); + ndst.ndts_allocs += st->allocs; + ndst.ndts_destroys += st->destroys; + ndst.ndts_hash_grows += st->hash_grows; + ndst.ndts_res_failed += st->res_failed; + ndst.ndts_lookups += st->lookups; + ndst.ndts_hits += st->hits; + ndst.ndts_rcv_probes_mcast += st->rcv_probes_mcast; + ndst.ndts_rcv_probes_ucast += st->rcv_probes_ucast; + ndst.ndts_periodic_gc_runs += st->periodic_gc_runs; + ndst.ndts_forced_gc_runs += st->forced_gc_runs; + } + + RTA_PUT(skb, NDTA_STATS, sizeof(ndst), &ndst); + } + + BUG_ON(tbl->parms.dev); + if (neightbl_fill_parms(skb, &tbl->parms) < 0) + goto rtattr_failure; + + read_unlock_bh(&tbl->lock); + return NLMSG_END(skb, nlh); + +rtattr_failure: + read_unlock_bh(&tbl->lock); + return NLMSG_CANCEL(skb, nlh); + +nlmsg_failure: + return -1; +} + +static int neightbl_fill_param_info(struct neigh_table *tbl, + struct neigh_parms *parms, + struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct ndtmsg *ndtmsg; + struct nlmsghdr *nlh; + + nlh = NLMSG_PUT_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg)); + ndtmsg = NLMSG_DATA(nlh); + + NLMSG_SET_MULTIPART(nlh); + + read_lock_bh(&tbl->lock); + ndtmsg->ndtm_family = tbl->family; + RTA_PUT_STRING(skb, NDTA_NAME, tbl->id); + + if (neightbl_fill_parms(skb, parms) < 0) + goto rtattr_failure; + + read_unlock_bh(&tbl->lock); + return NLMSG_END(skb, nlh); + +rtattr_failure: + read_unlock_bh(&tbl->lock); + return NLMSG_CANCEL(skb, nlh); + +nlmsg_failure: + return -1; +} + +static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl, + int ifindex) +{ + struct neigh_parms *p; + + for (p = &tbl->parms; p; p = p->next) + if ((p->dev && p->dev->ifindex == ifindex) || + (!p->dev && !ifindex)) + return p; + + return NULL; +} + +int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) +{ + struct neigh_table *tbl; + struct ndtmsg *ndtmsg = NLMSG_DATA(nlh); + struct rtattr **tb = arg; + int err = -EINVAL; + + if (!tb[NDTA_NAME - 1] || !RTA_PAYLOAD(tb[NDTA_NAME - 1])) + return -EINVAL; + + read_lock(&neigh_tbl_lock); + for (tbl = neigh_tables; tbl; tbl = tbl->next) { + if (ndtmsg->ndtm_family && tbl->family != ndtmsg->ndtm_family) + continue; + + if (!rtattr_strcmp(tb[NDTA_NAME - 1], tbl->id)) + break; + } + + if (tbl == NULL) { + err = -ENOENT; + goto errout; + } + + /* + * We acquire tbl->lock to be nice to the periodic timers and + * make sure they always see a consistent set of values. + */ + write_lock_bh(&tbl->lock); + + if (tb[NDTA_THRESH1 - 1]) + tbl->gc_thresh1 = RTA_GET_U32(tb[NDTA_THRESH1 - 1]); + + if (tb[NDTA_THRESH2 - 1]) + tbl->gc_thresh2 = RTA_GET_U32(tb[NDTA_THRESH2 - 1]); + + if (tb[NDTA_THRESH3 - 1]) + tbl->gc_thresh3 = RTA_GET_U32(tb[NDTA_THRESH3 - 1]); + + if (tb[NDTA_GC_INTERVAL - 1]) + tbl->gc_interval = RTA_GET_MSECS(tb[NDTA_GC_INTERVAL - 1]); + + if (tb[NDTA_PARMS - 1]) { + struct rtattr *tbp[NDTPA_MAX]; + struct neigh_parms *p; + u32 ifindex = 0; + + if (rtattr_parse_nested(tbp, NDTPA_MAX, tb[NDTA_PARMS - 1]) < 0) + goto rtattr_failure; + + if (tbp[NDTPA_IFINDEX - 1]) + ifindex = RTA_GET_U32(tbp[NDTPA_IFINDEX - 1]); + + p = lookup_neigh_params(tbl, ifindex); + if (p == NULL) { + err = -ENOENT; + goto rtattr_failure; + } + + if (tbp[NDTPA_QUEUE_LEN - 1]) + p->queue_len = RTA_GET_U32(tbp[NDTPA_QUEUE_LEN - 1]); + + if (tbp[NDTPA_PROXY_QLEN - 1]) + p->proxy_qlen = RTA_GET_U32(tbp[NDTPA_PROXY_QLEN - 1]); + + if (tbp[NDTPA_APP_PROBES - 1]) + p->app_probes = RTA_GET_U32(tbp[NDTPA_APP_PROBES - 1]); + + if (tbp[NDTPA_UCAST_PROBES - 1]) + p->ucast_probes = + RTA_GET_U32(tbp[NDTPA_UCAST_PROBES - 1]); + + if (tbp[NDTPA_MCAST_PROBES - 1]) + p->mcast_probes = + RTA_GET_U32(tbp[NDTPA_MCAST_PROBES - 1]); + + if (tbp[NDTPA_BASE_REACHABLE_TIME - 1]) + p->base_reachable_time = + RTA_GET_MSECS(tbp[NDTPA_BASE_REACHABLE_TIME - 1]); + + if (tbp[NDTPA_GC_STALETIME - 1]) + p->gc_staletime = + RTA_GET_MSECS(tbp[NDTPA_GC_STALETIME - 1]); + + if (tbp[NDTPA_DELAY_PROBE_TIME - 1]) + p->delay_probe_time = + RTA_GET_MSECS(tbp[NDTPA_DELAY_PROBE_TIME - 1]); + + if (tbp[NDTPA_RETRANS_TIME - 1]) + p->retrans_time = + RTA_GET_MSECS(tbp[NDTPA_RETRANS_TIME - 1]); + + if (tbp[NDTPA_ANYCAST_DELAY - 1]) + p->anycast_delay = + RTA_GET_MSECS(tbp[NDTPA_ANYCAST_DELAY - 1]); + + if (tbp[NDTPA_PROXY_DELAY - 1]) + p->proxy_delay = + RTA_GET_MSECS(tbp[NDTPA_PROXY_DELAY - 1]); + + if (tbp[NDTPA_LOCKTIME - 1]) + p->locktime = RTA_GET_MSECS(tbp[NDTPA_LOCKTIME - 1]); + } + + err = 0; + +rtattr_failure: + write_unlock_bh(&tbl->lock); +errout: + read_unlock(&neigh_tbl_lock); + return err; +} + +int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) +{ + int idx, family; + int s_idx = cb->args[0]; + struct neigh_table *tbl; + + family = ((struct rtgenmsg *)NLMSG_DATA(cb->nlh))->rtgen_family; + + read_lock(&neigh_tbl_lock); + for (tbl = neigh_tables, idx = 0; tbl; tbl = tbl->next) { + struct neigh_parms *p; + + if (idx < s_idx || (family && tbl->family != family)) + continue; + + if (neightbl_fill_info(tbl, skb, cb) <= 0) + break; + + for (++idx, p = tbl->parms.next; p; p = p->next, idx++) { + if (idx < s_idx) + continue; + + if (neightbl_fill_param_info(tbl, p, skb, cb) <= 0) + goto out; + } + + } +out: + read_unlock(&neigh_tbl_lock); + cb->args[0] = idx; + + return skb->len; +} static int neigh_fill_info(struct sk_buff *skb, struct neighbour *n, u32 pid, u32 seq, int event) @@ -2352,6 +2661,8 @@ EXPORT_SYMBOL(neigh_update); EXPORT_SYMBOL(neigh_update_hhs); EXPORT_SYMBOL(pneigh_enqueue); EXPORT_SYMBOL(pneigh_lookup); +EXPORT_SYMBOL(neightbl_dump_info); +EXPORT_SYMBOL(neightbl_set); #ifdef CONFIG_ARPD EXPORT_SYMBOL(neigh_app_ns); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 00caf4b318b2..56a20f014b8a 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -100,6 +100,7 @@ static const int rtm_min[RTM_NR_FAMILIES] = [RTM_FAM(RTM_NEWPREFIX)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETANYCAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), + [RTM_FAM(RTM_NEWNEIGHTBL)] = NLMSG_LENGTH(sizeof(struct ndtmsg)), }; static const int rta_max[RTM_NR_FAMILIES] = @@ -113,6 +114,7 @@ static const int rta_max[RTM_NR_FAMILIES] = [RTM_FAM(RTM_NEWTCLASS)] = TCA_MAX, [RTM_FAM(RTM_NEWTFILTER)] = TCA_MAX, [RTM_FAM(RTM_NEWACTION)] = TCAA_MAX, + [RTM_FAM(RTM_NEWNEIGHTBL)] = NDTA_MAX, }; void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data) @@ -649,14 +651,16 @@ static void rtnetlink_rcv(struct sock *sk, int len) static struct rtnetlink_link link_rtnetlink_table[RTM_NR_MSGTYPES] = { - [RTM_GETLINK - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo }, - [RTM_SETLINK - RTM_BASE] = { .doit = do_setlink }, - [RTM_GETADDR - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, - [RTM_GETROUTE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, - [RTM_NEWNEIGH - RTM_BASE] = { .doit = neigh_add }, - [RTM_DELNEIGH - RTM_BASE] = { .doit = neigh_delete }, - [RTM_GETNEIGH - RTM_BASE] = { .dumpit = neigh_dump_info }, - [RTM_GETRULE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_GETLINK - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo }, + [RTM_SETLINK - RTM_BASE] = { .doit = do_setlink }, + [RTM_GETADDR - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_GETROUTE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_NEWNEIGH - RTM_BASE] = { .doit = neigh_add }, + [RTM_DELNEIGH - RTM_BASE] = { .doit = neigh_delete }, + [RTM_GETNEIGH - RTM_BASE] = { .dumpit = neigh_dump_info }, + [RTM_GETRULE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_GETNEIGHTBL - RTM_BASE] = { .dumpit = neightbl_dump_info }, + [RTM_SETNEIGHTBL - RTM_BASE] = { .doit = neightbl_set }, }; static int rtnetlink_event(struct notifier_block *this, unsigned long event, void *ptr) diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index f0fb6d76f7c5..92b057becb4b 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -63,6 +63,8 @@ static struct nlmsg_perm nlmsg_route_perms[] = { RTM_GETPREFIX, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETMULTICAST, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETANYCAST, NETLINK_ROUTE_SOCKET__NLMSG_READ }, + { RTM_GETNEIGHTBL, NETLINK_ROUTE_SOCKET__NLMSG_READ }, + { RTM_SETNEIGHTBL, NETLINK_ROUTE_SOCKET__NLMSG_WRITE }, }; static struct nlmsg_perm nlmsg_firewall_perms[] = -- cgit v1.2.3 From c52a3f89f882b84fc422000655c023fe73e701cf Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:51:26 -0700 Subject: [NETLINK]: Fix RTA_NEST_CANCEL(). Only skb_trim() if 'start' is non-NULL. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 5a5cda160267..6fff8c4c99c7 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -924,7 +924,8 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi (skb)->len; }) #define RTA_NEST_CANCEL(skb, start) \ -({ skb_trim(skb, (unsigned char *) (start) - (skb)->data); \ +({ if (start) \ + skb_trim(skb, (unsigned char *) (start) - (skb)->data); \ -1; }) #define RTA_GET_U32(rta) \ -- cgit v1.2.3 From 8f48bcd4ef11a69add178fc3111a77e7ee95bacd Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:52:36 -0700 Subject: [RTNETLINK]: Add RTA_(PUT|GET) shortcuts for u8, u16, and flag Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) (limited to 'include/linux') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 6fff8c4c99c7..e68dbf0bf579 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -897,6 +897,14 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi goto rtattr_failure; \ memcpy(skb_put(skb, RTA_ALIGN(attrlen)), data, attrlen); }) +#define RTA_PUT_U8(skb, attrtype, value) \ +({ u8 _tmp = (value); \ + RTA_PUT(skb, attrtype, sizeof(u8), &_tmp); }) + +#define RTA_PUT_U16(skb, attrtype, value) \ +({ u16 _tmp = (value); \ + RTA_PUT(skb, attrtype, sizeof(u16), &_tmp); }) + #define RTA_PUT_U32(skb, attrtype, value) \ ({ u32 _tmp = (value); \ RTA_PUT(skb, attrtype, sizeof(u32), &_tmp); }) @@ -914,6 +922,9 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi #define RTA_PUT_STRING(skb, attrtype, value) \ RTA_PUT(skb, attrtype, strlen(value) + 1, value) +#define RTA_PUT_FLAG(skb, attrtype) \ + RTA_PUT(skb, attrtype, 0, NULL); + #define RTA_NEST(skb, type) \ ({ struct rtattr *__start = (struct rtattr *) (skb)->tail; \ RTA_PUT(skb, type, 0, NULL); \ @@ -928,6 +939,16 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi skb_trim(skb, (unsigned char *) (start) - (skb)->data); \ -1; }) +#define RTA_GET_U8(rta) \ +({ if (!rta || RTA_PAYLOAD(rta) < sizeof(u8)) \ + goto rtattr_failure; \ + *(u8 *) RTA_DATA(rta); }) + +#define RTA_GET_U16(rta) \ +({ if (!rta || RTA_PAYLOAD(rta) < sizeof(u16)) \ + goto rtattr_failure; \ + *(u16 *) RTA_DATA(rta); }) + #define RTA_GET_U32(rta) \ ({ if (!rta || RTA_PAYLOAD(rta) < sizeof(u32)) \ goto rtattr_failure; \ @@ -940,6 +961,8 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi memcpy(&_tmp, RTA_DATA(rta), sizeof(_tmp)); \ _tmp; }) +#define RTA_GET_FLAG(rta) (!!(rta)) + #define RTA_GET_SECS(rta) ((unsigned long) RTA_GET_U64(rta) * HZ) #define RTA_GET_MSECS(rta) (msecs_to_jiffies((unsigned long) RTA_GET_U64(rta))) -- cgit v1.2.3 From 1797754ea7ee5e0d859b0a32506ff999f8d5fb71 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:53:48 -0700 Subject: [NETLINK]: Introduce NLMSG_NEW macro to better handle netlink flags Introduces a new macro NLMSG_NEW which extends NLMSG_PUT but takes a flags argument. NLMSG_PUT stays there for compatibility but now calls NLMSG_NEW with flags == 0. NLMSG_PUT_ANSWER is renamed to NLMSG_NEW_ANSWER which now also takes a flags argument. Also converts the users of NLMSG_PUT_ANSWER to use NLMSG_NEW_ANSWER and fixes the two direct users of __nlmsg_put to either provide the flags or use NLMSG_NEW(_ANSWER). Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/netlink.h | 17 ++++++++++------- net/core/neighbour.c | 8 ++++---- net/netlink/af_netlink.c | 8 +++++--- 3 files changed, 19 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 8d1cb419a930..e38407a23d04 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -156,7 +156,7 @@ struct netlink_notify }; static __inline__ struct nlmsghdr * -__nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len) +__nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len, int flags) { struct nlmsghdr *nlh; int size = NLMSG_LENGTH(len); @@ -164,20 +164,23 @@ __nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len) nlh = (struct nlmsghdr*)skb_put(skb, NLMSG_ALIGN(size)); nlh->nlmsg_type = type; nlh->nlmsg_len = size; - nlh->nlmsg_flags = 0; + nlh->nlmsg_flags = flags; nlh->nlmsg_pid = pid; nlh->nlmsg_seq = seq; return nlh; } -#define NLMSG_PUT(skb, pid, seq, type, len) \ +#define NLMSG_NEW(skb, pid, seq, type, len, flags) \ ({ if (skb_tailroom(skb) < (int)NLMSG_SPACE(len)) \ goto nlmsg_failure; \ - __nlmsg_put(skb, pid, seq, type, len); }) + __nlmsg_put(skb, pid, seq, type, len, flags); }) + +#define NLMSG_PUT(skb, pid, seq, type, len) \ + NLMSG_NEW(skb, pid, seq, type, len, 0) -#define NLMSG_PUT_ANSWER(skb, cb, type, len) \ - NLMSG_PUT(skb, NETLINK_CB((cb)->skb).pid, \ - (cb)->nlh->nlmsg_seq, type, len) +#define NLMSG_NEW_ANSWER(skb, cb, type, len, flags) \ + NLMSG_NEW(skb, NETLINK_CB((cb)->skb).pid, \ + (cb)->nlh->nlmsg_seq, type, len, flags) #define NLMSG_END(skb, nlh) \ ({ (nlh)->nlmsg_len = (skb)->tail - (unsigned char *) (nlh); \ diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 2296a145fb78..0fb742e228cc 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1590,8 +1590,8 @@ static int neightbl_fill_info(struct neigh_table *tbl, struct sk_buff *skb, struct nlmsghdr *nlh; struct ndtmsg *ndtmsg; - nlh = NLMSG_PUT_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg)); - nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg), + NLM_F_MULTI); ndtmsg = NLMSG_DATA(nlh); @@ -1675,8 +1675,8 @@ static int neightbl_fill_param_info(struct neigh_table *tbl, struct ndtmsg *ndtmsg; struct nlmsghdr *nlh; - nlh = NLMSG_PUT_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg)); - nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg), + NLM_F_MULTI); ndtmsg = NLMSG_DATA(nlh); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index e41ce458c2a9..70bcd4744d93 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1095,8 +1095,7 @@ static int netlink_dump(struct sock *sk) return 0; } - nlh = __nlmsg_put(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, NLMSG_DONE, sizeof(int)); - nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW_ANSWER(skb, cb, NLMSG_DONE, sizeof(len), NLM_F_MULTI); memcpy(NLMSG_DATA(nlh), &len, sizeof(len)); skb_queue_tail(&sk->sk_receive_queue, skb); sk->sk_data_ready(sk, skb->len); @@ -1107,6 +1106,9 @@ static int netlink_dump(struct sock *sk) netlink_destroy_callback(cb); return 0; + +nlmsg_failure: + return -ENOBUFS; } int netlink_dump_start(struct sock *ssk, struct sk_buff *skb, @@ -1178,7 +1180,7 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err) } rep = __nlmsg_put(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, - NLMSG_ERROR, sizeof(struct nlmsgerr)); + NLMSG_ERROR, sizeof(struct nlmsgerr), 0); errmsg = NLMSG_DATA(rep); errmsg->error = err; memcpy(&errmsg->msg, nlh, err ? nlh->nlmsg_len : sizeof(struct nlmsghdr)); -- cgit v1.2.3 From 0603eac0d6b77acac5924a2734228cbaf072f993 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sat, 18 Jun 2005 22:54:36 -0700 Subject: [IPSEC]: Add XFRMA_SA/XFRMA_POLICY for delete notification This patch changes the format of the XFRM_MSG_DELSA and XFRM_MSG_DELPOLICY notification so that the main message sent is of the same format as that received by the kernel if the original message was via netlink. This also means that we won't lose the byid information carried in km_event. Since this user interface is introduced by Jamal's patch we can still afford to change it. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/xfrm.h | 2 ++ net/xfrm/xfrm_user.c | 47 +++++++++++++++++++++++++++++++++++++++++++---- 2 files changed, 45 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index 03bc600516ea..d68391a9b9f3 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -174,6 +174,8 @@ enum xfrm_attr_type_t { XFRMA_ALG_COMP, /* struct xfrm_algo */ XFRMA_ENCAP, /* struct xfrm_algo + struct xfrm_encap_tmpl */ XFRMA_TMPL, /* 1 or more struct xfrm_user_tmpl */ + XFRMA_SA, + XFRMA_POLICY, __XFRMA_MAX #define XFRMA_MAX (__XFRMA_MAX - 1) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index ffe1b217347c..5ce8558eac91 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1168,7 +1168,7 @@ nlmsg_failure: static int inline xfrm_sa_len(struct xfrm_state *x) { - int l = NLMSG_LENGTH(sizeof(struct xfrm_usersa_info)); + int l = 0; if (x->aalg) l += RTA_SPACE(sizeof(*x->aalg) + (x->aalg->alg_key_len+7)/8); if (x->ealg) @@ -1184,20 +1184,39 @@ static int inline xfrm_sa_len(struct xfrm_state *x) static int xfrm_notify_sa(struct xfrm_state *x, struct km_event *c) { struct xfrm_usersa_info *p; + struct xfrm_usersa_id *id; struct nlmsghdr *nlh; struct sk_buff *skb; unsigned char *b; int len = xfrm_sa_len(x); + int headlen; + + headlen = sizeof(*p); + if (c->event == XFRM_MSG_DELSA) { + len += RTA_SPACE(headlen); + headlen = sizeof(*id); + } + len += NLMSG_SPACE(headlen); skb = alloc_skb(len, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; b = skb->tail; - nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, sizeof(*p)); + nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, headlen); nlh->nlmsg_flags = 0; p = NLMSG_DATA(nlh); + if (c->event == XFRM_MSG_DELSA) { + id = NLMSG_DATA(nlh); + memcpy(&id->daddr, &x->id.daddr, sizeof(id->daddr)); + id->spi = x->id.spi; + id->family = x->props.family; + id->proto = x->id.proto; + + p = RTA_DATA(__RTA_PUT(skb, XFRMA_SA, sizeof(*p))); + } + copy_to_user_state(x, p); if (x->aalg) @@ -1398,20 +1417,39 @@ static int xfrm_exp_policy_notify(struct xfrm_policy *xp, int dir, struct km_eve static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c) { struct xfrm_userpolicy_info *p; + struct xfrm_userpolicy_id *id; struct nlmsghdr *nlh; struct sk_buff *skb; unsigned char *b; int len = RTA_SPACE(sizeof(struct xfrm_user_tmpl) * xp->xfrm_nr); - len += NLMSG_SPACE(sizeof(struct xfrm_userpolicy_info)); + int headlen; + + headlen = sizeof(*p); + if (c->event == XFRM_MSG_DELPOLICY) { + len += RTA_SPACE(headlen); + headlen = sizeof(*id); + } + len += NLMSG_SPACE(headlen); skb = alloc_skb(len, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; b = skb->tail; - nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, sizeof(*p)); + nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, headlen); p = NLMSG_DATA(nlh); + if (c->event == XFRM_MSG_DELPOLICY) { + id = NLMSG_DATA(nlh); + memset(id, 0, sizeof(*id)); + id->dir = dir; + if (c->data.byid) + id->index = xp->index; + else + memcpy(&id->sel, &xp->selector, sizeof(id->sel)); + + p = RTA_DATA(__RTA_PUT(skb, XFRMA_POLICY, sizeof(*p))); + } nlh->nlmsg_flags = 0; @@ -1424,6 +1462,7 @@ static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event * return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_POLICY, GFP_ATOMIC); nlmsg_failure: +rtattr_failure: kfree_skb(skb); return -1; } -- cgit v1.2.3 From dd87147eed934eaff92869f3d158697c7239d1d2 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 20 Jun 2005 13:21:43 -0700 Subject: [IPSEC]: Add XFRM_STATE_NOPMTUDISC flag This patch adds the flag XFRM_STATE_NOPMTUDISC for xfrm states. It is similar to the nopmtudisc on IPIP/GRE tunnels. It only has an effect on IPv4 tunnel mode states. For these states, it will ensure that the DF flag is always cleared. This is primarily useful to work around ICMP blackholes. In future this flag could also allow a larger MTU to be set within the tunnel just like IPIP/GRE tunnels. This could be useful for short haul tunnels where temporary fragmentation outside the tunnel is desired over smaller fragments inside the tunnel. Signed-off-by: Herbert Xu Acked-by: James Morris Signed-off-by: David S. Miller --- include/linux/pfkeyv2.h | 1 + include/linux/xfrm.h | 1 + net/ipv4/xfrm4_output.c | 8 ++++++-- net/ipv4/xfrm4_state.c | 9 +++++++++ net/key/af_key.c | 4 ++++ 5 files changed, 21 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pfkeyv2.h b/include/linux/pfkeyv2.h index e6b519220245..724066778aff 100644 --- a/include/linux/pfkeyv2.h +++ b/include/linux/pfkeyv2.h @@ -245,6 +245,7 @@ struct sadb_x_nat_t_port { /* Security Association flags */ #define SADB_SAFLAGS_PFS 1 +#define SADB_SAFLAGS_NOPMTUDISC 0x20000000 #define SADB_SAFLAGS_DECAP_DSCP 0x40000000 #define SADB_SAFLAGS_NOECN 0x80000000 diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index d68391a9b9f3..f0d423300d84 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -196,6 +196,7 @@ struct xfrm_usersa_info { __u8 flags; #define XFRM_STATE_NOECN 1 #define XFRM_STATE_DECAP_DSCP 2 +#define XFRM_STATE_NOPMTUDISC 4 }; struct xfrm_usersa_id { diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c index af2392ae5769..66620a95942a 100644 --- a/net/ipv4/xfrm4_output.c +++ b/net/ipv4/xfrm4_output.c @@ -33,6 +33,7 @@ static void xfrm4_encap(struct sk_buff *skb) struct dst_entry *dst = skb->dst; struct xfrm_state *x = dst->xfrm; struct iphdr *iph, *top_iph; + int flags; iph = skb->nh.iph; skb->h.ipiph = iph; @@ -51,10 +52,13 @@ static void xfrm4_encap(struct sk_buff *skb) /* DS disclosed */ top_iph->tos = INET_ECN_encapsulate(iph->tos, iph->tos); - if (x->props.flags & XFRM_STATE_NOECN) + + flags = x->props.flags; + if (flags & XFRM_STATE_NOECN) IP_ECN_clear(top_iph); - top_iph->frag_off = iph->frag_off & htons(IP_DF); + top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ? + 0 : (iph->frag_off & htons(IP_DF)); if (!top_iph->frag_off) __ip_select_ident(top_iph, dst, 0); diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c index 223a2e83853f..050611d7a967 100644 --- a/net/ipv4/xfrm4_state.c +++ b/net/ipv4/xfrm4_state.c @@ -7,12 +7,20 @@ * */ +#include #include #include #include static struct xfrm_state_afinfo xfrm4_state_afinfo; +static int xfrm4_init_flags(struct xfrm_state *x) +{ + if (ipv4_config.no_pmtu_disc) + x->props.flags |= XFRM_STATE_NOPMTUDISC; + return 0; +} + static void __xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl, struct xfrm_tmpl *tmpl, @@ -109,6 +117,7 @@ __xfrm4_find_acq(u8 mode, u32 reqid, u8 proto, static struct xfrm_state_afinfo xfrm4_state_afinfo = { .family = AF_INET, .lock = RW_LOCK_UNLOCKED, + .init_flags = xfrm4_init_flags, .init_tempsel = __xfrm4_init_tempsel, .state_lookup = __xfrm4_state_lookup, .find_acq = __xfrm4_find_acq, diff --git a/net/key/af_key.c b/net/key/af_key.c index 652dd09ccd3a..4879743b945a 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -690,6 +690,8 @@ static struct sk_buff * pfkey_xfrm_state2msg(struct xfrm_state *x, int add_keys, sa->sadb_sa_flags |= SADB_SAFLAGS_NOECN; if (x->props.flags & XFRM_STATE_DECAP_DSCP) sa->sadb_sa_flags |= SADB_SAFLAGS_DECAP_DSCP; + if (x->props.flags & XFRM_STATE_NOPMTUDISC) + sa->sadb_sa_flags |= SADB_SAFLAGS_NOPMTUDISC; /* hard time */ if (hsc & 2) { @@ -974,6 +976,8 @@ static struct xfrm_state * pfkey_msg2xfrm_state(struct sadb_msg *hdr, x->props.flags |= XFRM_STATE_NOECN; if (sa->sadb_sa_flags & SADB_SAFLAGS_DECAP_DSCP) x->props.flags |= XFRM_STATE_DECAP_DSCP; + if (sa->sadb_sa_flags & SADB_SAFLAGS_NOPMTUDISC) + x->props.flags |= XFRM_STATE_NOPMTUDISC; lifetime = (struct sadb_lifetime*) ext_hdrs[SADB_EXT_LIFETIME_HARD-1]; if (lifetime != NULL) { -- cgit v1.2.3 From f6e276ee67c0ac9efafd24bc6f7a84aa359656df Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 20 Jun 2005 13:32:05 -0700 Subject: [ATALK]: endian annotations Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/linux/atalk.h | 26 +++++++++++++------------- net/appletalk/aarp.c | 2 +- net/appletalk/ddp.c | 2 +- 3 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/atalk.h b/include/linux/atalk.h index 31d3fc25ccbd..09a1451c1159 100644 --- a/include/linux/atalk.h +++ b/include/linux/atalk.h @@ -20,7 +20,7 @@ #define SIOCATALKDIFADDR (SIOCPROTOPRIVATE + 0) struct atalk_addr { - __u16 s_net; + __be16 s_net; __u8 s_node; }; @@ -33,8 +33,8 @@ struct sockaddr_at { struct atalk_netrange { __u8 nr_phase; - __u16 nr_firstnet; - __u16 nr_lastnet; + __be16 nr_firstnet; + __be16 nr_lastnet; }; #ifdef __KERNEL__ @@ -70,8 +70,8 @@ struct atalk_iface { struct atalk_sock { /* struct sock has to be the first member of atalk_sock */ struct sock sk; - unsigned short dest_net; - unsigned short src_net; + __be16 dest_net; + __be16 src_net; unsigned char dest_node; unsigned char src_node; unsigned char dest_port; @@ -95,9 +95,9 @@ struct ddpehdr { deh_hops:4, deh_len:10; #endif - __u16 deh_sum; - __u16 deh_dnet; - __u16 deh_snet; + __be16 deh_sum; + __be16 deh_dnet; + __be16 deh_snet; __u8 deh_dnode; __u8 deh_snode; __u8 deh_dport; @@ -142,24 +142,24 @@ struct ddpshdr { /* AppleTalk AARP headers */ struct elapaarp { - __u16 hw_type; + __be16 hw_type; #define AARP_HW_TYPE_ETHERNET 1 #define AARP_HW_TYPE_TOKENRING 2 - __u16 pa_type; + __be16 pa_type; __u8 hw_len; __u8 pa_len; #define AARP_PA_ALEN 4 - __u16 function; + __be16 function; #define AARP_REQUEST 1 #define AARP_REPLY 2 #define AARP_PROBE 3 __u8 hw_src[ETH_ALEN] __attribute__ ((packed)); __u8 pa_src_zero __attribute__ ((packed)); - __u16 pa_src_net __attribute__ ((packed)); + __be16 pa_src_net __attribute__ ((packed)); __u8 pa_src_node __attribute__ ((packed)); __u8 hw_dst[ETH_ALEN] __attribute__ ((packed)); __u8 pa_dst_zero __attribute__ ((packed)); - __u16 pa_dst_net __attribute__ ((packed)); + __be16 pa_dst_net __attribute__ ((packed)); __u8 pa_dst_node __attribute__ ((packed)); }; diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c index 54640c01b50c..10d040461021 100644 --- a/net/appletalk/aarp.c +++ b/net/appletalk/aarp.c @@ -565,7 +565,7 @@ int aarp_send_ddp(struct net_device *dev, struct sk_buff *skb, * numbers we just happen to need. Now put the * length in the lower two. */ - *((__u16 *)skb->data) = htons(skb->len); + *((__be16 *)skb->data) = htons(skb->len); ft = 1; } /* diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 876dbac71060..192b529f86a4 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -401,7 +401,7 @@ out_err: } /* Find a match for a specific network:node pair */ -static struct atalk_iface *atalk_find_interface(int net, int node) +static struct atalk_iface *atalk_find_interface(__be16 net, int node) { struct atalk_iface *iface; -- cgit v1.2.3 From 246955fe4c38bd706ae30e37c64892c94213775d Mon Sep 17 00:00:00 2001 From: Robert Olsson Date: Mon, 20 Jun 2005 13:36:39 -0700 Subject: [NETLINK]: fib_lookup() via netlink Below is a more generic patch to do fib_lookup via netlink. For others we should say that we discussed this as a way to verify route selection. It's also possible there are others uses for this. In short the fist half of struct fib_result_nl is filled in by caller and netlink call fills in the other half and returns it. In case anyone is interested there is a corresponding user app to compare the full routing table this was used to test implementation of the LC-trie. Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 + include/net/ip_fib.h | 14 +++++++++++++ net/ipv4/fib_frontend.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 70 insertions(+) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index e38407a23d04..561d4dc75836 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -14,6 +14,7 @@ #define NETLINK_SELINUX 7 /* SELinux event notifications */ #define NETLINK_ARPD 8 #define NETLINK_AUDIT 9 /* auditing */ +#define NETLINK_FIB_LOOKUP 10 #define NETLINK_ROUTE6 11 /* af_inet6 route comm channel */ #define NETLINK_IP6_FW 13 #define NETLINK_DNRTMSG 14 /* DECnet routing messages */ diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index e5a5f6b62f88..a4208a336ac0 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -109,6 +109,20 @@ struct fib_result { #endif }; +struct fib_result_nl { + u32 fl_addr; /* To be looked up*/ + u32 fl_fwmark; + unsigned char fl_tos; + unsigned char fl_scope; + unsigned char tb_id_in; + + unsigned char tb_id; /* Results */ + unsigned char prefixlen; + unsigned char nh_sel; + unsigned char type; + unsigned char scope; + int err; +}; #ifdef CONFIG_IP_ROUTE_MULTIPATH diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 563e7d612706..cd8e45ab9580 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -516,6 +516,60 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa) #undef BRD1_OK } +static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb ) +{ + + struct fib_result res; + struct flowi fl = { .nl_u = { .ip4_u = { .daddr = frn->fl_addr, + .fwmark = frn->fl_fwmark, + .tos = frn->fl_tos, + .scope = frn->fl_scope } } }; + if (tb) { + local_bh_disable(); + + frn->tb_id = tb->tb_id; + frn->err = tb->tb_lookup(tb, &fl, &res); + + if (!frn->err) { + frn->prefixlen = res.prefixlen; + frn->nh_sel = res.nh_sel; + frn->type = res.type; + frn->scope = res.scope; + } + local_bh_enable(); + } +} + +static void nl_fib_input(struct sock *sk, int len) +{ + struct sk_buff *skb = NULL; + struct nlmsghdr *nlh = NULL; + struct fib_result_nl *frn; + int err; + u32 pid; + struct fib_table *tb; + + skb = skb_recv_datagram(sk, 0, 0, &err); + nlh = (struct nlmsghdr *)skb->data; + + frn = (struct fib_result_nl *) NLMSG_DATA(nlh); + tb = fib_get_table(frn->tb_id_in); + + nl_fib_lookup(frn, tb); + + pid = nlh->nlmsg_pid; /*pid of sending process */ + NETLINK_CB(skb).groups = 0; /* not in mcast group */ + NETLINK_CB(skb).pid = 0; /* from kernel */ + NETLINK_CB(skb).dst_pid = pid; + NETLINK_CB(skb).dst_groups = 0; /* unicast */ + netlink_unicast(sk, skb, pid, MSG_DONTWAIT); +} + +static void nl_fib_lookup_init(void) +{ + netlink_kernel_create(NETLINK_FIB_LOOKUP, nl_fib_input); +} + static void fib_disable_ip(struct net_device *dev, int force) { if (fib_sync_down(0, dev, force)) @@ -604,6 +658,7 @@ void __init ip_fib_init(void) register_netdevice_notifier(&fib_netdev_notifier); register_inetaddr_notifier(&fib_inetaddr_notifier); + nl_fib_lookup_init(); } EXPORT_SYMBOL(inet_addr_type); -- cgit v1.2.3 From e3a15db2415579d5136b9ba9b52fe27c66da8780 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 26 Apr 2005 02:31:08 -0500 Subject: [PATCH] sysfs_{create|remove}_link should take const char * sysfs: make sysfs_{create|remove}_link to take const char * name. Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- fs/sysfs/symlink.c | 8 ++++---- include/linux/sysfs.h | 10 +++++----- 2 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c index dfdf70174354..fae57c83a722 100644 --- a/fs/sysfs/symlink.c +++ b/fs/sysfs/symlink.c @@ -43,7 +43,7 @@ static void fill_object_path(struct kobject * kobj, char * buffer, int length) } } -static int sysfs_add_link(struct dentry * parent, char * name, struct kobject * target) +static int sysfs_add_link(struct dentry * parent, const char * name, struct kobject * target) { struct sysfs_dirent * parent_sd = parent->d_fsdata; struct sysfs_symlink * sl; @@ -79,7 +79,7 @@ exit1: * @target: object we're pointing to. * @name: name of the symlink. */ -int sysfs_create_link(struct kobject * kobj, struct kobject * target, char * name) +int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name) { struct dentry * dentry = kobj->dentry; int error = 0; @@ -99,13 +99,13 @@ int sysfs_create_link(struct kobject * kobj, struct kobject * target, char * nam * @name: name of the symlink to remove. */ -void sysfs_remove_link(struct kobject * kobj, char * name) +void sysfs_remove_link(struct kobject * kobj, const char * name) { sysfs_hash_and_remove(kobj->dentry,name); } static int sysfs_get_target_path(struct kobject * kobj, struct kobject * target, - char *path) + char *path) { char * s; int depth, size; diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index 38b58b30814a..931b5aaaf380 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -105,11 +105,11 @@ sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode); extern void sysfs_remove_file(struct kobject *, const struct attribute *); -extern int -sysfs_create_link(struct kobject * kobj, struct kobject * target, char * name); +extern int +sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name); extern void -sysfs_remove_link(struct kobject *, char * name); +sysfs_remove_link(struct kobject *, const char * name); int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr); int sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr); @@ -153,12 +153,12 @@ static inline void sysfs_remove_file(struct kobject * k, const struct attribute ; } -static inline int sysfs_create_link(struct kobject * k, struct kobject * t, char * n) +static inline int sysfs_create_link(struct kobject * k, struct kobject * t, const char * n) { return 0; } -static inline void sysfs_remove_link(struct kobject * k, char * name) +static inline void sysfs_remove_link(struct kobject * k, const char * name) { ; } -- cgit v1.2.3 From f3b4f3c6dec04c6c8261fe22645f07b39976595a Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 26 Apr 2005 02:32:00 -0500 Subject: [PATCH] Make kobject's name be const char * kobject: make kobject's name const char * since users should not attempt to change it (except by calling kobject_rename). Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- include/linux/kobject.h | 6 +++--- lib/kobject.c | 2 +- lib/kobject_uevent.c | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kobject.h b/include/linux/kobject.h index 765d660d3bea..76dc67245c0c 100644 --- a/include/linux/kobject.h +++ b/include/linux/kobject.h @@ -33,7 +33,7 @@ extern u64 hotplug_seqnum; struct kobject { - char * k_name; + const char * k_name; char name[KOBJ_NAME_LEN]; struct kref kref; struct list_head entry; @@ -46,7 +46,7 @@ struct kobject { extern int kobject_set_name(struct kobject *, const char *, ...) __attribute__((format(printf,2,3))); -static inline char * kobject_name(struct kobject * kobj) +static inline const char * kobject_name(const struct kobject * kobj) { return kobj->k_name; } @@ -57,7 +57,7 @@ extern void kobject_cleanup(struct kobject *); extern int kobject_add(struct kobject *); extern void kobject_del(struct kobject *); -extern int kobject_rename(struct kobject *, char *new_name); +extern int kobject_rename(struct kobject *, const char *new_name); extern int kobject_register(struct kobject *); extern void kobject_unregister(struct kobject *); diff --git a/lib/kobject.c b/lib/kobject.c index 94048826624c..dd0917dd9fa9 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -279,7 +279,7 @@ EXPORT_SYMBOL(kobject_set_name); * @new_name: object's new name */ -int kobject_rename(struct kobject * kobj, char *new_name) +int kobject_rename(struct kobject * kobj, const char *new_name) { int error = 0; diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c index 1ede5aa33376..8e49d21057e4 100644 --- a/lib/kobject_uevent.c +++ b/lib/kobject_uevent.c @@ -197,7 +197,7 @@ void kobject_hotplug(struct kobject *kobj, enum kobject_action action) int i = 0; int retval; char *kobj_path = NULL; - char *name = NULL; + const char *name = NULL; char *action_string; u64 seq; struct kobject *top_kobj = kobj; @@ -249,7 +249,7 @@ void kobject_hotplug(struct kobject *kobj, enum kobject_action action) name = kobject_name(&kset->kobj); argv [0] = hotplug_path; - argv [1] = name; + argv [1] = (char *)name; /* won't be changed but 'const' has to go */ argv [2] = NULL; /* minimal command environment */ -- cgit v1.2.3 From 419cab3fc69588ebe35b845cc3a584ae172463de Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 26 Apr 2005 02:32:54 -0500 Subject: [PATCH] kset_hotplug_ops->name shoudl return const char * kobject: change name() method in kset_hotplug_ops return const char * since users shoudl not try to modify returned data. Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/base/class.c | 2 +- drivers/base/core.c | 2 +- include/linux/kobject.h | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/class.c b/drivers/base/class.c index d2a2f8f2b4ed..ff33ebbf1dc7 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -262,7 +262,7 @@ static int class_hotplug_filter(struct kset *kset, struct kobject *kobj) return 0; } -static char *class_hotplug_name(struct kset *kset, struct kobject *kobj) +static const char *class_hotplug_name(struct kset *kset, struct kobject *kobj) { struct class_device *class_dev = to_class_dev(kobj); diff --git a/drivers/base/core.c b/drivers/base/core.c index fbc223486f81..32f3f09a42e0 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -102,7 +102,7 @@ static int dev_hotplug_filter(struct kset *kset, struct kobject *kobj) return 0; } -static char *dev_hotplug_name(struct kset *kset, struct kobject *kobj) +static const char *dev_hotplug_name(struct kset *kset, struct kobject *kobj) { struct device *dev = to_dev(kobj); diff --git a/include/linux/kobject.h b/include/linux/kobject.h index 76dc67245c0c..3b22304f12fd 100644 --- a/include/linux/kobject.h +++ b/include/linux/kobject.h @@ -94,7 +94,7 @@ struct kobj_type { */ struct kset_hotplug_ops { int (*filter)(struct kset *kset, struct kobject *kobj); - char *(*name)(struct kset *kset, struct kobject *kobj); + const char *(*name)(struct kset *kset, struct kobject *kobj); int (*hotplug)(struct kset *kset, struct kobject *kobj, char **envp, int num_envp, char *buffer, int buffer_size); }; -- cgit v1.2.3 From 8d790d74085833ba2a3e84b5bcd683be4981c29a Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 26 Apr 2005 02:34:05 -0500 Subject: [PATCH] make driver's name be const char * Driver core: change driver's, bus's, class's and platform device's names to be const char * so one can use const char *drv_name = "asdfg"; when initializing structures. Also kill couple of whitespaces. Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/devices.c | 2 +- include/linux/device.h | 12 ++++++------ 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/devices.c b/drivers/usb/core/devices.c index ef0b35731ff0..83e815d3cd52 100644 --- a/drivers/usb/core/devices.c +++ b/drivers/usb/core/devices.c @@ -239,7 +239,7 @@ static char *usb_dump_interface_descriptor(char *start, char *end, int setno) { const struct usb_interface_descriptor *desc = &intfc->altsetting[setno].desc; - char *driver_name = ""; + const char *driver_name = ""; if (start > end) return start; diff --git a/include/linux/device.h b/include/linux/device.h index df94c0de53f2..fa9e6ca08f5a 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -47,7 +47,7 @@ struct class_device; struct class_simple; struct bus_type { - char * name; + const char * name; struct subsystem subsys; struct kset drivers; @@ -98,17 +98,17 @@ extern int bus_create_file(struct bus_type *, struct bus_attribute *); extern void bus_remove_file(struct bus_type *, struct bus_attribute *); struct device_driver { - char * name; + const char * name; struct bus_type * bus; struct completion unloaded; struct kobject kobj; struct list_head devices; - struct module * owner; + struct module * owner; int (*probe) (struct device * dev); - int (*remove) (struct device * dev); + int (*remove) (struct device * dev); void (*shutdown) (struct device * dev); int (*suspend) (struct device * dev, pm_message_t state, u32 level); int (*resume) (struct device * dev, u32 level); @@ -142,7 +142,7 @@ extern void driver_remove_file(struct device_driver *, struct driver_attribute * * device classes */ struct class { - char * name; + const char * name; struct subsystem subsys; struct list_head children; @@ -366,7 +366,7 @@ extern struct device *device_find(const char *name, struct bus_type *bus); /* drivers/base/platform.c */ struct platform_device { - char * name; + const char * name; u32 id; struct device dev; u32 num_resources; -- cgit v1.2.3 From d48593bf208e0d046c35fb0707ae5b23fef8c4ff Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Fri, 29 Apr 2005 00:58:46 -0500 Subject: [PATCH] Make attributes names const char * sysfs: make attributes and attribute_group's names const char * Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/infiniband/core/sysfs.c | 120 ++++++++++++++++------------------------ drivers/pci/pci-sysfs.c | 13 +++-- include/linux/sysfs.h | 4 +- 3 files changed, 58 insertions(+), 79 deletions(-) (limited to 'include/linux') diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c index 3a413f72ff6d..5febd6d8b885 100644 --- a/drivers/infiniband/core/sysfs.c +++ b/drivers/infiniband/core/sysfs.c @@ -40,9 +40,7 @@ struct ib_port { struct kobject kobj; struct ib_device *ibdev; struct attribute_group gid_group; - struct attribute **gid_attr; struct attribute_group pkey_group; - struct attribute **pkey_attr; u8 port_num; }; @@ -60,8 +58,9 @@ struct port_attribute port_attr_##_name = __ATTR(_name, _mode, _show, _store) struct port_attribute port_attr_##_name = __ATTR_RO(_name) struct port_table_attribute { - struct port_attribute attr; - int index; + struct port_attribute attr; + char name[8]; + int index; }; static ssize_t port_attr_show(struct kobject *kobj, @@ -398,17 +397,16 @@ static void ib_port_release(struct kobject *kobj) struct attribute *a; int i; - for (i = 0; (a = p->gid_attr[i]); ++i) { - kfree(a->name); + for (i = 0; (a = p->gid_group.attrs[i]); ++i) kfree(a); - } - for (i = 0; (a = p->pkey_attr[i]); ++i) { - kfree(a->name); + kfree(p->gid_group.attrs); + + for (i = 0; (a = p->pkey_group.attrs[i]); ++i) kfree(a); - } - kfree(p->gid_attr); + kfree(p->pkey_group.attrs); + kfree(p); } @@ -449,58 +447,45 @@ static int ib_device_hotplug(struct class_device *cdev, char **envp, return 0; } -static int alloc_group(struct attribute ***attr, - ssize_t (*show)(struct ib_port *, - struct port_attribute *, char *buf), - int len) +static struct attribute ** +alloc_group_attrs(ssize_t (*show)(struct ib_port *, + struct port_attribute *, char *buf), + int len) { - struct port_table_attribute ***tab_attr = - (struct port_table_attribute ***) attr; + struct attribute **tab_attr; + struct port_table_attribute *element; int i; - int ret; - - *tab_attr = kmalloc((1 + len) * sizeof *tab_attr, GFP_KERNEL); - if (!*tab_attr) - return -ENOMEM; - memset(*tab_attr, 0, (1 + len) * sizeof *tab_attr); + tab_attr = kcalloc(1 + len, sizeof(struct attribute *), GFP_KERNEL); + if (!tab_attr) + return NULL; - for (i = 0; i < len; ++i) { - (*tab_attr)[i] = kmalloc(sizeof *(*tab_attr)[i], GFP_KERNEL); - if (!(*tab_attr)[i]) { - ret = -ENOMEM; + for (i = 0; i < len; i++) { + element = kcalloc(1, sizeof(struct port_table_attribute), + GFP_KERNEL); + if (!element) goto err; - } - memset((*tab_attr)[i], 0, sizeof *(*tab_attr)[i]); - (*tab_attr)[i]->attr.attr.name = kmalloc(8, GFP_KERNEL); - if (!(*tab_attr)[i]->attr.attr.name) { - ret = -ENOMEM; - goto err; - } - if (snprintf((*tab_attr)[i]->attr.attr.name, 8, "%d", i) >= 8) { - ret = -ENOMEM; + if (snprintf(element->name, sizeof(element->name), + "%d", i) >= sizeof(element->name)) goto err; - } - (*tab_attr)[i]->attr.attr.mode = S_IRUGO; - (*tab_attr)[i]->attr.attr.owner = THIS_MODULE; - (*tab_attr)[i]->attr.show = show; - (*tab_attr)[i]->index = i; - } - - return 0; + element->attr.attr.name = element->name; + element->attr.attr.mode = S_IRUGO; + element->attr.attr.owner = THIS_MODULE; + element->attr.show = show; + element->index = i; -err: - for (i = 0; i < len; ++i) { - if ((*tab_attr)[i]) - kfree((*tab_attr)[i]->attr.attr.name); - kfree((*tab_attr)[i]); + tab_attr[i] = &element->attr.attr; } - kfree(*tab_attr); + return tab_attr; - return ret; +err: + while (--i >= 0) + kfree(tab_attr[i]); + kfree(tab_attr); + return NULL; } static int add_port(struct ib_device *device, int port_num) @@ -541,23 +526,20 @@ static int add_port(struct ib_device *device, int port_num) if (ret) goto err_put; - ret = alloc_group(&p->gid_attr, show_port_gid, attr.gid_tbl_len); - if (ret) - goto err_remove_pma; - p->gid_group.name = "gids"; - p->gid_group.attrs = p->gid_attr; + p->gid_group.attrs = alloc_group_attrs(show_port_gid, attr.gid_tbl_len); + if (!p->gid_group.attrs) + goto err_remove_pma; ret = sysfs_create_group(&p->kobj, &p->gid_group); if (ret) goto err_free_gid; - ret = alloc_group(&p->pkey_attr, show_port_pkey, attr.pkey_tbl_len); - if (ret) - goto err_remove_gid; - p->pkey_group.name = "pkeys"; - p->pkey_group.attrs = p->pkey_attr; + p->pkey_group.attrs = alloc_group_attrs(show_port_pkey, + attr.pkey_tbl_len); + if (!p->pkey_group.attrs) + goto err_remove_gid; ret = sysfs_create_group(&p->kobj, &p->pkey_group); if (ret) @@ -568,23 +550,19 @@ static int add_port(struct ib_device *device, int port_num) return 0; err_free_pkey: - for (i = 0; i < attr.pkey_tbl_len; ++i) { - kfree(p->pkey_attr[i]->name); - kfree(p->pkey_attr[i]); - } + for (i = 0; i < attr.pkey_tbl_len; ++i) + kfree(p->pkey_group.attrs[i]); - kfree(p->pkey_attr); + kfree(p->pkey_group.attrs); err_remove_gid: sysfs_remove_group(&p->kobj, &p->gid_group); err_free_gid: - for (i = 0; i < attr.gid_tbl_len; ++i) { - kfree(p->gid_attr[i]->name); - kfree(p->gid_attr[i]); - } + for (i = 0; i < attr.gid_tbl_len; ++i) + kfree(p->gid_group.attrs[i]); - kfree(p->gid_attr); + kfree(p->gid_group.attrs); err_remove_pma: sysfs_remove_group(&p->kobj, &pma_group); diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 6ca0061137a6..e8aad151175f 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -339,16 +339,17 @@ pci_create_resource_files(struct pci_dev *pdev) if (!pci_resource_len(pdev, i)) continue; - res_attr = kmalloc(sizeof(*res_attr) + 10, GFP_ATOMIC); + /* allocate attribute structure, piggyback attribute name */ + res_attr = kcalloc(1, sizeof(*res_attr) + 10, GFP_ATOMIC); if (res_attr) { - memset(res_attr, 0, sizeof(*res_attr) + 10); + char *res_attr_name = (char *)(res_attr + 1); + pdev->res_attr[i] = res_attr; - /* Allocated above after the res_attr struct */ - res_attr->attr.name = (char *)(res_attr + 1); - sprintf(res_attr->attr.name, "resource%d", i); - res_attr->size = pci_resource_len(pdev, i); + sprintf(res_attr_name, "resource%d", i); + res_attr->attr.name = res_attr_name; res_attr->attr.mode = S_IRUSR | S_IWUSR; res_attr->attr.owner = THIS_MODULE; + res_attr->size = pci_resource_len(pdev, i); res_attr->mmap = pci_mmap_resource; res_attr->private = &pdev->resource[i]; sysfs_create_bin_file(&pdev->dev.kobj, res_attr); diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index 931b5aaaf380..d9cd2d31d377 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -16,13 +16,13 @@ struct kobject; struct module; struct attribute { - char * name; + const char * name; struct module * owner; mode_t mode; }; struct attribute_group { - char * name; + const char * name; struct attribute ** attrs; }; -- cgit v1.2.3 From e9ba6365fd4f0d9e7d022c883bd044fbaa48257f Mon Sep 17 00:00:00 2001 From: "gregkh@suse.de" Date: Tue, 15 Mar 2005 11:54:21 -0800 Subject: [PATCH] CLASS: move a "simple" class logic into the class core. One step on improving the class api so that it can not be used incorrectly. This also fixes the module owner issue with the dev files that happened when the devt logic moved to the class core. Based on a patch originally written by Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/class.c | 145 +++++++++++++++++++++++++++++++++++++++++++++---- include/linux/device.h | 9 +++ 2 files changed, 143 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/class.c b/drivers/base/class.c index 344b8cd73901..a3b006b6f2ba 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -16,6 +16,7 @@ #include #include #include +#include #include "base.h" #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr) @@ -162,6 +163,51 @@ void class_unregister(struct class * cls) subsystem_unregister(&cls->subsys); } +static void class_create_release(struct class *cls) +{ + kfree(cls); +} + +static void class_device_create_release(struct class_device *class_dev) +{ + kfree(class_dev); +} + +struct class *class_create(struct module *owner, char *name) +{ + struct class *cls; + int retval; + + cls = kmalloc(sizeof(struct class), GFP_KERNEL); + if (!cls) { + retval = -ENOMEM; + goto error; + } + memset(cls, 0x00, sizeof(struct class)); + + cls->name = name; + cls->owner = owner; + cls->class_release = class_create_release; + cls->release = class_device_create_release; + + retval = class_register(cls); + if (retval) + goto error; + + return cls; + +error: + kfree(cls); + return ERR_PTR(retval); +} + +void class_destroy(struct class *cls) +{ + if ((cls == NULL) || (IS_ERR(cls))) + return; + + class_unregister(cls); +} /* Class Device Stuff */ @@ -375,7 +421,6 @@ static ssize_t show_dev(struct class_device *class_dev, char *buf) { return print_dev_t(buf, class_dev->devt); } -static CLASS_DEVICE_ATTR(dev, S_IRUGO, show_dev, NULL); void class_device_initialize(struct class_device *class_dev) { @@ -412,7 +457,31 @@ int class_device_add(struct class_device *class_dev) if ((error = kobject_add(&class_dev->kobj))) goto register_done; - /* now take care of our own registration */ + /* add the needed attributes to this device */ + if (MAJOR(class_dev->devt)) { + struct class_device_attribute *attr; + attr = kmalloc(sizeof(*attr), GFP_KERNEL); + if (!attr) { + error = -ENOMEM; + kobject_del(&class_dev->kobj); + goto register_done; + } + memset(attr, sizeof(*attr), 0x00); + attr->attr.name = "dev"; + attr->attr.mode = S_IRUGO; + attr->attr.owner = parent->owner; + attr->show = show_dev; + attr->store = NULL; + class_device_create_file(class_dev, attr); + class_dev->devt_attr = attr; + } + + class_device_add_attrs(class_dev); + if (class_dev->dev) + sysfs_create_link(&class_dev->kobj, + &class_dev->dev->kobj, "device"); + + /* notify any interfaces this device is now here */ if (parent) { down(&parent->sem); list_add_tail(&class_dev->node, &parent->children); @@ -421,16 +490,8 @@ int class_device_add(struct class_device *class_dev) class_intf->add(class_dev); up(&parent->sem); } - - if (MAJOR(class_dev->devt)) - class_device_create_file(class_dev, &class_device_attr_dev); - - class_device_add_attrs(class_dev); - if (class_dev->dev) - sysfs_create_link(&class_dev->kobj, - &class_dev->dev->kobj, "device"); - kobject_hotplug(&class_dev->kobj, KOBJ_ADD); + register_done: if (error && parent) class_put(parent); @@ -444,6 +505,41 @@ int class_device_register(struct class_device *class_dev) return class_device_add(class_dev); } +struct class_device *class_device_create(struct class *cls, dev_t devt, + struct device *device, char *fmt, ...) +{ + va_list args; + struct class_device *class_dev = NULL; + int retval = -ENODEV; + + if (cls == NULL || IS_ERR(cls)) + goto error; + + class_dev = kmalloc(sizeof(struct class_device), GFP_KERNEL); + if (!class_dev) { + retval = -ENOMEM; + goto error; + } + memset(class_dev, 0x00, sizeof(struct class_device)); + + class_dev->devt = devt; + class_dev->dev = device; + class_dev->class = cls; + + va_start(args, fmt); + vsnprintf(class_dev->class_id, BUS_ID_SIZE, fmt, args); + va_end(args); + retval = class_device_register(class_dev); + if (retval) + goto error; + + return class_dev; + +error: + kfree(class_dev); + return ERR_PTR(retval); +} + void class_device_del(struct class_device *class_dev) { struct class * parent = class_dev->class; @@ -460,6 +556,11 @@ void class_device_del(struct class_device *class_dev) if (class_dev->dev) sysfs_remove_link(&class_dev->kobj, "device"); + if (class_dev->devt_attr) { + class_device_remove_file(class_dev, class_dev->devt_attr); + kfree(class_dev->devt_attr); + class_dev->devt_attr = NULL; + } class_device_remove_attrs(class_dev); kobject_hotplug(&class_dev->kobj, KOBJ_REMOVE); @@ -477,6 +578,24 @@ void class_device_unregister(struct class_device *class_dev) class_device_put(class_dev); } +void class_device_destroy(struct class *cls, dev_t devt) +{ + struct class_device *class_dev = NULL; + struct class_device *class_dev_tmp; + + down(&cls->sem); + list_for_each_entry(class_dev_tmp, &cls->children, node) { + if (class_dev_tmp->devt == devt) { + class_dev = class_dev_tmp; + break; + } + } + up(&cls->sem); + + if (class_dev) + class_device_unregister(class_dev); +} + int class_device_rename(struct class_device *class_dev, char *new_name) { int error = 0; @@ -576,6 +695,8 @@ EXPORT_SYMBOL_GPL(class_register); EXPORT_SYMBOL_GPL(class_unregister); EXPORT_SYMBOL_GPL(class_get); EXPORT_SYMBOL_GPL(class_put); +EXPORT_SYMBOL_GPL(class_create); +EXPORT_SYMBOL_GPL(class_destroy); EXPORT_SYMBOL_GPL(class_device_register); EXPORT_SYMBOL_GPL(class_device_unregister); @@ -584,6 +705,8 @@ EXPORT_SYMBOL_GPL(class_device_add); EXPORT_SYMBOL_GPL(class_device_del); EXPORT_SYMBOL_GPL(class_device_get); EXPORT_SYMBOL_GPL(class_device_put); +EXPORT_SYMBOL_GPL(class_device_create); +EXPORT_SYMBOL_GPL(class_device_destroy); EXPORT_SYMBOL_GPL(class_device_create_file); EXPORT_SYMBOL_GPL(class_device_remove_file); EXPORT_SYMBOL_GPL(class_device_create_bin_file); diff --git a/include/linux/device.h b/include/linux/device.h index fa9e6ca08f5a..73250d01c01f 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -143,6 +143,7 @@ extern void driver_remove_file(struct device_driver *, struct driver_attribute * */ struct class { const char * name; + struct module * owner; struct subsystem subsys; struct list_head children; @@ -185,6 +186,7 @@ struct class_device { struct kobject kobj; struct class * class; /* required */ dev_t devt; /* dev_t, creates the sysfs "dev" */ + struct class_device_attribute *devt_attr; struct device * dev; /* not necessary, but nice to have */ void * class_data; /* class-specific data */ @@ -245,6 +247,13 @@ struct class_interface { extern int class_interface_register(struct class_interface *); extern void class_interface_unregister(struct class_interface *); +extern struct class *class_create(struct module *owner, char *name); +extern void class_destroy(struct class *cls); +extern struct class_device *class_device_create(struct class *cls, dev_t devt, + struct device *device, char *fmt, ...) + __attribute__((format(printf,4,5))); +extern void class_device_destroy(struct class *cls, dev_t devt); + /* interface for class simple stuff */ extern struct class_simple *class_simple_create(struct module *owner, char *name); extern void class_simple_destroy(struct class_simple *cs); -- cgit v1.2.3 From 1235686f6e67cf30c460eb77d90a6cb4be57b92f Mon Sep 17 00:00:00 2001 From: "gregkh@suse.de" Date: Tue, 15 Mar 2005 14:26:30 -0800 Subject: [PATCH] INPUT: move to use the new class code, instead of class_simple Signed-off-by: Greg Kroah-Hartman --- drivers/input/evdev.c | 9 +++++---- drivers/input/input.c | 10 +++++----- drivers/input/joydev.c | 8 ++++---- drivers/input/mousedev.c | 16 +++++++++------- drivers/input/tsdev.c | 9 +++++---- include/linux/input.h | 2 +- 6 files changed, 29 insertions(+), 25 deletions(-) (limited to 'include/linux') diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c index 17552a29978b..556264b43425 100644 --- a/drivers/input/evdev.c +++ b/drivers/input/evdev.c @@ -431,9 +431,9 @@ static struct input_handle *evdev_connect(struct input_handler *handler, struct devfs_mk_cdev(MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + minor), S_IFCHR|S_IRUGO|S_IWUSR, "input/event%d", minor); - class_simple_device_add(input_class, - MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + minor), - dev->dev, "event%d", minor); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + minor), + dev->dev, "event%d", minor); return &evdev->handle; } @@ -443,7 +443,8 @@ static void evdev_disconnect(struct input_handle *handle) struct evdev *evdev = handle->private; struct evdev_list *list; - class_simple_device_remove(MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + evdev->minor)); + class_device_destroy(input_class, + MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + evdev->minor)); devfs_remove("input/event%d", evdev->minor); evdev->exist = 0; diff --git a/drivers/input/input.c b/drivers/input/input.c index 3385dd03abfc..83c77c990dda 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -702,13 +702,13 @@ static int __init input_proc_init(void) static inline int input_proc_init(void) { return 0; } #endif -struct class_simple *input_class; +struct class *input_class; static int __init input_init(void) { int retval = -ENOMEM; - input_class = class_simple_create(THIS_MODULE, "input"); + input_class = class_create(THIS_MODULE, "input"); if (IS_ERR(input_class)) return PTR_ERR(input_class); input_proc_init(); @@ -718,7 +718,7 @@ static int __init input_init(void) remove_proc_entry("devices", proc_bus_input_dir); remove_proc_entry("handlers", proc_bus_input_dir); remove_proc_entry("input", proc_bus); - class_simple_destroy(input_class); + class_destroy(input_class); return retval; } @@ -728,7 +728,7 @@ static int __init input_init(void) remove_proc_entry("handlers", proc_bus_input_dir); remove_proc_entry("input", proc_bus); unregister_chrdev(INPUT_MAJOR, "input"); - class_simple_destroy(input_class); + class_destroy(input_class); } return retval; } @@ -741,7 +741,7 @@ static void __exit input_exit(void) devfs_remove("input"); unregister_chrdev(INPUT_MAJOR, "input"); - class_simple_destroy(input_class); + class_destroy(input_class); } subsys_initcall(input_init); diff --git a/drivers/input/joydev.c b/drivers/input/joydev.c index 627d343dfba1..39775fc380c7 100644 --- a/drivers/input/joydev.c +++ b/drivers/input/joydev.c @@ -452,9 +452,9 @@ static struct input_handle *joydev_connect(struct input_handler *handler, struct devfs_mk_cdev(MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + minor), S_IFCHR|S_IRUGO|S_IWUSR, "input/js%d", minor); - class_simple_device_add(input_class, - MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + minor), - dev->dev, "js%d", minor); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + minor), + dev->dev, "js%d", minor); return &joydev->handle; } @@ -464,7 +464,7 @@ static void joydev_disconnect(struct input_handle *handle) struct joydev *joydev = handle->private; struct joydev_list *list; - class_simple_device_remove(MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + joydev->minor)); + class_device_destroy(input_class, MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + joydev->minor)); devfs_remove("input/js%d", joydev->minor); joydev->exist = 0; diff --git a/drivers/input/mousedev.c b/drivers/input/mousedev.c index 96fb9870834a..062848ac7e6b 100644 --- a/drivers/input/mousedev.c +++ b/drivers/input/mousedev.c @@ -647,9 +647,9 @@ static struct input_handle *mousedev_connect(struct input_handler *handler, stru devfs_mk_cdev(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + minor), S_IFCHR|S_IRUGO|S_IWUSR, "input/mouse%d", minor); - class_simple_device_add(input_class, - MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + minor), - dev->dev, "mouse%d", minor); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + minor), + dev->dev, "mouse%d", minor); return &mousedev->handle; } @@ -659,7 +659,8 @@ static void mousedev_disconnect(struct input_handle *handle) struct mousedev *mousedev = handle->private; struct mousedev_list *list; - class_simple_device_remove(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + mousedev->minor)); + class_device_destroy(input_class, + MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + mousedev->minor)); devfs_remove("input/mouse%d", mousedev->minor); mousedev->exist = 0; @@ -735,8 +736,8 @@ static int __init mousedev_init(void) devfs_mk_cdev(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX), S_IFCHR|S_IRUGO|S_IWUSR, "input/mice"); - class_simple_device_add(input_class, MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX), - NULL, "mice"); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX), NULL, "mice"); #ifdef CONFIG_INPUT_MOUSEDEV_PSAUX if (!(psaux_registered = !misc_register(&psaux_mouse))) @@ -755,7 +756,8 @@ static void __exit mousedev_exit(void) misc_deregister(&psaux_mouse); #endif devfs_remove("input/mice"); - class_simple_device_remove(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX)); + class_device_destroy(input_class, + MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX)); input_unregister_handler(&mousedev_handler); } diff --git a/drivers/input/tsdev.c b/drivers/input/tsdev.c index d0afba85720b..50c63a155156 100644 --- a/drivers/input/tsdev.c +++ b/drivers/input/tsdev.c @@ -414,9 +414,9 @@ static struct input_handle *tsdev_connect(struct input_handler *handler, S_IFCHR|S_IRUGO|S_IWUSR, "input/ts%d", minor); devfs_mk_cdev(MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + minor + TSDEV_MINORS/2), S_IFCHR|S_IRUGO|S_IWUSR, "input/tsraw%d", minor); - class_simple_device_add(input_class, - MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + minor), - dev->dev, "ts%d", minor); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + minor), + dev->dev, "ts%d", minor); return &tsdev->handle; } @@ -426,7 +426,8 @@ static void tsdev_disconnect(struct input_handle *handle) struct tsdev *tsdev = handle->private; struct tsdev_list *list; - class_simple_device_remove(MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + tsdev->minor)); + class_device_destroy(input_class, + MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + tsdev->minor)); devfs_remove("input/ts%d", tsdev->minor); devfs_remove("input/tsraw%d", tsdev->minor); tsdev->exist = 0; diff --git a/include/linux/input.h b/include/linux/input.h index 72731d7d189e..9d9598ed760d 100644 --- a/include/linux/input.h +++ b/include/linux/input.h @@ -1015,7 +1015,7 @@ static inline void input_set_abs_params(struct input_dev *dev, int axis, int min dev->absbit[LONG(axis)] |= BIT(axis); } -extern struct class_simple *input_class; +extern struct class *input_class; #endif #endif -- cgit v1.2.3 From 8561b10f6e7ef0a085709ffc844f74130a067abe Mon Sep 17 00:00:00 2001 From: "gregkh@suse.de" Date: Tue, 15 Mar 2005 15:10:13 -0800 Subject: [PATCH] USB: move the usb hcd code to use the new class code. This moves a kref into the main hcd structure, which detaches it from the class device structure. Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/hcd.c | 61 ++++++++++++++++++++------------------------- drivers/usb/host/ehci-dbg.c | 10 ++++---- drivers/usb/host/ohci-dbg.c | 10 ++++---- include/linux/usb.h | 5 ++-- 4 files changed, 39 insertions(+), 47 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c index 266e9e06a9f5..d041782e0c8b 100644 --- a/drivers/usb/core/hcd.c +++ b/drivers/usb/core/hcd.c @@ -651,50 +651,45 @@ static int usb_rh_urb_dequeue (struct usb_hcd *hcd, struct urb *urb) /*-------------------------------------------------------------------------*/ /* exported only within usbcore */ -struct usb_bus *usb_bus_get (struct usb_bus *bus) +struct usb_bus *usb_bus_get(struct usb_bus *bus) { - struct class_device *tmp; + if (bus) + kref_get(&bus->kref); + return bus; +} - if (!bus) - return NULL; +static void usb_host_release(struct kref *kref) +{ + struct usb_bus *bus = container_of(kref, struct usb_bus, kref); - tmp = class_device_get(&bus->class_dev); - if (tmp) - return to_usb_bus(tmp); - else - return NULL; + if (bus->release) + bus->release(bus); } /* exported only within usbcore */ -void usb_bus_put (struct usb_bus *bus) +void usb_bus_put(struct usb_bus *bus) { if (bus) - class_device_put(&bus->class_dev); + kref_put(&bus->kref, usb_host_release); } /*-------------------------------------------------------------------------*/ -static void usb_host_release(struct class_device *class_dev) -{ - struct usb_bus *bus = to_usb_bus(class_dev); - - if (bus->release) - bus->release(bus); -} - -static struct class usb_host_class = { - .name = "usb_host", - .release = &usb_host_release, -}; +static struct class *usb_host_class; int usb_host_init(void) { - return class_register(&usb_host_class); + int retval = 0; + + usb_host_class = class_create(THIS_MODULE, "usb_host"); + if (IS_ERR(usb_host_class)) + retval = PTR_ERR(usb_host_class); + return retval; } void usb_host_cleanup(void) { - class_unregister(&usb_host_class); + class_destroy(usb_host_class); } /** @@ -719,8 +714,7 @@ static void usb_bus_init (struct usb_bus *bus) INIT_LIST_HEAD (&bus->bus_list); - class_device_initialize(&bus->class_dev); - bus->class_dev.class = &usb_host_class; + kref_init(&bus->kref); } /** @@ -761,7 +755,6 @@ struct usb_bus *usb_alloc_bus (struct usb_operations *op) static int usb_register_bus(struct usb_bus *bus) { int busnum; - int retval; down (&usb_bus_list_lock); busnum = find_next_zero_bit (busmap.busmap, USB_MAXBUS, 1); @@ -774,15 +767,15 @@ static int usb_register_bus(struct usb_bus *bus) return -E2BIG; } - snprintf(bus->class_dev.class_id, BUS_ID_SIZE, "usb%d", busnum); - bus->class_dev.dev = bus->controller; - retval = class_device_add(&bus->class_dev); - if (retval) { + bus->class_dev = class_device_create(usb_host_class, MKDEV(0,0), bus->controller, "usb%d", busnum); + if (IS_ERR(bus->class_dev)) { clear_bit(busnum, busmap.busmap); up(&usb_bus_list_lock); - return retval; + return PTR_ERR(bus->class_dev); } + class_set_devdata(bus->class_dev, bus); + /* Add it to the local list of buses */ list_add (&bus->bus_list, &usb_bus_list); up (&usb_bus_list_lock); @@ -820,7 +813,7 @@ static void usb_deregister_bus (struct usb_bus *bus) clear_bit (bus->busnum, busmap.busmap); - class_device_del(&bus->class_dev); + class_device_unregister(bus->class_dev); } /** diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c index 9b347d765383..2ff11d53567b 100644 --- a/drivers/usb/host/ehci-dbg.c +++ b/drivers/usb/host/ehci-dbg.c @@ -450,7 +450,7 @@ show_async (struct class_device *class_dev, char *buf) *buf = 0; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ehci = hcd_to_ehci (hcd); next = buf; @@ -496,7 +496,7 @@ show_periodic (struct class_device *class_dev, char *buf) return 0; seen_count = 0; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ehci = hcd_to_ehci (hcd); next = buf; @@ -633,7 +633,7 @@ show_registers (struct class_device *class_dev, char *buf) static char fmt [] = "%*s\n"; static char label [] = ""; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ehci = hcd_to_ehci (hcd); next = buf; @@ -735,7 +735,7 @@ static CLASS_DEVICE_ATTR (registers, S_IRUGO, show_registers, NULL); static inline void create_debug_files (struct ehci_hcd *ehci) { - struct class_device *cldev = &ehci_to_hcd(ehci)->self.class_dev; + struct class_device *cldev = ehci_to_hcd(ehci)->self.class_dev; class_device_create_file(cldev, &class_device_attr_async); class_device_create_file(cldev, &class_device_attr_periodic); @@ -744,7 +744,7 @@ static inline void create_debug_files (struct ehci_hcd *ehci) static inline void remove_debug_files (struct ehci_hcd *ehci) { - struct class_device *cldev = &ehci_to_hcd(ehci)->self.class_dev; + struct class_device *cldev = ehci_to_hcd(ehci)->self.class_dev; class_device_remove_file(cldev, &class_device_attr_async); class_device_remove_file(cldev, &class_device_attr_periodic); diff --git a/drivers/usb/host/ohci-dbg.c b/drivers/usb/host/ohci-dbg.c index 62f53a213808..c58408c95c3d 100644 --- a/drivers/usb/host/ohci-dbg.c +++ b/drivers/usb/host/ohci-dbg.c @@ -481,7 +481,7 @@ show_async (struct class_device *class_dev, char *buf) size_t temp; unsigned long flags; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ohci = hcd_to_ohci(hcd); @@ -514,7 +514,7 @@ show_periodic (struct class_device *class_dev, char *buf) return 0; seen_count = 0; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ohci = hcd_to_ohci(hcd); next = buf; @@ -611,7 +611,7 @@ show_registers (struct class_device *class_dev, char *buf) char *next; u32 rdata; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ohci = hcd_to_ohci(hcd); regs = ohci->regs; @@ -684,7 +684,7 @@ static CLASS_DEVICE_ATTR (registers, S_IRUGO, show_registers, NULL); static inline void create_debug_files (struct ohci_hcd *ohci) { - struct class_device *cldev = &ohci_to_hcd(ohci)->self.class_dev; + struct class_device *cldev = ohci_to_hcd(ohci)->self.class_dev; class_device_create_file(cldev, &class_device_attr_async); class_device_create_file(cldev, &class_device_attr_periodic); @@ -694,7 +694,7 @@ static inline void create_debug_files (struct ohci_hcd *ohci) static inline void remove_debug_files (struct ohci_hcd *ohci) { - struct class_device *cldev = &ohci_to_hcd(ohci)->self.class_dev; + struct class_device *cldev = ohci_to_hcd(ohci)->self.class_dev; class_device_remove_file(cldev, &class_device_attr_async); class_device_remove_file(cldev, &class_device_attr_periodic); diff --git a/include/linux/usb.h b/include/linux/usb.h index 2d1ac5058534..3d508bf08402 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -287,15 +287,14 @@ struct usb_bus { struct dentry *usbfs_dentry; /* usbfs dentry entry for the bus */ - struct class_device class_dev; /* class device for this bus */ + struct class_device *class_dev; /* class device for this bus */ + struct kref kref; /* handles reference counting this bus */ void (*release)(struct usb_bus *bus); /* function to destroy this bus's memory */ #if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE) struct mon_bus *mon_bus; /* non-null when associated */ int monitored; /* non-zero when monitored */ #endif }; -#define to_usb_bus(d) container_of(d, struct usb_bus, class_dev) - /* -------------------------------------------------------------------------- */ -- cgit v1.2.3 From cd987d38cc59053e0bab8150ffaca33b109067f3 Mon Sep 17 00:00:00 2001 From: "gregkh@suse.de" Date: Wed, 23 Mar 2005 11:12:38 -0800 Subject: [PATCH] class: remove class_simple code, as no one in the tree is using it anymore. Signed-off-by: Greg Kroah-Hartman --- drivers/base/Makefile | 2 +- drivers/base/class_simple.c | 199 -------------------------------------------- include/linux/device.h | 10 --- 3 files changed, 1 insertion(+), 210 deletions(-) delete mode 100644 drivers/base/class_simple.c (limited to 'include/linux') diff --git a/drivers/base/Makefile b/drivers/base/Makefile index a47928a2e575..771f6c80e824 100644 --- a/drivers/base/Makefile +++ b/drivers/base/Makefile @@ -1,7 +1,7 @@ # Makefile for the Linux device tree obj-y := core.o sys.o bus.o \ - driver.o class.o class_simple.o platform.o \ + driver.o class.o platform.o \ cpu.o firmware.o init.o map.o dmapool.o \ attribute_container.o transport_class.o obj-y += power/ diff --git a/drivers/base/class_simple.c b/drivers/base/class_simple.c deleted file mode 100644 index 27699eb20a37..000000000000 --- a/drivers/base/class_simple.c +++ /dev/null @@ -1,199 +0,0 @@ -/* - * class_simple.c - a "simple" interface for classes for simple char devices. - * - * Copyright (c) 2003-2004 Greg Kroah-Hartman - * Copyright (c) 2003-2004 IBM Corp. - * - * This file is released under the GPLv2 - * - */ - -#include -#include -#include - -struct class_simple { - struct class class; -}; -#define to_class_simple(d) container_of(d, struct class_simple, class) - -struct simple_dev { - struct list_head node; - struct class_device class_dev; -}; -#define to_simple_dev(d) container_of(d, struct simple_dev, class_dev) - -static LIST_HEAD(simple_dev_list); -static DEFINE_SPINLOCK(simple_dev_list_lock); - -static void release_simple_dev(struct class_device *class_dev) -{ - struct simple_dev *s_dev = to_simple_dev(class_dev); - kfree(s_dev); -} - -static void class_simple_release(struct class *class) -{ - struct class_simple *cs = to_class_simple(class); - kfree(cs); -} - -/** - * class_simple_create - create a struct class_simple structure - * @owner: pointer to the module that is to "own" this struct class_simple - * @name: pointer to a string for the name of this class. - * - * This is used to create a struct class_simple pointer that can then be used - * in calls to class_simple_device_add(). This is used when you do not wish to - * create a full blown class support for a type of char devices. - * - * Note, the pointer created here is to be destroyed when finished by making a - * call to class_simple_destroy(). - */ -struct class_simple *class_simple_create(struct module *owner, char *name) -{ - struct class_simple *cs; - int retval; - - cs = kmalloc(sizeof(*cs), GFP_KERNEL); - if (!cs) { - retval = -ENOMEM; - goto error; - } - memset(cs, 0x00, sizeof(*cs)); - - cs->class.name = name; - cs->class.class_release = class_simple_release; - cs->class.release = release_simple_dev; - - retval = class_register(&cs->class); - if (retval) - goto error; - - return cs; - -error: - kfree(cs); - return ERR_PTR(retval); -} -EXPORT_SYMBOL(class_simple_create); - -/** - * class_simple_destroy - destroys a struct class_simple structure - * @cs: pointer to the struct class_simple that is to be destroyed - * - * Note, the pointer to be destroyed must have been created with a call to - * class_simple_create(). - */ -void class_simple_destroy(struct class_simple *cs) -{ - if ((cs == NULL) || (IS_ERR(cs))) - return; - - class_unregister(&cs->class); -} -EXPORT_SYMBOL(class_simple_destroy); - -/** - * class_simple_device_add - adds a class device to sysfs for a character driver - * @cs: pointer to the struct class_simple that this device should be registered to. - * @dev: the dev_t for the device to be added. - * @device: a pointer to a struct device that is assiociated with this class device. - * @fmt: string for the class device's name - * - * This function can be used by simple char device classes that do not - * implement their own class device registration. A struct class_device will - * be created in sysfs, registered to the specified class. A "dev" file will - * be created, showing the dev_t for the device. The pointer to the struct - * class_device will be returned from the call. Any further sysfs files that - * might be required can be created using this pointer. - * Note: the struct class_simple passed to this function must have previously been - * created with a call to class_simple_create(). - */ -struct class_device *class_simple_device_add(struct class_simple *cs, dev_t dev, struct device *device, const char *fmt, ...) -{ - va_list args; - struct simple_dev *s_dev = NULL; - int retval; - - if ((cs == NULL) || (IS_ERR(cs))) { - retval = -ENODEV; - goto error; - } - - s_dev = kmalloc(sizeof(*s_dev), GFP_KERNEL); - if (!s_dev) { - retval = -ENOMEM; - goto error; - } - memset(s_dev, 0x00, sizeof(*s_dev)); - - s_dev->class_dev.devt = dev; - s_dev->class_dev.dev = device; - s_dev->class_dev.class = &cs->class; - - va_start(args, fmt); - vsnprintf(s_dev->class_dev.class_id, BUS_ID_SIZE, fmt, args); - va_end(args); - retval = class_device_register(&s_dev->class_dev); - if (retval) - goto error; - - spin_lock(&simple_dev_list_lock); - list_add(&s_dev->node, &simple_dev_list); - spin_unlock(&simple_dev_list_lock); - - return &s_dev->class_dev; - -error: - kfree(s_dev); - return ERR_PTR(retval); -} -EXPORT_SYMBOL(class_simple_device_add); - -/** - * class_simple_set_hotplug - set the hotplug callback in the embedded struct class - * @cs: pointer to the struct class_simple to hold the pointer - * @hotplug: function pointer to the hotplug function - * - * Implement and set a hotplug function to add environment variables specific to this - * class on the hotplug event. - */ -int class_simple_set_hotplug(struct class_simple *cs, - int (*hotplug)(struct class_device *dev, char **envp, int num_envp, char *buffer, int buffer_size)) -{ - if ((cs == NULL) || (IS_ERR(cs))) - return -ENODEV; - cs->class.hotplug = hotplug; - return 0; -} -EXPORT_SYMBOL(class_simple_set_hotplug); - -/** - * class_simple_device_remove - removes a class device that was created with class_simple_device_add() - * @dev: the dev_t of the device that was previously registered. - * - * This call unregisters and cleans up a class device that was created with a - * call to class_device_simple_add() - */ -void class_simple_device_remove(dev_t dev) -{ - struct simple_dev *s_dev = NULL; - int found = 0; - - spin_lock(&simple_dev_list_lock); - list_for_each_entry(s_dev, &simple_dev_list, node) { - if (s_dev->class_dev.devt == dev) { - found = 1; - break; - } - } - if (found) { - list_del(&s_dev->node); - spin_unlock(&simple_dev_list_lock); - class_device_unregister(&s_dev->class_dev); - } else { - spin_unlock(&simple_dev_list_lock); - } -} -EXPORT_SYMBOL(class_simple_device_remove); diff --git a/include/linux/device.h b/include/linux/device.h index 73250d01c01f..68a95358013d 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -44,7 +44,6 @@ struct device; struct device_driver; struct class; struct class_device; -struct class_simple; struct bus_type { const char * name; @@ -254,15 +253,6 @@ extern struct class_device *class_device_create(struct class *cls, dev_t devt, __attribute__((format(printf,4,5))); extern void class_device_destroy(struct class *cls, dev_t devt); -/* interface for class simple stuff */ -extern struct class_simple *class_simple_create(struct module *owner, char *name); -extern void class_simple_destroy(struct class_simple *cs); -extern struct class_device *class_simple_device_add(struct class_simple *cs, dev_t dev, struct device *device, const char *fmt, ...) - __attribute__((format(printf,4,5))); -extern int class_simple_set_hotplug(struct class_simple *, - int (*hotplug)(struct class_device *dev, char **envp, int num_envp, char *buffer, int buffer_size)); -extern void class_simple_device_remove(dev_t dev); - struct device { struct list_head node; /* node in sibling list */ -- cgit v1.2.3 From af70316af182f4716cc5eec7e0d27fc731d164bd Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 10:41:04 -0800 Subject: [PATCH] Add a semaphore to struct device to synchronize calls to its driver. This adds a per-device semaphore that is taken before every call from the core to a driver method. This prevents e.g. simultaneous calls to the ->suspend() or ->resume() and ->probe() or ->release(), potentially saving a whole lot of headaches. It also moves us a step closer to removing the bus rwsem, since it protects the fields in struct device that are modified by the core. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 14 +++++++++++--- drivers/base/core.c | 1 + drivers/base/power/resume.c | 8 ++++++-- drivers/base/power/suspend.c | 3 ++- include/linux/device.h | 4 ++++ 5 files changed, 24 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 80ce88de56fa..aa27f76d28cd 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -283,18 +283,22 @@ void device_bind_driver(struct device * dev) */ int driver_probe_device(struct device_driver * drv, struct device * dev) { + int error = 0; + if (drv->bus->match && !drv->bus->match(dev, drv)) return -ENODEV; + down(&dev->sem); dev->driver = drv; if (drv->probe) { - int error = drv->probe(dev); + error = drv->probe(dev); if (error) { dev->driver = NULL; + up(&dev->sem); return error; } } - + up(&dev->sem); device_bind_driver(dev); return 0; } @@ -385,7 +389,10 @@ void driver_attach(struct device_driver * drv) void device_release_driver(struct device * dev) { - struct device_driver * drv = dev->driver; + struct device_driver * drv; + + down(&dev->sem); + drv = dev->driver; if (drv) { sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); sysfs_remove_link(&dev->kobj, "driver"); @@ -394,6 +401,7 @@ void device_release_driver(struct device * dev) drv->remove(dev); dev->driver = NULL; } + up(&dev->sem); } diff --git a/drivers/base/core.c b/drivers/base/core.c index a293a788abd4..93440824b800 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -212,6 +212,7 @@ void device_initialize(struct device *dev) INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->bus_list); INIT_LIST_HEAD(&dev->dma_pools); + init_MUTEX(&dev->sem); } /** diff --git a/drivers/base/power/resume.c b/drivers/base/power/resume.c index 26468971ef5a..bdd96b03b885 100644 --- a/drivers/base/power/resume.c +++ b/drivers/base/power/resume.c @@ -22,6 +22,9 @@ extern int sysdev_resume(void); int resume_device(struct device * dev) { + int error = 0; + + down(&dev->sem); if (dev->power.pm_parent && dev->power.pm_parent->power.power_state) { dev_err(dev, "PM: resume from %d, parent %s still %d\n", @@ -31,9 +34,10 @@ int resume_device(struct device * dev) } if (dev->bus && dev->bus->resume) { dev_dbg(dev,"resuming\n"); - return dev->bus->resume(dev); + error = dev->bus->resume(dev); } - return 0; + up(&dev->sem); + return error; } diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c index 0ec44ef840be..807e13fb205b 100644 --- a/drivers/base/power/suspend.c +++ b/drivers/base/power/suspend.c @@ -39,6 +39,7 @@ int suspend_device(struct device * dev, pm_message_t state) { int error = 0; + down(&dev->sem); if (dev->power.power_state) { dev_dbg(dev, "PM: suspend %d-->%d\n", dev->power.power_state, state); @@ -58,7 +59,7 @@ int suspend_device(struct device * dev, pm_message_t state) dev_dbg(dev, "suspending\n"); error = dev->bus->suspend(dev, state); } - + up(&dev->sem); return error; } diff --git a/include/linux/device.h b/include/linux/device.h index 68a95358013d..195b327f3e19 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -264,6 +264,10 @@ struct device { struct kobject kobj; char bus_id[BUS_ID_SIZE]; /* position on parent bus */ + struct semaphore sem; /* semaphore to synchronize calls to + * its driver. + */ + struct bus_type * bus; /* type of bus device is on */ struct device_driver *driver; /* which driver has allocated this device */ -- cgit v1.2.3 From fae3cd00255e3e51ffd59fedb1bdb91ec96be395 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 10:59:56 -0800 Subject: [PATCH] Add driver_for_each_device(). Now there's an iterator for accessing each device bound to a driver. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman Index: linux-2.6.12-rc2/drivers/base/driver.c =================================================================== --- drivers/base/driver.c | 35 +++++++++++++++++++++++++++++++++++ include/linux/device.h | 3 +++ 2 files changed, 38 insertions(+) (limited to 'include/linux') diff --git a/drivers/base/driver.c b/drivers/base/driver.c index 3b269f7e5213..484fed1985aa 100644 --- a/drivers/base/driver.c +++ b/drivers/base/driver.c @@ -18,6 +18,41 @@ #define to_dev(node) container_of(node, struct device, driver_list) #define to_drv(obj) container_of(obj, struct device_driver, kobj) + +/** + * driver_for_each_device - Iterator for devices bound to a driver. + * @drv: Driver we're iterating. + * @data: Data to pass to the callback. + * @fn: Function to call for each device. + * + * Take the bus's rwsem and iterate over the @drv's list of devices, + * calling @fn for each one. + */ + +int driver_for_each_device(struct device_driver * drv, struct device * start, + void * data, int (*fn)(struct device *, void *)) +{ + struct list_head * head; + struct device * dev; + int error = 0; + + down_read(&drv->bus->subsys.rwsem); + head = &drv->devices; + dev = list_prepare_entry(start, head, driver_list); + list_for_each_entry_continue(dev, head, driver_list) { + get_device(dev); + error = fn(dev, data); + put_device(dev); + if (error) + break; + } + up_read(&drv->bus->subsys.rwsem); + return error; +} + +EXPORT_SYMBOL(driver_for_each_device); + + /** * driver_create_file - create sysfs file for driver. * @drv: driver. diff --git a/include/linux/device.h b/include/linux/device.h index 195b327f3e19..10f5aa20e041 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -136,6 +136,9 @@ struct driver_attribute driver_attr_##_name = __ATTR(_name,_mode,_show,_store) extern int driver_create_file(struct device_driver *, struct driver_attribute *); extern void driver_remove_file(struct device_driver *, struct driver_attribute *); +extern int driver_for_each_device(struct device_driver * drv, struct device * start, + void * data, int (*fn)(struct device *, void *)); + /* * device classes -- cgit v1.2.3 From 9a19fea43616066561e221359596ce532e631395 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 11:45:16 -0800 Subject: [PATCH] Add initial implementation of klist helpers. This klist interface provides a couple of structures that wrap around struct list_head to provide explicit list "head" (struct klist) and list "node" (struct klist_node) objects. For struct klist, a spinlock is included that protects access to the actual list itself. struct klist_node provides a pointer to the klist that owns it and a kref reference count that indicates the number of current users of that node in the list. The entire point is to provide an interface for iterating over a list that is safe and allows for modification of the list during the iteration (e.g. insertion and removal), including modification of the current node on the list. It works using a 3rd object type - struct klist_iter - that is declared and initialized before an iteration. klist_next() is used to acquire the next element in the list. It returns NULL if there are no more items. This klist interface provides a couple of structures that wrap around struct list_head to provide explicit list "head" (struct klist) and list "node" (struct klist_node) objects. For struct klist, a spinlock is included that protects access to the actual list itself. struct klist_node provides a pointer to the klist that owns it and a kref reference count that indicates the number of current users of that node in the list. The entire point is to provide an interface for iterating over a list that is safe and allows for modification of the list during the iteration (e.g. insertion and removal), including modification of the current node on the list. It works using a 3rd object type - struct klist_iter - that is declared and initialized before an iteration. klist_next() is used to acquire the next element in the list. It returns NULL if there are no more items. Internally, that routine takes the klist's lock, decrements the reference count of the previous klist_node and increments the count of the next klist_node. It then drops the lock and returns. There are primitives for adding and removing nodes to/from a klist. When deleting, klist_del() will simply decrement the reference count. Only when the count goes to 0 is the node removed from the list. klist_remove() will try to delete the node from the list and block until it is actually removed. This is useful for objects (like devices) that have been removed from the system and must be freed (but must wait until all accessors have finished). Internally, that routine takes the klist's lock, decrements the reference count of the previous klist_node and increments the count of the next klist_node. It then drops the lock and returns. There are primitives for adding and removing nodes to/from a klist. When deleting, klist_del() will simply decrement the reference count. Only when the count goes to 0 is the node removed from the list. klist_remove() will try to delete the node from the list and block until it is actually removed. This is useful for objects (like devices) that have been removed from the system and must be freed (but must wait until all accessors have finished). Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman diff -Nru a/include/linux/klist.h b/include/linux/klist.h --- include/linux/klist.h | 53 +++++++++++ lib/Makefile | 7 +- lib/klist.c | 248 ++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 305 insertions(+), 3 deletions(-) create mode 100644 include/linux/klist.h create mode 100644 lib/klist.c (limited to 'include/linux') diff --git a/include/linux/klist.h b/include/linux/klist.h new file mode 100644 index 000000000000..fb52f9d9d611 --- /dev/null +++ b/include/linux/klist.h @@ -0,0 +1,53 @@ +/* + * klist.h - Some generic list helpers, extending struct list_head a bit. + * + * Implementations are found in lib/klist.c + * + * + * Copyright (C) 2005 Patrick Mochel + * + * This file is rleased under the GPL v2. + */ + +#include +#include +#include +#include + + +struct klist { + spinlock_t k_lock; + struct list_head k_list; +}; + + +extern void klist_init(struct klist * k); + + +struct klist_node { + struct klist * n_klist; + struct list_head n_node; + struct kref n_ref; + struct completion n_removed; +}; + +extern void klist_add_tail(struct klist * k, struct klist_node * n); +extern void klist_add_head(struct klist * k, struct klist_node * n); + +extern void klist_del(struct klist_node * n); +extern void klist_remove(struct klist_node * n); + + +struct klist_iter { + struct klist * i_klist; + struct list_head * i_head; + struct klist_node * i_cur; +}; + + +extern void klist_iter_init(struct klist * k, struct klist_iter * i); +extern void klist_iter_init_node(struct klist * k, struct klist_iter * i, + struct klist_node * n); +extern void klist_iter_exit(struct klist_iter * i); +extern struct klist_node * klist_next(struct klist_iter * i); + diff --git a/lib/Makefile b/lib/Makefile index 7c70db79c0e0..9eccea9429a7 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -4,9 +4,10 @@ lib-y := errno.o ctype.o string.o vsprintf.o cmdline.o \ bust_spinlocks.o rbtree.o radix-tree.o dump_stack.o \ - kobject.o kref.o idr.o div64.o int_sqrt.o \ - bitmap.o extable.o kobject_uevent.o prio_tree.o sha1.o \ - halfmd4.o + idr.o div64.o int_sqrt.o bitmap.o extable.o prio_tree.o \ + sha1.o halfmd4.o + +lib-y += kobject.o kref.o kobject_uevent.o klist.o obj-y += sort.o parser.o diff --git a/lib/klist.c b/lib/klist.c new file mode 100644 index 000000000000..6f760424648b --- /dev/null +++ b/lib/klist.c @@ -0,0 +1,248 @@ +/* + * klist.c - Routines for manipulating klists. + * + * + * This klist interface provides a couple of structures that wrap around + * struct list_head to provide explicit list "head" (struct klist) and + * list "node" (struct klist_node) objects. For struct klist, a spinlock + * is included that protects access to the actual list itself. struct + * klist_node provides a pointer to the klist that owns it and a kref + * reference count that indicates the number of current users of that node + * in the list. + * + * The entire point is to provide an interface for iterating over a list + * that is safe and allows for modification of the list during the + * iteration (e.g. insertion and removal), including modification of the + * current node on the list. + * + * It works using a 3rd object type - struct klist_iter - that is declared + * and initialized before an iteration. klist_next() is used to acquire the + * next element in the list. It returns NULL if there are no more items. + * Internally, that routine takes the klist's lock, decrements the reference + * count of the previous klist_node and increments the count of the next + * klist_node. It then drops the lock and returns. + * + * There are primitives for adding and removing nodes to/from a klist. + * When deleting, klist_del() will simply decrement the reference count. + * Only when the count goes to 0 is the node removed from the list. + * klist_remove() will try to delete the node from the list and block + * until it is actually removed. This is useful for objects (like devices) + * that have been removed from the system and must be freed (but must wait + * until all accessors have finished). + * + * Copyright (C) 2005 Patrick Mochel + * + * This file is released under the GPL v2. + */ + +#include +#include + + +/** + * klist_init - Initialize a klist structure. + * @k: The klist we're initializing. + */ + +void klist_init(struct klist * k) +{ + INIT_LIST_HEAD(&k->k_list); + spin_lock_init(&k->k_lock); +} + +EXPORT_SYMBOL_GPL(klist_init); + + +static void add_head(struct klist * k, struct klist_node * n) +{ + spin_lock(&k->k_lock); + list_add(&n->n_node, &k->k_list); + spin_unlock(&k->k_lock); +} + +static void add_tail(struct klist * k, struct klist_node * n) +{ + spin_lock(&k->k_lock); + list_add_tail(&n->n_node, &k->k_list); + spin_unlock(&k->k_lock); +} + + +static void klist_node_init(struct klist * k, struct klist_node * n) +{ + INIT_LIST_HEAD(&n->n_node); + init_completion(&n->n_removed); + kref_init(&n->n_ref); + n->n_klist = k; +} + + +/** + * klist_add_head - Initialize a klist_node and add it to front. + * @k: klist it's going on. + * @n: node we're adding. + */ + +void klist_add_head(struct klist * k, struct klist_node * n) +{ + klist_node_init(k, n); + add_head(k, n); +} + +EXPORT_SYMBOL_GPL(klist_add_head); + + +/** + * klist_add_tail - Initialize a klist_node and add it to back. + * @k: klist it's going on. + * @n: node we're adding. + */ + +void klist_add_tail(struct klist * k, struct klist_node * n) +{ + klist_node_init(k, n); + add_tail(k, n); +} + +EXPORT_SYMBOL_GPL(klist_add_tail); + + +static void klist_release(struct kref * kref) +{ + struct klist_node * n = container_of(kref, struct klist_node, n_ref); + list_del(&n->n_node); + complete(&n->n_removed); +} + +static int klist_dec_and_del(struct klist_node * n) +{ + return kref_put(&n->n_ref, klist_release); +} + + +/** + * klist_del - Decrement the reference count of node and try to remove. + * @n: node we're deleting. + */ + +void klist_del(struct klist_node * n) +{ + struct klist * k = n->n_klist; + + spin_lock(&k->k_lock); + klist_dec_and_del(n); + spin_unlock(&k->k_lock); +} + +EXPORT_SYMBOL_GPL(klist_del); + + +/** + * klist_remove - Decrement the refcount of node and wait for it to go away. + * @n: node we're removing. + */ + +void klist_remove(struct klist_node * n) +{ + spin_lock(&n->n_klist->k_lock); + klist_dec_and_del(n); + spin_unlock(&n->n_klist->k_lock); + wait_for_completion(&n->n_removed); +} + +EXPORT_SYMBOL_GPL(klist_remove); + + +/** + * klist_iter_init_node - Initialize a klist_iter structure. + * @k: klist we're iterating. + * @i: klist_iter we're filling. + * @n: node to start with. + * + * Similar to klist_iter_init(), but starts the action off with @n, + * instead of with the list head. + */ + +void klist_iter_init_node(struct klist * k, struct klist_iter * i, struct klist_node * n) +{ + i->i_klist = k; + i->i_head = &k->k_list; + i->i_cur = n; +} + +EXPORT_SYMBOL_GPL(klist_iter_init_node); + + +/** + * klist_iter_init - Iniitalize a klist_iter structure. + * @k: klist we're iterating. + * @i: klist_iter structure we're filling. + * + * Similar to klist_iter_init_node(), but start with the list head. + */ + +void klist_iter_init(struct klist * k, struct klist_iter * i) +{ + klist_iter_init_node(k, i, NULL); +} + +EXPORT_SYMBOL_GPL(klist_iter_init); + + +/** + * klist_iter_exit - Finish a list iteration. + * @i: Iterator structure. + * + * Must be called when done iterating over list, as it decrements the + * refcount of the current node. Necessary in case iteration exited before + * the end of the list was reached, and always good form. + */ + +void klist_iter_exit(struct klist_iter * i) +{ + if (i->i_cur) { + klist_del(i->i_cur); + i->i_cur = NULL; + } +} + +EXPORT_SYMBOL_GPL(klist_iter_exit); + + +static struct klist_node * to_klist_node(struct list_head * n) +{ + return container_of(n, struct klist_node, n_node); +} + + +/** + * klist_next - Ante up next node in list. + * @i: Iterator structure. + * + * First grab list lock. Decrement the reference count of the previous + * node, if there was one. Grab the next node, increment its reference + * count, drop the lock, and return that next node. + */ + +struct klist_node * klist_next(struct klist_iter * i) +{ + struct list_head * next; + struct klist_node * knode = NULL; + + spin_lock(&i->i_klist->k_lock); + if (i->i_cur) { + next = i->i_cur->n_node.next; + klist_dec_and_del(i->i_cur); + } else + next = i->i_head->next; + + if (next != i->i_head) { + knode = to_klist_node(next); + kref_get(&knode->n_ref); + } + i->i_cur = knode; + spin_unlock(&i->i_klist->k_lock); + return knode; +} + +EXPORT_SYMBOL_GPL(klist_next); -- cgit v1.2.3 From 465c7a3a3a5aabcedd2e43612cac5a12f23da19a Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 11:49:14 -0800 Subject: [PATCH] Add a klist to struct bus_type for its devices. - Use it for bus_for_each_dev(). - Use the klist spinlock instead of the bus rwsem. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 54 ++++++++++++++++++++++---------------------------- include/linux/device.h | 3 +++ 2 files changed, 27 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 1b05c1399e37..c8da37230b3e 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -134,28 +134,6 @@ static struct kobj_type ktype_bus = { decl_subsys(bus, &ktype_bus, NULL); -static int __bus_for_each_dev(struct bus_type *bus, struct device *start, - void *data, int (*fn)(struct device *, void *)) -{ - struct list_head *head; - struct device *dev; - int error = 0; - - if (!(bus = get_bus(bus))) - return -EINVAL; - - head = &bus->devices.list; - dev = list_prepare_entry(start, head, bus_list); - list_for_each_entry_continue(dev, head, bus_list) { - get_device(dev); - error = fn(dev, data); - put_device(dev); - if (error) - break; - } - put_bus(bus); - return error; -} static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start, void * data, int (*fn)(struct device_driver *, void *)) @@ -180,6 +158,13 @@ static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start, return error; } + +static struct device * next_device(struct klist_iter * i) +{ + struct klist_node * n = klist_next(i); + return n ? container_of(n, struct device, knode_bus) : NULL; +} + /** * bus_for_each_dev - device iterator. * @bus: bus type. @@ -203,12 +188,19 @@ static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start, int bus_for_each_dev(struct bus_type * bus, struct device * start, void * data, int (*fn)(struct device *, void *)) { - int ret; + struct klist_iter i; + struct device * dev; + int error = 0; - down_read(&bus->subsys.rwsem); - ret = __bus_for_each_dev(bus, start, data, fn); - up_read(&bus->subsys.rwsem); - return ret; + if (!bus) + return -EINVAL; + + klist_iter_init_node(&bus->klist_devices, &i, + (start ? &start->knode_bus : NULL)); + while ((dev = next_device(&i)) && !error) + error = fn(dev, data); + klist_iter_exit(&i); + return error; } /** @@ -293,6 +285,7 @@ int bus_add_device(struct device * dev) list_add_tail(&dev->bus_list, &dev->bus->devices.list); device_attach(dev); up_write(&dev->bus->subsys.rwsem); + klist_add_tail(&bus->klist_devices, &dev->knode_bus); device_add_attrs(bus, dev); sysfs_create_link(&bus->devices.kobj, &dev->kobj, dev->bus_id); sysfs_create_link(&dev->kobj, &dev->bus->subsys.kset.kobj, "bus"); @@ -315,6 +308,7 @@ void bus_remove_device(struct device * dev) sysfs_remove_link(&dev->kobj, "bus"); sysfs_remove_link(&dev->bus->devices.kobj, dev->bus_id); device_remove_attrs(dev->bus, dev); + klist_remove(&dev->knode_bus); down_write(&dev->bus->subsys.rwsem); pr_debug("bus %s: remove device %s\n", dev->bus->name, dev->bus_id); device_release_driver(dev); @@ -439,9 +433,7 @@ int bus_rescan_devices(struct bus_type * bus) { int count = 0; - down_write(&bus->subsys.rwsem); - __bus_for_each_dev(bus, NULL, &count, bus_rescan_devices_helper); - up_write(&bus->subsys.rwsem); + bus_for_each_dev(bus, NULL, &count, bus_rescan_devices_helper); return count; } @@ -542,6 +534,8 @@ int bus_register(struct bus_type * bus) retval = kset_register(&bus->drivers); if (retval) goto bus_drivers_fail; + + klist_init(&bus->klist_devices); bus_add_attrs(bus); pr_debug("bus type '%s' registered\n", bus->name); diff --git a/include/linux/device.h b/include/linux/device.h index 10f5aa20e041..e36953cf7f14 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -51,6 +52,7 @@ struct bus_type { struct subsystem subsys; struct kset drivers; struct kset devices; + struct klist klist_devices; struct bus_attribute * bus_attrs; struct device_attribute * dev_attrs; @@ -262,6 +264,7 @@ struct device { struct list_head bus_list; /* node in bus's list */ struct list_head driver_list; struct list_head children; + struct klist_node knode_bus; struct device * parent; struct kobject kobj; -- cgit v1.2.3 From 38fdac3cdce276554b4484a41f8ec2daf81cb2ff Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 12:00:18 -0800 Subject: [PATCH] Add a klist to struct bus_type for its drivers. - Use it in bus_for_each_drv(). - Use the klist spinlock instead of the bus rwsem. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 52 ++++++++++++++++++++++---------------------------- include/linux/device.h | 2 ++ 2 files changed, 25 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/bus.c b/drivers/base/bus.c index c8da37230b3e..736dc1f5a316 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -135,30 +135,6 @@ static struct kobj_type ktype_bus = { decl_subsys(bus, &ktype_bus, NULL); -static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start, - void * data, int (*fn)(struct device_driver *, void *)) -{ - struct list_head *head; - struct device_driver *drv; - int error = 0; - - if (!(bus = get_bus(bus))) - return -EINVAL; - - head = &bus->drivers.list; - drv = list_prepare_entry(start, head, kobj.entry); - list_for_each_entry_continue(drv, head, kobj.entry) { - get_driver(drv); - error = fn(drv, data); - put_driver(drv); - if (error) - break; - } - put_bus(bus); - return error; -} - - static struct device * next_device(struct klist_iter * i) { struct klist_node * n = klist_next(i); @@ -203,6 +179,14 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start, return error; } + + +static struct device_driver * next_driver(struct klist_iter * i) +{ + struct klist_node * n = klist_next(i); + return n ? container_of(n, struct device_driver, knode_bus) : NULL; +} + /** * bus_for_each_drv - driver iterator * @bus: bus we're dealing with. @@ -226,12 +210,19 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start, int bus_for_each_drv(struct bus_type * bus, struct device_driver * start, void * data, int (*fn)(struct device_driver *, void *)) { - int ret; + struct klist_iter i; + struct device_driver * drv; + int error = 0; - down_read(&bus->subsys.rwsem); - ret = __bus_for_each_drv(bus, start, data, fn); - up_read(&bus->subsys.rwsem); - return ret; + if (!bus) + return -EINVAL; + + klist_iter_init_node(&bus->klist_drivers, &i, + start ? &start->knode_bus : NULL); + while ((drv = next_driver(&i)) && !error) + error = fn(drv, data); + klist_iter_exit(&i); + return error; } static int device_add_attrs(struct bus_type * bus, struct device * dev) @@ -376,6 +367,7 @@ int bus_add_driver(struct device_driver * drv) down_write(&bus->subsys.rwsem); driver_attach(drv); up_write(&bus->subsys.rwsem); + klist_add_tail(&bus->klist_drivers, &drv->knode_bus); module_add_driver(drv->owner, drv); driver_add_attrs(bus, drv); @@ -397,6 +389,7 @@ void bus_remove_driver(struct device_driver * drv) { if (drv->bus) { driver_remove_attrs(drv->bus, drv); + klist_remove(&drv->knode_bus); down_write(&drv->bus->subsys.rwsem); pr_debug("bus %s: remove driver %s\n", drv->bus->name, drv->name); driver_detach(drv); @@ -536,6 +529,7 @@ int bus_register(struct bus_type * bus) goto bus_drivers_fail; klist_init(&bus->klist_devices); + klist_init(&bus->klist_drivers); bus_add_attrs(bus); pr_debug("bus type '%s' registered\n", bus->name); diff --git a/include/linux/device.h b/include/linux/device.h index e36953cf7f14..ea9ab33dfe71 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -53,6 +53,7 @@ struct bus_type { struct kset drivers; struct kset devices; struct klist klist_devices; + struct klist klist_drivers; struct bus_attribute * bus_attrs; struct device_attribute * dev_attrs; @@ -105,6 +106,7 @@ struct device_driver { struct completion unloaded; struct kobject kobj; struct list_head devices; + struct klist_node knode_bus; struct module * owner; -- cgit v1.2.3 From 94e7b1c5ff2055571703e38b059afffe17658432 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 12:25:36 -0800 Subject: [PATCH] Add a klist to struct device_driver for the devices bound to it. - Use it in driver_for_each_device() instead of the regular list_head and stop using the bus's rwsem for protection. - Use driver_for_each_device() in driver_detach() so we don't deadlock on the bus's rwsem. - Remove ->devices. - Move klist access and sysfs link access out from under device's semaphore, since they're synchronized through other means. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 2 +- drivers/base/dd.c | 34 +++++++++++++++++++--------------- drivers/base/driver.c | 27 +++++++++++++++------------ include/linux/device.h | 3 ++- 4 files changed, 37 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/core.c b/drivers/base/core.c index 93440824b800..bc5bec61a01a 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -209,8 +209,8 @@ void device_initialize(struct device *dev) kobject_init(&dev->kobj); INIT_LIST_HEAD(&dev->node); INIT_LIST_HEAD(&dev->children); - INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->bus_list); + INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->dma_pools); init_MUTEX(&dev->sem); } diff --git a/drivers/base/dd.c b/drivers/base/dd.c index b709b1e0cb2a..47cbb5641235 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -40,7 +40,7 @@ void device_bind_driver(struct device * dev) { pr_debug("bound device '%s' to driver '%s'\n", dev->bus_id, dev->driver->name); - list_add_tail(&dev->driver_list, &dev->driver->devices); + klist_add_tail(&dev->driver->klist_devices, &dev->knode_driver); sysfs_create_link(&dev->driver->kobj, &dev->kobj, kobject_name(&dev->kobj)); sysfs_create_link(&dev->kobj, &dev->driver->kobj, "driver"); @@ -164,31 +164,35 @@ void driver_attach(struct device_driver * drv) */ void device_release_driver(struct device * dev) { - struct device_driver * drv; + struct device_driver * drv = dev->driver; + + if (!drv) + return; + + sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); + sysfs_remove_link(&dev->kobj, "driver"); + klist_remove(&dev->knode_driver); down(&dev->sem); - drv = dev->driver; - if (drv) { - sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); - sysfs_remove_link(&dev->kobj, "driver"); - list_del_init(&dev->driver_list); - if (drv->remove) - drv->remove(dev); - dev->driver = NULL; - } + if (drv->remove) + drv->remove(dev); + dev->driver = NULL; up(&dev->sem); } +static int __remove_driver(struct device * dev, void * unused) +{ + device_release_driver(dev); + return 0; +} + /** * driver_detach - detach driver from all devices it controls. * @drv: driver. */ void driver_detach(struct device_driver * drv) { - while (!list_empty(&drv->devices)) { - struct device * dev = container_of(drv->devices.next, struct device, driver_list); - device_release_driver(dev); - } + driver_for_each_device(drv, NULL, NULL, __remove_driver); } diff --git a/drivers/base/driver.c b/drivers/base/driver.c index 484fed1985aa..34bd38aa7eb8 100644 --- a/drivers/base/driver.c +++ b/drivers/base/driver.c @@ -19,6 +19,12 @@ #define to_drv(obj) container_of(obj, struct device_driver, kobj) +static struct device * next_device(struct klist_iter * i) +{ + struct klist_node * n = klist_next(i); + return n ? container_of(n, struct device, knode_driver) : NULL; +} + /** * driver_for_each_device - Iterator for devices bound to a driver. * @drv: Driver we're iterating. @@ -32,21 +38,18 @@ int driver_for_each_device(struct device_driver * drv, struct device * start, void * data, int (*fn)(struct device *, void *)) { - struct list_head * head; + struct klist_iter i; struct device * dev; int error = 0; - down_read(&drv->bus->subsys.rwsem); - head = &drv->devices; - dev = list_prepare_entry(start, head, driver_list); - list_for_each_entry_continue(dev, head, driver_list) { - get_device(dev); + if (!drv) + return -EINVAL; + + klist_iter_init_node(&drv->klist_devices, &i, + start ? &start->knode_driver : NULL); + while ((dev = next_device(&i)) && !error) error = fn(dev, data); - put_device(dev); - if (error) - break; - } - up_read(&drv->bus->subsys.rwsem); + klist_iter_exit(&i); return error; } @@ -120,7 +123,7 @@ void put_driver(struct device_driver * drv) */ int driver_register(struct device_driver * drv) { - INIT_LIST_HEAD(&drv->devices); + klist_init(&drv->klist_devices); init_completion(&drv->unloaded); return bus_add_driver(drv); } diff --git a/include/linux/device.h b/include/linux/device.h index ea9ab33dfe71..96c71b59fdef 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -105,7 +105,7 @@ struct device_driver { struct completion unloaded; struct kobject kobj; - struct list_head devices; + struct klist klist_devices; struct klist_node knode_bus; struct module * owner; @@ -266,6 +266,7 @@ struct device { struct list_head bus_list; /* node in bus's list */ struct list_head driver_list; struct list_head children; + struct klist_node knode_driver; struct klist_node knode_bus; struct device * parent; -- cgit v1.2.3 From cb85b6f1cc811ecb9ed4b950206d8941ba710e68 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 10:48:35 -0800 Subject: [PATCH] Remove the unused device_find(). Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 19 ------------------- include/linux/device.h | 1 - 2 files changed, 20 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/core.c b/drivers/base/core.c index bc5bec61a01a..0eb1d424bd1d 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -401,24 +401,6 @@ int device_for_each_child(struct device * dev, void * data, return error; } -/** - * device_find - locate device on a bus by name. - * @name: name of the device. - * @bus: bus to scan for the device. - * - * Call kset_find_obj() to iterate over list of devices on - * a bus to find device by name. Return device if found. - * - * Note that kset_find_obj increments device's reference count. - */ -struct device *device_find(const char *name, struct bus_type *bus) -{ - struct kobject *k = kset_find_obj(&bus->devices, name); - if (k) - return to_dev(k); - return NULL; -} - int __init devices_init(void) { return subsystem_register(&devices_subsys); @@ -434,7 +416,6 @@ EXPORT_SYMBOL_GPL(device_del); EXPORT_SYMBOL_GPL(device_unregister); EXPORT_SYMBOL_GPL(get_device); EXPORT_SYMBOL_GPL(put_device); -EXPORT_SYMBOL_GPL(device_find); EXPORT_SYMBOL_GPL(device_create_file); EXPORT_SYMBOL_GPL(device_remove_file); diff --git a/include/linux/device.h b/include/linux/device.h index 96c71b59fdef..f1c38d869ac9 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -372,7 +372,6 @@ extern int (*platform_notify_remove)(struct device * dev); */ extern struct device * get_device(struct device * dev); extern void put_device(struct device * dev); -extern struct device *device_find(const char *name, struct bus_type *bus); /* drivers/base/platform.c */ -- cgit v1.2.3 From 8b0c250be489dcbf1a3a33bb4ec4c7f33735a365 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 12:58:57 -0800 Subject: [PATCH] add klist_node_attached() to determine if a node is on a list or not. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman diff -Nru a/include/linux/klist.h b/include/linux/klist.h --- include/linux/klist.h | 2 ++ lib/klist.c | 16 ++++++++++++++++ 2 files changed, 18 insertions(+) (limited to 'include/linux') diff --git a/include/linux/klist.h b/include/linux/klist.h index fb52f9d9d611..eebf5e5696ec 100644 --- a/include/linux/klist.h +++ b/include/linux/klist.h @@ -37,6 +37,8 @@ extern void klist_add_head(struct klist * k, struct klist_node * n); extern void klist_del(struct klist_node * n); extern void klist_remove(struct klist_node * n); +extern int klist_node_attached(struct klist_node * n); + struct klist_iter { struct klist * i_klist; diff --git a/lib/klist.c b/lib/klist.c index 6f760424648b..02177d72dc89 100644 --- a/lib/klist.c +++ b/lib/klist.c @@ -112,6 +112,7 @@ static void klist_release(struct kref * kref) struct klist_node * n = container_of(kref, struct klist_node, n_ref); list_del(&n->n_node); complete(&n->n_removed); + n->n_klist = NULL; } static int klist_dec_and_del(struct klist_node * n) @@ -153,6 +154,19 @@ void klist_remove(struct klist_node * n) EXPORT_SYMBOL_GPL(klist_remove); +/** + * klist_node_attached - Say whether a node is bound to a list or not. + * @n: Node that we're testing. + */ + +int klist_node_attached(struct klist_node * n) +{ + return (n->n_klist != NULL); +} + +EXPORT_SYMBOL_GPL(klist_node_attached); + + /** * klist_iter_init_node - Initialize a klist_iter structure. * @k: klist we're iterating. @@ -246,3 +260,5 @@ struct klist_node * klist_next(struct klist_iter * i) } EXPORT_SYMBOL_GPL(klist_next); + + -- cgit v1.2.3 From 7dc35cdf69149a7f2b5216ada9bafe359746ac1c Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 13:03:35 -0800 Subject: [PATCH] Remove struct device::bus_list. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 1 - include/linux/device.h | 1 - 2 files changed, 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/core.c b/drivers/base/core.c index 0eb1d424bd1d..f258a21ed78a 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -209,7 +209,6 @@ void device_initialize(struct device *dev) kobject_init(&dev->kobj); INIT_LIST_HEAD(&dev->node); INIT_LIST_HEAD(&dev->children); - INIT_LIST_HEAD(&dev->bus_list); INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->dma_pools); init_MUTEX(&dev->sem); diff --git a/include/linux/device.h b/include/linux/device.h index f1c38d869ac9..13f65853e583 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -263,7 +263,6 @@ extern void class_device_destroy(struct class *cls, dev_t devt); struct device { struct list_head node; /* node in sibling list */ - struct list_head bus_list; /* node in bus's list */ struct list_head driver_list; struct list_head children; struct klist_node knode_driver; -- cgit v1.2.3 From 63c4f204ffc8219696bda88eb20c9873d007a2fc Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 13:08:05 -0800 Subject: [PATCH] Remove struct device::driver_list. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 1 - include/linux/device.h | 1 - 2 files changed, 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/core.c b/drivers/base/core.c index f258a21ed78a..bfa4268289ea 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -209,7 +209,6 @@ void device_initialize(struct device *dev) kobject_init(&dev->kobj); INIT_LIST_HEAD(&dev->node); INIT_LIST_HEAD(&dev->children); - INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->dma_pools); init_MUTEX(&dev->sem); } diff --git a/include/linux/device.h b/include/linux/device.h index 13f65853e583..d2434934d091 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -263,7 +263,6 @@ extern void class_device_destroy(struct class *cls, dev_t devt); struct device { struct list_head node; /* node in sibling list */ - struct list_head driver_list; struct list_head children; struct klist_node knode_driver; struct klist_node knode_bus; -- cgit v1.2.3 From 36239577cfb6b9a7c111209536b54200b0252ebf Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 19:08:30 -0800 Subject: [PATCH] Use a klist for device child lists. - Use klist iterator in device_for_each_child(), making it safe to use for removing devices. - Remove unused list_to_dev() function. - Kills all usage of devices_subsys.rwsem. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 30 +++++++++++++++--------------- include/linux/device.h | 10 ++-------- 2 files changed, 17 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/core.c b/drivers/base/core.c index bfa4268289ea..1d8c7790b55a 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -207,8 +207,7 @@ void device_initialize(struct device *dev) { kobj_set_kset_s(dev, devices_subsys); kobject_init(&dev->kobj); - INIT_LIST_HEAD(&dev->node); - INIT_LIST_HEAD(&dev->children); + klist_init(&dev->klist_children); INIT_LIST_HEAD(&dev->dma_pools); init_MUTEX(&dev->sem); } @@ -249,10 +248,8 @@ int device_add(struct device *dev) goto PMError; if ((error = bus_add_device(dev))) goto BusError; - down_write(&devices_subsys.rwsem); if (parent) - list_add_tail(&dev->node, &parent->children); - up_write(&devices_subsys.rwsem); + klist_add_tail(&parent->klist_children, &dev->knode_parent); /* notify platform of device entry */ if (platform_notify) @@ -335,10 +332,8 @@ void device_del(struct device * dev) { struct device * parent = dev->parent; - down_write(&devices_subsys.rwsem); if (parent) - list_del_init(&dev->node); - up_write(&devices_subsys.rwsem); + klist_remove(&dev->knode_parent); /* Notify the platform of the removal, in case they * need to do anything... @@ -372,6 +367,12 @@ void device_unregister(struct device * dev) } +static struct device * next_device(struct klist_iter * i) +{ + struct klist_node * n = klist_next(i); + return n ? container_of(n, struct device, knode_parent) : NULL; +} + /** * device_for_each_child - device child iterator. * @dev: parent struct device. @@ -384,18 +385,17 @@ void device_unregister(struct device * dev) * We check the return of @fn each time. If it returns anything * other than 0, we break out and return that value. */ -int device_for_each_child(struct device * dev, void * data, +int device_for_each_child(struct device * parent, void * data, int (*fn)(struct device *, void *)) { + struct klist_iter i; struct device * child; int error = 0; - down_read(&devices_subsys.rwsem); - list_for_each_entry(child, &dev->children, node) { - if((error = fn(child, data))) - break; - } - up_read(&devices_subsys.rwsem); + klist_iter_init(&parent->klist_children, &i); + while ((child = next_device(&i)) && !error) + error = fn(child, data); + klist_iter_exit(&i); return error; } diff --git a/include/linux/device.h b/include/linux/device.h index d2434934d091..43249260cd1c 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -262,8 +262,8 @@ extern void class_device_destroy(struct class *cls, dev_t devt); struct device { - struct list_head node; /* node in sibling list */ - struct list_head children; + struct klist klist_children; + struct klist_node knode_parent; /* node in sibling list */ struct klist_node knode_driver; struct klist_node knode_bus; struct device * parent; @@ -298,12 +298,6 @@ struct device { void (*release)(struct device * dev); }; -static inline struct device * -list_to_dev(struct list_head *node) -{ - return list_entry(node, struct device, node); -} - static inline void * dev_get_drvdata (struct device *dev) { -- cgit v1.2.3 From 0d3e5a2e39b6ba2974e9e7c2a429018c45de8e76 Mon Sep 17 00:00:00 2001 From: Patrick Mochel Date: Tue, 5 Apr 2005 23:46:33 -0700 Subject: [PATCH] Driver Core: fix bk-driver-core kills ppc64 There's no check to see if the device is already bound to a driver, which could do bad things. The first thing to go wrong is that it will try to match a driver with a device already bound to one. In some cases (it appears with USB with drivers/usb/core/usb.c::usb_match_id()), some drivers will match a device based on the class type, so it would be common (especially for HID devices) to match a device that is already bound. The fun comes when ->probe() is called, it fails, then driver_probe_device() does this: dev->driver = NULL; Later on, that pointer could be be dereferenced without checking and cause hell to break loose. This problem could be nasty. It's very hardware dependent, since some devices could have a different set of matching qualifiers than others. Now, I don't quite see exactly where/how you were getting that crash. You're dereferencing bad memory, but I'm not sure which pointer was bad and where it came from, but it could have come from a couple of different places. The patch below will hopefully fix it all up for you. It's against 2.6.12-rc2-mm1, and does the following: - Move logic to driver_probe_device() and comments uncommon returns: 1 - If device is bound 0 - If device not bound, and no error error - If there was an error. - Move locking to caller of that function, since we want to lock a device for the entire time we're trying to bind it to a driver (to prevent against a driver being loaded at the same time). - Update __device_attach() and __driver_attach() to do that locking. - Check if device is already bound in __driver_attach() - Update the converse device_release_driver() so it locks the device around all of the operations. - Mark driver_probe_device() as static and remove export. It's an internal function, it should stay that way, and there are no other callers. If there is ever a need to export it, we can audit it as necessary. Signed-off-by: Andrew Morton --- drivers/base/dd.c | 142 ++++++++++++++++++++++++++----------------------- include/linux/device.h | 1 - 2 files changed, 74 insertions(+), 69 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index dd2a8a79c121..8510918109e0 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -35,6 +35,8 @@ * nor take the bus's rwsem. Please verify those are accounted * for before calling this. (It is ok to call with no other effort * from a driver's probe() method.) + * + * This function must be called with @dev->sem held. */ void device_bind_driver(struct device * dev) { @@ -57,54 +59,56 @@ void device_bind_driver(struct device * dev) * because we don't know the format of the ID structures, nor what * is to be considered a match and what is not. * - * If we find a match, we call @drv->probe(@dev) if it exists, and - * call device_bind_driver() above. + * + * This function returns 1 if a match is found, an error if one + * occurs (that is not -ENODEV or -ENXIO), and 0 otherwise. + * + * This function must be called with @dev->sem held. */ -int driver_probe_device(struct device_driver * drv, struct device * dev) +static int driver_probe_device(struct device_driver * drv, struct device * dev) { - int error = 0; + int ret = 0; if (drv->bus->match && !drv->bus->match(dev, drv)) - return -ENODEV; + goto Done; - down(&dev->sem); + pr_debug("%s: Matched Device %s with Driver %s\n", + drv->bus->name, dev->bus_id, drv->name); dev->driver = drv; if (drv->probe) { - error = drv->probe(dev); - if (error) { + ret = drv->probe(dev); + if (ret) { dev->driver = NULL; - up(&dev->sem); - return error; + goto ProbeFailed; } } - up(&dev->sem); device_bind_driver(dev); - return 0; + ret = 1; + pr_debug("%s: Bound Device %s to Driver %s\n", + drv->bus->name, dev->bus_id, drv->name); + goto Done; + + ProbeFailed: + if (ret == -ENODEV || ret == -ENXIO) { + /* Driver matched, but didn't support device + * or device not found. + * Not an error; keep going. + */ + ret = 0; + } else { + /* driver matched but the probe failed */ + printk(KERN_WARNING + "%s: probe of %s failed with error %d\n", + drv->name, dev->bus_id, ret); + } + Done: + return ret; } static int __device_attach(struct device_driver * drv, void * data) { struct device * dev = data; - int error; - - error = driver_probe_device(drv, dev); - if (error) { - if ((error == -ENODEV) || (error == -ENXIO)) { - /* Driver matched, but didn't support device - * or device not found. - * Not an error; keep going. - */ - error = 0; - } else { - /* driver matched but the probe failed */ - printk(KERN_WARNING - "%s: probe of %s failed with error %d\n", - drv->name, dev->bus_id, error); - } - return error; - } - /* stop looking, this device is attached */ - return 1; + return driver_probe_device(drv, dev); } /** @@ -114,37 +118,43 @@ static int __device_attach(struct device_driver * drv, void * data) * Walk the list of drivers that the bus has and call * driver_probe_device() for each pair. If a compatible * pair is found, break out and return. + * + * Returns 1 if the device was bound to a driver; 0 otherwise. */ int device_attach(struct device * dev) { + int ret = 0; + + down(&dev->sem); if (dev->driver) { device_bind_driver(dev); - return 1; - } - - return bus_for_each_drv(dev->bus, NULL, dev, __device_attach); + ret = 1; + } else + ret = bus_for_each_drv(dev->bus, NULL, dev, __device_attach); + up(&dev->sem); + return ret; } static int __driver_attach(struct device * dev, void * data) { struct device_driver * drv = data; - int error = 0; - - if (!dev->driver) { - error = driver_probe_device(drv, dev); - if (error) { - if (error != -ENODEV) { - /* driver matched but the probe failed */ - printk(KERN_WARNING - "%s: probe of %s failed with error %d\n", - drv->name, dev->bus_id, error); - } else - error = 0; - return error; - } - /* stop looking, this driver is attached */ - return 1; - } + + /* + * Lock device and try to bind to it. We drop the error + * here and always return 0, because we need to keep trying + * to bind to devices and some drivers will return an error + * simply if it didn't support the device. + * + * driver_probe_device() will spit a warning if there + * is an error. + */ + + down(&dev->sem); + if (!dev->driver) + driver_probe_device(drv, dev); + up(&dev->sem); + + return 0; } @@ -156,9 +166,6 @@ static int __driver_attach(struct device * dev, void * data) * match the driver with each one. If driver_probe_device() * returns 0 and the @dev->driver is set, we've found a * compatible pair. - * - * Note that we ignore the -ENODEV error from driver_probe_device(), - * since it's perfectly valid for a driver not to bind to any devices. */ void driver_attach(struct device_driver * drv) { @@ -176,19 +183,19 @@ void driver_attach(struct device_driver * drv) */ void device_release_driver(struct device * dev) { - struct device_driver * drv = dev->driver; - - if (!drv) - return; - - sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); - sysfs_remove_link(&dev->kobj, "driver"); - klist_del(&dev->knode_driver); + struct device_driver * drv; down(&dev->sem); - if (drv->remove) - drv->remove(dev); - dev->driver = NULL; + if (dev->driver) { + drv = dev->driver; + sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); + sysfs_remove_link(&dev->kobj, "driver"); + klist_del(&dev->knode_driver); + + if (drv->remove) + drv->remove(dev); + dev->driver = NULL; + } up(&dev->sem); } @@ -208,7 +215,6 @@ void driver_detach(struct device_driver * drv) } -EXPORT_SYMBOL_GPL(driver_probe_device); EXPORT_SYMBOL_GPL(device_bind_driver); EXPORT_SYMBOL_GPL(device_release_driver); EXPORT_SYMBOL_GPL(device_attach); diff --git a/include/linux/device.h b/include/linux/device.h index 43249260cd1c..91aac349b9a7 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -325,7 +325,6 @@ extern int device_for_each_child(struct device *, void *, * Manual binding of a device to driver. See drivers/base/bus.c * for information on use. */ -extern int driver_probe_device(struct device_driver * drv, struct device * dev); extern void device_bind_driver(struct device * dev); extern void device_release_driver(struct device * dev); extern int device_attach(struct device * dev); -- cgit v1.2.3 From 4b45099b75832434c5113b9aed1499f8a69d13d5 Mon Sep 17 00:00:00 2001 From: Keiichiro Tokunaga Date: Sun, 8 May 2005 21:28:53 +0900 Subject: [PATCH] Driver core: unregister_node() for hotplug use This adds a generic function 'unregister_node()'. It is used to remove objects of a node going away for hotplug. All the devices on the node must be unregistered before calling this function. Signed-off-by: Keiichiro Tokunaga Signed-off-by: Greg Kroah-Hartman diff -puN drivers/base/node.c~numa_hp_base drivers/base/node.c --- drivers/base/node.c | 20 ++++++++++++++++++-- include/linux/node.h | 1 + 2 files changed, 19 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/node.c b/drivers/base/node.c index 583d57ec49a8..5d4517ccc422 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -136,7 +136,7 @@ static SYSDEV_ATTR(distance, S_IRUGO, node_read_distance, NULL); * * Initialize and register the node device. */ -int __init register_node(struct node *node, int num, struct node *parent) +int register_node(struct node *node, int num, struct node *parent) { int error; @@ -153,8 +153,24 @@ int __init register_node(struct node *node, int num, struct node *parent) return error; } +/** + * unregister_node - unregister a node device + * @node: node going away + * + * Unregisters a node device @node. All the devices on the node must be + * unregistered before calling this function. + */ +void unregister_node(struct node *node) +{ + sysdev_remove_file(&node->sysdev, &attr_cpumap); + sysdev_remove_file(&node->sysdev, &attr_meminfo); + sysdev_remove_file(&node->sysdev, &attr_numastat); + sysdev_remove_file(&node->sysdev, &attr_distance); + + sysdev_unregister(&node->sysdev); +} -int __init register_node_type(void) +static int __init register_node_type(void) { return sysdev_class_register(&node_class); } diff --git a/include/linux/node.h b/include/linux/node.h index 6e0a697e594e..254dc3de650b 100644 --- a/include/linux/node.h +++ b/include/linux/node.h @@ -27,6 +27,7 @@ struct node { }; extern int register_node(struct node *, int, struct node *); +extern void unregister_node(struct node *node); #define to_node(sys_device) container_of(sys_device, struct node, sysdev) -- cgit v1.2.3 From acaefc25d21f850e47ecc5098d1e0bc442c526be Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 18 May 2005 14:40:59 +0200 Subject: [PATCH] libfs: add simple attribute files Based on the discussion about spufs attributes, this is my suggestion for a more generic attribute file support that can be used by both debugfs and spufs. Simple attribute files behave similarly to sequential files from a kernel programmers perspective in that a standard set of file operations is provided and only an open operation needs to be written that registers file specific get() and set() functions. These operations are defined as void foo_set(void *data, u64 val); and u64 foo_get(void *data); where data is the inode->u.generic_ip pointer of the file and the operations just need to make send of that pointer. The infrastructure makes sure this works correctly with concurrent access and partial read calls. A macro named DEFINE_SIMPLE_ATTRIBUTE is provided to further simplify using the attributes. This patch already contains the changes for debugfs to use attributes for its internal file operations. Signed-off-by: Arnd Bergmann Signed-off-by: Greg Kroah-Hartman --- fs/debugfs/file.c | 67 ++++++++++++++++------------------- fs/libfs.c | 100 +++++++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/fs.h | 46 ++++++++++++++++++++++++ 3 files changed, 175 insertions(+), 38 deletions(-) (limited to 'include/linux') diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 548556ff2506..efc97d9b7860 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -45,44 +45,15 @@ struct file_operations debugfs_file_operations = { .open = default_open, }; -#define simple_type(type, format, temptype, strtolfn) \ -static ssize_t read_file_##type(struct file *file, char __user *user_buf, \ - size_t count, loff_t *ppos) \ -{ \ - char buf[32]; \ - type *val = file->private_data; \ - \ - snprintf(buf, sizeof(buf), format "\n", *val); \ - return simple_read_from_buffer(user_buf, count, ppos, buf, strlen(buf));\ -} \ -static ssize_t write_file_##type(struct file *file, const char __user *user_buf,\ - size_t count, loff_t *ppos) \ -{ \ - char *endp; \ - char buf[32]; \ - int buf_size; \ - type *val = file->private_data; \ - temptype tmp; \ - \ - memset(buf, 0x00, sizeof(buf)); \ - buf_size = min(count, (sizeof(buf)-1)); \ - if (copy_from_user(buf, user_buf, buf_size)) \ - return -EFAULT; \ - \ - tmp = strtolfn(buf, &endp, 0); \ - if ((endp == buf) || ((type)tmp != tmp)) \ - return -EINVAL; \ - *val = tmp; \ - return count; \ -} \ -static struct file_operations fops_##type = { \ - .read = read_file_##type, \ - .write = write_file_##type, \ - .open = default_open, \ -}; -simple_type(u8, "%c", unsigned long, simple_strtoul); -simple_type(u16, "%hi", unsigned long, simple_strtoul); -simple_type(u32, "%i", unsigned long, simple_strtoul); +static void debugfs_u8_set(void *data, u64 val) +{ + *(u8 *)data = val; +} +static u64 debugfs_u8_get(void *data) +{ + return *(u8 *)data; +} +DEFINE_SIMPLE_ATTRIBUTE(fops_u8, debugfs_u8_get, debugfs_u8_set, "%llu\n"); /** * debugfs_create_u8 - create a file in the debugfs filesystem that is used to read and write a unsigned 8 bit value. @@ -116,6 +87,16 @@ struct dentry *debugfs_create_u8(const char *name, mode_t mode, } EXPORT_SYMBOL_GPL(debugfs_create_u8); +static void debugfs_u16_set(void *data, u64 val) +{ + *(u16 *)data = val; +} +static u64 debugfs_u16_get(void *data) +{ + return *(u16 *)data; +} +DEFINE_SIMPLE_ATTRIBUTE(fops_u16, debugfs_u16_get, debugfs_u16_set, "%llu\n"); + /** * debugfs_create_u16 - create a file in the debugfs filesystem that is used to read and write a unsigned 8 bit value. * @@ -148,6 +129,16 @@ struct dentry *debugfs_create_u16(const char *name, mode_t mode, } EXPORT_SYMBOL_GPL(debugfs_create_u16); +static void debugfs_u32_set(void *data, u64 val) +{ + *(u32 *)data = val; +} +static u64 debugfs_u32_get(void *data) +{ + return *(u32 *)data; +} +DEFINE_SIMPLE_ATTRIBUTE(fops_u32, debugfs_u32_get, debugfs_u32_set, "%llu\n"); + /** * debugfs_create_u32 - create a file in the debugfs filesystem that is used to read and write a unsigned 8 bit value. * diff --git a/fs/libfs.c b/fs/libfs.c index f90b29595927..5025563e7379 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -519,6 +519,102 @@ int simple_transaction_release(struct inode *inode, struct file *file) return 0; } +/* Simple attribute files */ + +struct simple_attr { + u64 (*get)(void *); + void (*set)(void *, u64); + char get_buf[24]; /* enough to store a u64 and "\n\0" */ + char set_buf[24]; + void *data; + const char *fmt; /* format for read operation */ + struct semaphore sem; /* protects access to these buffers */ +}; + +/* simple_attr_open is called by an actual attribute open file operation + * to set the attribute specific access operations. */ +int simple_attr_open(struct inode *inode, struct file *file, + u64 (*get)(void *), void (*set)(void *, u64), + const char *fmt) +{ + struct simple_attr *attr; + + attr = kmalloc(sizeof(*attr), GFP_KERNEL); + if (!attr) + return -ENOMEM; + + attr->get = get; + attr->set = set; + attr->data = inode->u.generic_ip; + attr->fmt = fmt; + init_MUTEX(&attr->sem); + + file->private_data = attr; + + return nonseekable_open(inode, file); +} + +int simple_attr_close(struct inode *inode, struct file *file) +{ + kfree(file->private_data); + return 0; +} + +/* read from the buffer that is filled with the get function */ +ssize_t simple_attr_read(struct file *file, char __user *buf, + size_t len, loff_t *ppos) +{ + struct simple_attr *attr; + size_t size; + ssize_t ret; + + attr = file->private_data; + + if (!attr->get) + return -EACCES; + + down(&attr->sem); + if (*ppos) /* continued read */ + size = strlen(attr->get_buf); + else /* first read */ + size = scnprintf(attr->get_buf, sizeof(attr->get_buf), + attr->fmt, + (unsigned long long)attr->get(attr->data)); + + ret = simple_read_from_buffer(buf, len, ppos, attr->get_buf, size); + up(&attr->sem); + return ret; +} + +/* interpret the buffer as a number to call the set function with */ +ssize_t simple_attr_write(struct file *file, const char __user *buf, + size_t len, loff_t *ppos) +{ + struct simple_attr *attr; + u64 val; + size_t size; + ssize_t ret; + + attr = file->private_data; + + if (!attr->set) + return -EACCES; + + down(&attr->sem); + ret = -EFAULT; + size = min(sizeof(attr->set_buf) - 1, len); + if (copy_from_user(attr->set_buf, buf, size)) + goto out; + + ret = len; /* claim we got the whole input */ + attr->set_buf[size] = '\0'; + val = simple_strtol(attr->set_buf, NULL, 0); + attr->set(attr->data, val); +out: + up(&attr->sem); + return ret; +} + EXPORT_SYMBOL(dcache_dir_close); EXPORT_SYMBOL(dcache_dir_lseek); EXPORT_SYMBOL(dcache_dir_open); @@ -547,3 +643,7 @@ EXPORT_SYMBOL(simple_read_from_buffer); EXPORT_SYMBOL(simple_transaction_get); EXPORT_SYMBOL(simple_transaction_read); EXPORT_SYMBOL(simple_transaction_release); +EXPORT_SYMBOL_GPL(simple_attr_open); +EXPORT_SYMBOL_GPL(simple_attr_close); +EXPORT_SYMBOL_GPL(simple_attr_read); +EXPORT_SYMBOL_GPL(simple_attr_write); diff --git a/include/linux/fs.h b/include/linux/fs.h index 0180102dace1..9b8b696d4f15 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1657,6 +1657,52 @@ static inline void simple_transaction_set(struct file *file, size_t n) ar->size = n; } +/* + * simple attribute files + * + * These attributes behave similar to those in sysfs: + * + * Writing to an attribute immediately sets a value, an open file can be + * written to multiple times. + * + * Reading from an attribute creates a buffer from the value that might get + * read with multiple read calls. When the attribute has been read + * completely, no further read calls are possible until the file is opened + * again. + * + * All attributes contain a text representation of a numeric value + * that are accessed with the get() and set() functions. + */ +#define DEFINE_SIMPLE_ATTRIBUTE(__fops, __get, __set, __fmt) \ +static int __fops ## _open(struct inode *inode, struct file *file) \ +{ \ + __simple_attr_check_format(__fmt, 0ull); \ + return simple_attr_open(inode, file, __get, __set, __fmt); \ +} \ +static struct file_operations __fops = { \ + .owner = THIS_MODULE, \ + .open = __fops ## _open, \ + .release = simple_attr_close, \ + .read = simple_attr_read, \ + .write = simple_attr_write, \ +}; + +static inline void __attribute__((format(printf, 1, 2))) +__simple_attr_check_format(const char *fmt, ...) +{ + /* don't do anything, just let the compiler check the arguments; */ +} + +int simple_attr_open(struct inode *inode, struct file *file, + u64 (*get)(void *), void (*set)(void *, u64), + const char *fmt); +int simple_attr_close(struct inode *inode, struct file *file); +ssize_t simple_attr_read(struct file *file, char __user *buf, + size_t len, loff_t *ppos); +ssize_t simple_attr_write(struct file *file, const char __user *buf, + size_t len, loff_t *ppos); + + #ifdef CONFIG_SECURITY static inline char *alloc_secdata(void) { -- cgit v1.2.3 From 54b6f35c99974e99e64c05c2895718355123c55f Mon Sep 17 00:00:00 2001 From: Yani Ioannou Date: Tue, 17 May 2005 06:39:34 -0400 Subject: [PATCH] Driver core: change device_attribute callbacks This patch adds the device_attribute paramerter to the device_attribute store and show sysfs callback functions, and passes a reference to the attribute when the callbacks are called. Signed-off-by: Yani Ioannou Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 4 ++-- include/linux/device.h | 6 ++++-- 2 files changed, 6 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/core.c b/drivers/base/core.c index 1d8c7790b55a..86d79755fbfb 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -39,7 +39,7 @@ dev_attr_show(struct kobject * kobj, struct attribute * attr, char * buf) ssize_t ret = -EIO; if (dev_attr->show) - ret = dev_attr->show(dev, buf); + ret = dev_attr->show(dev, dev_attr, buf); return ret; } @@ -52,7 +52,7 @@ dev_attr_store(struct kobject * kobj, struct attribute * attr, ssize_t ret = -EIO; if (dev_attr->store) - ret = dev_attr->store(dev, buf, count); + ret = dev_attr->store(dev, dev_attr, buf, count); return ret; } diff --git a/include/linux/device.h b/include/linux/device.h index 91aac349b9a7..7b781a72b293 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -335,8 +335,10 @@ extern void driver_attach(struct device_driver * drv); struct device_attribute { struct attribute attr; - ssize_t (*show)(struct device * dev, char * buf); - ssize_t (*store)(struct device * dev, const char * buf, size_t count); + ssize_t (*show)(struct device *dev, struct device_attribute *attr, + char *buf); + ssize_t (*store)(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count); }; #define DEVICE_ATTR(_name,_mode,_show,_store) \ -- cgit v1.2.3 From 0a3e7eeabc9f76b7496488aad2d11626ff6a2a4f Mon Sep 17 00:00:00 2001 From: Yani Ioannou Date: Tue, 17 May 2005 22:59:05 -0400 Subject: [PATCH] I2C: add i2c sensor_device_attribute and macros This patch creates a new header with a potential standard i2c sensor attribute type (which simply includes an int representing the sensor number/index) and the associated macros, SENSOR_DEVICE_ATTR to define a static attribute and to_sensor_dev_attr to get a sensor_device_attribute reference from an embedded device_attribute reference. Signed-off-by: Yani Ioannou --- include/linux/i2c-sysfs.h | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) create mode 100644 include/linux/i2c-sysfs.h (limited to 'include/linux') diff --git a/include/linux/i2c-sysfs.h b/include/linux/i2c-sysfs.h new file mode 100644 index 000000000000..d7bf6ce11679 --- /dev/null +++ b/include/linux/i2c-sysfs.h @@ -0,0 +1,36 @@ +/* + * i2c-sysfs.h - i2c chip driver sysfs defines + * + * Copyright (C) 2005 Yani Ioannou + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ +#ifndef _LINUX_I2C_SYSFS_H +#define _LINUX_I2C_SYSFS_H + +struct sensor_device_attribute{ + struct device_attribute dev_attr; + int index; +}; +#define to_sensor_dev_attr(_dev_attr) \ + container_of(_dev_attr, struct sensor_device_attribute, dev_attr) + +#define SENSOR_DEVICE_ATTR(_name,_mode,_show,_store,_index) \ +struct sensor_device_attribute sensor_dev_attr_##_name = { \ + .dev_attr = __ATTR(_name,_mode,_show,_store), \ + .index = _index, \ +} + +#endif /* _LINUX_I2C_SYSFS_H */ -- cgit v1.2.3 From 988d186de5b6966a71a8cc52e6cb4895fd2f7799 Mon Sep 17 00:00:00 2001 From: Maneesh Soni Date: Tue, 31 May 2005 10:39:14 +0530 Subject: [PATCH] sysfs-iattr: add sysfs_setattr o This adds ->i_op->setattr VFS method for sysfs inodes. The changed attribues are saved in the persistent sysfs_dirent structure as a pointer to struct iattr. The struct iattr is allocated only for those sysfs_dirent's for which default attributes are getting changed. Thanks to Jon Smirl for this suggestion. Signed-off-by: Maneesh Soni Signed-off-by: Greg Kroah-Hartman --- fs/sysfs/dir.c | 1 + fs/sysfs/inode.c | 65 +++++++++++++++++++++++++++++++++++++++++++++++++++ fs/sysfs/sysfs.h | 2 ++ include/linux/sysfs.h | 1 + 4 files changed, 69 insertions(+) (limited to 'include/linux') diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 5cf3270014c0..37d7a6875d86 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -233,6 +233,7 @@ static struct dentry * sysfs_lookup(struct inode *dir, struct dentry *dentry, struct inode_operations sysfs_dir_inode_operations = { .lookup = sysfs_lookup, + .setattr = sysfs_setattr, }; static void remove_dir(struct dentry * d) diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index aff7b2dfa8ee..400c90be5685 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -26,6 +26,71 @@ static struct backing_dev_info sysfs_backing_dev_info = { .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, }; +static struct inode_operations sysfs_inode_operations ={ + .setattr = sysfs_setattr, +}; + +int sysfs_setattr(struct dentry * dentry, struct iattr * iattr) +{ + struct inode * inode = dentry->d_inode; + struct sysfs_dirent * sd = dentry->d_fsdata; + struct iattr * sd_iattr; + unsigned int ia_valid = iattr->ia_valid; + int error; + + if (!sd) + return -EINVAL; + + sd_iattr = sd->s_iattr; + + error = inode_change_ok(inode, iattr); + if (error) + return error; + + error = inode_setattr(inode, iattr); + if (error) + return error; + + if (!sd_iattr) { + /* setting attributes for the first time, allocate now */ + sd_iattr = kmalloc(sizeof(struct iattr), GFP_KERNEL); + if (!sd_iattr) + return -ENOMEM; + /* assign default attributes */ + memset(sd_iattr, 0, sizeof(struct iattr)); + sd_iattr->ia_mode = sd->s_mode; + sd_iattr->ia_uid = 0; + sd_iattr->ia_gid = 0; + sd_iattr->ia_atime = sd_iattr->ia_mtime = sd_iattr->ia_ctime = CURRENT_TIME; + sd->s_iattr = sd_iattr; + } + + /* attributes were changed atleast once in past */ + + if (ia_valid & ATTR_UID) + sd_iattr->ia_uid = iattr->ia_uid; + if (ia_valid & ATTR_GID) + sd_iattr->ia_gid = iattr->ia_gid; + if (ia_valid & ATTR_ATIME) + sd_iattr->ia_atime = timespec_trunc(iattr->ia_atime, + inode->i_sb->s_time_gran); + if (ia_valid & ATTR_MTIME) + sd_iattr->ia_mtime = timespec_trunc(iattr->ia_mtime, + inode->i_sb->s_time_gran); + if (ia_valid & ATTR_CTIME) + sd_iattr->ia_ctime = timespec_trunc(iattr->ia_ctime, + inode->i_sb->s_time_gran); + if (ia_valid & ATTR_MODE) { + umode_t mode = iattr->ia_mode; + + if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID)) + mode &= ~S_ISGID; + sd_iattr->ia_mode = mode; + } + + return error; +} + struct inode * sysfs_new_inode(mode_t mode) { struct inode * inode = new_inode(sysfs_sb); diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index a8a24a0c0b3b..e8e9f0c2573e 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -17,6 +17,7 @@ extern void sysfs_remove_subdir(struct dentry *); extern const unsigned char * sysfs_get_name(struct sysfs_dirent *sd); extern void sysfs_drop_dentry(struct sysfs_dirent *sd, struct dentry *parent); +extern int sysfs_setattr(struct dentry *dentry, struct iattr *iattr); extern struct rw_semaphore sysfs_rename_sem; extern struct super_block * sysfs_sb; @@ -75,6 +76,7 @@ static inline void release_sysfs_dirent(struct sysfs_dirent * sd) kobject_put(sl->target_kobj); kfree(sl); } + kfree(sd->s_iattr); kmem_cache_free(sysfs_dir_cachep, sd); } diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index d9cd2d31d377..392da5a6dacb 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -73,6 +73,7 @@ struct sysfs_dirent { int s_type; umode_t s_mode; struct dentry * s_dentry; + struct iattr * s_iattr; }; #define SYSFS_ROOT 0x0001 -- cgit v1.2.3 From 18b504e25fd617bee8830d2cdcaff7fb7b5931bb Mon Sep 17 00:00:00 2001 From: Alexey Kuznetsov Date: Tue, 21 Jun 2005 12:38:48 -0700 Subject: [NETLINK]: netlink_callback structure needs 5 args not 4 net/ipv4/tcp_diag.c uses up to ->args[4] Signed-off-by: David S. Miller --- include/linux/netlink.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 561d4dc75836..3029cad63a01 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -147,7 +147,7 @@ struct netlink_callback int (*dump)(struct sk_buff * skb, struct netlink_callback *cb); int (*done)(struct netlink_callback *cb); int family; - long args[4]; + long args[5]; }; struct netlink_notify -- cgit v1.2.3 From e45b1be8bcb3643808975a426fa3e201a2588e87 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 21 Jun 2005 14:01:30 -0700 Subject: [NETFILTER]: Kill lockhelp.h Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_conntrack_core.h | 3 +- include/linux/netfilter_ipv4/ip_nat.h | 3 +- include/linux/netfilter_ipv4/listhelp.h | 1 - include/linux/netfilter_ipv4/lockhelp.h | 129 ----------------------- net/ipv4/netfilter/arp_tables.c | 1 - net/ipv4/netfilter/ip_conntrack_amanda.c | 7 +- net/ipv4/netfilter/ip_conntrack_core.c | 84 +++++++-------- net/ipv4/netfilter/ip_conntrack_ftp.c | 7 +- net/ipv4/netfilter/ip_conntrack_irc.c | 7 +- net/ipv4/netfilter/ip_conntrack_proto_sctp.c | 23 ++-- net/ipv4/netfilter/ip_conntrack_proto_tcp.c | 27 +++-- net/ipv4/netfilter/ip_conntrack_standalone.c | 22 ++-- net/ipv4/netfilter/ip_nat_core.c | 32 +++--- net/ipv4/netfilter/ip_nat_helper.c | 10 +- net/ipv4/netfilter/ip_nat_rule.c | 4 +- net/ipv4/netfilter/ip_nat_standalone.c | 5 +- net/ipv4/netfilter/ip_tables.c | 1 - net/ipv4/netfilter/ipt_CLUSTERIP.c | 49 ++++----- net/ipv4/netfilter/ipt_MASQUERADE.c | 10 +- net/ipv4/netfilter/ipt_ULOG.c | 15 ++- net/ipv4/netfilter/ipt_hashlimit.c | 17 ++- net/ipv4/netfilter/ipt_helper.c | 4 +- net/ipv6/netfilter/ip6_tables.c | 1 - 23 files changed, 160 insertions(+), 302 deletions(-) delete mode 100644 include/linux/netfilter_ipv4/lockhelp.h (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ip_conntrack_core.h b/include/linux/netfilter_ipv4/ip_conntrack_core.h index d84be02cb4fc..694aec9b4784 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_core.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_core.h @@ -1,7 +1,6 @@ #ifndef _IP_CONNTRACK_CORE_H #define _IP_CONNTRACK_CORE_H #include -#include /* This header is used to share core functionality between the standalone connection tracking module, and the compatibility layer's use @@ -47,6 +46,6 @@ static inline int ip_conntrack_confirm(struct sk_buff **pskb) extern struct list_head *ip_conntrack_hash; extern struct list_head ip_conntrack_expect_list; -DECLARE_RWLOCK_EXTERN(ip_conntrack_lock); +extern rwlock_t ip_conntrack_lock; #endif /* _IP_CONNTRACK_CORE_H */ diff --git a/include/linux/netfilter_ipv4/ip_nat.h b/include/linux/netfilter_ipv4/ip_nat.h index 2b72b86176f0..e201ec6e9905 100644 --- a/include/linux/netfilter_ipv4/ip_nat.h +++ b/include/linux/netfilter_ipv4/ip_nat.h @@ -50,10 +50,9 @@ struct ip_nat_multi_range_compat #ifdef __KERNEL__ #include -#include /* Protects NAT hash tables, and NAT-private part of conntracks. */ -DECLARE_RWLOCK_EXTERN(ip_nat_lock); +extern rwlock_t ip_nat_lock; /* The structure embedded in the conntrack structure. */ struct ip_nat_info diff --git a/include/linux/netfilter_ipv4/listhelp.h b/include/linux/netfilter_ipv4/listhelp.h index f2ae7c5e57bb..360429f48737 100644 --- a/include/linux/netfilter_ipv4/listhelp.h +++ b/include/linux/netfilter_ipv4/listhelp.h @@ -2,7 +2,6 @@ #define _LISTHELP_H #include #include -#include /* Header to do more comprehensive job than linux/list.h; assume list is first entry in structure. */ diff --git a/include/linux/netfilter_ipv4/lockhelp.h b/include/linux/netfilter_ipv4/lockhelp.h deleted file mode 100644 index a3288633ab46..000000000000 --- a/include/linux/netfilter_ipv4/lockhelp.h +++ /dev/null @@ -1,129 +0,0 @@ -#ifndef _LOCKHELP_H -#define _LOCKHELP_H -#include - -#include -#include -#include -#include - -/* Header to do help in lock debugging. */ - -#ifdef CONFIG_NETFILTER_DEBUG -struct spinlock_debug -{ - spinlock_t l; - atomic_t locked_by; -}; - -struct rwlock_debug -{ - rwlock_t l; - long read_locked_map; - long write_locked_map; -}; - -#define DECLARE_LOCK(l) \ -struct spinlock_debug l = { SPIN_LOCK_UNLOCKED, ATOMIC_INIT(-1) } -#define DECLARE_LOCK_EXTERN(l) \ -extern struct spinlock_debug l -#define DECLARE_RWLOCK(l) \ -struct rwlock_debug l = { RW_LOCK_UNLOCKED, 0, 0 } -#define DECLARE_RWLOCK_EXTERN(l) \ -extern struct rwlock_debug l - -#define MUST_BE_LOCKED(l) \ -do { if (atomic_read(&(l)->locked_by) != smp_processor_id()) \ - printk("ASSERT %s:%u %s unlocked\n", __FILE__, __LINE__, #l); \ -} while(0) - -#define MUST_BE_UNLOCKED(l) \ -do { if (atomic_read(&(l)->locked_by) == smp_processor_id()) \ - printk("ASSERT %s:%u %s locked\n", __FILE__, __LINE__, #l); \ -} while(0) - -/* Write locked OK as well. */ -#define MUST_BE_READ_LOCKED(l) \ -do { if (!((l)->read_locked_map & (1UL << smp_processor_id())) \ - && !((l)->write_locked_map & (1UL << smp_processor_id()))) \ - printk("ASSERT %s:%u %s not readlocked\n", __FILE__, __LINE__, #l); \ -} while(0) - -#define MUST_BE_WRITE_LOCKED(l) \ -do { if (!((l)->write_locked_map & (1UL << smp_processor_id()))) \ - printk("ASSERT %s:%u %s not writelocked\n", __FILE__, __LINE__, #l); \ -} while(0) - -#define MUST_BE_READ_WRITE_UNLOCKED(l) \ -do { if ((l)->read_locked_map & (1UL << smp_processor_id())) \ - printk("ASSERT %s:%u %s readlocked\n", __FILE__, __LINE__, #l); \ - else if ((l)->write_locked_map & (1UL << smp_processor_id())) \ - printk("ASSERT %s:%u %s writelocked\n", __FILE__, __LINE__, #l); \ -} while(0) - -#define LOCK_BH(lk) \ -do { \ - MUST_BE_UNLOCKED(lk); \ - spin_lock_bh(&(lk)->l); \ - atomic_set(&(lk)->locked_by, smp_processor_id()); \ -} while(0) - -#define UNLOCK_BH(lk) \ -do { \ - MUST_BE_LOCKED(lk); \ - atomic_set(&(lk)->locked_by, -1); \ - spin_unlock_bh(&(lk)->l); \ -} while(0) - -#define READ_LOCK(lk) \ -do { \ - MUST_BE_READ_WRITE_UNLOCKED(lk); \ - read_lock_bh(&(lk)->l); \ - set_bit(smp_processor_id(), &(lk)->read_locked_map); \ -} while(0) - -#define WRITE_LOCK(lk) \ -do { \ - MUST_BE_READ_WRITE_UNLOCKED(lk); \ - write_lock_bh(&(lk)->l); \ - set_bit(smp_processor_id(), &(lk)->write_locked_map); \ -} while(0) - -#define READ_UNLOCK(lk) \ -do { \ - if (!((lk)->read_locked_map & (1UL << smp_processor_id()))) \ - printk("ASSERT: %s:%u %s not readlocked\n", \ - __FILE__, __LINE__, #lk); \ - clear_bit(smp_processor_id(), &(lk)->read_locked_map); \ - read_unlock_bh(&(lk)->l); \ -} while(0) - -#define WRITE_UNLOCK(lk) \ -do { \ - MUST_BE_WRITE_LOCKED(lk); \ - clear_bit(smp_processor_id(), &(lk)->write_locked_map); \ - write_unlock_bh(&(lk)->l); \ -} while(0) - -#else -#define DECLARE_LOCK(l) spinlock_t l = SPIN_LOCK_UNLOCKED -#define DECLARE_LOCK_EXTERN(l) extern spinlock_t l -#define DECLARE_RWLOCK(l) rwlock_t l = RW_LOCK_UNLOCKED -#define DECLARE_RWLOCK_EXTERN(l) extern rwlock_t l - -#define MUST_BE_LOCKED(l) -#define MUST_BE_UNLOCKED(l) -#define MUST_BE_READ_LOCKED(l) -#define MUST_BE_WRITE_LOCKED(l) -#define MUST_BE_READ_WRITE_UNLOCKED(l) - -#define LOCK_BH(l) spin_lock_bh(l) -#define UNLOCK_BH(l) spin_unlock_bh(l) - -#define READ_LOCK(l) read_lock_bh(l) -#define WRITE_LOCK(l) write_lock_bh(l) -#define READ_UNLOCK(l) read_unlock_bh(l) -#define WRITE_UNLOCK(l) write_unlock_bh(l) -#endif /*CONFIG_NETFILTER_DEBUG*/ - -#endif /* _LOCKHELP_H */ diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index df79f5ed6a0a..fa1634256680 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -60,7 +60,6 @@ static DECLARE_MUTEX(arpt_mutex); #define ASSERT_READ_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0) #define ASSERT_WRITE_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0) -#include #include struct arpt_table_info { diff --git a/net/ipv4/netfilter/ip_conntrack_amanda.c b/net/ipv4/netfilter/ip_conntrack_amanda.c index 3dbddd062605..a78a320eee08 100644 --- a/net/ipv4/netfilter/ip_conntrack_amanda.c +++ b/net/ipv4/netfilter/ip_conntrack_amanda.c @@ -26,7 +26,6 @@ #include #include -#include #include #include @@ -42,7 +41,7 @@ static char *conns[] = { "DATA ", "MESG ", "INDEX " }; /* This is slow, but it's simple. --RR */ static char amanda_buffer[65536]; -static DECLARE_LOCK(amanda_buffer_lock); +static DEFINE_SPINLOCK(amanda_buffer_lock); unsigned int (*ip_nat_amanda_hook)(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, @@ -76,7 +75,7 @@ static int help(struct sk_buff **pskb, return NF_ACCEPT; } - LOCK_BH(&amanda_buffer_lock); + spin_lock_bh(&amanda_buffer_lock); skb_copy_bits(*pskb, dataoff, amanda_buffer, (*pskb)->len - dataoff); data = amanda_buffer; data_limit = amanda_buffer + (*pskb)->len - dataoff; @@ -134,7 +133,7 @@ static int help(struct sk_buff **pskb, } out: - UNLOCK_BH(&amanda_buffer_lock); + spin_unlock_bh(&amanda_buffer_lock); return ret; } diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index 09e824622977..a7377a331ade 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -38,10 +38,10 @@ #include #include -/* This rwlock protects the main hash table, protocol/helper/expected +/* ip_conntrack_lock protects the main hash table, protocol/helper/expected registrations, conntrack timers*/ -#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_conntrack_lock) -#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_conntrack_lock) +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) #include #include @@ -57,7 +57,7 @@ #define DEBUGP(format, args...) #endif -DECLARE_RWLOCK(ip_conntrack_lock); +DEFINE_RWLOCK(ip_conntrack_lock); /* ip_conntrack_standalone needs this */ atomic_t ip_conntrack_count = ATOMIC_INIT(0); @@ -147,7 +147,7 @@ static void destroy_expect(struct ip_conntrack_expect *exp) static void unlink_expect(struct ip_conntrack_expect *exp) { - MUST_BE_WRITE_LOCKED(&ip_conntrack_lock); + ASSERT_WRITE_LOCK(&ip_conntrack_lock); list_del(&exp->list); /* Logically in destroy_expect, but we hold the lock here. */ exp->master->expecting--; @@ -157,9 +157,9 @@ static void expectation_timed_out(unsigned long ul_expect) { struct ip_conntrack_expect *exp = (void *)ul_expect; - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); unlink_expect(exp); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); destroy_expect(exp); } @@ -209,7 +209,7 @@ clean_from_lists(struct ip_conntrack *ct) unsigned int ho, hr; DEBUGP("clean_from_lists(%p)\n", ct); - MUST_BE_WRITE_LOCKED(&ip_conntrack_lock); + ASSERT_WRITE_LOCK(&ip_conntrack_lock); ho = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); hr = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); @@ -240,7 +240,7 @@ destroy_conntrack(struct nf_conntrack *nfct) if (ip_conntrack_destroyed) ip_conntrack_destroyed(ct); - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); /* Expectations will have been removed in clean_from_lists, * except TFTP can create an expectation on the first packet, * before connection is in the list, so we need to clean here, @@ -254,7 +254,7 @@ destroy_conntrack(struct nf_conntrack *nfct) } CONNTRACK_STAT_INC(delete); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); if (ct->master) ip_conntrack_put(ct->master); @@ -268,12 +268,12 @@ static void death_by_timeout(unsigned long ul_conntrack) { struct ip_conntrack *ct = (void *)ul_conntrack; - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); /* Inside lock so preempt is disabled on module removal path. * Otherwise we can get spurious warnings. */ CONNTRACK_STAT_INC(delete_list); clean_from_lists(ct); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); ip_conntrack_put(ct); } @@ -282,7 +282,7 @@ conntrack_tuple_cmp(const struct ip_conntrack_tuple_hash *i, const struct ip_conntrack_tuple *tuple, const struct ip_conntrack *ignored_conntrack) { - MUST_BE_READ_LOCKED(&ip_conntrack_lock); + ASSERT_READ_LOCK(&ip_conntrack_lock); return tuplehash_to_ctrack(i) != ignored_conntrack && ip_ct_tuple_equal(tuple, &i->tuple); } @@ -294,7 +294,7 @@ __ip_conntrack_find(const struct ip_conntrack_tuple *tuple, struct ip_conntrack_tuple_hash *h; unsigned int hash = hash_conntrack(tuple); - MUST_BE_READ_LOCKED(&ip_conntrack_lock); + ASSERT_READ_LOCK(&ip_conntrack_lock); list_for_each_entry(h, &ip_conntrack_hash[hash], list) { if (conntrack_tuple_cmp(h, tuple, ignored_conntrack)) { CONNTRACK_STAT_INC(found); @@ -313,11 +313,11 @@ ip_conntrack_find_get(const struct ip_conntrack_tuple *tuple, { struct ip_conntrack_tuple_hash *h; - READ_LOCK(&ip_conntrack_lock); + read_lock_bh(&ip_conntrack_lock); h = __ip_conntrack_find(tuple, ignored_conntrack); if (h) atomic_inc(&tuplehash_to_ctrack(h)->ct_general.use); - READ_UNLOCK(&ip_conntrack_lock); + read_unlock_bh(&ip_conntrack_lock); return h; } @@ -352,7 +352,7 @@ __ip_conntrack_confirm(struct sk_buff **pskb) IP_NF_ASSERT(!is_confirmed(ct)); DEBUGP("Confirming conntrack %p\n", ct); - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); /* See if there's one in the list already, including reverse: NAT could have grabbed it without realizing, since we're @@ -380,12 +380,12 @@ __ip_conntrack_confirm(struct sk_buff **pskb) atomic_inc(&ct->ct_general.use); set_bit(IPS_CONFIRMED_BIT, &ct->status); CONNTRACK_STAT_INC(insert); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); return NF_ACCEPT; } CONNTRACK_STAT_INC(insert_failed); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); return NF_DROP; } @@ -398,9 +398,9 @@ ip_conntrack_tuple_taken(const struct ip_conntrack_tuple *tuple, { struct ip_conntrack_tuple_hash *h; - READ_LOCK(&ip_conntrack_lock); + read_lock_bh(&ip_conntrack_lock); h = __ip_conntrack_find(tuple, ignored_conntrack); - READ_UNLOCK(&ip_conntrack_lock); + read_unlock_bh(&ip_conntrack_lock); return h != NULL; } @@ -419,13 +419,13 @@ static int early_drop(struct list_head *chain) struct ip_conntrack *ct = NULL; int dropped = 0; - READ_LOCK(&ip_conntrack_lock); + read_lock_bh(&ip_conntrack_lock); h = LIST_FIND_B(chain, unreplied, struct ip_conntrack_tuple_hash *); if (h) { ct = tuplehash_to_ctrack(h); atomic_inc(&ct->ct_general.use); } - READ_UNLOCK(&ip_conntrack_lock); + read_unlock_bh(&ip_conntrack_lock); if (!ct) return dropped; @@ -508,7 +508,7 @@ init_conntrack(const struct ip_conntrack_tuple *tuple, conntrack->timeout.data = (unsigned long)conntrack; conntrack->timeout.function = death_by_timeout; - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); exp = find_expectation(tuple); if (exp) { @@ -532,7 +532,7 @@ init_conntrack(const struct ip_conntrack_tuple *tuple, list_add(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL].list, &unconfirmed); atomic_inc(&ip_conntrack_count); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); if (exp) { if (exp->expectfn) @@ -723,17 +723,17 @@ void ip_conntrack_unexpect_related(struct ip_conntrack_expect *exp) { struct ip_conntrack_expect *i; - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); /* choose the the oldest expectation to evict */ list_for_each_entry_reverse(i, &ip_conntrack_expect_list, list) { if (expect_matches(i, exp) && del_timer(&i->timeout)) { unlink_expect(i); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); destroy_expect(i); return; } } - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); } struct ip_conntrack_expect *ip_conntrack_expect_alloc(void) @@ -808,7 +808,7 @@ int ip_conntrack_expect_related(struct ip_conntrack_expect *expect) DEBUGP("tuple: "); DUMP_TUPLE(&expect->tuple); DEBUGP("mask: "); DUMP_TUPLE(&expect->mask); - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); list_for_each_entry(i, &ip_conntrack_expect_list, list) { if (expect_matches(i, expect)) { /* Refresh timer: if it's dying, ignore.. */ @@ -832,7 +832,7 @@ int ip_conntrack_expect_related(struct ip_conntrack_expect *expect) ip_conntrack_expect_insert(expect); ret = 0; out: - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); return ret; } @@ -841,7 +841,7 @@ out: void ip_conntrack_alter_reply(struct ip_conntrack *conntrack, const struct ip_conntrack_tuple *newreply) { - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); /* Should be unconfirmed, so not in hash table yet */ IP_NF_ASSERT(!is_confirmed(conntrack)); @@ -851,15 +851,15 @@ void ip_conntrack_alter_reply(struct ip_conntrack *conntrack, conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; if (!conntrack->master && conntrack->expecting == 0) conntrack->helper = ip_ct_find_helper(newreply); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); } int ip_conntrack_helper_register(struct ip_conntrack_helper *me) { BUG_ON(me->timeout == 0); - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); list_prepend(&helpers, me); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); return 0; } @@ -878,7 +878,7 @@ void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me) struct ip_conntrack_expect *exp, *tmp; /* Need write lock here, to delete helper. */ - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); LIST_DELETE(&helpers, me); /* Get rid of expectations */ @@ -893,7 +893,7 @@ void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me) for (i = 0; i < ip_conntrack_htable_size; i++) LIST_FIND_W(&ip_conntrack_hash[i], unhelp, struct ip_conntrack_tuple_hash *, me); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); /* Someone could be still looking at the helper in a bh. */ synchronize_net(); @@ -925,14 +925,14 @@ void ip_ct_refresh_acct(struct ip_conntrack *ct, ct->timeout.expires = extra_jiffies; ct_add_counters(ct, ctinfo, skb); } else { - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); /* Need del_timer for race avoidance (may already be dying). */ if (del_timer(&ct->timeout)) { ct->timeout.expires = jiffies + extra_jiffies; add_timer(&ct->timeout); } ct_add_counters(ct, ctinfo, skb); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); } } @@ -997,7 +997,7 @@ get_next_corpse(int (*iter)(struct ip_conntrack *i, void *data), { struct ip_conntrack_tuple_hash *h = NULL; - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); for (; *bucket < ip_conntrack_htable_size; (*bucket)++) { h = LIST_FIND_W(&ip_conntrack_hash[*bucket], do_iter, struct ip_conntrack_tuple_hash *, iter, data); @@ -1009,7 +1009,7 @@ get_next_corpse(int (*iter)(struct ip_conntrack *i, void *data), struct ip_conntrack_tuple_hash *, iter, data); if (h) atomic_inc(&tuplehash_to_ctrack(h)->ct_general.use); - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); return h; } @@ -1201,14 +1201,14 @@ int __init ip_conntrack_init(void) } /* Don't NEED lock here, but good form anyway. */ - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); for (i = 0; i < MAX_IP_CT_PROTO; i++) ip_ct_protos[i] = &ip_conntrack_generic_protocol; /* Sew in builtin protocols. */ ip_ct_protos[IPPROTO_TCP] = &ip_conntrack_protocol_tcp; ip_ct_protos[IPPROTO_UDP] = &ip_conntrack_protocol_udp; ip_ct_protos[IPPROTO_ICMP] = &ip_conntrack_protocol_icmp; - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); for (i = 0; i < ip_conntrack_htable_size; i++) INIT_LIST_HEAD(&ip_conntrack_hash[i]); diff --git a/net/ipv4/netfilter/ip_conntrack_ftp.c b/net/ipv4/netfilter/ip_conntrack_ftp.c index dd86503aa788..fea6dd2a00b6 100644 --- a/net/ipv4/netfilter/ip_conntrack_ftp.c +++ b/net/ipv4/netfilter/ip_conntrack_ftp.c @@ -16,7 +16,6 @@ #include #include -#include #include #include #include @@ -28,7 +27,7 @@ MODULE_DESCRIPTION("ftp connection tracking helper"); /* This is slow, but it's simple. --RR */ static char ftp_buffer[65536]; -static DECLARE_LOCK(ip_ftp_lock); +static DEFINE_SPINLOCK(ip_ftp_lock); #define MAX_PORTS 8 static int ports[MAX_PORTS]; @@ -319,7 +318,7 @@ static int help(struct sk_buff **pskb, } datalen = (*pskb)->len - dataoff; - LOCK_BH(&ip_ftp_lock); + spin_lock_bh(&ip_ftp_lock); fb_ptr = skb_header_pointer(*pskb, dataoff, (*pskb)->len - dataoff, ftp_buffer); BUG_ON(fb_ptr == NULL); @@ -442,7 +441,7 @@ out_update_nl: if (ends_in_nl) update_nl_seq(seq, ct_ftp_info,dir); out: - UNLOCK_BH(&ip_ftp_lock); + spin_unlock_bh(&ip_ftp_lock); return ret; } diff --git a/net/ipv4/netfilter/ip_conntrack_irc.c b/net/ipv4/netfilter/ip_conntrack_irc.c index 33cc7348b6ee..cd98772cc332 100644 --- a/net/ipv4/netfilter/ip_conntrack_irc.c +++ b/net/ipv4/netfilter/ip_conntrack_irc.c @@ -29,7 +29,6 @@ #include #include -#include #include #include #include @@ -41,7 +40,7 @@ static int max_dcc_channels = 8; static unsigned int dcc_timeout = 300; /* This is slow, but it's simple. --RR */ static char irc_buffer[65536]; -static DECLARE_LOCK(irc_buffer_lock); +static DEFINE_SPINLOCK(irc_buffer_lock); unsigned int (*ip_nat_irc_hook)(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, @@ -141,7 +140,7 @@ static int help(struct sk_buff **pskb, if (dataoff >= (*pskb)->len) return NF_ACCEPT; - LOCK_BH(&irc_buffer_lock); + spin_lock_bh(&irc_buffer_lock); ib_ptr = skb_header_pointer(*pskb, dataoff, (*pskb)->len - dataoff, irc_buffer); BUG_ON(ib_ptr == NULL); @@ -237,7 +236,7 @@ static int help(struct sk_buff **pskb, } /* while data < ... */ out: - UNLOCK_BH(&irc_buffer_lock); + spin_unlock_bh(&irc_buffer_lock); return ret; } diff --git a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c index ff8c34a860ff..31d75390bf12 100644 --- a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c +++ b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c @@ -26,7 +26,6 @@ #include #include -#include #if 0 #define DEBUGP(format, ...) printk(format, ## __VA_ARGS__) @@ -35,7 +34,7 @@ #endif /* Protects conntrack->proto.sctp */ -static DECLARE_RWLOCK(sctp_lock); +static DEFINE_RWLOCK(sctp_lock); /* FIXME: Examine ipfilter's timeouts and conntrack transitions more closely. They're more complex. --RR @@ -199,9 +198,9 @@ static int sctp_print_conntrack(struct seq_file *s, DEBUGP(__FUNCTION__); DEBUGP("\n"); - READ_LOCK(&sctp_lock); + read_lock_bh(&sctp_lock); state = conntrack->proto.sctp.state; - READ_UNLOCK(&sctp_lock); + read_unlock_bh(&sctp_lock); return seq_printf(s, "%s ", sctp_conntrack_names[state]); } @@ -343,13 +342,13 @@ static int sctp_packet(struct ip_conntrack *conntrack, oldsctpstate = newconntrack = SCTP_CONNTRACK_MAX; for_each_sctp_chunk (skb, sch, _sch, offset, count) { - WRITE_LOCK(&sctp_lock); + write_lock_bh(&sctp_lock); /* Special cases of Verification tag check (Sec 8.5.1) */ if (sch->type == SCTP_CID_INIT) { /* Sec 8.5.1 (A) */ if (sh->vtag != 0) { - WRITE_UNLOCK(&sctp_lock); + write_unlock_bh(&sctp_lock); return -1; } } else if (sch->type == SCTP_CID_ABORT) { @@ -357,7 +356,7 @@ static int sctp_packet(struct ip_conntrack *conntrack, if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)]) && !(sh->vtag == conntrack->proto.sctp.vtag [1 - CTINFO2DIR(ctinfo)])) { - WRITE_UNLOCK(&sctp_lock); + write_unlock_bh(&sctp_lock); return -1; } } else if (sch->type == SCTP_CID_SHUTDOWN_COMPLETE) { @@ -366,13 +365,13 @@ static int sctp_packet(struct ip_conntrack *conntrack, && !(sh->vtag == conntrack->proto.sctp.vtag [1 - CTINFO2DIR(ctinfo)] && (sch->flags & 1))) { - WRITE_UNLOCK(&sctp_lock); + write_unlock_bh(&sctp_lock); return -1; } } else if (sch->type == SCTP_CID_COOKIE_ECHO) { /* Sec 8.5.1 (D) */ if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)])) { - WRITE_UNLOCK(&sctp_lock); + write_unlock_bh(&sctp_lock); return -1; } } @@ -384,7 +383,7 @@ static int sctp_packet(struct ip_conntrack *conntrack, if (newconntrack == SCTP_CONNTRACK_MAX) { DEBUGP("ip_conntrack_sctp: Invalid dir=%i ctype=%u conntrack=%u\n", CTINFO2DIR(ctinfo), sch->type, oldsctpstate); - WRITE_UNLOCK(&sctp_lock); + write_unlock_bh(&sctp_lock); return -1; } @@ -396,7 +395,7 @@ static int sctp_packet(struct ip_conntrack *conntrack, ih = skb_header_pointer(skb, offset + sizeof(sctp_chunkhdr_t), sizeof(_inithdr), &_inithdr); if (ih == NULL) { - WRITE_UNLOCK(&sctp_lock); + write_unlock_bh(&sctp_lock); return -1; } DEBUGP("Setting vtag %x for dir %d\n", @@ -405,7 +404,7 @@ static int sctp_packet(struct ip_conntrack *conntrack, } conntrack->proto.sctp.state = newconntrack; - WRITE_UNLOCK(&sctp_lock); + write_unlock_bh(&sctp_lock); } ip_ct_refresh_acct(conntrack, ctinfo, skb, *sctp_timeouts[newconntrack]); diff --git a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c index 721ddbf522b4..809dfed766d4 100644 --- a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c +++ b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c @@ -36,7 +36,6 @@ #include #include #include -#include #if 0 #define DEBUGP printk @@ -46,7 +45,7 @@ #endif /* Protects conntrack->proto.tcp */ -static DECLARE_RWLOCK(tcp_lock); +static DEFINE_RWLOCK(tcp_lock); /* "Be conservative in what you do, be liberal in what you accept from others." @@ -330,9 +329,9 @@ static int tcp_print_conntrack(struct seq_file *s, { enum tcp_conntrack state; - READ_LOCK(&tcp_lock); + read_lock_bh(&tcp_lock); state = conntrack->proto.tcp.state; - READ_UNLOCK(&tcp_lock); + read_unlock_bh(&tcp_lock); return seq_printf(s, "%s ", tcp_conntrack_names[state]); } @@ -738,14 +737,14 @@ void ip_conntrack_tcp_update(struct sk_buff *skb, end = segment_seq_plus_len(ntohl(tcph->seq), skb->len, iph, tcph); - WRITE_LOCK(&tcp_lock); + write_lock_bh(&tcp_lock); /* * We have to worry for the ack in the reply packet only... */ if (after(end, conntrack->proto.tcp.seen[dir].td_end)) conntrack->proto.tcp.seen[dir].td_end = end; conntrack->proto.tcp.last_end = end; - WRITE_UNLOCK(&tcp_lock); + write_unlock_bh(&tcp_lock); DEBUGP("tcp_update: sender end=%u maxend=%u maxwin=%u scale=%i " "receiver end=%u maxend=%u maxwin=%u scale=%i\n", sender->td_end, sender->td_maxend, sender->td_maxwin, @@ -857,7 +856,7 @@ static int tcp_packet(struct ip_conntrack *conntrack, sizeof(_tcph), &_tcph); BUG_ON(th == NULL); - WRITE_LOCK(&tcp_lock); + write_lock_bh(&tcp_lock); old_state = conntrack->proto.tcp.state; dir = CTINFO2DIR(ctinfo); index = get_conntrack_index(th); @@ -879,7 +878,7 @@ static int tcp_packet(struct ip_conntrack *conntrack, * that the client cannot but retransmit its SYN and * thus initiate a clean new session. */ - WRITE_UNLOCK(&tcp_lock); + write_unlock_bh(&tcp_lock); if (LOG_INVALID(IPPROTO_TCP)) nf_log_packet(PF_INET, 0, skb, NULL, NULL, "ip_ct_tcp: killing out of sync session "); @@ -894,7 +893,7 @@ static int tcp_packet(struct ip_conntrack *conntrack, conntrack->proto.tcp.last_end = segment_seq_plus_len(ntohl(th->seq), skb->len, iph, th); - WRITE_UNLOCK(&tcp_lock); + write_unlock_bh(&tcp_lock); if (LOG_INVALID(IPPROTO_TCP)) nf_log_packet(PF_INET, 0, skb, NULL, NULL, "ip_ct_tcp: invalid packet ignored "); @@ -904,7 +903,7 @@ static int tcp_packet(struct ip_conntrack *conntrack, DEBUGP("ip_ct_tcp: Invalid dir=%i index=%u ostate=%u\n", dir, get_conntrack_index(th), old_state); - WRITE_UNLOCK(&tcp_lock); + write_unlock_bh(&tcp_lock); if (LOG_INVALID(IPPROTO_TCP)) nf_log_packet(PF_INET, 0, skb, NULL, NULL, "ip_ct_tcp: invalid state "); @@ -918,13 +917,13 @@ static int tcp_packet(struct ip_conntrack *conntrack, conntrack->proto.tcp.seen[dir].td_end)) { /* Attempt to reopen a closed connection. * Delete this connection and look up again. */ - WRITE_UNLOCK(&tcp_lock); + write_unlock_bh(&tcp_lock); if (del_timer(&conntrack->timeout)) conntrack->timeout.function((unsigned long) conntrack); return -NF_REPEAT; } else { - WRITE_UNLOCK(&tcp_lock); + write_unlock_bh(&tcp_lock); if (LOG_INVALID(IPPROTO_TCP)) nf_log_packet(PF_INET, 0, skb, NULL, NULL, "ip_ct_tcp: invalid SYN"); @@ -949,7 +948,7 @@ static int tcp_packet(struct ip_conntrack *conntrack, if (!tcp_in_window(&conntrack->proto.tcp, dir, index, skb, iph, th)) { - WRITE_UNLOCK(&tcp_lock); + write_unlock_bh(&tcp_lock); return -NF_ACCEPT; } in_window: @@ -972,7 +971,7 @@ static int tcp_packet(struct ip_conntrack *conntrack, timeout = conntrack->proto.tcp.retrans >= ip_ct_tcp_max_retrans && *tcp_timeouts[new_state] > ip_ct_tcp_timeout_max_retrans ? ip_ct_tcp_timeout_max_retrans : *tcp_timeouts[new_state]; - WRITE_UNLOCK(&tcp_lock); + write_unlock_bh(&tcp_lock); if (!test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status)) { /* If only reply is a RST, we can consider ourselves not to diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c index bc59f7b39805..42dc95102873 100644 --- a/net/ipv4/netfilter/ip_conntrack_standalone.c +++ b/net/ipv4/netfilter/ip_conntrack_standalone.c @@ -28,8 +28,8 @@ #include #include -#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_conntrack_lock) -#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_conntrack_lock) +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) #include #include @@ -119,7 +119,7 @@ static struct list_head *ct_get_idx(struct seq_file *seq, loff_t pos) static void *ct_seq_start(struct seq_file *seq, loff_t *pos) { - READ_LOCK(&ip_conntrack_lock); + read_lock_bh(&ip_conntrack_lock); return ct_get_idx(seq, *pos); } @@ -131,7 +131,7 @@ static void *ct_seq_next(struct seq_file *s, void *v, loff_t *pos) static void ct_seq_stop(struct seq_file *s, void *v) { - READ_UNLOCK(&ip_conntrack_lock); + read_unlock_bh(&ip_conntrack_lock); } static int ct_seq_show(struct seq_file *s, void *v) @@ -140,7 +140,7 @@ static int ct_seq_show(struct seq_file *s, void *v) const struct ip_conntrack *conntrack = tuplehash_to_ctrack(hash); struct ip_conntrack_protocol *proto; - MUST_BE_READ_LOCKED(&ip_conntrack_lock); + ASSERT_READ_LOCK(&ip_conntrack_lock); IP_NF_ASSERT(conntrack); /* we only want to print DIR_ORIGINAL */ @@ -239,7 +239,7 @@ static void *exp_seq_start(struct seq_file *s, loff_t *pos) /* strange seq_file api calls stop even if we fail, * thus we need to grab lock since stop unlocks */ - READ_LOCK(&ip_conntrack_lock); + read_lock_bh(&ip_conntrack_lock); if (list_empty(e)) return NULL; @@ -267,7 +267,7 @@ static void *exp_seq_next(struct seq_file *s, void *v, loff_t *pos) static void exp_seq_stop(struct seq_file *s, void *v) { - READ_UNLOCK(&ip_conntrack_lock); + read_unlock_bh(&ip_conntrack_lock); } static int exp_seq_show(struct seq_file *s, void *v) @@ -921,22 +921,22 @@ int ip_conntrack_protocol_register(struct ip_conntrack_protocol *proto) { int ret = 0; - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); if (ip_ct_protos[proto->proto] != &ip_conntrack_generic_protocol) { ret = -EBUSY; goto out; } ip_ct_protos[proto->proto] = proto; out: - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); return ret; } void ip_conntrack_protocol_unregister(struct ip_conntrack_protocol *proto) { - WRITE_LOCK(&ip_conntrack_lock); + write_lock_bh(&ip_conntrack_lock); ip_ct_protos[proto->proto] = &ip_conntrack_generic_protocol; - WRITE_UNLOCK(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); /* Somebody could be still looking at the proto in bh. */ synchronize_net(); diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c index 9fc6f93af0dd..739b6dde1c82 100644 --- a/net/ipv4/netfilter/ip_nat_core.c +++ b/net/ipv4/netfilter/ip_nat_core.c @@ -22,8 +22,8 @@ #include #include -#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock) -#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock) +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) #include #include @@ -41,7 +41,7 @@ #define DEBUGP(format, args...) #endif -DECLARE_RWLOCK(ip_nat_lock); +DEFINE_RWLOCK(ip_nat_lock); /* Calculated at init based on memory size */ static unsigned int ip_nat_htable_size; @@ -65,9 +65,9 @@ static void ip_nat_cleanup_conntrack(struct ip_conntrack *conn) if (!(conn->status & IPS_NAT_DONE_MASK)) return; - WRITE_LOCK(&ip_nat_lock); + write_lock_bh(&ip_nat_lock); list_del(&conn->nat.info.bysource); - WRITE_UNLOCK(&ip_nat_lock); + write_unlock_bh(&ip_nat_lock); } /* We do checksum mangling, so if they were wrong before they're still @@ -142,7 +142,7 @@ find_appropriate_src(const struct ip_conntrack_tuple *tuple, unsigned int h = hash_by_src(tuple); struct ip_conntrack *ct; - READ_LOCK(&ip_nat_lock); + read_lock_bh(&ip_nat_lock); list_for_each_entry(ct, &bysource[h], nat.info.bysource) { if (same_src(ct, tuple)) { /* Copy source part from reply tuple. */ @@ -151,12 +151,12 @@ find_appropriate_src(const struct ip_conntrack_tuple *tuple, result->dst = tuple->dst; if (in_range(result, range)) { - READ_UNLOCK(&ip_nat_lock); + read_unlock_bh(&ip_nat_lock); return 1; } } } - READ_UNLOCK(&ip_nat_lock); + read_unlock_bh(&ip_nat_lock); return 0; } @@ -297,9 +297,9 @@ ip_nat_setup_info(struct ip_conntrack *conntrack, unsigned int srchash = hash_by_src(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL] .tuple); - WRITE_LOCK(&ip_nat_lock); + write_lock_bh(&ip_nat_lock); list_add(&info->bysource, &bysource[srchash]); - WRITE_UNLOCK(&ip_nat_lock); + write_unlock_bh(&ip_nat_lock); } /* It's done. */ @@ -474,23 +474,23 @@ int ip_nat_protocol_register(struct ip_nat_protocol *proto) { int ret = 0; - WRITE_LOCK(&ip_nat_lock); + write_lock_bh(&ip_nat_lock); if (ip_nat_protos[proto->protonum] != &ip_nat_unknown_protocol) { ret = -EBUSY; goto out; } ip_nat_protos[proto->protonum] = proto; out: - WRITE_UNLOCK(&ip_nat_lock); + write_unlock_bh(&ip_nat_lock); return ret; } /* Noone stores the protocol anywhere; simply delete it. */ void ip_nat_protocol_unregister(struct ip_nat_protocol *proto) { - WRITE_LOCK(&ip_nat_lock); + write_lock_bh(&ip_nat_lock); ip_nat_protos[proto->protonum] = &ip_nat_unknown_protocol; - WRITE_UNLOCK(&ip_nat_lock); + write_unlock_bh(&ip_nat_lock); /* Someone could be still looking at the proto in a bh. */ synchronize_net(); @@ -509,13 +509,13 @@ int __init ip_nat_init(void) return -ENOMEM; /* Sew in builtin protocols. */ - WRITE_LOCK(&ip_nat_lock); + write_lock_bh(&ip_nat_lock); for (i = 0; i < MAX_IP_NAT_PROTO; i++) ip_nat_protos[i] = &ip_nat_unknown_protocol; ip_nat_protos[IPPROTO_TCP] = &ip_nat_protocol_tcp; ip_nat_protos[IPPROTO_UDP] = &ip_nat_protocol_udp; ip_nat_protos[IPPROTO_ICMP] = &ip_nat_protocol_icmp; - WRITE_UNLOCK(&ip_nat_lock); + write_unlock_bh(&ip_nat_lock); for (i = 0; i < ip_nat_htable_size; i++) { INIT_LIST_HEAD(&bysource[i]); diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c index 1637b96d8c01..9cd51f180dcf 100644 --- a/net/ipv4/netfilter/ip_nat_helper.c +++ b/net/ipv4/netfilter/ip_nat_helper.c @@ -28,8 +28,8 @@ #include #include -#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock) -#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock) +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) #include #include @@ -47,7 +47,7 @@ #define DUMP_OFFSET(x) #endif -static DECLARE_LOCK(ip_nat_seqofs_lock); +static DEFINE_SPINLOCK(ip_nat_seqofs_lock); /* Setup TCP sequence correction given this change at this sequence */ static inline void @@ -70,7 +70,7 @@ adjust_tcp_sequence(u32 seq, DEBUGP("ip_nat_resize_packet: Seq_offset before: "); DUMP_OFFSET(this_way); - LOCK_BH(&ip_nat_seqofs_lock); + spin_lock_bh(&ip_nat_seqofs_lock); /* SYN adjust. If it's uninitialized, or this is after last * correction, record it: we don't handle more than one @@ -82,7 +82,7 @@ adjust_tcp_sequence(u32 seq, this_way->offset_before = this_way->offset_after; this_way->offset_after += sizediff; } - UNLOCK_BH(&ip_nat_seqofs_lock); + spin_unlock_bh(&ip_nat_seqofs_lock); DEBUGP("ip_nat_resize_packet: Seq_offset after: "); DUMP_OFFSET(this_way); diff --git a/net/ipv4/netfilter/ip_nat_rule.c b/net/ipv4/netfilter/ip_nat_rule.c index 581f097f5a24..60d70fa41a15 100644 --- a/net/ipv4/netfilter/ip_nat_rule.c +++ b/net/ipv4/netfilter/ip_nat_rule.c @@ -19,8 +19,8 @@ #include #include -#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock) -#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock) +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) #include #include diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c index 79f56f662b33..bc59d0d6e89e 100644 --- a/net/ipv4/netfilter/ip_nat_standalone.c +++ b/net/ipv4/netfilter/ip_nat_standalone.c @@ -31,8 +31,8 @@ #include #include -#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock) -#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock) +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) #include #include @@ -373,7 +373,6 @@ static int init_or_cleanup(int init) cleanup_rule_init: ip_nat_rule_cleanup(); cleanup_nothing: - MUST_BE_READ_WRITE_UNLOCKED(&ip_nat_lock); return ret; } diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 8a54f92b8496..c88dfcd38c56 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -67,7 +67,6 @@ static DECLARE_MUTEX(ipt_mutex); /* Must have mutex */ #define ASSERT_READ_LOCK(x) IP_NF_ASSERT(down_trylock(&ipt_mutex) != 0) #define ASSERT_WRITE_LOCK(x) IP_NF_ASSERT(down_trylock(&ipt_mutex) != 0) -#include #include #if 0 diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 0f12e3a3dc73..dc4362b57cfa 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -29,7 +29,6 @@ #include #include #include -#include #define CLUSTERIP_VERSION "0.6" @@ -41,6 +40,8 @@ #define DEBUGP #endif +#define ASSERT_READ_LOCK(x) + MODULE_LICENSE("GPL"); MODULE_AUTHOR("Harald Welte "); MODULE_DESCRIPTION("iptables target for CLUSTERIP"); @@ -67,7 +68,7 @@ static LIST_HEAD(clusterip_configs); /* clusterip_lock protects the clusterip_configs list _AND_ the configurable * data within all structurses (num_local_nodes, local_nodes[]) */ -static DECLARE_RWLOCK(clusterip_lock); +static DEFINE_RWLOCK(clusterip_lock); #ifdef CONFIG_PROC_FS static struct file_operations clusterip_proc_fops; @@ -82,9 +83,9 @@ clusterip_config_get(struct clusterip_config *c) { static inline void clusterip_config_put(struct clusterip_config *c) { if (atomic_dec_and_test(&c->refcount)) { - WRITE_LOCK(&clusterip_lock); + write_lock_bh(&clusterip_lock); list_del(&c->list); - WRITE_UNLOCK(&clusterip_lock); + write_unlock_bh(&clusterip_lock); dev_mc_delete(c->dev, c->clustermac, ETH_ALEN, 0); dev_put(c->dev); kfree(c); @@ -97,7 +98,7 @@ __clusterip_config_find(u_int32_t clusterip) { struct list_head *pos; - MUST_BE_READ_LOCKED(&clusterip_lock); + ASSERT_READ_LOCK(&clusterip_lock); list_for_each(pos, &clusterip_configs) { struct clusterip_config *c = list_entry(pos, struct clusterip_config, list); @@ -114,14 +115,14 @@ clusterip_config_find_get(u_int32_t clusterip) { struct clusterip_config *c; - READ_LOCK(&clusterip_lock); + read_lock_bh(&clusterip_lock); c = __clusterip_config_find(clusterip); if (!c) { - READ_UNLOCK(&clusterip_lock); + read_unlock_bh(&clusterip_lock); return NULL; } atomic_inc(&c->refcount); - READ_UNLOCK(&clusterip_lock); + read_unlock_bh(&clusterip_lock); return c; } @@ -160,9 +161,9 @@ clusterip_config_init(struct ipt_clusterip_tgt_info *i, u_int32_t ip, c->pde->data = c; #endif - WRITE_LOCK(&clusterip_lock); + write_lock_bh(&clusterip_lock); list_add(&c->list, &clusterip_configs); - WRITE_UNLOCK(&clusterip_lock); + write_unlock_bh(&clusterip_lock); return c; } @@ -172,25 +173,25 @@ clusterip_add_node(struct clusterip_config *c, u_int16_t nodenum) { int i; - WRITE_LOCK(&clusterip_lock); + write_lock_bh(&clusterip_lock); if (c->num_local_nodes >= CLUSTERIP_MAX_NODES || nodenum > CLUSTERIP_MAX_NODES) { - WRITE_UNLOCK(&clusterip_lock); + write_unlock_bh(&clusterip_lock); return 1; } /* check if we alrady have this number in our array */ for (i = 0; i < c->num_local_nodes; i++) { if (c->local_nodes[i] == nodenum) { - WRITE_UNLOCK(&clusterip_lock); + write_unlock_bh(&clusterip_lock); return 1; } } c->local_nodes[c->num_local_nodes++] = nodenum; - WRITE_UNLOCK(&clusterip_lock); + write_unlock_bh(&clusterip_lock); return 0; } @@ -199,10 +200,10 @@ clusterip_del_node(struct clusterip_config *c, u_int16_t nodenum) { int i; - WRITE_LOCK(&clusterip_lock); + write_lock_bh(&clusterip_lock); if (c->num_local_nodes <= 1 || nodenum > CLUSTERIP_MAX_NODES) { - WRITE_UNLOCK(&clusterip_lock); + write_unlock_bh(&clusterip_lock); return 1; } @@ -211,12 +212,12 @@ clusterip_del_node(struct clusterip_config *c, u_int16_t nodenum) int size = sizeof(u_int16_t)*(c->num_local_nodes-(i+1)); memmove(&c->local_nodes[i], &c->local_nodes[i+1], size); c->num_local_nodes--; - WRITE_UNLOCK(&clusterip_lock); + write_unlock_bh(&clusterip_lock); return 0; } } - WRITE_UNLOCK(&clusterip_lock); + write_unlock_bh(&clusterip_lock); return 1; } @@ -286,21 +287,21 @@ clusterip_responsible(struct clusterip_config *config, u_int32_t hash) { int i; - READ_LOCK(&clusterip_lock); + read_lock_bh(&clusterip_lock); if (config->num_local_nodes == 0) { - READ_UNLOCK(&clusterip_lock); + read_unlock_bh(&clusterip_lock); return 0; } for (i = 0; i < config->num_local_nodes; i++) { if (config->local_nodes[i] == hash) { - READ_UNLOCK(&clusterip_lock); + read_unlock_bh(&clusterip_lock); return 1; } } - READ_UNLOCK(&clusterip_lock); + read_unlock_bh(&clusterip_lock); return 0; } @@ -578,7 +579,7 @@ static void *clusterip_seq_start(struct seq_file *s, loff_t *pos) struct clusterip_config *c = pde->data; unsigned int *nodeidx; - READ_LOCK(&clusterip_lock); + read_lock_bh(&clusterip_lock); if (*pos >= c->num_local_nodes) return NULL; @@ -608,7 +609,7 @@ static void clusterip_seq_stop(struct seq_file *s, void *v) { kfree(v); - READ_UNLOCK(&clusterip_lock); + read_unlock_bh(&clusterip_lock); } static int clusterip_seq_show(struct seq_file *s, void *v) diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c index 57e9f6cf1c36..91e74502c3d3 100644 --- a/net/ipv4/netfilter/ipt_MASQUERADE.c +++ b/net/ipv4/netfilter/ipt_MASQUERADE.c @@ -33,7 +33,7 @@ MODULE_DESCRIPTION("iptables MASQUERADE target module"); #endif /* Lock protects masq region inside conntrack */ -static DECLARE_RWLOCK(masq_lock); +static DEFINE_RWLOCK(masq_lock); /* FIXME: Multiple targets. --RR */ static int @@ -103,9 +103,9 @@ masquerade_target(struct sk_buff **pskb, return NF_DROP; } - WRITE_LOCK(&masq_lock); + write_lock_bh(&masq_lock); ct->nat.masq_index = out->ifindex; - WRITE_UNLOCK(&masq_lock); + write_unlock_bh(&masq_lock); /* Transfer from original range. */ newrange = ((struct ip_nat_range) @@ -122,9 +122,9 @@ device_cmp(struct ip_conntrack *i, void *ifindex) { int ret; - READ_LOCK(&masq_lock); + read_lock_bh(&masq_lock); ret = (i->nat.masq_index == (int)(long)ifindex); - READ_UNLOCK(&masq_lock); + read_unlock_bh(&masq_lock); return ret; } diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c index 6f2cefbe16cd..52a0076302a7 100644 --- a/net/ipv4/netfilter/ipt_ULOG.c +++ b/net/ipv4/netfilter/ipt_ULOG.c @@ -56,7 +56,6 @@ #include #include #include -#include #include #include @@ -99,8 +98,8 @@ typedef struct { static ulog_buff_t ulog_buffers[ULOG_MAXNLGROUPS]; /* array of buffers */ -static struct sock *nflognl; /* our socket */ -static DECLARE_LOCK(ulog_lock); /* spinlock */ +static struct sock *nflognl; /* our socket */ +static DEFINE_SPINLOCK(ulog_lock); /* spinlock */ /* send one ulog_buff_t to userspace */ static void ulog_send(unsigned int nlgroupnum) @@ -135,9 +134,9 @@ static void ulog_timer(unsigned long data) /* lock to protect against somebody modifying our structure * from ipt_ulog_target at the same time */ - LOCK_BH(&ulog_lock); + spin_lock_bh(&ulog_lock); ulog_send(data); - UNLOCK_BH(&ulog_lock); + spin_unlock_bh(&ulog_lock); } static struct sk_buff *ulog_alloc_skb(unsigned int size) @@ -193,7 +192,7 @@ static void ipt_ulog_packet(unsigned int hooknum, ub = &ulog_buffers[groupnum]; - LOCK_BH(&ulog_lock); + spin_lock_bh(&ulog_lock); if (!ub->skb) { if (!(ub->skb = ulog_alloc_skb(size))) @@ -278,7 +277,7 @@ static void ipt_ulog_packet(unsigned int hooknum, ulog_send(groupnum); } - UNLOCK_BH(&ulog_lock); + spin_unlock_bh(&ulog_lock); return; @@ -288,7 +287,7 @@ nlmsg_failure: alloc_failure: PRINTR("ipt_ULOG: Error building netlink message\n"); - UNLOCK_BH(&ulog_lock); + spin_unlock_bh(&ulog_lock); } static unsigned int ipt_ulog_target(struct sk_buff **pskb, diff --git a/net/ipv4/netfilter/ipt_hashlimit.c b/net/ipv4/netfilter/ipt_hashlimit.c index f1937190cd77..564b49bfebcf 100644 --- a/net/ipv4/netfilter/ipt_hashlimit.c +++ b/net/ipv4/netfilter/ipt_hashlimit.c @@ -37,7 +37,6 @@ #include #include -#include /* FIXME: this is just for IP_NF_ASSERRT */ #include @@ -92,7 +91,7 @@ struct ipt_hashlimit_htable { struct hlist_head hash[0]; /* hashtable itself */ }; -static DECLARE_LOCK(hashlimit_lock); /* protects htables list */ +static DEFINE_SPINLOCK(hashlimit_lock); /* protects htables list */ static DECLARE_MUTEX(hlimit_mutex); /* additional checkentry protection */ static HLIST_HEAD(hashlimit_htables); static kmem_cache_t *hashlimit_cachep; @@ -233,9 +232,9 @@ static int htable_create(struct ipt_hashlimit_info *minfo) hinfo->timer.function = htable_gc; add_timer(&hinfo->timer); - LOCK_BH(&hashlimit_lock); + spin_lock_bh(&hashlimit_lock); hlist_add_head(&hinfo->node, &hashlimit_htables); - UNLOCK_BH(&hashlimit_lock); + spin_unlock_bh(&hashlimit_lock); return 0; } @@ -301,15 +300,15 @@ static struct ipt_hashlimit_htable *htable_find_get(char *name) struct ipt_hashlimit_htable *hinfo; struct hlist_node *pos; - LOCK_BH(&hashlimit_lock); + spin_lock_bh(&hashlimit_lock); hlist_for_each_entry(hinfo, pos, &hashlimit_htables, node) { if (!strcmp(name, hinfo->pde->name)) { atomic_inc(&hinfo->use); - UNLOCK_BH(&hashlimit_lock); + spin_unlock_bh(&hashlimit_lock); return hinfo; } } - UNLOCK_BH(&hashlimit_lock); + spin_unlock_bh(&hashlimit_lock); return NULL; } @@ -317,9 +316,9 @@ static struct ipt_hashlimit_htable *htable_find_get(char *name) static void htable_put(struct ipt_hashlimit_htable *hinfo) { if (atomic_dec_and_test(&hinfo->use)) { - LOCK_BH(&hashlimit_lock); + spin_lock_bh(&hashlimit_lock); hlist_del(&hinfo->node); - UNLOCK_BH(&hashlimit_lock); + spin_unlock_bh(&hashlimit_lock); htable_destroy(hinfo); } } diff --git a/net/ipv4/netfilter/ipt_helper.c b/net/ipv4/netfilter/ipt_helper.c index 33fdf364d3d3..3e7dd014de43 100644 --- a/net/ipv4/netfilter/ipt_helper.c +++ b/net/ipv4/netfilter/ipt_helper.c @@ -53,7 +53,7 @@ match(const struct sk_buff *skb, return ret; } - READ_LOCK(&ip_conntrack_lock); + read_lock_bh(&ip_conntrack_lock); if (!ct->master->helper) { DEBUGP("ipt_helper: master ct %p has no helper\n", exp->expectant); @@ -69,7 +69,7 @@ match(const struct sk_buff *skb, ret ^= !strncmp(ct->master->helper->name, info->name, strlen(ct->master->helper->name)); out_unlock: - READ_UNLOCK(&ip_conntrack_lock); + read_unlock_bh(&ip_conntrack_lock); return ret; } diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index c735276fdd5f..73034511c8db 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -71,7 +71,6 @@ static DECLARE_MUTEX(ip6t_mutex); /* Must have mutex */ #define ASSERT_READ_LOCK(x) IP_NF_ASSERT(down_trylock(&ip6t_mutex) != 0) #define ASSERT_WRITE_LOCK(x) IP_NF_ASSERT(down_trylock(&ip6t_mutex) != 0) -#include #include #if 0 -- cgit v1.2.3 From 18b8afc771102b1b6af97962808291a7d27f52af Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 21 Jun 2005 14:01:57 -0700 Subject: [NETFILTER]: Kill nf_debug Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4.h | 6 -- include/linux/skbuff.h | 13 ---- net/bridge/br_forward.c | 3 - net/bridge/br_input.c | 4 - net/bridge/br_netfilter.c | 38 --------- net/core/netfilter.c | 138 --------------------------------- net/core/skbuff.c | 6 -- net/ipv4/ip_input.c | 4 - net/ipv4/ip_output.c | 11 --- net/ipv4/ipvs/ip_vs_xmit.c | 1 - net/ipv4/netfilter/ip_conntrack_core.c | 9 --- net/ipv4/netfilter/ip_nat_helper.c | 3 - net/ipv6/ip6_output.c | 3 - 13 files changed, 239 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h index 9e5750079e09..3ebc36afae1a 100644 --- a/include/linux/netfilter_ipv4.h +++ b/include/linux/netfilter_ipv4.h @@ -75,12 +75,6 @@ enum nf_ip_hook_priorities { #define SO_ORIGINAL_DST 80 #ifdef __KERNEL__ -#ifdef CONFIG_NETFILTER_DEBUG -void nf_debug_ip_local_deliver(struct sk_buff *skb); -void nf_debug_ip_loopback_xmit(struct sk_buff *newskb); -void nf_debug_ip_finish_output2(struct sk_buff *skb); -#endif /*CONFIG_NETFILTER_DEBUG*/ - extern int ip_route_me_harder(struct sk_buff **pskb); /* Call this before modifying an existing IP packet: ensures it is diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index cc04f5cd2286..d7c839a21842 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -193,7 +193,6 @@ struct skb_shared_info { * @nfcache: Cache info * @nfct: Associated connection, if any * @nfctinfo: Relationship of this skb to the connection - * @nf_debug: Netfilter debugging * @nf_bridge: Saved data about a bridged frame - see br_netfilter.c * @private: Data which is private to the HIPPI implementation * @tc_index: Traffic control index @@ -264,9 +263,6 @@ struct sk_buff { __u32 nfcache; __u32 nfctinfo; struct nf_conntrack *nfct; -#ifdef CONFIG_NETFILTER_DEBUG - unsigned int nf_debug; -#endif #ifdef CONFIG_BRIDGE_NETFILTER struct nf_bridge_info *nf_bridge; #endif @@ -1219,15 +1215,6 @@ static inline void nf_reset(struct sk_buff *skb) { nf_conntrack_put(skb->nfct); skb->nfct = NULL; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif -} -static inline void nf_reset_debug(struct sk_buff *skb) -{ -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif } #ifdef CONFIG_BRIDGE_NETFILTER diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index ef9f2095f96e..069253f830c1 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -57,9 +57,6 @@ int br_forward_finish(struct sk_buff *skb) static void __br_deliver(const struct net_bridge_port *to, struct sk_buff *skb) { skb->dev = to->dev; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif NF_HOOK(PF_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev, br_forward_finish); } diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 8f5f2e730992..9a45e6279c57 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -23,11 +23,7 @@ const unsigned char bridge_ula[6] = { 0x01, 0x80, 0xc2, 0x00, 0x00, 0x00 }; static int br_pass_frame_up_finish(struct sk_buff *skb) { -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif netif_receive_skb(skb); - return 0; } diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index be03d3ad2648..03ae4edddac3 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -102,10 +102,6 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb) { struct nf_bridge_info *nf_bridge = skb->nf_bridge; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug ^= (1 << NF_BR_PRE_ROUTING); -#endif - if (nf_bridge->mask & BRNF_PKT_TYPE) { skb->pkt_type = PACKET_OTHERHOST; nf_bridge->mask ^= BRNF_PKT_TYPE; @@ -182,10 +178,6 @@ static void __br_dnat_complain(void) * --Bart, 20021007 (updated) */ static int br_nf_pre_routing_finish_bridge(struct sk_buff *skb) { -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug |= (1 << NF_BR_PRE_ROUTING) | (1 << NF_BR_FORWARD); -#endif - if (skb->pkt_type == PACKET_OTHERHOST) { skb->pkt_type = PACKET_HOST; skb->nf_bridge->mask |= BRNF_PKT_TYPE; @@ -207,10 +199,6 @@ static int br_nf_pre_routing_finish(struct sk_buff *skb) struct iphdr *iph = skb->nh.iph; struct nf_bridge_info *nf_bridge = skb->nf_bridge; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug ^= (1 << NF_BR_PRE_ROUTING); -#endif - if (nf_bridge->mask & BRNF_PKT_TYPE) { skb->pkt_type = PACKET_OTHERHOST; nf_bridge->mask ^= BRNF_PKT_TYPE; @@ -382,9 +370,6 @@ static unsigned int br_nf_pre_routing_ipv6(unsigned int hook, if (hdr->nexthdr == NEXTHDR_HOP && check_hbh_len(skb)) goto inhdr_error; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug ^= (1 << NF_IP6_PRE_ROUTING); -#endif if ((nf_bridge = nf_bridge_alloc(skb)) == NULL) return NF_DROP; setup_pre_routing(skb); @@ -468,9 +453,6 @@ static unsigned int br_nf_pre_routing(unsigned int hook, struct sk_buff **pskb, skb->ip_summed = CHECKSUM_NONE; } -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug ^= (1 << NF_IP_PRE_ROUTING); -#endif if ((nf_bridge = nf_bridge_alloc(skb)) == NULL) return NF_DROP; setup_pre_routing(skb); @@ -517,10 +499,6 @@ static int br_nf_forward_finish(struct sk_buff *skb) struct net_device *in; struct vlan_ethhdr *hdr = vlan_eth_hdr(skb); -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug ^= (1 << NF_BR_FORWARD); -#endif - if (skb->protocol != __constant_htons(ETH_P_ARP) && !IS_VLAN_ARP) { in = nf_bridge->physindev; if (nf_bridge->mask & BRNF_PKT_TYPE) { @@ -566,9 +544,6 @@ static unsigned int br_nf_forward_ip(unsigned int hook, struct sk_buff **pskb, (*pskb)->nh.raw += VLAN_HLEN; } -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug ^= (1 << NF_BR_FORWARD); -#endif nf_bridge = skb->nf_bridge; if (skb->pkt_type == PACKET_OTHERHOST) { skb->pkt_type = PACKET_HOST; @@ -605,10 +580,6 @@ static unsigned int br_nf_forward_arp(unsigned int hook, struct sk_buff **pskb, (*pskb)->nh.raw += VLAN_HLEN; } -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug ^= (1 << NF_BR_FORWARD); -#endif - if (skb->nh.arph->ar_pln != 4) { if (IS_VLAN_ARP) { skb_push(*pskb, VLAN_HLEN); @@ -627,9 +598,6 @@ static unsigned int br_nf_forward_arp(unsigned int hook, struct sk_buff **pskb, /* PF_BRIDGE/LOCAL_OUT ***********************************************/ static int br_nf_local_out_finish(struct sk_buff *skb) { -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug &= ~(1 << NF_BR_LOCAL_OUT); -#endif if (skb->protocol == __constant_htons(ETH_P_8021Q)) { skb_push(skb, VLAN_HLEN); skb->nh.raw -= VLAN_HLEN; @@ -731,10 +699,6 @@ static unsigned int br_nf_local_out(unsigned int hook, struct sk_buff **pskb, realoutdev, br_nf_local_out_finish, NF_IP_PRI_BRIDGE_SABOTAGE_FORWARD + 1); } else { -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug ^= (1 << NF_IP_LOCAL_OUT); -#endif - NF_HOOK_THRESH(pf, NF_IP_LOCAL_OUT, skb, realindev, realoutdev, br_nf_local_out_finish, NF_IP_PRI_BRIDGE_SABOTAGE_LOCAL_OUT + 1); @@ -779,8 +743,6 @@ static unsigned int br_nf_post_routing(unsigned int hook, struct sk_buff **pskb, printk(KERN_CRIT "br_netfilter: skb->dst == NULL."); goto print_error; } - - skb->nf_debug ^= (1 << NF_IP_POST_ROUTING); #endif /* We assume any code from br_dev_queue_push_xmit onwards doesn't care diff --git a/net/core/netfilter.c b/net/core/netfilter.c index 22a8f127c4aa..076c156d5eda 100644 --- a/net/core/netfilter.c +++ b/net/core/netfilter.c @@ -141,136 +141,6 @@ void nf_unregister_sockopt(struct nf_sockopt_ops *reg) up(&nf_sockopt_mutex); } -#ifdef CONFIG_NETFILTER_DEBUG -#include -#include -#include - -static void debug_print_hooks_ip(unsigned int nf_debug) -{ - if (nf_debug & (1 << NF_IP_PRE_ROUTING)) { - printk("PRE_ROUTING "); - nf_debug ^= (1 << NF_IP_PRE_ROUTING); - } - if (nf_debug & (1 << NF_IP_LOCAL_IN)) { - printk("LOCAL_IN "); - nf_debug ^= (1 << NF_IP_LOCAL_IN); - } - if (nf_debug & (1 << NF_IP_FORWARD)) { - printk("FORWARD "); - nf_debug ^= (1 << NF_IP_FORWARD); - } - if (nf_debug & (1 << NF_IP_LOCAL_OUT)) { - printk("LOCAL_OUT "); - nf_debug ^= (1 << NF_IP_LOCAL_OUT); - } - if (nf_debug & (1 << NF_IP_POST_ROUTING)) { - printk("POST_ROUTING "); - nf_debug ^= (1 << NF_IP_POST_ROUTING); - } - if (nf_debug) - printk("Crap bits: 0x%04X", nf_debug); - printk("\n"); -} - -static void nf_dump_skb(int pf, struct sk_buff *skb) -{ - printk("skb: pf=%i %s dev=%s len=%u\n", - pf, - skb->sk ? "(owned)" : "(unowned)", - skb->dev ? skb->dev->name : "(no dev)", - skb->len); - switch (pf) { - case PF_INET: { - const struct iphdr *ip = skb->nh.iph; - __u32 *opt = (__u32 *) (ip + 1); - int opti; - __u16 src_port = 0, dst_port = 0; - - if (ip->protocol == IPPROTO_TCP - || ip->protocol == IPPROTO_UDP) { - struct tcphdr *tcp=(struct tcphdr *)((__u32 *)ip+ip->ihl); - src_port = ntohs(tcp->source); - dst_port = ntohs(tcp->dest); - } - - printk("PROTO=%d %u.%u.%u.%u:%hu %u.%u.%u.%u:%hu" - " L=%hu S=0x%2.2hX I=%hu F=0x%4.4hX T=%hu", - ip->protocol, NIPQUAD(ip->saddr), - src_port, NIPQUAD(ip->daddr), - dst_port, - ntohs(ip->tot_len), ip->tos, ntohs(ip->id), - ntohs(ip->frag_off), ip->ttl); - - for (opti = 0; opti < (ip->ihl - sizeof(struct iphdr) / 4); opti++) - printk(" O=0x%8.8X", *opt++); - printk("\n"); - } - } -} - -void nf_debug_ip_local_deliver(struct sk_buff *skb) -{ - /* If it's a loopback packet, it must have come through - * NF_IP_LOCAL_OUT, NF_IP_RAW_INPUT, NF_IP_PRE_ROUTING and - * NF_IP_LOCAL_IN. Otherwise, must have gone through - * NF_IP_RAW_INPUT and NF_IP_PRE_ROUTING. */ - if (!skb->dev) { - printk("ip_local_deliver: skb->dev is NULL.\n"); - } else { - if (skb->nf_debug != ((1<nf_debug); - nf_dump_skb(PF_INET, skb); - } - } -} - -void nf_debug_ip_loopback_xmit(struct sk_buff *newskb) -{ - if (newskb->nf_debug != ((1 << NF_IP_LOCAL_OUT) - | (1 << NF_IP_POST_ROUTING))) { - printk("ip_dev_loopback_xmit: bad owned skb = %p: ", - newskb); - debug_print_hooks_ip(newskb->nf_debug); - nf_dump_skb(PF_INET, newskb); - } -} - -void nf_debug_ip_finish_output2(struct sk_buff *skb) -{ - /* If it's owned, it must have gone through the - * NF_IP_LOCAL_OUT and NF_IP_POST_ROUTING. - * Otherwise, must have gone through - * NF_IP_PRE_ROUTING, NF_IP_FORWARD and NF_IP_POST_ROUTING. - */ - if (skb->sk) { - if (skb->nf_debug != ((1 << NF_IP_LOCAL_OUT) - | (1 << NF_IP_POST_ROUTING))) { - printk("ip_finish_output: bad owned skb = %p: ", skb); - debug_print_hooks_ip(skb->nf_debug); - nf_dump_skb(PF_INET, skb); - } - } else { - if (skb->nf_debug != ((1 << NF_IP_PRE_ROUTING) - | (1 << NF_IP_FORWARD) - | (1 << NF_IP_POST_ROUTING))) { - /* Fragments, entunnelled packets, TCP RSTs - generated by ipt_REJECT will have no - owners, but still may be local */ - if (skb->nf_debug != ((1 << NF_IP_LOCAL_OUT) - | (1 << NF_IP_POST_ROUTING))){ - printk("ip_finish_output:" - " bad unowned skb = %p: ",skb); - debug_print_hooks_ip(skb->nf_debug); - nf_dump_skb(PF_INET, skb); - } - } - } -} -#endif /*CONFIG_NETFILTER_DEBUG*/ - /* Call get/setsockopt() */ static int nf_sockopt(struct sock *sk, int pf, int val, char __user *opt, int *len, int get) @@ -488,14 +358,6 @@ int nf_hook_slow(int pf, unsigned int hook, struct sk_buff **pskb, /* We may already have this, but read-locks nest anyway */ rcu_read_lock(); -#ifdef CONFIG_NETFILTER_DEBUG - if (unlikely((*pskb)->nf_debug & (1 << hook))) { - printk("nf_hook: hook %i already set.\n", hook); - nf_dump_skb(pf, *pskb); - } - (*pskb)->nf_debug |= (1 << hook); -#endif - elem = &nf_hooks[pf][hook]; next_hook: verdict = nf_iterate(&nf_hooks[pf][hook], pskb, hook, indev, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index f65b3de590a9..6d68c03bc051 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -365,9 +365,6 @@ struct sk_buff *skb_clone(struct sk_buff *skb, int gfp_mask) C(nfct); nf_conntrack_get(skb->nfct); C(nfctinfo); -#ifdef CONFIG_NETFILTER_DEBUG - C(nf_debug); -#endif #ifdef CONFIG_BRIDGE_NETFILTER C(nf_bridge); nf_bridge_get(skb->nf_bridge); @@ -432,9 +429,6 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old) new->nfct = old->nfct; nf_conntrack_get(old->nfct); new->nfctinfo = old->nfctinfo; -#ifdef CONFIG_NETFILTER_DEBUG - new->nf_debug = old->nf_debug; -#endif #ifdef CONFIG_BRIDGE_NETFILTER new->nf_bridge = old->nf_bridge; nf_bridge_get(old->nf_bridge); diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 4e47a2658c7c..2b7485e65040 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -200,10 +200,6 @@ static inline int ip_local_deliver_finish(struct sk_buff *skb) { int ihl = skb->nh.iph->ihl*4; -#ifdef CONFIG_NETFILTER_DEBUG - nf_debug_ip_local_deliver(skb); -#endif /*CONFIG_NETFILTER_DEBUG*/ - __skb_pull(skb, ihl); /* Free reference early: we don't need it any more, and it may diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 760dc8238d65..ee07aec215a0 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -107,10 +107,6 @@ static int ip_dev_loopback_xmit(struct sk_buff *newskb) newskb->pkt_type = PACKET_LOOPBACK; newskb->ip_summed = CHECKSUM_UNNECESSARY; BUG_TRAP(newskb->dst); - -#ifdef CONFIG_NETFILTER_DEBUG - nf_debug_ip_loopback_xmit(newskb); -#endif nf_reset(newskb); netif_rx(newskb); return 0; @@ -192,10 +188,6 @@ static inline int ip_finish_output2(struct sk_buff *skb) skb = skb2; } -#ifdef CONFIG_NETFILTER_DEBUG - nf_debug_ip_finish_output2(skb); -#endif /*CONFIG_NETFILTER_DEBUG*/ - nf_reset(skb); if (hh) { @@ -415,9 +407,6 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) to->nf_bridge = from->nf_bridge; nf_bridge_get(to->nf_bridge); #endif -#ifdef CONFIG_NETFILTER_DEBUG - to->nf_debug = from->nf_debug; -#endif #endif } diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c index de21da00057f..a8512a3fd08a 100644 --- a/net/ipv4/ipvs/ip_vs_xmit.c +++ b/net/ipv4/ipvs/ip_vs_xmit.c @@ -127,7 +127,6 @@ ip_vs_dst_reset(struct ip_vs_dest *dest) #define IP_VS_XMIT(skb, rt) \ do { \ - nf_reset_debug(skb); \ (skb)->nfcache |= NFC_IPVS_PROPERTY; \ (skb)->ip_summed = CHECKSUM_NONE; \ NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, (skb), NULL, \ diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index a7377a331ade..ffba0ad3c880 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -940,10 +940,6 @@ void ip_ct_refresh_acct(struct ip_conntrack *ct, struct sk_buff * ip_ct_gather_frags(struct sk_buff *skb, u_int32_t user) { -#ifdef CONFIG_NETFILTER_DEBUG - unsigned int olddebug = skb->nf_debug; -#endif - skb_orphan(skb); local_bh_disable(); @@ -953,12 +949,7 @@ ip_ct_gather_frags(struct sk_buff *skb, u_int32_t user) if (skb) { ip_send_check(skb->nh.iph); skb->nfcache |= NFC_ALTERED; -#ifdef CONFIG_NETFILTER_DEBUG - /* Packet path as if nothing had happened. */ - skb->nf_debug = olddebug; -#endif } - return skb; } diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c index 9cd51f180dcf..158f34f32c04 100644 --- a/net/ipv4/netfilter/ip_nat_helper.c +++ b/net/ipv4/netfilter/ip_nat_helper.c @@ -142,9 +142,6 @@ static int enlarge_skb(struct sk_buff **pskb, unsigned int extra) /* Transfer socket to new skb. */ if ((*pskb)->sk) skb_set_owner_w(nskb, (*pskb)->sk); -#ifdef CONFIG_NETFILTER_DEBUG - nskb->nf_debug = (*pskb)->nf_debug; -#endif kfree_skb(*pskb); *pskb = nskb; return 1; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index b78a53586804..06e7cdaeedc5 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -484,9 +484,6 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) to->nf_bridge = from->nf_bridge; nf_bridge_get(to->nf_bridge); #endif -#ifdef CONFIG_NETFILTER_DEBUG - to->nf_debug = from->nf_debug; -#endif #endif } -- cgit v1.2.3 From 39c715b71740c4a78ba4769fb54826929bac03cb Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 21 Jun 2005 17:14:34 -0700 Subject: [PATCH] smp_processor_id() cleanup This patch implements a number of smp_processor_id() cleanup ideas that Arjan van de Ven and I came up with. The previous __smp_processor_id/_smp_processor_id/smp_processor_id API spaghetti was hard to follow both on the implementational and on the usage side. Some of the complexity arose from picking wrong names, some of the complexity comes from the fact that not all architectures defined __smp_processor_id. In the new code, there are two externally visible symbols: - smp_processor_id(): debug variant. - raw_smp_processor_id(): nondebug variant. Replaces all existing uses of _smp_processor_id() and __smp_processor_id(). Defined by every SMP architecture in include/asm-*/smp.h. There is one new internal symbol, dependent on DEBUG_PREEMPT: - debug_smp_processor_id(): internal debug variant, mapped to smp_processor_id(). Also, i moved debug_smp_processor_id() from lib/kernel_lock.c into a new lib/smp_processor_id.c file. All related comments got updated and/or clarified. I have build/boot tested the following 8 .config combinations on x86: {SMP,UP} x {PREEMPT,!PREEMPT} x {DEBUG_PREEMPT,!DEBUG_PREEMPT} I have also build/boot tested x64 on UP/PREEMPT/DEBUG_PREEMPT. (Other architectures are untested, but should work just fine.) Signed-off-by: Ingo Molnar Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/traps.c | 2 +- arch/i386/lib/delay.c | 2 +- arch/ppc/lib/locks.c | 4 +-- arch/ppc64/kernel/idle.c | 2 +- arch/sh/lib/delay.c | 2 +- arch/sparc64/lib/delay.c | 2 +- arch/x86_64/lib/delay.c | 2 +- drivers/acpi/processor_idle.c | 2 +- drivers/input/gameport/gameport.c | 2 +- drivers/oprofile/buffer_sync.c | 4 +-- fs/xfs/linux-2.6/xfs_linux.h | 6 ++--- include/asm-alpha/smp.h | 2 +- include/asm-arm/smp.h | 2 +- include/asm-i386/smp.h | 2 +- include/asm-ia64/smp.h | 2 +- include/asm-m32r/smp.h | 2 +- include/asm-mips/smp.h | 2 +- include/asm-parisc/smp.h | 2 +- include/asm-ppc/smp.h | 2 +- include/asm-ppc64/smp.h | 2 +- include/asm-s390/smp.h | 2 +- include/asm-sh/smp.h | 2 +- include/asm-sparc/smp.h | 2 +- include/asm-sparc64/smp.h | 2 +- include/asm-um/smp.h | 3 ++- include/asm-x86_64/smp.h | 2 +- include/linux/mmzone.h | 2 +- include/linux/smp.h | 40 ++++++++++++---------------- include/net/route.h | 2 +- include/net/snmp.h | 14 +++++----- kernel/module.c | 2 +- kernel/power/smp.c | 4 +-- kernel/sched.c | 4 +-- kernel/stop_machine.c | 4 +-- lib/Makefile | 1 + lib/kernel_lock.c | 55 --------------------------------------- lib/smp_processor_id.c | 55 +++++++++++++++++++++++++++++++++++++++ 37 files changed, 119 insertions(+), 125 deletions(-) create mode 100644 lib/smp_processor_id.c (limited to 'include/linux') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 00c63419c06f..83c579e82a81 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -306,7 +306,7 @@ void die(const char * str, struct pt_regs * regs, long err) }; static int die_counter; - if (die.lock_owner != _smp_processor_id()) { + if (die.lock_owner != raw_smp_processor_id()) { console_verbose(); spin_lock_irq(&die.lock); die.lock_owner = smp_processor_id(); diff --git a/arch/i386/lib/delay.c b/arch/i386/lib/delay.c index 080639f262b1..eb0cdfe9280f 100644 --- a/arch/i386/lib/delay.c +++ b/arch/i386/lib/delay.c @@ -34,7 +34,7 @@ inline void __const_udelay(unsigned long xloops) xloops *= 4; __asm__("mull %0" :"=d" (xloops), "=&a" (d0) - :"1" (xloops),"0" (cpu_data[_smp_processor_id()].loops_per_jiffy * (HZ/4))); + :"1" (xloops),"0" (cpu_data[raw_smp_processor_id()].loops_per_jiffy * (HZ/4))); __delay(++xloops); } diff --git a/arch/ppc/lib/locks.c b/arch/ppc/lib/locks.c index 694163d696d8..c450dc4b766e 100644 --- a/arch/ppc/lib/locks.c +++ b/arch/ppc/lib/locks.c @@ -130,7 +130,7 @@ void _raw_read_lock(rwlock_t *rw) while (!read_can_lock(rw)) { if (--stuck == 0) { printk("_read_lock(%p) CPU#%d lock %d\n", - rw, _smp_processor_id(), rw->lock); + rw, raw_smp_processor_id(), rw->lock); stuck = INIT_STUCK; } } @@ -158,7 +158,7 @@ void _raw_write_lock(rwlock_t *rw) while (!write_can_lock(rw)) { if (--stuck == 0) { printk("write_lock(%p) CPU#%d lock %d)\n", - rw, _smp_processor_id(), rw->lock); + rw, raw_smp_processor_id(), rw->lock); stuck = INIT_STUCK; } } diff --git a/arch/ppc64/kernel/idle.c b/arch/ppc64/kernel/idle.c index f24ce2b87200..ff8a7db142d3 100644 --- a/arch/ppc64/kernel/idle.c +++ b/arch/ppc64/kernel/idle.c @@ -292,7 +292,7 @@ static int native_idle(void) if (need_resched()) schedule(); - if (cpu_is_offline(_smp_processor_id()) && + if (cpu_is_offline(raw_smp_processor_id()) && system_state == SYSTEM_RUNNING) cpu_die(); } diff --git a/arch/sh/lib/delay.c b/arch/sh/lib/delay.c index 50b36037d86b..351714694d6d 100644 --- a/arch/sh/lib/delay.c +++ b/arch/sh/lib/delay.c @@ -24,7 +24,7 @@ inline void __const_udelay(unsigned long xloops) __asm__("dmulu.l %0, %2\n\t" "sts mach, %0" : "=r" (xloops) - : "0" (xloops), "r" (cpu_data[_smp_processor_id()].loops_per_jiffy) + : "0" (xloops), "r" (cpu_data[raw_smp_processor_id()].loops_per_jiffy) : "macl", "mach"); __delay(xloops * HZ); } diff --git a/arch/sparc64/lib/delay.c b/arch/sparc64/lib/delay.c index f6b4c784d53e..e8808727617a 100644 --- a/arch/sparc64/lib/delay.c +++ b/arch/sparc64/lib/delay.c @@ -31,7 +31,7 @@ void __const_udelay(unsigned long n) { n *= 4; - n *= (cpu_data(_smp_processor_id()).udelay_val * (HZ/4)); + n *= (cpu_data(raw_smp_processor_id()).udelay_val * (HZ/4)); n >>= 32; __delay(n + 1); diff --git a/arch/x86_64/lib/delay.c b/arch/x86_64/lib/delay.c index 6e2d66472eb1..aed61a668a1b 100644 --- a/arch/x86_64/lib/delay.c +++ b/arch/x86_64/lib/delay.c @@ -34,7 +34,7 @@ void __delay(unsigned long loops) inline void __const_udelay(unsigned long xloops) { - __delay(((xloops * cpu_data[_smp_processor_id()].loops_per_jiffy) >> 32) * HZ); + __delay(((xloops * cpu_data[raw_smp_processor_id()].loops_per_jiffy) >> 32) * HZ); } void __udelay(unsigned long usecs) diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index ff64d333e95f..c9d671cf7857 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -171,7 +171,7 @@ static void acpi_processor_idle (void) int sleep_ticks = 0; u32 t1, t2 = 0; - pr = processors[_smp_processor_id()]; + pr = processors[raw_smp_processor_id()]; if (!pr) return; diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c index 9b8ff396e6f8..e152d0fa0cdd 100644 --- a/drivers/input/gameport/gameport.c +++ b/drivers/input/gameport/gameport.c @@ -134,7 +134,7 @@ static int gameport_measure_speed(struct gameport *gameport) } gameport_close(gameport); - return (cpu_data[_smp_processor_id()].loops_per_jiffy * (unsigned long)HZ / (1000 / 50)) / (tx < 1 ? 1 : tx); + return (cpu_data[raw_smp_processor_id()].loops_per_jiffy * (unsigned long)HZ / (1000 / 50)) / (tx < 1 ? 1 : tx); #else diff --git a/drivers/oprofile/buffer_sync.c b/drivers/oprofile/buffer_sync.c index 55720dc6ec43..745a14183634 100644 --- a/drivers/oprofile/buffer_sync.c +++ b/drivers/oprofile/buffer_sync.c @@ -62,7 +62,7 @@ static int task_exit_notify(struct notifier_block * self, unsigned long val, voi /* To avoid latency problems, we only process the current CPU, * hoping that most samples for the task are on this CPU */ - sync_buffer(_smp_processor_id()); + sync_buffer(raw_smp_processor_id()); return 0; } @@ -86,7 +86,7 @@ static int munmap_notify(struct notifier_block * self, unsigned long val, void * /* To avoid latency problems, we only process the current CPU, * hoping that most samples for the task are on this CPU */ - sync_buffer(_smp_processor_id()); + sync_buffer(raw_smp_processor_id()); return 0; } diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h index 71bb41019a12..7d7c8788ea75 100644 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ b/fs/xfs/linux-2.6/xfs_linux.h @@ -145,10 +145,10 @@ static inline void set_buffer_unwritten_io(struct buffer_head *bh) #define xfs_inherit_nosymlinks xfs_params.inherit_nosym.val #define xfs_rotorstep xfs_params.rotorstep.val -#ifndef __smp_processor_id -#define __smp_processor_id() smp_processor_id() +#ifndef raw_smp_processor_id +#define raw_smp_processor_id() smp_processor_id() #endif -#define current_cpu() __smp_processor_id() +#define current_cpu() raw_smp_processor_id() #define current_pid() (current->pid) #define current_fsuid(cred) (current->fsuid) #define current_fsgid(cred) (current->fsgid) diff --git a/include/asm-alpha/smp.h b/include/asm-alpha/smp.h index cbc173ae45aa..9950706abdf8 100644 --- a/include/asm-alpha/smp.h +++ b/include/asm-alpha/smp.h @@ -43,7 +43,7 @@ extern struct cpuinfo_alpha cpu_data[NR_CPUS]; #define PROC_CHANGE_PENALTY 20 #define hard_smp_processor_id() __hard_smp_processor_id() -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) extern cpumask_t cpu_present_mask; extern cpumask_t cpu_online_map; diff --git a/include/asm-arm/smp.h b/include/asm-arm/smp.h index bd44f894690f..6c6c60adbbaa 100644 --- a/include/asm-arm/smp.h +++ b/include/asm-arm/smp.h @@ -21,7 +21,7 @@ # error " included in non-SMP build" #endif -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) extern cpumask_t cpu_present_mask; #define cpu_possible_map cpu_present_mask diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h index e03a206dfa36..55ef31f66bbe 100644 --- a/include/asm-i386/smp.h +++ b/include/asm-i386/smp.h @@ -51,7 +51,7 @@ extern u8 x86_cpu_to_apicid[]; * from the initial startup. We map APIC_BASE very early in page_setup(), * so this is correct in the x86 case. */ -#define __smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) extern cpumask_t cpu_callout_map; extern cpumask_t cpu_callin_map; diff --git a/include/asm-ia64/smp.h b/include/asm-ia64/smp.h index 3ba1a061e4ae..a3914352c995 100644 --- a/include/asm-ia64/smp.h +++ b/include/asm-ia64/smp.h @@ -46,7 +46,7 @@ ia64_get_lid (void) #define SMP_IRQ_REDIRECTION (1 << 0) #define SMP_IPI_REDIRECTION (1 << 1) -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) extern struct smp_boot_data { int cpu_count; diff --git a/include/asm-m32r/smp.h b/include/asm-m32r/smp.h index 8cd4d0da4be1..b9a20cdad65f 100644 --- a/include/asm-m32r/smp.h +++ b/include/asm-m32r/smp.h @@ -66,7 +66,7 @@ extern volatile int cpu_2_physid[NR_CPUS]; #define physid_to_cpu(physid) physid_2_cpu[physid] #define cpu_to_physid(cpu_id) cpu_2_physid[cpu_id] -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) extern cpumask_t cpu_callout_map; #define cpu_possible_map cpu_callout_map diff --git a/include/asm-mips/smp.h b/include/asm-mips/smp.h index 8ba370ecfd4c..5618f1e12f40 100644 --- a/include/asm-mips/smp.h +++ b/include/asm-mips/smp.h @@ -21,7 +21,7 @@ #include #include -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) /* Map from cpu id to sequential logical cpu number. This will only not be idempotent when cpus failed to come on-line. */ diff --git a/include/asm-parisc/smp.h b/include/asm-parisc/smp.h index fde77ac35463..9413f67a540b 100644 --- a/include/asm-parisc/smp.h +++ b/include/asm-parisc/smp.h @@ -51,7 +51,7 @@ extern void smp_send_reschedule(int cpu); extern unsigned long cpu_present_mask; -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) #endif /* CONFIG_SMP */ diff --git a/include/asm-ppc/smp.h b/include/asm-ppc/smp.h index ebfb614f55f6..17530c232c76 100644 --- a/include/asm-ppc/smp.h +++ b/include/asm-ppc/smp.h @@ -44,7 +44,7 @@ extern void smp_message_recv(int, struct pt_regs *); #define NO_PROC_ID 0xFF /* No processor magic marker */ #define PROC_CHANGE_PENALTY 20 -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) extern int __cpu_up(unsigned int cpu); diff --git a/include/asm-ppc64/smp.h b/include/asm-ppc64/smp.h index c8646fa999c2..8115ecb8feee 100644 --- a/include/asm-ppc64/smp.h +++ b/include/asm-ppc64/smp.h @@ -45,7 +45,7 @@ void generic_cpu_die(unsigned int cpu); void generic_mach_cpu_die(void); #endif -#define __smp_processor_id() (get_paca()->paca_index) +#define raw_smp_processor_id() (get_paca()->paca_index) #define hard_smp_processor_id() (get_paca()->hw_cpu_id) extern cpumask_t cpu_sibling_map[NR_CPUS]; diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h index 9473786387a3..dd50e57a928f 100644 --- a/include/asm-s390/smp.h +++ b/include/asm-s390/smp.h @@ -47,7 +47,7 @@ extern int smp_call_function_on(void (*func) (void *info), void *info, #define PROC_CHANGE_PENALTY 20 /* Schedule penalty */ -#define smp_processor_id() (S390_lowcore.cpu_data.cpu_nr) +#define raw_smp_processor_id() (S390_lowcore.cpu_data.cpu_nr) extern int smp_get_cpu(cpumask_t cpu_map); extern void smp_put_cpu(int cpu); diff --git a/include/asm-sh/smp.h b/include/asm-sh/smp.h index 38b54469d7d1..f19a8b3b69a6 100644 --- a/include/asm-sh/smp.h +++ b/include/asm-sh/smp.h @@ -25,7 +25,7 @@ extern cpumask_t cpu_possible_map; #define cpu_online(cpu) cpu_isset(cpu, cpu_online_map) -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) /* I've no idea what the real meaning of this is */ #define PROC_CHANGE_PENALTY 20 diff --git a/include/asm-sparc/smp.h b/include/asm-sparc/smp.h index f986c0d0922a..4f96d8333a12 100644 --- a/include/asm-sparc/smp.h +++ b/include/asm-sparc/smp.h @@ -148,7 +148,7 @@ extern __inline__ int hard_smp_processor_id(void) } #endif -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) #define prof_multiplier(__cpu) cpu_data(__cpu).multiplier #define prof_counter(__cpu) cpu_data(__cpu).counter diff --git a/include/asm-sparc64/smp.h b/include/asm-sparc64/smp.h index 5e3e06d908fe..110a2de89123 100644 --- a/include/asm-sparc64/smp.h +++ b/include/asm-sparc64/smp.h @@ -64,7 +64,7 @@ static __inline__ int hard_smp_processor_id(void) } } -#define smp_processor_id() (current_thread_info()->cpu) +#define raw_smp_processor_id() (current_thread_info()->cpu) #endif /* !(__ASSEMBLY__) */ diff --git a/include/asm-um/smp.h b/include/asm-um/smp.h index 4412d5d9c26b..d879eba2b52c 100644 --- a/include/asm-um/smp.h +++ b/include/asm-um/smp.h @@ -8,7 +8,8 @@ #include "asm/current.h" #include "linux/cpumask.h" -#define smp_processor_id() (current_thread->cpu) +#define raw_smp_processor_id() (current_thread->cpu) + #define cpu_logical_map(n) (n) #define cpu_number_map(n) (n) #define PROC_CHANGE_PENALTY 15 /* Pick a number, any number */ diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h index 96844fecbde8..a7425aa5a3b7 100644 --- a/include/asm-x86_64/smp.h +++ b/include/asm-x86_64/smp.h @@ -68,7 +68,7 @@ static inline int num_booting_cpus(void) return cpus_weight(cpu_callout_map); } -#define __smp_processor_id() read_pda(cpunumber) +#define raw_smp_processor_id() read_pda(cpunumber) extern __inline int hard_smp_processor_id(void) { diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index e530c6c092f1..beacd931b606 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -381,7 +381,7 @@ int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int, struct file *, #include /* Returns the number of the current Node. */ -#define numa_node_id() (cpu_to_node(_smp_processor_id())) +#define numa_node_id() (cpu_to_node(raw_smp_processor_id())) #ifndef CONFIG_DISCONTIGMEM diff --git a/include/linux/smp.h b/include/linux/smp.h index dcf1db3b35d3..9dfa3ee769ae 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -92,10 +92,7 @@ void smp_prepare_boot_cpu(void); /* * These macros fold the SMP functionality into a single CPU system */ - -#if !defined(__smp_processor_id) || !defined(CONFIG_PREEMPT) -# define smp_processor_id() 0 -#endif +#define raw_smp_processor_id() 0 #define hard_smp_processor_id() 0 #define smp_call_function(func,info,retry,wait) ({ 0; }) #define on_each_cpu(func,info,retry,wait) ({ func(info); 0; }) @@ -106,30 +103,25 @@ static inline void smp_send_reschedule(int cpu) { } #endif /* !SMP */ /* - * DEBUG_PREEMPT support: check whether smp_processor_id() is being - * used in a preemption-safe way. + * smp_processor_id(): get the current CPU ID. * - * An architecture has to enable this debugging code explicitly. - * It can do so by renaming the smp_processor_id() macro to - * __smp_processor_id(). This should only be done after some minimal - * testing, because usually there are a number of false positives - * that an architecture will trigger. + * if DEBUG_PREEMPT is enabled the we check whether it is + * used in a preemption-safe way. (smp_processor_id() is safe + * if it's used in a preemption-off critical section, or in + * a thread that is bound to the current CPU.) * - * To fix a false positive (i.e. smp_processor_id() use that the - * debugging code reports but which use for some reason is legal), - * change the smp_processor_id() reference to _smp_processor_id(), - * which is the nondebug variant. NOTE: don't use this to hack around - * real bugs. + * NOTE: raw_smp_processor_id() is for internal use only + * (smp_processor_id() is the preferred variant), but in rare + * instances it might also be used to turn off false positives + * (i.e. smp_processor_id() use that the debugging code reports but + * which use for some reason is legal). Don't use this to hack around + * the warning message, as your code might not work under PREEMPT. */ -#ifdef __smp_processor_id -# if defined(CONFIG_PREEMPT) && defined(CONFIG_DEBUG_PREEMPT) - extern unsigned int smp_processor_id(void); -# else -# define smp_processor_id() __smp_processor_id() -# endif -# define _smp_processor_id() __smp_processor_id() +#ifdef CONFIG_DEBUG_PREEMPT + extern unsigned int debug_smp_processor_id(void); +# define smp_processor_id() debug_smp_processor_id() #else -# define _smp_processor_id() smp_processor_id() +# define smp_processor_id() raw_smp_processor_id() #endif #define get_cpu() ({ preempt_disable(); smp_processor_id(); }) diff --git a/include/net/route.h b/include/net/route.h index d34ca8fc6756..c3cd069a9aca 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -107,7 +107,7 @@ struct rt_cache_stat extern struct rt_cache_stat *rt_cache_stat; #define RT_CACHE_STAT_INC(field) \ - (per_cpu_ptr(rt_cache_stat, _smp_processor_id())->field++) + (per_cpu_ptr(rt_cache_stat, raw_smp_processor_id())->field++) extern struct ip_rt_acct *ip_rt_acct; diff --git a/include/net/snmp.h b/include/net/snmp.h index a15ab256276e..a36bed8ea210 100644 --- a/include/net/snmp.h +++ b/include/net/snmp.h @@ -128,18 +128,18 @@ struct linux_mib { #define SNMP_STAT_USRPTR(name) (name[1]) #define SNMP_INC_STATS_BH(mib, field) \ - (per_cpu_ptr(mib[0], _smp_processor_id())->mibs[field]++) + (per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field]++) #define SNMP_INC_STATS_OFFSET_BH(mib, field, offset) \ - (per_cpu_ptr(mib[0], _smp_processor_id())->mibs[field + (offset)]++) + (per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field + (offset)]++) #define SNMP_INC_STATS_USER(mib, field) \ - (per_cpu_ptr(mib[1], _smp_processor_id())->mibs[field]++) + (per_cpu_ptr(mib[1], raw_smp_processor_id())->mibs[field]++) #define SNMP_INC_STATS(mib, field) \ - (per_cpu_ptr(mib[!in_softirq()], _smp_processor_id())->mibs[field]++) + (per_cpu_ptr(mib[!in_softirq()], raw_smp_processor_id())->mibs[field]++) #define SNMP_DEC_STATS(mib, field) \ - (per_cpu_ptr(mib[!in_softirq()], _smp_processor_id())->mibs[field]--) + (per_cpu_ptr(mib[!in_softirq()], raw_smp_processor_id())->mibs[field]--) #define SNMP_ADD_STATS_BH(mib, field, addend) \ - (per_cpu_ptr(mib[0], _smp_processor_id())->mibs[field] += addend) + (per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field] += addend) #define SNMP_ADD_STATS_USER(mib, field, addend) \ - (per_cpu_ptr(mib[1], _smp_processor_id())->mibs[field] += addend) + (per_cpu_ptr(mib[1], raw_smp_processor_id())->mibs[field] += addend) #endif diff --git a/kernel/module.c b/kernel/module.c index 83b3d376708c..a566745dde62 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -379,7 +379,7 @@ static void module_unload_init(struct module *mod) for (i = 0; i < NR_CPUS; i++) local_set(&mod->ref[i].count, 0); /* Hold reference count during initialization. */ - local_set(&mod->ref[_smp_processor_id()].count, 1); + local_set(&mod->ref[raw_smp_processor_id()].count, 1); /* Backwards compatibility macros put refcount during init. */ mod->waiter = current; } diff --git a/kernel/power/smp.c b/kernel/power/smp.c index cba3584b80fe..457c2302ed42 100644 --- a/kernel/power/smp.c +++ b/kernel/power/smp.c @@ -48,11 +48,11 @@ void disable_nonboot_cpus(void) { oldmask = current->cpus_allowed; set_cpus_allowed(current, cpumask_of_cpu(0)); - printk("Freezing CPUs (at %d)", _smp_processor_id()); + printk("Freezing CPUs (at %d)", raw_smp_processor_id()); current->state = TASK_INTERRUPTIBLE; schedule_timeout(HZ); printk("..."); - BUG_ON(_smp_processor_id() != 0); + BUG_ON(raw_smp_processor_id() != 0); /* FIXME: for this to work, all the CPUs must be running * "idle" thread (or we deadlock). Is that guaranteed? */ diff --git a/kernel/sched.c b/kernel/sched.c index f12a0c8a7d98..deca041fc364 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3814,7 +3814,7 @@ EXPORT_SYMBOL(yield); */ void __sched io_schedule(void) { - struct runqueue *rq = &per_cpu(runqueues, _smp_processor_id()); + struct runqueue *rq = &per_cpu(runqueues, raw_smp_processor_id()); atomic_inc(&rq->nr_iowait); schedule(); @@ -3825,7 +3825,7 @@ EXPORT_SYMBOL(io_schedule); long __sched io_schedule_timeout(long timeout) { - struct runqueue *rq = &per_cpu(runqueues, _smp_processor_id()); + struct runqueue *rq = &per_cpu(runqueues, raw_smp_processor_id()); long ret; atomic_inc(&rq->nr_iowait); diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 6116b25aa7cf..84a9d18aa8da 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -100,7 +100,7 @@ static int stop_machine(void) stopmachine_state = STOPMACHINE_WAIT; for_each_online_cpu(i) { - if (i == _smp_processor_id()) + if (i == raw_smp_processor_id()) continue; ret = kernel_thread(stopmachine, (void *)(long)i,CLONE_KERNEL); if (ret < 0) @@ -182,7 +182,7 @@ struct task_struct *__stop_machine_run(int (*fn)(void *), void *data, /* If they don't care which CPU fn runs on, bind to any online one. */ if (cpu == NR_CPUS) - cpu = _smp_processor_id(); + cpu = raw_smp_processor_id(); p = kthread_create(do_stop, &smdata, "kstopmachine"); if (!IS_ERR(p)) { diff --git a/lib/Makefile b/lib/Makefile index 9eccea9429a7..5f10cb898407 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -20,6 +20,7 @@ lib-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o lib-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem.o lib-$(CONFIG_GENERIC_FIND_NEXT_BIT) += find_next_bit.o obj-$(CONFIG_LOCK_KERNEL) += kernel_lock.o +obj-$(CONFIG_DEBUG_PREEMPT) += smp_processor_id.o ifneq ($(CONFIG_HAVE_DEC_LOCK),y) lib-y += dec_and_lock.o diff --git a/lib/kernel_lock.c b/lib/kernel_lock.c index 99b0ae3d51dd..bd2bc5d887b8 100644 --- a/lib/kernel_lock.c +++ b/lib/kernel_lock.c @@ -9,61 +9,6 @@ #include #include -#if defined(CONFIG_PREEMPT) && defined(__smp_processor_id) && \ - defined(CONFIG_DEBUG_PREEMPT) - -/* - * Debugging check. - */ -unsigned int smp_processor_id(void) -{ - unsigned long preempt_count = preempt_count(); - int this_cpu = __smp_processor_id(); - cpumask_t this_mask; - - if (likely(preempt_count)) - goto out; - - if (irqs_disabled()) - goto out; - - /* - * Kernel threads bound to a single CPU can safely use - * smp_processor_id(): - */ - this_mask = cpumask_of_cpu(this_cpu); - - if (cpus_equal(current->cpus_allowed, this_mask)) - goto out; - - /* - * It is valid to assume CPU-locality during early bootup: - */ - if (system_state != SYSTEM_RUNNING) - goto out; - - /* - * Avoid recursion: - */ - preempt_disable(); - - if (!printk_ratelimit()) - goto out_enable; - - printk(KERN_ERR "BUG: using smp_processor_id() in preemptible [%08x] code: %s/%d\n", preempt_count(), current->comm, current->pid); - print_symbol("caller is %s\n", (long)__builtin_return_address(0)); - dump_stack(); - -out_enable: - preempt_enable_no_resched(); -out: - return this_cpu; -} - -EXPORT_SYMBOL(smp_processor_id); - -#endif /* PREEMPT && __smp_processor_id && DEBUG_PREEMPT */ - #ifdef CONFIG_PREEMPT_BKL /* * The 'big kernel semaphore' diff --git a/lib/smp_processor_id.c b/lib/smp_processor_id.c new file mode 100644 index 000000000000..42c08ef828c5 --- /dev/null +++ b/lib/smp_processor_id.c @@ -0,0 +1,55 @@ +/* + * lib/smp_processor_id.c + * + * DEBUG_PREEMPT variant of smp_processor_id(). + */ +#include +#include + +unsigned int debug_smp_processor_id(void) +{ + unsigned long preempt_count = preempt_count(); + int this_cpu = raw_smp_processor_id(); + cpumask_t this_mask; + + if (likely(preempt_count)) + goto out; + + if (irqs_disabled()) + goto out; + + /* + * Kernel threads bound to a single CPU can safely use + * smp_processor_id(): + */ + this_mask = cpumask_of_cpu(this_cpu); + + if (cpus_equal(current->cpus_allowed, this_mask)) + goto out; + + /* + * It is valid to assume CPU-locality during early bootup: + */ + if (system_state != SYSTEM_RUNNING) + goto out; + + /* + * Avoid recursion: + */ + preempt_disable(); + + if (!printk_ratelimit()) + goto out_enable; + + printk(KERN_ERR "BUG: using smp_processor_id() in preemptible [%08x] code: %s/%d\n", preempt_count(), current->comm, current->pid); + print_symbol("caller is %s\n", (long)__builtin_return_address(0)); + dump_stack(); + +out_enable: + preempt_enable_no_resched(); +out: + return this_cpu; +} + +EXPORT_SYMBOL(debug_smp_processor_id); + -- cgit v1.2.3 From 753ee728964e5afb80c17659cc6c3a6fd0a42fe0 Mon Sep 17 00:00:00 2001 From: Martin Hicks Date: Tue, 21 Jun 2005 17:14:41 -0700 Subject: [PATCH] VM: early zone reclaim This is the core of the (much simplified) early reclaim. The goal of this patch is to reclaim some easily-freed pages from a zone before falling back onto another zone. One of the major uses of this is NUMA machines. With the default allocator behavior the allocator would look for memory in another zone, which might be off-node, before trying to reclaim from the current zone. This adds a zone tuneable to enable early zone reclaim. It is selected on a per-zone basis and is turned on/off via syscall. Adding some extra throttling on the reclaim was also required (patch 4/4). Without the machine would grind to a crawl when doing a "make -j" kernel build. Even with this patch the System Time is higher on average, but it seems tolerable. Here are some numbers for kernbench runs on a 2-node, 4cpu, 8Gig RAM Altix in the "make -j" run: wall user sys %cpu ctx sw. sleeps ---- ---- --- ---- ------ ------ No patch 1009 1384 847 258 298170 504402 w/patch, no reclaim 880 1376 667 288 254064 396745 w/patch & reclaim 1079 1385 926 252 291625 548873 These numbers are the average of 2 runs of 3 "make -j" runs done right after system boot. Run-to-run variability for "make -j" is huge, so these numbers aren't terribly useful except to seee that with reclaim the benchmark still finishes in a reasonable amount of time. I also looked at the NUMA hit/miss stats for the "make -j" runs and the reclaim doesn't make any difference when the machine is thrashing away. Doing a "make -j8" on a single node that is filled with page cache pages takes 700 seconds with reclaim turned on and 735 seconds without reclaim (due to remote memory accesses). The simple zone_reclaim syscall program is at http://www.bork.org/~mort/sgi/zone_reclaim.c Signed-off-by: Martin Hicks Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/syscall_table.S | 2 +- arch/ia64/kernel/entry.S | 2 +- include/asm-i386/unistd.h | 2 +- include/asm-ia64/unistd.h | 1 + include/linux/mmzone.h | 6 ++++ include/linux/swap.h | 1 + kernel/sys_ni.c | 1 + mm/page_alloc.c | 33 +++++++++++++++++---- mm/vmscan.c | 64 ++++++++++++++++++++++++++++++++++++++++ 9 files changed, 104 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S index 6cd1ed311f02..d408afaf6495 100644 --- a/arch/i386/kernel/syscall_table.S +++ b/arch/i386/kernel/syscall_table.S @@ -251,7 +251,7 @@ ENTRY(sys_call_table) .long sys_io_submit .long sys_io_cancel .long sys_fadvise64 /* 250 */ - .long sys_ni_syscall + .long sys_set_zone_reclaim .long sys_exit_group .long sys_lookup_dcookie .long sys_epoll_create diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index d99316c9be28..b1d5d3d5276c 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1579,7 +1579,7 @@ sys_call_table: data8 sys_keyctl data8 sys_ni_syscall data8 sys_ni_syscall // 1275 - data8 sys_ni_syscall + data8 sys_set_zone_reclaim data8 sys_ni_syscall data8 sys_ni_syscall data8 sys_ni_syscall diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h index 61bcc1b1e3f4..176413fb9ae3 100644 --- a/include/asm-i386/unistd.h +++ b/include/asm-i386/unistd.h @@ -256,7 +256,7 @@ #define __NR_io_submit 248 #define __NR_io_cancel 249 #define __NR_fadvise64 250 - +#define __NR_set_zone_reclaim 251 #define __NR_exit_group 252 #define __NR_lookup_dcookie 253 #define __NR_epoll_create 254 diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h index 33e26c557c5c..f7f43ec2483a 100644 --- a/include/asm-ia64/unistd.h +++ b/include/asm-ia64/unistd.h @@ -263,6 +263,7 @@ #define __NR_add_key 1271 #define __NR_request_key 1272 #define __NR_keyctl 1273 +#define __NR_set_zone_reclaim 1276 #ifdef __KERNEL__ diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index beacd931b606..dfc2452ccb10 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -144,6 +144,12 @@ struct zone { unsigned long pages_scanned; /* since last reclaim */ int all_unreclaimable; /* All pages pinned */ + /* + * Does the allocator try to reclaim pages from the zone as soon + * as it fails a watermark_ok() in __alloc_pages? + */ + int reclaim_pages; + /* * prev_priority holds the scanning priority for this zone. It is * defined as the scanning priority at which we achieved our reclaim diff --git a/include/linux/swap.h b/include/linux/swap.h index 3bbc41be9bd0..0d21e682d99d 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -173,6 +173,7 @@ extern void swap_setup(void); /* linux/mm/vmscan.c */ extern int try_to_free_pages(struct zone **, unsigned int, unsigned int); +extern int zone_reclaim(struct zone *, unsigned int, unsigned int); extern int shrink_all_memory(int); extern int vm_swappiness; diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 0dda70ed1f98..6f15bea7d1a8 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -77,6 +77,7 @@ cond_syscall(sys_request_key); cond_syscall(sys_keyctl); cond_syscall(compat_sys_keyctl); cond_syscall(compat_sys_socketcall); +cond_syscall(sys_set_zone_reclaim); /* arch-specific weak syscall entries */ cond_syscall(sys_pciconfig_read); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 40169f0b7e9e..3c0f69ded6b5 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -724,6 +724,14 @@ int zone_watermark_ok(struct zone *z, int order, unsigned long mark, return 1; } +static inline int +should_reclaim_zone(struct zone *z, unsigned int gfp_mask) +{ + if (!z->reclaim_pages) + return 0; + return 1; +} + /* * This is the 'heart' of the zoned buddy allocator. */ @@ -760,17 +768,32 @@ __alloc_pages(unsigned int __nocast gfp_mask, unsigned int order, classzone_idx = zone_idx(zones[0]); - restart: +restart: /* Go through the zonelist once, looking for a zone with enough free */ for (i = 0; (z = zones[i]) != NULL; i++) { - - if (!zone_watermark_ok(z, order, z->pages_low, - classzone_idx, 0, 0)) - continue; + int do_reclaim = should_reclaim_zone(z, gfp_mask); if (!cpuset_zone_allowed(z)) continue; + /* + * If the zone is to attempt early page reclaim then this loop + * will try to reclaim pages and check the watermark a second + * time before giving up and falling back to the next zone. + */ +zone_reclaim_retry: + if (!zone_watermark_ok(z, order, z->pages_low, + classzone_idx, 0, 0)) { + if (!do_reclaim) + continue; + else { + zone_reclaim(z, gfp_mask, order); + /* Only try reclaim once */ + do_reclaim = 0; + goto zone_reclaim_retry; + } + } + page = buffered_rmqueue(z, order, gfp_mask); if (page) goto got_pg; diff --git a/mm/vmscan.c b/mm/vmscan.c index 6379ddbffd9b..7da846960d8a 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1323,3 +1323,67 @@ static int __init kswapd_init(void) } module_init(kswapd_init) + + +/* + * Try to free up some pages from this zone through reclaim. + */ +int zone_reclaim(struct zone *zone, unsigned int gfp_mask, unsigned int order) +{ + struct scan_control sc; + int nr_pages = 1 << order; + int total_reclaimed = 0; + + /* The reclaim may sleep, so don't do it if sleep isn't allowed */ + if (!(gfp_mask & __GFP_WAIT)) + return 0; + if (zone->all_unreclaimable) + return 0; + + sc.gfp_mask = gfp_mask; + sc.may_writepage = 0; + sc.may_swap = 0; + sc.nr_mapped = read_page_state(nr_mapped); + sc.nr_scanned = 0; + sc.nr_reclaimed = 0; + /* scan at the highest priority */ + sc.priority = 0; + + if (nr_pages > SWAP_CLUSTER_MAX) + sc.swap_cluster_max = nr_pages; + else + sc.swap_cluster_max = SWAP_CLUSTER_MAX; + + shrink_zone(zone, &sc); + total_reclaimed = sc.nr_reclaimed; + + return total_reclaimed; +} + +asmlinkage long sys_set_zone_reclaim(unsigned int node, unsigned int zone, + unsigned int state) +{ + struct zone *z; + int i; + + if (node >= MAX_NUMNODES || !node_online(node)) + return -EINVAL; + + /* This will break if we ever add more zones */ + if (!(zone & (1<node_zones[i]; + + if (state) + z->reclaim_pages = 1; + else + z->reclaim_pages = 0; + } + + return 0; +} -- cgit v1.2.3 From 0c35bbadc59f5ed105c34471143eceb4c0dd9c95 Mon Sep 17 00:00:00 2001 From: Martin Hicks Date: Tue, 21 Jun 2005 17:14:42 -0700 Subject: [PATCH] VM: add __GFP_NORECLAIM When using the early zone reclaim, it was noticed that allocating new pages that should be spread across the whole system caused eviction of local pages. This adds a new GFP flag to prevent early reclaim from happening during certain allocation attempts. The example that is implemented here is for page cache pages. We want page cache pages to be spread across the whole system, and we don't want page cache pages to evict other pages to get local memory. Signed-off-by: Martin Hicks Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 3 ++- include/linux/pagemap.h | 4 ++-- mm/page_alloc.c | 2 ++ 3 files changed, 6 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gfp.h b/include/linux/gfp.h index af7407e8cfc5..208535fd4832 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -39,6 +39,7 @@ struct vm_area_struct; #define __GFP_COMP 0x4000u /* Add compound page metadata */ #define __GFP_ZERO 0x8000u /* Return zeroed page on success */ #define __GFP_NOMEMALLOC 0x10000u /* Don't use emergency reserves */ +#define __GFP_NORECLAIM 0x20000u /* No realy zone reclaim during allocation */ #define __GFP_BITS_SHIFT 20 /* Room for 20 __GFP_FOO bits */ #define __GFP_BITS_MASK ((1 << __GFP_BITS_SHIFT) - 1) @@ -47,7 +48,7 @@ struct vm_area_struct; #define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \ __GFP_COLD|__GFP_NOWARN|__GFP_REPEAT| \ __GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP| \ - __GFP_NOMEMALLOC) + __GFP_NOMEMALLOC|__GFP_NORECLAIM) #define GFP_ATOMIC (__GFP_HIGH) #define GFP_NOIO (__GFP_WAIT) diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index 0422031161ba..d9a25647a295 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -52,12 +52,12 @@ void release_pages(struct page **pages, int nr, int cold); static inline struct page *page_cache_alloc(struct address_space *x) { - return alloc_pages(mapping_gfp_mask(x), 0); + return alloc_pages(mapping_gfp_mask(x)|__GFP_NORECLAIM, 0); } static inline struct page *page_cache_alloc_cold(struct address_space *x) { - return alloc_pages(mapping_gfp_mask(x)|__GFP_COLD, 0); + return alloc_pages(mapping_gfp_mask(x)|__GFP_COLD|__GFP_NORECLAIM, 0); } typedef int filler_t(void *, struct page *); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 3c0f69ded6b5..a9da20bc2ed0 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -729,6 +729,8 @@ should_reclaim_zone(struct zone *z, unsigned int gfp_mask) { if (!z->reclaim_pages) return 0; + if (gfp_mask & __GFP_NORECLAIM) + return 0; return 1; } -- cgit v1.2.3 From 1e7e5a9048b30c57ba1ddaa6cdf59b21b65cde99 Mon Sep 17 00:00:00 2001 From: Martin Hicks Date: Tue, 21 Jun 2005 17:14:43 -0700 Subject: [PATCH] VM: rate limit early reclaim When early zone reclaim is turned on the LRU is scanned more frequently when a zone is low on memory. This limits when the zone reclaim can be called by skipping the scan if another thread (either via kswapd or sync reclaim) is already reclaiming from the zone. Signed-off-by: Martin Hicks Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 2 ++ mm/page_alloc.c | 1 + mm/vmscan.c | 10 ++++++++++ 3 files changed, 13 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index dfc2452ccb10..18fed8b67943 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -149,6 +149,8 @@ struct zone { * as it fails a watermark_ok() in __alloc_pages? */ int reclaim_pages; + /* A count of how many reclaimers are scanning this zone */ + atomic_t reclaim_in_progress; /* * prev_priority holds the scanning priority for this zone. It is diff --git a/mm/page_alloc.c b/mm/page_alloc.c index a9da20bc2ed0..2019c1b19254 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1738,6 +1738,7 @@ static void __init free_area_init_core(struct pglist_data *pgdat, zone->nr_scan_inactive = 0; zone->nr_active = 0; zone->nr_inactive = 0; + atomic_set(&zone->reclaim_in_progress, -1); if (!size) continue; diff --git a/mm/vmscan.c b/mm/vmscan.c index 7da846960d8a..24da725a30f0 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -900,7 +900,9 @@ shrink_caches(struct zone **zones, struct scan_control *sc) if (zone->all_unreclaimable && sc->priority != DEF_PRIORITY) continue; /* Let kswapd poll it */ + atomic_inc(&zone->reclaim_in_progress); shrink_zone(zone, sc); + atomic_dec(&zone->reclaim_in_progress); } } @@ -1111,7 +1113,9 @@ scan: sc.nr_reclaimed = 0; sc.priority = priority; sc.swap_cluster_max = nr_pages? nr_pages : SWAP_CLUSTER_MAX; + atomic_inc(&zone->reclaim_in_progress); shrink_zone(zone, &sc); + atomic_dec(&zone->reclaim_in_progress); reclaim_state->reclaimed_slab = 0; nr_slab = shrink_slab(sc.nr_scanned, GFP_KERNEL, lru_pages); @@ -1354,9 +1358,15 @@ int zone_reclaim(struct zone *zone, unsigned int gfp_mask, unsigned int order) else sc.swap_cluster_max = SWAP_CLUSTER_MAX; + /* Don't reclaim the zone if there are other reclaimers active */ + if (!atomic_inc_and_test(&zone->reclaim_in_progress)) + goto out; + shrink_zone(zone, &sc); total_reclaimed = sc.nr_reclaimed; + out: + atomic_dec(&zone->reclaim_in_progress); return total_reclaimed; } -- cgit v1.2.3 From 63551ae0feaaa23807ebea60de1901564bbef32e Mon Sep 17 00:00:00 2001 From: David Gibson Date: Tue, 21 Jun 2005 17:14:44 -0700 Subject: [PATCH] Hugepage consolidation A lot of the code in arch/*/mm/hugetlbpage.c is quite similar. This patch attempts to consolidate a lot of the code across the arch's, putting the combined version in mm/hugetlb.c. There are a couple of uglyish hacks in order to covert all the hugepage archs, but the result is a very large reduction in the total amount of code. It also means things like hugepage lazy allocation could be implemented in one place, instead of six. Tested, at least a little, on ppc64, i386 and x86_64. Notes: - this patch changes the meaning of set_huge_pte() to be more analagous to set_pte() - does SH4 need s special huge_ptep_get_and_clear()?? Acked-by: William Lee Irwin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/hugetlbpage.c | 170 ++---------------------------------- arch/ia64/mm/hugetlbpage.c | 158 +--------------------------------- arch/ppc64/mm/hugetlbpage.c | 180 +------------------------------------- arch/sh/mm/hugetlbpage.c | 196 ++++++------------------------------------ arch/sh64/mm/hugetlbpage.c | 18 +++- arch/sparc64/mm/hugetlbpage.c | 195 ++++++----------------------------------- include/asm-i386/page.h | 1 + include/asm-i386/pgtable.h | 2 +- include/asm-ia64/pgtable.h | 1 + include/asm-sh/page.h | 1 + include/asm-sh/pgtable.h | 1 + include/asm-sh64/page.h | 1 + include/asm-sh64/pgtable.h | 2 + include/asm-sparc64/page.h | 2 + include/asm-sparc64/pgtable.h | 1 + include/asm-x86_64/page.h | 1 + include/asm-x86_64/pgtable.h | 3 +- include/linux/hugetlb.h | 40 +++++++-- mm/hugetlb.c | 177 +++++++++++++++++++++++++++++++++++++- 19 files changed, 300 insertions(+), 850 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/mm/hugetlbpage.c b/arch/i386/mm/hugetlbpage.c index 171fc925e1e4..5aa06001a4bd 100644 --- a/arch/i386/mm/hugetlbpage.c +++ b/arch/i386/mm/hugetlbpage.c @@ -18,7 +18,7 @@ #include #include -static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; pud_t *pud; @@ -30,7 +30,7 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) return (pte_t *) pmd; } -static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; pud_t *pud; @@ -42,21 +42,6 @@ static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return (pte_t *) pmd; } -static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, struct page *page, pte_t * page_table, int write_access) -{ - pte_t entry; - - add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE); - if (write_access) { - entry = - pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot))); - } else - entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot)); - entry = pte_mkyoung(entry); - mk_pte_huge(entry); - set_pte(page_table, entry); -} - /* * This function checks for proper alignment of input addr and len parameters. */ @@ -69,77 +54,6 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len) return 0; } -int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, - struct vm_area_struct *vma) -{ - pte_t *src_pte, *dst_pte, entry; - struct page *ptepage; - unsigned long addr = vma->vm_start; - unsigned long end = vma->vm_end; - - while (addr < end) { - dst_pte = huge_pte_alloc(dst, addr); - if (!dst_pte) - goto nomem; - src_pte = huge_pte_offset(src, addr); - entry = *src_pte; - ptepage = pte_page(entry); - get_page(ptepage); - set_pte(dst_pte, entry); - add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE); - addr += HPAGE_SIZE; - } - return 0; - -nomem: - return -ENOMEM; -} - -int -follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, - struct page **pages, struct vm_area_struct **vmas, - unsigned long *position, int *length, int i) -{ - unsigned long vpfn, vaddr = *position; - int remainder = *length; - - WARN_ON(!is_vm_hugetlb_page(vma)); - - vpfn = vaddr/PAGE_SIZE; - while (vaddr < vma->vm_end && remainder) { - - if (pages) { - pte_t *pte; - struct page *page; - - pte = huge_pte_offset(mm, vaddr); - - /* hugetlb should be locked, and hence, prefaulted */ - WARN_ON(!pte || pte_none(*pte)); - - page = &pte_page(*pte)[vpfn % (HPAGE_SIZE/PAGE_SIZE)]; - - WARN_ON(!PageCompound(page)); - - get_page(page); - pages[i] = page; - } - - if (vmas) - vmas[i] = vma; - - vaddr += PAGE_SIZE; - ++vpfn; - --remainder; - ++i; - } - - *length = remainder; - *position = vaddr; - - return i; -} - #if 0 /* This is just for testing */ struct page * follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) @@ -204,83 +118,15 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address, } #endif -void unmap_hugepage_range(struct vm_area_struct *vma, - unsigned long start, unsigned long end) +void hugetlb_clean_stale_pgtable(pte_t *pte) { - struct mm_struct *mm = vma->vm_mm; - unsigned long address; - pte_t pte, *ptep; + pmd_t *pmd = (pmd_t *) pte; struct page *page; - BUG_ON(start & (HPAGE_SIZE - 1)); - BUG_ON(end & (HPAGE_SIZE - 1)); - - for (address = start; address < end; address += HPAGE_SIZE) { - ptep = huge_pte_offset(mm, address); - if (!ptep) - continue; - pte = ptep_get_and_clear(mm, address, ptep); - if (pte_none(pte)) - continue; - page = pte_page(pte); - put_page(page); - } - add_mm_counter(mm ,rss, -((end - start) >> PAGE_SHIFT)); - flush_tlb_range(vma, start, end); -} - -int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma) -{ - struct mm_struct *mm = current->mm; - unsigned long addr; - int ret = 0; - - BUG_ON(vma->vm_start & ~HPAGE_MASK); - BUG_ON(vma->vm_end & ~HPAGE_MASK); - - spin_lock(&mm->page_table_lock); - for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) { - unsigned long idx; - pte_t *pte = huge_pte_alloc(mm, addr); - struct page *page; - - if (!pte) { - ret = -ENOMEM; - goto out; - } - - if (!pte_none(*pte)) - continue; - - idx = ((addr - vma->vm_start) >> HPAGE_SHIFT) - + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); - page = find_get_page(mapping, idx); - if (!page) { - /* charge the fs quota first */ - if (hugetlb_get_quota(mapping)) { - ret = -ENOMEM; - goto out; - } - page = alloc_huge_page(); - if (!page) { - hugetlb_put_quota(mapping); - ret = -ENOMEM; - goto out; - } - ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC); - if (! ret) { - unlock_page(page); - } else { - hugetlb_put_quota(mapping); - free_huge_page(page); - goto out; - } - } - set_huge_pte(mm, vma, page, pte, vma->vm_flags & VM_WRITE); - } -out: - spin_unlock(&mm->page_table_lock); - return ret; + page = pmd_page(*pmd); + pmd_clear(pmd); + dec_page_state(nr_page_table_pages); + page_cache_release(page); } /* x86_64 also uses this file */ diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index df08ae7634b6..e0a776a3044c 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -24,7 +24,7 @@ unsigned int hpage_shift=HPAGE_SHIFT_DEFAULT; -static pte_t * +pte_t * huge_pte_alloc (struct mm_struct *mm, unsigned long addr) { unsigned long taddr = htlbpage_to_page(addr); @@ -43,7 +43,7 @@ huge_pte_alloc (struct mm_struct *mm, unsigned long addr) return pte; } -static pte_t * +pte_t * huge_pte_offset (struct mm_struct *mm, unsigned long addr) { unsigned long taddr = htlbpage_to_page(addr); @@ -67,23 +67,6 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr) #define mk_pte_huge(entry) { pte_val(entry) |= _PAGE_P; } -static void -set_huge_pte (struct mm_struct *mm, struct vm_area_struct *vma, - struct page *page, pte_t * page_table, int write_access) -{ - pte_t entry; - - add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE); - if (write_access) { - entry = - pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot))); - } else - entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot)); - entry = pte_mkyoung(entry); - mk_pte_huge(entry); - set_pte(page_table, entry); - return; -} /* * This function checks for proper alignment of input addr and len parameters. */ @@ -99,68 +82,6 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len) return 0; } -int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, - struct vm_area_struct *vma) -{ - pte_t *src_pte, *dst_pte, entry; - struct page *ptepage; - unsigned long addr = vma->vm_start; - unsigned long end = vma->vm_end; - - while (addr < end) { - dst_pte = huge_pte_alloc(dst, addr); - if (!dst_pte) - goto nomem; - src_pte = huge_pte_offset(src, addr); - entry = *src_pte; - ptepage = pte_page(entry); - get_page(ptepage); - set_pte(dst_pte, entry); - add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE); - addr += HPAGE_SIZE; - } - return 0; -nomem: - return -ENOMEM; -} - -int -follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, - struct page **pages, struct vm_area_struct **vmas, - unsigned long *st, int *length, int i) -{ - pte_t *ptep, pte; - unsigned long start = *st; - unsigned long pstart; - int len = *length; - struct page *page; - - do { - pstart = start & HPAGE_MASK; - ptep = huge_pte_offset(mm, start); - pte = *ptep; - -back1: - page = pte_page(pte); - if (pages) { - page += ((start & ~HPAGE_MASK) >> PAGE_SHIFT); - get_page(page); - pages[i] = page; - } - if (vmas) - vmas[i] = vma; - i++; - len--; - start += PAGE_SIZE; - if (((start & HPAGE_MASK) == pstart) && len && - (start < vma->vm_end)) - goto back1; - } while (len && start < vma->vm_end); - *length = len; - *st = start; - return i; -} - struct page *follow_huge_addr(struct mm_struct *mm, unsigned long addr, int write) { struct page *page; @@ -212,81 +133,6 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb, free_pgd_range(tlb, addr, end, floor, ceiling); } -void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, unsigned long end) -{ - struct mm_struct *mm = vma->vm_mm; - unsigned long address; - pte_t *pte; - struct page *page; - - BUG_ON(start & (HPAGE_SIZE - 1)); - BUG_ON(end & (HPAGE_SIZE - 1)); - - for (address = start; address < end; address += HPAGE_SIZE) { - pte = huge_pte_offset(mm, address); - if (pte_none(*pte)) - continue; - page = pte_page(*pte); - put_page(page); - pte_clear(mm, address, pte); - } - add_mm_counter(mm, rss, - ((end - start) >> PAGE_SHIFT)); - flush_tlb_range(vma, start, end); -} - -int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma) -{ - struct mm_struct *mm = current->mm; - unsigned long addr; - int ret = 0; - - BUG_ON(vma->vm_start & ~HPAGE_MASK); - BUG_ON(vma->vm_end & ~HPAGE_MASK); - - spin_lock(&mm->page_table_lock); - for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) { - unsigned long idx; - pte_t *pte = huge_pte_alloc(mm, addr); - struct page *page; - - if (!pte) { - ret = -ENOMEM; - goto out; - } - if (!pte_none(*pte)) - continue; - - idx = ((addr - vma->vm_start) >> HPAGE_SHIFT) - + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); - page = find_get_page(mapping, idx); - if (!page) { - /* charge the fs quota first */ - if (hugetlb_get_quota(mapping)) { - ret = -ENOMEM; - goto out; - } - page = alloc_huge_page(); - if (!page) { - hugetlb_put_quota(mapping); - ret = -ENOMEM; - goto out; - } - ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC); - if (! ret) { - unlock_page(page); - } else { - hugetlb_put_quota(mapping); - page_cache_release(page); - goto out; - } - } - set_huge_pte(mm, vma, page, pte, vma->vm_flags & VM_WRITE); - } -out: - spin_unlock(&mm->page_table_lock); - return ret; -} - unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { diff --git a/arch/ppc64/mm/hugetlbpage.c b/arch/ppc64/mm/hugetlbpage.c index d3bf86a5c1ad..b4ab766f5980 100644 --- a/arch/ppc64/mm/hugetlbpage.c +++ b/arch/ppc64/mm/hugetlbpage.c @@ -121,7 +121,7 @@ static pte_t *hugepte_alloc(struct mm_struct *mm, pud_t *dir, unsigned long addr return hugepte_offset(dir, addr); } -static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) { pud_t *pud; @@ -134,7 +134,7 @@ static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return hugepte_offset(pud, addr); } -static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { pud_t *pud; @@ -147,25 +147,6 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) return hugepte_alloc(mm, pud, addr); } -static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, - unsigned long addr, struct page *page, - pte_t *ptep, int write_access) -{ - pte_t entry; - - add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE); - if (write_access) { - entry = - pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot))); - } else { - entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot)); - } - entry = pte_mkyoung(entry); - entry = pte_mkhuge(entry); - - set_pte_at(mm, addr, ptep, entry); -} - /* * This function checks for proper alignment of input addr and len parameters. */ @@ -259,80 +240,6 @@ int prepare_hugepage_range(unsigned long addr, unsigned long len) return -EINVAL; } -int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, - struct vm_area_struct *vma) -{ - pte_t *src_pte, *dst_pte, entry; - struct page *ptepage; - unsigned long addr = vma->vm_start; - unsigned long end = vma->vm_end; - int err = -ENOMEM; - - while (addr < end) { - dst_pte = huge_pte_alloc(dst, addr); - if (!dst_pte) - goto out; - - src_pte = huge_pte_offset(src, addr); - entry = *src_pte; - - ptepage = pte_page(entry); - get_page(ptepage); - add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE); - set_pte_at(dst, addr, dst_pte, entry); - - addr += HPAGE_SIZE; - } - - err = 0; - out: - return err; -} - -int -follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, - struct page **pages, struct vm_area_struct **vmas, - unsigned long *position, int *length, int i) -{ - unsigned long vpfn, vaddr = *position; - int remainder = *length; - - WARN_ON(!is_vm_hugetlb_page(vma)); - - vpfn = vaddr/PAGE_SIZE; - while (vaddr < vma->vm_end && remainder) { - if (pages) { - pte_t *pte; - struct page *page; - - pte = huge_pte_offset(mm, vaddr); - - /* hugetlb should be locked, and hence, prefaulted */ - WARN_ON(!pte || pte_none(*pte)); - - page = &pte_page(*pte)[vpfn % (HPAGE_SIZE/PAGE_SIZE)]; - - WARN_ON(!PageCompound(page)); - - get_page(page); - pages[i] = page; - } - - if (vmas) - vmas[i] = vma; - - vaddr += PAGE_SIZE; - ++vpfn; - --remainder; - ++i; - } - - *length = remainder; - *position = vaddr; - - return i; -} - struct page * follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { @@ -363,89 +270,6 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address, return NULL; } -void unmap_hugepage_range(struct vm_area_struct *vma, - unsigned long start, unsigned long end) -{ - struct mm_struct *mm = vma->vm_mm; - unsigned long addr; - pte_t *ptep; - struct page *page; - - WARN_ON(!is_vm_hugetlb_page(vma)); - BUG_ON((start % HPAGE_SIZE) != 0); - BUG_ON((end % HPAGE_SIZE) != 0); - - for (addr = start; addr < end; addr += HPAGE_SIZE) { - pte_t pte; - - ptep = huge_pte_offset(mm, addr); - if (!ptep || pte_none(*ptep)) - continue; - - pte = *ptep; - page = pte_page(pte); - pte_clear(mm, addr, ptep); - - put_page(page); - } - add_mm_counter(mm, rss, -((end - start) >> PAGE_SHIFT)); - flush_tlb_pending(); -} - -int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma) -{ - struct mm_struct *mm = current->mm; - unsigned long addr; - int ret = 0; - - WARN_ON(!is_vm_hugetlb_page(vma)); - BUG_ON((vma->vm_start % HPAGE_SIZE) != 0); - BUG_ON((vma->vm_end % HPAGE_SIZE) != 0); - - spin_lock(&mm->page_table_lock); - for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) { - unsigned long idx; - pte_t *pte = huge_pte_alloc(mm, addr); - struct page *page; - - if (!pte) { - ret = -ENOMEM; - goto out; - } - if (! pte_none(*pte)) - continue; - - idx = ((addr - vma->vm_start) >> HPAGE_SHIFT) - + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); - page = find_get_page(mapping, idx); - if (!page) { - /* charge the fs quota first */ - if (hugetlb_get_quota(mapping)) { - ret = -ENOMEM; - goto out; - } - page = alloc_huge_page(); - if (!page) { - hugetlb_put_quota(mapping); - ret = -ENOMEM; - goto out; - } - ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC); - if (! ret) { - unlock_page(page); - } else { - hugetlb_put_quota(mapping); - free_huge_page(page); - goto out; - } - } - set_huge_pte(mm, vma, addr, page, pte, vma->vm_flags & VM_WRITE); - } -out: - spin_unlock(&mm->page_table_lock); - return ret; -} - /* Because we have an exclusive hugepage region which lies within the * normal user address space, we have to take special measures to make * non-huge mmap()s evade the hugepage reserved regions. */ diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c index 1f897bab2318..95bb1a6c6060 100644 --- a/arch/sh/mm/hugetlbpage.c +++ b/arch/sh/mm/hugetlbpage.c @@ -24,7 +24,7 @@ #include #include -static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; pmd_t *pmd; @@ -39,7 +39,7 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) return pte; } -static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; pmd_t *pmd; @@ -56,28 +56,34 @@ static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) #define mk_pte_huge(entry) do { pte_val(entry) |= _PAGE_SZHUGE; } while (0) -static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, - struct page *page, pte_t * page_table, int write_access) +void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t entry) { - unsigned long i; - pte_t entry; + int i; - add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE); + for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { + set_pte_at(mm, addr, ptep, entry); + ptep++; + addr += PAGE_SIZE; + pte_val(entry) += PAGE_SIZE; + } +} - if (write_access) - entry = pte_mkwrite(pte_mkdirty(mk_pte(page, - vma->vm_page_prot))); - else - entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot)); - entry = pte_mkyoung(entry); - mk_pte_huge(entry); +pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) +{ + pte_t entry; + int i; - for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { - set_pte(page_table, entry); - page_table++; + entry = *ptep; - pte_val(entry) += PAGE_SIZE; + for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { + pte_clear(mm, addr, ptep); + addr += PAGE_SIZE; + ptep++; } + + return entry; } /* @@ -92,79 +98,6 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len) return 0; } -int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, - struct vm_area_struct *vma) -{ - pte_t *src_pte, *dst_pte, entry; - struct page *ptepage; - unsigned long addr = vma->vm_start; - unsigned long end = vma->vm_end; - int i; - - while (addr < end) { - dst_pte = huge_pte_alloc(dst, addr); - if (!dst_pte) - goto nomem; - src_pte = huge_pte_offset(src, addr); - BUG_ON(!src_pte || pte_none(*src_pte)); - entry = *src_pte; - ptepage = pte_page(entry); - get_page(ptepage); - for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { - set_pte(dst_pte, entry); - pte_val(entry) += PAGE_SIZE; - dst_pte++; - } - add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE); - addr += HPAGE_SIZE; - } - return 0; - -nomem: - return -ENOMEM; -} - -int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, - struct page **pages, struct vm_area_struct **vmas, - unsigned long *position, int *length, int i) -{ - unsigned long vaddr = *position; - int remainder = *length; - - WARN_ON(!is_vm_hugetlb_page(vma)); - - while (vaddr < vma->vm_end && remainder) { - if (pages) { - pte_t *pte; - struct page *page; - - pte = huge_pte_offset(mm, vaddr); - - /* hugetlb should be locked, and hence, prefaulted */ - BUG_ON(!pte || pte_none(*pte)); - - page = pte_page(*pte); - - WARN_ON(!PageCompound(page)); - - get_page(page); - pages[i] = page; - } - - if (vmas) - vmas[i] = vma; - - vaddr += PAGE_SIZE; - --remainder; - ++i; - } - - *length = remainder; - *position = vaddr; - - return i; -} - struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { @@ -181,84 +114,3 @@ struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, { return NULL; } - -void unmap_hugepage_range(struct vm_area_struct *vma, - unsigned long start, unsigned long end) -{ - struct mm_struct *mm = vma->vm_mm; - unsigned long address; - pte_t *pte; - struct page *page; - int i; - - BUG_ON(start & (HPAGE_SIZE - 1)); - BUG_ON(end & (HPAGE_SIZE - 1)); - - for (address = start; address < end; address += HPAGE_SIZE) { - pte = huge_pte_offset(mm, address); - BUG_ON(!pte); - if (pte_none(*pte)) - continue; - page = pte_page(*pte); - put_page(page); - for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { - pte_clear(mm, address+(i*PAGE_SIZE), pte); - pte++; - } - } - add_mm_counter(mm, rss, -((end - start) >> PAGE_SHIFT)); - flush_tlb_range(vma, start, end); -} - -int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma) -{ - struct mm_struct *mm = current->mm; - unsigned long addr; - int ret = 0; - - BUG_ON(vma->vm_start & ~HPAGE_MASK); - BUG_ON(vma->vm_end & ~HPAGE_MASK); - - spin_lock(&mm->page_table_lock); - for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) { - unsigned long idx; - pte_t *pte = huge_pte_alloc(mm, addr); - struct page *page; - - if (!pte) { - ret = -ENOMEM; - goto out; - } - if (!pte_none(*pte)) - continue; - - idx = ((addr - vma->vm_start) >> HPAGE_SHIFT) - + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); - page = find_get_page(mapping, idx); - if (!page) { - /* charge the fs quota first */ - if (hugetlb_get_quota(mapping)) { - ret = -ENOMEM; - goto out; - } - page = alloc_huge_page(); - if (!page) { - hugetlb_put_quota(mapping); - ret = -ENOMEM; - goto out; - } - ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC); - if (! ret) { - unlock_page(page); - } else { - hugetlb_put_quota(mapping); - free_huge_page(page); - goto out; - } - } - set_huge_pte(mm, vma, page, pte, vma->vm_flags & VM_WRITE); - } -out: - spin_unlock(&mm->page_table_lock); - return ret; -} diff --git a/arch/sh64/mm/hugetlbpage.c b/arch/sh64/mm/hugetlbpage.c index bcad2aefa4ee..dcd9c8a8baf8 100644 --- a/arch/sh64/mm/hugetlbpage.c +++ b/arch/sh64/mm/hugetlbpage.c @@ -24,7 +24,7 @@ #include #include -static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; pmd_t *pmd; @@ -39,7 +39,7 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) return pte; } -static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; pmd_t *pmd; @@ -80,6 +80,20 @@ static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, } } +pte_t huge_ptep_get_and_clear(pte_t *ptep) +{ + pte_t entry; + + entry = *ptep; + + for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { + pte_clear(pte); + pte++; + } + + return entry; +} + /* * This function checks for proper alignment of input addr and len parameters. */ diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c index 5a1f831b2de1..625cbb336a23 100644 --- a/arch/sparc64/mm/hugetlbpage.c +++ b/arch/sparc64/mm/hugetlbpage.c @@ -22,7 +22,7 @@ #include #include -static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; pud_t *pud; @@ -41,7 +41,7 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) return pte; } -static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; pud_t *pud; @@ -62,30 +62,34 @@ static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) #define mk_pte_huge(entry) do { pte_val(entry) |= _PAGE_SZHUGE; } while (0) -static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, - unsigned long addr, - struct page *page, pte_t * page_table, int write_access) +void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t entry) { - unsigned long i; - pte_t entry; + int i; + + for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { + set_pte_at(mm, addr, ptep, entry); + ptep++; + addr += PAGE_SIZE; + pte_val(entry) += PAGE_SIZE; + } +} - add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE); +pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) +{ + pte_t entry; + int i; - if (write_access) - entry = pte_mkwrite(pte_mkdirty(mk_pte(page, - vma->vm_page_prot))); - else - entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot)); - entry = pte_mkyoung(entry); - mk_pte_huge(entry); + entry = *ptep; for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { - set_pte_at(mm, addr, page_table, entry); - page_table++; + pte_clear(mm, addr, ptep); addr += PAGE_SIZE; - - pte_val(entry) += PAGE_SIZE; + ptep++; } + + return entry; } /* @@ -100,79 +104,6 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len) return 0; } -int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, - struct vm_area_struct *vma) -{ - pte_t *src_pte, *dst_pte, entry; - struct page *ptepage; - unsigned long addr = vma->vm_start; - unsigned long end = vma->vm_end; - int i; - - while (addr < end) { - dst_pte = huge_pte_alloc(dst, addr); - if (!dst_pte) - goto nomem; - src_pte = huge_pte_offset(src, addr); - BUG_ON(!src_pte || pte_none(*src_pte)); - entry = *src_pte; - ptepage = pte_page(entry); - get_page(ptepage); - for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { - set_pte_at(dst, addr, dst_pte, entry); - pte_val(entry) += PAGE_SIZE; - dst_pte++; - addr += PAGE_SIZE; - } - add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE); - } - return 0; - -nomem: - return -ENOMEM; -} - -int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, - struct page **pages, struct vm_area_struct **vmas, - unsigned long *position, int *length, int i) -{ - unsigned long vaddr = *position; - int remainder = *length; - - WARN_ON(!is_vm_hugetlb_page(vma)); - - while (vaddr < vma->vm_end && remainder) { - if (pages) { - pte_t *pte; - struct page *page; - - pte = huge_pte_offset(mm, vaddr); - - /* hugetlb should be locked, and hence, prefaulted */ - BUG_ON(!pte || pte_none(*pte)); - - page = pte_page(*pte); - - WARN_ON(!PageCompound(page)); - - get_page(page); - pages[i] = page; - } - - if (vmas) - vmas[i] = vma; - - vaddr += PAGE_SIZE; - --remainder; - ++i; - } - - *length = remainder; - *position = vaddr; - - return i; -} - struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { @@ -190,34 +121,6 @@ struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, return NULL; } -void unmap_hugepage_range(struct vm_area_struct *vma, - unsigned long start, unsigned long end) -{ - struct mm_struct *mm = vma->vm_mm; - unsigned long address; - pte_t *pte; - struct page *page; - int i; - - BUG_ON(start & (HPAGE_SIZE - 1)); - BUG_ON(end & (HPAGE_SIZE - 1)); - - for (address = start; address < end; address += HPAGE_SIZE) { - pte = huge_pte_offset(mm, address); - BUG_ON(!pte); - if (pte_none(*pte)) - continue; - page = pte_page(*pte); - put_page(page); - for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) { - pte_clear(mm, address+(i*PAGE_SIZE), pte); - pte++; - } - } - add_mm_counter(mm, rss, -((end - start) >> PAGE_SHIFT)); - flush_tlb_range(vma, start, end); -} - static void context_reload(void *__data) { struct mm_struct *mm = __data; @@ -226,12 +129,8 @@ static void context_reload(void *__data) load_secondary_context(mm); } -int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma) +void hugetlb_prefault_arch_hook(struct mm_struct *mm) { - struct mm_struct *mm = current->mm; - unsigned long addr; - int ret = 0; - /* On UltraSPARC-III+ and later, configure the second half of * the Data-TLB for huge pages. */ @@ -261,50 +160,4 @@ int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma) } spin_unlock(&ctx_alloc_lock); } - - BUG_ON(vma->vm_start & ~HPAGE_MASK); - BUG_ON(vma->vm_end & ~HPAGE_MASK); - - spin_lock(&mm->page_table_lock); - for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) { - unsigned long idx; - pte_t *pte = huge_pte_alloc(mm, addr); - struct page *page; - - if (!pte) { - ret = -ENOMEM; - goto out; - } - if (!pte_none(*pte)) - continue; - - idx = ((addr - vma->vm_start) >> HPAGE_SHIFT) - + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); - page = find_get_page(mapping, idx); - if (!page) { - /* charge the fs quota first */ - if (hugetlb_get_quota(mapping)) { - ret = -ENOMEM; - goto out; - } - page = alloc_huge_page(); - if (!page) { - hugetlb_put_quota(mapping); - ret = -ENOMEM; - goto out; - } - ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC); - if (! ret) { - unlock_page(page); - } else { - hugetlb_put_quota(mapping); - free_huge_page(page); - goto out; - } - } - set_huge_pte(mm, vma, addr, page, pte, vma->vm_flags & VM_WRITE); - } -out: - spin_unlock(&mm->page_table_lock); - return ret; } diff --git a/include/asm-i386/page.h b/include/asm-i386/page.h index ed13969fa2d6..41400d342d44 100644 --- a/include/asm-i386/page.h +++ b/include/asm-i386/page.h @@ -68,6 +68,7 @@ typedef struct { unsigned long pgprot; } pgprot_t; #define HPAGE_MASK (~(HPAGE_SIZE - 1)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) #define HAVE_ARCH_HUGETLB_UNMAPPED_AREA +#define ARCH_HAS_HUGETLB_CLEAN_STALE_PGTABLE #endif #define pgd_val(x) ((x).pgd) diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h index 8d60c2b4b003..e9efe148fdf7 100644 --- a/include/asm-i386/pgtable.h +++ b/include/asm-i386/pgtable.h @@ -236,6 +236,7 @@ static inline pte_t pte_mkexec(pte_t pte) { (pte).pte_low |= _PAGE_USER; return static inline pte_t pte_mkdirty(pte_t pte) { (pte).pte_low |= _PAGE_DIRTY; return pte; } static inline pte_t pte_mkyoung(pte_t pte) { (pte).pte_low |= _PAGE_ACCESSED; return pte; } static inline pte_t pte_mkwrite(pte_t pte) { (pte).pte_low |= _PAGE_RW; return pte; } +static inline pte_t pte_mkhuge(pte_t pte) { (pte).pte_low |= _PAGE_PRESENT | _PAGE_PSE; return pte; } #ifdef CONFIG_X86_PAE # include @@ -275,7 +276,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, */ #define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot)) -#define mk_pte_huge(entry) ((entry).pte_low |= _PAGE_PRESENT | _PAGE_PSE) static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) { diff --git a/include/asm-ia64/pgtable.h b/include/asm-ia64/pgtable.h index fcc9c3344ab4..48586e08f432 100644 --- a/include/asm-ia64/pgtable.h +++ b/include/asm-ia64/pgtable.h @@ -283,6 +283,7 @@ ia64_phys_addr_valid (unsigned long addr) #define pte_mkyoung(pte) (__pte(pte_val(pte) | _PAGE_A)) #define pte_mkclean(pte) (__pte(pte_val(pte) & ~_PAGE_D)) #define pte_mkdirty(pte) (__pte(pte_val(pte) | _PAGE_D)) +#define pte_mkhuge(pte) (__pte(pte_val(pte) | _PAGE_P)) /* * Macro to a page protection value as "uncacheable". Note that "protection" is really a diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h index 4c6d129e7d91..180467be8e7b 100644 --- a/include/asm-sh/page.h +++ b/include/asm-sh/page.h @@ -31,6 +31,7 @@ #define HPAGE_SIZE (1UL << HPAGE_SHIFT) #define HPAGE_MASK (~(HPAGE_SIZE-1)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT-PAGE_SHIFT) +#define ARCH_HAS_SETCLEAR_HUGE_PTE #endif #ifdef __KERNEL__ diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h index cd847a47a9aa..ecb909572d3f 100644 --- a/include/asm-sh/pgtable.h +++ b/include/asm-sh/pgtable.h @@ -196,6 +196,7 @@ static inline pte_t pte_mkexec(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _ static inline pte_t pte_mkdirty(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; } static inline pte_t pte_mkyoung(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; } static inline pte_t pte_mkwrite(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW)); return pte; } +static inline pte_t pte_mkhuge(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_SZHUGE)); return pte; } /* * Macro and implementation to make a page protection as uncachable. diff --git a/include/asm-sh64/page.h b/include/asm-sh64/page.h index e1f7f5a41210..d6167f1c0e99 100644 --- a/include/asm-sh64/page.h +++ b/include/asm-sh64/page.h @@ -41,6 +41,7 @@ #define HPAGE_SIZE (1UL << HPAGE_SHIFT) #define HPAGE_MASK (~(HPAGE_SIZE-1)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT-PAGE_SHIFT) +#define ARCH_HAS_SETCLEAR_HUGE_PTE #endif #ifdef __KERNEL__ diff --git a/include/asm-sh64/pgtable.h b/include/asm-sh64/pgtable.h index 525e1523ef5f..78ac6be2d9ef 100644 --- a/include/asm-sh64/pgtable.h +++ b/include/asm-sh64/pgtable.h @@ -430,6 +430,8 @@ extern inline pte_t pte_mkwrite(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | extern inline pte_t pte_mkexec(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_EXECUTE)); return pte; } extern inline pte_t pte_mkdirty(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; } extern inline pte_t pte_mkyoung(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; } +extern inline pte_t pte_mkhuge(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_SZHUGE)); return pte; } + /* * Conversion functions: convert a page and protection to a page entry. diff --git a/include/asm-sparc64/page.h b/include/asm-sparc64/page.h index 219ea043a14a..b87dbbd64bc9 100644 --- a/include/asm-sparc64/page.h +++ b/include/asm-sparc64/page.h @@ -95,6 +95,8 @@ typedef unsigned long pgprot_t; #define HPAGE_SIZE (_AC(1,UL) << HPAGE_SHIFT) #define HPAGE_MASK (~(HPAGE_SIZE - 1UL)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) +#define ARCH_HAS_SETCLEAR_HUGE_PTE +#define ARCH_HAS_HUGETLB_PREFAULT_HOOK #endif #define TASK_UNMAPPED_BASE (test_thread_flag(TIF_32BIT) ? \ diff --git a/include/asm-sparc64/pgtable.h b/include/asm-sparc64/pgtable.h index ae2cd5b09a7c..1ae00c5087f1 100644 --- a/include/asm-sparc64/pgtable.h +++ b/include/asm-sparc64/pgtable.h @@ -286,6 +286,7 @@ static inline pte_t pte_modify(pte_t orig_pte, pgprot_t new_prot) #define pte_mkyoung(pte) (__pte(pte_val(pte) | _PAGE_ACCESSED | _PAGE_R)) #define pte_mkwrite(pte) (__pte(pte_val(pte) | _PAGE_WRITE)) #define pte_mkdirty(pte) (__pte(pte_val(pte) | _PAGE_MODIFIED | _PAGE_W)) +#define pte_mkhuge(pte) (__pte(pte_val(pte) | _PAGE_SZHUGE)) /* to find an entry in a page-table-directory. */ #define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1)) diff --git a/include/asm-x86_64/page.h b/include/asm-x86_64/page.h index f43048035a03..9ce338c3a71e 100644 --- a/include/asm-x86_64/page.h +++ b/include/asm-x86_64/page.h @@ -28,6 +28,7 @@ #define HPAGE_SIZE ((1UL) << HPAGE_SHIFT) #define HPAGE_MASK (~(HPAGE_SIZE - 1)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) +#define ARCH_HAS_HUGETLB_CLEAN_STALE_PGTABLE #ifdef __KERNEL__ #ifndef __ASSEMBLY__ diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h index db2a0efbf573..4eec176c3c39 100644 --- a/include/asm-x86_64/pgtable.h +++ b/include/asm-x86_64/pgtable.h @@ -253,6 +253,7 @@ extern inline int pte_young(pte_t pte) { return pte_val(pte) & _PAGE_ACCESSED; extern inline int pte_write(pte_t pte) { return pte_val(pte) & _PAGE_RW; } static inline int pte_file(pte_t pte) { return pte_val(pte) & _PAGE_FILE; } +#define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT) extern inline pte_t pte_rdprotect(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_USER)); return pte; } extern inline pte_t pte_exprotect(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_USER)); return pte; } extern inline pte_t pte_mkclean(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_DIRTY)); return pte; } @@ -263,6 +264,7 @@ extern inline pte_t pte_mkexec(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _ extern inline pte_t pte_mkdirty(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; } extern inline pte_t pte_mkyoung(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; } extern inline pte_t pte_mkwrite(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW)); return pte; } +extern inline pte_t pte_mkhuge(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | __LARGE_PTE)); return pte; } struct vm_area_struct; @@ -290,7 +292,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, */ #define pgprot_noncached(prot) (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT)) -#define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT) static inline int pmd_large(pmd_t pte) { return (pmd_val(pte) & __LARGE_PTE) == __LARGE_PTE; } diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 6af1ae4a8211..f529d1442815 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -4,6 +4,7 @@ #ifdef CONFIG_HUGETLB_PAGE #include +#include struct ctl_table; @@ -22,12 +23,6 @@ int hugetlb_report_meminfo(char *); int hugetlb_report_node_meminfo(int, char *); int is_hugepage_mem_enough(size_t); unsigned long hugetlb_total_pages(void); -struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, - int write); -struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, - pmd_t *pmd, int write); -int is_aligned_hugepage_range(unsigned long addr, unsigned long len); -int pmd_huge(pmd_t pmd); struct page *alloc_huge_page(void); void free_huge_page(struct page *); @@ -35,6 +30,17 @@ extern unsigned long max_huge_pages; extern const unsigned long hugetlb_zero, hugetlb_infinity; extern int sysctl_hugetlb_shm_group; +/* arch callbacks */ + +pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr); +pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr); +struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, + int write); +struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, + pmd_t *pmd, int write); +int is_aligned_hugepage_range(unsigned long addr, unsigned long len); +int pmd_huge(pmd_t pmd); + #ifndef ARCH_HAS_HUGEPAGE_ONLY_RANGE #define is_hugepage_only_range(mm, addr, len) 0 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) \ @@ -48,6 +54,28 @@ extern int sysctl_hugetlb_shm_group; int prepare_hugepage_range(unsigned long addr, unsigned long len); #endif +#ifndef ARCH_HAS_SETCLEAR_HUGE_PTE +#define set_huge_pte_at(mm, addr, ptep, pte) set_pte_at(mm, addr, ptep, pte) +#define huge_ptep_get_and_clear(mm, addr, ptep) ptep_get_and_clear(mm, addr, ptep) +#else +void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pte); +pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, + pte_t *ptep); +#endif + +#ifndef ARCH_HAS_HUGETLB_PREFAULT_HOOK +#define hugetlb_prefault_arch_hook(mm) do { } while (0) +#else +void hugetlb_prefault_arch_hook(struct mm_struct *mm); +#endif + +#ifndef ARCH_HAS_HUGETLB_CLEAN_STALE_PGTABLE +#define hugetlb_clean_stale_pgtable(pte) BUG() +#else +void hugetlb_clean_stale_pgtable(pte_t *pte); +#endif + #else /* !CONFIG_HUGETLB_PAGE */ static inline int is_vm_hugetlb_page(struct vm_area_struct *vma) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 4eb5ae3fbe10..fbd1111ea119 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -7,10 +7,14 @@ #include #include #include -#include #include #include #include +#include +#include +#include + +#include const unsigned long hugetlb_zero = 0, hugetlb_infinity = ~0UL; static unsigned long nr_huge_pages, free_huge_pages; @@ -249,6 +253,72 @@ struct vm_operations_struct hugetlb_vm_ops = { .nopage = hugetlb_nopage, }; +static pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page) +{ + pte_t entry; + + if (vma->vm_flags & VM_WRITE) { + entry = + pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot))); + } else { + entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot)); + } + entry = pte_mkyoung(entry); + entry = pte_mkhuge(entry); + + return entry; +} + +int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, + struct vm_area_struct *vma) +{ + pte_t *src_pte, *dst_pte, entry; + struct page *ptepage; + unsigned long addr = vma->vm_start; + unsigned long end = vma->vm_end; + + while (addr < end) { + dst_pte = huge_pte_alloc(dst, addr); + if (!dst_pte) + goto nomem; + src_pte = huge_pte_offset(src, addr); + BUG_ON(!src_pte || pte_none(*src_pte)); /* prefaulted */ + entry = *src_pte; + ptepage = pte_page(entry); + get_page(ptepage); + add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE); + set_huge_pte_at(dst, addr, dst_pte, entry); + addr += HPAGE_SIZE; + } + return 0; + +nomem: + return -ENOMEM; +} + +void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + struct mm_struct *mm = vma->vm_mm; + unsigned long address; + pte_t pte; + struct page *page; + + WARN_ON(!is_vm_hugetlb_page(vma)); + BUG_ON(start & ~HPAGE_MASK); + BUG_ON(end & ~HPAGE_MASK); + + for (address = start; address < end; address += HPAGE_SIZE) { + pte = huge_ptep_get_and_clear(mm, address, huge_pte_offset(mm, address)); + if (pte_none(pte)) + continue; + page = pte_page(pte); + put_page(page); + } + add_mm_counter(mm, rss, -((end - start) >> PAGE_SHIFT)); + flush_tlb_range(vma, start, end); +} + void zap_hugepage_range(struct vm_area_struct *vma, unsigned long start, unsigned long length) { @@ -258,3 +328,108 @@ void zap_hugepage_range(struct vm_area_struct *vma, unmap_hugepage_range(vma, start, start + length); spin_unlock(&mm->page_table_lock); } + +int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma) +{ + struct mm_struct *mm = current->mm; + unsigned long addr; + int ret = 0; + + WARN_ON(!is_vm_hugetlb_page(vma)); + BUG_ON(vma->vm_start & ~HPAGE_MASK); + BUG_ON(vma->vm_end & ~HPAGE_MASK); + + hugetlb_prefault_arch_hook(mm); + + spin_lock(&mm->page_table_lock); + for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) { + unsigned long idx; + pte_t *pte = huge_pte_alloc(mm, addr); + struct page *page; + + if (!pte) { + ret = -ENOMEM; + goto out; + } + if (! pte_none(*pte)) + hugetlb_clean_stale_pgtable(pte); + + idx = ((addr - vma->vm_start) >> HPAGE_SHIFT) + + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); + page = find_get_page(mapping, idx); + if (!page) { + /* charge the fs quota first */ + if (hugetlb_get_quota(mapping)) { + ret = -ENOMEM; + goto out; + } + page = alloc_huge_page(); + if (!page) { + hugetlb_put_quota(mapping); + ret = -ENOMEM; + goto out; + } + ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC); + if (! ret) { + unlock_page(page); + } else { + hugetlb_put_quota(mapping); + free_huge_page(page); + goto out; + } + } + add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE); + set_huge_pte_at(mm, addr, pte, make_huge_pte(vma, page)); + } +out: + spin_unlock(&mm->page_table_lock); + return ret; +} + +int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, + struct page **pages, struct vm_area_struct **vmas, + unsigned long *position, int *length, int i) +{ + unsigned long vpfn, vaddr = *position; + int remainder = *length; + + BUG_ON(!is_vm_hugetlb_page(vma)); + + vpfn = vaddr/PAGE_SIZE; + while (vaddr < vma->vm_end && remainder) { + + if (pages) { + pte_t *pte; + struct page *page; + + /* Some archs (sparc64, sh*) have multiple + * pte_ts to each hugepage. We have to make + * sure we get the first, for the page + * indexing below to work. */ + pte = huge_pte_offset(mm, vaddr & HPAGE_MASK); + + /* hugetlb should be locked, and hence, prefaulted */ + WARN_ON(!pte || pte_none(*pte)); + + page = &pte_page(*pte)[vpfn % (HPAGE_SIZE/PAGE_SIZE)]; + + WARN_ON(!PageCompound(page)); + + get_page(page); + pages[i] = page; + } + + if (vmas) + vmas[i] = vma; + + vaddr += PAGE_SIZE; + ++vpfn; + --remainder; + ++i; + } + + *length = remainder; + *position = vaddr; + + return i; +} -- cgit v1.2.3 From e7c8d5c9955a4d2e88e36b640563f5d6d5aba48a Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Tue, 21 Jun 2005 17:14:47 -0700 Subject: [PATCH] node local per-cpu-pages This patch modifies the way pagesets in struct zone are managed. Each zone has a per-cpu array of pagesets. So any particular CPU has some memory in each zone structure which belongs to itself. Even if that CPU is not local to that zone. So the patch relocates the pagesets for each cpu to the node that is nearest to the cpu instead of allocating the pagesets in the (possibly remote) target zone. This means that the operations to manage pages on remote zone can be done with information available locally. We play a macro trick so that non-NUMA pmachines avoid the additional pointer chase on the page allocator fastpath. AIM7 benchmark on a 32 CPU SGI Altix w/o patches: Tasks jobs/min jti jobs/min/task real cpu 1 484.68 100 484.6769 12.01 1.97 Fri Mar 25 11:01:42 2005 100 27140.46 89 271.4046 21.44 148.71 Fri Mar 25 11:02:04 2005 200 30792.02 82 153.9601 37.80 296.72 Fri Mar 25 11:02:42 2005 300 32209.27 81 107.3642 54.21 451.34 Fri Mar 25 11:03:37 2005 400 34962.83 78 87.4071 66.59 588.97 Fri Mar 25 11:04:44 2005 500 31676.92 75 63.3538 91.87 742.71 Fri Mar 25 11:06:16 2005 600 36032.69 73 60.0545 96.91 885.44 Fri Mar 25 11:07:54 2005 700 35540.43 77 50.7720 114.63 1024.28 Fri Mar 25 11:09:49 2005 800 33906.70 74 42.3834 137.32 1181.65 Fri Mar 25 11:12:06 2005 900 34120.67 73 37.9119 153.51 1325.26 Fri Mar 25 11:14:41 2005 1000 34802.37 74 34.8024 167.23 1465.26 Fri Mar 25 11:17:28 2005 with slab API changes and pageset patch: Tasks jobs/min jti jobs/min/task real cpu 1 485.00 100 485.0000 12.00 1.96 Fri Mar 25 11:46:18 2005 100 28000.96 89 280.0096 20.79 150.45 Fri Mar 25 11:46:39 2005 200 32285.80 79 161.4290 36.05 293.37 Fri Mar 25 11:47:16 2005 300 40424.15 84 134.7472 43.19 438.42 Fri Mar 25 11:47:59 2005 400 39155.01 79 97.8875 59.46 590.05 Fri Mar 25 11:48:59 2005 500 37881.25 82 75.7625 76.82 730.19 Fri Mar 25 11:50:16 2005 600 39083.14 78 65.1386 89.35 872.79 Fri Mar 25 11:51:46 2005 700 38627.83 77 55.1826 105.47 1022.46 Fri Mar 25 11:53:32 2005 800 39631.94 78 49.5399 117.48 1169.94 Fri Mar 25 11:55:30 2005 900 36903.70 79 41.0041 141.94 1310.78 Fri Mar 25 11:57:53 2005 1000 36201.23 77 36.2012 160.77 1458.31 Fri Mar 25 12:00:34 2005 Signed-off-by: Christoph Lameter Signed-off-by: Shobhit Dayal Signed-off-by: Shai Fultheim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 2 +- include/linux/mm.h | 6 ++ include/linux/mmzone.h | 11 ++- init/main.c | 1 + mm/mempolicy.c | 2 +- mm/page_alloc.c | 211 +++++++++++++++++++++++++++++++++++++++++-------- 6 files changed, 195 insertions(+), 38 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/node.c b/drivers/base/node.c index 5d4517ccc422..904b27caf697 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -87,7 +87,7 @@ static ssize_t node_read_numastat(struct sys_device * dev, char * buf) for (i = 0; i < MAX_NR_ZONES; i++) { struct zone *z = &pg->node_zones[i]; for (cpu = 0; cpu < NR_CPUS; cpu++) { - struct per_cpu_pageset *ps = &z->pageset[cpu]; + struct per_cpu_pageset *ps = zone_pcp(z,cpu); numa_hit += ps->numa_hit; numa_miss += ps->numa_miss; numa_foreign += ps->numa_foreign; diff --git a/include/linux/mm.h b/include/linux/mm.h index 17518fe0b311..1813b162b0a8 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -691,6 +691,12 @@ extern void show_mem(void); extern void si_meminfo(struct sysinfo * val); extern void si_meminfo_node(struct sysinfo *val, int nid); +#ifdef CONFIG_NUMA +extern void setup_per_cpu_pageset(void); +#else +static inline void setup_per_cpu_pageset(void) {} +#endif + /* prio_tree.c */ void vma_prio_tree_add(struct vm_area_struct *, struct vm_area_struct *old); void vma_prio_tree_insert(struct vm_area_struct *, struct prio_tree_root *); diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 18fed8b67943..4733d35d8223 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -63,6 +63,12 @@ struct per_cpu_pageset { #endif } ____cacheline_aligned_in_smp; +#ifdef CONFIG_NUMA +#define zone_pcp(__z, __cpu) ((__z)->pageset[(__cpu)]) +#else +#define zone_pcp(__z, __cpu) (&(__z)->pageset[(__cpu)]) +#endif + #define ZONE_DMA 0 #define ZONE_NORMAL 1 #define ZONE_HIGHMEM 2 @@ -122,8 +128,11 @@ struct zone { */ unsigned long lowmem_reserve[MAX_NR_ZONES]; +#ifdef CONFIG_NUMA + struct per_cpu_pageset *pageset[NR_CPUS]; +#else struct per_cpu_pageset pageset[NR_CPUS]; - +#endif /* * free areas of different sizes */ diff --git a/init/main.c b/init/main.c index 40bf367ffdf1..d324801729ba 100644 --- a/init/main.c +++ b/init/main.c @@ -490,6 +490,7 @@ asmlinkage void __init start_kernel(void) vfs_caches_init_early(); mem_init(); kmem_cache_init(); + setup_per_cpu_pageset(); numa_policy_init(); if (late_time_init) late_time_init(); diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 08c41da429cf..39252c732db2 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -721,7 +721,7 @@ static struct page *alloc_page_interleave(unsigned int __nocast gfp, unsigned or zl = NODE_DATA(nid)->node_zonelists + (gfp & GFP_ZONEMASK); page = __alloc_pages(gfp, order, zl); if (page && page_zone(page) == zl->zones[0]) { - zl->zones[0]->pageset[get_cpu()].interleave_hit++; + zone_pcp(zl->zones[0],get_cpu())->interleave_hit++; put_cpu(); } return page; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 2019c1b19254..95cbd30a67b8 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -71,6 +71,11 @@ EXPORT_SYMBOL(nr_swap_pages); struct zone *zone_table[1 << (ZONES_SHIFT + NODES_SHIFT)]; EXPORT_SYMBOL(zone_table); +#ifdef CONFIG_NUMA +static struct per_cpu_pageset + pageset_table[MAX_NR_ZONES*MAX_NUMNODES*NR_CPUS] __initdata; +#endif + static char *zone_names[MAX_NR_ZONES] = { "DMA", "Normal", "HighMem" }; int min_free_kbytes = 1024; @@ -520,7 +525,7 @@ static void __drain_pages(unsigned int cpu) for_each_zone(zone) { struct per_cpu_pageset *pset; - pset = &zone->pageset[cpu]; + pset = zone_pcp(zone, cpu); for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) { struct per_cpu_pages *pcp; @@ -583,12 +588,12 @@ static void zone_statistics(struct zonelist *zonelist, struct zone *z) local_irq_save(flags); cpu = smp_processor_id(); - p = &z->pageset[cpu]; + p = zone_pcp(z,cpu); if (pg == orig) { - z->pageset[cpu].numa_hit++; + p->numa_hit++; } else { p->numa_miss++; - zonelist->zones[0]->pageset[cpu].numa_foreign++; + zone_pcp(zonelist->zones[0], cpu)->numa_foreign++; } if (pg == NODE_DATA(numa_node_id())) p->local_node++; @@ -615,7 +620,7 @@ static void fastcall free_hot_cold_page(struct page *page, int cold) if (PageAnon(page)) page->mapping = NULL; free_pages_check(__FUNCTION__, page); - pcp = &zone->pageset[get_cpu()].pcp[cold]; + pcp = &zone_pcp(zone, get_cpu())->pcp[cold]; local_irq_save(flags); if (pcp->count >= pcp->high) pcp->count -= free_pages_bulk(zone, pcp->batch, &pcp->list, 0); @@ -659,7 +664,7 @@ buffered_rmqueue(struct zone *zone, int order, unsigned int __nocast gfp_flags) if (order == 0) { struct per_cpu_pages *pcp; - pcp = &zone->pageset[get_cpu()].pcp[cold]; + pcp = &zone_pcp(zone, get_cpu())->pcp[cold]; local_irq_save(flags); if (pcp->count <= pcp->low) pcp->count += rmqueue_bulk(zone, 0, @@ -1262,7 +1267,7 @@ void show_free_areas(void) if (!cpu_possible(cpu)) continue; - pageset = zone->pageset + cpu; + pageset = zone_pcp(zone, cpu); for (temperature = 0; temperature < 2; temperature++) printk("cpu %d %s: low %d, high %d, batch %d\n", @@ -1645,6 +1650,157 @@ void zone_init_free_lists(struct pglist_data *pgdat, struct zone *zone, memmap_init_zone((size), (nid), (zone), (start_pfn)) #endif +static int __devinit zone_batchsize(struct zone *zone) +{ + int batch; + + /* + * The per-cpu-pages pools are set to around 1000th of the + * size of the zone. But no more than 1/4 of a meg - there's + * no point in going beyond the size of L2 cache. + * + * OK, so we don't know how big the cache is. So guess. + */ + batch = zone->present_pages / 1024; + if (batch * PAGE_SIZE > 256 * 1024) + batch = (256 * 1024) / PAGE_SIZE; + batch /= 4; /* We effectively *= 4 below */ + if (batch < 1) + batch = 1; + + /* + * Clamp the batch to a 2^n - 1 value. Having a power + * of 2 value was found to be more likely to have + * suboptimal cache aliasing properties in some cases. + * + * For example if 2 tasks are alternately allocating + * batches of pages, one task can end up with a lot + * of pages of one half of the possible page colors + * and the other with pages of the other colors. + */ + batch = (1 << fls(batch + batch/2)) - 1; + return batch; +} + +#ifdef CONFIG_NUMA +/* + * Dynamicaly allocate memory for the + * per cpu pageset array in struct zone. + */ +static int __devinit process_zones(int cpu) +{ + struct zone *zone, *dzone; + int i; + + for_each_zone(zone) { + struct per_cpu_pageset *npageset = NULL; + + npageset = kmalloc_node(sizeof(struct per_cpu_pageset), + GFP_KERNEL, cpu_to_node(cpu)); + if (!npageset) { + zone->pageset[cpu] = NULL; + goto bad; + } + + if (zone->pageset[cpu]) { + memcpy(npageset, zone->pageset[cpu], + sizeof(struct per_cpu_pageset)); + + /* Relocate lists */ + for (i = 0; i < 2; i++) { + INIT_LIST_HEAD(&npageset->pcp[i].list); + list_splice(&zone->pageset[cpu]->pcp[i].list, + &npageset->pcp[i].list); + } + } else { + struct per_cpu_pages *pcp; + unsigned long batch; + + batch = zone_batchsize(zone); + + pcp = &npageset->pcp[0]; /* hot */ + pcp->count = 0; + pcp->low = 2 * batch; + pcp->high = 6 * batch; + pcp->batch = 1 * batch; + INIT_LIST_HEAD(&pcp->list); + + pcp = &npageset->pcp[1]; /* cold*/ + pcp->count = 0; + pcp->low = 0; + pcp->high = 2 * batch; + pcp->batch = 1 * batch; + INIT_LIST_HEAD(&pcp->list); + } + zone->pageset[cpu] = npageset; + } + + return 0; +bad: + for_each_zone(dzone) { + if (dzone == zone) + break; + kfree(dzone->pageset[cpu]); + dzone->pageset[cpu] = NULL; + } + return -ENOMEM; +} + +static inline void free_zone_pagesets(int cpu) +{ +#ifdef CONFIG_NUMA + struct zone *zone; + + for_each_zone(zone) { + struct per_cpu_pageset *pset = zone_pcp(zone, cpu); + + zone_pcp(zone, cpu) = NULL; + kfree(pset); + } +#endif +} + +static int __devinit pageset_cpuup_callback(struct notifier_block *nfb, + unsigned long action, + void *hcpu) +{ + int cpu = (long)hcpu; + int ret = NOTIFY_OK; + + switch (action) { + case CPU_UP_PREPARE: + if (process_zones(cpu)) + ret = NOTIFY_BAD; + break; +#ifdef CONFIG_HOTPLUG_CPU + case CPU_DEAD: + free_zone_pagesets(cpu); + break; +#endif + default: + break; + } + return ret; +} + +static struct notifier_block pageset_notifier = + { &pageset_cpuup_callback, NULL, 0 }; + +void __init setup_per_cpu_pageset() +{ + int err; + + /* Initialize per_cpu_pageset for cpu 0. + * A cpuup callback will do this for every cpu + * as it comes online + */ + err = process_zones(smp_processor_id()); + BUG_ON(err); + register_cpu_notifier(&pageset_notifier); +} + +#endif + /* * Set up the zone data structures: * - mark all pages reserved @@ -1687,43 +1843,28 @@ static void __init free_area_init_core(struct pglist_data *pgdat, zone->temp_priority = zone->prev_priority = DEF_PRIORITY; - /* - * The per-cpu-pages pools are set to around 1000th of the - * size of the zone. But no more than 1/4 of a meg - there's - * no point in going beyond the size of L2 cache. - * - * OK, so we don't know how big the cache is. So guess. - */ - batch = zone->present_pages / 1024; - if (batch * PAGE_SIZE > 256 * 1024) - batch = (256 * 1024) / PAGE_SIZE; - batch /= 4; /* We effectively *= 4 below */ - if (batch < 1) - batch = 1; - - /* - * Clamp the batch to a 2^n - 1 value. Having a power - * of 2 value was found to be more likely to have - * suboptimal cache aliasing properties in some cases. - * - * For example if 2 tasks are alternately allocating - * batches of pages, one task can end up with a lot - * of pages of one half of the possible page colors - * and the other with pages of the other colors. - */ - batch = (1 << fls(batch + batch/2)) - 1; + batch = zone_batchsize(zone); for (cpu = 0; cpu < NR_CPUS; cpu++) { struct per_cpu_pages *pcp; +#ifdef CONFIG_NUMA + struct per_cpu_pageset *pgset; + pgset = &pageset_table[nid*MAX_NR_ZONES*NR_CPUS + + (j * NR_CPUS) + cpu]; + + zone->pageset[cpu] = pgset; +#else + struct per_cpu_pageset *pgset = zone_pcp(zone, cpu); +#endif - pcp = &zone->pageset[cpu].pcp[0]; /* hot */ + pcp = &pgset->pcp[0]; /* hot */ pcp->count = 0; pcp->low = 2 * batch; pcp->high = 6 * batch; pcp->batch = 1 * batch; INIT_LIST_HEAD(&pcp->list); - pcp = &zone->pageset[cpu].pcp[1]; /* cold */ + pcp = &pgset->pcp[1]; /* cold */ pcp->count = 0; pcp->low = 0; pcp->high = 2 * batch; @@ -1929,7 +2070,7 @@ static int zoneinfo_show(struct seq_file *m, void *arg) struct per_cpu_pageset *pageset; int j; - pageset = &zone->pageset[i]; + pageset = zone_pcp(zone, i); for (j = 0; j < ARRAY_SIZE(pageset->pcp); j++) { if (pageset->pcp[j].count) break; -- cgit v1.2.3 From 1363c3cd8603a913a27e2995dccbd70d5312d8e6 Mon Sep 17 00:00:00 2001 From: Wolfgang Wander Date: Tue, 21 Jun 2005 17:14:49 -0700 Subject: [PATCH] Avoiding mmap fragmentation Ingo recently introduced a great speedup for allocating new mmaps using the free_area_cache pointer which boosts the specweb SSL benchmark by 4-5% and causes huge performance increases in thread creation. The downside of this patch is that it does lead to fragmentation in the mmap-ed areas (visible via /proc/self/maps), such that some applications that work fine under 2.4 kernels quickly run out of memory on any 2.6 kernel. The problem is twofold: 1) the free_area_cache is used to continue a search for memory where the last search ended. Before the change new areas were always searched from the base address on. So now new small areas are cluttering holes of all sizes throughout the whole mmap-able region whereas before small holes tended to close holes near the base leaving holes far from the base large and available for larger requests. 2) the free_area_cache also is set to the location of the last munmap-ed area so in scenarios where we allocate e.g. five regions of 1K each, then free regions 4 2 3 in this order the next request for 1K will be placed in the position of the old region 3, whereas before we appended it to the still active region 1, placing it at the location of the old region 2. Before we had 1 free region of 2K, now we only get two free regions of 1K -> fragmentation. The patch addresses thes issues by introducing yet another cache descriptor cached_hole_size that contains the largest known hole size below the current free_area_cache. If a new request comes in the size is compared against the cached_hole_size and if the request can be filled with a hole below free_area_cache the search is started from the base instead. The results look promising: Whereas 2.6.12-rc4 fragments quickly and my (earlier posted) leakme.c test program terminates after 50000+ iterations with 96 distinct and fragmented maps in /proc/self/maps it performs nicely (as expected) with thread creation, Ingo's test_str02 with 20000 threads requires 0.7s system time. Taking out Ingo's patch (un-patch available per request) by basically deleting all mentions of free_area_cache from the kernel and starting the search for new memory always at the respective bases we observe: leakme terminates successfully with 11 distinctive hardly fragmented areas in /proc/self/maps but thread creating is gringdingly slow: 30+s(!) system time for Ingo's test_str02 with 20000 threads. Now - drumroll ;-) the appended patch works fine with leakme: it ends with only 7 distinct areas in /proc/self/maps and also thread creation seems sufficiently fast with 0.71s for 20000 threads. Signed-off-by: Wolfgang Wander Credit-to: "Richard Purdie" Signed-off-by: Ken Chen Acked-by: Ingo Molnar (partly) Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/mm/mmap.c | 10 +++++++- arch/i386/mm/hugetlbpage.c | 34 ++++++++++++++++++++++---- arch/ppc64/mm/hugetlbpage.c | 34 ++++++++++++++++++++++---- arch/sh/kernel/sys_sh.c | 8 +++++++ arch/sparc64/kernel/sys_sparc.c | 8 +++++++ arch/x86_64/ia32/ia32_aout.c | 1 + arch/x86_64/kernel/sys_x86_64.c | 9 +++++++ fs/binfmt_aout.c | 1 + fs/binfmt_elf.c | 1 + fs/hugetlbfs/inode.c | 3 +++ include/linux/sched.h | 11 +++++---- kernel/fork.c | 2 ++ mm/mmap.c | 53 +++++++++++++++++++++++++++++++---------- mm/nommu.c | 2 +- 14 files changed, 147 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/arch/arm/mm/mmap.c b/arch/arm/mm/mmap.c index 32c4b0e35b37..3de7f84b53c2 100644 --- a/arch/arm/mm/mmap.c +++ b/arch/arm/mm/mmap.c @@ -73,7 +73,12 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, (!vma || addr + len <= vma->vm_start)) return addr; } - start_addr = addr = mm->free_area_cache; + if (len > mm->cached_hole_size) { + start_addr = addr = mm->free_area_cache; + } else { + start_addr = addr = TASK_UNMAPPED_BASE; + mm->cached_hole_size = 0; + } full_search: if (do_align) @@ -90,6 +95,7 @@ full_search: */ if (start_addr != TASK_UNMAPPED_BASE) { start_addr = addr = TASK_UNMAPPED_BASE; + mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; @@ -101,6 +107,8 @@ full_search: mm->free_area_cache = addr + len; return addr; } + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; addr = vma->vm_end; if (do_align) addr = COLOUR_ALIGN(addr, pgoff); diff --git a/arch/i386/mm/hugetlbpage.c b/arch/i386/mm/hugetlbpage.c index 5aa06001a4bd..3b099f32b948 100644 --- a/arch/i386/mm/hugetlbpage.c +++ b/arch/i386/mm/hugetlbpage.c @@ -140,7 +140,12 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file, struct vm_area_struct *vma; unsigned long start_addr; - start_addr = mm->free_area_cache; + if (len > mm->cached_hole_size) { + start_addr = mm->free_area_cache; + } else { + start_addr = TASK_UNMAPPED_BASE; + mm->cached_hole_size = 0; + } full_search: addr = ALIGN(start_addr, HPAGE_SIZE); @@ -154,6 +159,7 @@ full_search: */ if (start_addr != TASK_UNMAPPED_BASE) { start_addr = TASK_UNMAPPED_BASE; + mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; @@ -162,6 +168,8 @@ full_search: mm->free_area_cache = addr + len; return addr; } + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; addr = ALIGN(vma->vm_end, HPAGE_SIZE); } } @@ -173,12 +181,17 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev_vma; unsigned long base = mm->mmap_base, addr = addr0; + unsigned long largest_hole = mm->cached_hole_size; int first_time = 1; /* don't allow allocations above current base */ if (mm->free_area_cache > base) mm->free_area_cache = base; + if (len <= largest_hole) { + largest_hole = 0; + mm->free_area_cache = base; + } try_again: /* make sure it can fit in the remaining address space */ if (mm->free_area_cache < len) @@ -199,13 +212,21 @@ try_again: * vma->vm_start, use it: */ if (addr + len <= vma->vm_start && - (!prev_vma || (addr >= prev_vma->vm_end))) + (!prev_vma || (addr >= prev_vma->vm_end))) { /* remember the address as a hint for next time */ - return (mm->free_area_cache = addr); - else + mm->cached_hole_size = largest_hole; + return (mm->free_area_cache = addr); + } else { /* pull free_area_cache down to the first hole */ - if (mm->free_area_cache == vma->vm_end) + if (mm->free_area_cache == vma->vm_end) { mm->free_area_cache = vma->vm_start; + mm->cached_hole_size = largest_hole; + } + } + + /* remember the largest hole we saw so far */ + if (addr + largest_hole < vma->vm_start) + largest_hole = vma->vm_start - addr; /* try just below the current vma->vm_start */ addr = (vma->vm_start - len) & HPAGE_MASK; @@ -218,6 +239,7 @@ fail: */ if (first_time) { mm->free_area_cache = base; + largest_hole = 0; first_time = 0; goto try_again; } @@ -228,6 +250,7 @@ fail: * allocations. */ mm->free_area_cache = TASK_UNMAPPED_BASE; + mm->cached_hole_size = ~0UL; addr = hugetlb_get_unmapped_area_bottomup(file, addr0, len, pgoff, flags); @@ -235,6 +258,7 @@ fail: * Restore the topdown base: */ mm->free_area_cache = base; + mm->cached_hole_size = ~0UL; return addr; } diff --git a/arch/ppc64/mm/hugetlbpage.c b/arch/ppc64/mm/hugetlbpage.c index b4ab766f5980..fdcfe97c75c1 100644 --- a/arch/ppc64/mm/hugetlbpage.c +++ b/arch/ppc64/mm/hugetlbpage.c @@ -292,7 +292,12 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, && !is_hugepage_only_range(mm, addr,len)) return addr; } - start_addr = addr = mm->free_area_cache; + if (len > mm->cached_hole_size) { + start_addr = addr = mm->free_area_cache; + } else { + start_addr = addr = TASK_UNMAPPED_BASE; + mm->cached_hole_size = 0; + } full_search: vma = find_vma(mm, addr); @@ -316,6 +321,8 @@ full_search: mm->free_area_cache = addr + len; return addr; } + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; addr = vma->vm_end; vma = vma->vm_next; } @@ -323,6 +330,7 @@ full_search: /* Make sure we didn't miss any holes */ if (start_addr != TASK_UNMAPPED_BASE) { start_addr = addr = TASK_UNMAPPED_BASE; + mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; @@ -344,6 +352,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, struct vm_area_struct *vma, *prev_vma; struct mm_struct *mm = current->mm; unsigned long base = mm->mmap_base, addr = addr0; + unsigned long largest_hole = mm->cached_hole_size; int first_time = 1; /* requested length too big for entire address space */ @@ -364,6 +373,10 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, return addr; } + if (len <= largest_hole) { + largest_hole = 0; + mm->free_area_cache = base; + } try_again: /* make sure it can fit in the remaining address space */ if (mm->free_area_cache < len) @@ -392,13 +405,21 @@ hugepage_recheck: * vma->vm_start, use it: */ if (addr+len <= vma->vm_start && - (!prev_vma || (addr >= prev_vma->vm_end))) + (!prev_vma || (addr >= prev_vma->vm_end))) { /* remember the address as a hint for next time */ - return (mm->free_area_cache = addr); - else + mm->cached_hole_size = largest_hole; + return (mm->free_area_cache = addr); + } else { /* pull free_area_cache down to the first hole */ - if (mm->free_area_cache == vma->vm_end) + if (mm->free_area_cache == vma->vm_end) { mm->free_area_cache = vma->vm_start; + mm->cached_hole_size = largest_hole; + } + } + + /* remember the largest hole we saw so far */ + if (addr + largest_hole < vma->vm_start) + largest_hole = vma->vm_start - addr; /* try just below the current vma->vm_start */ addr = vma->vm_start-len; @@ -411,6 +432,7 @@ fail: */ if (first_time) { mm->free_area_cache = base; + largest_hole = 0; first_time = 0; goto try_again; } @@ -421,11 +443,13 @@ fail: * allocations. */ mm->free_area_cache = TASK_UNMAPPED_BASE; + mm->cached_hole_size = ~0UL; addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); /* * Restore the topdown base: */ mm->free_area_cache = base; + mm->cached_hole_size = ~0UL; return addr; } diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c index df5ac294c379..917b2f32f260 100644 --- a/arch/sh/kernel/sys_sh.c +++ b/arch/sh/kernel/sys_sh.c @@ -79,6 +79,10 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, (!vma || addr + len <= vma->vm_start)) return addr; } + if (len <= mm->cached_hole_size) { + mm->cached_hole_size = 0; + mm->free_area_cache = TASK_UNMAPPED_BASE; + } if (flags & MAP_PRIVATE) addr = PAGE_ALIGN(mm->free_area_cache); else @@ -95,6 +99,7 @@ full_search: */ if (start_addr != TASK_UNMAPPED_BASE) { start_addr = addr = TASK_UNMAPPED_BASE; + mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; @@ -106,6 +111,9 @@ full_search: mm->free_area_cache = addr + len; return addr; } + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; + addr = vma->vm_end; if (!(flags & MAP_PRIVATE)) addr = COLOUR_ALIGN(addr); diff --git a/arch/sparc64/kernel/sys_sparc.c b/arch/sparc64/kernel/sys_sparc.c index 0077f02f4b37..5f8c822a2b4a 100644 --- a/arch/sparc64/kernel/sys_sparc.c +++ b/arch/sparc64/kernel/sys_sparc.c @@ -84,6 +84,10 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi return addr; } + if (len <= mm->cached_hole_size) { + mm->cached_hole_size = 0; + mm->free_area_cache = TASK_UNMAPPED_BASE; + } start_addr = addr = mm->free_area_cache; task_size -= len; @@ -103,6 +107,7 @@ full_search: if (task_size < addr) { if (start_addr != TASK_UNMAPPED_BASE) { start_addr = addr = TASK_UNMAPPED_BASE; + mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; @@ -114,6 +119,9 @@ full_search: mm->free_area_cache = addr + len; return addr; } + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; + addr = vma->vm_end; if (do_color_align) addr = COLOUR_ALIGN(addr, pgoff); diff --git a/arch/x86_64/ia32/ia32_aout.c b/arch/x86_64/ia32/ia32_aout.c index 1965efc974dc..c12edf5d97f0 100644 --- a/arch/x86_64/ia32/ia32_aout.c +++ b/arch/x86_64/ia32/ia32_aout.c @@ -312,6 +312,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) current->mm->brk = ex.a_bss + (current->mm->start_brk = N_BSSADDR(ex)); current->mm->free_area_cache = TASK_UNMAPPED_BASE; + current->mm->cached_hole_size = 0; set_mm_counter(current->mm, rss, 0); current->mm->mmap = NULL; diff --git a/arch/x86_64/kernel/sys_x86_64.c b/arch/x86_64/kernel/sys_x86_64.c index d9798dd433fc..cc7821c68851 100644 --- a/arch/x86_64/kernel/sys_x86_64.c +++ b/arch/x86_64/kernel/sys_x86_64.c @@ -105,6 +105,11 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, (!vma || addr + len <= vma->vm_start)) return addr; } + if (((flags & MAP_32BIT) || test_thread_flag(TIF_IA32)) + && len <= mm->cached_hole_size) { + mm->cached_hole_size = 0; + mm->free_area_cache = begin; + } addr = mm->free_area_cache; if (addr < begin) addr = begin; @@ -120,6 +125,7 @@ full_search: */ if (start_addr != begin) { start_addr = addr = begin; + mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; @@ -131,6 +137,9 @@ full_search: mm->free_area_cache = addr + len; return addr; } + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; + addr = vma->vm_end; } } diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index 009b8920c1ff..dd9baabaf016 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -316,6 +316,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) current->mm->brk = ex.a_bss + (current->mm->start_brk = N_BSSADDR(ex)); current->mm->free_area_cache = current->mm->mmap_base; + current->mm->cached_hole_size = 0; set_mm_counter(current->mm, rss, 0); current->mm->mmap = NULL; diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index f8f6b6b76179..7976a238f0a3 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -775,6 +775,7 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs) change some of these later */ set_mm_counter(current->mm, rss, 0); current->mm->free_area_cache = current->mm->mmap_base; + current->mm->cached_hole_size = 0; retval = setup_arg_pages(bprm, randomize_stack_top(STACK_TOP), executable_stack); if (retval < 0) { diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 2af3338f891b..3a9b6d179cbd 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -122,6 +122,9 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, start_addr = mm->free_area_cache; + if (len <= mm->cached_hole_size) + start_addr = TASK_UNMAPPED_BASE; + full_search: addr = ALIGN(start_addr, HPAGE_SIZE); diff --git a/include/linux/sched.h b/include/linux/sched.h index 4dbb109022f3..b58afd97a180 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -201,8 +201,8 @@ extern unsigned long arch_get_unmapped_area_topdown(struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags); -extern void arch_unmap_area(struct vm_area_struct *area); -extern void arch_unmap_area_topdown(struct vm_area_struct *area); +extern void arch_unmap_area(struct mm_struct *, unsigned long); +extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long); #define set_mm_counter(mm, member, value) (mm)->_##member = (value) #define get_mm_counter(mm, member) ((mm)->_##member) @@ -218,9 +218,10 @@ struct mm_struct { unsigned long (*get_unmapped_area) (struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags); - void (*unmap_area) (struct vm_area_struct *area); - unsigned long mmap_base; /* base of mmap area */ - unsigned long free_area_cache; /* first hole */ + void (*unmap_area) (struct mm_struct *mm, unsigned long addr); + unsigned long mmap_base; /* base of mmap area */ + unsigned long cached_hole_size; /* if non-zero, the largest hole below free_area_cache */ + unsigned long free_area_cache; /* first hole of size cached_hole_size or larger */ pgd_t * pgd; atomic_t mm_users; /* How many users with user space? */ atomic_t mm_count; /* How many references to "struct mm_struct" (users count as 1) */ diff --git a/kernel/fork.c b/kernel/fork.c index f42a17f88699..876b31cd822d 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -194,6 +194,7 @@ static inline int dup_mmap(struct mm_struct * mm, struct mm_struct * oldmm) mm->mmap = NULL; mm->mmap_cache = NULL; mm->free_area_cache = oldmm->mmap_base; + mm->cached_hole_size = ~0UL; mm->map_count = 0; set_mm_counter(mm, rss, 0); set_mm_counter(mm, anon_rss, 0); @@ -322,6 +323,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm) mm->ioctx_list = NULL; mm->default_kioctx = (struct kioctx)INIT_KIOCTX(mm->default_kioctx, *mm); mm->free_area_cache = TASK_UNMAPPED_BASE; + mm->cached_hole_size = ~0UL; if (likely(!mm_alloc_pgd(mm))) { mm->def_flags = 0; diff --git a/mm/mmap.c b/mm/mmap.c index de54acd9942f..9da23c1ef9dc 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1175,7 +1175,12 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, (!vma || addr + len <= vma->vm_start)) return addr; } - start_addr = addr = mm->free_area_cache; + if (len > mm->cached_hole_size) { + start_addr = addr = mm->free_area_cache; + } else { + start_addr = addr = TASK_UNMAPPED_BASE; + mm->cached_hole_size = 0; + } full_search: for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { @@ -1186,7 +1191,9 @@ full_search: * some holes. */ if (start_addr != TASK_UNMAPPED_BASE) { - start_addr = addr = TASK_UNMAPPED_BASE; + addr = TASK_UNMAPPED_BASE; + start_addr = addr; + mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; @@ -1198,19 +1205,22 @@ full_search: mm->free_area_cache = addr + len; return addr; } + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; addr = vma->vm_end; } } #endif -void arch_unmap_area(struct vm_area_struct *area) +void arch_unmap_area(struct mm_struct *mm, unsigned long addr) { /* * Is this a new hole at the lowest possible address? */ - if (area->vm_start >= TASK_UNMAPPED_BASE && - area->vm_start < area->vm_mm->free_area_cache) - area->vm_mm->free_area_cache = area->vm_start; + if (addr >= TASK_UNMAPPED_BASE && addr < mm->free_area_cache) { + mm->free_area_cache = addr; + mm->cached_hole_size = ~0UL; + } } /* @@ -1240,6 +1250,12 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, return addr; } + /* check if free_area_cache is useful for us */ + if (len <= mm->cached_hole_size) { + mm->cached_hole_size = 0; + mm->free_area_cache = mm->mmap_base; + } + /* either no address requested or can't fit in requested address hole */ addr = mm->free_area_cache; @@ -1264,6 +1280,10 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, /* remember the address as a hint for next time */ return (mm->free_area_cache = addr); + /* remember the largest hole we saw so far */ + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; + /* try just below the current vma->vm_start */ addr = vma->vm_start-len; } while (len < vma->vm_start); @@ -1274,28 +1294,30 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, * can happen with large stack limits and large mmap() * allocations. */ - mm->free_area_cache = TASK_UNMAPPED_BASE; + mm->cached_hole_size = ~0UL; + mm->free_area_cache = TASK_UNMAPPED_BASE; addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); /* * Restore the topdown base: */ mm->free_area_cache = mm->mmap_base; + mm->cached_hole_size = ~0UL; return addr; } #endif -void arch_unmap_area_topdown(struct vm_area_struct *area) +void arch_unmap_area_topdown(struct mm_struct *mm, unsigned long addr) { /* * Is this a new hole at the highest possible address? */ - if (area->vm_end > area->vm_mm->free_area_cache) - area->vm_mm->free_area_cache = area->vm_end; + if (addr > mm->free_area_cache) + mm->free_area_cache = addr; /* dont allow allocations above current base */ - if (area->vm_mm->free_area_cache > area->vm_mm->mmap_base) - area->vm_mm->free_area_cache = area->vm_mm->mmap_base; + if (mm->free_area_cache > mm->mmap_base) + mm->free_area_cache = mm->mmap_base; } unsigned long @@ -1595,7 +1617,6 @@ static void unmap_vma(struct mm_struct *mm, struct vm_area_struct *area) if (area->vm_flags & VM_LOCKED) area->vm_mm->locked_vm -= len >> PAGE_SHIFT; vm_stat_unaccount(area); - area->vm_mm->unmap_area(area); remove_vm_struct(area); } @@ -1649,6 +1670,7 @@ detach_vmas_to_be_unmapped(struct mm_struct *mm, struct vm_area_struct *vma, { struct vm_area_struct **insertion_point; struct vm_area_struct *tail_vma = NULL; + unsigned long addr; insertion_point = (prev ? &prev->vm_next : &mm->mmap); do { @@ -1659,6 +1681,11 @@ detach_vmas_to_be_unmapped(struct mm_struct *mm, struct vm_area_struct *vma, } while (vma && vma->vm_start < end); *insertion_point = vma; tail_vma->vm_next = NULL; + if (mm->unmap_area == arch_unmap_area) + addr = prev ? prev->vm_end : mm->mmap_base; + else + addr = vma ? vma->vm_start : mm->mmap_base; + mm->unmap_area(mm, addr); mm->mmap_cache = NULL; /* Kill the cache. */ } diff --git a/mm/nommu.c b/mm/nommu.c index c53e9c8f6b4a..ce74452c02d9 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -1067,7 +1067,7 @@ unsigned long arch_get_unmapped_area(struct file *file, unsigned long addr, return -ENOMEM; } -void arch_unmap_area(struct vm_area_struct *area) +void arch_unmap_area(struct mm_struct *mm, unsigned long addr) { } -- cgit v1.2.3 From cbe37d093707762fc0abb280781e6a82a9d8d568 Mon Sep 17 00:00:00 2001 From: Badari Pulavarty Date: Tue, 21 Jun 2005 17:14:52 -0700 Subject: [PATCH] mm: remove PG_highmem Remove PG_highmem, to save a page flag. Use is_highmem() instead. It'll generate a little more code, but we don't use PageHigheMem() in many places. Signed-off-by: Badari Pulavarty Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/mm/init.c | 1 - arch/i386/mm/init.c | 1 - arch/mips/mm/init.c | 1 - arch/ppc/mm/init.c | 1 - arch/sparc/mm/init.c | 1 - arch/um/kernel/mem.c | 1 - include/linux/page-flags.h | 31 +++++++++++++++---------------- 7 files changed, 15 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/arch/frv/mm/init.c b/arch/frv/mm/init.c index 41958f57c838..79433159b5f0 100644 --- a/arch/frv/mm/init.c +++ b/arch/frv/mm/init.c @@ -169,7 +169,6 @@ void __init mem_init(void) struct page *page = &mem_map[pfn]; ClearPageReserved(page); - set_bit(PG_highmem, &page->flags); set_page_count(page, 1); __free_page(page); totalram_pages++; diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c index 7a7ea3737265..8766c771bb45 100644 --- a/arch/i386/mm/init.c +++ b/arch/i386/mm/init.c @@ -269,7 +269,6 @@ void __init one_highpage_init(struct page *page, int pfn, int bad_ppro) { if (page_is_ram(pfn) && !(bad_ppro && page_kills_ppro(pfn))) { ClearPageReserved(page); - set_bit(PG_highmem, &page->flags); set_page_count(page, 1); __free_page(page); totalhigh_pages++; diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index b027ce7efbc6..73843c528778 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -232,7 +232,6 @@ void __init mem_init(void) #ifdef CONFIG_LIMITED_DMA set_page_address(page, lowmem_page_address(page)); #endif - set_bit(PG_highmem, &page->flags); set_page_count(page, 1); __free_page(page); totalhigh_pages++; diff --git a/arch/ppc/mm/init.c b/arch/ppc/mm/init.c index 363c157e3617..c8529d004cfa 100644 --- a/arch/ppc/mm/init.c +++ b/arch/ppc/mm/init.c @@ -469,7 +469,6 @@ void __init mem_init(void) struct page *page = mem_map + pfn; ClearPageReserved(page); - set_bit(PG_highmem, &page->flags); set_page_count(page, 1); __free_page(page); totalhigh_pages++; diff --git a/arch/sparc/mm/init.c b/arch/sparc/mm/init.c index a2dea69b2f07..ec2e05028a10 100644 --- a/arch/sparc/mm/init.c +++ b/arch/sparc/mm/init.c @@ -384,7 +384,6 @@ void map_high_region(unsigned long start_pfn, unsigned long end_pfn) struct page *page = pfn_to_page(tmp); ClearPageReserved(page); - set_bit(PG_highmem, &page->flags); set_page_count(page, 1); __free_page(page); totalhigh_pages++; diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index c22825f13e40..5597bd39e6b5 100644 --- a/arch/um/kernel/mem.c +++ b/arch/um/kernel/mem.c @@ -53,7 +53,6 @@ static void setup_highmem(unsigned long highmem_start, for(i = 0; i < highmem_len >> PAGE_SHIFT; i++){ page = &mem_map[highmem_pfn + i]; ClearPageReserved(page); - set_bit(PG_highmem, &page->flags); set_page_count(page, 1); __free_page(page); } diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 39ab8c6b5652..df313891db10 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -61,21 +61,20 @@ #define PG_active 6 #define PG_slab 7 /* slab debug (Suparna wants this) */ -#define PG_highmem 8 -#define PG_checked 9 /* kill me in 2.5.. */ -#define PG_arch_1 10 -#define PG_reserved 11 - -#define PG_private 12 /* Has something at ->private */ -#define PG_writeback 13 /* Page is under writeback */ -#define PG_nosave 14 /* Used for system suspend/resume */ -#define PG_compound 15 /* Part of a compound page */ - -#define PG_swapcache 16 /* Swap page: swp_entry_t in private */ -#define PG_mappedtodisk 17 /* Has blocks allocated on-disk */ -#define PG_reclaim 18 /* To be reclaimed asap */ -#define PG_nosave_free 19 /* Free, should not be written */ -#define PG_uncached 20 /* Page has been mapped as uncached */ +#define PG_checked 8 /* kill me in 2.5.. */ +#define PG_arch_1 9 +#define PG_reserved 10 +#define PG_private 11 /* Has something at ->private */ + +#define PG_writeback 12 /* Page is under writeback */ +#define PG_nosave 13 /* Used for system suspend/resume */ +#define PG_compound 14 /* Part of a compound page */ +#define PG_swapcache 15 /* Swap page: swp_entry_t in private */ + +#define PG_mappedtodisk 16 /* Has blocks allocated on-disk */ +#define PG_reclaim 17 /* To be reclaimed asap */ +#define PG_nosave_free 18 /* Free, should not be written */ +#define PG_uncached 19 /* Page has been mapped as uncached */ /* * Global page accounting. One instance per CPU. Only unsigned longs are @@ -215,7 +214,7 @@ extern void __mod_page_state(unsigned offset, unsigned long delta); #define TestSetPageSlab(page) test_and_set_bit(PG_slab, &(page)->flags) #ifdef CONFIG_HIGHMEM -#define PageHighMem(page) test_bit(PG_highmem, &(page)->flags) +#define PageHighMem(page) is_highmem(page_zone(page)) #else #define PageHighMem(page) 0 /* needed to optimize away at compile time */ #endif -- cgit v1.2.3 From 1ad539b2bd89bf2e129123eb24d5bcc4484a35de Mon Sep 17 00:00:00 2001 From: Darren Hart Date: Tue, 21 Jun 2005 17:14:53 -0700 Subject: [PATCH] vm: try_to_free_pages unused argument try_to_free_pages accepts a third argument, order, but hasn't used it since before 2.6.0. The following patch removes the argument and updates all the calls to try_to_free_pages. Signed-off-by: Darren Hart Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 2 +- include/linux/swap.h | 2 +- mm/page_alloc.c | 2 +- mm/vmscan.c | 3 +-- 4 files changed, 4 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/fs/buffer.c b/fs/buffer.c index 7e9e409feaa7..0befa724ab98 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -528,7 +528,7 @@ static void free_more_memory(void) for_each_pgdat(pgdat) { zones = pgdat->node_zonelists[GFP_NOFS&GFP_ZONEMASK].zones; if (*zones) - try_to_free_pages(zones, GFP_NOFS, 0); + try_to_free_pages(zones, GFP_NOFS); } } diff --git a/include/linux/swap.h b/include/linux/swap.h index 0d21e682d99d..2343f999e6e1 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -172,7 +172,7 @@ extern int rotate_reclaimable_page(struct page *page); extern void swap_setup(void); /* linux/mm/vmscan.c */ -extern int try_to_free_pages(struct zone **, unsigned int, unsigned int); +extern int try_to_free_pages(struct zone **, unsigned int); extern int zone_reclaim(struct zone *, unsigned int, unsigned int); extern int shrink_all_memory(int); extern int vm_swappiness; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 95cbd30a67b8..70f710286ced 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -859,7 +859,7 @@ rebalance: reclaim_state.reclaimed_slab = 0; p->reclaim_state = &reclaim_state; - did_some_progress = try_to_free_pages(zones, gfp_mask, order); + did_some_progress = try_to_free_pages(zones, gfp_mask); p->reclaim_state = NULL; p->flags &= ~PF_MEMALLOC; diff --git a/mm/vmscan.c b/mm/vmscan.c index 24da725a30f0..4b8e62a19370 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -919,8 +919,7 @@ shrink_caches(struct zone **zones, struct scan_control *sc) * holds filesystem locks which prevent writeout this might not work, and the * allocation attempt will fail. */ -int try_to_free_pages(struct zone **zones, - unsigned int gfp_mask, unsigned int order) +int try_to_free_pages(struct zone **zones, unsigned int gfp_mask) { int priority; int ret = 0; -- cgit v1.2.3 From 83e5d8f7253cb7b14472385a6d57df1e9f848e8e Mon Sep 17 00:00:00 2001 From: Benjamin LaHaise Date: Tue, 21 Jun 2005 17:14:54 -0700 Subject: [PATCH] __mod_page_state(): pass unsigned long instead of unsigned By making the offset argument of __mod_page_state an unsigned long instead of unsigned, we can avoid forcing the compiler to sign extend a usually constant argument. This saves 1 instruction on x86-64. Signed-off-by: Benjamin LaHaise Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page-flags.h | 2 +- mm/page_alloc.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index df313891db10..f2ee9b2332e3 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -136,7 +136,7 @@ struct page_state { extern void get_page_state(struct page_state *ret); extern void get_full_page_state(struct page_state *ret); extern unsigned long __read_page_state(unsigned offset); -extern void __mod_page_state(unsigned offset, unsigned long delta); +extern void __mod_page_state(unsigned long offset, unsigned long delta); #define read_page_state(member) \ __read_page_state(offsetof(struct page_state, member)) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 70f710286ced..c976628f9d95 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1158,7 +1158,7 @@ unsigned long __read_page_state(unsigned offset) return ret; } -void __mod_page_state(unsigned offset, unsigned long delta) +void __mod_page_state(unsigned long offset, unsigned long delta) { unsigned long flags; void* ptr; -- cgit v1.2.3 From c2f29ea111e3344ed48257c2a142c3db514e1529 Mon Sep 17 00:00:00 2001 From: Benjamin LaHaise Date: Tue, 21 Jun 2005 17:14:55 -0700 Subject: [PATCH] __read_page_state(): pass unsigned long instead of unsigned By making the offset argument of __read_page_state an unsigned long instead of unsigned, we can avoid forcing the compiler to sign extend a usually constant argument. This saves 1 instruction on x86-64. Signed-off-by: Benjamin LaHaise Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page-flags.h | 2 +- mm/page_alloc.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index f2ee9b2332e3..f5a6695d4d21 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -135,7 +135,7 @@ struct page_state { extern void get_page_state(struct page_state *ret); extern void get_full_page_state(struct page_state *ret); -extern unsigned long __read_page_state(unsigned offset); +extern unsigned long __read_page_state(unsigned long offset); extern void __mod_page_state(unsigned long offset, unsigned long delta); #define read_page_state(member) \ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index c976628f9d95..10446a646374 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1144,7 +1144,7 @@ void get_full_page_state(struct page_state *ret) __get_page_state(ret, sizeof(*ret) / sizeof(unsigned long)); } -unsigned long __read_page_state(unsigned offset) +unsigned long __read_page_state(unsigned long offset) { unsigned long ret = 0; int cpu; -- cgit v1.2.3 From 4ae7c03943fca73f23bc0cdb938070f41b98101f Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Tue, 21 Jun 2005 17:14:57 -0700 Subject: [PATCH] Periodically drain non local pagesets The pageset array can potentially acquire a huge amount of memory on large NUMA systems. F.e. on a system with 512 processors and 256 nodes there will be 256*512 pagesets. If each pageset only holds 5 pages then we are talking about 655360 pages.With a 16K page size on IA64 this results in potentially 10 Gigabytes of memory being trapped in pagesets. The typical cases are much less for smaller systems but there is still the potential of memory being trapped in off node pagesets. Off node memory may be rarely used if local memory is available and so we may potentially have memory in seldom used pagesets without this patch. The slab allocator flushes its per cpu caches every 2 seconds. The following patch flushes the off node pageset caches in the same way by tying into the slab flush. The patch also changes /proc/zoneinfo to include the number of pages currently in each pageset. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 5 +++++ mm/page_alloc.c | 35 +++++++++++++++++++++++++++++++++-- mm/slab.c | 1 + 3 files changed, 39 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 208535fd4832..8d6bf608b199 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -133,5 +133,10 @@ extern void FASTCALL(free_cold_page(struct page *page)); #define free_page(addr) free_pages((addr),0) void page_alloc_init(void); +#ifdef CONFIG_NUMA +void drain_remote_pages(void); +#else +static inline void drain_remote_pages(void) { }; +#endif #endif /* __LINUX_GFP_H */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index be05d17bd7df..a95e72d7f945 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -516,6 +516,36 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order, return allocated; } +#ifdef CONFIG_NUMA +/* Called from the slab reaper to drain remote pagesets */ +void drain_remote_pages(void) +{ + struct zone *zone; + int i; + unsigned long flags; + + local_irq_save(flags); + for_each_zone(zone) { + struct per_cpu_pageset *pset; + + /* Do not drain local pagesets */ + if (zone->zone_pgdat->node_id == numa_node_id()) + continue; + + pset = zone->pageset[smp_processor_id()]; + for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) { + struct per_cpu_pages *pcp; + + pcp = &pset->pcp[i]; + if (pcp->count) + pcp->count -= free_pages_bulk(zone, pcp->count, + &pcp->list, 0); + } + } + local_irq_restore(flags); +} +#endif + #if defined(CONFIG_PM) || defined(CONFIG_HOTPLUG_CPU) static void __drain_pages(unsigned int cpu) { @@ -1271,12 +1301,13 @@ void show_free_areas(void) pageset = zone_pcp(zone, cpu); for (temperature = 0; temperature < 2; temperature++) - printk("cpu %d %s: low %d, high %d, batch %d\n", + printk("cpu %d %s: low %d, high %d, batch %d used:%d\n", cpu, temperature ? "cold" : "hot", pageset->pcp[temperature].low, pageset->pcp[temperature].high, - pageset->pcp[temperature].batch); + pageset->pcp[temperature].batch, + pageset->pcp[temperature].count); } } diff --git a/mm/slab.c b/mm/slab.c index c78d343b3c5f..93cbbbb39f42 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2851,6 +2851,7 @@ next: } check_irq_on(); up(&cache_chain_sem); + drain_remote_pages(); /* Setup the next iteration */ schedule_delayed_work(&__get_cpu_var(reap_work), REAPTIMEOUT_CPUC + smp_processor_id()); } -- cgit v1.2.3 From f14f75b81187cdbe10cc53a521bf9fdf97b59f8c Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Tue, 21 Jun 2005 17:15:02 -0700 Subject: [PATCH] ia64 uncached alloc This patch contains the ia64 uncached page allocator and the generic allocator (genalloc). The uncached allocator was formerly part of the SN2 mspec driver but there are several other users of it so it has been split off from the driver. The generic allocator can be used by device driver to manage special memory etc. The generic allocator is based on the allocator from the sym53c8xx_2 driver. Various users on ia64 needs uncached memory. The SGI SN architecture requires it for inter-partition communication between partitions within a large NUMA cluster. The specific user for this is the XPC code. Another application is large MPI style applications which use it for synchronization, on SN this can be done using special 'fetchop' operations but it also benefits non SN hardware which may use regular uncached memory for this purpose. Performance of doing this through uncached vs cached memory is pretty substantial. This is handled by the mspec driver which I will push out in a seperate patch. Rather than creating a specific allocator for just uncached memory I came up with genalloc which is a generic purpose allocator that can be used by device drivers and other subsystems as they please. For instance to handle onboard device memory. It was derived from the sym53c7xx_2 driver's allocator which is also an example of a potential user (I am refraining from modifying sym2 right now as it seems to have been under fairly heavy development recently). On ia64 memory has various properties within a granule, ie. it isn't safe to access memory as uncached within the same granule as currently has memory accessed in cached mode. The regular system therefore doesn't utilize memory in the lower granules which is mixed in with device PAL code etc. The uncached driver walks the EFI memmap and pulls out the spill uncached pages and sticks them into the uncached pool. Only after these chunks have been utilized, will it start converting regular cached memory into uncached memory. Hence the reason for the EFI related code additions. Signed-off-by: Jes Sorensen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/Kconfig | 4 + arch/ia64/kernel/Makefile | 1 + arch/ia64/kernel/efi.c | 32 ++++++ arch/ia64/kernel/uncached.c | 246 ++++++++++++++++++++++++++++++++++++++++++++ include/asm-ia64/uncached.h | 12 +++ include/linux/genalloc.h | 40 +++++++ lib/Kconfig | 6 ++ lib/Makefile | 1 + lib/genalloc.c | 188 +++++++++++++++++++++++++++++++++ 9 files changed, 530 insertions(+) create mode 100644 arch/ia64/kernel/uncached.c create mode 100644 include/asm-ia64/uncached.h create mode 100644 include/linux/genalloc.h create mode 100644 lib/genalloc.c (limited to 'include/linux') diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 3ad2c4af099c..295b5abee72f 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -50,6 +50,10 @@ config SCHED_NO_NO_OMIT_FRAME_POINTER bool default y +config IA64_UNCACHED_ALLOCATOR + bool + select GENERIC_ALLOCATOR + choice prompt "System type" default IA64_GENERIC diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile index c1a02bbc252c..4c73d8ba2e3d 100644 --- a/arch/ia64/kernel/Makefile +++ b/arch/ia64/kernel/Makefile @@ -20,6 +20,7 @@ obj-$(CONFIG_SMP) += smp.o smpboot.o domain.o obj-$(CONFIG_PERFMON) += perfmon_default_smpl.o obj-$(CONFIG_IA64_CYCLONE) += cyclone.o obj-$(CONFIG_IA64_MCA_RECOVERY) += mca_recovery.o +obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR) += uncached.o mca_recovery-y += mca_drv.o mca_drv_asm.o # The gate DSO image is built using a special linker script. diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c index 4a3b1aac43e7..179f230816ed 100644 --- a/arch/ia64/kernel/efi.c +++ b/arch/ia64/kernel/efi.c @@ -409,6 +409,38 @@ efi_memmap_walk (efi_freemem_callback_t callback, void *arg) } } +/* + * Walk the EFI memory map to pull out leftover pages in the lower + * memory regions which do not end up in the regular memory map and + * stick them into the uncached allocator + * + * The regular walk function is significantly more complex than the + * uncached walk which means it really doesn't make sense to try and + * marge the two. + */ +void __init +efi_memmap_walk_uc (efi_freemem_callback_t callback) +{ + void *efi_map_start, *efi_map_end, *p; + efi_memory_desc_t *md; + u64 efi_desc_size, start, end; + + efi_map_start = __va(ia64_boot_param->efi_memmap); + efi_map_end = efi_map_start + ia64_boot_param->efi_memmap_size; + efi_desc_size = ia64_boot_param->efi_memdesc_size; + + for (p = efi_map_start; p < efi_map_end; p += efi_desc_size) { + md = p; + if (md->attribute == EFI_MEMORY_UC) { + start = PAGE_ALIGN(md->phys_addr); + end = PAGE_ALIGN((md->phys_addr+(md->num_pages << EFI_PAGE_SHIFT)) & PAGE_MASK); + if ((*callback)(start, end, NULL) < 0) + return; + } + } +} + + /* * Look for the PAL_CODE region reported by EFI and maps it using an * ITR to enable safe PAL calls in virtual mode. See IA-64 Processor diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c new file mode 100644 index 000000000000..490dfc9ab47f --- /dev/null +++ b/arch/ia64/kernel/uncached.c @@ -0,0 +1,246 @@ +/* + * Copyright (C) 2001-2005 Silicon Graphics, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + * + * A simple uncached page allocator using the generic allocator. This + * allocator first utilizes the spare (spill) pages found in the EFI + * memmap and will then start converting cached pages to uncached ones + * at a granule at a time. Node awareness is implemented by having a + * pool of pages per node. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define DEBUG 0 + +#if DEBUG +#define dprintk printk +#else +#define dprintk(x...) do { } while (0) +#endif + +void __init efi_memmap_walk_uc (efi_freemem_callback_t callback); + +#define MAX_UNCACHED_GRANULES 5 +static int allocated_granules; + +struct gen_pool *uncached_pool[MAX_NUMNODES]; + + +static void uncached_ipi_visibility(void *data) +{ + int status; + + status = ia64_pal_prefetch_visibility(PAL_VISIBILITY_PHYSICAL); + if ((status != PAL_VISIBILITY_OK) && + (status != PAL_VISIBILITY_OK_REMOTE_NEEDED)) + printk(KERN_DEBUG "pal_prefetch_visibility() returns %i on " + "CPU %i\n", status, get_cpu()); +} + + +static void uncached_ipi_mc_drain(void *data) +{ + int status; + status = ia64_pal_mc_drain(); + if (status) + printk(KERN_WARNING "ia64_pal_mc_drain() failed with %i on " + "CPU %i\n", status, get_cpu()); +} + + +static unsigned long +uncached_get_new_chunk(struct gen_pool *poolp) +{ + struct page *page; + void *tmp; + int status, i; + unsigned long addr, node; + + if (allocated_granules >= MAX_UNCACHED_GRANULES) + return 0; + + node = poolp->private; + page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, + IA64_GRANULE_SHIFT-PAGE_SHIFT); + + dprintk(KERN_INFO "get_new_chunk page %p, addr %lx\n", + page, (unsigned long)(page-vmem_map) << PAGE_SHIFT); + + /* + * Do magic if no mem on local node! XXX + */ + if (!page) + return 0; + tmp = page_address(page); + + /* + * There's a small race here where it's possible for someone to + * access the page through /dev/mem halfway through the conversion + * to uncached - not sure it's really worth bothering about + */ + for (i = 0; i < (IA64_GRANULE_SIZE / PAGE_SIZE); i++) + SetPageUncached(&page[i]); + + flush_tlb_kernel_range(tmp, tmp + IA64_GRANULE_SIZE); + + status = ia64_pal_prefetch_visibility(PAL_VISIBILITY_PHYSICAL); + + dprintk(KERN_INFO "pal_prefetch_visibility() returns %i on cpu %i\n", + status, get_cpu()); + + if (!status) { + status = smp_call_function(uncached_ipi_visibility, NULL, 0, 1); + if (status) + printk(KERN_WARNING "smp_call_function failed for " + "uncached_ipi_visibility! (%i)\n", status); + } + + if (ia64_platform_is("sn2")) + sn_flush_all_caches((unsigned long)tmp, IA64_GRANULE_SIZE); + else + flush_icache_range((unsigned long)tmp, + (unsigned long)tmp+IA64_GRANULE_SIZE); + + ia64_pal_mc_drain(); + status = smp_call_function(uncached_ipi_mc_drain, NULL, 0, 1); + if (status) + printk(KERN_WARNING "smp_call_function failed for " + "uncached_ipi_mc_drain! (%i)\n", status); + + addr = (unsigned long)tmp - PAGE_OFFSET + __IA64_UNCACHED_OFFSET; + + allocated_granules++; + return addr; +} + + +/* + * uncached_alloc_page + * + * Allocate 1 uncached page. Allocates on the requested node. If no + * uncached pages are available on the requested node, roundrobin starting + * with higher nodes. + */ +unsigned long +uncached_alloc_page(int nid) +{ + unsigned long maddr; + + maddr = gen_pool_alloc(uncached_pool[nid], PAGE_SIZE); + + dprintk(KERN_DEBUG "uncached_alloc_page returns %lx on node %i\n", + maddr, nid); + + /* + * If no memory is availble on our local node, try the + * remaining nodes in the system. + */ + if (!maddr) { + int i; + + for (i = MAX_NUMNODES - 1; i >= 0; i--) { + if (i == nid || !node_online(i)) + continue; + maddr = gen_pool_alloc(uncached_pool[i], PAGE_SIZE); + dprintk(KERN_DEBUG "uncached_alloc_page alternate search " + "returns %lx on node %i\n", maddr, i); + if (maddr) { + break; + } + } + } + + return maddr; +} +EXPORT_SYMBOL(uncached_alloc_page); + + +/* + * uncached_free_page + * + * Free a single uncached page. + */ +void +uncached_free_page(unsigned long maddr) +{ + int node; + + node = nasid_to_cnodeid(NASID_GET(maddr)); + + dprintk(KERN_DEBUG "uncached_free_page(%lx) on node %i\n", maddr, node); + + if ((maddr & (0XFUL << 60)) != __IA64_UNCACHED_OFFSET) + panic("uncached_free_page invalid address %lx\n", maddr); + + gen_pool_free(uncached_pool[node], maddr, PAGE_SIZE); +} +EXPORT_SYMBOL(uncached_free_page); + + +/* + * uncached_build_memmap, + * + * Called at boot time to build a map of pages that can be used for + * memory special operations. + */ +static int __init +uncached_build_memmap(unsigned long start, unsigned long end, void *arg) +{ + long length; + unsigned long vstart, vend; + int node; + + length = end - start; + vstart = start + __IA64_UNCACHED_OFFSET; + vend = end + __IA64_UNCACHED_OFFSET; + + dprintk(KERN_ERR "uncached_build_memmap(%lx %lx)\n", start, end); + + memset((char *)vstart, 0, length); + + node = nasid_to_cnodeid(NASID_GET(start)); + + for (; vstart < vend ; vstart += PAGE_SIZE) { + dprintk(KERN_INFO "sticking %lx into the pool!\n", vstart); + gen_pool_free(uncached_pool[node], vstart, PAGE_SIZE); + } + + return 0; +} + + +static int __init uncached_init(void) { + int i; + + for (i = 0; i < MAX_NUMNODES; i++) { + if (!node_online(i)) + continue; + uncached_pool[i] = gen_pool_create(0, IA64_GRANULE_SHIFT, + &uncached_get_new_chunk, i); + } + + efi_memmap_walk_uc(uncached_build_memmap); + + return 0; +} + +__initcall(uncached_init); diff --git a/include/asm-ia64/uncached.h b/include/asm-ia64/uncached.h new file mode 100644 index 000000000000..b82d923b73c1 --- /dev/null +++ b/include/asm-ia64/uncached.h @@ -0,0 +1,12 @@ +/* + * Copyright (C) 2001-2005 Silicon Graphics, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + * + * Prototypes for the uncached page allocator + */ + +extern unsigned long uncached_alloc_page(int nid); +extern void uncached_free_page(unsigned long); diff --git a/include/linux/genalloc.h b/include/linux/genalloc.h new file mode 100644 index 000000000000..7fd0576a4454 --- /dev/null +++ b/include/linux/genalloc.h @@ -0,0 +1,40 @@ +/* + * Basic general purpose allocator for managing special purpose memory + * not managed by the regular kmalloc/kfree interface. + * Uses for this includes on-device special memory, uncached memory + * etc. + * + * This code is based on the buddy allocator found in the sym53c8xx_2 + * driver, adapted for general purpose use. + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include + +#define ALLOC_MIN_SHIFT 5 /* 32 bytes minimum */ +/* + * Link between free memory chunks of a given size. + */ +struct gen_pool_link { + struct gen_pool_link *next; +}; + +/* + * Memory pool descriptor. + */ +struct gen_pool { + spinlock_t lock; + unsigned long (*get_new_chunk)(struct gen_pool *); + struct gen_pool *next; + struct gen_pool_link *h; + unsigned long private; + int max_chunk_shift; +}; + +unsigned long gen_pool_alloc(struct gen_pool *poolp, int size); +void gen_pool_free(struct gen_pool *mp, unsigned long ptr, int size); +struct gen_pool *gen_pool_create(int nr_chunks, int max_chunk_shift, + unsigned long (*fp)(struct gen_pool *), + unsigned long data); diff --git a/lib/Kconfig b/lib/Kconfig index eeb45225248f..2d4d4e3bc4aa 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -39,6 +39,12 @@ config ZLIB_INFLATE config ZLIB_DEFLATE tristate +# +# Generic allocator support is selected if needed +# +config GENERIC_ALLOCATOR + boolean + # # reed solomon support is select'ed if needed # diff --git a/lib/Makefile b/lib/Makefile index 5f10cb898407..dcb4231916e2 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -30,6 +30,7 @@ obj-$(CONFIG_CRC_CCITT) += crc-ccitt.o obj-$(CONFIG_CRC32) += crc32.o obj-$(CONFIG_LIBCRC32C) += libcrc32c.o obj-$(CONFIG_GENERIC_IOMAP) += iomap.o +obj-$(CONFIG_GENERIC_ALLOCATOR) += genalloc.o obj-$(CONFIG_ZLIB_INFLATE) += zlib_inflate/ obj-$(CONFIG_ZLIB_DEFLATE) += zlib_deflate/ diff --git a/lib/genalloc.c b/lib/genalloc.c new file mode 100644 index 000000000000..d6d30d2e7166 --- /dev/null +++ b/lib/genalloc.c @@ -0,0 +1,188 @@ +/* + * Basic general purpose allocator for managing special purpose memory + * not managed by the regular kmalloc/kfree interface. + * Uses for this includes on-device special memory, uncached memory + * etc. + * + * This code is based on the buddy allocator found in the sym53c8xx_2 + * driver Copyright (C) 1999-2001 Gerard Roudier , + * and adapted for general purpose use. + * + * Copyright 2005 (C) Jes Sorensen + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + + +struct gen_pool *gen_pool_create(int nr_chunks, int max_chunk_shift, + unsigned long (*fp)(struct gen_pool *), + unsigned long data) +{ + struct gen_pool *poolp; + unsigned long tmp; + int i; + + /* + * This is really an arbitrary limit, +10 is enough for + * IA64_GRANULE_SHIFT, aka 16MB. If anyone needs a large limit + * this can be increased without problems. + */ + if ((max_chunk_shift > (PAGE_SHIFT + 10)) || + ((max_chunk_shift < ALLOC_MIN_SHIFT) && max_chunk_shift)) + return NULL; + + if (!max_chunk_shift) + max_chunk_shift = PAGE_SHIFT; + + poolp = kmalloc(sizeof(struct gen_pool), GFP_KERNEL); + if (!poolp) + return NULL; + memset(poolp, 0, sizeof(struct gen_pool)); + poolp->h = kmalloc(sizeof(struct gen_pool_link) * + (max_chunk_shift - ALLOC_MIN_SHIFT + 1), + GFP_KERNEL); + if (!poolp->h) { + printk(KERN_WARNING "gen_pool_alloc() failed to allocate\n"); + kfree(poolp); + return NULL; + } + memset(poolp->h, 0, sizeof(struct gen_pool_link) * + (max_chunk_shift - ALLOC_MIN_SHIFT + 1)); + + spin_lock_init(&poolp->lock); + poolp->get_new_chunk = fp; + poolp->max_chunk_shift = max_chunk_shift; + poolp->private = data; + + for (i = 0; i < nr_chunks; i++) { + tmp = poolp->get_new_chunk(poolp); + printk(KERN_INFO "allocated %lx\n", tmp); + if (!tmp) + break; + gen_pool_free(poolp, tmp, (1 << poolp->max_chunk_shift)); + } + + return poolp; +} +EXPORT_SYMBOL(gen_pool_create); + + +/* + * Simple power of two buddy-like generic allocator. + * Provides naturally aligned memory chunks. + */ +unsigned long gen_pool_alloc(struct gen_pool *poolp, int size) +{ + int j, i, s, max_chunk_size; + unsigned long a, flags; + struct gen_pool_link *h = poolp->h; + + max_chunk_size = 1 << poolp->max_chunk_shift; + + if (size > max_chunk_size) + return 0; + + i = 0; + + size = max(size, 1 << ALLOC_MIN_SHIFT); + s = roundup_pow_of_two(size); + + j = i; + + spin_lock_irqsave(&poolp->lock, flags); + while (!h[j].next) { + if (s == max_chunk_size) { + struct gen_pool_link *ptr; + spin_unlock_irqrestore(&poolp->lock, flags); + ptr = (struct gen_pool_link *)poolp->get_new_chunk(poolp); + spin_lock_irqsave(&poolp->lock, flags); + h[j].next = ptr; + if (h[j].next) + h[j].next->next = NULL; + break; + } + j++; + s <<= 1; + } + a = (unsigned long) h[j].next; + if (a) { + h[j].next = h[j].next->next; + /* + * This should be split into a seperate function doing + * the chunk split in order to support custom + * handling memory not physically accessible by host + */ + while (j > i) { + j -= 1; + s >>= 1; + h[j].next = (struct gen_pool_link *) (a + s); + h[j].next->next = NULL; + } + } + spin_unlock_irqrestore(&poolp->lock, flags); + return a; +} +EXPORT_SYMBOL(gen_pool_alloc); + + +/* + * Counter-part of the generic allocator. + */ +void gen_pool_free(struct gen_pool *poolp, unsigned long ptr, int size) +{ + struct gen_pool_link *q; + struct gen_pool_link *h = poolp->h; + unsigned long a, b, flags; + int i, s, max_chunk_size; + + max_chunk_size = 1 << poolp->max_chunk_shift; + + if (size > max_chunk_size) + return; + + i = 0; + + size = max(size, 1 << ALLOC_MIN_SHIFT); + s = roundup_pow_of_two(size); + + a = ptr; + + spin_lock_irqsave(&poolp->lock, flags); + while (1) { + if (s == max_chunk_size) { + ((struct gen_pool_link *)a)->next = h[i].next; + h[i].next = (struct gen_pool_link *)a; + break; + } + b = a ^ s; + q = &h[i]; + + while (q->next && q->next != (struct gen_pool_link *)b) + q = q->next; + + if (!q->next) { + ((struct gen_pool_link *)a)->next = h[i].next; + h[i].next = (struct gen_pool_link *)a; + break; + } + q->next = q->next->next; + a = a & b; + s <<= 1; + i++; + } + spin_unlock_irqrestore(&poolp->lock, flags); +} +EXPORT_SYMBOL(gen_pool_free); -- cgit v1.2.3 From 5b37b700f7c491a9320f4e29472bbaf23dded8fd Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Tue, 21 Jun 2005 17:15:18 -0700 Subject: [PATCH] ppc32: Added support for new MPC8548 family of PowerQUICC III processors Added descriptions of the new MPC8548 family processors, e500 core and peripherals. Signed-off-by: Kumar Gala Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc/kernel/cputable.c | 14 +++ arch/ppc/syslib/mpc85xx_devices.c | 185 ++++++++++++++++++++++++++++++++++++++ arch/ppc/syslib/mpc85xx_sys.c | 105 ++++++++++++++++++++++ include/asm-ppc/irq.h | 6 ++ include/asm-ppc/mpc85xx.h | 7 +- include/linux/fsl_devices.h | 8 +- 6 files changed, 323 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/arch/ppc/kernel/cputable.c b/arch/ppc/kernel/cputable.c index d44b7dc5390a..01c226008dbf 100644 --- a/arch/ppc/kernel/cputable.c +++ b/arch/ppc/kernel/cputable.c @@ -918,6 +918,20 @@ struct cpu_spec cpu_specs[] = { .dcache_bsize = 32, .num_pmcs = 4, }, + { /* e500v2 */ + .pvr_mask = 0xffff0000, + .pvr_value = 0x80210000, + .cpu_name = "e500v2", + /* xxx - galak: add CPU_FTR_MAYBE_CAN_DOZE */ + .cpu_features = CPU_FTR_SPLIT_ID_CACHE | + CPU_FTR_USE_TB | CPU_FTR_BIG_PHYS, + .cpu_user_features = PPC_FEATURE_32 | + PPC_FEATURE_HAS_MMU | PPC_FEATURE_SPE_COMP | + PPC_FEATURE_HAS_EFP_SINGLE | PPC_FEATURE_HAS_EFP_DOUBLE, + .icache_bsize = 32, + .dcache_bsize = 32, + .num_pmcs = 4, + }, #endif #if !CLASSIC_PPC { /* default match */ diff --git a/arch/ppc/syslib/mpc85xx_devices.c b/arch/ppc/syslib/mpc85xx_devices.c index 1e658ef57e75..b30c57d4518f 100644 --- a/arch/ppc/syslib/mpc85xx_devices.c +++ b/arch/ppc/syslib/mpc85xx_devices.c @@ -40,6 +40,42 @@ static struct gianfar_platform_data mpc85xx_tsec2_pdata = { .phy_reg_addr = MPC85xx_ENET1_OFFSET, }; +static struct gianfar_platform_data mpc85xx_etsec1_pdata = { + .device_flags = FSL_GIANFAR_DEV_HAS_GIGABIT | + FSL_GIANFAR_DEV_HAS_COALESCE | FSL_GIANFAR_DEV_HAS_RMON | + FSL_GIANFAR_DEV_HAS_MULTI_INTR | + FSL_GIANFAR_DEV_HAS_CSUM | FSL_GIANFAR_DEV_HAS_VLAN | + FSL_GIANFAR_DEV_HAS_EXTENDED_HASH, + .phy_reg_addr = MPC85xx_ENET1_OFFSET, +}; + +static struct gianfar_platform_data mpc85xx_etsec2_pdata = { + .device_flags = FSL_GIANFAR_DEV_HAS_GIGABIT | + FSL_GIANFAR_DEV_HAS_COALESCE | FSL_GIANFAR_DEV_HAS_RMON | + FSL_GIANFAR_DEV_HAS_MULTI_INTR | + FSL_GIANFAR_DEV_HAS_CSUM | FSL_GIANFAR_DEV_HAS_VLAN | + FSL_GIANFAR_DEV_HAS_EXTENDED_HASH, + .phy_reg_addr = MPC85xx_ENET1_OFFSET, +}; + +static struct gianfar_platform_data mpc85xx_etsec3_pdata = { + .device_flags = FSL_GIANFAR_DEV_HAS_GIGABIT | + FSL_GIANFAR_DEV_HAS_COALESCE | FSL_GIANFAR_DEV_HAS_RMON | + FSL_GIANFAR_DEV_HAS_MULTI_INTR | + FSL_GIANFAR_DEV_HAS_CSUM | FSL_GIANFAR_DEV_HAS_VLAN | + FSL_GIANFAR_DEV_HAS_EXTENDED_HASH, + .phy_reg_addr = MPC85xx_ENET1_OFFSET, +}; + +static struct gianfar_platform_data mpc85xx_etsec4_pdata = { + .device_flags = FSL_GIANFAR_DEV_HAS_GIGABIT | + FSL_GIANFAR_DEV_HAS_COALESCE | FSL_GIANFAR_DEV_HAS_RMON | + FSL_GIANFAR_DEV_HAS_MULTI_INTR | + FSL_GIANFAR_DEV_HAS_CSUM | FSL_GIANFAR_DEV_HAS_VLAN | + FSL_GIANFAR_DEV_HAS_EXTENDED_HASH, + .phy_reg_addr = MPC85xx_ENET1_OFFSET, +}; + static struct gianfar_platform_data mpc85xx_fec_pdata = { .phy_reg_addr = MPC85xx_ENET1_OFFSET, }; @@ -48,6 +84,10 @@ static struct fsl_i2c_platform_data mpc85xx_fsl_i2c_pdata = { .device_flags = FSL_I2C_DEV_SEPARATE_DFSRR, }; +static struct fsl_i2c_platform_data mpc85xx_fsl_i2c2_pdata = { + .device_flags = FSL_I2C_DEV_SEPARATE_DFSRR, +}; + static struct plat_serial8250_port serial_platform_data[] = { [0] = { .mapbase = 0x4500, @@ -536,6 +576,151 @@ struct platform_device ppc_sys_platform_devices[] = { }, }, #endif /* CONFIG_CPM2 */ + [MPC85xx_eTSEC1] = { + .name = "fsl-gianfar", + .id = 1, + .dev.platform_data = &mpc85xx_etsec1_pdata, + .num_resources = 4, + .resource = (struct resource[]) { + { + .start = MPC85xx_ENET1_OFFSET, + .end = MPC85xx_ENET1_OFFSET + + MPC85xx_ENET1_SIZE - 1, + .flags = IORESOURCE_MEM, + }, + { + .name = "tx", + .start = MPC85xx_IRQ_TSEC1_TX, + .end = MPC85xx_IRQ_TSEC1_TX, + .flags = IORESOURCE_IRQ, + }, + { + .name = "rx", + .start = MPC85xx_IRQ_TSEC1_RX, + .end = MPC85xx_IRQ_TSEC1_RX, + .flags = IORESOURCE_IRQ, + }, + { + .name = "error", + .start = MPC85xx_IRQ_TSEC1_ERROR, + .end = MPC85xx_IRQ_TSEC1_ERROR, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC85xx_eTSEC2] = { + .name = "fsl-gianfar", + .id = 2, + .dev.platform_data = &mpc85xx_etsec2_pdata, + .num_resources = 4, + .resource = (struct resource[]) { + { + .start = MPC85xx_ENET2_OFFSET, + .end = MPC85xx_ENET2_OFFSET + + MPC85xx_ENET2_SIZE - 1, + .flags = IORESOURCE_MEM, + }, + { + .name = "tx", + .start = MPC85xx_IRQ_TSEC2_TX, + .end = MPC85xx_IRQ_TSEC2_TX, + .flags = IORESOURCE_IRQ, + }, + { + .name = "rx", + .start = MPC85xx_IRQ_TSEC2_RX, + .end = MPC85xx_IRQ_TSEC2_RX, + .flags = IORESOURCE_IRQ, + }, + { + .name = "error", + .start = MPC85xx_IRQ_TSEC2_ERROR, + .end = MPC85xx_IRQ_TSEC2_ERROR, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC85xx_eTSEC3] = { + .name = "fsl-gianfar", + .id = 3, + .dev.platform_data = &mpc85xx_etsec3_pdata, + .num_resources = 4, + .resource = (struct resource[]) { + { + .start = MPC85xx_ENET3_OFFSET, + .end = MPC85xx_ENET3_OFFSET + + MPC85xx_ENET3_SIZE - 1, + .flags = IORESOURCE_MEM, + }, + { + .name = "tx", + .start = MPC85xx_IRQ_TSEC3_TX, + .end = MPC85xx_IRQ_TSEC3_TX, + .flags = IORESOURCE_IRQ, + }, + { + .name = "rx", + .start = MPC85xx_IRQ_TSEC3_RX, + .end = MPC85xx_IRQ_TSEC3_RX, + .flags = IORESOURCE_IRQ, + }, + { + .name = "error", + .start = MPC85xx_IRQ_TSEC3_ERROR, + .end = MPC85xx_IRQ_TSEC3_ERROR, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC85xx_eTSEC4] = { + .name = "fsl-gianfar", + .id = 4, + .dev.platform_data = &mpc85xx_etsec4_pdata, + .num_resources = 4, + .resource = (struct resource[]) { + { + .start = 0x27000, + .end = 0x27fff, + .flags = IORESOURCE_MEM, + }, + { + .name = "tx", + .start = MPC85xx_IRQ_TSEC4_TX, + .end = MPC85xx_IRQ_TSEC4_TX, + .flags = IORESOURCE_IRQ, + }, + { + .name = "rx", + .start = MPC85xx_IRQ_TSEC4_RX, + .end = MPC85xx_IRQ_TSEC4_RX, + .flags = IORESOURCE_IRQ, + }, + { + .name = "error", + .start = MPC85xx_IRQ_TSEC4_ERROR, + .end = MPC85xx_IRQ_TSEC4_ERROR, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC85xx_IIC2] = { + .name = "fsl-i2c", + .id = 2, + .dev.platform_data = &mpc85xx_fsl_i2c2_pdata, + .num_resources = 2, + .resource = (struct resource[]) { + { + .start = 0x03100, + .end = 0x031ff, + .flags = IORESOURCE_MEM, + }, + { + .start = MPC85xx_IRQ_IIC1, + .end = MPC85xx_IRQ_IIC1, + .flags = IORESOURCE_IRQ, + }, + }, + }, }; static int __init mach_mpc85xx_fixup(struct platform_device *pdev) diff --git a/arch/ppc/syslib/mpc85xx_sys.c b/arch/ppc/syslib/mpc85xx_sys.c index d806a92a9401..6e3184ab354f 100644 --- a/arch/ppc/syslib/mpc85xx_sys.c +++ b/arch/ppc/syslib/mpc85xx_sys.c @@ -110,6 +110,111 @@ struct ppc_sys_spec ppc_sys_specs[] = { MPC85xx_CPM_USB, }, }, + /* SVRs on 8548 rev1.0 matches for 8548/8547/8545 */ + { + .ppc_sys_name = "8548E", + .mask = 0xFFFF00F0, + .value = 0x80390010, + .num_devices = 13, + .device_list = (enum ppc_sys_devices[]) + { + MPC85xx_eTSEC1, MPC85xx_eTSEC2, MPC85xx_eTSEC3, + MPC85xx_eTSEC4, MPC85xx_IIC1, MPC85xx_IIC2, + MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3, + MPC85xx_PERFMON, MPC85xx_DUART, MPC85xx_SEC2, + }, + }, + { + .ppc_sys_name = "8548", + .mask = 0xFFFF00F0, + .value = 0x80310010, + .num_devices = 12, + .device_list = (enum ppc_sys_devices[]) + { + MPC85xx_eTSEC1, MPC85xx_eTSEC2, MPC85xx_eTSEC3, + MPC85xx_eTSEC4, MPC85xx_IIC1, MPC85xx_IIC2, + MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3, + MPC85xx_PERFMON, MPC85xx_DUART, + }, + }, + { + .ppc_sys_name = "8547E", + .mask = 0xFFFF00F0, + .value = 0x80390010, + .num_devices = 13, + .device_list = (enum ppc_sys_devices[]) + { + MPC85xx_eTSEC1, MPC85xx_eTSEC2, MPC85xx_eTSEC3, + MPC85xx_eTSEC4, MPC85xx_IIC1, MPC85xx_IIC2, + MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3, + MPC85xx_PERFMON, MPC85xx_DUART, MPC85xx_SEC2, + }, + }, + { + .ppc_sys_name = "8547", + .mask = 0xFFFF00F0, + .value = 0x80310010, + .num_devices = 12, + .device_list = (enum ppc_sys_devices[]) + { + MPC85xx_eTSEC1, MPC85xx_eTSEC2, MPC85xx_eTSEC3, + MPC85xx_eTSEC4, MPC85xx_IIC1, MPC85xx_IIC2, + MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3, + MPC85xx_PERFMON, MPC85xx_DUART, + }, + }, + { + .ppc_sys_name = "8545E", + .mask = 0xFFFF00F0, + .value = 0x80390010, + .num_devices = 11, + .device_list = (enum ppc_sys_devices[]) + { + MPC85xx_eTSEC1, MPC85xx_eTSEC2, + MPC85xx_IIC1, MPC85xx_IIC2, + MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3, + MPC85xx_PERFMON, MPC85xx_DUART, MPC85xx_SEC2, + }, + }, + { + .ppc_sys_name = "8545", + .mask = 0xFFFF00F0, + .value = 0x80310010, + .num_devices = 10, + .device_list = (enum ppc_sys_devices[]) + { + MPC85xx_eTSEC1, MPC85xx_eTSEC2, + MPC85xx_IIC1, MPC85xx_IIC2, + MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3, + MPC85xx_PERFMON, MPC85xx_DUART, + }, + }, + { + .ppc_sys_name = "8543E", + .mask = 0xFFFF00F0, + .value = 0x803A0010, + .num_devices = 11, + .device_list = (enum ppc_sys_devices[]) + { + MPC85xx_eTSEC1, MPC85xx_eTSEC2, + MPC85xx_IIC1, MPC85xx_IIC2, + MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3, + MPC85xx_PERFMON, MPC85xx_DUART, MPC85xx_SEC2, + }, + }, + { + .ppc_sys_name = "8543", + .mask = 0xFFFF00F0, + .value = 0x80320010, + .num_devices = 10, + .device_list = (enum ppc_sys_devices[]) + { + MPC85xx_eTSEC1, MPC85xx_eTSEC2, + MPC85xx_IIC1, MPC85xx_IIC2, + MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3, + MPC85xx_PERFMON, MPC85xx_DUART, + }, + }, { /* default match */ .ppc_sys_name = "", .mask = 0x00000000, diff --git a/include/asm-ppc/irq.h b/include/asm-ppc/irq.h index 06b86be61ed1..434fc515ba64 100644 --- a/include/asm-ppc/irq.h +++ b/include/asm-ppc/irq.h @@ -223,9 +223,15 @@ static __inline__ int irq_canonicalize(int irq) #define MPC85xx_IRQ_RIO_RX (12 + MPC85xx_OPENPIC_IRQ_OFFSET) #define MPC85xx_IRQ_TSEC1_TX (13 + MPC85xx_OPENPIC_IRQ_OFFSET) #define MPC85xx_IRQ_TSEC1_RX (14 + MPC85xx_OPENPIC_IRQ_OFFSET) +#define MPC85xx_IRQ_TSEC3_TX (15 + MPC85xx_OPENPIC_IRQ_OFFSET) +#define MPC85xx_IRQ_TSEC3_RX (16 + MPC85xx_OPENPIC_IRQ_OFFSET) +#define MPC85xx_IRQ_TSEC3_ERROR (17 + MPC85xx_OPENPIC_IRQ_OFFSET) #define MPC85xx_IRQ_TSEC1_ERROR (18 + MPC85xx_OPENPIC_IRQ_OFFSET) #define MPC85xx_IRQ_TSEC2_TX (19 + MPC85xx_OPENPIC_IRQ_OFFSET) #define MPC85xx_IRQ_TSEC2_RX (20 + MPC85xx_OPENPIC_IRQ_OFFSET) +#define MPC85xx_IRQ_TSEC4_TX (21 + MPC85xx_OPENPIC_IRQ_OFFSET) +#define MPC85xx_IRQ_TSEC4_RX (22 + MPC85xx_OPENPIC_IRQ_OFFSET) +#define MPC85xx_IRQ_TSEC4_ERROR (23 + MPC85xx_OPENPIC_IRQ_OFFSET) #define MPC85xx_IRQ_TSEC2_ERROR (24 + MPC85xx_OPENPIC_IRQ_OFFSET) #define MPC85xx_IRQ_FEC (25 + MPC85xx_OPENPIC_IRQ_OFFSET) #define MPC85xx_IRQ_DUART (26 + MPC85xx_OPENPIC_IRQ_OFFSET) diff --git a/include/asm-ppc/mpc85xx.h b/include/asm-ppc/mpc85xx.h index 22713e331585..51b61d3305f7 100644 --- a/include/asm-ppc/mpc85xx.h +++ b/include/asm-ppc/mpc85xx.h @@ -74,7 +74,7 @@ extern unsigned char __res[]; #define MPC85xx_GUTS_OFFSET (0xe0000) #define MPC85xx_GUTS_SIZE (0x01000) #define MPC85xx_IIC1_OFFSET (0x03000) -#define MPC85xx_IIC1_SIZE (0x01000) +#define MPC85xx_IIC1_SIZE (0x00100) #define MPC85xx_OPENPIC_OFFSET (0x40000) #define MPC85xx_OPENPIC_SIZE (0x40000) #define MPC85xx_PCI1_OFFSET (0x08000) @@ -127,6 +127,11 @@ enum ppc_sys_devices { MPC85xx_CPM_MCC2, MPC85xx_CPM_SMC1, MPC85xx_CPM_SMC2, + MPC85xx_eTSEC1, + MPC85xx_eTSEC2, + MPC85xx_eTSEC3, + MPC85xx_eTSEC4, + MPC85xx_IIC2, }; #endif /* CONFIG_85xx */ diff --git a/include/linux/fsl_devices.h b/include/linux/fsl_devices.h index faaff4c64559..70f54af87b9f 100644 --- a/include/linux/fsl_devices.h +++ b/include/linux/fsl_devices.h @@ -51,6 +51,7 @@ struct gianfar_platform_data { /* board specific information */ u32 board_flags; + u32 phy_flags; u32 phyid; u32 interruptPHY; u8 mac_addr[6]; @@ -61,9 +62,14 @@ struct gianfar_platform_data { #define FSL_GIANFAR_DEV_HAS_COALESCE 0x00000002 #define FSL_GIANFAR_DEV_HAS_RMON 0x00000004 #define FSL_GIANFAR_DEV_HAS_MULTI_INTR 0x00000008 +#define FSL_GIANFAR_DEV_HAS_CSUM 0x00000010 +#define FSL_GIANFAR_DEV_HAS_VLAN 0x00000020 +#define FSL_GIANFAR_DEV_HAS_EXTENDED_HASH 0x00000040 +#define FSL_GIANFAR_DEV_HAS_PADDING 0x00000080 /* Flags in gianfar_platform_data */ -#define FSL_GIANFAR_BRD_HAS_PHY_INTR 0x00000001 /* if not set use a timer */ +#define FSL_GIANFAR_BRD_HAS_PHY_INTR 0x00000001 /* set or use a timer */ +#define FSL_GIANFAR_BRD_IS_REDUCED 0x00000002 /* Set if RGMII, RMII */ struct fsl_i2c_platform_data { /* device specific information */ -- cgit v1.2.3 From 22329b511a97557b293583194037d1f4c71e1504 Mon Sep 17 00:00:00 2001 From: Brent Casavant Date: Tue, 21 Jun 2005 17:15:59 -0700 Subject: [PATCH] ioc4: Core driver rewrite This series of patches reworks the configuration and internal structure of the SGI IOC4 I/O controller device drivers. These changes are motivated by several factors: - The IOC4 chip PCI resources are of mixed use between functions (i.e. multiple functions are handled in the same address range, sometimes within the same register), muddling resource ownership and initialization issues. Centralizing this ownership in a core driver is desirable. - The IOC4 chip implements multiple functions (serial, IDE, others not yet implemented in the mainline kernel) but is not a multifunction PCI device. In order to properly handle device addition and removal as well as module insertion and deletion, an intermediary IOC4-specific driver layer is needed to handle these operations cleanly. - All IOC4 drivers are currently enabled by a single CONFIG value. As not all systems need all IOC4 functions, it is desireable to enable these drivers independently. - The current IOC4 core driver will trigger loading of all function-level drivers, as it makes direct calls to them. This situation should be reversed (i.e. function-level drivers cause loading of core driver) in order to maintain a clear and least-surprise driver loading model. - IOC4 hardware design necessitates some driver-level dependency on the PCI bus clock speed. Current code assumes a 66MHz bus, but the speed should be autodetected and appropriate compensation taken. This patch series effects the above changes by a newly and better designed IOC4 core driver with which the function-level drivers can register and deregister themselves upon module insertion/removal. By tracking these modules, device addition/removal is also handled properly. PCI resource management and ownership issues are centralized in this core driver, and IOC4-wide configuration actions such as bus speed detection are also handled in this core driver. This patch: The SGI IOC4 I/O controller chip implements multiple functions, though it is not a multi-function PCI device. Additionally, various PCI resources of the IOC4 are shared by multiple hardware functions, and thus resource ownership by driver is not clearly delineated. Due to the current driver design, all core and subordinate drivers must be loaded, or none, which is undesirable if not all IOC4 hardware features are being used. This patch reorganizes the IOC4 drivers so that the core driver provides a subdriver registration service. Through appropriate callbacks the subdrivers can now handle device addition and removal, as well as module insertion and deletion (though the IOC4 IDE driver requires further work before module deletion will work). The core driver now takes care of allocating PCI resources and data which must be shared between subdrivers, to clearly delineate module ownership of these items. Signed-off-by: Brent Casavant Acked-by: Pat Gefre Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/sgi-ioc4.txt | 45 +++++++ drivers/ide/pci/sgiioc4.c | 30 ++++- drivers/serial/ioc4_serial.c | 311 ++++++++++++++++++------------------------- drivers/sn/ioc4.c | 290 ++++++++++++++++++++++++++++++++++++---- include/linux/ioc4.h | 156 ++++++++++++++++++++++ include/linux/ioc4_common.h | 21 --- 6 files changed, 617 insertions(+), 236 deletions(-) create mode 100644 Documentation/sgi-ioc4.txt create mode 100644 include/linux/ioc4.h delete mode 100644 include/linux/ioc4_common.h (limited to 'include/linux') diff --git a/Documentation/sgi-ioc4.txt b/Documentation/sgi-ioc4.txt new file mode 100644 index 000000000000..876c96ae38db --- /dev/null +++ b/Documentation/sgi-ioc4.txt @@ -0,0 +1,45 @@ +The SGI IOC4 PCI device is a bit of a strange beast, so some notes on +it are in order. + +First, even though the IOC4 performs multiple functions, such as an +IDE controller, a serial controller, a PS/2 keyboard/mouse controller, +and an external interrupt mechanism, it's not implemented as a +multifunction device. The consequence of this from a software +standpoint is that all these functions share a single IRQ, and +they can't all register to own the same PCI device ID. To make +matters a bit worse, some of the register blocks (and even registers +themselves) present in IOC4 are mixed-purpose between these several +functions, meaning that there's no clear "owning" device driver. + +The solution is to organize the IOC4 driver into several independent +drivers, "ioc4", "sgiioc4", and "ioc4_serial". Note that there is no +PS/2 controller driver as this functionality has never been wired up +on a shipping IO card. + +ioc4 +==== +This is the core (or shim) driver for IOC4. It is responsible for +initializing the basic functionality of the chip, and allocating +the PCI resources that are shared between the IOC4 functions. + +This driver also provides registration functions that the other +IOC4 drivers can call to make their presence known. Each driver +needs to provide a probe and remove function, which are invoked +by the core driver at appropriate times. The interface of these +IOC4 function probe and remove operations isn't precisely the same +as PCI device probe and remove operations, but is logically the +same operation. + +sgiioc4 +======= +This is the IDE driver for IOC4. Its name isn't very descriptive +simply for historical reasons (it used to be the only IOC4 driver +component). There's not much to say about it other than it hooks +up to the ioc4 driver via the appropriate registration, probe, and +remove functions. + +ioc4_serial +=========== +This is the serial driver for IOC4. There's not much to say about it +other than it hooks up to the ioc4 driver via the appropriate registration, +probe, and remove functions. diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c index 4651a22bf12e..af526b671c4e 100644 --- a/drivers/ide/pci/sgiioc4.c +++ b/drivers/ide/pci/sgiioc4.c @@ -34,7 +34,7 @@ #include #include #include -#include +#include #include #include @@ -715,14 +715,34 @@ static ide_pci_device_t sgiioc4_chipsets[] __devinitdata = { }; int -ioc4_ide_attach_one(struct pci_dev *dev, const struct pci_device_id *id) +ioc4_ide_attach_one(struct ioc4_driver_data *idd) { - return pci_init_sgiioc4(dev, &sgiioc4_chipsets[id->driver_data]); + return pci_init_sgiioc4(idd->idd_pdev, + &sgiioc4_chipsets[idd->idd_pci_id->driver_data]); } +static struct ioc4_submodule ioc4_ide_submodule = { + .is_name = "IOC4_ide", + .is_owner = THIS_MODULE, + .is_probe = ioc4_ide_attach_one, +/* .is_remove = ioc4_ide_remove_one, */ +}; + +static int __devinit +ioc4_ide_init(void) +{ + return ioc4_register_submodule(&ioc4_ide_submodule); +} + +static void __devexit +ioc4_ide_exit(void) +{ + ioc4_unregister_submodule(&ioc4_ide_submodule); +} + +module_init(ioc4_ide_init); +module_exit(ioc4_ide_exit); MODULE_AUTHOR("Aniket Malatpure - Silicon Graphics Inc. (SGI)"); MODULE_DESCRIPTION("IDE PCI driver module for SGI IOC4 Base-IO Card"); MODULE_LICENSE("GPL"); - -EXPORT_SYMBOL(ioc4_ide_attach_one); diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index ba4e13a22a50..da5f10eb4845 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -20,7 +20,7 @@ #include #include #include -#include +#include #include /* @@ -130,12 +130,19 @@ IOC4_SIO_IR_S3_TX_EXPLICIT) /* Bitmasks for IOC4_OTHER_IR, IOC4_OTHER_IEC, and IOC4_OTHER_IES */ -#define IOC4_OTHER_IR_ATA_INT 0x00000001 /* ATAPI intr pass-thru */ -#define IOC4_OTHER_IR_ATA_MEMERR 0x00000002 /* ATAPI DMA PCI error */ -#define IOC4_OTHER_IR_S0_MEMERR 0x00000004 /* Port 0 PCI error */ -#define IOC4_OTHER_IR_S1_MEMERR 0x00000008 /* Port 1 PCI error */ -#define IOC4_OTHER_IR_S2_MEMERR 0x00000010 /* Port 2 PCI error */ -#define IOC4_OTHER_IR_S3_MEMERR 0x00000020 /* Port 3 PCI error */ +#define IOC4_OTHER_IR_ATA_INT 0x00000001 /* ATAPI intr pass-thru */ +#define IOC4_OTHER_IR_ATA_MEMERR 0x00000002 /* ATAPI DMA PCI error */ +#define IOC4_OTHER_IR_S0_MEMERR 0x00000004 /* Port 0 PCI error */ +#define IOC4_OTHER_IR_S1_MEMERR 0x00000008 /* Port 1 PCI error */ +#define IOC4_OTHER_IR_S2_MEMERR 0x00000010 /* Port 2 PCI error */ +#define IOC4_OTHER_IR_S3_MEMERR 0x00000020 /* Port 3 PCI error */ +#define IOC4_OTHER_IR_KBD_INT 0x00000040 /* Keyboard/mouse */ +#define IOC4_OTHER_IR_RESERVED 0x007fff80 /* Reserved */ +#define IOC4_OTHER_IR_RT_INT 0x00800000 /* INT_OUT section output */ +#define IOC4_OTHER_IR_GEN_INT 0xff000000 /* Generic pins */ + +#define IOC4_OTHER_IR_SER_MEMERR (IOC4_OTHER_IR_S0_MEMERR | IOC4_OTHER_IR_S1_MEMERR | \ + IOC4_OTHER_IR_S2_MEMERR | IOC4_OTHER_IR_S3_MEMERR) /* Bitmasks for IOC4_SIO_CR */ #define IOC4_SIO_CR_CMD_PULSE_SHIFT 0 /* byte bus strobe shift */ @@ -274,67 +281,22 @@ struct ioc4_uartregs { #define i4u_dlm u2.dlm #define i4u_fcr u3.fcr -/* PCI memory space register map addressed using pci_bar0 */ -struct ioc4_memregs { - struct ioc4_mem { - /* Miscellaneous IOC4 registers */ - uint32_t pci_err_addr_l; - uint32_t pci_err_addr_h; - uint32_t sio_ir; - uint32_t other_ir; - - /* These registers are read-only for general kernel code. */ - uint32_t sio_ies_ro; - uint32_t other_ies_ro; - uint32_t sio_iec_ro; - uint32_t other_iec_ro; - uint32_t sio_cr; - uint32_t misc_fill1; - uint32_t int_out; - uint32_t misc_fill2; - uint32_t gpcr_s; - uint32_t gpcr_c; - uint32_t gpdr; - uint32_t misc_fill3; - uint32_t gppr_0; - uint32_t gppr_1; - uint32_t gppr_2; - uint32_t gppr_3; - uint32_t gppr_4; - uint32_t gppr_5; - uint32_t gppr_6; - uint32_t gppr_7; - } ioc4_mem; - - char misc_fill4[0x100 - 0x5C - 4]; - - /* ATA/ATAP registers */ - uint32_t ata_notused[9]; - char ata_fill1[0x140 - 0x120 - 4]; - uint32_t ata_notused1[8]; - char ata_fill2[0x200 - 0x15C - 4]; - - /* Keyboard and mouse registers */ - uint32_t km_notused[5];; - char km_fill1[0x300 - 0x210 - 4]; - - /* Serial port registers used for DMA serial I/O */ - struct ioc4_serial { - uint32_t sbbr01_l; - uint32_t sbbr01_h; - uint32_t sbbr23_l; - uint32_t sbbr23_h; - - struct ioc4_serialregs port_0; - struct ioc4_serialregs port_1; - struct ioc4_serialregs port_2; - struct ioc4_serialregs port_3; - struct ioc4_uartregs uart_0; - struct ioc4_uartregs uart_1; - struct ioc4_uartregs uart_2; - struct ioc4_uartregs uart_3; - } ioc4_serial; -}; +/* Serial port registers used for DMA serial I/O */ +struct ioc4_serial { + uint32_t sbbr01_l; + uint32_t sbbr01_h; + uint32_t sbbr23_l; + uint32_t sbbr23_h; + + struct ioc4_serialregs port_0; + struct ioc4_serialregs port_1; + struct ioc4_serialregs port_2; + struct ioc4_serialregs port_3; + struct ioc4_uartregs uart_0; + struct ioc4_uartregs uart_1; + struct ioc4_uartregs uart_2; + struct ioc4_uartregs uart_3; +} ioc4_serial; /* UART clock speed */ #define IOC4_SER_XIN_CLK IOC4_SER_XIN_CLK_66 @@ -412,8 +374,8 @@ enum sio_proto { | UART_LCR_WLEN7 | UART_LCR_WLEN8) #define LCR_MASK_STOP_BITS (UART_LCR_STOP) -#define PENDING(_p) (readl(&(_p)->ip_mem->sio_ir) & _p->ip_ienb) -#define READ_SIO_IR(_p) readl(&(_p)->ip_mem->sio_ir) +#define PENDING(_p) (readl(&(_p)->ip_mem->sio_ir.raw) & _p->ip_ienb) +#define READ_SIO_IR(_p) readl(&(_p)->ip_mem->sio_ir.raw) /* Default to 4k buffers */ #ifdef IOC4_1K_BUFFERS @@ -447,7 +409,7 @@ struct ioc4_control { */ #define MAX_IOC4_INTR_ENTS (8 * sizeof(uint32_t)) struct ioc4_soft { - struct ioc4_mem __iomem *is_ioc4_mem_addr; + struct ioc4_misc_regs __iomem *is_ioc4_misc_addr; struct ioc4_serial __iomem *is_ioc4_serial_addr; /* Each interrupt type has an entry in the array */ @@ -486,7 +448,7 @@ struct ioc4_port { struct ioc4_soft *ip_ioc4_soft; /* pci mem addresses */ - struct ioc4_mem __iomem *ip_mem; + struct ioc4_misc_regs __iomem *ip_mem; struct ioc4_serial __iomem *ip_serial; struct ioc4_serialregs __iomem *ip_serial_regs; struct ioc4_uartregs __iomem *ip_uart_regs; @@ -553,7 +515,7 @@ struct hooks { uint32_t intr_dma_error; uint32_t intr_clear; uint32_t intr_all; - char rs422_select_pin; + int rs422_select_pin; }; static struct hooks hooks_array[IOC4_NUM_SERIAL_PORTS] = { @@ -669,7 +631,7 @@ static void handle_intr(void *arg, uint32_t sio_ir); static inline void write_ireg(struct ioc4_soft *ioc4_soft, uint32_t val, int which, int type) { - struct ioc4_mem __iomem *mem = ioc4_soft->is_ioc4_mem_addr; + struct ioc4_misc_regs __iomem *mem = ioc4_soft->is_ioc4_misc_addr; unsigned long flags; spin_lock_irqsave(&ioc4_soft->is_ir_lock, flags); @@ -678,11 +640,11 @@ write_ireg(struct ioc4_soft *ioc4_soft, uint32_t val, int which, int type) case IOC4_SIO_INTR_TYPE: switch (which) { case IOC4_W_IES: - writel(val, &mem->sio_ies_ro); + writel(val, &mem->sio_ies.raw); break; case IOC4_W_IEC: - writel(val, &mem->sio_iec_ro); + writel(val, &mem->sio_iec.raw); break; } break; @@ -690,11 +652,11 @@ write_ireg(struct ioc4_soft *ioc4_soft, uint32_t val, int which, int type) case IOC4_OTHER_INTR_TYPE: switch (which) { case IOC4_W_IES: - writel(val, &mem->other_ies_ro); + writel(val, &mem->other_ies.raw); break; case IOC4_W_IEC: - writel(val, &mem->other_iec_ro); + writel(val, &mem->other_iec.raw); break; } break; @@ -747,7 +709,8 @@ static int set_baud(struct ioc4_port *port, int baud) */ static struct ioc4_port *get_ioc4_port(struct uart_port *the_port) { - struct ioc4_control *control = dev_get_drvdata(the_port->dev); + struct ioc4_driver_data *idd = dev_get_drvdata(the_port->dev); + struct ioc4_control *control = idd->idd_serial_data; int ii; if (control) { @@ -782,7 +745,7 @@ static struct ioc4_port *get_ioc4_port(struct uart_port *the_port) static inline uint32_t pending_intrs(struct ioc4_soft *soft, int type) { - struct ioc4_mem __iomem *mem = soft->is_ioc4_mem_addr; + struct ioc4_misc_regs __iomem *mem = soft->is_ioc4_misc_addr; unsigned long flag; uint32_t intrs = 0; @@ -793,11 +756,11 @@ pending_intrs(struct ioc4_soft *soft, int type) switch (type) { case IOC4_SIO_INTR_TYPE: - intrs = readl(&mem->sio_ir) & readl(&mem->sio_ies_ro); + intrs = readl(&mem->sio_ir.raw) & readl(&mem->sio_ies.raw); break; case IOC4_OTHER_INTR_TYPE: - intrs = readl(&mem->other_ir) & readl(&mem->other_ies_ro); + intrs = readl(&mem->other_ir.raw) & readl(&mem->other_ies.raw); /* Don't process any ATA interrupte */ intrs &= ~(IOC4_OTHER_IR_ATA_INT | IOC4_OTHER_IR_ATA_MEMERR); @@ -826,7 +789,7 @@ static int inline port_init(struct ioc4_port *port) /* Wait until any pending bus activity for this port has ceased */ do - sio_cr = readl(&port->ip_mem->sio_cr); + sio_cr = readl(&port->ip_mem->sio_cr.raw); while (!(sio_cr & IOC4_SIO_CR_SIO_DIAG_IDLE)); /* Finish reset sequence */ @@ -899,7 +862,7 @@ static int inline port_init(struct ioc4_port *port) write_ireg(port->ip_ioc4_soft, hooks->intr_clear, IOC4_W_IEC, IOC4_SIO_INTR_TYPE); port->ip_ienb &= ~hooks->intr_clear; - writel(hooks->intr_clear, &port->ip_mem->sio_ir); + writel(hooks->intr_clear, &port->ip_mem->sio_ir.raw); return 0; } @@ -918,23 +881,23 @@ static void handle_dma_error_intr(void *arg, uint32_t other_ir) spin_lock_irqsave(&port->ip_lock, flags); /* ACK the interrupt */ - writel(hooks->intr_dma_error, &port->ip_mem->other_ir); + writel(hooks->intr_dma_error, &port->ip_mem->other_ir.raw); - if (readl(&port->ip_mem->pci_err_addr_l) & IOC4_PCI_ERR_ADDR_VLD) { + if (readl(&port->ip_mem->pci_err_addr_l.raw) & IOC4_PCI_ERR_ADDR_VLD) { printk(KERN_ERR "PCI error address is 0x%lx, " "master is serial port %c %s\n", (((uint64_t)readl(&port->ip_mem->pci_err_addr_h) << 32) - | readl(&port->ip_mem->pci_err_addr_l)) + | readl(&port->ip_mem->pci_err_addr_l.raw)) & IOC4_PCI_ERR_ADDR_ADDR_MSK, '1' + - ((char)(readl(&port->ip_mem-> pci_err_addr_l) & + ((char)(readl(&port->ip_mem->pci_err_addr_l.raw) & IOC4_PCI_ERR_ADDR_MST_NUM_MSK) >> 1), - (readl(&port->ip_mem->pci_err_addr_l) + (readl(&port->ip_mem->pci_err_addr_l.raw) & IOC4_PCI_ERR_ADDR_MST_TYP_MSK) ? "RX" : "TX"); - if (readl(&port->ip_mem->pci_err_addr_l) + if (readl(&port->ip_mem->pci_err_addr_l.raw) & IOC4_PCI_ERR_ADDR_MUL_ERR) { printk(KERN_ERR "Multiple errors occurred\n"); @@ -1018,26 +981,26 @@ static irqreturn_t ioc4_intr(int irq, void *arg, struct pt_regs *regs) "other_ies = 0x%x\n", (intr_type == IOC4_SIO_INTR_TYPE) ? "sio" : "other", this_ir, - readl(&soft->is_ioc4_mem_addr->sio_ir), - readl(&soft->is_ioc4_mem_addr->sio_ies_ro), - readl(&soft->is_ioc4_mem_addr->other_ir), - readl(&soft->is_ioc4_mem_addr->other_ies_ro)); + readl(&soft->is_ioc4_misc_addr->sio_ir.raw), + readl(&soft->is_ioc4_misc_addr->sio_ies.raw), + readl(&soft->is_ioc4_misc_addr->other_ir.raw), + readl(&soft->is_ioc4_misc_addr->other_ies.raw)); } } #ifdef DEBUG_INTERRUPTS { - struct ioc4_mem __iomem *mem = soft->is_ioc4_mem_addr; + struct ioc4_misc_regs __iomem *mem = soft->is_ioc4_misc_addr; spinlock_t *lp = &soft->is_ir_lock; unsigned long flag; spin_lock_irqsave(&soft->is_ir_lock, flag); - printk ("%s : %d : mem 0x%p sio_ir 0x%x sio_ies_ro 0x%x " - "other_ir 0x%x other_ies_ro 0x%x mask 0x%x\n", + printk ("%s : %d : mem 0x%p sio_ir 0x%x sio_ies 0x%x " + "other_ir 0x%x other_ies 0x%x mask 0x%x\n", __FUNCTION__, __LINE__, - (void *)mem, readl(&mem->sio_ir), - readl(&mem->sio_ies_ro), - readl(&mem->other_ir), - readl(&mem->other_ies_ro), + (void *)mem, readl(&mem->sio_ir.raw), + readl(&mem->sio_ies.raw), + readl(&mem->other_ir.raw), + readl(&mem->other_ies.raw), IOC4_OTHER_IR_ATA_INT | IOC4_OTHER_IR_ATA_MEMERR); spin_unlock_irqrestore(&soft->is_ir_lock, flag); } @@ -1056,7 +1019,7 @@ static irqreturn_t ioc4_intr(int irq, void *arg, struct pt_regs *regs) */ static int inline ioc4_attach_local(struct pci_dev *pdev, struct ioc4_control *control, - struct ioc4_soft *soft, void __iomem *ioc4_mem, + struct ioc4_soft *soft, void __iomem *ioc4_misc, void __iomem *ioc4_serial) { struct ioc4_port *port; @@ -1076,7 +1039,7 @@ static int inline ioc4_attach_local(struct pci_dev *pdev, ioc4_revid, ioc4_revid_min); return -EPERM; } - BUG_ON(ioc4_mem == NULL); + BUG_ON(ioc4_misc == NULL); BUG_ON(ioc4_serial == NULL); /* Create port structures for each port */ @@ -1103,7 +1066,7 @@ static int inline ioc4_attach_local(struct pci_dev *pdev, port->ip_pci_bus_speed = IOC4_SER_XIN_CLK; port->ip_baud = 9600; port->ip_control = control; - port->ip_mem = ioc4_mem; + port->ip_mem = ioc4_misc; port->ip_serial = ioc4_serial; /* point to the right hook */ @@ -1604,14 +1567,12 @@ static int ioc4_set_proto(struct ioc4_port *port, enum sio_proto proto) switch (proto) { case PROTO_RS232: /* Clear the appropriate GIO pin */ - writel(0, (&port->ip_mem->gppr_0 + - hooks->rs422_select_pin)); + writel(0, (&port->ip_mem->gppr[hooks->rs422_select_pin].raw)); break; case PROTO_RS422: /* Set the appropriate GIO pin */ - writel(1, (&port->ip_mem->gppr_0 + - hooks->rs422_select_pin)); + writel(1, (&port->ip_mem->gppr[hooks->rs422_select_pin].raw)); break; default: @@ -1885,7 +1846,7 @@ static void handle_intr(void *arg, uint32_t sio_ir) if (sio_ir & hooks->intr_delta_dcd) { /* ACK the interrupt */ writel(hooks->intr_delta_dcd, - &port->ip_mem->sio_ir); + &port->ip_mem->sio_ir.raw); shadow = readl(&port->ip_serial_regs->shadow); @@ -1907,7 +1868,7 @@ static void handle_intr(void *arg, uint32_t sio_ir) if (sio_ir & hooks->intr_delta_cts) { /* ACK the interrupt */ writel(hooks->intr_delta_cts, - &port->ip_mem->sio_ir); + &port->ip_mem->sio_ir.raw); shadow = readl(&port->ip_serial_regs->shadow); @@ -1928,7 +1889,7 @@ static void handle_intr(void *arg, uint32_t sio_ir) if (sio_ir & hooks->intr_rx_timer) { /* ACK the interrupt */ writel(hooks->intr_rx_timer, - &port->ip_mem->sio_ir); + &port->ip_mem->sio_ir.raw); if ((port->ip_notify & N_DATA_READY) && (port->ip_port)) { @@ -1974,7 +1935,7 @@ static void handle_intr(void *arg, uint32_t sio_ir) /* ACK the interrupt */ writel(hooks->intr_tx_explicit, - &port->ip_mem->sio_ir); + &port->ip_mem->sio_ir.raw); if (port->ip_notify & N_OUTPUT_LOWAT) ioc4_cb_output_lowat(port); @@ -2634,7 +2595,8 @@ ioc4_serial_core_attach(struct pci_dev *pdev) { struct ioc4_port *port; struct uart_port *the_port; - struct ioc4_control *control = pci_get_drvdata(pdev); + struct ioc4_driver_data *idd = pci_get_drvdata(pdev); + struct ioc4_control *control = idd->idd_serial_data; int ii; DPRINT_CONFIG(("%s: attach pdev 0x%p - control 0x%p\n", @@ -2680,55 +2642,29 @@ ioc4_serial_core_attach(struct pci_dev *pdev) /** * ioc4_serial_attach_one - register attach function - * called per card found from ioc4_serial_detect as part - * of module_init(). - * @pdev: handle for this card - * @pci_id: pci id for this card + * called per card found from IOC4 master module. + * @idd: Master module data for this IOC4 */ int -ioc4_serial_attach_one(struct pci_dev *pdev, const struct pci_device_id *pci_id) +ioc4_serial_attach_one(struct ioc4_driver_data *idd) { - struct ioc4_mem __iomem *mem; - unsigned long tmp_addr, tmp_addr1; + unsigned long tmp_addr1; struct ioc4_serial __iomem *serial; struct ioc4_soft *soft; struct ioc4_control *control; - int tmp, ret = 0; + int ret = 0; - DPRINT_CONFIG(("%s (0x%p, 0x%p)\n", __FUNCTION__, pdev, pci_id)); - - /* Map in the ioc4 memory */ - tmp_addr = pci_resource_start(pdev, 0); - if (!tmp_addr) { - printk(KERN_WARNING - "ioc4 (%p) : unable to get PIO mapping for " - "MEM space\n", (void *)pdev); - return -ENODEV; - } - if (!request_region(tmp_addr, sizeof(struct ioc4_mem), "sioc4_mem")) { - printk(KERN_ALERT - "ioc4 (%p): unable to get request region for " - "MEM space\n", (void *)pdev); - return -ENODEV; - } - mem = ioremap(tmp_addr, sizeof(struct ioc4_mem)); - if (!mem) { - printk(KERN_WARNING - "ioc4 (%p) : unable to remap ioc4 memory\n", - (void *)pdev); - ret = -ENODEV; - goto out1; - } + DPRINT_CONFIG(("%s (0x%p, 0x%p)\n", __FUNCTION__, idd->idd_pdev, idd->idd_pci_id)); /* request serial registers */ - tmp_addr1 = pci_resource_start(pdev, 0) + IOC4_SERIAL_OFFSET; + tmp_addr1 = idd->idd_bar0 + IOC4_SERIAL_OFFSET; if (!request_region(tmp_addr1, sizeof(struct ioc4_serial), "sioc4_uart")) { printk(KERN_WARNING "ioc4 (%p): unable to get request region for " - "uart space\n", (void *)pdev); + "uart space\n", (void *)idd->idd_pdev); ret = -ENODEV; goto out1; } @@ -2736,12 +2672,12 @@ ioc4_serial_attach_one(struct pci_dev *pdev, const struct pci_device_id *pci_id) if (!serial) { printk(KERN_WARNING "ioc4 (%p) : unable to remap ioc4 serial register\n", - (void *)pdev); + (void *)idd->idd_pdev); ret = -ENODEV; goto out2; } DPRINT_CONFIG(("%s : mem 0x%p, serial 0x%p\n", - __FUNCTION__, (void *)mem, (void *)serial)); + __FUNCTION__, (void *)idd->idd_misc_regs, (void *)serial)); /* Get memory for the new card */ control = kmalloc(sizeof(struct ioc4_control) * IOC4_NUM_SERIAL_PORTS, @@ -2754,59 +2690,57 @@ ioc4_serial_attach_one(struct pci_dev *pdev, const struct pci_device_id *pci_id) goto out2; } memset(control, 0, sizeof(struct ioc4_control)); - pci_set_drvdata(pdev, control); + idd->idd_serial_data = control; /* Allocate the soft structure */ soft = kmalloc(sizeof(struct ioc4_soft), GFP_KERNEL); if (!soft) { printk(KERN_WARNING "ioc4 (%p): unable to get memory for the soft struct\n", - (void *)pdev); + (void *)idd->idd_pdev); ret = -ENOMEM; goto out3; } memset(soft, 0, sizeof(struct ioc4_soft)); spin_lock_init(&soft->is_ir_lock); - soft->is_ioc4_mem_addr = mem; + soft->is_ioc4_misc_addr = idd->idd_misc_regs; soft->is_ioc4_serial_addr = serial; /* Init the IOC4 */ - pci_read_config_dword(pdev, PCI_COMMAND, &tmp); - pci_write_config_dword(pdev, PCI_COMMAND, - tmp | PCI_COMMAND_PARITY | PCI_COMMAND_SERR); - - writel(0xf << IOC4_SIO_CR_CMD_PULSE_SHIFT, &mem->sio_cr); + writel(0xf << IOC4_SIO_CR_CMD_PULSE_SHIFT, + &idd->idd_misc_regs->sio_cr.raw); /* Enable serial port mode select generic PIO pins as outputs */ writel(IOC4_GPCR_UART0_MODESEL | IOC4_GPCR_UART1_MODESEL | IOC4_GPCR_UART2_MODESEL | IOC4_GPCR_UART3_MODESEL, - &mem->gpcr_s); + &idd->idd_misc_regs->gpcr_s.raw); - /* Clear and disable all interrupts */ + /* Clear and disable all serial interrupts */ write_ireg(soft, ~0, IOC4_W_IEC, IOC4_SIO_INTR_TYPE); - writel(~0, &mem->sio_ir); - write_ireg(soft, ~(IOC4_OTHER_IR_ATA_INT | IOC4_OTHER_IR_ATA_MEMERR), - IOC4_W_IEC, IOC4_OTHER_INTR_TYPE); - writel(~(IOC4_OTHER_IR_ATA_MEMERR | IOC4_OTHER_IR_ATA_MEMERR), - &mem->other_ir); + writel(~0, &idd->idd_misc_regs->sio_ir.raw); + write_ireg(soft, IOC4_OTHER_IR_SER_MEMERR, IOC4_W_IEC, + IOC4_OTHER_INTR_TYPE); + writel(IOC4_OTHER_IR_SER_MEMERR, &idd->idd_misc_regs->other_ir.raw); control->ic_soft = soft; - if (!request_irq(pdev->irq, ioc4_intr, SA_SHIRQ, + + /* Hook up interrupt handler */ + if (!request_irq(idd->idd_pdev->irq, ioc4_intr, SA_SHIRQ, "sgi-ioc4serial", (void *)soft)) { - control->ic_irq = pdev->irq; + control->ic_irq = idd->idd_pdev->irq; } else { printk(KERN_WARNING "%s : request_irq fails for IRQ 0x%x\n ", - __FUNCTION__, pdev->irq); + __FUNCTION__, idd->idd_pdev->irq); } - if ((ret = ioc4_attach_local(pdev, control, soft, - soft->is_ioc4_mem_addr, + if ((ret = ioc4_attach_local(idd->idd_pdev, control, soft, + soft->is_ioc4_misc_addr, soft->is_ioc4_serial_addr))) goto out4; /* register port with the serial core */ - if ((ret = ioc4_serial_core_attach(pdev))) + if ((ret = ioc4_serial_core_attach(idd->idd_pdev))) goto out4; return ret; @@ -2819,7 +2753,6 @@ out3: out2: release_region(tmp_addr1, sizeof(struct ioc4_serial)); out1: - release_region(tmp_addr, sizeof(struct ioc4_mem)); return ret; } @@ -2828,11 +2761,10 @@ out1: /** * ioc4_serial_remove_one - detach function * - * @pdev: handle for this card + * @idd: IOC4 master module data for this IOC4 */ -#if 0 -void ioc4_serial_remove_one(struct pci_dev *pdev) +int ioc4_serial_remove_one(struct ioc4_driver_data *idd) { int ii; struct ioc4_control *control; @@ -2840,7 +2772,7 @@ void ioc4_serial_remove_one(struct pci_dev *pdev) struct ioc4_port *port; struct ioc4_soft *soft; - control = pci_get_drvdata(pdev); + control = idd->idd_serial_data; for (ii = 0; ii < IOC4_NUM_SERIAL_PORTS; ii++) { the_port = &control->ic_port[ii].icp_uart_port; @@ -2867,10 +2799,17 @@ void ioc4_serial_remove_one(struct pci_dev *pdev) kfree(soft); } kfree(control); - pci_set_drvdata(pdev, NULL); - uart_unregister_driver(&ioc4_uart); + idd->idd_serial_data = NULL; + + return 0; } -#endif + +static struct ioc4_submodule ioc4_serial_submodule = { + .is_name = "IOC4_serial", + .is_owner = THIS_MODULE, + .is_probe = ioc4_serial_attach_one, + .is_remove = ioc4_serial_remove_one, +}; /** * ioc4_serial_init - module init @@ -2886,12 +2825,20 @@ int ioc4_serial_init(void) __FUNCTION__); return ret; } - return 0; + + /* register with IOC4 main module */ + return ioc4_register_submodule(&ioc4_serial_submodule); } +static void __devexit ioc4_serial_exit(void) +{ + ioc4_unregister_submodule(&ioc4_serial_submodule); + uart_unregister_driver(&ioc4_uart); +} + +module_init(ioc4_serial_init); +module_exit(ioc4_serial_exit); + MODULE_AUTHOR("Pat Gefre - Silicon Graphics Inc. (SGI) "); MODULE_DESCRIPTION("Serial PCI driver module for SGI IOC4 Base-IO Card"); MODULE_LICENSE("GPL"); - -EXPORT_SYMBOL(ioc4_serial_init); -EXPORT_SYMBOL(ioc4_serial_attach_one); diff --git a/drivers/sn/ioc4.c b/drivers/sn/ioc4.c index d9e4ee280e5f..70862d72ea9d 100644 --- a/drivers/sn/ioc4.c +++ b/drivers/sn/ioc4.c @@ -6,60 +6,294 @@ * Copyright (C) 2005 Silicon Graphics, Inc. All Rights Reserved. */ -/* - * This file contains a shim driver for the IOC4 IDE and serial drivers. +/* This file contains the master driver module for use by SGI IOC4 subdrivers. + * + * It allocates any resources shared between multiple subdevices, and + * provides accessor functions (where needed) and the like for those + * resources. It also provides a mechanism for the subdevice modules + * to support loading and unloading. + * + * Non-shared resources (e.g. external interrupt A_INT_OUT register page + * alias, serial port and UART registers) are handled by the subdevice + * modules themselves. + * + * This is all necessary because IOC4 is not implemented as a multi-function + * PCI device, but an amalgamation of disparate registers for several + * types of device (ATA, serial, external interrupts). The normal + * resource management in the kernel doesn't have quite the right interfaces + * to handle this situation (e.g. multiple modules can't claim the same + * PCI ID), thus this IOC4 master module. */ #include #include #include -#include -#include +#include +#include +/************************ + * Submodule management * + ************************/ -static int __devinit -ioc4_probe_one(struct pci_dev *pdev, const struct pci_device_id *pci_id) +static LIST_HEAD(ioc4_devices); +static DECLARE_RWSEM(ioc4_devices_rwsem); + +static LIST_HEAD(ioc4_submodules); +static DECLARE_RWSEM(ioc4_submodules_rwsem); + +/* Register an IOC4 submodule */ +int +ioc4_register_submodule(struct ioc4_submodule *is) +{ + struct ioc4_driver_data *idd; + + down_write(&ioc4_submodules_rwsem); + list_add(&is->is_list, &ioc4_submodules); + up_write(&ioc4_submodules_rwsem); + + /* Initialize submodule for each IOC4 */ + if (!is->is_probe) + return 0; + + down_read(&ioc4_devices_rwsem); + list_for_each_entry(idd, &ioc4_devices, idd_list) { + if (is->is_probe(idd)) { + printk(KERN_WARNING + "%s: IOC4 submodule %s probe failed " + "for pci_dev %s", + __FUNCTION__, module_name(is->is_owner), + pci_name(idd->idd_pdev)); + } + } + up_read(&ioc4_devices_rwsem); + + return 0; +} + +/* Unregister an IOC4 submodule */ +void +ioc4_unregister_submodule(struct ioc4_submodule *is) { + struct ioc4_driver_data *idd; + + down_write(&ioc4_submodules_rwsem); + list_del(&is->is_list); + up_write(&ioc4_submodules_rwsem); + + /* Remove submodule for each IOC4 */ + if (!is->is_remove) + return; + + down_read(&ioc4_devices_rwsem); + list_for_each_entry(idd, &ioc4_devices, idd_list) { + if (is->is_remove(idd)) { + printk(KERN_WARNING + "%s: IOC4 submodule %s remove failed " + "for pci_dev %s.\n", + __FUNCTION__, module_name(is->is_owner), + pci_name(idd->idd_pdev)); + } + } + up_read(&ioc4_devices_rwsem); +} + +/********************* + * Device management * + *********************/ + +/* Adds a new instance of an IOC4 card */ +static int +ioc4_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id) +{ + struct ioc4_driver_data *idd; + struct ioc4_submodule *is; + uint32_t pcmd; int ret; + /* Enable IOC4 and take ownership of it */ if ((ret = pci_enable_device(pdev))) { printk(KERN_WARNING - "%s: Failed to enable device with " - "pci_dev 0x%p... returning\n", - __FUNCTION__, (void *)pdev); - return ret; + "%s: Failed to enable IOC4 device for pci_dev %s.\n", + __FUNCTION__, pci_name(pdev)); + goto out; } pci_set_master(pdev); - /* attach each sub-device */ - ret = ioc4_ide_attach_one(pdev, pci_id); - if (ret) - return ret; - return ioc4_serial_attach_one(pdev, pci_id); + /* Set up per-IOC4 data */ + idd = kmalloc(sizeof(struct ioc4_driver_data), GFP_KERNEL); + if (!idd) { + printk(KERN_WARNING + "%s: Failed to allocate IOC4 data for pci_dev %s.\n", + __FUNCTION__, pci_name(pdev)); + ret = -ENODEV; + goto out_idd; + } + idd->idd_pdev = pdev; + idd->idd_pci_id = pci_id; + + /* Map IOC4 misc registers. These are shared between subdevices + * so the main IOC4 module manages them. + */ + idd->idd_bar0 = pci_resource_start(idd->idd_pdev, 0); + if (!idd->idd_bar0) { + printk(KERN_WARNING + "%s: Unable to find IOC4 misc resource " + "for pci_dev %s.\n", + __FUNCTION__, pci_name(idd->idd_pdev)); + ret = -ENODEV; + goto out_pci; + } + if (!request_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs), + "ioc4_misc")) { + printk(KERN_WARNING + "%s: Unable to request IOC4 misc region " + "for pci_dev %s.\n", + __FUNCTION__, pci_name(idd->idd_pdev)); + ret = -ENODEV; + goto out_pci; + } + idd->idd_misc_regs = ioremap(idd->idd_bar0, + sizeof(struct ioc4_misc_regs)); + if (!idd->idd_misc_regs) { + printk(KERN_WARNING + "%s: Unable to remap IOC4 misc region " + "for pci_dev %s.\n", + __FUNCTION__, pci_name(idd->idd_pdev)); + ret = -ENODEV; + goto out_misc_region; + } + + /* Failsafe portion of per-IOC4 initialization */ + + /* Initialize IOC4 */ + pci_read_config_dword(idd->idd_pdev, PCI_COMMAND, &pcmd); + pci_write_config_dword(idd->idd_pdev, PCI_COMMAND, + pcmd | PCI_COMMAND_PARITY | PCI_COMMAND_SERR); + + /* Disable/clear all interrupts. Need to do this here lest + * one submodule request the shared IOC4 IRQ, but interrupt + * is generated by a different subdevice. + */ + /* Disable */ + writel(~0, &idd->idd_misc_regs->other_iec.raw); + writel(~0, &idd->idd_misc_regs->sio_iec); + /* Clear (i.e. acknowledge) */ + writel(~0, &idd->idd_misc_regs->other_ir.raw); + writel(~0, &idd->idd_misc_regs->sio_ir); + + /* Track PCI-device specific data */ + idd->idd_serial_data = NULL; + pci_set_drvdata(idd->idd_pdev, idd); + down_write(&ioc4_devices_rwsem); + list_add(&idd->idd_list, &ioc4_devices); + up_write(&ioc4_devices_rwsem); + + /* Add this IOC4 to all submodules */ + down_read(&ioc4_submodules_rwsem); + list_for_each_entry(is, &ioc4_submodules, is_list) { + if (is->is_probe && is->is_probe(idd)) { + printk(KERN_WARNING + "%s: IOC4 submodule 0x%s probe failed " + "for pci_dev %s.\n", + __FUNCTION__, module_name(is->is_owner), + pci_name(idd->idd_pdev)); + } + } + up_read(&ioc4_submodules_rwsem); + + return 0; + +out_misc_region: + release_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs)); +out_pci: + kfree(idd); +out_idd: + pci_disable_device(pdev); +out: + return ret; } -/* pci device struct */ -static struct pci_device_id ioc4_s_id_table[] = { +/* Removes a particular instance of an IOC4 card. */ +static void +ioc4_remove(struct pci_dev *pdev) +{ + struct ioc4_submodule *is; + struct ioc4_driver_data *idd; + + idd = pci_get_drvdata(pdev); + + /* Remove this IOC4 from all submodules */ + down_read(&ioc4_submodules_rwsem); + list_for_each_entry(is, &ioc4_submodules, is_list) { + if (is->is_remove && is->is_remove(idd)) { + printk(KERN_WARNING + "%s: IOC4 submodule 0x%s remove failed " + "for pci_dev %s.\n", + __FUNCTION__, module_name(is->is_owner), + pci_name(idd->idd_pdev)); + } + } + up_read(&ioc4_submodules_rwsem); + + /* Release resources */ + iounmap(idd->idd_misc_regs); + if (!idd->idd_bar0) { + printk(KERN_WARNING + "%s: Unable to get IOC4 misc mapping for pci_dev %s. " + "Device removal may be incomplete.\n", + __FUNCTION__, pci_name(idd->idd_pdev)); + } + release_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs)); + + /* Disable IOC4 and relinquish */ + pci_disable_device(pdev); + + /* Remove and free driver data */ + down_write(&ioc4_devices_rwsem); + list_del(&idd->idd_list); + up_write(&ioc4_devices_rwsem); + kfree(idd); +} + +static struct pci_device_id ioc4_id_table[] = { {PCI_VENDOR_ID_SGI, PCI_DEVICE_ID_SGI_IOC4, PCI_ANY_ID, PCI_ANY_ID, 0x0b4000, 0xFFFFFF}, {0} }; -MODULE_DEVICE_TABLE(pci, ioc4_s_id_table); -static struct pci_driver __devinitdata ioc4_s_driver = { - .name = "IOC4", - .id_table = ioc4_s_id_table, - .probe = ioc4_probe_one, +static struct pci_driver __devinitdata ioc4_driver = { + .name = "IOC4", + .id_table = ioc4_id_table, + .probe = ioc4_probe, + .remove = ioc4_remove, }; -static int __devinit ioc4_detect(void) +MODULE_DEVICE_TABLE(pci, ioc4_id_table); + +/********************* + * Module management * + *********************/ + +/* Module load */ +static int __devinit +ioc4_init(void) { - ioc4_serial_init(); + return pci_register_driver(&ioc4_driver); +} - return pci_register_driver(&ioc4_s_driver); +/* Module unload */ +static void __devexit +ioc4_exit(void) +{ + pci_unregister_driver(&ioc4_driver); } -module_init(ioc4_detect); -MODULE_AUTHOR("Pat Gefre - Silicon Graphics Inc. (SGI) "); -MODULE_DESCRIPTION("PCI driver module for SGI IOC4 Base-IO Card"); +module_init(ioc4_init); +module_exit(ioc4_exit); + +MODULE_AUTHOR("Brent Casavant - Silicon Graphics, Inc. "); +MODULE_DESCRIPTION("PCI driver master module for SGI IOC4 Base-IO Card"); MODULE_LICENSE("GPL"); + +EXPORT_SYMBOL(ioc4_register_submodule); +EXPORT_SYMBOL(ioc4_unregister_submodule); diff --git a/include/linux/ioc4.h b/include/linux/ioc4.h new file mode 100644 index 000000000000..729bfa4c4ac5 --- /dev/null +++ b/include/linux/ioc4.h @@ -0,0 +1,156 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2005 Silicon Graphics, Inc. All Rights Reserved. + */ + +#ifndef _LINUX_IOC4_H +#define _LINUX_IOC4_H + +#include + +/*********************************** + * Structures needed by subdrivers * + ***********************************/ + +/* This structure fully describes the IOC4 miscellaneous registers which + * appear at bar[0]+0x00000 through bar[0]+0x0005c. The corresponding + * PCI resource is managed by the main IOC4 driver because it contains + * registers of interest to many different IOC4 subdrivers. + */ +struct ioc4_misc_regs { + /* Miscellaneous IOC4 registers */ + union ioc4_pci_err_addr_l { + uint32_t raw; + struct { + uint32_t valid:1; /* Address captured */ + uint32_t master_id:4; /* Unit causing error + * 0/1: Serial port 0 TX/RX + * 2/3: Serial port 1 TX/RX + * 4/5: Serial port 2 TX/RX + * 6/7: Serial port 3 TX/RX + * 8: ATA/ATAPI + * 9-15: Undefined + */ + uint32_t mul_err:1; /* Multiple errors occurred */ + uint32_t addr:26; /* Bits 31-6 of error addr */ + } fields; + } pci_err_addr_l; + uint32_t pci_err_addr_h; /* Bits 63-32 of error addr */ + union ioc4_sio_int { + uint32_t raw; + struct { + uint8_t tx_mt:1; /* TX ring buffer empty */ + uint8_t rx_full:1; /* RX ring buffer full */ + uint8_t rx_high:1; /* RX high-water exceeded */ + uint8_t rx_timer:1; /* RX timer has triggered */ + uint8_t delta_dcd:1; /* DELTA_DCD seen */ + uint8_t delta_cts:1; /* DELTA_CTS seen */ + uint8_t intr_pass:1; /* Interrupt pass-through */ + uint8_t tx_explicit:1; /* TX, MCW, or delay complete */ + } fields[4]; + } sio_ir; /* Serial interrupt state */ + union ioc4_other_int { + uint32_t raw; + struct { + uint32_t ata_int:1; /* ATA port passthru */ + uint32_t ata_memerr:1; /* ATA halted by mem error */ + uint32_t memerr:4; /* Serial halted by mem err */ + uint32_t kbd_int:1; /* kbd/mouse intr asserted */ + uint32_t reserved:16; /* zero */ + uint32_t rt_int:1; /* INT_OUT section latch */ + uint32_t gen_int:8; /* Intr. from generic pins */ + } fields; + } other_ir; /* Other interrupt state */ + union ioc4_sio_int sio_ies; /* Serial interrupt enable set */ + union ioc4_other_int other_ies; /* Other interrupt enable set */ + union ioc4_sio_int sio_iec; /* Serial interrupt enable clear */ + union ioc4_other_int other_iec; /* Other interrupt enable clear */ + union ioc4_sio_cr { + uint32_t raw; + struct { + uint32_t cmd_pulse:4; /* Bytebus strobe width */ + uint32_t arb_diag:3; /* PCI bus requester */ + uint32_t sio_diag_idle:1; /* Active ser req? */ + uint32_t ata_diag_idle:1; /* Active ATA req? */ + uint32_t ata_diag_active:1; /* ATA req is winner */ + uint32_t reserved:22; /* zero */ + } fields; + } sio_cr; + uint32_t unused1; + union ioc4_int_out { + uint32_t raw; + struct { + uint32_t count:16; /* Period control */ + uint32_t mode:3; /* Output signal shape */ + uint32_t reserved:11; /* zero */ + uint32_t diag:1; /* Timebase control */ + uint32_t int_out:1; /* Current value */ + } fields; + } int_out; /* External interrupt output control */ + uint32_t unused2; + union ioc4_gpcr { + uint32_t raw; + struct { + uint32_t dir:8; /* Pin direction */ + uint32_t edge:8; /* Edge/level mode */ + uint32_t reserved1:4; /* zero */ + uint32_t int_out_en:1; /* INT_OUT enable */ + uint32_t reserved2:11; /* zero */ + } fields; + } gpcr_s; /* Generic PIO control set */ + union ioc4_gpcr gpcr_c; /* Generic PIO control clear */ + union ioc4_gpdr { + uint32_t raw; + struct { + uint32_t gen_pin:8; /* State of pins */ + uint32_t reserved:24; + } fields; + } gpdr; /* Generic PIO data */ + uint32_t unused3; + union ioc4_gppr { + uint32_t raw; + struct { + uint32_t gen_pin:1; /* Single pin state */ + uint32_t reserved:31; + } fields; + } gppr[8]; /* Generic PIO pins */ +}; + +/* One of these per IOC4 + * + * The idd_serial_data field is present here, even though it's used + * solely by the serial subdriver, because the main IOC4 module + * properly owns pci_{get,set}_drvdata functionality. This field + * allows that subdriver to stash its own drvdata somewhere. + */ +struct ioc4_driver_data { + struct list_head idd_list; + unsigned long idd_bar0; + struct pci_dev *idd_pdev; + const struct pci_device_id *idd_pci_id; + struct __iomem ioc4_misc_regs *idd_misc_regs; + void *idd_serial_data; +}; + +/* One per submodule */ +struct ioc4_submodule { + struct list_head is_list; + char *is_name; + struct module *is_owner; + int (*is_probe) (struct ioc4_driver_data *); + int (*is_remove) (struct ioc4_driver_data *); +}; + +#define IOC4_NUM_CARDS 8 /* max cards per partition */ + +/********************************** + * Functions needed by submodules * + **********************************/ + +extern int ioc4_register_submodule(struct ioc4_submodule *); +extern void ioc4_unregister_submodule(struct ioc4_submodule *); + +#endif /* _LINUX_IOC4_H */ diff --git a/include/linux/ioc4_common.h b/include/linux/ioc4_common.h deleted file mode 100644 index b03bcc46df55..000000000000 --- a/include/linux/ioc4_common.h +++ /dev/null @@ -1,21 +0,0 @@ -/* - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - * Copyright (c) 2005 Silicon Graphics, Inc. All Rights Reserved. - */ - -#ifndef _LINUX_IOC4_COMMON_H -#define _LINUX_IOC4_COMMON_H - -/* prototypes */ - -int ioc4_serial_init(void); - -int ioc4_serial_attach_one(struct pci_dev *pdev, const struct - pci_device_id *pci_id); -int ioc4_ide_attach_one(struct pci_dev *pdev, const struct - pci_device_id *pci_id); - -#endif /* _LINUX_IOC4_COMMON_H */ -- cgit v1.2.3 From d4c477ca5448f19afaaf6c0cfd655009ea9e614d Mon Sep 17 00:00:00 2001 From: Brent Casavant Date: Tue, 21 Jun 2005 17:16:01 -0700 Subject: [PATCH] ioc4: PCI bus speed detection Several hardware features of SGI's IOC4 I/O controller chip require timing-related driver calculations dependent upon the PCI bus speed. This patch enables the core IOC4 driver code to detect the actual bus speed and store a value that can later be used by the IOC4 subdrivers as needed. Signed-off-by: Brent Casavant Acked-by: Pat Gefre Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/ioc4_serial.c | 31 ++++++----- drivers/sn/Kconfig | 2 +- drivers/sn/ioc4.c | 128 +++++++++++++++++++++++++++++++++++++++++++ include/linux/ioc4.h | 37 ++++++++++--- 4 files changed, 177 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index da5f10eb4845..793c3a7cbe47 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -299,7 +299,6 @@ struct ioc4_serial { } ioc4_serial; /* UART clock speed */ -#define IOC4_SER_XIN_CLK IOC4_SER_XIN_CLK_66 #define IOC4_SER_XIN_CLK_66 66666667 #define IOC4_SER_XIN_CLK_33 33333333 @@ -1012,21 +1011,20 @@ static irqreturn_t ioc4_intr(int irq, void *arg, struct pt_regs *regs) * ioc4_attach_local - Device initialization. * Called at *_attach() time for each * IOC4 with serial ports in the system. - * @control: ioc4_control ptr - * @pdev: PCI handle for this device - * @soft: soft struct for this device - * @ioc4: ioc4 mem space + * @idd: Master module data for this IOC4 */ -static int inline ioc4_attach_local(struct pci_dev *pdev, - struct ioc4_control *control, - struct ioc4_soft *soft, void __iomem *ioc4_misc, - void __iomem *ioc4_serial) +static int inline ioc4_attach_local(struct ioc4_driver_data *idd) { struct ioc4_port *port; struct ioc4_port *ports[IOC4_NUM_SERIAL_PORTS]; int port_number; uint16_t ioc4_revid_min = 62; uint16_t ioc4_revid; + struct pci_dev *pdev = idd->idd_pdev; + struct ioc4_control* control = idd->idd_serial_data; + struct ioc4_soft *soft = control->ic_soft; + void __iomem *ioc4_misc = idd->idd_misc_regs; + void __iomem *ioc4_serial = soft->is_ioc4_serial_addr; /* IOC4 firmware must be at least rev 62 */ pci_read_config_word(pdev, PCI_COMMAND_SPECIAL, &ioc4_revid); @@ -1063,7 +1061,15 @@ static int inline ioc4_attach_local(struct pci_dev *pdev, port->ip_ioc4_soft = soft; port->ip_pdev = pdev; port->ip_ienb = 0; - port->ip_pci_bus_speed = IOC4_SER_XIN_CLK; + /* Use baud rate calculations based on detected PCI + * bus speed. Simply test whether the PCI clock is + * running closer to 66MHz or 33MHz. + */ + if (idd->count_period/IOC4_EXTINT_COUNT_DIVISOR < 20) { + port->ip_pci_bus_speed = IOC4_SER_XIN_CLK_66; + } else { + port->ip_pci_bus_speed = IOC4_SER_XIN_CLK_33; + } port->ip_baud = 9600; port->ip_control = control; port->ip_mem = ioc4_misc; @@ -2733,9 +2739,8 @@ ioc4_serial_attach_one(struct ioc4_driver_data *idd) "%s : request_irq fails for IRQ 0x%x\n ", __FUNCTION__, idd->idd_pdev->irq); } - if ((ret = ioc4_attach_local(idd->idd_pdev, control, soft, - soft->is_ioc4_misc_addr, - soft->is_ioc4_serial_addr))) + ret = ioc4_attach_local(idd); + if (ret) goto out4; /* register port with the serial core */ diff --git a/drivers/sn/Kconfig b/drivers/sn/Kconfig index 20f7515ab830..13b8d249da5c 100644 --- a/drivers/sn/Kconfig +++ b/drivers/sn/Kconfig @@ -6,7 +6,7 @@ menu "SN Devices" config SGI_IOC4 tristate "SGI IOC4 Base IO support" - depends on IA64_GENERIC || IA64_SGI_SN2 + depends on (IA64_GENERIC || IA64_SGI_SN2) && MMTIMER default m ---help--- This option enables basic support for the SGI IOC4-based Base IO diff --git a/drivers/sn/ioc4.c b/drivers/sn/ioc4.c index 70862d72ea9d..ea75b3d0612b 100644 --- a/drivers/sn/ioc4.c +++ b/drivers/sn/ioc4.c @@ -29,7 +29,26 @@ #include #include #include +#include +#include #include +#include +#include +#include + +/*************** + * Definitions * + ***************/ + +/* Tweakable values */ + +/* PCI bus speed detection/calibration */ +#define IOC4_CALIBRATE_COUNT 63 /* Calibration cycle period */ +#define IOC4_CALIBRATE_CYCLES 256 /* Average over this many cycles */ +#define IOC4_CALIBRATE_DISCARD 2 /* Discard first few cycles */ +#define IOC4_CALIBRATE_LOW_MHZ 25 /* Lower bound on bus speed sanity */ +#define IOC4_CALIBRATE_HIGH_MHZ 75 /* Upper bound on bus speed sanity */ +#define IOC4_CALIBRATE_DEFAULT_MHZ 66 /* Assumed if sanity check fails */ /************************ * Submodule management * @@ -101,6 +120,112 @@ ioc4_unregister_submodule(struct ioc4_submodule *is) * Device management * *********************/ +#define IOC4_CALIBRATE_LOW_LIMIT \ + (1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_LOW_MHZ) +#define IOC4_CALIBRATE_HIGH_LIMIT \ + (1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_HIGH_MHZ) +#define IOC4_CALIBRATE_DEFAULT \ + (1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_DEFAULT_MHZ) + +#define IOC4_CALIBRATE_END \ + (IOC4_CALIBRATE_CYCLES + IOC4_CALIBRATE_DISCARD) + +#define IOC4_INT_OUT_MODE_TOGGLE 0x7 /* Toggle INT_OUT every COUNT+1 ticks */ + +/* Determines external interrupt output clock period of the PCI bus an + * IOC4 is attached to. This value can be used to determine the PCI + * bus speed. + * + * IOC4 has a design feature that various internal timers are derived from + * the PCI bus clock. This causes IOC4 device drivers to need to take the + * bus speed into account when setting various register values (e.g. INT_OUT + * register COUNT field, UART divisors, etc). Since this information is + * needed by several subdrivers, it is determined by the main IOC4 driver, + * even though the following code utilizes external interrupt registers + * to perform the speed calculation. + */ +static void +ioc4_clock_calibrate(struct ioc4_driver_data *idd) +{ + extern unsigned long sn_rtc_cycles_per_second; + union ioc4_int_out int_out; + union ioc4_gpcr gpcr; + unsigned int state, last_state = 1; + uint64_t start = 0, end, period; + unsigned int count = 0; + + /* Enable output */ + gpcr.raw = 0; + gpcr.fields.dir = IOC4_GPCR_DIR_0; + gpcr.fields.int_out_en = 1; + writel(gpcr.raw, &idd->idd_misc_regs->gpcr_s.raw); + + /* Reset to power-on state */ + writel(0, &idd->idd_misc_regs->int_out.raw); + mmiowb(); + + printk(KERN_INFO + "%s: Calibrating PCI bus speed " + "for pci_dev %s ... ", __FUNCTION__, pci_name(idd->idd_pdev)); + /* Set up square wave */ + int_out.raw = 0; + int_out.fields.count = IOC4_CALIBRATE_COUNT; + int_out.fields.mode = IOC4_INT_OUT_MODE_TOGGLE; + int_out.fields.diag = 0; + writel(int_out.raw, &idd->idd_misc_regs->int_out.raw); + mmiowb(); + + /* Check square wave period averaged over some number of cycles */ + do { + int_out.raw = readl(&idd->idd_misc_regs->int_out.raw); + state = int_out.fields.int_out; + if (!last_state && state) { + count++; + if (count == IOC4_CALIBRATE_END) { + end = rtc_time(); + break; + } else if (count == IOC4_CALIBRATE_DISCARD) + start = rtc_time(); + } + last_state = state; + } while (1); + + /* Calculation rearranged to preserve intermediate precision. + * Logically: + * 1. "end - start" gives us number of RTC cycles over all the + * square wave cycles measured. + * 2. Divide by number of square wave cycles to get number of + * RTC cycles per square wave cycle. + * 3. Divide by 2*(int_out.fields.count+1), which is the formula + * by which the IOC4 generates the square wave, to get the + * number of RTC cycles per IOC4 INT_OUT count. + * 4. Divide by sn_rtc_cycles_per_second to get seconds per + * count. + * 5. Multiply by 1E9 to get nanoseconds per count. + */ + period = ((end - start) * 1000000000) / + (IOC4_CALIBRATE_CYCLES * 2 * (IOC4_CALIBRATE_COUNT + 1) + * sn_rtc_cycles_per_second); + + /* Bounds check the result. */ + if (period > IOC4_CALIBRATE_LOW_LIMIT || + period < IOC4_CALIBRATE_HIGH_LIMIT) { + printk("failed. Assuming PCI clock ticks are %d ns.\n", + IOC4_CALIBRATE_DEFAULT / IOC4_EXTINT_COUNT_DIVISOR); + period = IOC4_CALIBRATE_DEFAULT; + } else { + printk("succeeded. PCI clock ticks are %ld ns.\n", + period / IOC4_EXTINT_COUNT_DIVISOR); + } + + /* Remember results. We store the extint clock period rather + * than the PCI clock period so that greater precision is + * retained. Divide by IOC4_EXTINT_COUNT_DIVISOR to get + * PCI clock period. + */ + idd->count_period = period; +} + /* Adds a new instance of an IOC4 card */ static int ioc4_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id) @@ -170,6 +295,9 @@ ioc4_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id) pci_write_config_dword(idd->idd_pdev, PCI_COMMAND, pcmd | PCI_COMMAND_PARITY | PCI_COMMAND_SERR); + /* Determine PCI clock */ + ioc4_clock_calibrate(idd); + /* Disable/clear all interrupts. Need to do this here lest * one submodule request the shared IOC4 IRQ, but interrupt * is generated by a different subdevice. diff --git a/include/linux/ioc4.h b/include/linux/ioc4.h index 729bfa4c4ac5..3dd18b785ebd 100644 --- a/include/linux/ioc4.h +++ b/include/linux/ioc4.h @@ -11,6 +11,14 @@ #include +/*************** + * Definitions * + ***************/ + +/* Miscellaneous values inherent to hardware */ + +#define IOC4_EXTINT_COUNT_DIVISOR 520 /* PCI clocks per COUNT tick */ + /*********************************** * Structures needed by subdrivers * ***********************************/ @@ -119,19 +127,34 @@ struct ioc4_misc_regs { } gppr[8]; /* Generic PIO pins */ }; -/* One of these per IOC4 - * - * The idd_serial_data field is present here, even though it's used - * solely by the serial subdriver, because the main IOC4 module - * properly owns pci_{get,set}_drvdata functionality. This field - * allows that subdriver to stash its own drvdata somewhere. - */ +/* Masks for GPCR DIR pins */ +#define IOC4_GPCR_DIR_0 0x01 /* External interrupt output */ +#define IOC4_GPCR_DIR_1 0x02 /* External interrupt input */ +#define IOC4_GPCR_DIR_2 0x04 +#define IOC4_GPCR_DIR_3 0x08 /* Keyboard/mouse presence */ +#define IOC4_GPCR_DIR_4 0x10 /* Ser. port 0 xcvr select (0=232, 1=422) */ +#define IOC4_GPCR_DIR_5 0x20 /* Ser. port 1 xcvr select (0=232, 1=422) */ +#define IOC4_GPCR_DIR_6 0x40 /* Ser. port 2 xcvr select (0=232, 1=422) */ +#define IOC4_GPCR_DIR_7 0x80 /* Ser. port 3 xcvr select (0=232, 1=422) */ + +/* Masks for GPCR EDGE pins */ +#define IOC4_GPCR_EDGE_0 0x01 +#define IOC4_GPCR_EDGE_1 0x02 /* External interrupt input */ +#define IOC4_GPCR_EDGE_2 0x04 +#define IOC4_GPCR_EDGE_3 0x08 +#define IOC4_GPCR_EDGE_4 0x10 +#define IOC4_GPCR_EDGE_5 0x20 +#define IOC4_GPCR_EDGE_6 0x40 +#define IOC4_GPCR_EDGE_7 0x80 + +/* One of these per IOC4 */ struct ioc4_driver_data { struct list_head idd_list; unsigned long idd_bar0; struct pci_dev *idd_pdev; const struct pci_device_id *idd_pci_id; struct __iomem ioc4_misc_regs *idd_misc_regs; + unsigned long count_period; void *idd_serial_data; }; -- cgit v1.2.3 From dbce706e2550253c5ab6043f4f5dfde0cd02470f Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Tue, 21 Jun 2005 17:16:19 -0700 Subject: [PATCH] uml: add and use generic hw_controller_type->release With Chris Wedgwood Currently UML must explicitly call the UML-specific free_irq_by_irq_and_dev() for each free_irq call it's done. This is needed because ->shutdown and/or ->disable are only called when the last "action" for that irq is removed. Instead, for UML shared IRQs (UML IRQs are very often, if not always, shared), for each dev_id some setup is done, which must be cleared on the release of that fd. For instance, for each open console a new instance (i.e. new dev_id) of the same IRQ is requested(). Exactly, a fd is stored in an array (pollfds), which is after read by a host thread and passed to poll(). Each event registered by poll() triggers an interrupt. So, for each free_irq() we must remove the corresponding host fd from the table, which we do via this -release() method. In this patch we add an appropriate hook for this, and remove all uses of it by pointing the hook to the said procedure; this is safe to do since the said procedure. Also some cosmetic improvements are included. This is heavily based on some work by Chris Wedgwood, which however didn't get the patch merged for something I'd call a "misunderstanding" (the need for this patch wasn't cleanly explained, thus adding the generic hook was felt as undesirable). Signed-off-by: Paolo 'Blaisorblade' Giarrusso CC: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/line.c | 2 -- arch/um/drivers/net_kern.c | 1 - arch/um/drivers/port_kern.c | 1 - arch/um/drivers/xterm_kern.c | 1 - arch/um/kernel/irq.c | 11 +++++++---- arch/um/kernel/irq_user.c | 2 -- include/linux/irq.h | 1 + kernel/irq/manage.c | 4 ++++ 8 files changed, 12 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index 025d3be8aca4..562f864254ba 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -406,14 +406,12 @@ void line_disable(struct tty_struct *tty, int current_irq) if(line->driver->read_irq == current_irq) free_irq_later(line->driver->read_irq, tty); else { - free_irq_by_irq_and_dev(line->driver->read_irq, tty); free_irq(line->driver->read_irq, tty); } if(line->driver->write_irq == current_irq) free_irq_later(line->driver->write_irq, tty); else { - free_irq_by_irq_and_dev(line->driver->write_irq, tty); free_irq(line->driver->write_irq, tty); } diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c index 4eeaf88c1e97..5388a7428691 100644 --- a/arch/um/drivers/net_kern.c +++ b/arch/um/drivers/net_kern.c @@ -146,7 +146,6 @@ static int uml_net_close(struct net_device *dev) netif_stop_queue(dev); spin_lock(&lp->lock); - free_irq_by_irq_and_dev(dev->irq, dev); free_irq(dev->irq, dev); if(lp->close != NULL) (*lp->close)(lp->fd, &lp->user); diff --git a/arch/um/drivers/port_kern.c b/arch/um/drivers/port_kern.c index b5ee07472f79..c41efd207fcc 100644 --- a/arch/um/drivers/port_kern.c +++ b/arch/um/drivers/port_kern.c @@ -257,7 +257,6 @@ int port_wait(void *data) * connection. Then we loop here throwing out failed * connections until a good one is found. */ - free_irq_by_irq_and_dev(TELNETD_IRQ, conn); free_irq(TELNETD_IRQ, conn); if(conn->fd >= 0) break; diff --git a/arch/um/drivers/xterm_kern.c b/arch/um/drivers/xterm_kern.c index a4fdf3584ad2..d269a80f4b0c 100644 --- a/arch/um/drivers/xterm_kern.c +++ b/arch/um/drivers/xterm_kern.c @@ -69,7 +69,6 @@ int xterm_fd(int socket, int *pid_out) * isn't set) this will hang... */ wait_for_completion(&data->ready); - free_irq_by_irq_and_dev(XTERM_IRQ, data); free_irq(XTERM_IRQ, data); ret = data->new_fd; diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c index d44fb5282547..9f18061ef4c9 100644 --- a/arch/um/kernel/irq.c +++ b/arch/um/kernel/irq.c @@ -124,14 +124,16 @@ void irq_unlock(unsigned long flags) spin_unlock_irqrestore(&irq_spinlock, flags); } -/* presently hw_interrupt_type must define (startup || enable) && - * disable && end */ +/* hw_interrupt_type must define (startup || enable) && + * (shutdown || disable) && end */ static void dummy(unsigned int irq) { } -static struct hw_interrupt_type SIGIO_irq_type = { +/* This is used for everything else than the timer. */ +static struct hw_interrupt_type normal_irq_type = { .typename = "SIGIO", + .release = free_irq_by_irq_and_dev, .disable = dummy, .enable = dummy, .ack = dummy, @@ -140,6 +142,7 @@ static struct hw_interrupt_type SIGIO_irq_type = { static struct hw_interrupt_type SIGVTALRM_irq_type = { .typename = "SIGVTALRM", + .release = free_irq_by_irq_and_dev, .shutdown = dummy, /* never called */ .disable = dummy, .enable = dummy, @@ -160,7 +163,7 @@ void __init init_IRQ(void) irq_desc[i].status = IRQ_DISABLED; irq_desc[i].action = NULL; irq_desc[i].depth = 1; - irq_desc[i].handler = &SIGIO_irq_type; + irq_desc[i].handler = &normal_irq_type; enable_irq(i); } } diff --git a/arch/um/kernel/irq_user.c b/arch/um/kernel/irq_user.c index b3074cbaa479..c3ccaf24f3e0 100644 --- a/arch/um/kernel/irq_user.c +++ b/arch/um/kernel/irq_user.c @@ -85,8 +85,6 @@ void sigio_handler(int sig, union uml_pt_regs *regs) next = irq_fd->next; if(irq_fd->freed){ free_irq(irq_fd->irq, irq_fd->id); - free_irq_by_irq_and_dev(irq_fd->irq, - irq_fd->id); } } } diff --git a/include/linux/irq.h b/include/linux/irq.h index c3ff4d101667..b68ad80e2464 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -47,6 +47,7 @@ struct hw_interrupt_type { void (*ack)(unsigned int irq); void (*end)(unsigned int irq); void (*set_affinity)(unsigned int irq, cpumask_t dest); + void (*release)(unsigned int irq, void *dev_id); }; typedef struct hw_interrupt_type hw_irq_controller; diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 5202e4c4a5b6..5fde8177eedf 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -255,6 +255,10 @@ void free_irq(unsigned int irq, void *dev_id) /* Found it - now remove it from the list of entries */ *pp = action->next; + + if (desc->handler->release) + desc->handler->release(irq, dev_id); + if (!desc->action) { desc->status |= IRQ_DISABLED; if (desc->handler->shutdown) -- cgit v1.2.3 From b77d6adc922b8bbf8b16b67f567958c42962cf88 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Tue, 21 Jun 2005 17:16:24 -0700 Subject: [PATCH] uml: make hw_controller_type->release exist only for archs needing it With Chris Wedgwood As suggested by Chris, we can make the "just added" method ->release conditional to UML only (better: to archs requesting it, i.e. only UML currently), so that other archs don't get this unneeded crud, and if UML won't need it any more we can kill this. Signed-off-by: Paolo 'Blaisorblade' Giarrusso CC: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/Kconfig | 5 +++++ include/linux/irq.h | 3 +++ kernel/irq/manage.c | 4 ++++ 3 files changed, 12 insertions(+) (limited to 'include/linux') diff --git a/arch/um/Kconfig b/arch/um/Kconfig index c5292181a664..b8e952c88fd1 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -35,6 +35,11 @@ config GENERIC_CALIBRATE_DELAY bool default y +# Used in kernel/irq/manage.c and include/linux/irq.h +config IRQ_RELEASE_METHOD + bool + default y + menu "UML-specific options" config MODE_TT diff --git a/include/linux/irq.h b/include/linux/irq.h index b68ad80e2464..7fc1022be9ee 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -47,7 +47,10 @@ struct hw_interrupt_type { void (*ack)(unsigned int irq); void (*end)(unsigned int irq); void (*set_affinity)(unsigned int irq, cpumask_t dest); + /* Currently used only by UML, might disappear one day.*/ +#ifdef CONFIG_IRQ_RELEASE_METHOD void (*release)(unsigned int irq, void *dev_id); +#endif }; typedef struct hw_interrupt_type hw_irq_controller; diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 5fde8177eedf..ac6700985705 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -6,6 +6,7 @@ * This file contains driver APIs to the irq subsystem. */ +#include #include #include #include @@ -256,8 +257,11 @@ void free_irq(unsigned int irq, void *dev_id) /* Found it - now remove it from the list of entries */ *pp = action->next; + /* Currently used only by UML, might disappear one day.*/ +#ifdef CONFIG_IRQ_RELEASE_METHOD if (desc->handler->release) desc->handler->release(irq, dev_id); +#endif if (!desc->action) { desc->status |= IRQ_DISABLED; -- cgit v1.2.3 From 8a96619145840c6eb50d85759f72d9337db411f7 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Tue, 21 Jun 2005 17:16:41 -0700 Subject: [PATCH] autofs4: subversion bump to identify these changes Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/auto_fs4.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h index a1657fb99516..9343c89d843c 100644 --- a/include/linux/auto_fs4.h +++ b/include/linux/auto_fs4.h @@ -23,7 +23,7 @@ #define AUTOFS_MIN_PROTO_VERSION 3 #define AUTOFS_MAX_PROTO_VERSION 4 -#define AUTOFS_PROTO_SUBVERSION 6 +#define AUTOFS_PROTO_SUBVERSION 7 /* Mask for expire behaviour */ #define AUTOFS_EXP_IMMEDIATE 1 -- cgit v1.2.3 From f5a9951c94e7a285a3d00648e3d790a7f016bd11 Mon Sep 17 00:00:00 2001 From: James Simmons Date: Tue, 21 Jun 2005 17:16:58 -0700 Subject: [PATCH] fbdev: iomove removal Since no one is using the inbuf, outbuf of struct fb_pixmap I removed their use in the framebuffer console. The idea is instead move the pixmap functionality below the accelerated functions intead of on top as the way it is now. If there is no objection please apply. This is against Linus latestr GIT tree. Thank you. Signed-off-by: James Simmons Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/console/bitblit.c | 18 ++----------- drivers/video/fbmem.c | 57 ++++------------------------------------- drivers/video/softcursor.c | 8 ++---- include/linux/fb.h | 12 +++------ 4 files changed, 12 insertions(+), 83 deletions(-) (limited to 'include/linux') diff --git a/drivers/video/console/bitblit.c b/drivers/video/console/bitblit.c index b28a4b0e395e..4eef20790c3e 100644 --- a/drivers/video/console/bitblit.c +++ b/drivers/video/console/bitblit.c @@ -107,13 +107,6 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info, const unsigned short *s, int count, int yy, int xx, int fg, int bg) { - void (*move_unaligned)(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 idx, - u32 height, u32 shift_high, u32 shift_low, - u32 mod); - void (*move_aligned)(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, - u32 height); unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff; unsigned int width = (vc->vc_font.width + 7) >> 3; unsigned int cellsize = vc->vc_font.height * width; @@ -141,13 +134,6 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info, image.height = vc->vc_font.height; image.depth = 1; - if (info->pixmap.outbuf && info->pixmap.inbuf) { - move_aligned = fb_iomove_buf_aligned; - move_unaligned = fb_iomove_buf_unaligned; - } else { - move_aligned = fb_sysmove_buf_aligned; - move_unaligned = fb_sysmove_buf_unaligned; - } while (count) { if (count > maxcnt) cnt = k = maxcnt; @@ -171,7 +157,7 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info, src = buf; } - move_unaligned(info, &info->pixmap, dst, pitch, + fb_sysmove_buf_unaligned(info, &info->pixmap, dst, pitch, src, idx, image.height, shift_high, shift_low, mod); shift_low += mod; @@ -189,7 +175,7 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info, src = buf; } - move_aligned(info, &info->pixmap, dst, pitch, + fb_sysmove_buf_aligned(info, &info->pixmap, dst, pitch, src, idx, image.height); dst += width; } diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 8cef020d1801..78907a873493 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -76,65 +76,21 @@ int fb_get_color_depth(struct fb_var_screeninfo *var) EXPORT_SYMBOL(fb_get_color_depth); /* - * Drawing helpers. + * Data padding functions. */ -void fb_iomove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, - u32 height) -{ - int i; - - for (i = height; i--; ) { - buf->outbuf(info, dst, src, s_pitch); - src += s_pitch; - dst += d_pitch; - } -} - void fb_sysmove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf, u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, u32 height) { - int i, j; + int i; for (i = height; i--; ) { - for (j = 0; j < s_pitch; j++) - dst[j] = src[j]; + memcpy(dst, src, s_pitch); src += s_pitch; dst += d_pitch; } } - -void fb_iomove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 idx, - u32 height, u32 shift_high, u32 shift_low, - u32 mod) -{ - u8 mask = (u8) (0xfff << shift_high), tmp; - int i, j; - - for (i = height; i--; ) { - for (j = 0; j < idx; j++) { - tmp = buf->inbuf(info, dst+j); - tmp &= mask; - tmp |= *src >> shift_low; - buf->outbuf(info, dst+j, &tmp, 1); - tmp = *src << shift_high; - buf->outbuf(info, dst+j+1, &tmp, 1); - src++; - } - tmp = buf->inbuf(info, dst+idx); - tmp &= mask; - tmp |= *src >> shift_low; - buf->outbuf(info, dst+idx, &tmp, 1); - if (shift_high < mod) { - tmp = *src << shift_high; - buf->outbuf(info, dst+idx+1, &tmp, 1); - } - src++; - dst += d_pitch; - } -} +EXPORT_SYMBOL(fb_sysmove_buf_aligned); void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf, u8 *dst, u32 d_pitch, u8 *src, u32 idx, @@ -166,6 +122,7 @@ void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf, dst += d_pitch; } } +EXPORT_SYMBOL(fb_sysmove_buf_unaligned); /* * we need to lock this section since fb_cursor @@ -1357,10 +1314,6 @@ EXPORT_SYMBOL(fb_set_var); EXPORT_SYMBOL(fb_blank); EXPORT_SYMBOL(fb_pan_display); EXPORT_SYMBOL(fb_get_buffer_offset); -EXPORT_SYMBOL(fb_iomove_buf_unaligned); -EXPORT_SYMBOL(fb_iomove_buf_aligned); -EXPORT_SYMBOL(fb_sysmove_buf_unaligned); -EXPORT_SYMBOL(fb_sysmove_buf_aligned); EXPORT_SYMBOL(fb_set_suspend); EXPORT_SYMBOL(fb_register_client); EXPORT_SYMBOL(fb_unregister_client); diff --git a/drivers/video/softcursor.c b/drivers/video/softcursor.c index 13a4511539a1..a6c5ca88d6b0 100644 --- a/drivers/video/softcursor.c +++ b/drivers/video/softcursor.c @@ -58,12 +58,8 @@ int soft_cursor(struct fb_info *info, struct fb_cursor *cursor) } else memcpy(src, image->data, dsize); - if (info->pixmap.outbuf) - fb_iomove_buf_aligned(info, &info->pixmap, dst, d_pitch, src, - s_pitch, image->height); - else - fb_sysmove_buf_aligned(info, &info->pixmap, dst, d_pitch, src, - s_pitch, image->height); + fb_sysmove_buf_aligned(info, &info->pixmap, dst, d_pitch, src, + s_pitch, image->height); image->data = dst; info->fbops->fb_imageblit(info, image); diff --git a/include/linux/fb.h b/include/linux/fb.h index b468bf496547..e14942805dbd 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -524,11 +524,11 @@ struct fb_pixmap { u32 offset; /* current offset to buffer */ u32 buf_align; /* byte alignment of each bitmap */ u32 scan_align; /* alignment per scanline */ - u32 access_align; /* alignment per read/write */ + u32 access_align; /* alignment per read/write (bits) */ u32 flags; /* see FB_PIXMAP_* */ /* access methods */ - void (*outbuf)(struct fb_info *info, u8 *addr, u8 *src, unsigned int size); - u8 (*inbuf) (struct fb_info *info, u8 *addr); + void (*writeio)(struct fb_info *info, void __iomem *dst, void *src, unsigned int size); + void (*readio) (struct fb_info *info, void *dst, void __iomem *src, unsigned int size); }; @@ -816,12 +816,6 @@ extern int unregister_framebuffer(struct fb_info *fb_info); extern int fb_prepare_logo(struct fb_info *fb_info); extern int fb_show_logo(struct fb_info *fb_info); extern char* fb_get_buffer_offset(struct fb_info *info, struct fb_pixmap *buf, u32 size); -extern void fb_iomove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 idx, - u32 height, u32 shift_high, u32 shift_low, u32 mod); -extern void fb_iomove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, - u32 height); extern void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf, u8 *dst, u32 d_pitch, u8 *src, u32 idx, u32 height, u32 shift_high, u32 shift_low, u32 mod); -- cgit v1.2.3 From 1154ea7dcd8eed758fb5ec47393a79d5a1f0bc43 Mon Sep 17 00:00:00 2001 From: Jaya Kumar Date: Tue, 21 Jun 2005 17:17:04 -0700 Subject: [PATCH] Framebuffer driver for Arc LCD board Add support for the Arc monochrome LCD board. The board uses KS108 controllers to drive individual 64x64 LCD matrices. The board can be paneled in a variety of setups such as 2x1=128x64, 4x4=256x256 and so on. The board/host interface is through GPIO. Signed-off-by: Jaya Kumar Cc: "Antonino A. Daplas" Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- CREDITS | 7 + MAINTAINERS | 5 + drivers/video/Kconfig | 16 ++ drivers/video/Makefile | 1 + drivers/video/arcfb.c | 684 +++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/arcfb.h | 8 + 6 files changed, 721 insertions(+) create mode 100644 drivers/video/arcfb.c create mode 100644 include/linux/arcfb.h (limited to 'include/linux') diff --git a/CREDITS b/CREDITS index d65ffe5a4d08..2993348ce6ce 100644 --- a/CREDITS +++ b/CREDITS @@ -1880,6 +1880,13 @@ S: Schlehenweg 9 S: D-91080 Uttenreuth S: Germany +N: Jaya Kumar +E: jayalk@intworks.biz +W: http://www.intworks.biz +D: Arc monochrome LCD framebuffer driver, x86 reboot fixups +S: Gurgaon, India +S: Kuala Lumpur, Malaysia + N: Gabor Kuti M: seasons@falcon.sch.bme.hu M: seasons@makosteszta.sote.hu diff --git a/MAINTAINERS b/MAINTAINERS index e3f0f3f157c9..0f88a70a15d8 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -265,6 +265,11 @@ P: Arnaldo Carvalho de Melo M: acme@conectiva.com.br S: Maintained +ARC FRAMEBUFFER DRIVER +P: Jaya Kumar +M: jayalk@intworks.biz +S: Maintained + ARM26 ARCHITECTURE P: Ian Molton M: spyro@f2s.com diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig index 6be8fbec0a0e..04d3120f7236 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -322,6 +322,22 @@ config FB_FM2 This is the frame buffer device driver for the Amiga FrameMaster card from BSC (exhibited 1992 but not shipped as a CBM product). +config FB_ARC + tristate "Arc Monochrome LCD board support" + depends on FB && X86 + select FB_CFB_FILLRECT + select FB_CFB_COPYAREA + select FB_CFB_IMAGEBLIT + select FB_SOFT_CURSOR + help + This enables support for the Arc Monochrome LCD board. The board + is based on the KS-108 lcd controller and is typically a matrix + of 2*n chips. This driver was tested with a 128x64 panel. This + driver supports it for use with x86 SBCs through a 16 bit GPIO + interface (8 bit data, 8 bit control). If you anticpate using + this driver, say Y or M; otherwise say N. You must specify the + GPIO IO address to be used for setting control and data. + config FB_ATARI bool "Atari native chipset support" depends on (FB = y) && ATARI && BROKEN diff --git a/drivers/video/Makefile b/drivers/video/Makefile index bd8dc0ffe723..b018df4e95c8 100644 --- a/drivers/video/Makefile +++ b/drivers/video/Makefile @@ -21,6 +21,7 @@ obj-$(CONFIG_FB_MACMODES) += macmodes.o # Hardware specific drivers go first obj-$(CONFIG_FB_RETINAZ3) += retz3fb.o obj-$(CONFIG_FB_AMIGA) += amifb.o c2p.o +obj-$(CONFIG_FB_ARC) += arcfb.o obj-$(CONFIG_FB_CLPS711X) += clps711xfb.o obj-$(CONFIG_FB_CYBER) += cyberfb.o obj-$(CONFIG_FB_CYBER2000) += cyber2000fb.o diff --git a/drivers/video/arcfb.c b/drivers/video/arcfb.c new file mode 100644 index 000000000000..d28457e0c063 --- /dev/null +++ b/drivers/video/arcfb.c @@ -0,0 +1,684 @@ +/* + * linux/drivers/video/arcfb.c -- FB driver for Arc monochrome LCD board + * + * Copyright (C) 2005, Jaya Kumar + * http://www.intworks.biz/arclcd + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file COPYING in the main directory of this archive for + * more details. + * + * Layout is based on skeletonfb.c by James Simmons and Geert Uytterhoeven. + * + * This driver was written to be used with the Arc LCD board. Arc uses a + * set of KS108 chips that control individual 64x64 LCD matrices. The board + * can be paneled in a variety of setups such as 2x1=128x64, 4x4=256x256 and + * so on. The interface between the board and the host is TTL based GPIO. The + * GPIO requirements are 8 writable data lines and 4+n lines for control. On a + * GPIO-less system, the board can be tested by connecting the respective sigs + * up to a parallel port connector. The driver requires the IO addresses for + * data and control GPIO at load time. It is unable to probe for the + * existence of the LCD so it must be told at load time whether it should + * be enabled or not. + * + * Todo: + * - testing with 4x4 + * - testing with interrupt hw + * + * General notes: + * - User must set tuhold. It's in microseconds. According to the 108 spec, + * the hold time is supposed to be at least 1 microsecond. + * - User must set num_cols=x num_rows=y, eg: x=2 means 128 + * - User must set arcfb_enable=1 to enable it + * - User must set dio_addr=0xIOADDR cio_addr=0xIOADDR + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#define floor8(a) (a&(~0x07)) +#define floorXres(a,xres) (a&(~(xres - 1))) +#define iceil8(a) (((int)((a+7)/8))*8) +#define ceil64(a) (a|0x3F) +#define ceilXres(a,xres) (a|(xres - 1)) + +/* ks108 chipset specific defines and code */ + +#define KS_SET_DPY_START_LINE 0xC0 +#define KS_SET_PAGE_NUM 0xB8 +#define KS_SET_X 0x40 +#define KS_CEHI 0x01 +#define KS_CELO 0x00 +#define KS_SEL_CMD 0x08 +#define KS_SEL_DATA 0x00 +#define KS_DPY_ON 0x3F +#define KS_DPY_OFF 0x3E +#define KS_INTACK 0x40 +#define KS_CLRINT 0x02 + +struct arcfb_par { + unsigned long dio_addr; + unsigned long cio_addr; + unsigned long c2io_addr; + atomic_t ref_count; + unsigned char cslut[9]; + struct fb_info *info; + unsigned int irq; + spinlock_t lock; +}; + +static struct fb_fix_screeninfo arcfb_fix __initdata = { + .id = "arcfb", + .type = FB_TYPE_PACKED_PIXELS, + .visual = FB_VISUAL_MONO01, + .xpanstep = 0, + .ypanstep = 1, + .ywrapstep = 0, + .accel = FB_ACCEL_NONE, +}; + +static struct fb_var_screeninfo arcfb_var __initdata = { + .xres = 128, + .yres = 64, + .xres_virtual = 128, + .yres_virtual = 64, + .bits_per_pixel = 1, + .nonstd = 1, +}; + +static unsigned long num_cols; +static unsigned long num_rows; +static unsigned long dio_addr; +static unsigned long cio_addr; +static unsigned long c2io_addr; +static unsigned long splashval; +static unsigned long tuhold; +static unsigned int nosplash; +static unsigned int arcfb_enable; +static unsigned int irq; + +static DECLARE_WAIT_QUEUE_HEAD(arcfb_waitq); + +static void ks108_writeb_ctl(struct arcfb_par *par, + unsigned int chipindex, unsigned char value) +{ + unsigned char chipselval = par->cslut[chipindex]; + + outb(chipselval|KS_CEHI|KS_SEL_CMD, par->cio_addr); + outb(value, par->dio_addr); + udelay(tuhold); + outb(chipselval|KS_CELO|KS_SEL_CMD, par->cio_addr); +} + +static void ks108_writeb_mainctl(struct arcfb_par *par, unsigned char value) +{ + + outb(value, par->cio_addr); + udelay(tuhold); +} + +static unsigned char ks108_readb_ctl2(struct arcfb_par *par) +{ + return inb(par->c2io_addr); +} + +static void ks108_writeb_data(struct arcfb_par *par, + unsigned int chipindex, unsigned char value) +{ + unsigned char chipselval = par->cslut[chipindex]; + + outb(chipselval|KS_CEHI|KS_SEL_DATA, par->cio_addr); + outb(value, par->dio_addr); + udelay(tuhold); + outb(chipselval|KS_CELO|KS_SEL_DATA, par->cio_addr); +} + +static void ks108_set_start_line(struct arcfb_par *par, + unsigned int chipindex, unsigned char y) +{ + ks108_writeb_ctl(par, chipindex, KS_SET_DPY_START_LINE|y); +} + +static void ks108_set_yaddr(struct arcfb_par *par, + unsigned int chipindex, unsigned char y) +{ + ks108_writeb_ctl(par, chipindex, KS_SET_PAGE_NUM|y); +} + +static void ks108_set_xaddr(struct arcfb_par *par, + unsigned int chipindex, unsigned char x) +{ + ks108_writeb_ctl(par, chipindex, KS_SET_X|x); +} + +static void ks108_clear_lcd(struct arcfb_par *par, unsigned int chipindex) +{ + int i,j; + + for (i = 0; i <= 8; i++) { + ks108_set_yaddr(par, chipindex, i); + ks108_set_xaddr(par, chipindex, 0); + for (j = 0; j < 64; j++) { + ks108_writeb_data(par, chipindex, + (unsigned char) splashval); + } + } +} + +/* main arcfb functions */ + +static int arcfb_open(struct fb_info *info, int user) +{ + struct arcfb_par *par = info->par; + + atomic_inc(&par->ref_count); + return 0; +} + +static int arcfb_release(struct fb_info *info, int user) +{ + struct arcfb_par *par = info->par; + int count = atomic_read(&par->ref_count); + + if (!count) + return -EINVAL; + atomic_dec(&par->ref_count); + return 0; +} + +static int arcfb_pan_display(struct fb_var_screeninfo *var, + struct fb_info *info) +{ + int i; + struct arcfb_par *par = info->par; + + if ((var->vmode & FB_VMODE_YWRAP) && (var->yoffset < 64) + && (info->var.yres <= 64)) { + for (i = 0; i < num_cols; i++) { + ks108_set_start_line(par, i, var->yoffset); + } + info->var.yoffset = var->yoffset; + return 0; + } + + return -EINVAL; +} + +static irqreturn_t arcfb_interrupt(int vec, void *dev_instance, + struct pt_regs *regs) +{ + struct fb_info *info = dev_instance; + unsigned char ctl2status; + struct arcfb_par *par = info->par; + + ctl2status = ks108_readb_ctl2(par); + + if (!(ctl2status & KS_INTACK)) /* not arc generated interrupt */ + return IRQ_NONE; + + ks108_writeb_mainctl(par, KS_CLRINT); + + spin_lock(&par->lock); + if (waitqueue_active(&arcfb_waitq)) { + wake_up(&arcfb_waitq); + } + spin_unlock(&par->lock); + + return IRQ_HANDLED; +} + +/* + * here we handle a specific page on the lcd. the complexity comes from + * the fact that the fb is laidout in 8xX vertical columns. we extract + * each write of 8 vertical pixels. then we shift out as we move along + * X. That's what rightshift does. bitmask selects the desired input bit. + */ +static void arcfb_lcd_update_page(struct arcfb_par *par, unsigned int upper, + unsigned int left, unsigned int right, unsigned int distance) +{ + unsigned char *src; + unsigned int xindex, yindex, chipindex, linesize; + int i, count; + unsigned char val; + unsigned char bitmask, rightshift; + + xindex = left >> 6; + yindex = upper >> 6; + chipindex = (xindex + (yindex*num_cols)); + + ks108_set_yaddr(par, chipindex, upper/8); + + linesize = par->info->var.xres/8; + src = par->info->screen_base + (left/8) + (upper * linesize); + ks108_set_xaddr(par, chipindex, left); + + bitmask=1; + rightshift=0; + while (left <= right) { + val = 0; + for (i = 0; i < 8; i++) { + if ( i > rightshift) { + val |= (*(src + (i*linesize)) & bitmask) + << (i - rightshift); + } else { + val |= (*(src + (i*linesize)) & bitmask) + >> (rightshift - i); + } + } + ks108_writeb_data(par, chipindex, val); + left++; + count++; + if (bitmask == 0x80) { + bitmask = 1; + src++; + rightshift=0; + } else { + bitmask <<= 1; + rightshift++; + } + } +} + +/* + * here we handle the entire vertical page of the update. we write across + * lcd chips. update_page uses the upper/left values to decide which + * chip to select for the right. upper is needed for setting the page + * desired for the write. + */ +static void arcfb_lcd_update_vert(struct arcfb_par *par, unsigned int top, + unsigned int bottom, unsigned int left, unsigned int right) +{ + unsigned int distance, upper, lower; + + distance = (bottom - top) + 1; + upper = top; + lower = top + 7; + + while (distance > 0) { + distance -= 8; + arcfb_lcd_update_page(par, upper, left, right, 8); + upper = lower + 1; + lower = upper + 7; + } +} + +/* + * here we handle horizontal blocks for the update. update_vert will + * handle spaning multiple pages. we break out each horizontal + * block in to individual blocks no taller than 64 pixels. + */ +static void arcfb_lcd_update_horiz(struct arcfb_par *par, unsigned int left, + unsigned int right, unsigned int top, unsigned int h) +{ + unsigned int distance, upper, lower; + + distance = h; + upper = floor8(top); + lower = min(upper + distance - 1, ceil64(upper)); + + while (distance > 0) { + distance -= ((lower - upper) + 1 ); + arcfb_lcd_update_vert(par, upper, lower, left, right); + upper = lower + 1; + lower = min(upper + distance - 1, ceil64(upper)); + } +} + +/* + * here we start the process of spliting out the fb update into + * individual blocks of pixels. we end up spliting into 64x64 blocks + * and finally down to 64x8 pages. + */ +static void arcfb_lcd_update(struct arcfb_par *par, unsigned int dx, + unsigned int dy, unsigned int w, unsigned int h) +{ + unsigned int left, right, distance, y; + + /* align the request first */ + y = floor8(dy); + h += dy - y; + h = iceil8(h); + + distance = w; + left = dx; + right = min(left + w - 1, ceil64(left)); + + while (distance > 0) { + arcfb_lcd_update_horiz(par, left, right, y, h); + distance -= ((right - left) + 1); + left = right + 1; + right = min(left + distance - 1, ceil64(left)); + } +} + +void arcfb_fillrect(struct fb_info *info, const struct fb_fillrect *rect) +{ + struct arcfb_par *par = info->par; + + cfb_fillrect(info, rect); + + /* update the physical lcd */ + arcfb_lcd_update(par, rect->dx, rect->dy, rect->width, rect->height); +} + +void arcfb_copyarea(struct fb_info *info, const struct fb_copyarea *area) +{ + struct arcfb_par *par = info->par; + + cfb_copyarea(info, area); + + /* update the physical lcd */ + arcfb_lcd_update(par, area->dx, area->dy, area->width, area->height); +} + +void arcfb_imageblit(struct fb_info *info, const struct fb_image *image) +{ + struct arcfb_par *par = info->par; + + cfb_imageblit(info, image); + + /* update the physical lcd */ + arcfb_lcd_update(par, image->dx, image->dy, image->width, + image->height); +} + +static int arcfb_ioctl(struct inode *inode, struct file *file, + unsigned int cmd, unsigned long arg, + struct fb_info *info) +{ + void __user *argp = (void __user *)arg; + struct arcfb_par *par = info->par; + unsigned long flags; + + switch (cmd) { + case FBIO_WAITEVENT: + { + DEFINE_WAIT(wait); + /* illegal to wait on arc if no irq will occur */ + if (!par->irq) + return -EINVAL; + + /* wait until the Arc has generated an interrupt + * which will wake us up */ + spin_lock_irqsave(&par->lock, flags); + prepare_to_wait(&arcfb_waitq, &wait, + TASK_INTERRUPTIBLE); + spin_unlock_irqrestore(&par->lock, flags); + schedule(); + finish_wait(&arcfb_waitq, &wait); + } + case FBIO_GETCONTROL2: + { + unsigned char ctl2; + + ctl2 = ks108_readb_ctl2(info->par); + if (copy_to_user(argp, &ctl2, sizeof(ctl2))) + return -EFAULT; + return 0; + } + default: + return -EINVAL; + } +} + +/* + * this is the access path from userspace. they can seek and write to + * the fb. it's inefficient for them to do anything less than 64*8 + * writes since we update the lcd in each write() anyway. + */ +static ssize_t arcfb_write(struct file *file, const char *buf, size_t count, + loff_t *ppos) +{ + /* modded from epson 1355 */ + + struct inode *inode; + int fbidx; + struct fb_info *info; + unsigned long p; + int err=-EINVAL; + unsigned int fbmemlength,x,y,w,h, bitppos, startpos, endpos, bitcount; + struct arcfb_par *par; + unsigned int xres; + + p = *ppos; + inode = file->f_dentry->d_inode; + fbidx = iminor(inode); + info = registered_fb[fbidx]; + par = info->par; + + if (!info || !info->screen_base) + return -ENODEV; + + xres = info->var.xres; + fbmemlength = (xres * info->var.yres)/8; + + if (p > fbmemlength) + return -ENOSPC; + + err = 0; + if ((count + p) > fbmemlength) { + count = fbmemlength - p; + err = -ENOSPC; + } + + if (count) { + char *base_addr; + + base_addr = info->screen_base; + count -= copy_from_user(base_addr + p, buf, count); + *ppos += count; + err = -EFAULT; + } + + + bitppos = p*8; + startpos = floorXres(bitppos, xres); + endpos = ceilXres((bitppos + (count*8)), xres); + bitcount = endpos - startpos; + + x = startpos % xres; + y = startpos / xres; + w = xres; + h = bitcount / xres; + arcfb_lcd_update(par, x, y, w, h); + + if (count) + return count; + return err; +} + +static void arcfb_platform_release(struct device *device) +{ +} + +static struct fb_ops arcfb_ops = { + .owner = THIS_MODULE, + .fb_open = arcfb_open, + .fb_write = arcfb_write, + .fb_release = arcfb_release, + .fb_pan_display = arcfb_pan_display, + .fb_fillrect = arcfb_fillrect, + .fb_copyarea = arcfb_copyarea, + .fb_imageblit = arcfb_imageblit, + .fb_cursor = soft_cursor, + .fb_ioctl = arcfb_ioctl, +}; + +static int __init arcfb_probe(struct device *device) +{ + struct platform_device *dev = to_platform_device(device); + struct fb_info *info; + int retval = -ENOMEM; + int videomemorysize; + unsigned char *videomemory; + struct arcfb_par *par; + int i; + + videomemorysize = (((64*64)*num_cols)*num_rows)/8; + + /* We need a flat backing store for the Arc's + less-flat actual paged framebuffer */ + if (!(videomemory = vmalloc(videomemorysize))) + return retval; + + memset(videomemory, 0, videomemorysize); + + info = framebuffer_alloc(sizeof(struct arcfb_par), &dev->dev); + if (!info) + goto err; + + info->screen_base = (char __iomem *)videomemory; + info->fbops = &arcfb_ops; + + info->var = arcfb_var; + info->fix = arcfb_fix; + par = info->par; + par->info = info; + + if (!dio_addr || !cio_addr || !c2io_addr) { + printk(KERN_WARNING "no IO addresses supplied\n"); + goto err1; + } + par->dio_addr = dio_addr; + par->cio_addr = cio_addr; + par->c2io_addr = c2io_addr; + par->cslut[0] = 0x00; + par->cslut[1] = 0x06; + info->flags = FBINFO_FLAG_DEFAULT; + spin_lock_init(&par->lock); + retval = register_framebuffer(info); + if (retval < 0) + goto err1; + dev_set_drvdata(&dev->dev, info); + if (irq) { + par->irq = irq; + if (request_irq(par->irq, &arcfb_interrupt, SA_SHIRQ, + "arcfb", info)) { + printk(KERN_INFO + "arcfb: Failed req IRQ %d\n", par->irq); + goto err1; + } + } + printk(KERN_INFO + "fb%d: Arc frame buffer device, using %dK of video memory\n", + info->node, videomemorysize >> 10); + + /* this inits the lcd but doesn't clear dirty pixels */ + for (i = 0; i < num_cols * num_rows; i++) { + ks108_writeb_ctl(par, i, KS_DPY_OFF); + ks108_set_start_line(par, i, 0); + ks108_set_yaddr(par, i, 0); + ks108_set_xaddr(par, i, 0); + ks108_writeb_ctl(par, i, KS_DPY_ON); + } + + /* if we were told to splash the screen, we just clear it */ + if (!nosplash) { + for (i = 0; i < num_cols * num_rows; i++) { + printk(KERN_INFO "fb%d: splashing lcd %d\n", + info->node, i); + ks108_set_start_line(par, i, 0); + ks108_clear_lcd(par, i); + } + } + + return 0; +err1: + framebuffer_release(info); +err: + vfree(videomemory); + return retval; +} + +static int arcfb_remove(struct device *device) +{ + struct fb_info *info = dev_get_drvdata(device); + + if (info) { + unregister_framebuffer(info); + vfree(info->screen_base); + framebuffer_release(info); + } + return 0; +} + +static struct device_driver arcfb_driver = { + .name = "arcfb", + .bus = &platform_bus_type, + .probe = arcfb_probe, + .remove = arcfb_remove, +}; + +static struct platform_device arcfb_device = { + .name = "arcfb", + .id = 0, + .dev = { + .release = arcfb_platform_release, + } +}; + +static int __init arcfb_init(void) +{ + int ret; + + if (!arcfb_enable) + return -ENXIO; + + ret = driver_register(&arcfb_driver); + if (!ret) { + ret = platform_device_register(&arcfb_device); + if (ret) + driver_unregister(&arcfb_driver); + } + return ret; + +} + +static void __exit arcfb_exit(void) +{ + platform_device_unregister(&arcfb_device); + driver_unregister(&arcfb_driver); +} + +module_param(num_cols, ulong, 0); +MODULE_PARM_DESC(num_cols, "Num horiz panels, eg: 2 = 128 bit wide"); +module_param(num_rows, ulong, 0); +MODULE_PARM_DESC(num_rows, "Num vert panels, eg: 1 = 64 bit high"); +module_param(nosplash, uint, 0); +MODULE_PARM_DESC(nosplash, "Disable doing the splash screen"); +module_param(arcfb_enable, uint, 0); +MODULE_PARM_DESC(arcfb_enable, "Enable communication with Arc board"); +module_param(dio_addr, ulong, 0); +MODULE_PARM_DESC(dio_addr, "IO address for data, eg: 0x480"); +module_param(cio_addr, ulong, 0); +MODULE_PARM_DESC(cio_addr, "IO address for control, eg: 0x400"); +module_param(c2io_addr, ulong, 0); +MODULE_PARM_DESC(c2io_addr, "IO address for secondary control, eg: 0x408"); +module_param(splashval, ulong, 0); +MODULE_PARM_DESC(splashval, "Splash pattern: 0xFF is black, 0x00 is green"); +module_param(tuhold, ulong, 0); +MODULE_PARM_DESC(tuhold, "Time to hold between strobing data to Arc board"); +module_param(irq, uint, 0); +MODULE_PARM_DESC(irq, "IRQ for the Arc board"); + +module_init(arcfb_init); +module_exit(arcfb_exit); + +MODULE_DESCRIPTION("fbdev driver for Arc monochrome LCD board"); +MODULE_AUTHOR("Jaya Kumar"); +MODULE_LICENSE("GPL"); + diff --git a/include/linux/arcfb.h b/include/linux/arcfb.h new file mode 100644 index 000000000000..721e7654daeb --- /dev/null +++ b/include/linux/arcfb.h @@ -0,0 +1,8 @@ +#ifndef __LINUX_ARCFB_H__ +#define __LINUX_ARCFB_H__ + +#define FBIO_WAITEVENT _IO('F', 0x88) +#define FBIO_GETCONTROL2 _IOR('F', 0x89, size_t) + +#endif + -- cgit v1.2.3 From d5881eb4883ef7dd28a4dcea237714817c4b2f8e Mon Sep 17 00:00:00 2001 From: James Simmons Date: Tue, 21 Jun 2005 17:17:05 -0700 Subject: [PATCH] fbdev: new pci id for chipsfb Patch adds pci ID for CT 69000 chipset. Signed-off-by: James Simmons Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pci_ids.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index b8b4ebf9abf1..63e89e47b8e9 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -575,6 +575,7 @@ #define PCI_DEVICE_ID_CT_65550 0x00e0 #define PCI_DEVICE_ID_CT_65554 0x00e4 #define PCI_DEVICE_ID_CT_65555 0x00e5 +#define PCI_DEVICE_ID_CT_69000 0x00c0 #define PCI_VENDOR_ID_MIRO 0x1031 #define PCI_DEVICE_ID_MIRO_36050 0x5601 -- cgit v1.2.3 From 303b86d9913eca0cbfc3c5cb41e7006f6e13b755 Mon Sep 17 00:00:00 2001 From: Jurriaan Date: Tue, 21 Jun 2005 17:17:06 -0700 Subject: [PATCH] New framebuffer fonts + updated 12x22 font available Improve the fonts for use with the framebuffer. I've added all the characters marked 'FIXME' in the sun12x22 font and created a 10x18 font (based on the sun12x22 font) and a 7x14 font (based on the vga8x16 font). This patch is non-intrusive, no options are enabled by default so most users won't notice a thing. I am placing my changes under the GPL, however, I've not seen any copyright notices on the sun12x22 font and the vga8x16 font which I derived my new fonts from so I don't know what the copyright status is. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/console/Kconfig | 17 + drivers/video/console/Makefile | 2 + drivers/video/console/font_10x18.c | 5146 +++++++++++++++++++++++++++++++++ drivers/video/console/font_7x14.c | 4118 ++++++++++++++++++++++++++ drivers/video/console/font_sun12x22.c | 1579 +++++----- drivers/video/console/fonts.c | 8 + include/linux/font.h | 26 +- 7 files changed, 10068 insertions(+), 828 deletions(-) create mode 100644 drivers/video/console/font_10x18.c create mode 100644 drivers/video/console/font_7x14.c (limited to 'include/linux') diff --git a/drivers/video/console/Kconfig b/drivers/video/console/Kconfig index ccf55811d24f..e4b91a4b936c 100644 --- a/drivers/video/console/Kconfig +++ b/drivers/video/console/Kconfig @@ -153,6 +153,15 @@ config FONT_6x11 Small console font with Macintosh-style high-half glyphs. Some Mac framebuffer drivers don't support this one at all. +config FONT_7x14 + bool "console 7x14 font (not supported by all drivers)" if FONTS + depends on FRAMEBUFFER_CONSOLE + default y if !SPARC32 && !SPARC64 && !FONTS + help + Console font with characters just a bit smaller than the default. + If the standard 8x16 font is a little too big for you, say Y. + Otherwise, say N. + config FONT_PEARL_8x8 bool "Pearl (old m68k) console 8x8 font" if FONTS depends on FRAMEBUFFER_CONSOLE @@ -187,5 +196,13 @@ config FONT_SUN12x22 big letters (like the letters used in the SPARC PROM). If the standard font is unreadable for you, say Y, otherwise say N. +config FONT_10x18 + bool "console 10x18 font (not supported by all drivers)" + depends on FONTS + help + This is a high resolution console font for machines with very + big letters. It fits between the sun 12x22 and the normal 8x16 font. + If other fonts are too big or too small for you, say Y, otherwise say N. + endmenu diff --git a/drivers/video/console/Makefile b/drivers/video/console/Makefile index 33516447f9f2..b562f6bb9d31 100644 --- a/drivers/video/console/Makefile +++ b/drivers/video/console/Makefile @@ -10,6 +10,8 @@ font-objs-$(CONFIG_FONT_SUN12x22) += font_sun12x22.o font-objs-$(CONFIG_FONT_8x8) += font_8x8.o font-objs-$(CONFIG_FONT_8x16) += font_8x16.o font-objs-$(CONFIG_FONT_6x11) += font_6x11.o +font-objs-$(CONFIG_FONT_7x14) += font_7x14.o +font-objs-$(CONFIG_FONT_10x18) += font_10x18.o font-objs-$(CONFIG_FONT_PEARL_8x8) += font_pearl_8x8.o font-objs-$(CONFIG_FONT_ACORN_8x8) += font_acorn_8x8.o font-objs-$(CONFIG_FONT_MINI_4x6) += font_mini_4x6.o diff --git a/drivers/video/console/font_10x18.c b/drivers/video/console/font_10x18.c new file mode 100644 index 000000000000..ff0af96e4dfc --- /dev/null +++ b/drivers/video/console/font_10x18.c @@ -0,0 +1,5146 @@ +/******************************** + * adapted from font_sun12x22.c * + * by Jurriaan Kalkman 06-2005 * + ********************************/ + +#include + +#define FONTDATAMAX 9216 + +static unsigned char fontdata_10x18[FONTDATAMAX] = { + + /* 0 0x00 '^@' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 1 0x01 '^A' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3f, 0x80, /* 0011111110 */ + 0x40, 0x40, /* 0100000001 */ + 0x5b, 0x40, /* 0101101101 */ + 0x40, 0x40, /* 0100000001 */ + 0x44, 0x40, /* 0100010001 */ + 0x44, 0x40, /* 0100010001 */ + 0x51, 0x40, /* 0101000101 */ + 0x4e, 0x40, /* 0100111001 */ + 0x40, 0x40, /* 0100000001 */ + 0x3f, 0x80, /* 0011111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 2 0x02 '^B' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3f, 0x80, /* 0011111110 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x64, 0xc0, /* 0110010011 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x7b, 0xc0, /* 0111101111 */ + 0x7b, 0xc0, /* 0111101111 */ + 0x6e, 0xc0, /* 0110111011 */ + 0x71, 0xc0, /* 0111000111 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x3f, 0x80, /* 0011111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 3 0x03 '^C' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x11, 0x00, /* 0001000100 */ + 0x3b, 0x80, /* 0011101110 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x3f, 0x80, /* 0011111110 */ + 0x3f, 0x80, /* 0011111110 */ + 0x1f, 0x00, /* 0001111100 */ + 0x1f, 0x00, /* 0001111100 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x04, 0x00, /* 0000010000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 4 0x04 '^D' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x1f, 0x00, /* 0001111100 */ + 0x3f, 0x80, /* 0011111110 */ + 0x1f, 0x00, /* 0001111100 */ + 0x1f, 0x00, /* 0001111100 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x04, 0x00, /* 0000010000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 5 0x05 '^E' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x31, 0x80, /* 0011000110 */ + 0x7b, 0xc0, /* 0111101111 */ + 0x35, 0x80, /* 0011010110 */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 6 0x06 '^F' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x1f, 0x00, /* 0001111100 */ + 0x3f, 0x80, /* 0011111110 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x35, 0x80, /* 0011010110 */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 7 0x07 '^G' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 8 0x08 '^H' */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xf3, 0xc0, /* 1111001111 */ + 0xe1, 0xc0, /* 1110000111 */ + 0xe1, 0xc0, /* 1110000111 */ + 0xc0, 0xc0, /* 1100000011 */ + 0xc0, 0xc0, /* 1100000011 */ + 0xe1, 0xc0, /* 1110000111 */ + 0xe1, 0xc0, /* 1110000111 */ + 0xf3, 0xc0, /* 1111001111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + + /* 9 0x09 '^I' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x12, 0x00, /* 0001001000 */ + 0x12, 0x00, /* 0001001000 */ + 0x21, 0x00, /* 0010000100 */ + 0x21, 0x00, /* 0010000100 */ + 0x12, 0x00, /* 0001001000 */ + 0x12, 0x00, /* 0001001000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 10 0x0a '^J' */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xf3, 0xc0, /* 1111001111 */ + 0xed, 0xc0, /* 1110110111 */ + 0xed, 0xc0, /* 1110110111 */ + 0xde, 0xc0, /* 1101111011 */ + 0xde, 0xc0, /* 1101111011 */ + 0xed, 0xc0, /* 1110110111 */ + 0xed, 0xc0, /* 1110110111 */ + 0xf3, 0xc0, /* 1111001111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + + /* 11 0x0b '^K' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x03, 0xc0, /* 0000001111 */ + 0x06, 0xc0, /* 0000011011 */ + 0x0c, 0xc0, /* 0000110011 */ + 0x3c, 0x00, /* 0011110000 */ + 0x66, 0x00, /* 0110011000 */ + 0xc3, 0x00, /* 1100001100 */ + 0xc3, 0x00, /* 1100001100 */ + 0xc3, 0x00, /* 1100001100 */ + 0x66, 0x00, /* 0110011000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 12 0x0c '^L' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x33, 0x00, /* 0011001100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 13 0x0d '^M' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0f, 0x80, /* 0000111110 */ + 0x08, 0x80, /* 0000100010 */ + 0x0f, 0x80, /* 0000111110 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x38, 0x00, /* 0011100000 */ + 0x78, 0x00, /* 0111100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 14 0x0e '^N' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x80, /* 0001111110 */ + 0x10, 0x80, /* 0001000010 */ + 0x1f, 0x80, /* 0001111110 */ + 0x10, 0x80, /* 0001000010 */ + 0x10, 0x80, /* 0001000010 */ + 0x10, 0x80, /* 0001000010 */ + 0x10, 0x80, /* 0001000010 */ + 0x13, 0x80, /* 0001001110 */ + 0x17, 0x80, /* 0001011110 */ + 0x73, 0x00, /* 0111001100 */ + 0xf0, 0x00, /* 1111000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 15 0x0f '^O' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x04, 0x00, /* 0000010000 */ + 0x24, 0x80, /* 0010010010 */ + 0x15, 0x00, /* 0001010100 */ + 0x55, 0x40, /* 0101010101 */ + 0x3f, 0x80, /* 0011111110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x3f, 0x80, /* 0011111110 */ + 0x55, 0x40, /* 0101010101 */ + 0x15, 0x00, /* 0001010100 */ + 0x24, 0x80, /* 0010010010 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 16 0x10 '^P' */ + 0x00, 0x80, /* 0000000010 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x80, /* 0000001110 */ + 0x07, 0x80, /* 0000011110 */ + 0x0f, 0x80, /* 0000111110 */ + 0x1f, 0x80, /* 0001111110 */ + 0x3f, 0x80, /* 0011111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0xff, 0x80, /* 1111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x3f, 0x80, /* 0011111110 */ + 0x1f, 0x80, /* 0001111110 */ + 0x0f, 0x80, /* 0000111110 */ + 0x07, 0x80, /* 0000011110 */ + 0x03, 0x80, /* 0000001110 */ + 0x01, 0x80, /* 0000000110 */ + 0x00, 0x80, /* 0000000010 */ + 0x00, 0x00, /* 0000000000 */ + + /* 17 0x11 '^Q' */ + 0x40, 0x00, /* 0100000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x70, 0x00, /* 0111000000 */ + 0x78, 0x00, /* 0111100000 */ + 0x7c, 0x00, /* 0111110000 */ + 0x7e, 0x00, /* 0111111000 */ + 0x7f, 0x00, /* 0111111100 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x00, /* 0111111100 */ + 0x7e, 0x00, /* 0111111000 */ + 0x7c, 0x00, /* 0111110000 */ + 0x78, 0x00, /* 0111100000 */ + 0x70, 0x00, /* 0111000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x40, 0x00, /* 0100000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 18 0x12 '^R' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x7f, 0x80, /* 0111111110 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x3f, 0x00, /* 0011111100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 19 0x13 '^S' */ + 0x00, 0x00, /* 0000000000 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 20 0x14 '^T' */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x39, 0x80, /* 0011100110 */ + 0x79, 0x80, /* 0111100110 */ + 0x79, 0x80, /* 0111100110 */ + 0x79, 0x80, /* 0111100110 */ + 0x39, 0x80, /* 0011100110 */ + 0x19, 0x80, /* 0001100110 */ + 0x19, 0x80, /* 0001100110 */ + 0x19, 0x80, /* 0001100110 */ + 0x19, 0x80, /* 0001100110 */ + 0x19, 0x80, /* 0001100110 */ + 0x19, 0x80, /* 0001100110 */ + 0x39, 0xc0, /* 0011100111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 21 0x15 '^U' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x3e, 0x00, /* 0011111000 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x3e, 0x00, /* 0011111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 22 0x16 '^V' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 23 0x17 '^W' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x7f, 0x80, /* 0111111110 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x3f, 0x00, /* 0011111100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 24 0x18 '^X' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x7f, 0x80, /* 0111111110 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 25 0x19 '^Y' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x3f, 0x00, /* 0011111100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 26 0x1a '^Z' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x04, 0x00, /* 0000010000 */ + 0x06, 0x00, /* 0000011000 */ + 0x07, 0x00, /* 0000011100 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x07, 0x00, /* 0000011100 */ + 0x06, 0x00, /* 0000011000 */ + 0x04, 0x00, /* 0000010000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 27 0x1b '^[' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x08, 0x00, /* 0000100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x38, 0x00, /* 0011100000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x38, 0x00, /* 0011100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 28 0x1c '^\' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 29 0x1d '^]' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x12, 0x00, /* 0001001000 */ + 0x33, 0x00, /* 0011001100 */ + 0x73, 0x80, /* 0111001110 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x73, 0x80, /* 0111001110 */ + 0x33, 0x00, /* 0011001100 */ + 0x12, 0x00, /* 0001001000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 30 0x1e '^^' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x04, 0x00, /* 0000010000 */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x1f, 0x00, /* 0001111100 */ + 0x3f, 0x80, /* 0011111110 */ + 0x3f, 0x80, /* 0011111110 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 31 0x1f '^_' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x3f, 0x80, /* 0011111110 */ + 0x3f, 0x80, /* 0011111110 */ + 0x1f, 0x00, /* 0001111100 */ + 0x1f, 0x00, /* 0001111100 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x04, 0x00, /* 0000010000 */ + 0x04, 0x00, /* 0000010000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 32 0x20 ' ' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 33 0x21 '!' */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 34 0x22 '"' */ + 0x00, 0x00, /* 0000000000 */ + 0x63, 0x00, /* 0110001100 */ + 0xf7, 0x80, /* 1111011110 */ + 0xf7, 0x80, /* 1111011110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x63, 0x00, /* 0110001100 */ + 0x42, 0x00, /* 0100001000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 35 0x23 '#' */ + 0x00, 0x00, /* 0000000000 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 36 0x24 '$' */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x6f, 0x80, /* 0110111110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6c, 0x80, /* 0110110010 */ + 0x3c, 0x00, /* 0011110000 */ + 0x0f, 0x00, /* 0000111100 */ + 0x0d, 0x80, /* 0000110110 */ + 0x4d, 0x80, /* 0100110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x7f, 0x00, /* 0111111100 */ + 0x3e, 0x00, /* 0011111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 37 0x25 '%' */ + 0x00, 0x00, /* 0000000000 */ + 0x31, 0x80, /* 0011000110 */ + 0x7b, 0x00, /* 0111101100 */ + 0x7b, 0x00, /* 0111101100 */ + 0x36, 0x00, /* 0011011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x1b, 0x00, /* 0001101100 */ + 0x37, 0x80, /* 0011011110 */ + 0x37, 0x80, /* 0011011110 */ + 0x63, 0x00, /* 0110001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 38 0x26 '&' */ + 0x00, 0x00, /* 0000000000 */ + 0x07, 0x00, /* 0000011100 */ + 0x0f, 0x80, /* 0000111110 */ + 0x19, 0x80, /* 0001100110 */ + 0x19, 0x80, /* 0001100110 */ + 0x0f, 0x80, /* 0000111110 */ + 0x1e, 0x00, /* 0001111000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x76, 0x00, /* 0111011000 */ + 0x66, 0x40, /* 0110011001 */ + 0x63, 0xc0, /* 0110001111 */ + 0x63, 0x80, /* 0110001110 */ + 0x63, 0x00, /* 0110001100 */ + 0x3f, 0x80, /* 0011111110 */ + 0x1c, 0xc0, /* 0001110011 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 39 0x27 ''' */ + 0x00, 0x00, /* 0000000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x78, 0x00, /* 0111100000 */ + 0x78, 0x00, /* 0111100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x20, 0x00, /* 0010000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 40 0x28 '(' */ + 0x00, 0x00, /* 0000000000 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x06, 0x00, /* 0000011000 */ + 0x03, 0x00, /* 0000001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 41 0x29 ')' */ + 0x00, 0x00, /* 0000000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 42 0x2a '*' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x4c, 0x80, /* 0100110010 */ + 0x6d, 0x80, /* 0110110110 */ + 0x3f, 0x00, /* 0011111100 */ + 0x7f, 0x80, /* 0111111110 */ + 0x3f, 0x00, /* 0011111100 */ + 0x6d, 0x80, /* 0110110110 */ + 0x4c, 0x80, /* 0100110010 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 43 0x2b '+' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 44 0x2c ',' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x78, 0x00, /* 0111100000 */ + 0x78, 0x00, /* 0111100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x40, 0x00, /* 0100000000 */ + + /* 45 0x2d '-' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 46 0x2e '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 47 0x2f '/' */ + 0x00, 0x00, /* 0000000000 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 48 0x30 '0' */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x23, 0x00, /* 0010001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x63, 0x80, /* 0110001110 */ + 0x65, 0x80, /* 0110010110 */ + 0x65, 0x80, /* 0110010110 */ + 0x69, 0x80, /* 0110100110 */ + 0x69, 0x80, /* 0110100110 */ + 0x71, 0x80, /* 0111000110 */ + 0x61, 0x00, /* 0110000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x3e, 0x00, /* 0011111000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 49 0x31 '1' */ + 0x00, 0x00, /* 0000000000 */ + 0x04, 0x00, /* 0000010000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 50 0x32 '2' */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x63, 0x80, /* 0110001110 */ + 0x41, 0x80, /* 0100000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x80, /* 0011000010 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 51 0x33 '3' */ + 0x00, 0x00, /* 0000000000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x47, 0x00, /* 0100011100 */ + 0x03, 0x00, /* 0000001100 */ + 0x07, 0x00, /* 0000011100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x07, 0x00, /* 0000011100 */ + 0x03, 0x00, /* 0000001100 */ + 0x01, 0x80, /* 0000000110 */ + 0x41, 0x80, /* 0100000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x3f, 0x00, /* 0011111100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 52 0x34 '4' */ + 0x00, 0x00, /* 0000000000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x36, 0x00, /* 0011011000 */ + 0x36, 0x00, /* 0011011000 */ + 0x66, 0x00, /* 0110011000 */ + 0x66, 0x00, /* 0110011000 */ + 0xc6, 0x00, /* 1100011000 */ + 0xc6, 0x00, /* 1100011000 */ + 0xff, 0x80, /* 1111111110 */ + 0xff, 0x80, /* 1111111110 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 53 0x35 '5' */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x1f, 0x00, /* 0001111100 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x7e, 0x00, /* 0111111000 */ + 0x67, 0x00, /* 0110011100 */ + 0x03, 0x80, /* 0000001110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x41, 0x80, /* 0100000110 */ + 0x63, 0x00, /* 0110001100 */ + 0x3e, 0x00, /* 0011111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 54 0x36 '6' */ + 0x00, 0x00, /* 0000000000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x6e, 0x00, /* 0110111000 */ + 0x7f, 0x00, /* 0111111100 */ + 0x73, 0x80, /* 0111001110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x71, 0x00, /* 0111000100 */ + 0x3e, 0x00, /* 0011111000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 55 0x37 '7' */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x80, /* 0001111110 */ + 0x3f, 0x80, /* 0011111110 */ + 0x61, 0x80, /* 0110000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 56 0x38 '8' */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x23, 0x00, /* 0010001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x31, 0x00, /* 0011000100 */ + 0x1a, 0x00, /* 0001101000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x16, 0x00, /* 0001011000 */ + 0x23, 0x00, /* 0010001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x31, 0x00, /* 0011000100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 57 0x39 '9' */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x17, 0x00, /* 0001011100 */ + 0x23, 0x80, /* 0010001110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x3d, 0x80, /* 0011110110 */ + 0x19, 0x80, /* 0001100110 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 58 0x3a ':' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 59 0x3b ';' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x20, 0x00, /* 0010000000 */ + + /* 60 0x3c '<' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x06, 0x00, /* 0000011000 */ + 0x03, 0x00, /* 0000001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 61 0x3d '=' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 62 0x3e '>' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x06, 0x00, /* 0000011000 */ + 0x03, 0x00, /* 0000001100 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 63 0x3f '?' */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x3b, 0x80, /* 0011101110 */ + 0x21, 0x80, /* 0010000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 64 0x40 '@' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x65, 0x80, /* 0110010110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6f, 0x80, /* 0110111110 */ + 0x60, 0x00, /* 0110000000 */ + 0x31, 0x80, /* 0011000110 */ + 0x3f, 0x80, /* 0011111110 */ + 0x0f, 0x00, /* 0000111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 65 0x41 'A' */ + 0x00, 0x00, /* 0000000000 */ + 0x04, 0x00, /* 0000010000 */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x19, 0x80, /* 0001100110 */ + 0x31, 0x80, /* 0011000110 */ + 0x3f, 0x80, /* 0011111110 */ + 0x31, 0x80, /* 0011000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x60, 0xc0, /* 0110000011 */ + 0x60, 0xc0, /* 0110000011 */ + 0xf1, 0xc0, /* 1111000111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 66 0x42 'B' */ + 0x00, 0x00, /* 0000000000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x62, 0x00, /* 0110001000 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x66, 0x00, /* 0110011000 */ + 0x7e, 0x00, /* 0111111000 */ + 0x63, 0x00, /* 0110001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x63, 0x00, /* 0110001100 */ + 0xfe, 0x00, /* 1111111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 67 0x43 'C' */ + 0x00, 0x00, /* 0000000000 */ + 0x0f, 0x00, /* 0000111100 */ + 0x11, 0x80, /* 0001000110 */ + 0x20, 0x80, /* 0010000010 */ + 0x20, 0x00, /* 0010000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x20, 0x00, /* 0010000000 */ + 0x30, 0x80, /* 0011000010 */ + 0x19, 0x00, /* 0001100100 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 68 0x44 'D' */ + 0x00, 0x00, /* 0000000000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x67, 0x00, /* 0110011100 */ + 0x63, 0x00, /* 0110001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x00, /* 0110000100 */ + 0x66, 0x00, /* 0110011000 */ + 0xf8, 0x00, /* 1111100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 69 0x45 'E' */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x30, 0x80, /* 0011000010 */ + 0x30, 0x80, /* 0011000010 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x31, 0x00, /* 0011000100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x31, 0x00, /* 0011000100 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x80, /* 0011000010 */ + 0x30, 0x80, /* 0011000010 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 70 0x46 'F' */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x30, 0x80, /* 0011000010 */ + 0x30, 0x80, /* 0011000010 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x31, 0x00, /* 0011000100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x31, 0x00, /* 0011000100 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x78, 0x00, /* 0111100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 71 0x47 'G' */ + 0x00, 0x00, /* 0000000000 */ + 0x0f, 0x00, /* 0000111100 */ + 0x11, 0x80, /* 0001000110 */ + 0x20, 0x80, /* 0010000010 */ + 0x20, 0x00, /* 0010000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x67, 0xc0, /* 0110011111 */ + 0x61, 0x80, /* 0110000110 */ + 0x21, 0x80, /* 0010000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x19, 0x80, /* 0001100110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 72 0x48 'H' */ + 0x00, 0x00, /* 0000000000 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 73 0x49 'I' */ + 0x00, 0x00, /* 0000000000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 74 0x4a 'J' */ + 0x00, 0x00, /* 0000000000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x08, 0x00, /* 0000100000 */ + 0x70, 0x00, /* 0111000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 75 0x4b 'K' */ + 0x00, 0x00, /* 0000000000 */ + 0xf1, 0x80, /* 1111000110 */ + 0x63, 0x00, /* 0110001100 */ + 0x66, 0x00, /* 0110011000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x78, 0x00, /* 0111100000 */ + 0x70, 0x00, /* 0111000000 */ + 0x70, 0x00, /* 0111000000 */ + 0x78, 0x00, /* 0111100000 */ + 0x78, 0x00, /* 0111100000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x66, 0x00, /* 0110011000 */ + 0x63, 0x00, /* 0110001100 */ + 0x61, 0x80, /* 0110000110 */ + 0xf0, 0xc0, /* 1111000011 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 76 0x4c 'L' */ + 0x00, 0x00, /* 0000000000 */ + 0x78, 0x00, /* 0111100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x80, /* 0011000010 */ + 0x30, 0x80, /* 0011000010 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 77 0x4d 'M' */ + 0x00, 0x00, /* 0000000000 */ + 0xe0, 0xc0, /* 1110000011 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x73, 0x80, /* 0111001110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 78 0x4e 'N' */ + 0x00, 0x00, /* 0000000000 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x71, 0x80, /* 0111000110 */ + 0x79, 0x80, /* 0111100110 */ + 0x79, 0x80, /* 0111100110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x67, 0x80, /* 0110011110 */ + 0x67, 0x80, /* 0110011110 */ + 0x63, 0x80, /* 0110001110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 79 0x4f 'O' */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x17, 0x00, /* 0001011100 */ + 0x23, 0x00, /* 0010001100 */ + 0x21, 0x80, /* 0010000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x21, 0x00, /* 0010000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x1a, 0x00, /* 0001101000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 80 0x50 'P' */ + 0x00, 0x00, /* 0000000000 */ + 0xfe, 0x00, /* 1111111000 */ + 0x63, 0x00, /* 0110001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x63, 0x00, /* 0110001100 */ + 0x7e, 0x00, /* 0111111000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0xf0, 0x00, /* 1111000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 81 0x51 'Q' */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x13, 0x00, /* 0001001100 */ + 0x23, 0x00, /* 0010001100 */ + 0x21, 0x80, /* 0010000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x3b, 0x00, /* 0011101100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x26, 0x00, /* 0010011000 */ + 0x03, 0x80, /* 0000001110 */ + 0x00, 0x00, /* 0000000000 */ + + /* 82 0x52 'R' */ + 0x00, 0x00, /* 0000000000 */ + 0xfe, 0x00, /* 1111111000 */ + 0x63, 0x00, /* 0110001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x00, /* 0110000100 */ + 0x7e, 0x00, /* 0111111000 */ + 0x78, 0x00, /* 0111100000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x6e, 0x00, /* 0110111000 */ + 0x67, 0x00, /* 0110011100 */ + 0x63, 0x80, /* 0110001110 */ + 0x61, 0xc0, /* 0110000111 */ + 0xf0, 0xc0, /* 1111000011 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 83 0x53 'S' */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x06, 0x00, /* 0000011000 */ + 0x03, 0x00, /* 0000001100 */ + 0x01, 0x80, /* 0000000110 */ + 0x41, 0x80, /* 0100000110 */ + 0x63, 0x00, /* 0110001100 */ + 0x3e, 0x00, /* 0011111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 84 0x54 'T' */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x4c, 0x80, /* 0100110010 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 85 0x55 'U' */ + 0x00, 0x00, /* 0000000000 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x00, /* 0111001100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 86 0x56 'V' */ + 0x00, 0x00, /* 0000000000 */ + 0xe1, 0xc0, /* 1110000111 */ + 0xc0, 0xc0, /* 1100000011 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x12, 0x00, /* 0001001000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 87 0x57 'W' */ + 0x00, 0x00, /* 0000000000 */ + 0xe1, 0xc0, /* 1110000111 */ + 0xc0, 0xc0, /* 1100000011 */ + 0xc0, 0xc0, /* 1100000011 */ + 0xc0, 0xc0, /* 1100000011 */ + 0xe0, 0xc0, /* 1110000011 */ + 0x61, 0x80, /* 0110000110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x77, 0x00, /* 0111011100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 88 0x58 'X' */ + 0x00, 0x00, /* 0000000000 */ + 0xf7, 0x80, /* 1111011110 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x36, 0x00, /* 0011011000 */ + 0x36, 0x00, /* 0011011000 */ + 0x36, 0x00, /* 0011011000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x36, 0x00, /* 0011011000 */ + 0x36, 0x00, /* 0011011000 */ + 0x36, 0x00, /* 0011011000 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0xf7, 0x80, /* 1111011110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 89 0x59 'Y' */ + 0x00, 0x00, /* 0000000000 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 90 0x5a 'Z' */ + 0x00, 0x00, /* 0000000000 */ + 0x3f, 0x80, /* 0011111110 */ + 0x21, 0x80, /* 0010000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x80, /* 0011000010 */ + 0x3f, 0x80, /* 0011111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 91 0x5b '[' */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x1f, 0x00, /* 0001111100 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x1f, 0x00, /* 0001111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 92 0x5c '\' */ + 0x00, 0x00, /* 0000000000 */ + 0xc0, 0x00, /* 1100000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x00, 0xc0, /* 0000000011 */ + 0x00, 0x00, /* 0000000000 */ + + /* 93 0x5d ']' */ + 0x00, 0x00, /* 0000000000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 94 0x5e '^' */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 95 0x5f '_' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + + /* 96 0x60 '`' */ + 0x04, 0x00, /* 0000010000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 97 0x61 'a' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x21, 0x80, /* 0010000110 */ + 0x07, 0x80, /* 0000011110 */ + 0x39, 0x80, /* 0011100110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x3d, 0xc0, /* 0011110111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 98 0x62 'b' */ + 0x20, 0x00, /* 0010000000 */ + 0x60, 0x00, /* 0110000000 */ + 0xe0, 0x00, /* 1110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x66, 0x00, /* 0110011000 */ + 0x6f, 0x00, /* 0110111100 */ + 0x73, 0x80, /* 0111001110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x71, 0x80, /* 0111000110 */ + 0x7b, 0x00, /* 0111101100 */ + 0x4e, 0x00, /* 0100111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 99 0x63 'c' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x37, 0x00, /* 0011011100 */ + 0x23, 0x00, /* 0010001100 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x71, 0x00, /* 0111000100 */ + 0x33, 0x00, /* 0011001100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 100 0x64 'd' */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x80, /* 0000001110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x0d, 0x80, /* 0000110110 */ + 0x37, 0x80, /* 0011011110 */ + 0x23, 0x80, /* 0010001110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x35, 0x80, /* 0011010110 */ + 0x19, 0xc0, /* 0001100111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 101 0x65 'e' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x19, 0x80, /* 0001100110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 102 0x66 'f' */ + 0x07, 0x00, /* 0000011100 */ + 0x09, 0x80, /* 0000100110 */ + 0x09, 0x80, /* 0000100110 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x7f, 0x00, /* 0111111100 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 103 0x67 'g' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1c, 0x80, /* 0001110010 */ + 0x37, 0x80, /* 0011011110 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x36, 0x00, /* 0011011000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x60, 0x00, /* 0110000000 */ + 0x7f, 0x00, /* 0111111100 */ + 0x3f, 0x80, /* 0011111110 */ + 0x21, 0x80, /* 0010000110 */ + 0x40, 0x80, /* 0100000010 */ + 0x7f, 0x00, /* 0111111100 */ + 0x3e, 0x00, /* 0011111000 */ + + /* 104 0x68 'h' */ + 0x10, 0x00, /* 0001000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x70, 0x00, /* 0111000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x37, 0x00, /* 0011011100 */ + 0x3b, 0x80, /* 0011101110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x7b, 0xc0, /* 0111101111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 105 0x69 'i' */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 106 0x6a 'j' */ + 0x00, 0x00, /* 0000000000 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x07, 0x80, /* 0000011110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x41, 0x80, /* 0100000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x71, 0x80, /* 0111000110 */ + 0x3f, 0x00, /* 0011111100 */ + 0x1c, 0x00, /* 0001110000 */ + + /* 107 0x6b 'k' */ + 0x60, 0x00, /* 0110000000 */ + 0xe0, 0x00, /* 1110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x63, 0x80, /* 0110001110 */ + 0x66, 0x00, /* 0110011000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x78, 0x00, /* 0111100000 */ + 0x70, 0x00, /* 0111000000 */ + 0x78, 0x00, /* 0111100000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x6e, 0x00, /* 0110111000 */ + 0x67, 0x00, /* 0110011100 */ + 0xf3, 0x80, /* 1111001110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 108 0x6c 'l' */ + 0x3c, 0x00, /* 0011110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 109 0x6d 'm' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xdb, 0x80, /* 1101101110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0xed, 0xc0, /* 1110110111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 110 0x6e 'n' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x6f, 0x00, /* 0110111100 */ + 0x7b, 0x80, /* 0111101110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x7b, 0xc0, /* 0111101111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 111 0x6f 'o' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x66, 0x00, /* 0110011000 */ + 0xc3, 0x00, /* 1100001100 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xe1, 0x80, /* 1110000110 */ + 0x73, 0x00, /* 0111001100 */ + 0x3c, 0x00, /* 0011110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 112 0x70 'p' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xde, 0x00, /* 1101111000 */ + 0x76, 0x00, /* 0111011000 */ + 0x63, 0x00, /* 0110001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x71, 0x80, /* 0111000110 */ + 0x7b, 0x00, /* 0111101100 */ + 0x7e, 0x00, /* 0111111000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0xf0, 0x00, /* 1111000000 */ + + /* 113 0x71 'q' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0xc0, /* 0000111011 */ + 0x1b, 0x80, /* 0001101110 */ + 0x33, 0x80, /* 0011001110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x71, 0x80, /* 0111000110 */ + 0x3b, 0x80, /* 0011101110 */ + 0x1f, 0x80, /* 0001111110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0xc0, /* 0000001111 */ + + /* 114 0x72 'r' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x73, 0x00, /* 0111001100 */ + 0x35, 0x80, /* 0011010110 */ + 0x39, 0x80, /* 0011100110 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x78, 0x00, /* 0111100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 115 0x73 's' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x63, 0x00, /* 0110001100 */ + 0x61, 0x00, /* 0110000100 */ + 0x70, 0x00, /* 0111000000 */ + 0x38, 0x00, /* 0011100000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x07, 0x00, /* 0000011100 */ + 0x43, 0x00, /* 0100001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x7e, 0x00, /* 0111111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 116 0x74 't' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x1c, 0x80, /* 0001110010 */ + 0x0f, 0x00, /* 0000111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 117 0x75 'u' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xf7, 0x80, /* 1111011110 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x77, 0x00, /* 0111011100 */ + 0x3d, 0x80, /* 0011110110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 118 0x76 'v' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xf1, 0xc0, /* 1111000111 */ + 0x60, 0xc0, /* 0110000011 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x19, 0x80, /* 0001100110 */ + 0x1b, 0x00, /* 0001101100 */ + 0x0f, 0x00, /* 0000111100 */ + 0x0f, 0x00, /* 0000111100 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 119 0x77 'w' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xe3, 0xc0, /* 1110001111 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0x6b, 0x00, /* 0110101100 */ + 0x6b, 0x00, /* 0110101100 */ + 0x7e, 0x00, /* 0111111000 */ + 0x36, 0x00, /* 0011011000 */ + 0x36, 0x00, /* 0011011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 120 0x78 'x' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xf7, 0x80, /* 1111011110 */ + 0x63, 0x00, /* 0110001100 */ + 0x36, 0x00, /* 0011011000 */ + 0x36, 0x00, /* 0011011000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x36, 0x00, /* 0011011000 */ + 0x66, 0x00, /* 0110011000 */ + 0x63, 0x00, /* 0110001100 */ + 0xf7, 0x80, /* 1111011110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 121 0x79 'y' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x61, 0x80, /* 0110000110 */ + 0x33, 0x00, /* 0011001100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x78, 0x00, /* 0111100000 */ + 0x70, 0x00, /* 0111000000 */ + + /* 122 0x7a 'z' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x61, 0x80, /* 0110000110 */ + 0x43, 0x00, /* 0100001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x60, 0x80, /* 0110000010 */ + 0x61, 0x80, /* 0110000110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 123 0x7b '{' */ + 0x07, 0x00, /* 0000011100 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x70, 0x00, /* 0111000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x07, 0x00, /* 0000011100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 124 0x7c '|' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 125 0x7d '}' */ + 0x38, 0x00, /* 0011100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x06, 0x00, /* 0000011000 */ + 0x03, 0x80, /* 0000001110 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x38, 0x00, /* 0011100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 126 0x7e '~' */ + 0x00, 0x00, /* 0000000000 */ + 0x18, 0x80, /* 0001100010 */ + 0x3d, 0x80, /* 0011110110 */ + 0x6f, 0x00, /* 0110111100 */ + 0x46, 0x00, /* 0100011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 127 0x7f '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x12, 0x00, /* 0001001000 */ + 0x21, 0x00, /* 0010000100 */ + 0x40, 0x80, /* 0100000010 */ + 0x40, 0x80, /* 0100000010 */ + 0x40, 0x80, /* 0100000010 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 128 0x80 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x21, 0x80, /* 0010000110 */ + 0x40, 0x80, /* 0100000010 */ + 0x40, 0x00, /* 0100000000 */ + 0x40, 0x00, /* 0100000000 */ + 0x40, 0x00, /* 0100000000 */ + 0x40, 0x00, /* 0100000000 */ + 0x40, 0x00, /* 0100000000 */ + 0x40, 0x00, /* 0100000000 */ + 0x60, 0x80, /* 0110000010 */ + 0x31, 0x00, /* 0011000100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x08, 0x00, /* 0000100000 */ + 0x04, 0x00, /* 0000010000 */ + 0x02, 0x00, /* 0000001000 */ + 0x02, 0x00, /* 0000001000 */ + 0x1c, 0x00, /* 0001110000 */ + + /* 129 0x81 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7b, 0x80, /* 0111101110 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x3b, 0x00, /* 0011101100 */ + 0x1c, 0x80, /* 0001110010 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 130 0x82 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x01, 0x00, /* 0000000100 */ + 0x02, 0x00, /* 0000001000 */ + 0x04, 0x00, /* 0000010000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x19, 0x80, /* 0001100110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 131 0x83 '.' */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1b, 0x00, /* 0001101100 */ + 0x31, 0x80, /* 0011000110 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x21, 0x80, /* 0010000110 */ + 0x07, 0x80, /* 0000011110 */ + 0x39, 0x80, /* 0011100110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x3d, 0xc0, /* 0011110111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 132 0x84 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x21, 0x80, /* 0010000110 */ + 0x07, 0x80, /* 0000011110 */ + 0x39, 0x80, /* 0011100110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x3d, 0xc0, /* 0011110111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 133 0x85 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x06, 0x00, /* 0000011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x21, 0x80, /* 0010000110 */ + 0x07, 0x80, /* 0000011110 */ + 0x39, 0x80, /* 0011100110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x3d, 0xc0, /* 0011110111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 134 0x86 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1b, 0x00, /* 0001101100 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x21, 0x80, /* 0010000110 */ + 0x07, 0x80, /* 0000011110 */ + 0x39, 0x80, /* 0011100110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x3d, 0xc0, /* 0011110111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 135 0x87 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x20, 0x80, /* 0010000010 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x70, 0x80, /* 0111000010 */ + 0x30, 0x80, /* 0011000010 */ + 0x1f, 0x00, /* 0001111100 */ + 0x04, 0x00, /* 0000010000 */ + 0x02, 0x00, /* 0000001000 */ + 0x01, 0x00, /* 0000000100 */ + 0x0e, 0x00, /* 0000111000 */ + + /* 136 0x88 '.' */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1b, 0x00, /* 0001101100 */ + 0x31, 0x80, /* 0011000110 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x19, 0x80, /* 0001100110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 137 0x89 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x36, 0x00, /* 0011011000 */ + 0x36, 0x00, /* 0011011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x19, 0x80, /* 0001100110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 138 0x8a '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x19, 0x80, /* 0001100110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 139 0x8b '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x36, 0x00, /* 0011011000 */ + 0x36, 0x00, /* 0011011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 140 0x8c '.' */ + 0x08, 0x00, /* 0000100000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x36, 0x00, /* 0011011000 */ + 0x63, 0x00, /* 0110001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 141 0x8d '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 142 0x8e '.' */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x04, 0x00, /* 0000010000 */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x19, 0x00, /* 0001100100 */ + 0x19, 0x00, /* 0001100100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x80, /* 0110000010 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 143 0x8f '.' */ + 0x04, 0x00, /* 0000010000 */ + 0x0a, 0x00, /* 0000101000 */ + 0x0a, 0x00, /* 0000101000 */ + 0x04, 0x00, /* 0000010000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x19, 0x00, /* 0001100100 */ + 0x19, 0x00, /* 0001100100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x80, /* 0110000010 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 144 0x90 '.' */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x30, 0x80, /* 0011000010 */ + 0x30, 0x80, /* 0011000010 */ + 0x30, 0x00, /* 0011000000 */ + 0x31, 0x00, /* 0011000100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x31, 0x00, /* 0011000100 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x80, /* 0011000010 */ + 0x30, 0x80, /* 0011000010 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 145 0x91 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3b, 0x80, /* 0011101110 */ + 0x6c, 0xc0, /* 0110110011 */ + 0x4c, 0xc0, /* 0100110011 */ + 0x0c, 0xc0, /* 0000110011 */ + 0x3f, 0xc0, /* 0011111111 */ + 0x6c, 0x00, /* 0110110000 */ + 0xcc, 0x00, /* 1100110000 */ + 0xcc, 0x00, /* 1100110000 */ + 0xee, 0xc0, /* 1110111011 */ + 0x7b, 0x80, /* 0111101110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 146 0x92 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x07, 0xc0, /* 0000011111 */ + 0x0e, 0x40, /* 0000111001 */ + 0x0e, 0x40, /* 0000111001 */ + 0x0e, 0x00, /* 0000111000 */ + 0x16, 0x00, /* 0001011000 */ + 0x16, 0x80, /* 0001011010 */ + 0x17, 0x80, /* 0001011110 */ + 0x16, 0x80, /* 0001011010 */ + 0x3e, 0x00, /* 0011111000 */ + 0x26, 0x00, /* 0010011000 */ + 0x26, 0x00, /* 0010011000 */ + 0x46, 0x40, /* 0100011001 */ + 0x46, 0x40, /* 0100011001 */ + 0xef, 0xc0, /* 1110111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 147 0x93 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x08, 0x00, /* 0000100000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x36, 0x00, /* 0011011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x66, 0x00, /* 0110011000 */ + 0xc3, 0x00, /* 1100001100 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xe1, 0x80, /* 1110000110 */ + 0x73, 0x00, /* 0111001100 */ + 0x3c, 0x00, /* 0011110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 148 0x94 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x66, 0x00, /* 0110011000 */ + 0xc3, 0x00, /* 1100001100 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xe1, 0x80, /* 1110000110 */ + 0x73, 0x00, /* 0111001100 */ + 0x3c, 0x00, /* 0011110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 149 0x95 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x06, 0x00, /* 0000011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x66, 0x00, /* 0110011000 */ + 0xc3, 0x00, /* 1100001100 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xe1, 0x80, /* 1110000110 */ + 0x73, 0x00, /* 0111001100 */ + 0x3c, 0x00, /* 0011110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 150 0x96 '.' */ + 0x08, 0x00, /* 0000100000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x36, 0x00, /* 0011011000 */ + 0x63, 0x00, /* 0110001100 */ + 0x00, 0x00, /* 0000000000 */ + 0xf7, 0x80, /* 1111011110 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x77, 0x00, /* 0111011100 */ + 0x3d, 0x80, /* 0011110110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 151 0x97 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x00, 0x00, /* 0000000000 */ + 0xf7, 0x80, /* 1111011110 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x77, 0x00, /* 0111011100 */ + 0x3d, 0x80, /* 0011110110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 152 0x98 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x61, 0x80, /* 0110000110 */ + 0x33, 0x00, /* 0011001100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x78, 0x00, /* 0111100000 */ + 0x70, 0x00, /* 0111000000 */ + + /* 153 0x99 '.' */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x0c, 0x00, /* 0000110000 */ + 0x17, 0x00, /* 0001011100 */ + 0x23, 0x00, /* 0010001100 */ + 0x21, 0x80, /* 0010000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x21, 0x00, /* 0010000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x1a, 0x00, /* 0001101000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 154 0x9a '.' */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x00, 0x00, /* 0000000000 */ + 0xf1, 0xc0, /* 1111000111 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x80, /* 0110000010 */ + 0x60, 0x80, /* 0110000010 */ + 0x71, 0x00, /* 0111000100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 155 0x9b '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x1f, 0x80, /* 0001111110 */ + 0x36, 0x80, /* 0011011010 */ + 0x26, 0x00, /* 0010011000 */ + 0x66, 0x00, /* 0110011000 */ + 0x66, 0x00, /* 0110011000 */ + 0x66, 0x00, /* 0110011000 */ + 0x66, 0x00, /* 0110011000 */ + 0x76, 0x00, /* 0111011000 */ + 0x36, 0x80, /* 0011011010 */ + 0x1f, 0x80, /* 0001111110 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 156 0x9c '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x3b, 0x00, /* 0011101100 */ + 0x33, 0x00, /* 0011001100 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x7e, 0x00, /* 0111111000 */ + 0x7e, 0x00, /* 0111111000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x7c, 0x80, /* 0111110010 */ + 0x7f, 0x80, /* 0111111110 */ + 0x43, 0x00, /* 0100001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 157 0x9d '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x40, 0x80, /* 0100000010 */ + 0x40, 0x80, /* 0100000010 */ + 0x21, 0x00, /* 0010000100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 158 0x9e '.' */ + 0x00, 0x00, /* 0000000000 */ + 0xbf, 0x00, /* 1011111100 */ + 0x40, 0x80, /* 0100000010 */ + 0x40, 0x80, /* 0100000010 */ + 0x7f, 0x00, /* 0111111100 */ + 0x40, 0x00, /* 0100000000 */ + 0x48, 0x00, /* 0100100000 */ + 0x48, 0x00, /* 0100100000 */ + 0x5e, 0x00, /* 0101111000 */ + 0x48, 0x00, /* 0100100000 */ + 0x48, 0x00, /* 0100100000 */ + 0x48, 0x00, /* 0100100000 */ + 0x48, 0x80, /* 0100100010 */ + 0x47, 0x00, /* 0100011100 */ + 0xe0, 0x00, /* 1110000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 159 0x9f '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x03, 0x00, /* 0000001100 */ + 0x04, 0x80, /* 0000010010 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x09, 0x00, /* 0000100100 */ + 0x3e, 0x00, /* 0011111000 */ + 0x48, 0x00, /* 0100100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x08, 0x00, /* 0000100000 */ + 0x90, 0x00, /* 1001000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 160 0xa0 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x21, 0x80, /* 0010000110 */ + 0x07, 0x80, /* 0000011110 */ + 0x39, 0x80, /* 0011100110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x3d, 0xc0, /* 0011110111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 161 0xa1 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x03, 0x00, /* 0000001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 162 0xa2 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x66, 0x00, /* 0110011000 */ + 0xc3, 0x00, /* 1100001100 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xc1, 0x80, /* 1100000110 */ + 0xe1, 0x80, /* 1110000110 */ + 0x73, 0x00, /* 0111001100 */ + 0x3c, 0x00, /* 0011110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 163 0xa3 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x00, 0x00, /* 0000000000 */ + 0xf7, 0x80, /* 1111011110 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x77, 0x00, /* 0111011100 */ + 0x3d, 0x80, /* 0011110110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 164 0xa4 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x38, 0x80, /* 0011100010 */ + 0x7f, 0x80, /* 0111111110 */ + 0x47, 0x00, /* 0100011100 */ + 0x00, 0x00, /* 0000000000 */ + 0x6f, 0x00, /* 0110111100 */ + 0x7b, 0x80, /* 0111101110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x7b, 0xc0, /* 0111101111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 165 0xa5 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x38, 0x80, /* 0011100010 */ + 0x7f, 0x80, /* 0111111110 */ + 0x47, 0x00, /* 0100011100 */ + 0x00, 0x00, /* 0000000000 */ + 0xe3, 0xc0, /* 1110001111 */ + 0x71, 0x80, /* 0111000110 */ + 0x79, 0x80, /* 0111100110 */ + 0x79, 0x80, /* 0111100110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x67, 0x80, /* 0110011110 */ + 0x63, 0x80, /* 0110001110 */ + 0x61, 0x80, /* 0110000110 */ + 0xf0, 0xc0, /* 1111000011 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 166 0xa6 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x63, 0x00, /* 0110001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x0f, 0x00, /* 0000111100 */ + 0x33, 0x00, /* 0011001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x67, 0x00, /* 0110011100 */ + 0x3b, 0x80, /* 0011101110 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 167 0xa7 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x21, 0x80, /* 0010000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x00, /* 0110000100 */ + 0x33, 0x00, /* 0011001100 */ + 0x1c, 0x00, /* 0001110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 168 0xa8 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x00, 0x00, /* 0000000000 */ + 0x06, 0x00, /* 0000011000 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x80, /* 0110000010 */ + 0x73, 0x80, /* 0111001110 */ + 0x3f, 0x00, /* 0011111100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 169 0xa9 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 170 0xaa '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 171 0xab '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x20, 0x00, /* 0010000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x20, 0x00, /* 0010000000 */ + 0x20, 0x80, /* 0010000010 */ + 0x21, 0x00, /* 0010000100 */ + 0x22, 0x00, /* 0010001000 */ + 0x74, 0x00, /* 0111010000 */ + 0x08, 0x00, /* 0000100000 */ + 0x17, 0x00, /* 0001011100 */ + 0x28, 0x80, /* 0010100010 */ + 0x43, 0x00, /* 0100001100 */ + 0x04, 0x00, /* 0000010000 */ + 0x08, 0x00, /* 0000100000 */ + 0x0f, 0x80, /* 0000111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 172 0xac '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x20, 0x00, /* 0010000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x20, 0x00, /* 0010000000 */ + 0x20, 0x80, /* 0010000010 */ + 0x21, 0x00, /* 0010000100 */ + 0x22, 0x00, /* 0010001000 */ + 0x74, 0x00, /* 0111010000 */ + 0x09, 0x00, /* 0000100100 */ + 0x13, 0x00, /* 0001001100 */ + 0x25, 0x00, /* 0010010100 */ + 0x49, 0x00, /* 0100100100 */ + 0x1f, 0x80, /* 0001111110 */ + 0x01, 0x00, /* 0000000100 */ + 0x01, 0x00, /* 0000000100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 173 0xad '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 174 0xae '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0d, 0x80, /* 0000110110 */ + 0x1b, 0x00, /* 0001101100 */ + 0x36, 0x00, /* 0011011000 */ + 0x6c, 0x00, /* 0110110000 */ + 0xd8, 0x00, /* 1101100000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x36, 0x00, /* 0011011000 */ + 0x1b, 0x00, /* 0001101100 */ + 0x0d, 0x80, /* 0000110110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 175 0xaf '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x36, 0x00, /* 0011011000 */ + 0x1b, 0x00, /* 0001101100 */ + 0x0d, 0x80, /* 0000110110 */ + 0x06, 0xc0, /* 0000011011 */ + 0x0d, 0x80, /* 0000110110 */ + 0x1b, 0x00, /* 0001101100 */ + 0x36, 0x00, /* 0011011000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 176 0xb0 '.' */ + 0xc3, 0x00, /* 1100001100 */ + 0x41, 0x00, /* 0100000100 */ + 0x18, 0x40, /* 0001100001 */ + 0x10, 0x40, /* 0001000001 */ + 0xc3, 0x00, /* 1100001100 */ + 0x41, 0x00, /* 0100000100 */ + 0x18, 0x40, /* 0001100001 */ + 0x10, 0x40, /* 0001000001 */ + 0xc3, 0x00, /* 1100001100 */ + 0x41, 0x00, /* 0100000100 */ + 0x18, 0x40, /* 0001100001 */ + 0x10, 0x40, /* 0001000001 */ + 0xc3, 0x00, /* 1100001100 */ + 0x41, 0x00, /* 0100000100 */ + 0x18, 0x40, /* 0001100001 */ + 0x10, 0x40, /* 0001000001 */ + 0xc3, 0x00, /* 1100001100 */ + 0x41, 0x00, /* 0100000100 */ + + /* 177 0xb1 '.' */ + 0x11, 0x00, /* 0001000100 */ + 0xbb, 0x80, /* 1011101110 */ + 0x11, 0x00, /* 0001000100 */ + 0x44, 0x40, /* 0100010001 */ + 0xee, 0xc0, /* 1110111011 */ + 0x44, 0x40, /* 0100010001 */ + 0x11, 0x00, /* 0001000100 */ + 0xbb, 0x80, /* 1011101110 */ + 0x11, 0x00, /* 0001000100 */ + 0x44, 0x40, /* 0100010001 */ + 0xee, 0xc0, /* 1110111011 */ + 0x44, 0x40, /* 0100010001 */ + 0x11, 0x00, /* 0001000100 */ + 0xbb, 0x80, /* 1011101110 */ + 0x11, 0x00, /* 0001000100 */ + 0x44, 0x40, /* 0100010001 */ + 0xee, 0xc0, /* 1110111011 */ + 0x44, 0x40, /* 0100010001 */ + + /* 178 0xb2 '.' */ + 0x3c, 0xc0, /* 0011110011 */ + 0xbe, 0xc0, /* 1011111011 */ + 0xe7, 0x80, /* 1110011110 */ + 0xef, 0x80, /* 1110111110 */ + 0x3c, 0xc0, /* 0011110011 */ + 0xbe, 0xc0, /* 1011111011 */ + 0xe7, 0x80, /* 1110011110 */ + 0xef, 0x80, /* 1110111110 */ + 0x3c, 0xc0, /* 0011110011 */ + 0xbe, 0xc0, /* 1011111011 */ + 0xe7, 0x80, /* 1110011110 */ + 0xef, 0x80, /* 1110111110 */ + 0x3c, 0xc0, /* 0011110011 */ + 0xbe, 0xc0, /* 1011111011 */ + 0xe7, 0x80, /* 1110011110 */ + 0xef, 0x80, /* 1110111110 */ + 0x3c, 0xc0, /* 0011110011 */ + 0xbe, 0xc0, /* 1011111011 */ + + /* 179 0xb3 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 180 0xb4 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 181 0xb5 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 182 0xb6 '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0xfb, 0x00, /* 1111101100 */ + 0xfb, 0x00, /* 1111101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 183 0xb7 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0x00, /* 1111111100 */ + 0xff, 0x00, /* 1111111100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 184 0xb8 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xfc, 0x00, /* 1111110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 185 0xb9 '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0xfb, 0x00, /* 1111101100 */ + 0xfb, 0x00, /* 1111101100 */ + 0x03, 0x00, /* 0000001100 */ + 0xfb, 0x00, /* 1111101100 */ + 0xfb, 0x00, /* 1111101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 186 0xba '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 187 0xbb '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0x00, /* 1111111100 */ + 0xff, 0x00, /* 1111111100 */ + 0x03, 0x00, /* 0000001100 */ + 0xfb, 0x00, /* 1111101100 */ + 0xfb, 0x00, /* 1111101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 188 0xbc '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0xfb, 0x00, /* 1111101100 */ + 0xfb, 0x00, /* 1111101100 */ + 0x03, 0x00, /* 0000001100 */ + 0xff, 0x00, /* 1111111100 */ + 0xff, 0x00, /* 1111111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 189 0xbd '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0xff, 0x00, /* 1111111100 */ + 0xff, 0x00, /* 1111111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 190 0xbe '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 191 0xbf '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xfc, 0x00, /* 1111110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 192 0xc0 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 193 0xc1 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 194 0xc2 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 195 0xc3 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 196 0xc4 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 197 0xc5 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 198 0xc6 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 199 0xc7 '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 200 0xc8 '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x18, 0x00, /* 0001100000 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 201 0xc9 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x18, 0x00, /* 0001100000 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 202 0xca '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0xfb, 0xc0, /* 1111101111 */ + 0xfb, 0xc0, /* 1111101111 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 203 0xcb '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0xfb, 0xc0, /* 1111101111 */ + 0xfb, 0xc0, /* 1111101111 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 204 0xcc '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x18, 0x00, /* 0001100000 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x1b, 0xc0, /* 0001101111 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 205 0xcd '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 206 0xce '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0xfb, 0xc0, /* 1111101111 */ + 0xfb, 0xc0, /* 1111101111 */ + 0x00, 0x00, /* 0000000000 */ + 0xfb, 0xc0, /* 1111101111 */ + 0xfb, 0xc0, /* 1111101111 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 207 0xcf '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 208 0xd0 '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 209 0xd1 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 210 0xd2 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 211 0xd3 '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 212 0xd4 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 213 0xd5 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 214 0xd6 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 215 0xd7 '.' */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + 0x1b, 0x00, /* 0001101100 */ + + /* 216 0xd8 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 217 0xd9 '.' */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0xfc, 0x00, /* 1111110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 218 0xda '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + + /* 219 0xdb '.' */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + + /* 220 0xdc '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + + /* 221 0xdd '.' */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + 0xf8, 0x00, /* 1111100000 */ + + /* 222 0xde '.' */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + 0x07, 0xc0, /* 0000011111 */ + + /* 223 0xdf '.' */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0xff, 0xc0, /* 1111111111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 224 0xe0 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1c, 0x80, /* 0001110010 */ + 0x35, 0x80, /* 0011010110 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x63, 0x00, /* 0110001100 */ + 0x37, 0x80, /* 0011011110 */ + 0x1c, 0x80, /* 0001110010 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 225 0xe1 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x63, 0x00, /* 0110001100 */ + 0x6f, 0x00, /* 0110111100 */ + 0x63, 0x00, /* 0110001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x63, 0x00, /* 0110001100 */ + 0x6e, 0x00, /* 0110111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 226 0xe2 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 227 0xe3 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 228 0xe4 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0x80, /* 1111111110 */ + 0x60, 0x00, /* 0110000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x80, /* 0011000010 */ + 0x61, 0x80, /* 0110000110 */ + 0xff, 0x80, /* 1111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 229 0xe5 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1f, 0xc0, /* 0001111111 */ + 0x36, 0x00, /* 0011011000 */ + 0x63, 0x00, /* 0110001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x33, 0x00, /* 0011001100 */ + 0x3e, 0x00, /* 0011111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 230 0xe6 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x73, 0x80, /* 0111001110 */ + 0x6d, 0x80, /* 0110110110 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0xc0, 0x00, /* 1100000000 */ + + /* 231 0xe7 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x01, 0x80, /* 0000000110 */ + 0x36, 0x40, /* 0011011001 */ + 0x5e, 0x00, /* 0101111000 */ + 0x8c, 0x00, /* 1000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 232 0xe8 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x33, 0x00, /* 0011001100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 233 0xe9 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x1f, 0x00, /* 0001111100 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x60, 0xc0, /* 0110000011 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x7f, 0xc0, /* 0111111111 */ + 0x60, 0xc0, /* 0110000011 */ + 0x31, 0x80, /* 0011000110 */ + 0x31, 0x80, /* 0011000110 */ + 0x1f, 0x00, /* 0001111100 */ + 0x0e, 0x00, /* 0000111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 234 0xea '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0xc0, 0xc0, /* 1100000011 */ + 0xc0, 0xc0, /* 1100000011 */ + 0xc0, 0xc0, /* 1100000011 */ + 0x61, 0x80, /* 0110000110 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0xf3, 0xc0, /* 1111001111 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 235 0xeb '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x07, 0x00, /* 0000011100 */ + 0x1f, 0x80, /* 0001111110 */ + 0x30, 0xc0, /* 0011000011 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x3e, 0x00, /* 0011111000 */ + 0x66, 0x00, /* 0110011000 */ + 0xc3, 0x00, /* 1100001100 */ + 0xc3, 0x00, /* 1100001100 */ + 0xc3, 0x00, /* 1100001100 */ + 0x66, 0x00, /* 0110011000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 236 0xec '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x33, 0x00, /* 0011001100 */ + 0x6d, 0x80, /* 0110110110 */ + 0xcc, 0xc0, /* 1100110011 */ + 0xcc, 0xc0, /* 1100110011 */ + 0xcc, 0xc0, /* 1100110011 */ + 0xcc, 0xc0, /* 1100110011 */ + 0x6d, 0x80, /* 0110110110 */ + 0x33, 0x00, /* 0011001100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 237 0xed '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x01, 0x80, /* 0000000110 */ + 0x01, 0x80, /* 0000000110 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x37, 0x00, /* 0011011100 */ + 0x6d, 0x80, /* 0110110110 */ + 0xcc, 0xc0, /* 1100110011 */ + 0xcc, 0xc0, /* 1100110011 */ + 0xcc, 0xc0, /* 1100110011 */ + 0xcc, 0xc0, /* 1100110011 */ + 0x6d, 0x80, /* 0110110110 */ + 0x3b, 0x00, /* 0011101100 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x60, 0x00, /* 0110000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 238 0xee '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x03, 0x80, /* 0000001110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x3f, 0x80, /* 0011111110 */ + 0x3f, 0x80, /* 0011111110 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x18, 0x00, /* 0001100000 */ + 0x18, 0x00, /* 0001100000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x03, 0x80, /* 0000001110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 239 0xef '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x61, 0x80, /* 0110000110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 240 0xf0 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 241 0xf1 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 242 0xf2 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xe0, 0x00, /* 1110000000 */ + 0x38, 0x00, /* 0011100000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x03, 0x80, /* 0000001110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x38, 0x00, /* 0011100000 */ + 0xe0, 0x00, /* 1110000000 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0x00, /* 1111111100 */ + 0xff, 0x00, /* 1111111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 243 0xf3 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x03, 0x80, /* 0000001110 */ + 0x0e, 0x00, /* 0000111000 */ + 0x38, 0x00, /* 0011100000 */ + 0xe0, 0x00, /* 1110000000 */ + 0x38, 0x00, /* 0011100000 */ + 0x0e, 0x00, /* 0000111000 */ + 0x03, 0x80, /* 0000001110 */ + 0x00, 0x00, /* 0000000000 */ + 0xff, 0x80, /* 1111111110 */ + 0xff, 0x80, /* 1111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 244 0xf4 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x30, 0x00, /* 0011000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 245 0xf5 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x03, 0x00, /* 0000001100 */ + 0x00, 0x00, /* 0000000000 */ + + /* 246 0xf6 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 247 0xf7 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x38, 0x00, /* 0011100000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x06, 0xc0, /* 0000011011 */ + 0x03, 0x80, /* 0000001110 */ + 0x38, 0x00, /* 0011100000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x06, 0xc0, /* 0000011011 */ + 0x03, 0x80, /* 0000001110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 248 0xf8 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x33, 0x00, /* 0011001100 */ + 0x33, 0x00, /* 0011001100 */ + 0x1e, 0x00, /* 0001111000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 249 0xf9 '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 250 0xfa '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 251 0xfb '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0f, 0xc0, /* 0000111111 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0xcc, 0x00, /* 1100110000 */ + 0x6c, 0x00, /* 0110110000 */ + 0x3c, 0x00, /* 0011110000 */ + 0x1c, 0x00, /* 0001110000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 252 0xfc '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x27, 0x00, /* 0010011100 */ + 0x7b, 0x00, /* 0111101100 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x31, 0x00, /* 0011000100 */ + 0x7b, 0x80, /* 0111101110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 253 0xfd '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x1e, 0x00, /* 0001111000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x63, 0x00, /* 0110001100 */ + 0x43, 0x00, /* 0100001100 */ + 0x06, 0x00, /* 0000011000 */ + 0x0c, 0x00, /* 0000110000 */ + 0x18, 0x00, /* 0001100000 */ + 0x30, 0x80, /* 0011000010 */ + 0x7f, 0x80, /* 0111111110 */ + 0x7f, 0x80, /* 0111111110 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 254 0xfe '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x3f, 0x00, /* 0011111100 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + + /* 255 0xff '.' */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + 0x00, 0x00, /* 0000000000 */ + +}; + + +struct font_desc font_10x18 = { + FONT10x18_IDX, + "10x18", + 10, + 18, + fontdata_10x18, +#ifdef __sparc__ + 5 +#else + -1 +#endif +}; diff --git a/drivers/video/console/font_7x14.c b/drivers/video/console/font_7x14.c new file mode 100644 index 000000000000..1fa7fcf2ff72 --- /dev/null +++ b/drivers/video/console/font_7x14.c @@ -0,0 +1,4118 @@ +/**************************************/ +/* this file adapted from font_8x16.c */ +/* by Jurriaan Kalkman 05-2005 */ +/**************************************/ + +#include + +#define FONTDATAMAX 3584 + +static unsigned char fontdata_7x14[FONTDATAMAX] = { + + /* 0 0x00 '^@' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 1 0x01 '^A' */ + 0x00, /* 0000000 */ + 0x7c, /* 0111110 */ + 0x82, /* 1000001 */ + 0xaa, /* 1010101 */ + 0x82, /* 1000001 */ + 0x82, /* 1000001 */ + 0xba, /* 1011101 */ + 0x92, /* 1001001 */ + 0x82, /* 1000001 */ + 0x7c, /* 0111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 2 0x02 '^B' */ + 0x00, /* 0000000 */ + 0x7c, /* 0111110 */ + 0xfe, /* 1111111 */ + 0xd6, /* 1101011 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xc6, /* 1100011 */ + 0xee, /* 1110111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0x7c, /* 0111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 3 0x03 '^C' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x6c, /* 0110110 */ + 0x7c, /* 0111110 */ + 0xfe, /* 1111111 */ + 0x7c, /* 0111110 */ + 0x38, /* 0011100 */ + 0x18, /* 0001100 */ + 0x10, /* 0001000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 4 0x04 '^D' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x10, /* 0001000 */ + 0x38, /* 0011100 */ + 0x7c, /* 0111110 */ + 0xfe, /* 1111111 */ + 0x7c, /* 0111110 */ + 0x38, /* 0011100 */ + 0x10, /* 0001000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 5 0x05 '^E' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x38, /* 0011100 */ + 0x38, /* 0011100 */ + 0xee, /* 1110111 */ + 0xee, /* 1110111 */ + 0xee, /* 1110111 */ + 0x10, /* 0001000 */ + 0x10, /* 0001000 */ + 0x38, /* 0011100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 6 0x06 '^F' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x10, /* 0001000 */ + 0x38, /* 0011100 */ + 0x7c, /* 0111110 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0x7c, /* 0111110 */ + 0x10, /* 0001000 */ + 0x10, /* 0001000 */ + 0x38, /* 0011100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 7 0x07 '^G' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x3c, /* 0011110 */ + 0x3c, /* 0011110 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 8 0x08 '^H' */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xe6, /* 1110011 */ + 0xc2, /* 1100001 */ + 0xc2, /* 1100001 */ + 0xe6, /* 1110011 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + + /* 9 0x09 '^I' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x44, /* 0100010 */ + 0x6c, /* 0110110 */ + 0x38, /* 0011100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 10 0x0a '^J' */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xc6, /* 1100011 */ + 0x92, /* 1001001 */ + 0xba, /* 1011101 */ + 0x92, /* 1001001 */ + 0xc6, /* 1100011 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + + /* 11 0x0b '^K' */ + 0x00, /* 0000000 */ + 0x1e, /* 0001111 */ + 0x0e, /* 0000111 */ + 0x1a, /* 0001101 */ + 0x1a, /* 0001101 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 12 0x0c '^L' */ + 0x00, /* 0000000 */ + 0x3c, /* 0011110 */ + 0x66, /* 0110011 */ + 0x66, /* 0110011 */ + 0x66, /* 0110011 */ + 0x66, /* 0110011 */ + 0x3c, /* 0011110 */ + 0x18, /* 0001100 */ + 0x7e, /* 0111111 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 13 0x0d '^M' */ + 0x00, /* 0000000 */ + 0x3e, /* 0011111 */ + 0x36, /* 0011011 */ + 0x3e, /* 0011111 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x70, /* 0111000 */ + 0xf0, /* 1111000 */ + 0xe0, /* 1110000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 14 0x0e '^N' */ + 0x00, /* 0000000 */ + 0x7e, /* 0111111 */ + 0x66, /* 0110011 */ + 0x7e, /* 0111111 */ + 0x66, /* 0110011 */ + 0x66, /* 0110011 */ + 0x66, /* 0110011 */ + 0x66, /* 0110011 */ + 0x6e, /* 0110111 */ + 0xee, /* 1110111 */ + 0xec, /* 1110110 */ + 0xc0, /* 1100000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 15 0x0f '^O' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x10, /* 0001000 */ + 0x10, /* 0001000 */ + 0xd6, /* 1101011 */ + 0x38, /* 0011100 */ + 0xee, /* 1110111 */ + 0x38, /* 0011100 */ + 0xd6, /* 1101011 */ + 0x10, /* 0001000 */ + 0x10, /* 0001000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 16 0x10 '^P' */ + 0x00, /* 0000000 */ + 0x80, /* 1000000 */ + 0xc0, /* 1100000 */ + 0xe0, /* 1110000 */ + 0xf0, /* 1111000 */ + 0xfc, /* 1111110 */ + 0xf0, /* 1111000 */ + 0xe0, /* 1110000 */ + 0xc0, /* 1100000 */ + 0x80, /* 1000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 17 0x11 '^Q' */ + 0x00, /* 0000000 */ + 0x04, /* 0000010 */ + 0x0c, /* 0000110 */ + 0x1c, /* 0001110 */ + 0x3c, /* 0011110 */ + 0xfc, /* 1111110 */ + 0x3c, /* 0011110 */ + 0x1c, /* 0001110 */ + 0x0c, /* 0000110 */ + 0x04, /* 0000010 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 18 0x12 '^R' */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x3c, /* 0011110 */ + 0x7e, /* 0111111 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x7e, /* 0111111 */ + 0x3c, /* 0011110 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 19 0x13 '^S' */ + 0x00, /* 0000000 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 20 0x14 '^T' */ + 0x00, /* 0000000 */ + 0x7e, /* 0111111 */ + 0xd4, /* 1101010 */ + 0xd4, /* 1101010 */ + 0xd4, /* 1101010 */ + 0x74, /* 0111010 */ + 0x14, /* 0001010 */ + 0x14, /* 0001010 */ + 0x14, /* 0001010 */ + 0x14, /* 0001010 */ + 0x16, /* 0001011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 21 0x15 '^U' */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0x60, /* 0110000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0x6c, /* 0110110 */ + 0x38, /* 0011100 */ + 0x18, /* 0001100 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 22 0x16 '^V' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xfc, /* 1111110 */ + 0xfc, /* 1111110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 23 0x17 '^W' */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x3c, /* 0011110 */ + 0x7e, /* 0111111 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x7e, /* 0111111 */ + 0x3c, /* 0011110 */ + 0x18, /* 0001100 */ + 0x7e, /* 0111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 24 0x18 '^X' */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x3c, /* 0011110 */ + 0x7e, /* 0111111 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 25 0x19 '^Y' */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x7e, /* 0111111 */ + 0x3c, /* 0011110 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 26 0x1a '^Z' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0xfc, /* 1111110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 27 0x1b '^[' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xfc, /* 1111110 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 28 0x1c '^\' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 29 0x1d '^]' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x28, /* 0010100 */ + 0x6c, /* 0110110 */ + 0xfe, /* 1111111 */ + 0x6c, /* 0110110 */ + 0x28, /* 0010100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 30 0x1e '^^' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0x78, /* 0111100 */ + 0xfc, /* 1111110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 31 0x1f '^_' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xfc, /* 1111110 */ + 0x78, /* 0111100 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 32 0x20 ' ' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 33 0x21 '!' */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x3c, /* 0011110 */ + 0x3c, /* 0011110 */ + 0x3c, /* 0011110 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 34 0x22 '"' */ + 0x00, /* 0000000 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x28, /* 0010100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 35 0x23 '#' */ + 0x00, /* 0000000 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 36 0x24 '$' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xc4, /* 1100010 */ + 0xc0, /* 1100000 */ + 0x78, /* 0111100 */ + 0x0c, /* 0000110 */ + 0x8c, /* 1000110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + + /* 37 0x25 '%' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xc0, /* 1100000 */ + 0xc4, /* 1100010 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xcc, /* 1100110 */ + 0x8c, /* 1000110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 38 0x26 '&' */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x38, /* 0011100 */ + 0x78, /* 0111100 */ + 0xde, /* 1101111 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xdc, /* 1101110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 39 0x27 ''' */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 40 0x28 '(' */ + 0x00, /* 0000000 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x0c, /* 0000110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 41 0x29 ')' */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 42 0x2a '*' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x6c, /* 0110110 */ + 0x38, /* 0011100 */ + 0xfe, /* 1111111 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 43 0x2b '+' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x10, /* 0001000 */ + 0x10, /* 0001000 */ + 0x7c, /* 0111110 */ + 0x10, /* 0001000 */ + 0x10, /* 0001000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 44 0x2c ',' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 45 0x2d '-' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 46 0x2e '.' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 47 0x2f '/' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x04, /* 0000010 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xc0, /* 1100000 */ + 0x80, /* 1000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 48 0x30 '0' */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xdc, /* 1101110 */ + 0xec, /* 1110110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 49 0x31 '1' */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x38, /* 0011100 */ + 0x78, /* 0111100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x7c, /* 0111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 50 0x32 '2' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 51 0x33 '3' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x38, /* 0011100 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 52 0x34 '4' */ + 0x00, /* 0000000 */ + 0x0c, /* 0000110 */ + 0x1c, /* 0001110 */ + 0x3c, /* 0011110 */ + 0x6c, /* 0110110 */ + 0xcc, /* 1100110 */ + 0xfe, /* 1111111 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 53 0x35 '5' */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xf8, /* 1111100 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 54 0x36 '6' */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xf8, /* 1111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 55 0x37 '7' */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 56 0x38 '8' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 57 0x39 '9' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x7c, /* 0111110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x70, /* 0111000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 58 0x3a ':' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 59 0x3b ';' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 60 0x3c '<' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x04, /* 0000010 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x0c, /* 0000110 */ + 0x04, /* 0000010 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 61 0x3d '=' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x7c, /* 0111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x7c, /* 0111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 62 0x3e '>' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x40, /* 0100000 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0x40, /* 0100000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 63 0x3f '?' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 64 0x40 '@' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xdc, /* 1101110 */ + 0xdc, /* 1101110 */ + 0xd8, /* 1101100 */ + 0xc0, /* 1100000 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 65 0x41 'A' */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 66 0x42 'B' */ + 0x00, /* 0000000 */ + 0xf8, /* 1111100 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x78, /* 0111100 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xf8, /* 1111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 67 0x43 'C' */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0xc4, /* 1100010 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc4, /* 1100010 */ + 0x6c, /* 0110110 */ + 0x38, /* 0011100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 68 0x44 'D' */ + 0x00, /* 0000000 */ + 0xf0, /* 1111000 */ + 0xd8, /* 1101100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xd8, /* 1101100 */ + 0xf0, /* 1111000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 69 0x45 'E' */ + 0x00, /* 0000000 */ + 0x7c, /* 0111110 */ + 0x6c, /* 0110110 */ + 0x64, /* 0110010 */ + 0x68, /* 0110100 */ + 0x78, /* 0111100 */ + 0x68, /* 0110100 */ + 0x60, /* 0110000 */ + 0x64, /* 0110010 */ + 0x6c, /* 0110110 */ + 0x7c, /* 0111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 70 0x46 'F' */ + 0x00, /* 0000000 */ + 0x7c, /* 0111110 */ + 0x64, /* 0110010 */ + 0x60, /* 0110000 */ + 0x68, /* 0110100 */ + 0x78, /* 0111100 */ + 0x68, /* 0110100 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x70, /* 0111000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 71 0x47 'G' */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0xc4, /* 1100010 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xdc, /* 1101110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x6c, /* 0110110 */ + 0x34, /* 0011010 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 72 0x48 'H' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 73 0x49 'I' */ + 0x00, /* 0000000 */ + 0x3c, /* 0011110 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x3c, /* 0011110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 74 0x4a 'J' */ + 0x00, /* 0000000 */ + 0x1c, /* 0001110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 75 0x4b 'K' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xd8, /* 1101100 */ + 0xf0, /* 1111000 */ + 0xf0, /* 1111000 */ + 0xd8, /* 1101100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 76 0x4c 'L' */ + 0x00, /* 0000000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc4, /* 1100010 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 77 0x4d 'M' */ + 0x00, /* 0000000 */ + 0xc6, /* 1100011 */ + 0xee, /* 1110111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xd6, /* 1101011 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 78 0x4e 'N' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xec, /* 1110110 */ + 0xec, /* 1110110 */ + 0xfc, /* 1111110 */ + 0xdc, /* 1101110 */ + 0xdc, /* 1101110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 79 0x4f 'O' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 80 0x50 'P' */ + 0x00, /* 0000000 */ + 0xf8, /* 1111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xf8, /* 1111100 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 81 0x51 'Q' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xdc, /* 1101110 */ + 0x78, /* 0111100 */ + 0x18, /* 0001100 */ + 0x1c, /* 0001110 */ + 0x00, /* 0000000 */ + + /* 82 0x52 'R' */ + 0x00, /* 0000000 */ + 0xf8, /* 1111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xf8, /* 1111100 */ + 0xd8, /* 1101100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 83 0x53 'S' */ + 0x00, /* 0000000 */ + 0x7c, /* 0111110 */ + 0xc4, /* 1100010 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0x60, /* 0110000 */ + 0x38, /* 0011100 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x8c, /* 1000110 */ + 0xf8, /* 1111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 84 0x54 'T' */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xfc, /* 1111110 */ + 0xb4, /* 1011010 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 85 0x55 'U' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 86 0x56 'V' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 87 0x57 'W' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xfc, /* 1111110 */ + 0x48, /* 0100100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 88 0x58 'X' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 89 0x59 'Y' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 90 0x5a 'Z' */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0x8c, /* 1000110 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xc4, /* 1100010 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 91 0x5b '[' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 92 0x5c '\' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x80, /* 1000000 */ + 0xc0, /* 1100000 */ + 0xe0, /* 1110000 */ + 0x70, /* 0111000 */ + 0x38, /* 0011100 */ + 0x1c, /* 0001110 */ + 0x0c, /* 0000110 */ + 0x04, /* 0000010 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 93 0x5d ']' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 94 0x5e '^' */ + 0x10, /* 0001000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0xc6, /* 1100011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 95 0x5f '_' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + + /* 96 0x60 '`' */ + 0x00, /* 0000000 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 97 0x61 'a' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0x0c, /* 0000110 */ + 0x7c, /* 0111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 98 0x62 'b' */ + 0x00, /* 0000000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xf0, /* 1111000 */ + 0xd8, /* 1101100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xf8, /* 1111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 99 0x63 'c' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 100 0x64 'd' */ + 0x00, /* 0000000 */ + 0x1c, /* 0001110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x3c, /* 0011110 */ + 0x6c, /* 0110110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 101 0x65 'e' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 102 0x66 'f' */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x64, /* 0110010 */ + 0x60, /* 0110000 */ + 0xf0, /* 1111000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0xf0, /* 1111000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 103 0x67 'g' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x76, /* 0111011 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x7c, /* 0111110 */ + 0x0c, /* 0000110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + + /* 104 0x68 'h' */ + 0x00, /* 0000000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xd8, /* 1101100 */ + 0xec, /* 1110110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 105 0x69 'i' */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x70, /* 0111000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 106 0x6a 'j' */ + 0x00, /* 0000000 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x00, /* 0000000 */ + 0x1c, /* 0001110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + + /* 107 0x6b 'k' */ + 0x00, /* 0000000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0xd8, /* 1101100 */ + 0xf0, /* 1111000 */ + 0xf0, /* 1111000 */ + 0xd8, /* 1101100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 108 0x6c 'l' */ + 0x00, /* 0000000 */ + 0x70, /* 0111000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 109 0x6d 'm' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xec, /* 1110110 */ + 0xfe, /* 1111111 */ + 0xd6, /* 1101011 */ + 0xd6, /* 1101011 */ + 0xd6, /* 1101011 */ + 0xd6, /* 1101011 */ + 0xd6, /* 1101011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 110 0x6e 'n' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xb8, /* 1011100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 111 0x6f 'o' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 112 0x70 'p' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xb8, /* 1011100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xf8, /* 1111100 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + + /* 113 0x71 'q' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x74, /* 0111010 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x7c, /* 0111110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + + /* 114 0x72 'r' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xb8, /* 1011100 */ + 0xec, /* 1110110 */ + 0xcc, /* 1100110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 115 0x73 's' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 116 0x74 't' */ + 0x00, /* 0000000 */ + 0x10, /* 0001000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xfc, /* 1111110 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x36, /* 0011011 */ + 0x1c, /* 0001110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 117 0x75 'u' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 118 0x76 'v' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 119 0x77 'w' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0xd6, /* 1101011 */ + 0xd6, /* 1101011 */ + 0xd6, /* 1101011 */ + 0xfe, /* 1111111 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 120 0x78 'x' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 121 0x79 'y' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x7c, /* 0111110 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0xf0, /* 1111000 */ + + /* 122 0x7a 'z' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 123 0x7b '{' */ + 0x00, /* 0000000 */ + 0x1c, /* 0001110 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xe0, /* 1110000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x1c, /* 0001110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 124 0x7c '|' */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 125 0x7d '}' */ + 0x00, /* 0000000 */ + 0x70, /* 0111000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x0e, /* 0000111 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x70, /* 0111000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 126 0x7e '~' */ + 0x00, /* 0000000 */ + 0xec, /* 1110110 */ + 0xb8, /* 1011100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 127 0x7f '' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x10, /* 0001000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 128 0x80 '€' */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0xc4, /* 1100010 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc4, /* 1100010 */ + 0x6c, /* 0110110 */ + 0x38, /* 0011100 */ + 0x18, /* 0001100 */ + 0x70, /* 0111000 */ + 0x00, /* 0000000 */ + + /* 129 0x81 '' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 130 0x82 '‚' */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 131 0x83 'ƒ' */ + 0x10, /* 0001000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0x0c, /* 0000110 */ + 0x7c, /* 0111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 132 0x84 '„' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0x0c, /* 0000110 */ + 0x7c, /* 0111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 133 0x85 '…' */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0x0c, /* 0000110 */ + 0x7c, /* 0111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 134 0x86 '†' */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x38, /* 0011100 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0x0c, /* 0000110 */ + 0x7c, /* 0111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 135 0x87 '‡' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xe0, /* 1110000 */ + + /* 136 0x88 'ˆ' */ + 0x10, /* 0001000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 137 0x89 '‰' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 138 0x8a 'Š' */ + 0xc0, /* 1100000 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 139 0x8b '‹' */ + 0x00, /* 0000000 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x3c, /* 0011110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 140 0x8c 'Œ' */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x70, /* 0111000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 141 0x8d '' */ + 0xc0, /* 1100000 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x70, /* 0111000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 142 0x8e 'Ž' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 143 0x8f '' */ + 0x30, /* 0011000 */ + 0x48, /* 0100100 */ + 0x48, /* 0100100 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 144 0x90 '' */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0xc4, /* 1100010 */ + 0xd0, /* 1101000 */ + 0xf0, /* 1111000 */ + 0xd0, /* 1101000 */ + 0xc4, /* 1100010 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 145 0x91 '‘' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xec, /* 1110110 */ + 0x36, /* 0011011 */ + 0x36, /* 0011011 */ + 0x7e, /* 0111111 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0x6e, /* 0110111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 146 0x92 '’' */ + 0x00, /* 0000000 */ + 0x3e, /* 0011111 */ + 0x6c, /* 0110110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xfe, /* 1111111 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xce, /* 1100111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 147 0x93 '“' */ + 0x10, /* 0001000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 148 0x94 '”' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 149 0x95 '•' */ + 0xc0, /* 1100000 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 150 0x96 '–' */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 151 0x97 '—' */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 152 0x98 '˜' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x7c, /* 0111110 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x70, /* 0111000 */ + + /* 153 0x99 '™' */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 154 0x9a 'š' */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 155 0x9b '›' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x7c, /* 0111110 */ + 0xcc, /* 1100110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0x7c, /* 0111110 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 156 0x9c 'œ' */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x64, /* 0110010 */ + 0x60, /* 0110000 */ + 0xf0, /* 1111000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0xe6, /* 1110011 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 157 0x9d '' */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0xfc, /* 1111110 */ + 0x30, /* 0011000 */ + 0xfc, /* 1111110 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 158 0x9e 'ž' */ + 0xf8, /* 1111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xf8, /* 1111100 */ + 0xc4, /* 1100010 */ + 0xcc, /* 1100110 */ + 0xde, /* 1101111 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xc6, /* 1100011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 159 0x9f 'Ÿ' */ + 0x1c, /* 0001110 */ + 0x36, /* 0011011 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xfc, /* 1111110 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xb0, /* 1011000 */ + 0xe0, /* 1110000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 160 0xa0 ' ' */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0x0c, /* 0000110 */ + 0x7c, /* 0111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 161 0xa1 '¡' */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0x00, /* 0000000 */ + 0x70, /* 0111000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 162 0xa2 '¢' */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 163 0xa3 '£' */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 164 0xa4 '¤' */ + 0x00, /* 0000000 */ + 0x76, /* 0111011 */ + 0xdc, /* 1101110 */ + 0x00, /* 0000000 */ + 0xb8, /* 1011100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 165 0xa5 '¥' */ + 0x76, /* 0111011 */ + 0xdc, /* 1101110 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xec, /* 1110110 */ + 0xec, /* 1110110 */ + 0xfc, /* 1111110 */ + 0xdc, /* 1101110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 166 0xa6 '¦' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0x7c, /* 0111110 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 167 0xa7 '§' */ + 0x00, /* 0000000 */ + 0x70, /* 0111000 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0x70, /* 0111000 */ + 0x00, /* 0000000 */ + 0xf8, /* 1111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 168 0xa8 '¨' */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xc0, /* 1100000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 169 0xa9 '©' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 170 0xaa 'ª' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 171 0xab '«' */ + 0x60, /* 0110000 */ + 0xe0, /* 1110000 */ + 0x62, /* 0110001 */ + 0x66, /* 0110011 */ + 0x6c, /* 0110110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xc0, /* 1100000 */ + 0xb8, /* 1011100 */ + 0x4c, /* 0100110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x7c, /* 0111110 */ + + /* 172 0xac '¬' */ + 0x60, /* 0110000 */ + 0xe0, /* 1110000 */ + 0x62, /* 0110001 */ + 0x66, /* 0110011 */ + 0x6c, /* 0110110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x6c, /* 0110110 */ + 0xdc, /* 1101110 */ + 0xb4, /* 1011010 */ + 0x7e, /* 0111111 */ + 0x0c, /* 0000110 */ + 0x0c, /* 0000110 */ + 0x00, /* 0000000 */ + + /* 173 0xad '­' */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0x78, /* 0111100 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 174 0xae '®' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x36, /* 0011011 */ + 0x6c, /* 0110110 */ + 0xd8, /* 1101100 */ + 0x6c, /* 0110110 */ + 0x36, /* 0011011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 175 0xaf '¯' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xd8, /* 1101100 */ + 0x6c, /* 0110110 */ + 0x36, /* 0011011 */ + 0x6c, /* 0110110 */ + 0xd8, /* 1101100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 176 0xb0 '°' */ + 0x88, /* 1000100 */ + 0x22, /* 0010001 */ + 0x88, /* 1000100 */ + 0x22, /* 0010001 */ + 0x88, /* 1000100 */ + 0x22, /* 0010001 */ + 0x88, /* 1000100 */ + 0x22, /* 0010001 */ + 0x88, /* 1000100 */ + 0x22, /* 0010001 */ + 0x88, /* 1000100 */ + 0x22, /* 0010001 */ + 0x88, /* 1000100 */ + 0x22, /* 0010001 */ + + /* 177 0xb1 '±' */ + 0x54, /* 0101010 */ + 0xaa, /* 1010101 */ + 0x54, /* 0101010 */ + 0xaa, /* 1010101 */ + 0x54, /* 0101010 */ + 0xaa, /* 1010101 */ + 0x54, /* 0101010 */ + 0xaa, /* 1010101 */ + 0x54, /* 0101010 */ + 0xaa, /* 1010101 */ + 0x54, /* 0101010 */ + 0xaa, /* 1010101 */ + 0x54, /* 0101010 */ + 0xaa, /* 1010101 */ + + /* 178 0xb2 '²' */ + 0xee, /* 1110111 */ + 0xba, /* 1011101 */ + 0xee, /* 1110111 */ + 0xba, /* 1011101 */ + 0xee, /* 1110111 */ + 0xba, /* 1011101 */ + 0xee, /* 1110111 */ + 0xba, /* 1011101 */ + 0xee, /* 1110111 */ + 0xba, /* 1011101 */ + 0xee, /* 1110111 */ + 0xba, /* 1011101 */ + 0xee, /* 1110111 */ + 0xba, /* 1011101 */ + + /* 179 0xb3 '³' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 180 0xb4 '´' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xf0, /* 1111000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 181 0xb5 'µ' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xf0, /* 1111000 */ + 0x30, /* 0011000 */ + 0xf0, /* 1111000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 182 0xb6 '¶' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xec, /* 1110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 183 0xb7 '·' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 184 0xb8 '¸' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xf0, /* 1111000 */ + 0x30, /* 0011000 */ + 0xf0, /* 1111000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 185 0xb9 '¹' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xec, /* 1110110 */ + 0x0c, /* 0000110 */ + 0xec, /* 1110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 186 0xba 'º' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 187 0xbb '»' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x0c, /* 0000110 */ + 0xec, /* 1110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 188 0xbc '¼' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xec, /* 1110110 */ + 0x0c, /* 0000110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 189 0xbd '½' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 190 0xbe '¾' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xf0, /* 1111000 */ + 0x30, /* 0011000 */ + 0xf0, /* 1111000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 191 0xbf '¿' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xf0, /* 1111000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 192 0xc0 'À' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x3e, /* 0011111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 193 0xc1 'Á' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 194 0xc2 'Â' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 195 0xc3 'Ã' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x3e, /* 0011111 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 196 0xc4 'Ä' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 197 0xc5 'Å' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xfe, /* 1111111 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 198 0xc6 'Æ' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x3e, /* 0011111 */ + 0x30, /* 0011000 */ + 0x3e, /* 0011111 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 199 0xc7 'Ç' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6e, /* 0110111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 200 0xc8 'È' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6e, /* 0110111 */ + 0x60, /* 0110000 */ + 0x7e, /* 0111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 201 0xc9 'É' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x7e, /* 0111111 */ + 0x60, /* 0110000 */ + 0x6e, /* 0110111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 202 0xca 'Ê' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xee, /* 1110111 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 203 0xcb 'Ë' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0xee, /* 1110111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 204 0xcc 'Ì' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6e, /* 0110111 */ + 0x60, /* 0110000 */ + 0x6e, /* 0110111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 205 0xcd 'Í' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 206 0xce 'Î' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xee, /* 1110111 */ + 0x00, /* 0000000 */ + 0xee, /* 1110111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 207 0xcf 'Ï' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 208 0xd0 'Ð' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 209 0xd1 'Ñ' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 210 0xd2 'Ò' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 211 0xd3 'Ó' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x7e, /* 0111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 212 0xd4 'Ô' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x3e, /* 0011111 */ + 0x30, /* 0011000 */ + 0x3e, /* 0011111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 213 0xd5 'Õ' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x3e, /* 0011111 */ + 0x30, /* 0011000 */ + 0x3e, /* 0011111 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 214 0xd6 'Ö' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x7e, /* 0111111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 215 0xd7 '×' */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xfe, /* 1111111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + + /* 216 0xd8 'Ø' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xfe, /* 1111111 */ + 0x30, /* 0011000 */ + 0xfe, /* 1111111 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 217 0xd9 'Ù' */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xf0, /* 1111000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 218 0xda 'Ú' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x3e, /* 0011111 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 219 0xdb 'Û' */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + + /* 220 0xdc 'Ü' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + + /* 221 0xdd 'Ý' */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + 0xe0, /* 1110000 */ + + /* 222 0xde 'Þ' */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + 0x1e, /* 0001111 */ + + /* 223 0xdf 'ß' */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 224 0xe0 'à' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x76, /* 0111011 */ + 0xdc, /* 1101110 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0xdc, /* 1101110 */ + 0x76, /* 0111011 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 225 0xe1 'á' */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xd8, /* 1101100 */ + 0xcc, /* 1100110 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 226 0xe2 'â' */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 227 0xe3 'ã' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfe, /* 1111111 */ + 0xfe, /* 1111111 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 228 0xe4 'ä' */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 229 0xe5 'å' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x7e, /* 0111111 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0x70, /* 0111000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 230 0xe6 'æ' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xf8, /* 1111100 */ + 0xc0, /* 1100000 */ + 0xc0, /* 1100000 */ + 0x80, /* 1000000 */ + + /* 231 0xe7 'ç' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x76, /* 0111011 */ + 0xdc, /* 1101110 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 232 0xe8 'è' */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x30, /* 0011000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x30, /* 0011000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 233 0xe9 'é' */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xfc, /* 1111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x6c, /* 0110110 */ + 0x38, /* 0011100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 234 0xea 'ê' */ + 0x00, /* 0000000 */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0xc6, /* 1100011 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0xee, /* 1110111 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 235 0xeb 'ë' */ + 0x00, /* 0000000 */ + 0x3c, /* 0011110 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x7c, /* 0111110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x78, /* 0111100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 236 0xec 'ì' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x7c, /* 0111110 */ + 0xd6, /* 1101011 */ + 0xd6, /* 1101011 */ + 0xd6, /* 1101011 */ + 0x7c, /* 0111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 237 0xed 'í' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x06, /* 0000011 */ + 0x0c, /* 0000110 */ + 0x7c, /* 0111110 */ + 0xd6, /* 1101011 */ + 0xd6, /* 1101011 */ + 0xe6, /* 1110011 */ + 0x7c, /* 0111110 */ + 0x60, /* 0110000 */ + 0xc0, /* 1100000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 238 0xee 'î' */ + 0x00, /* 0000000 */ + 0x1c, /* 0001110 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x7c, /* 0111110 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x1c, /* 0001110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 239 0xef 'ï' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0xcc, /* 1100110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 240 0xf0 'ð' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 241 0xf1 'ñ' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0xfc, /* 1111110 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 242 0xf2 'ò' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x0c, /* 0000110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 243 0xf3 'ó' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x60, /* 0110000 */ + 0xc0, /* 1100000 */ + 0x60, /* 0110000 */ + 0x30, /* 0011000 */ + 0x18, /* 0001100 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 244 0xf4 'ô' */ + 0x00, /* 0000000 */ + 0x1c, /* 0001110 */ + 0x36, /* 0011011 */ + 0x36, /* 0011011 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + + /* 245 0xf5 'õ' */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0x70, /* 0111000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 246 0xf6 'ö' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 247 0xf7 '÷' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x76, /* 0111011 */ + 0xdc, /* 1101110 */ + 0x00, /* 0000000 */ + 0x76, /* 0111011 */ + 0xdc, /* 1101110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 248 0xf8 'ø' */ + 0x38, /* 0011100 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x38, /* 0011100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 249 0xf9 'ù' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 250 0xfa 'ú' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x30, /* 0011000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 251 0xfb 'û' */ + 0x1e, /* 0001111 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0x18, /* 0001100 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0xd8, /* 1101100 */ + 0x78, /* 0111100 */ + 0x38, /* 0011100 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 252 0xfc 'ü' */ + 0xd8, /* 1101100 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x6c, /* 0110110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 253 0xfd 'ý' */ + 0x78, /* 0111100 */ + 0xcc, /* 1100110 */ + 0x18, /* 0001100 */ + 0x30, /* 0011000 */ + 0x64, /* 0110010 */ + 0xfc, /* 1111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 254 0xfe 'þ' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x7c, /* 0111110 */ + 0x7c, /* 0111110 */ + 0x7c, /* 0111110 */ + 0x7c, /* 0111110 */ + 0x7c, /* 0111110 */ + 0x7c, /* 0111110 */ + 0x7c, /* 0111110 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + + /* 255 0xff 'ÿ' */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + 0x00, /* 0000000 */ + +}; + + +struct font_desc font_7x14 = { + FONT7x14_IDX, + "7x14", + 7, + 14, + fontdata_7x14, + 0 +}; diff --git a/drivers/video/console/font_sun12x22.c b/drivers/video/console/font_sun12x22.c index 05215d0c3e09..c7bd967ea100 100644 --- a/drivers/video/console/font_sun12x22.c +++ b/drivers/video/console/font_sun12x22.c @@ -29,24 +29,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 1 0x01 '^A' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x1f, 0xc0, /* 000111111100 */ + 0x30, 0x60, /* 001100000110 */ + 0x65, 0x30, /* 011001010011 */ + 0x6d, 0xb0, /* 011011011011 */ + 0x60, 0x30, /* 011000000011 */ + 0x62, 0x30, /* 011000100011 */ + 0x62, 0x30, /* 011000100011 */ + 0x60, 0x30, /* 011000000011 */ + 0x6f, 0xb0, /* 011011111011 */ + 0x67, 0x30, /* 011001110011 */ + 0x30, 0x60, /* 001100000110 */ + 0x1f, 0xc0, /* 000111111100 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -54,24 +53,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 2 0x02 '^B' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x1f, 0xc0, /* 000111111100 */ + 0x3f, 0xe0, /* 001111111110 */ + 0x7a, 0xf0, /* 011110101111 */ + 0x72, 0x70, /* 011100100111 */ + 0x7f, 0xf0, /* 011111111111 */ + 0x7d, 0xf0, /* 011111011111 */ + 0x7d, 0xf0, /* 011111011111 */ + 0x7f, 0xf0, /* 011111111111 */ + 0x70, 0x70, /* 011100000111 */ + 0x78, 0xf0, /* 011110001111 */ + 0x3f, 0xe0, /* 001111111110 */ + 0x1f, 0xc0, /* 000111111100 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -79,24 +77,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 3 0x03 '^C' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x19, 0x80, /* 000110011000 */ 0x3f, 0xc0, /* 001111111100 */ + 0x7f, 0xe0, /* 011111111110 */ 0x3f, 0xc0, /* 001111111100 */ 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x1f, 0x80, /* 000111111000 */ + 0x1f, 0x80, /* 000111111000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x06, 0x00, /* 000001100000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -104,24 +101,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 4 0x04 '^D' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x02, 0x00, /* 000000100000 */ + 0x07, 0x00, /* 000001110000 */ + 0x0f, 0x80, /* 000011111000 */ + 0x0f, 0x80, /* 000011111000 */ + 0x1f, 0xc0, /* 000111111100 */ + 0x1f, 0xc0, /* 000111111100 */ + 0x3f, 0xe0, /* 001111111110 */ + 0x1f, 0xc0, /* 000111111100 */ + 0x1f, 0xc0, /* 000111111100 */ + 0x0f, 0x80, /* 000011111000 */ + 0x0f, 0x80, /* 000011111000 */ + 0x07, 0x00, /* 000001110000 */ + 0x02, 0x00, /* 000000100000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -129,24 +125,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 5 0x05 '^E' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x02, 0x00, /* 000000100000 */ + 0x07, 0x00, /* 000001110000 */ + 0x07, 0x00, /* 000001110000 */ + 0x02, 0x00, /* 000000100000 */ + 0x18, 0xc0, /* 000110001100 */ + 0x3d, 0xe0, /* 001111011110 */ + 0x3d, 0xe0, /* 001111011110 */ + 0x1a, 0xc0, /* 000110101100 */ + 0x02, 0x00, /* 000000100000 */ + 0x07, 0x00, /* 000001110000 */ + 0x0f, 0x80, /* 000011111000 */ + 0x1f, 0xc0, /* 000111111100 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -154,23 +149,22 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 6 0x06 '^F' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x06, 0x00, /* 000001100000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x1f, 0x80, /* 000111111000 */ + 0x1f, 0x80, /* 000111111000 */ 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x36, 0xc0, /* 001101101100 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x1f, 0x80, /* 000111111000 */ 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -179,24 +173,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 7 0x07 '^G' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x06, 0x00, /* 000001100000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x1f, 0x80, /* 000111111000 */ + 0x1f, 0x80, /* 000111111000 */ 0x3f, 0xc0, /* 001111111100 */ 0x3f, 0xc0, /* 001111111100 */ + 0x1f, 0x80, /* 000111111000 */ + 0x1f, 0x80, /* 000111111000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x06, 0x00, /* 000001100000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -204,49 +197,47 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 8 0x08 '^H' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xf9, 0xf0, /* 111110011111 */ + 0xf0, 0xf0, /* 111100001111 */ + 0xf0, 0xf0, /* 111100001111 */ + 0xe0, 0x70, /* 111000000111 */ + 0xe0, 0x70, /* 111000000111 */ + 0xc0, 0x30, /* 110000000011 */ + 0xc0, 0x30, /* 110000000011 */ + 0xe0, 0x70, /* 111000000111 */ + 0xe0, 0x70, /* 111000000111 */ + 0xf0, 0xf0, /* 111100001111 */ + 0xf0, 0xf0, /* 111100001111 */ + 0xf9, 0xf0, /* 111110011111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ /* 9 0x09 '^I' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x06, 0x00, /* 000001100000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x06, 0x00, /* 000001100000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -254,74 +245,71 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 10 0x0a '^J' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xf9, 0xf0, /* 111110011111 */ + 0xf0, 0xf0, /* 111100001111 */ + 0xf0, 0xf0, /* 111100001111 */ + 0xe6, 0x70, /* 111001100111 */ + 0xe6, 0x70, /* 111001100111 */ + 0xcf, 0x30, /* 110011110011 */ + 0xcf, 0x30, /* 110011110011 */ + 0xe6, 0x70, /* 111001100111 */ + 0xe6, 0x70, /* 111001100111 */ + 0xf0, 0xf0, /* 111100001111 */ + 0xf0, 0xf0, /* 111100001111 */ + 0xf9, 0xf0, /* 111110011111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ /* 11 0x0b '^K' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x0f, 0xe0, /* 000011111110 */ + 0x0f, 0xe0, /* 000011111110 */ + 0x01, 0xe0, /* 000000011110 */ + 0x03, 0x60, /* 000000110110 */ + 0x06, 0x60, /* 000001100110 */ + 0x1e, 0x00, /* 000111100000 */ + 0x33, 0x00, /* 001100110000 */ + 0x33, 0x00, /* 001100110000 */ + 0x61, 0x80, /* 011000011000 */ + 0x61, 0x80, /* 011000011000 */ + 0x33, 0x00, /* 001100110000 */ + 0x33, 0x00, /* 001100110000 */ + 0x1e, 0x00, /* 000111100000 */ + 0x0c, 0x00, /* 000011000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ /* 12 0x0c '^L' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x06, 0x00, /* 000001100000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x19, 0x80, /* 000110011000 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x19, 0x80, /* 000110011000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ 0x3f, 0xc0, /* 001111111100 */ 0x3f, 0xc0, /* 001111111100 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -329,149 +317,143 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 13 0x0d '^M' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x0f, 0xe0, /* 000011111110 */ + 0x0c, 0x60, /* 000011000110 */ + 0x0c, 0x60, /* 000011000110 */ + 0x0f, 0xe0, /* 000011111110 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x3c, 0x00, /* 001111000000 */ + 0x7c, 0x00, /* 011111000000 */ + 0x78, 0x00, /* 011110000000 */ + 0x30, 0x00, /* 001100000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ /* 14 0x0e '^N' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x1f, 0xe0, /* 000111111110 */ + 0x18, 0x60, /* 000110000110 */ + 0x18, 0x60, /* 000110000110 */ + 0x1f, 0xe0, /* 000111111110 */ + 0x18, 0x60, /* 000110000110 */ + 0x18, 0x60, /* 000110000110 */ + 0x18, 0x60, /* 000110000110 */ + 0x18, 0x60, /* 000110000110 */ + 0x18, 0x60, /* 000110000110 */ + 0x18, 0x60, /* 000110000110 */ + 0x19, 0xe0, /* 000110011110 */ + 0x1b, 0xe0, /* 000110111110 */ + 0x1b, 0xc0, /* 000110111100 */ + 0x79, 0x80, /* 011110011000 */ + 0xf8, 0x00, /* 111110000000 */ + 0xf0, 0x00, /* 111100000000 */ + 0x60, 0x00, /* 011000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ /* 15 0x0f '^O' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x18, 0xc0, /* 000110001100 */ + 0x0d, 0x80, /* 000011011000 */ + 0x6d, 0xb0, /* 011011011011 */ + 0x3d, 0xe0, /* 001111011110 */ 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - - /* 16 0x10 '^P' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ + 0x3d, 0xe0, /* 001111011110 */ + 0x6d, 0xb0, /* 011011011011 */ + 0x0d, 0x80, /* 000011011000 */ + 0x18, 0xc0, /* 000110001100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + + /* 16 0x10 '^P' */ 0x00, 0x00, /* 000000000000 */ + 0x00, 0x20, /* 000000000010 */ + 0x00, 0x60, /* 000000000110 */ + 0x00, 0xe0, /* 000000001110 */ + 0x01, 0xe0, /* 000000011110 */ + 0x03, 0xe0, /* 000000111110 */ + 0x07, 0xe0, /* 000001111110 */ + 0x0f, 0xe0, /* 000011111110 */ + 0x1f, 0xe0, /* 000111111110 */ + 0x3f, 0xe0, /* 001111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x3f, 0xe0, /* 001111111110 */ + 0x1f, 0xe0, /* 000111111110 */ + 0x0f, 0xe0, /* 000011111110 */ + 0x07, 0xe0, /* 000001111110 */ + 0x03, 0xe0, /* 000000111110 */ + 0x01, 0xe0, /* 000000011110 */ + 0x00, 0xe0, /* 000000001110 */ + 0x00, 0x60, /* 000000000110 */ + 0x00, 0x20, /* 000000000010 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ /* 17 0x11 '^Q' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x40, 0x00, /* 010000000000 */ + 0x60, 0x00, /* 011000000000 */ + 0x70, 0x00, /* 011100000000 */ + 0x78, 0x00, /* 011110000000 */ + 0x7c, 0x00, /* 011111000000 */ + 0x7e, 0x00, /* 011111100000 */ + 0x7f, 0x00, /* 011111110000 */ + 0x7f, 0x80, /* 011111111000 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x7f, 0x80, /* 011111111000 */ + 0x7f, 0x00, /* 011111110000 */ + 0x7e, 0x00, /* 011111100000 */ + 0x7c, 0x00, /* 011111000000 */ + 0x78, 0x00, /* 011110000000 */ + 0x70, 0x00, /* 011100000000 */ + 0x60, 0x00, /* 011000000000 */ + 0x40, 0x00, /* 010000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ /* 18 0x12 '^R' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x04, 0x00, /* 000001000000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x3f, 0x80, /* 001111111000 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x3f, 0x80, /* 001111111000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x04, 0x00, /* 000001000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -551,99 +533,95 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 22 0x16 '^V' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - - /* 23 0x17 '^W' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - /* 24 0x18 '^X' */ - /* FIXME */ + /* 23 0x17 '^W' */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x04, 0x00, /* 000001000000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x3f, 0x80, /* 001111111000 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x3f, 0x80, /* 001111111000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x04, 0x00, /* 000001000000 */ 0x00, 0x00, /* 000000000000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + + /* 24 0x18 '^X' */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - - /* 25 0x19 '^Y' */ - /* FIXME */ + 0x04, 0x00, /* 000001000000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x3f, 0x80, /* 001111111000 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x00, 0x00, /* 000000000000 */ + + /* 25 0x19 '^Y' */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x3f, 0x80, /* 001111111000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x04, 0x00, /* 000001000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -651,24 +629,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 26 0x1a '^Z' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x00, 0x00, /* 000000000000 */ + 0x08, 0x00, /* 000010000000 */ + 0x18, 0x00, /* 000110000000 */ + 0x38, 0x00, /* 001110000000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0xff, 0xe0, /* 111111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x38, 0x00, /* 001110000000 */ + 0x18, 0x00, /* 000110000000 */ + 0x08, 0x00, /* 000010000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -676,24 +653,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 27 0x1b '^[' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x00, 0x00, /* 000000000000 */ + 0x01, 0x00, /* 000000010000 */ + 0x01, 0x80, /* 000000011000 */ + 0x01, 0xc0, /* 000000011100 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xf0, /* 011111111111 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x01, 0xc0, /* 000000011100 */ + 0x01, 0x80, /* 000000011000 */ + 0x01, 0x00, /* 000000010000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -701,24 +677,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 28 0x1c '^\' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x3f, 0xe0, /* 001111111110 */ + 0x3f, 0xe0, /* 001111111110 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -726,24 +701,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 29 0x1d '^]' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x00, 0x00, /* 000000000000 */ + 0x09, 0x00, /* 000010010000 */ + 0x19, 0x80, /* 000110011000 */ + 0x39, 0xc0, /* 001110011100 */ + 0x7f, 0xe0, /* 011111111110 */ + 0xff, 0xf0, /* 111111111111 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x39, 0xc0, /* 001110011100 */ + 0x19, 0x80, /* 000110011000 */ + 0x09, 0x00, /* 000010010000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -751,24 +725,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 30 0x1e '^^' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x00, 0x00, /* 000000000000 */ + 0x04, 0x00, /* 000001000000 */ + 0x04, 0x00, /* 000001000000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x3f, 0x80, /* 001111111000 */ + 0x3f, 0x80, /* 001111111000 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -776,24 +749,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 31 0x1f '^_' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x00, 0x00, /* 000000000000 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x3f, 0x80, /* 001111111000 */ + 0x3f, 0x80, /* 001111111000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x04, 0x00, /* 000001000000 */ + 0x04, 0x00, /* 000001000000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -3081,29 +3053,28 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 127 0x7f '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0xff, 0xf0, /* 111111111111 */ + 0xff, 0xf0, /* 111111111111 */ + 0x00, 0x00, /* 000000000000 */ /* 128 0x80 '.' */ 0x00, 0x00, /* 000000000000 */ @@ -3826,24 +3797,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 158 0x9e '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x7f, 0x80, /* 011111111000 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0x60, /* 001100000110 */ + 0x30, 0x60, /* 001100000110 */ + 0x30, 0x60, /* 001100000110 */ + 0x30, 0xc0, /* 001100001100 */ + 0x37, 0x80, /* 001101111000 */ + 0x30, 0x00, /* 001100000000 */ + 0x33, 0x00, /* 001100110000 */ + 0x37, 0x80, /* 001101111000 */ + 0x33, 0x00, /* 001100110000 */ + 0x33, 0x00, /* 001100110000 */ + 0x33, 0x30, /* 001100110011 */ + 0x31, 0xe0, /* 001100011110 */ + 0x78, 0xc0, /* 011110001100 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -3851,28 +3821,27 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 159 0x9f '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x00, 0xc0, /* 000000001100 */ + 0x01, 0xe0, /* 000000011110 */ + 0x03, 0x30, /* 000000110011 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x3f, 0xe0, /* 001111111110 */ + 0x7f, 0xc0, /* 011111111100 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0xcc, 0x00, /* 110011000000 */ + 0x78, 0x00, /* 011110000000 */ + 0x30, 0x00, /* 001100000000 */ 0x00, 0x00, /* 000000000000 */ /* 160 0xa0 '.' */ @@ -4092,24 +4061,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 169 0xa9 '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x3f, 0xc0, /* 001111111100 */ 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5413,24 +5381,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 224 0xe0 '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x00, 0x00, /* 000000000000 */ + 0x0f, 0x60, /* 000011110110 */ + 0x13, 0xe0, /* 000100111110 */ + 0x21, 0xc0, /* 001000011100 */ + 0x60, 0xc0, /* 011000001100 */ + 0x60, 0xc0, /* 011000001100 */ + 0x60, 0xc0, /* 011000001100 */ + 0x60, 0xc0, /* 011000001100 */ + 0x70, 0x80, /* 011100001000 */ + 0x39, 0xc0, /* 001110011100 */ + 0x1f, 0x60, /* 000111110110 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5462,24 +5429,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 226 0xe2 '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x3f, 0xe0, /* 001111111110 */ + 0x3f, 0xe0, /* 001111111110 */ + 0x30, 0x60, /* 001100000110 */ + 0x30, 0x60, /* 001100000110 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5487,49 +5453,47 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 227 0xe3 '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - - /* 228 0xe4 '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + + /* 228 0xe4 '.' */ 0x00, 0x00, /* 000000000000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x60, 0x60, /* 011000000110 */ + 0x30, 0x60, /* 001100000110 */ + 0x30, 0x00, /* 001100000000 */ + 0x18, 0x00, /* 000110000000 */ + 0x18, 0x00, /* 000110000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x18, 0x00, /* 000110000000 */ + 0x18, 0x00, /* 000110000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x60, /* 001100000110 */ + 0x60, 0x60, /* 011000000110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5537,24 +5501,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 229 0xe5 '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x07, 0xe0, /* 000001111110 */ + 0x0f, 0xe0, /* 000011111110 */ + 0x13, 0x80, /* 000100111000 */ + 0x21, 0xc0, /* 001000011100 */ + 0x60, 0xc0, /* 011000001100 */ + 0x60, 0xc0, /* 011000001100 */ + 0x60, 0xc0, /* 011000001100 */ + 0x60, 0xc0, /* 011000001100 */ + 0x70, 0x80, /* 011100001000 */ + 0x39, 0x00, /* 001110010000 */ + 0x1e, 0x00, /* 000111100000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5586,24 +5549,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 231 0xe7 '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x19, 0x80, /* 000110011000 */ 0x3f, 0xc0, /* 001111111100 */ + 0x66, 0x60, /* 011001100110 */ + 0x66, 0x60, /* 011001100110 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5611,24 +5573,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 232 0xe8 '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x19, 0x80, /* 000110011000 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x19, 0x80, /* 000110011000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5636,24 +5597,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 233 0xe9 '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x0f, 0x00, /* 000011110000 */ + 0x1f, 0x80, /* 000111111000 */ + 0x30, 0xc0, /* 001100001100 */ + 0x60, 0x60, /* 011000000110 */ + 0x60, 0x60, /* 011000000110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x60, 0x60, /* 011000000110 */ + 0x60, 0x60, /* 011000000110 */ + 0x30, 0xc0, /* 001100001100 */ + 0x1f, 0x80, /* 000111111000 */ + 0x0f, 0x00, /* 000011110000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5661,24 +5621,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 234 0xea '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x1f, 0x00, /* 000111110000 */ + 0x31, 0x80, /* 001100011000 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x60, 0x60, /* 011000000110 */ + 0x60, 0x60, /* 011000000110 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x19, 0x80, /* 000110011000 */ + 0x19, 0x80, /* 000110011000 */ + 0xd9, 0xb0, /* 110110011011 */ + 0x79, 0xe0, /* 011110011110 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5686,24 +5645,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 235 0xeb '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x07, 0x80, /* 000001111000 */ + 0x0c, 0xc0, /* 000011001100 */ + 0x18, 0x60, /* 000110000110 */ + 0x18, 0x00, /* 000110000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x06, 0x00, /* 000001100000 */ + 0x03, 0x00, /* 000000110000 */ + 0x0f, 0x80, /* 000011111000 */ + 0x11, 0xc0, /* 000100011100 */ + 0x20, 0xe0, /* 001000001110 */ + 0x60, 0x60, /* 011000000110 */ + 0x60, 0x60, /* 011000000110 */ + 0x70, 0x40, /* 011100000100 */ + 0x38, 0x80, /* 001110001000 */ + 0x1f, 0x00, /* 000111110000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5711,74 +5669,71 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 236 0xec '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x39, 0xc0, /* 001110011100 */ + 0x6f, 0x60, /* 011011110110 */ + 0x66, 0x60, /* 011001100110 */ + 0xc6, 0x30, /* 110001100011 */ + 0xc6, 0x30, /* 110001100011 */ + 0x66, 0x60, /* 011001100110 */ + 0x6f, 0x60, /* 011011110110 */ + 0x39, 0xc0, /* 001110011100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - - /* 237 0xed '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + + /* 237 0xed '.' */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x00, 0xc0, /* 000000001100 */ + 0x00, 0xc0, /* 000000001100 */ + 0x01, 0x80, /* 000000011000 */ + 0x01, 0x80, /* 000000011000 */ + 0x3b, 0xc0, /* 001110111100 */ + 0x6f, 0x60, /* 011011110110 */ + 0x66, 0x60, /* 011001100110 */ + 0xc6, 0x30, /* 110001100011 */ + 0xc6, 0x30, /* 110001100011 */ + 0x66, 0x60, /* 011001100110 */ + 0x6f, 0x60, /* 011011110110 */ + 0x3d, 0xc0, /* 001111011100 */ + 0x18, 0x00, /* 000110000000 */ + 0x18, 0x00, /* 000110000000 */ + 0x30, 0x00, /* 001100000000 */ + 0x30, 0x00, /* 001100000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ /* 238 0xee '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x01, 0xc0, /* 000000011100 */ + 0x03, 0x00, /* 000000110000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x18, 0x00, /* 000110000000 */ + 0x1f, 0xc0, /* 000111111100 */ + 0x18, 0x00, /* 000110000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x03, 0x00, /* 000000110000 */ + 0x01, 0xc0, /* 000000011100 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5786,24 +5741,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 239 0xef '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x0f, 0x00, /* 000011110000 */ + 0x1f, 0x80, /* 000111111000 */ + 0x39, 0xc0, /* 001110011100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ + 0x30, 0xc0, /* 001100001100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5811,24 +5765,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 240 0xf0 '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5860,24 +5813,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 242 0xf2 '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x60, 0x00, /* 011000000000 */ + 0x38, 0x00, /* 001110000000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x03, 0x80, /* 000000111000 */ + 0x00, 0xe0, /* 000000001110 */ + 0x00, 0xe0, /* 000000001110 */ + 0x03, 0x80, /* 000000111000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x38, 0x00, /* 001110000000 */ + 0x60, 0x00, /* 011000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5885,24 +5837,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 243 0xf3 '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x00, 0x60, /* 000000000110 */ + 0x01, 0xc0, /* 000000011100 */ + 0x07, 0x00, /* 000001110000 */ + 0x1c, 0x00, /* 000111000000 */ + 0x70, 0x00, /* 011100000000 */ + 0x70, 0x00, /* 011100000000 */ + 0x1c, 0x00, /* 000111000000 */ + 0x07, 0x00, /* 000001110000 */ + 0x01, 0xc0, /* 000000011100 */ + 0x00, 0x60, /* 000000000110 */ 0x00, 0x00, /* 000000000000 */ + 0x7f, 0xe0, /* 011111111110 */ + 0x7f, 0xe0, /* 011111111110 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -5910,54 +5861,52 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 244 0xf4 '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x03, 0x80, /* 000000111000 */ + 0x07, 0xc0, /* 000001111100 */ + 0x0c, 0x60, /* 000011000110 */ + 0x0c, 0x60, /* 000011000110 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ + 0x0c, 0x00, /* 000011000000 */ /* 245 0xf5 '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x1c, 0x00, /* 000111000000 */ + 0x3e, 0x00, /* 001111100000 */ + 0x63, 0x00, /* 011000110000 */ + 0x63, 0x00, /* 011000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ + 0x03, 0x00, /* 000000110000 */ /* 246 0xf6 '.' */ 0x00, 0x00, /* 000000000000 */ @@ -5984,24 +5933,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 247 0xf7 '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x38, 0x00, /* 001110000000 */ + 0x6c, 0x00, /* 011011000000 */ + 0x06, 0x30, /* 000001100011 */ + 0x03, 0x60, /* 000000110110 */ + 0x39, 0xc0, /* 001110011100 */ + 0x6c, 0x00, /* 011011000000 */ + 0x06, 0x30, /* 000001100011 */ + 0x03, 0x60, /* 000000110110 */ + 0x01, 0xc0, /* 000000011100 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -6033,44 +5981,31 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ /* 249 0xf9 '.' */ - /* FIXME */ - 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - - /* 250 0xfa '.' */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x1c, 0x00, /* 000111000000 */ + 0x3e, 0x00, /* 001111100000 */ + 0x3e, 0x00, /* 001111100000 */ + 0x3e, 0x00, /* 001111100000 */ + 0x1c, 0x00, /* 000111000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + + /* 250 0xfa '.' */ 0x00, 0x00, /* 000000000000 */ - 0x06, 0x00, /* 000001100000 */ - 0x0f, 0x00, /* 000011110000 */ - 0x0f, 0x00, /* 000011110000 */ - 0x06, 0x00, /* 000001100000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ @@ -6080,51 +6015,61 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = { 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - - /* 251 0xfb '.' */ - /* FIXME */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + 0x18, 0x00, /* 000110000000 */ + 0x3c, 0x00, /* 001111000000 */ + 0x3c, 0x00, /* 001111000000 */ + 0x18, 0x00, /* 000110000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ + + /* 251 0xfb '.' */ 0x00, 0x00, /* 000000000000 */ + 0x07, 0xe0, /* 000001111110 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x06, 0x00, /* 000001100000 */ + 0xc6, 0x00, /* 110001100000 */ + 0x66, 0x00, /* 011001100000 */ + 0x36, 0x00, /* 001101100000 */ + 0x1e, 0x00, /* 000111100000 */ + 0x0e, 0x00, /* 000011100000 */ + 0x06, 0x00, /* 000001100000 */ + 0x02, 0x00, /* 000000100000 */ 0x00, 0x00, /* 000000000000 */ /* 252 0xfc '.' */ - /* FIXME */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x13, 0x80, /* 000100111000 */ + 0x3d, 0xc0, /* 001111011100 */ + 0x18, 0xc0, /* 000110001100 */ + 0x18, 0xc0, /* 000110001100 */ + 0x18, 0xc0, /* 000110001100 */ + 0x18, 0xc0, /* 000110001100 */ + 0x3d, 0xe0, /* 001111011110 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ + 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ - 0x3f, 0xc0, /* 001111111100 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ 0x00, 0x00, /* 000000000000 */ diff --git a/drivers/video/console/fonts.c b/drivers/video/console/fonts.c index 465d678230ae..e79b29702649 100644 --- a/drivers/video/console/fonts.c +++ b/drivers/video/console/fonts.c @@ -36,6 +36,10 @@ static struct font_desc *fonts[] = { #undef NO_FONTS &font_vga_6x11, #endif +#ifdef CONFIG_FONT_7x14 +#undef NO_FONTS + &font_7x14, +#endif #ifdef CONFIG_FONT_SUN8x16 #undef NO_FONTS &font_sun_8x16, @@ -44,6 +48,10 @@ static struct font_desc *fonts[] = { #undef NO_FONTS &font_sun_12x22, #endif +#ifdef CONFIG_FONT_10x18 +#undef NO_FONTS + &font_10x18, +#endif #ifdef CONFIG_FONT_ACORN_8x8 #undef NO_FONTS &font_acorn_8x8, diff --git a/include/linux/font.h b/include/linux/font.h index fc2d690c9d5f..8fc80a7d78ac 100644 --- a/include/linux/font.h +++ b/include/linux/font.h @@ -25,19 +25,23 @@ struct font_desc { #define VGA8x16_IDX 1 #define PEARL8x8_IDX 2 #define VGA6x11_IDX 3 -#define SUN8x16_IDX 4 -#define SUN12x22_IDX 5 -#define ACORN8x8_IDX 6 -#define MINI4x6_IDX 7 +#define FONT7x14_IDX 4 +#define FONT10x18_IDX 5 +#define SUN8x16_IDX 6 +#define SUN12x22_IDX 7 +#define ACORN8x8_IDX 8 +#define MINI4x6_IDX 9 extern struct font_desc font_vga_8x8, - font_vga_8x16, - font_pearl_8x8, - font_vga_6x11, - font_sun_8x16, - font_sun_12x22, - font_acorn_8x8, - font_mini_4x6; + font_vga_8x16, + font_pearl_8x8, + font_vga_6x11, + font_7x14, + font_10x18, + font_sun_8x16, + font_sun_12x22, + font_acorn_8x8, + font_mini_4x6; /* Find a font with a specific name */ -- cgit v1.2.3 From f1ab5dac251bb4514607918b0019a3b3f5f5fb48 Mon Sep 17 00:00:00 2001 From: James Simmons Date: Tue, 21 Jun 2005 17:17:07 -0700 Subject: [PATCH] fbdev: stack reduction Shrink the stack when calling the drawing alignment functions. Signed-off-by: James Simmons Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/console/bitblit.c | 9 ++++----- drivers/video/fbmem.c | 14 +++++--------- drivers/video/nvidia/nvidia.c | 17 +++++++---------- drivers/video/riva/fbdev.c | 18 +++++++----------- drivers/video/softcursor.c | 5 +---- include/linux/fb.h | 7 ++----- 6 files changed, 26 insertions(+), 44 deletions(-) (limited to 'include/linux') diff --git a/drivers/video/console/bitblit.c b/drivers/video/console/bitblit.c index 4eef20790c3e..3c731577fed6 100644 --- a/drivers/video/console/bitblit.c +++ b/drivers/video/console/bitblit.c @@ -157,9 +157,9 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info, src = buf; } - fb_sysmove_buf_unaligned(info, &info->pixmap, dst, pitch, - src, idx, image.height, - shift_high, shift_low, mod); + fb_pad_unaligned_buffer(dst, pitch, src, idx, + image.height, shift_high, + shift_low, mod); shift_low += mod; dst += (shift_low >= 8) ? width : width - 1; shift_low &= 7; @@ -175,8 +175,7 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info, src = buf; } - fb_sysmove_buf_aligned(info, &info->pixmap, dst, pitch, - src, idx, image.height); + fb_pad_aligned_buffer(dst, pitch, src, idx, image.height); dst += width; } } diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 78907a873493..6f871a8b9058 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -78,9 +78,7 @@ EXPORT_SYMBOL(fb_get_color_depth); /* * Data padding functions. */ -void fb_sysmove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, - u32 height) +void fb_pad_aligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, u32 height) { int i; @@ -90,12 +88,10 @@ void fb_sysmove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf, dst += d_pitch; } } -EXPORT_SYMBOL(fb_sysmove_buf_aligned); +EXPORT_SYMBOL(fb_pad_aligned_buffer); -void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 idx, - u32 height, u32 shift_high, u32 shift_low, - u32 mod) +void fb_pad_unaligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 idx, u32 height, + u32 shift_high, u32 shift_low, u32 mod) { u8 mask = (u8) (0xfff << shift_high), tmp; int i, j; @@ -122,7 +118,7 @@ void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf, dst += d_pitch; } } -EXPORT_SYMBOL(fb_sysmove_buf_unaligned); +EXPORT_SYMBOL(fb_pad_unaligned_buffer); /* * we need to lock this section since fb_cursor diff --git a/drivers/video/nvidia/nvidia.c b/drivers/video/nvidia/nvidia.c index 47733f58153b..266ff5d7ac36 100644 --- a/drivers/video/nvidia/nvidia.c +++ b/drivers/video/nvidia/nvidia.c @@ -516,9 +516,9 @@ static struct backlight_controller nvidia_backlight_controller = { static void nvidiafb_load_cursor_image(struct nvidia_par *par, u8 * data8, u16 bg, u16 fg, u32 w, u32 h) { + u32 *data = (u32 *) data8; int i, j, k = 0; u32 b, tmp; - u32 *data = (u32 *) data8; w = (w + 1) & ~1; @@ -890,11 +890,11 @@ static int nvidiafb_cursor(struct fb_info *info, struct fb_cursor *cursor) { struct nvidia_par *par = info->par; u8 data[MAX_CURS * MAX_CURS / 8]; - u16 fg, bg; int i, set = cursor->set; + u16 fg, bg; if (cursor->image.width > MAX_CURS || cursor->image.height > MAX_CURS) - return soft_cursor(info, cursor); + return -ENXIO; NVShowHideCursor(par, 0); @@ -931,21 +931,18 @@ static int nvidiafb_cursor(struct fb_info *info, struct fb_cursor *cursor) if (src) { switch (cursor->rop) { case ROP_XOR: - for (i = 0; i < s_pitch * cursor->image.height; - i++) + for (i = 0; i < s_pitch * cursor->image.height; i++) src[i] = dat[i] ^ msk[i]; break; case ROP_COPY: default: - for (i = 0; i < s_pitch * cursor->image.height; - i++) + for (i = 0; i < s_pitch * cursor->image.height; i++) src[i] = dat[i] & msk[i]; break; } - fb_sysmove_buf_aligned(info, &info->pixmap, data, - d_pitch, src, s_pitch, - cursor->image.height); + fb_pad_aligned_buffer(data, d_pitch, src, s_pitch, + cursor->image.height); bg = ((info->cmap.red[bg_idx] & 0xf8) << 7) | ((info->cmap.green[bg_idx] & 0xf8) << 2) | diff --git a/drivers/video/riva/fbdev.c b/drivers/video/riva/fbdev.c index b0c886de0404..7540f60af948 100644 --- a/drivers/video/riva/fbdev.c +++ b/drivers/video/riva/fbdev.c @@ -1582,12 +1582,11 @@ static int rivafb_cursor(struct fb_info *info, struct fb_cursor *cursor) { struct riva_par *par = (struct riva_par *) info->par; u8 data[MAX_CURS * MAX_CURS/8]; - u16 fg, bg; int i, set = cursor->set; + u16 fg, bg; - if (cursor->image.width > MAX_CURS || - cursor->image.height > MAX_CURS) - return soft_cursor(info, cursor); + if (cursor->image.width > MAX_CURS || cursor->image.height > MAX_CURS) + return -ENXIO; par->riva.ShowHideCursor(&par->riva, 0); @@ -1625,21 +1624,18 @@ static int rivafb_cursor(struct fb_info *info, struct fb_cursor *cursor) if (src) { switch (cursor->rop) { case ROP_XOR: - for (i = 0; i < s_pitch * cursor->image.height; - i++) + for (i = 0; i < s_pitch * cursor->image.height; i++) src[i] = dat[i] ^ msk[i]; break; case ROP_COPY: default: - for (i = 0; i < s_pitch * cursor->image.height; - i++) + for (i = 0; i < s_pitch * cursor->image.height; i++) src[i] = dat[i] & msk[i]; break; } - fb_sysmove_buf_aligned(info, &info->pixmap, data, - d_pitch, src, s_pitch, - cursor->image.height); + fb_pad_aligned_buffer(data, d_pitch, src, s_pitch, + cursor->image.height); bg = ((info->cmap.red[bg_idx] & 0xf8) << 7) | ((info->cmap.green[bg_idx] & 0xf8) << 2) | diff --git a/drivers/video/softcursor.c b/drivers/video/softcursor.c index a6c5ca88d6b0..229c4bc35079 100644 --- a/drivers/video/softcursor.c +++ b/drivers/video/softcursor.c @@ -58,13 +58,10 @@ int soft_cursor(struct fb_info *info, struct fb_cursor *cursor) } else memcpy(src, image->data, dsize); - fb_sysmove_buf_aligned(info, &info->pixmap, dst, d_pitch, src, - s_pitch, image->height); - + fb_pad_aligned_buffer(dst, d_pitch, src, s_pitch, image->height); image->data = dst; info->fbops->fb_imageblit(info, image); kfree(src); - return 0; } diff --git a/include/linux/fb.h b/include/linux/fb.h index e14942805dbd..bc24beeed971 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -816,12 +816,9 @@ extern int unregister_framebuffer(struct fb_info *fb_info); extern int fb_prepare_logo(struct fb_info *fb_info); extern int fb_show_logo(struct fb_info *fb_info); extern char* fb_get_buffer_offset(struct fb_info *info, struct fb_pixmap *buf, u32 size); -extern void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 idx, +extern void fb_pad_unaligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 idx, u32 height, u32 shift_high, u32 shift_low, u32 mod); -extern void fb_sysmove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf, - u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, - u32 height); +extern void fb_pad_aligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, u32 height); extern void fb_set_suspend(struct fb_info *info, int state); extern int fb_get_color_depth(struct fb_var_screeninfo *var); extern int fb_get_options(char *name, char **option); -- cgit v1.2.3 From 06d91a5fe0b50c9060e70bdf7786f8a3c66249db Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:12 -0700 Subject: [PATCH] md: improve locking on 'safemode' and move superblock writes When md marks the superblock dirty before a write, it calls generic_make_request (to write the superblock) from within generic_make_request (to write the first dirty block), which could cause problems later. With this patch, the superblock write is always done by the helper thread, and write request are delayed until that write completes. Also, the locking around marking the array dirty and writing the superblock is improved to avoid possible races. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 73 ++++++++++++++++++++++++++++++++++++++--------- drivers/md/raid1.c | 4 ++- drivers/md/raid10.c | 5 +++- drivers/md/raid5.c | 6 ++-- drivers/md/raid6main.c | 6 ++-- include/linux/raid/md.h | 2 +- include/linux/raid/md_k.h | 7 +++++ 7 files changed, 82 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/md.c b/drivers/md/md.c index c842e34d850e..177d2a7d7cea 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -218,6 +218,8 @@ static mddev_t * mddev_find(dev_t unit) INIT_LIST_HEAD(&new->all_mddevs); init_timer(&new->safemode_timer); atomic_set(&new->active, 1); + bio_list_init(&new->write_list); + spin_lock_init(&new->write_lock); new->queue = blk_alloc_queue(GFP_KERNEL); if (!new->queue) { @@ -1251,9 +1253,11 @@ static void md_update_sb(mddev_t * mddev) int err, count = 100; struct list_head *tmp; mdk_rdev_t *rdev; + int sync_req; - mddev->sb_dirty = 0; repeat: + spin_lock(&mddev->write_lock); + sync_req = mddev->in_sync; mddev->utime = get_seconds(); mddev->events ++; @@ -1272,8 +1276,12 @@ repeat: * do not write anything to disk if using * nonpersistent superblocks */ - if (!mddev->persistent) + if (!mddev->persistent) { + mddev->sb_dirty = 0; + spin_unlock(&mddev->write_lock); return; + } + spin_unlock(&mddev->write_lock); dprintk(KERN_INFO "md: updating %s RAID superblock on device (in sync %d)\n", @@ -1304,6 +1312,15 @@ repeat: printk(KERN_ERR \ "md: excessive errors occurred during superblock update, exiting\n"); } + spin_lock(&mddev->write_lock); + if (mddev->in_sync != sync_req) { + /* have to write it out again */ + spin_unlock(&mddev->write_lock); + goto repeat; + } + mddev->sb_dirty = 0; + spin_unlock(&mddev->write_lock); + } /* @@ -3178,19 +3195,31 @@ void md_done_sync(mddev_t *mddev, int blocks, int ok) } -void md_write_start(mddev_t *mddev) +/* md_write_start(mddev, bi) + * If we need to update some array metadata (e.g. 'active' flag + * in superblock) before writing, queue bi for later writing + * and return 0, else return 1 and it will be written now + */ +int md_write_start(mddev_t *mddev, struct bio *bi) { - if (!atomic_read(&mddev->writes_pending)) { - mddev_lock_uninterruptible(mddev); - if (mddev->in_sync) { - mddev->in_sync = 0; - del_timer(&mddev->safemode_timer); - md_update_sb(mddev); - } - atomic_inc(&mddev->writes_pending); - mddev_unlock(mddev); - } else - atomic_inc(&mddev->writes_pending); + if (bio_data_dir(bi) != WRITE) + return 1; + + atomic_inc(&mddev->writes_pending); + spin_lock(&mddev->write_lock); + if (mddev->in_sync == 0 && mddev->sb_dirty == 0) { + spin_unlock(&mddev->write_lock); + return 1; + } + bio_list_add(&mddev->write_list, bi); + + if (mddev->in_sync) { + mddev->in_sync = 0; + mddev->sb_dirty = 1; + } + spin_unlock(&mddev->write_lock); + md_wakeup_thread(mddev->thread); + return 0; } void md_write_end(mddev_t *mddev) @@ -3472,6 +3501,7 @@ void md_check_recovery(mddev_t *mddev) mddev->sb_dirty || test_bit(MD_RECOVERY_NEEDED, &mddev->recovery) || test_bit(MD_RECOVERY_DONE, &mddev->recovery) || + mddev->write_list.head || (mddev->safemode == 1) || (mddev->safemode == 2 && ! atomic_read(&mddev->writes_pending) && !mddev->in_sync && mddev->recovery_cp == MaxSector) @@ -3480,7 +3510,9 @@ void md_check_recovery(mddev_t *mddev) if (mddev_trylock(mddev)==0) { int spares =0; + struct bio *blist; + spin_lock(&mddev->write_lock); if (mddev->safemode && !atomic_read(&mddev->writes_pending) && !mddev->in_sync && mddev->recovery_cp == MaxSector) { mddev->in_sync = 1; @@ -3488,9 +3520,22 @@ void md_check_recovery(mddev_t *mddev) } if (mddev->safemode == 1) mddev->safemode = 0; + blist = bio_list_get(&mddev->write_list); + spin_unlock(&mddev->write_lock); if (mddev->sb_dirty) md_update_sb(mddev); + + while (blist) { + struct bio *b = blist; + blist = blist->bi_next; + b->bi_next = NULL; + generic_make_request(b); + /* we already counted this, so need to un-count */ + md_write_end(mddev); + } + + if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) && !test_bit(MD_RECOVERY_DONE, &mddev->recovery)) { /* resync/recovery still happening */ diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index b34ad56362df..3f1280bbaf39 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -530,6 +530,8 @@ static int make_request(request_queue_t *q, struct bio * bio) * thread has put up a bar for new requests. * Continue immediately if no resync is active currently. */ + if (md_write_start(mddev, bio)==0) + return 0; spin_lock_irq(&conf->resync_lock); wait_event_lock_irq(conf->wait_resume, !conf->barrier, conf->resync_lock, ); conf->nr_pending++; @@ -611,7 +613,7 @@ static int make_request(request_queue_t *q, struct bio * bio) rcu_read_unlock(); atomic_set(&r1_bio->remaining, 1); - md_write_start(mddev); + for (i = 0; i < disks; i++) { struct bio *mbio; if (!r1_bio->bios[i]) diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 9ae21504db8a..bfc9f52f0ecf 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -700,6 +700,9 @@ static int make_request(request_queue_t *q, struct bio * bio) return 0; } + if (md_write_start(mddev, bio) == 0) + return 0; + /* * Register the new request and wait if the reconstruction * thread has put up a bar for new requests. @@ -774,7 +777,7 @@ static int make_request(request_queue_t *q, struct bio * bio) rcu_read_unlock(); atomic_set(&r10_bio->remaining, 1); - md_write_start(mddev); + for (i = 0; i < conf->copies; i++) { struct bio *mbio; int d = r10_bio->devs[i].devnum; diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 63b1c59d36ff..677ce49078da 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -1411,6 +1411,9 @@ static int make_request (request_queue_t *q, struct bio * bi) sector_t logical_sector, last_sector; struct stripe_head *sh; + if (md_write_start(mddev, bi)==0) + return 0; + if (bio_data_dir(bi)==WRITE) { disk_stat_inc(mddev->gendisk, writes); disk_stat_add(mddev->gendisk, write_sectors, bio_sectors(bi)); @@ -1423,8 +1426,7 @@ static int make_request (request_queue_t *q, struct bio * bi) last_sector = bi->bi_sector + (bi->bi_size>>9); bi->bi_next = NULL; bi->bi_phys_segments = 1; /* over-loaded to count active stripes */ - if ( bio_data_dir(bi) == WRITE ) - md_write_start(mddev); + for (;logical_sector < last_sector; logical_sector += STRIPE_SECTORS) { DEFINE_WAIT(w); diff --git a/drivers/md/raid6main.c b/drivers/md/raid6main.c index 9d0e0e42a3be..fede16c4e8f3 100644 --- a/drivers/md/raid6main.c +++ b/drivers/md/raid6main.c @@ -1570,6 +1570,9 @@ static int make_request (request_queue_t *q, struct bio * bi) sector_t logical_sector, last_sector; struct stripe_head *sh; + if (md_write_start(mddev, bi)==0) + return 0; + if (bio_data_dir(bi)==WRITE) { disk_stat_inc(mddev->gendisk, writes); disk_stat_add(mddev->gendisk, write_sectors, bio_sectors(bi)); @@ -1583,8 +1586,7 @@ static int make_request (request_queue_t *q, struct bio * bi) bi->bi_next = NULL; bi->bi_phys_segments = 1; /* over-loaded to count active stripes */ - if ( bio_data_dir(bi) == WRITE ) - md_write_start(mddev); + for (;logical_sector < last_sector; logical_sector += STRIPE_SECTORS) { DEFINE_WAIT(w); diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h index a6a67d102bfa..cfde8f497d6d 100644 --- a/include/linux/raid/md.h +++ b/include/linux/raid/md.h @@ -69,7 +69,7 @@ extern mdk_thread_t * md_register_thread (void (*run) (mddev_t *mddev), extern void md_unregister_thread (mdk_thread_t *thread); extern void md_wakeup_thread(mdk_thread_t *thread); extern void md_check_recovery(mddev_t *mddev); -extern void md_write_start(mddev_t *mddev); +extern int md_write_start(mddev_t *mddev, struct bio *bi); extern void md_write_end(mddev_t *mddev); extern void md_handle_safemode(mddev_t *mddev); extern void md_done_sync(mddev_t *mddev, int blocks, int ok); diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h index c9a0d4013be7..d92db54255a3 100644 --- a/include/linux/raid/md_k.h +++ b/include/linux/raid/md_k.h @@ -15,6 +15,9 @@ #ifndef _MD_K_H #define _MD_K_H +/* and dm-bio-list.h is not under include/linux because.... ??? */ +#include "../../../drivers/md/dm-bio-list.h" + #define MD_RESERVED 0UL #define LINEAR 1UL #define RAID0 2UL @@ -252,6 +255,10 @@ struct mddev_s atomic_t recovery_active; /* blocks scheduled, but not written */ wait_queue_head_t recovery_wait; sector_t recovery_cp; + + spinlock_t write_lock; + struct bio_list write_list; + unsigned int safemode; /* if set, update "clean" superblock * when no writes pending. */ -- cgit v1.2.3 From 57afd89f98a990747445f01c458ecae64263b2f8 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:13 -0700 Subject: [PATCH] md: improve the interface to sync_request 1/ change the return value (which is number-of-sectors synced) from 'int' to 'sector_t'. The number of sectors is usually easily small enough to fit in an int, but if resync needs to abort, it may want to return the total number of remaining sectors, which could be large. Also errors cannot be returned as negative numbers now, so use 0 instead 2/ Add a 'skipped' return parameter to allow the array to report that it skipped the sectors. This allows md to take this into account in the speed calculations. Currently there is no important skipping, but the bitmap-based-resync that is coming will use this. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 35 +++++++++++++++++++++++------------ drivers/md/raid1.c | 8 ++++---- drivers/md/raid10.c | 19 ++++++++++++------- drivers/md/raid5.c | 6 +++--- drivers/md/raid6main.c | 6 +++--- include/linux/raid/md_k.h | 2 +- 6 files changed, 46 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/md.c b/drivers/md/md.c index 177d2a7d7cea..fa608a1a5c20 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -3241,12 +3241,13 @@ static void md_do_sync(mddev_t *mddev) mddev_t *mddev2; unsigned int currspeed = 0, window; - sector_t max_sectors,j; + sector_t max_sectors,j, io_sectors; unsigned long mark[SYNC_MARKS]; sector_t mark_cnt[SYNC_MARKS]; int last_mark,m; struct list_head *tmp; sector_t last_check; + int skipped = 0; /* just incase thread restarts... */ if (test_bit(MD_RECOVERY_DONE, &mddev->recovery)) @@ -3312,7 +3313,7 @@ static void md_do_sync(mddev_t *mddev) if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) /* resync follows the size requested by the personality, - * which default to physical size, but can be virtual size + * which defaults to physical size, but can be virtual size */ max_sectors = mddev->resync_max_sectors; else @@ -3331,9 +3332,10 @@ static void md_do_sync(mddev_t *mddev) j = mddev->recovery_cp; else j = 0; + io_sectors = 0; for (m = 0; m < SYNC_MARKS; m++) { mark[m] = jiffies; - mark_cnt[m] = j; + mark_cnt[m] = io_sectors; } last_mark = 0; mddev->resync_mark = mark[last_mark]; @@ -3358,21 +3360,29 @@ static void md_do_sync(mddev_t *mddev) } while (j < max_sectors) { - int sectors; + sector_t sectors; - sectors = mddev->pers->sync_request(mddev, j, currspeed < sysctl_speed_limit_min); - if (sectors < 0) { + skipped = 0; + sectors = mddev->pers->sync_request(mddev, j, &skipped, + currspeed < sysctl_speed_limit_min); + if (sectors == 0) { set_bit(MD_RECOVERY_ERR, &mddev->recovery); goto out; } - atomic_add(sectors, &mddev->recovery_active); + + if (!skipped) { /* actual IO requested */ + io_sectors += sectors; + atomic_add(sectors, &mddev->recovery_active); + } + j += sectors; if (j>1) mddev->curr_resync = j; - if (last_check + window > j || j == max_sectors) + + if (last_check + window > io_sectors || j == max_sectors) continue; - last_check = j; + last_check = io_sectors; if (test_bit(MD_RECOVERY_INTR, &mddev->recovery) || test_bit(MD_RECOVERY_ERR, &mddev->recovery)) @@ -3386,7 +3396,7 @@ static void md_do_sync(mddev_t *mddev) mddev->resync_mark = mark[next]; mddev->resync_mark_cnt = mark_cnt[next]; mark[next] = jiffies; - mark_cnt[next] = j - atomic_read(&mddev->recovery_active); + mark_cnt[next] = io_sectors - atomic_read(&mddev->recovery_active); last_mark = next; } @@ -3413,7 +3423,8 @@ static void md_do_sync(mddev_t *mddev) mddev->queue->unplug_fn(mddev->queue); cond_resched(); - currspeed = ((unsigned long)(j-mddev->resync_mark_cnt))/2/((jiffies-mddev->resync_mark)/HZ +1) +1; + currspeed = ((unsigned long)(io_sectors-mddev->resync_mark_cnt))/2 + /((jiffies-mddev->resync_mark)/HZ +1) +1; if (currspeed > sysctl_speed_limit_min) { if ((currspeed > sysctl_speed_limit_max) || @@ -3433,7 +3444,7 @@ static void md_do_sync(mddev_t *mddev) wait_event(mddev->recovery_wait, !atomic_read(&mddev->recovery_active)); /* tell personality that we are finished */ - mddev->pers->sync_request(mddev, max_sectors, 1); + mddev->pers->sync_request(mddev, max_sectors, &skipped, 1); if (!test_bit(MD_RECOVERY_ERR, &mddev->recovery) && mddev->curr_resync > 2 && diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 3f1280bbaf39..3c5c916cb09e 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -1010,7 +1010,7 @@ static int init_resync(conf_t *conf) * that can be installed to exclude normal IO requests. */ -static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster) +static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster) { conf_t *conf = mddev_to_conf(mddev); mirror_info_t *mirror; @@ -1023,7 +1023,7 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster) if (!conf->r1buf_pool) if (init_resync(conf)) - return -ENOMEM; + return 0; max_sector = mddev->size << 1; if (sector_nr >= max_sector) { @@ -1107,8 +1107,8 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster) /* There is nowhere to write, so all non-sync * drives must be failed - so we are finished */ - int rv = max_sector - sector_nr; - md_done_sync(mddev, rv, 1); + sector_t rv = max_sector - sector_nr; + *skipped = 1; put_buf(r1_bio); rdev_dec_pending(conf->mirrors[disk].rdev, mddev); return rv; diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index bfc9f52f0ecf..8476515bfdc7 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -1321,7 +1321,7 @@ static int init_resync(conf_t *conf) * */ -static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster) +static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster) { conf_t *conf = mddev_to_conf(mddev); r10bio_t *r10_bio; @@ -1335,7 +1335,7 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster) if (!conf->r10buf_pool) if (init_resync(conf)) - return -ENOMEM; + return 0; skipped: max_sector = mddev->size << 1; @@ -1343,15 +1343,15 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster) max_sector = mddev->resync_max_sectors; if (sector_nr >= max_sector) { close_sync(conf); + *skipped = 1; return sectors_skipped; } if (chunks_skipped >= conf->raid_disks) { /* if there has been nothing to do on any drive, * then there is nothing to do at all.. */ - sector_t sec = max_sector - sector_nr; - md_done_sync(mddev, sec, 1); - return sec + sectors_skipped; + *skipped = 1; + return (max_sector - sector_nr) + sectors_skipped; } /* make sure whole request will fit in a chunk - if chunks @@ -1565,17 +1565,22 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster) } } + if (sectors_skipped) + /* pretend they weren't skipped, it makes + * no important difference in this case + */ + md_done_sync(mddev, sectors_skipped, 1); + return sectors_skipped + nr_sectors; giveup: /* There is nowhere to write, so all non-sync * drives must be failed, so try the next chunk... */ { - int sec = max_sector - sector_nr; + sector_t sec = max_sector - sector_nr; sectors_skipped += sec; chunks_skipped ++; sector_nr = max_sector; - md_done_sync(mddev, sec, 1); goto skipped; } } diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 677ce49078da..1ce3f5aaa984 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -1477,7 +1477,7 @@ static int make_request (request_queue_t *q, struct bio * bi) } /* FIXME go_faster isn't used */ -static int sync_request (mddev_t *mddev, sector_t sector_nr, int go_faster) +static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster) { raid5_conf_t *conf = (raid5_conf_t *) mddev->private; struct stripe_head *sh; @@ -1500,8 +1500,8 @@ static int sync_request (mddev_t *mddev, sector_t sector_nr, int go_faster) * nothing we can do. */ if (mddev->degraded >= 1 && test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) { - int rv = (mddev->size << 1) - sector_nr; - md_done_sync(mddev, rv, 1); + sector_t rv = (mddev->size << 1) - sector_nr; + *skipped = 1; return rv; } diff --git a/drivers/md/raid6main.c b/drivers/md/raid6main.c index fede16c4e8f3..d9c385496dc5 100644 --- a/drivers/md/raid6main.c +++ b/drivers/md/raid6main.c @@ -1636,7 +1636,7 @@ static int make_request (request_queue_t *q, struct bio * bi) } /* FIXME go_faster isn't used */ -static int sync_request (mddev_t *mddev, sector_t sector_nr, int go_faster) +static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster) { raid6_conf_t *conf = (raid6_conf_t *) mddev->private; struct stripe_head *sh; @@ -1659,8 +1659,8 @@ static int sync_request (mddev_t *mddev, sector_t sector_nr, int go_faster) * nothing we can do. */ if (mddev->degraded >= 2 && test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) { - int rv = (mddev->size << 1) - sector_nr; - md_done_sync(mddev, rv, 1); + sector_t rv = (mddev->size << 1) - sector_nr; + *skipped = 1; return rv; } diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h index d92db54255a3..bce0032decff 100644 --- a/include/linux/raid/md_k.h +++ b/include/linux/raid/md_k.h @@ -298,7 +298,7 @@ struct mdk_personality_s int (*hot_add_disk) (mddev_t *mddev, mdk_rdev_t *rdev); int (*hot_remove_disk) (mddev_t *mddev, int number); int (*spare_active) (mddev_t *mddev); - int (*sync_request)(mddev_t *mddev, sector_t sector_nr, int go_faster); + sector_t (*sync_request)(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster); int (*resize) (mddev_t *mddev, sector_t sectors); int (*reshape) (mddev_t *mddev, int raid_disks); int (*reconfig) (mddev_t *mddev, int layout, int chunk_size); -- cgit v1.2.3 From 32a7627cf3a35396a8e834faf34e38ae9f3b1309 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:14 -0700 Subject: [PATCH] md: optimised resync using Bitmap based intent logging With this patch, the intent to write to some block in the array can be logged to a bitmap file. Each bit represents some number of sectors and is set before any update happens, and only cleared when all writes relating to all sectors are complete. After an unclean shutdown, information in this bitmap can be used to optimise resync - only sectors which could be out-of-sync need to be updated. Also if a drive is removed and then added back into an array, the recovery can make use of the bitmap to optimise reconstruction. This is not implemented in this patch. Currently the bitmap is stored in a file which must (obviously) be stored on a separate device. The patch only provided infrastructure. It does not update any personalities to bitmap intent logging. Md arrays can still be used with no bitmap file. This patch has minimal impact on such arrays. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/Makefile | 3 +- drivers/md/bitmap.c | 1519 +++++++++++++++++++++++++++++++++++++++++++ drivers/md/md.c | 172 ++++- include/linux/raid/bitmap.h | 280 ++++++++ include/linux/raid/md_k.h | 4 + include/linux/raid/md_u.h | 7 + 6 files changed, 1970 insertions(+), 15 deletions(-) create mode 100644 drivers/md/bitmap.c create mode 100644 include/linux/raid/bitmap.h (limited to 'include/linux') diff --git a/drivers/md/Makefile b/drivers/md/Makefile index 90de9c146a5f..d3efedf6a6ad 100644 --- a/drivers/md/Makefile +++ b/drivers/md/Makefile @@ -7,6 +7,7 @@ dm-mod-objs := dm.o dm-table.o dm-target.o dm-linear.o dm-stripe.o \ dm-multipath-objs := dm-hw-handler.o dm-path-selector.o dm-mpath.o dm-snapshot-objs := dm-snap.o dm-exception-store.o dm-mirror-objs := dm-log.o dm-raid1.o +md-mod-objs := md.o bitmap.o raid6-objs := raid6main.o raid6algos.o raid6recov.o raid6tables.o \ raid6int1.o raid6int2.o raid6int4.o \ raid6int8.o raid6int16.o raid6int32.o \ @@ -28,7 +29,7 @@ obj-$(CONFIG_MD_RAID5) += raid5.o xor.o obj-$(CONFIG_MD_RAID6) += raid6.o xor.o obj-$(CONFIG_MD_MULTIPATH) += multipath.o obj-$(CONFIG_MD_FAULTY) += faulty.o -obj-$(CONFIG_BLK_DEV_MD) += md.o +obj-$(CONFIG_BLK_DEV_MD) += md-mod.o obj-$(CONFIG_BLK_DEV_DM) += dm-mod.o obj-$(CONFIG_DM_CRYPT) += dm-crypt.o obj-$(CONFIG_DM_MULTIPATH) += dm-multipath.o dm-round-robin.o diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c new file mode 100644 index 000000000000..34ffc133db05 --- /dev/null +++ b/drivers/md/bitmap.c @@ -0,0 +1,1519 @@ +/* + * bitmap.c two-level bitmap (C) Peter T. Breuer (ptb@ot.uc3m.es) 2003 + * + * bitmap_create - sets up the bitmap structure + * bitmap_destroy - destroys the bitmap structure + * + * additions, Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc.: + * - added disk storage for bitmap + * - changes to allow various bitmap chunk sizes + * - added bitmap daemon (to asynchronously clear bitmap bits from disk) + */ + +/* + * Still to do: + * + * flush after percent set rather than just time based. (maybe both). + * wait if count gets too high, wake when it drops to half. + * allow bitmap to be mirrored with superblock (before or after...) + * allow hot-add to re-instate a current device. + * allow hot-add of bitmap after quiessing device + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* debug macros */ + +#define DEBUG 0 + +#if DEBUG +/* these are for debugging purposes only! */ + +/* define one and only one of these */ +#define INJECT_FAULTS_1 0 /* cause bitmap_alloc_page to fail always */ +#define INJECT_FAULTS_2 0 /* cause bitmap file to be kicked when first bit set*/ +#define INJECT_FAULTS_3 0 /* treat bitmap file as kicked at init time */ +#define INJECT_FAULTS_4 0 /* undef */ +#define INJECT_FAULTS_5 0 /* undef */ +#define INJECT_FAULTS_6 0 + +/* if these are defined, the driver will fail! debug only */ +#define INJECT_FATAL_FAULT_1 0 /* fail kmalloc, causing bitmap_create to fail */ +#define INJECT_FATAL_FAULT_2 0 /* undef */ +#define INJECT_FATAL_FAULT_3 0 /* undef */ +#endif + +//#define DPRINTK PRINTK /* set this NULL to avoid verbose debug output */ +#define DPRINTK(x...) do { } while(0) + +#ifndef PRINTK +# if DEBUG > 0 +# define PRINTK(x...) printk(KERN_DEBUG x) +# else +# define PRINTK(x...) +# endif +#endif + +static inline char * bmname(struct bitmap *bitmap) +{ + return bitmap->mddev ? mdname(bitmap->mddev) : "mdX"; +} + + +/* + * test if the bitmap is active + */ +int bitmap_active(struct bitmap *bitmap) +{ + unsigned long flags; + int res = 0; + + if (!bitmap) + return res; + spin_lock_irqsave(&bitmap->lock, flags); + res = bitmap->flags & BITMAP_ACTIVE; + spin_unlock_irqrestore(&bitmap->lock, flags); + return res; +} + +#define WRITE_POOL_SIZE 256 +/* mempool for queueing pending writes on the bitmap file */ +static void *write_pool_alloc(unsigned int gfp_flags, void *data) +{ + return kmalloc(sizeof(struct page_list), gfp_flags); +} + +static void write_pool_free(void *ptr, void *data) +{ + kfree(ptr); +} + +/* + * just a placeholder - calls kmalloc for bitmap pages + */ +static unsigned char *bitmap_alloc_page(struct bitmap *bitmap) +{ + unsigned char *page; + +#if INJECT_FAULTS_1 + page = NULL; +#else + page = kmalloc(PAGE_SIZE, GFP_NOIO); +#endif + if (!page) + printk("%s: bitmap_alloc_page FAILED\n", bmname(bitmap)); + else + printk("%s: bitmap_alloc_page: allocated page at %p\n", + bmname(bitmap), page); + return page; +} + +/* + * for now just a placeholder -- just calls kfree for bitmap pages + */ +static void bitmap_free_page(struct bitmap *bitmap, unsigned char *page) +{ + PRINTK("%s: bitmap_free_page: free page %p\n", bmname(bitmap), page); + kfree(page); +} + +/* + * check a page and, if necessary, allocate it (or hijack it if the alloc fails) + * + * 1) check to see if this page is allocated, if it's not then try to alloc + * 2) if the alloc fails, set the page's hijacked flag so we'll use the + * page pointer directly as a counter + * + * if we find our page, we increment the page's refcount so that it stays + * allocated while we're using it + */ +static int bitmap_checkpage(struct bitmap *bitmap, unsigned long page, int create) +{ + unsigned char *mappage; + + if (page >= bitmap->pages) { + printk(KERN_ALERT + "%s: invalid bitmap page request: %lu (> %lu)\n", + bmname(bitmap), page, bitmap->pages-1); + return -EINVAL; + } + + + if (bitmap->bp[page].hijacked) /* it's hijacked, don't try to alloc */ + return 0; + + if (bitmap->bp[page].map) /* page is already allocated, just return */ + return 0; + + if (!create) + return -ENOENT; + + spin_unlock_irq(&bitmap->lock); + + /* this page has not been allocated yet */ + + if ((mappage = bitmap_alloc_page(bitmap)) == NULL) { + PRINTK("%s: bitmap map page allocation failed, hijacking\n", + bmname(bitmap)); + /* failed - set the hijacked flag so that we can use the + * pointer as a counter */ + spin_lock_irq(&bitmap->lock); + if (!bitmap->bp[page].map) + bitmap->bp[page].hijacked = 1; + goto out; + } + + /* got a page */ + + spin_lock_irq(&bitmap->lock); + + /* recheck the page */ + + if (bitmap->bp[page].map || bitmap->bp[page].hijacked) { + /* somebody beat us to getting the page */ + bitmap_free_page(bitmap, mappage); + return 0; + } + + /* no page was in place and we have one, so install it */ + + memset(mappage, 0, PAGE_SIZE); + bitmap->bp[page].map = mappage; + bitmap->missing_pages--; +out: + return 0; +} + + +/* if page is completely empty, put it back on the free list, or dealloc it */ +/* if page was hijacked, unmark the flag so it might get alloced next time */ +/* Note: lock should be held when calling this */ +static inline void bitmap_checkfree(struct bitmap *bitmap, unsigned long page) +{ + char *ptr; + + if (bitmap->bp[page].count) /* page is still busy */ + return; + + /* page is no longer in use, it can be released */ + + if (bitmap->bp[page].hijacked) { /* page was hijacked, undo this now */ + bitmap->bp[page].hijacked = 0; + bitmap->bp[page].map = NULL; + return; + } + + /* normal case, free the page */ + +#if 0 +/* actually ... let's not. We will probably need the page again exactly when + * memory is tight and we are flusing to disk + */ + return; +#else + ptr = bitmap->bp[page].map; + bitmap->bp[page].map = NULL; + bitmap->missing_pages++; + bitmap_free_page(bitmap, ptr); + return; +#endif +} + + +/* + * bitmap file handling - read and write the bitmap file and its superblock + */ + +/* copy the pathname of a file to a buffer */ +char *file_path(struct file *file, char *buf, int count) +{ + struct dentry *d; + struct vfsmount *v; + + if (!buf) + return NULL; + + d = file->f_dentry; + v = file->f_vfsmnt; + + buf = d_path(d, v, buf, count); + + return IS_ERR(buf) ? NULL : buf; +} + +/* + * basic page I/O operations + */ + +/* + * write out a page + */ +static int write_page(struct page *page, int wait) +{ + int ret = -ENOMEM; + + lock_page(page); + + if (page->mapping == NULL) + goto unlock_out; + else if (i_size_read(page->mapping->host) < page->index << PAGE_SHIFT) { + ret = -ENOENT; + goto unlock_out; + } + + ret = page->mapping->a_ops->prepare_write(NULL, page, 0, PAGE_SIZE); + if (!ret) + ret = page->mapping->a_ops->commit_write(NULL, page, 0, + PAGE_SIZE); + if (ret) { +unlock_out: + unlock_page(page); + return ret; + } + + set_page_dirty(page); /* force it to be written out */ + return write_one_page(page, wait); +} + +/* read a page from a file, pinning it into cache, and return bytes_read */ +static struct page *read_page(struct file *file, unsigned long index, + unsigned long *bytes_read) +{ + struct inode *inode = file->f_mapping->host; + struct page *page = NULL; + loff_t isize = i_size_read(inode); + unsigned long end_index = isize >> PAGE_CACHE_SHIFT; + + PRINTK("read bitmap file (%dB @ %Lu)\n", (int)PAGE_CACHE_SIZE, + (unsigned long long)index << PAGE_CACHE_SHIFT); + + page = read_cache_page(inode->i_mapping, index, + (filler_t *)inode->i_mapping->a_ops->readpage, file); + if (IS_ERR(page)) + goto out; + wait_on_page_locked(page); + if (!PageUptodate(page) || PageError(page)) { + page_cache_release(page); + page = ERR_PTR(-EIO); + goto out; + } + + if (index > end_index) /* we have read beyond EOF */ + *bytes_read = 0; + else if (index == end_index) /* possible short read */ + *bytes_read = isize & ~PAGE_CACHE_MASK; + else + *bytes_read = PAGE_CACHE_SIZE; /* got a full page */ +out: + if (IS_ERR(page)) + printk(KERN_ALERT "md: bitmap read error: (%dB @ %Lu): %ld\n", + (int)PAGE_CACHE_SIZE, + (unsigned long long)index << PAGE_CACHE_SHIFT, + PTR_ERR(page)); + return page; +} + +/* + * bitmap file superblock operations + */ + +/* update the event counter and sync the superblock to disk */ +int bitmap_update_sb(struct bitmap *bitmap) +{ + bitmap_super_t *sb; + unsigned long flags; + + if (!bitmap || !bitmap->mddev) /* no bitmap for this array */ + return 0; + spin_lock_irqsave(&bitmap->lock, flags); + if (!bitmap->sb_page) { /* no superblock */ + spin_unlock_irqrestore(&bitmap->lock, flags); + return 0; + } + page_cache_get(bitmap->sb_page); + spin_unlock_irqrestore(&bitmap->lock, flags); + sb = (bitmap_super_t *)kmap(bitmap->sb_page); + sb->events = cpu_to_le64(bitmap->mddev->events); + if (!bitmap->mddev->degraded) + sb->events_cleared = cpu_to_le64(bitmap->mddev->events); + kunmap(bitmap->sb_page); + write_page(bitmap->sb_page, 0); + return 0; +} + +/* print out the bitmap file superblock */ +void bitmap_print_sb(struct bitmap *bitmap) +{ + bitmap_super_t *sb; + + if (!bitmap || !bitmap->sb_page) + return; + sb = (bitmap_super_t *)kmap(bitmap->sb_page); + printk(KERN_DEBUG "%s: bitmap file superblock:\n", bmname(bitmap)); + printk(KERN_DEBUG " magic: %08x\n", le32_to_cpu(sb->magic)); + printk(KERN_DEBUG " version: %d\n", le32_to_cpu(sb->version)); + printk(KERN_DEBUG " uuid: %08x.%08x.%08x.%08x\n", + *(__u32 *)(sb->uuid+0), + *(__u32 *)(sb->uuid+4), + *(__u32 *)(sb->uuid+8), + *(__u32 *)(sb->uuid+12)); + printk(KERN_DEBUG " events: %llu\n", + (unsigned long long) le64_to_cpu(sb->events)); + printk(KERN_DEBUG "events_clred: %llu\n", + (unsigned long long) le64_to_cpu(sb->events_cleared)); + printk(KERN_DEBUG " state: %08x\n", le32_to_cpu(sb->state)); + printk(KERN_DEBUG " chunksize: %d B\n", le32_to_cpu(sb->chunksize)); + printk(KERN_DEBUG "daemon sleep: %ds\n", le32_to_cpu(sb->daemon_sleep)); + printk(KERN_DEBUG " sync size: %llu KB\n", le64_to_cpu(sb->sync_size)); + kunmap(bitmap->sb_page); +} + +/* read the superblock from the bitmap file and initialize some bitmap fields */ +static int bitmap_read_sb(struct bitmap *bitmap) +{ + char *reason = NULL; + bitmap_super_t *sb; + unsigned long chunksize, daemon_sleep; + unsigned long bytes_read; + unsigned long long events; + int err = -EINVAL; + + /* page 0 is the superblock, read it... */ + bitmap->sb_page = read_page(bitmap->file, 0, &bytes_read); + if (IS_ERR(bitmap->sb_page)) { + err = PTR_ERR(bitmap->sb_page); + bitmap->sb_page = NULL; + return err; + } + + sb = (bitmap_super_t *)kmap(bitmap->sb_page); + + if (bytes_read < sizeof(*sb)) { /* short read */ + printk(KERN_INFO "%s: bitmap file superblock truncated\n", + bmname(bitmap)); + err = -ENOSPC; + goto out; + } + + chunksize = le32_to_cpu(sb->chunksize); + daemon_sleep = le32_to_cpu(sb->daemon_sleep); + + /* verify that the bitmap-specific fields are valid */ + if (sb->magic != cpu_to_le32(BITMAP_MAGIC)) + reason = "bad magic"; + else if (sb->version != cpu_to_le32(BITMAP_MAJOR)) + reason = "unrecognized superblock version"; + else if (chunksize < 512 || chunksize > (1024 * 1024 * 4)) + reason = "bitmap chunksize out of range (512B - 4MB)"; + else if ((1 << ffz(~chunksize)) != chunksize) + reason = "bitmap chunksize not a power of 2"; + else if (daemon_sleep < 1 || daemon_sleep > 15) + reason = "daemon sleep period out of range"; + if (reason) { + printk(KERN_INFO "%s: invalid bitmap file superblock: %s\n", + bmname(bitmap), reason); + goto out; + } + + /* keep the array size field of the bitmap superblock up to date */ + sb->sync_size = cpu_to_le64(bitmap->mddev->resync_max_sectors); + + if (!bitmap->mddev->persistent) + goto success; + + /* + * if we have a persistent array superblock, compare the + * bitmap's UUID and event counter to the mddev's + */ + if (memcmp(sb->uuid, bitmap->mddev->uuid, 16)) { + printk(KERN_INFO "%s: bitmap superblock UUID mismatch\n", + bmname(bitmap)); + goto out; + } + events = le64_to_cpu(sb->events); + if (events < bitmap->mddev->events) { + printk(KERN_INFO "%s: bitmap file is out of date (%llu < %llu) " + "-- forcing full recovery\n", bmname(bitmap), events, + (unsigned long long) bitmap->mddev->events); + sb->state |= BITMAP_STALE; + } +success: + /* assign fields using values from superblock */ + bitmap->chunksize = chunksize; + bitmap->daemon_sleep = daemon_sleep; + bitmap->flags |= sb->state; + bitmap->events_cleared = le64_to_cpu(sb->events_cleared); + err = 0; +out: + kunmap(bitmap->sb_page); + if (err) + bitmap_print_sb(bitmap); + return err; +} + +enum bitmap_mask_op { + MASK_SET, + MASK_UNSET +}; + +/* record the state of the bitmap in the superblock */ +static void bitmap_mask_state(struct bitmap *bitmap, enum bitmap_state bits, + enum bitmap_mask_op op) +{ + bitmap_super_t *sb; + unsigned long flags; + + spin_lock_irqsave(&bitmap->lock, flags); + if (!bitmap || !bitmap->sb_page) { /* can't set the state */ + spin_unlock_irqrestore(&bitmap->lock, flags); + return; + } + page_cache_get(bitmap->sb_page); + spin_unlock_irqrestore(&bitmap->lock, flags); + sb = (bitmap_super_t *)kmap(bitmap->sb_page); + switch (op) { + case MASK_SET: sb->state |= bits; + break; + case MASK_UNSET: sb->state &= ~bits; + break; + default: BUG(); + } + kunmap(bitmap->sb_page); + page_cache_release(bitmap->sb_page); +} + +/* + * general bitmap file operations + */ + +/* calculate the index of the page that contains this bit */ +static inline unsigned long file_page_index(unsigned long chunk) +{ + return CHUNK_BIT_OFFSET(chunk) >> PAGE_BIT_SHIFT; +} + +/* calculate the (bit) offset of this bit within a page */ +static inline unsigned long file_page_offset(unsigned long chunk) +{ + return CHUNK_BIT_OFFSET(chunk) & (PAGE_BITS - 1); +} + +/* + * return a pointer to the page in the filemap that contains the given bit + * + * this lookup is complicated by the fact that the bitmap sb might be exactly + * 1 page (e.g., x86) or less than 1 page -- so the bitmap might start on page + * 0 or page 1 + */ +static inline struct page *filemap_get_page(struct bitmap *bitmap, + unsigned long chunk) +{ + return bitmap->filemap[file_page_index(chunk) - file_page_index(0)]; +} + + +static void bitmap_file_unmap(struct bitmap *bitmap) +{ + struct page **map, *sb_page; + unsigned long *attr; + int pages; + unsigned long flags; + + spin_lock_irqsave(&bitmap->lock, flags); + map = bitmap->filemap; + bitmap->filemap = NULL; + attr = bitmap->filemap_attr; + bitmap->filemap_attr = NULL; + pages = bitmap->file_pages; + bitmap->file_pages = 0; + sb_page = bitmap->sb_page; + bitmap->sb_page = NULL; + spin_unlock_irqrestore(&bitmap->lock, flags); + + while (pages--) + if (map[pages]->index != 0) /* 0 is sb_page, release it below */ + page_cache_release(map[pages]); + kfree(map); + kfree(attr); + + if (sb_page) + page_cache_release(sb_page); +} + +static void bitmap_stop_daemons(struct bitmap *bitmap); + +/* dequeue the next item in a page list -- don't call from irq context */ +static struct page_list *dequeue_page(struct bitmap *bitmap, + struct list_head *head) +{ + struct page_list *item = NULL; + + spin_lock(&bitmap->write_lock); + if (list_empty(head)) + goto out; + item = list_entry(head->prev, struct page_list, list); + list_del(head->prev); +out: + spin_unlock(&bitmap->write_lock); + return item; +} + +static void drain_write_queues(struct bitmap *bitmap) +{ + struct list_head *queues[] = { &bitmap->complete_pages, NULL }; + struct list_head *head; + struct page_list *item; + int i; + + for (i = 0; queues[i]; i++) { + head = queues[i]; + while ((item = dequeue_page(bitmap, head))) { + page_cache_release(item->page); + mempool_free(item, bitmap->write_pool); + } + } + + spin_lock(&bitmap->write_lock); + bitmap->writes_pending = 0; /* make sure waiters continue */ + wake_up(&bitmap->write_wait); + spin_unlock(&bitmap->write_lock); +} + +static void bitmap_file_put(struct bitmap *bitmap) +{ + struct file *file; + struct inode *inode; + unsigned long flags; + + spin_lock_irqsave(&bitmap->lock, flags); + file = bitmap->file; + bitmap->file = NULL; + spin_unlock_irqrestore(&bitmap->lock, flags); + + bitmap_stop_daemons(bitmap); + + drain_write_queues(bitmap); + + bitmap_file_unmap(bitmap); + + if (file) { + inode = file->f_mapping->host; + spin_lock(&inode->i_lock); + atomic_set(&inode->i_writecount, 1); /* allow writes again */ + spin_unlock(&inode->i_lock); + fput(file); + } +} + + +/* + * bitmap_file_kick - if an error occurs while manipulating the bitmap file + * then it is no longer reliable, so we stop using it and we mark the file + * as failed in the superblock + */ +static void bitmap_file_kick(struct bitmap *bitmap) +{ + char *path, *ptr = NULL; + + bitmap_mask_state(bitmap, BITMAP_STALE, MASK_SET); + bitmap_update_sb(bitmap); + + path = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (path) + ptr = file_path(bitmap->file, path, PAGE_SIZE); + + printk(KERN_ALERT "%s: kicking failed bitmap file %s from array!\n", + bmname(bitmap), ptr ? ptr : ""); + + kfree(path); + + bitmap_file_put(bitmap); + + return; +} + +enum bitmap_page_attr { + BITMAP_PAGE_DIRTY = 1, // there are set bits that need to be synced + BITMAP_PAGE_CLEAN = 2, // there are bits that might need to be cleared + BITMAP_PAGE_NEEDWRITE=4, // there are cleared bits that need to be synced +}; + +static inline void set_page_attr(struct bitmap *bitmap, struct page *page, + enum bitmap_page_attr attr) +{ + bitmap->filemap_attr[page->index] |= attr; +} + +static inline void clear_page_attr(struct bitmap *bitmap, struct page *page, + enum bitmap_page_attr attr) +{ + bitmap->filemap_attr[page->index] &= ~attr; +} + +static inline unsigned long get_page_attr(struct bitmap *bitmap, struct page *page) +{ + return bitmap->filemap_attr[page->index]; +} + +/* + * bitmap_file_set_bit -- called before performing a write to the md device + * to set (and eventually sync) a particular bit in the bitmap file + * + * we set the bit immediately, then we record the page number so that + * when an unplug occurs, we can flush the dirty pages out to disk + */ +static void bitmap_file_set_bit(struct bitmap *bitmap, sector_t block) +{ + unsigned long bit; + struct page *page; + void *kaddr; + unsigned long chunk = block >> CHUNK_BLOCK_SHIFT(bitmap); + + if (!bitmap->file || !bitmap->filemap) { + return; + } + + page = filemap_get_page(bitmap, chunk); + bit = file_page_offset(chunk); + + + /* make sure the page stays cached until it gets written out */ + if (! (get_page_attr(bitmap, page) & BITMAP_PAGE_DIRTY)) + page_cache_get(page); + + /* set the bit */ + kaddr = kmap_atomic(page, KM_USER0); + set_bit(bit, kaddr); + kunmap_atomic(kaddr, KM_USER0); + PRINTK("set file bit %lu page %lu\n", bit, page->index); + + /* record page number so it gets flushed to disk when unplug occurs */ + set_page_attr(bitmap, page, BITMAP_PAGE_DIRTY); + +} + +/* this gets called when the md device is ready to unplug its underlying + * (slave) device queues -- before we let any writes go down, we need to + * sync the dirty pages of the bitmap file to disk */ +int bitmap_unplug(struct bitmap *bitmap) +{ + unsigned long i, attr, flags; + struct page *page; + int wait = 0; + + if (!bitmap) + return 0; + + /* look at each page to see if there are any set bits that need to be + * flushed out to disk */ + for (i = 0; i < bitmap->file_pages; i++) { + spin_lock_irqsave(&bitmap->lock, flags); + if (!bitmap->file || !bitmap->filemap) { + spin_unlock_irqrestore(&bitmap->lock, flags); + return 0; + } + page = bitmap->filemap[i]; + attr = get_page_attr(bitmap, page); + clear_page_attr(bitmap, page, BITMAP_PAGE_DIRTY); + clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE); + if ((attr & BITMAP_PAGE_DIRTY)) + wait = 1; + spin_unlock_irqrestore(&bitmap->lock, flags); + + if (attr & (BITMAP_PAGE_DIRTY | BITMAP_PAGE_NEEDWRITE)) + write_page(page, 0); + } + if (wait) { /* if any writes were performed, we need to wait on them */ + spin_lock_irq(&bitmap->write_lock); + wait_event_lock_irq(bitmap->write_wait, + bitmap->writes_pending == 0, bitmap->write_lock, + wake_up_process(bitmap->writeback_daemon->tsk)); + spin_unlock_irq(&bitmap->write_lock); + } + return 0; +} + +static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset, + unsigned long sectors, int set); +/* * bitmap_init_from_disk -- called at bitmap_create time to initialize + * the in-memory bitmap from the on-disk bitmap -- also, sets up the + * memory mapping of the bitmap file + * Special cases: + * if there's no bitmap file, or if the bitmap file had been + * previously kicked from the array, we mark all the bits as + * 1's in order to cause a full resync. + */ +static int bitmap_init_from_disk(struct bitmap *bitmap) +{ + unsigned long i, chunks, index, oldindex, bit; + struct page *page = NULL, *oldpage = NULL; + unsigned long num_pages, bit_cnt = 0; + struct file *file; + unsigned long bytes, offset, dummy; + int outofdate; + int ret = -ENOSPC; + + chunks = bitmap->chunks; + file = bitmap->file; + + if (!file) { /* no file, dirty all the in-memory bits */ + printk(KERN_INFO "%s: no bitmap file, doing full recovery\n", + bmname(bitmap)); + bitmap_set_memory_bits(bitmap, 0, + chunks << CHUNK_BLOCK_SHIFT(bitmap), 1); + return 0; + } + +#if INJECT_FAULTS_3 + outofdate = 1; +#else + outofdate = bitmap->flags & BITMAP_STALE; +#endif + if (outofdate) + printk(KERN_INFO "%s: bitmap file is out of date, doing full " + "recovery\n", bmname(bitmap)); + + bytes = (chunks + 7) / 8; + num_pages = (bytes + PAGE_SIZE - 1) / PAGE_SIZE; + if (i_size_read(file->f_mapping->host) < bytes + sizeof(bitmap_super_t)) { + printk(KERN_INFO "%s: bitmap file too short %lu < %lu\n", + bmname(bitmap), + (unsigned long) i_size_read(file->f_mapping->host), + bytes + sizeof(bitmap_super_t)); + goto out; + } + num_pages++; + bitmap->filemap = kmalloc(sizeof(struct page *) * num_pages, GFP_KERNEL); + if (!bitmap->filemap) { + ret = -ENOMEM; + goto out; + } + + bitmap->filemap_attr = kmalloc(sizeof(long) * num_pages, GFP_KERNEL); + if (!bitmap->filemap_attr) { + ret = -ENOMEM; + goto out; + } + + memset(bitmap->filemap_attr, 0, sizeof(long) * num_pages); + + oldindex = ~0L; + + for (i = 0; i < chunks; i++) { + index = file_page_index(i); + bit = file_page_offset(i); + if (index != oldindex) { /* this is a new page, read it in */ + /* unmap the old page, we're done with it */ + if (oldpage != NULL) + kunmap(oldpage); + if (index == 0) { + /* + * if we're here then the superblock page + * contains some bits (PAGE_SIZE != sizeof sb) + * we've already read it in, so just use it + */ + page = bitmap->sb_page; + offset = sizeof(bitmap_super_t); + } else { + page = read_page(file, index, &dummy); + if (IS_ERR(page)) { /* read error */ + ret = PTR_ERR(page); + goto out; + } + offset = 0; + } + oldindex = index; + oldpage = page; + kmap(page); + + if (outofdate) { + /* + * if bitmap is out of date, dirty the + * whole page and write it out + */ + memset(page_address(page) + offset, 0xff, + PAGE_SIZE - offset); + ret = write_page(page, 1); + if (ret) { + kunmap(page); + /* release, page not in filemap yet */ + page_cache_release(page); + goto out; + } + } + + bitmap->filemap[bitmap->file_pages++] = page; + } + if (test_bit(bit, page_address(page))) { + /* if the disk bit is set, set the memory bit */ + bitmap_set_memory_bits(bitmap, + i << CHUNK_BLOCK_SHIFT(bitmap), 1, 1); + bit_cnt++; + } +#if 0 + else + bitmap_set_memory_bits(bitmap, + i << CHUNK_BLOCK_SHIFT(bitmap), 1, 0); +#endif + } + + /* everything went OK */ + ret = 0; + bitmap_mask_state(bitmap, BITMAP_STALE, MASK_UNSET); + + if (page) /* unmap the last page */ + kunmap(page); + + if (bit_cnt) { /* Kick recovery if any bits were set */ + set_bit(MD_RECOVERY_NEEDED, &bitmap->mddev->recovery); + md_wakeup_thread(bitmap->mddev->thread); + } + +out: + printk(KERN_INFO "%s: bitmap initialized from disk: " + "read %lu/%lu pages, set %lu bits, status: %d\n", + bmname(bitmap), bitmap->file_pages, num_pages, bit_cnt, ret); + + return ret; +} + + +static void bitmap_count_page(struct bitmap *bitmap, sector_t offset, int inc) +{ + sector_t chunk = offset >> CHUNK_BLOCK_SHIFT(bitmap); + unsigned long page = chunk >> PAGE_COUNTER_SHIFT; + bitmap->bp[page].count += inc; +/* + if (page == 0) printk("count page 0, offset %llu: %d gives %d\n", + (unsigned long long)offset, inc, bitmap->bp[page].count); +*/ + bitmap_checkfree(bitmap, page); +} +static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap, + sector_t offset, int *blocks, + int create); + +/* + * bitmap daemon -- periodically wakes up to clean bits and flush pages + * out to disk + */ + +int bitmap_daemon_work(struct bitmap *bitmap) +{ + unsigned long bit, j; + unsigned long flags; + struct page *page = NULL, *lastpage = NULL; + int err = 0; + int blocks; + int attr; + + if (bitmap == NULL) + return 0; + if (time_before(jiffies, bitmap->daemon_lastrun + bitmap->daemon_sleep*HZ)) + return 0; + bitmap->daemon_lastrun = jiffies; + + for (j = 0; j < bitmap->chunks; j++) { + bitmap_counter_t *bmc; + spin_lock_irqsave(&bitmap->lock, flags); + if (!bitmap->file || !bitmap->filemap) { + /* error or shutdown */ + spin_unlock_irqrestore(&bitmap->lock, flags); + break; + } + + page = filemap_get_page(bitmap, j); + /* skip this page unless it's marked as needing cleaning */ + if (!((attr=get_page_attr(bitmap, page)) & BITMAP_PAGE_CLEAN)) { + if (attr & BITMAP_PAGE_NEEDWRITE) { + page_cache_get(page); + clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE); + } + spin_unlock_irqrestore(&bitmap->lock, flags); + if (attr & BITMAP_PAGE_NEEDWRITE) { + if (write_page(page, 0)) + bitmap_file_kick(bitmap); + page_cache_release(page); + } + continue; + } + + bit = file_page_offset(j); + + if (page != lastpage) { + /* grab the new page, sync and release the old */ + page_cache_get(page); + if (lastpage != NULL) { + if (get_page_attr(bitmap, lastpage) & BITMAP_PAGE_NEEDWRITE) { + clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE); + spin_unlock_irqrestore(&bitmap->lock, flags); + write_page(lastpage, 0); + } else { + set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE); + spin_unlock_irqrestore(&bitmap->lock, flags); + } + kunmap(lastpage); + page_cache_release(lastpage); + if (err) + bitmap_file_kick(bitmap); + } else + spin_unlock_irqrestore(&bitmap->lock, flags); + lastpage = page; + kmap(page); +/* + printk("bitmap clean at page %lu\n", j); +*/ + spin_lock_irqsave(&bitmap->lock, flags); + clear_page_attr(bitmap, page, BITMAP_PAGE_CLEAN); + } + bmc = bitmap_get_counter(bitmap, j << CHUNK_BLOCK_SHIFT(bitmap), + &blocks, 0); + if (bmc) { +/* + if (j < 100) printk("bitmap: j=%lu, *bmc = 0x%x\n", j, *bmc); +*/ + if (*bmc == 2) { + *bmc=1; /* maybe clear the bit next time */ + set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN); + } else if (*bmc == 1) { + /* we can clear the bit */ + *bmc = 0; + bitmap_count_page(bitmap, j << CHUNK_BLOCK_SHIFT(bitmap), + -1); + + /* clear the bit */ + clear_bit(bit, page_address(page)); + } + } + spin_unlock_irqrestore(&bitmap->lock, flags); + } + + /* now sync the final page */ + if (lastpage != NULL) { + kunmap(lastpage); + spin_lock_irqsave(&bitmap->lock, flags); + if (get_page_attr(bitmap, lastpage) &BITMAP_PAGE_NEEDWRITE) { + clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE); + spin_unlock_irqrestore(&bitmap->lock, flags); + write_page(lastpage, 0); + } else { + set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE); + spin_unlock_irqrestore(&bitmap->lock, flags); + } + + page_cache_release(lastpage); + } + + return err; +} + +static void daemon_exit(struct bitmap *bitmap, mdk_thread_t **daemon) +{ + mdk_thread_t *dmn; + unsigned long flags; + + /* if no one is waiting on us, we'll free the md thread struct + * and exit, otherwise we let the waiter clean things up */ + spin_lock_irqsave(&bitmap->lock, flags); + if ((dmn = *daemon)) { /* no one is waiting, cleanup and exit */ + *daemon = NULL; + spin_unlock_irqrestore(&bitmap->lock, flags); + kfree(dmn); + complete_and_exit(NULL, 0); /* do_exit not exported */ + } + spin_unlock_irqrestore(&bitmap->lock, flags); +} + +static void bitmap_writeback_daemon(mddev_t *mddev) +{ + struct bitmap *bitmap = mddev->bitmap; + struct page *page; + struct page_list *item; + int err = 0; + + while (1) { + PRINTK("%s: bitmap writeback daemon waiting...\n", bmname(bitmap)); + down_interruptible(&bitmap->write_done); + if (signal_pending(current)) { + printk(KERN_INFO + "%s: bitmap writeback daemon got signal, exiting...\n", + bmname(bitmap)); + break; + } + + PRINTK("%s: bitmap writeback daemon woke up...\n", bmname(bitmap)); + /* wait on bitmap page writebacks */ + while ((item = dequeue_page(bitmap, &bitmap->complete_pages))) { + page = item->page; + mempool_free(item, bitmap->write_pool); + PRINTK("wait on page writeback: %p %lu\n", page, bitmap->writes_pending); + wait_on_page_writeback(page); + PRINTK("finished page writeback: %p %lu\n", page, bitmap->writes_pending); + spin_lock(&bitmap->write_lock); + if (!--bitmap->writes_pending) + wake_up(&bitmap->write_wait); + spin_unlock(&bitmap->write_lock); + err = PageError(page); + page_cache_release(page); + if (err) { + printk(KERN_WARNING "%s: bitmap file writeback " + "failed (page %lu): %d\n", + bmname(bitmap), page->index, err); + bitmap_file_kick(bitmap); + goto out; + } + } + } +out: + if (err) { + printk(KERN_INFO "%s: bitmap writeback daemon exiting (%d)\n", + bmname(bitmap), err); + daemon_exit(bitmap, &bitmap->writeback_daemon); + } + return; +} + +static int bitmap_start_daemon(struct bitmap *bitmap, mdk_thread_t **ptr, + void (*func)(mddev_t *), char *name) +{ + mdk_thread_t *daemon; + unsigned long flags; + char namebuf[32]; + + spin_lock_irqsave(&bitmap->lock, flags); + *ptr = NULL; + if (!bitmap->file) /* no need for daemon if there's no backing file */ + goto out_unlock; + + spin_unlock_irqrestore(&bitmap->lock, flags); + +#if INJECT_FATAL_FAULT_2 + daemon = NULL; +#else + sprintf(namebuf, "%%s_%s", name); + daemon = md_register_thread(func, bitmap->mddev, namebuf); +#endif + if (!daemon) { + printk(KERN_ERR "%s: failed to start bitmap daemon\n", + bmname(bitmap)); + return -ECHILD; + } + + spin_lock_irqsave(&bitmap->lock, flags); + *ptr = daemon; + + md_wakeup_thread(daemon); /* start it running */ + + PRINTK("%s: %s daemon (pid %d) started...\n", + bmname(bitmap), name, bitmap->daemon->tsk->pid); +out_unlock: + spin_unlock_irqrestore(&bitmap->lock, flags); + return 0; +} + +static int bitmap_start_daemons(struct bitmap *bitmap) +{ + int err = bitmap_start_daemon(bitmap, &bitmap->writeback_daemon, + bitmap_writeback_daemon, "bitmap_wb"); + return err; +} + +static void bitmap_stop_daemon(struct bitmap *bitmap, mdk_thread_t **ptr) +{ + mdk_thread_t *daemon; + unsigned long flags; + + spin_lock_irqsave(&bitmap->lock, flags); + daemon = *ptr; + *ptr = NULL; + spin_unlock_irqrestore(&bitmap->lock, flags); + if (daemon) + md_unregister_thread(daemon); /* destroy the thread */ +} + +static void bitmap_stop_daemons(struct bitmap *bitmap) +{ + /* the daemons can't stop themselves... they'll just exit instead... */ + if (bitmap->writeback_daemon && + current->pid != bitmap->writeback_daemon->tsk->pid) + bitmap_stop_daemon(bitmap, &bitmap->writeback_daemon); +} + +static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap, + sector_t offset, int *blocks, + int create) +{ + /* If 'create', we might release the lock and reclaim it. + * The lock must have been taken with interrupts enabled. + * If !create, we don't release the lock. + */ + sector_t chunk = offset >> CHUNK_BLOCK_SHIFT(bitmap); + unsigned long page = chunk >> PAGE_COUNTER_SHIFT; + unsigned long pageoff = (chunk & PAGE_COUNTER_MASK) << COUNTER_BYTE_SHIFT; + sector_t csize; + + if (bitmap_checkpage(bitmap, page, create) < 0) { + csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap)); + *blocks = csize - (offset & (csize- 1)); + return NULL; + } + /* now locked ... */ + + if (bitmap->bp[page].hijacked) { /* hijacked pointer */ + /* should we use the first or second counter field + * of the hijacked pointer? */ + int hi = (pageoff > PAGE_COUNTER_MASK); + csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap) + + PAGE_COUNTER_SHIFT - 1); + *blocks = csize - (offset & (csize- 1)); + return &((bitmap_counter_t *) + &bitmap->bp[page].map)[hi]; + } else { /* page is allocated */ + csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap)); + *blocks = csize - (offset & (csize- 1)); + return (bitmap_counter_t *) + &(bitmap->bp[page].map[pageoff]); + } +} + +int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors) +{ + if (!bitmap) return 0; + while (sectors) { + int blocks; + bitmap_counter_t *bmc; + + spin_lock_irq(&bitmap->lock); + bmc = bitmap_get_counter(bitmap, offset, &blocks, 1); + if (!bmc) { + spin_unlock_irq(&bitmap->lock); + return 0; + } + + switch(*bmc) { + case 0: + bitmap_file_set_bit(bitmap, offset); + bitmap_count_page(bitmap,offset, 1); + blk_plug_device(bitmap->mddev->queue); + /* fall through */ + case 1: + *bmc = 2; + } + if ((*bmc & COUNTER_MAX) == COUNTER_MAX) BUG(); + (*bmc)++; + + spin_unlock_irq(&bitmap->lock); + + offset += blocks; + if (sectors > blocks) + sectors -= blocks; + else sectors = 0; + } + return 0; +} + +void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors, + int success) +{ + if (!bitmap) return; + while (sectors) { + int blocks; + unsigned long flags; + bitmap_counter_t *bmc; + + spin_lock_irqsave(&bitmap->lock, flags); + bmc = bitmap_get_counter(bitmap, offset, &blocks, 0); + if (!bmc) { + spin_unlock_irqrestore(&bitmap->lock, flags); + return; + } + + if (!success && ! (*bmc & NEEDED_MASK)) + *bmc |= NEEDED_MASK; + + (*bmc)--; + if (*bmc <= 2) { + set_page_attr(bitmap, + filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap)), + BITMAP_PAGE_CLEAN); + } + spin_unlock_irqrestore(&bitmap->lock, flags); + offset += blocks; + if (sectors > blocks) + sectors -= blocks; + else sectors = 0; + } +} + +int bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks) +{ + bitmap_counter_t *bmc; + int rv; + if (bitmap == NULL) {/* FIXME or bitmap set as 'failed' */ + *blocks = 1024; + return 1; /* always resync if no bitmap */ + } + spin_lock_irq(&bitmap->lock); + bmc = bitmap_get_counter(bitmap, offset, blocks, 0); + rv = 0; + if (bmc) { + /* locked */ + if (RESYNC(*bmc)) + rv = 1; + else if (NEEDED(*bmc)) { + rv = 1; + *bmc |= RESYNC_MASK; + *bmc &= ~NEEDED_MASK; + } + } + spin_unlock_irq(&bitmap->lock); + return rv; +} + +void bitmap_end_sync(struct bitmap *bitmap, sector_t offset, int *blocks, int aborted) +{ + bitmap_counter_t *bmc; + unsigned long flags; +/* + if (offset == 0) printk("bitmap_end_sync 0 (%d)\n", aborted); +*/ if (bitmap == NULL) { + *blocks = 1024; + return; + } + spin_lock_irqsave(&bitmap->lock, flags); + bmc = bitmap_get_counter(bitmap, offset, blocks, 0); + if (bmc == NULL) + goto unlock; + /* locked */ +/* + if (offset == 0) printk("bitmap_end sync found 0x%x, blocks %d\n", *bmc, *blocks); +*/ + if (RESYNC(*bmc)) { + *bmc &= ~RESYNC_MASK; + + if (!NEEDED(*bmc) && aborted) + *bmc |= NEEDED_MASK; + else { + if (*bmc <= 2) { + set_page_attr(bitmap, + filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap)), + BITMAP_PAGE_CLEAN); + } + } + } + unlock: + spin_unlock_irqrestore(&bitmap->lock, flags); +} + +void bitmap_close_sync(struct bitmap *bitmap) +{ + /* Sync has finished, and any bitmap chunks that weren't synced + * properly have been aborted. It remains to us to clear the + * RESYNC bit wherever it is still on + */ + sector_t sector = 0; + int blocks; + if (!bitmap) return; + while (sector < bitmap->mddev->resync_max_sectors) { + bitmap_end_sync(bitmap, sector, &blocks, 0); +/* + if (sector < 500) printk("bitmap_close_sync: sec %llu blks %d\n", + (unsigned long long)sector, blocks); +*/ sector += blocks; + } +} + +static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset, + unsigned long sectors, int set) +{ + /* For each chunk covered by any of these sectors, set the + * resync needed bit, and the counter to 1. They should all + * be 0 at this point + */ + while (sectors) { + int secs; + bitmap_counter_t *bmc; + spin_lock_irq(&bitmap->lock); + bmc = bitmap_get_counter(bitmap, offset, &secs, 1); + if (!bmc) { + spin_unlock_irq(&bitmap->lock); + return; + } + if (set && !NEEDED(*bmc)) { + BUG_ON(*bmc); + *bmc = NEEDED_MASK | 1; + bitmap_count_page(bitmap, offset, 1); + } + spin_unlock_irq(&bitmap->lock); + if (sectors > secs) + sectors -= secs; + else + sectors = 0; + } +} + +/* dirty the entire bitmap */ +int bitmap_setallbits(struct bitmap *bitmap) +{ + unsigned long flags; + unsigned long j; + + /* dirty the in-memory bitmap */ + bitmap_set_memory_bits(bitmap, 0, bitmap->chunks << CHUNK_BLOCK_SHIFT(bitmap), 1); + + /* dirty the bitmap file */ + for (j = 0; j < bitmap->file_pages; j++) { + struct page *page = bitmap->filemap[j]; + + spin_lock_irqsave(&bitmap->lock, flags); + page_cache_get(page); + spin_unlock_irqrestore(&bitmap->lock, flags); + memset(kmap(page), 0xff, PAGE_SIZE); + kunmap(page); + write_page(page, 0); + } + + return 0; +} + +/* + * free memory that was allocated + */ +void bitmap_destroy(mddev_t *mddev) +{ + unsigned long k, pages; + struct bitmap_page *bp; + struct bitmap *bitmap = mddev->bitmap; + + if (!bitmap) /* there was no bitmap */ + return; + + mddev->bitmap = NULL; /* disconnect from the md device */ + + /* release the bitmap file and kill the daemon */ + bitmap_file_put(bitmap); + + bp = bitmap->bp; + pages = bitmap->pages; + + /* free all allocated memory */ + + mempool_destroy(bitmap->write_pool); + + if (bp) /* deallocate the page memory */ + for (k = 0; k < pages; k++) + if (bp[k].map && !bp[k].hijacked) + kfree(bp[k].map); + kfree(bp); + kfree(bitmap); +} + +/* + * initialize the bitmap structure + * if this returns an error, bitmap_destroy must be called to do clean up + */ +int bitmap_create(mddev_t *mddev) +{ + struct bitmap *bitmap; + unsigned long blocks = mddev->resync_max_sectors; + unsigned long chunks; + unsigned long pages; + struct file *file = mddev->bitmap_file; + int err; + + BUG_ON(sizeof(bitmap_super_t) != 256); + + if (!file) /* bitmap disabled, nothing to do */ + return 0; + + bitmap = kmalloc(sizeof(*bitmap), GFP_KERNEL); + if (!bitmap) + return -ENOMEM; + + memset(bitmap, 0, sizeof(*bitmap)); + + spin_lock_init(&bitmap->lock); + bitmap->mddev = mddev; + mddev->bitmap = bitmap; + + spin_lock_init(&bitmap->write_lock); + init_MUTEX_LOCKED(&bitmap->write_done); + INIT_LIST_HEAD(&bitmap->complete_pages); + init_waitqueue_head(&bitmap->write_wait); + bitmap->write_pool = mempool_create(WRITE_POOL_SIZE, write_pool_alloc, + write_pool_free, NULL); + if (!bitmap->write_pool) + return -ENOMEM; + + bitmap->file = file; + get_file(file); + /* read superblock from bitmap file (this sets bitmap->chunksize) */ + err = bitmap_read_sb(bitmap); + if (err) + return err; + + bitmap->chunkshift = find_first_bit(&bitmap->chunksize, + sizeof(bitmap->chunksize)); + + /* now that chunksize and chunkshift are set, we can use these macros */ + chunks = (blocks + CHUNK_BLOCK_RATIO(bitmap) - 1) / + CHUNK_BLOCK_RATIO(bitmap); + pages = (chunks + PAGE_COUNTER_RATIO - 1) / PAGE_COUNTER_RATIO; + + BUG_ON(!pages); + + bitmap->chunks = chunks; + bitmap->pages = pages; + bitmap->missing_pages = pages; + bitmap->counter_bits = COUNTER_BITS; + + bitmap->syncchunk = ~0UL; + +#if INJECT_FATAL_FAULT_1 + bitmap->bp = NULL; +#else + bitmap->bp = kmalloc(pages * sizeof(*bitmap->bp), GFP_KERNEL); +#endif + if (!bitmap->bp) + return -ENOMEM; + memset(bitmap->bp, 0, pages * sizeof(*bitmap->bp)); + + bitmap->flags |= BITMAP_ACTIVE; + + /* now that we have some pages available, initialize the in-memory + * bitmap from the on-disk bitmap */ + err = bitmap_init_from_disk(bitmap); + if (err) + return err; + + printk(KERN_INFO "created bitmap (%lu pages) for device %s\n", + pages, bmname(bitmap)); + + /* kick off the bitmap daemons */ + err = bitmap_start_daemons(bitmap); + if (err) + return err; + return bitmap_update_sb(bitmap); +} + +/* the bitmap API -- for raid personalities */ +EXPORT_SYMBOL(bitmap_startwrite); +EXPORT_SYMBOL(bitmap_endwrite); +EXPORT_SYMBOL(bitmap_start_sync); +EXPORT_SYMBOL(bitmap_end_sync); +EXPORT_SYMBOL(bitmap_unplug); +EXPORT_SYMBOL(bitmap_close_sync); +EXPORT_SYMBOL(bitmap_daemon_work); diff --git a/drivers/md/md.c b/drivers/md/md.c index fa608a1a5c20..c402f6cc7047 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -19,6 +19,9 @@ Neil Brown . + - persistent bitmap code + Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc. + This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2, or (at your option) @@ -33,6 +36,7 @@ #include #include #include +#include #include #include #include /* for invalidate_bdev */ @@ -40,6 +44,8 @@ #include +#include + #ifdef CONFIG_KMOD #include #endif @@ -1198,8 +1204,11 @@ void md_print_devices(void) printk("md: * *\n"); printk("md: **********************************\n"); ITERATE_MDDEV(mddev,tmp) { - printk("%s: ", mdname(mddev)); + if (mddev->bitmap) + bitmap_print_sb(mddev->bitmap); + else + printk("%s: ", mdname(mddev)); ITERATE_RDEV(mddev,rdev,tmp2) printk("<%s>", bdevname(rdev->bdev,b)); printk("\n"); @@ -1287,7 +1296,7 @@ repeat: "md: updating %s RAID superblock on device (in sync %d)\n", mdname(mddev),mddev->in_sync); - err = 0; + err = bitmap_update_sb(mddev->bitmap); ITERATE_RDEV(mddev,rdev,tmp) { char b[BDEVNAME_SIZE]; dprintk(KERN_INFO "md: "); @@ -1624,12 +1633,19 @@ static int do_md_run(mddev_t * mddev) mddev->resync_max_sectors = mddev->size << 1; /* may be over-ridden by personality */ - err = mddev->pers->run(mddev); + /* before we start the array running, initialise the bitmap */ + err = bitmap_create(mddev); + if (err) + printk(KERN_ERR "%s: failed to create bitmap (%d)\n", + mdname(mddev), err); + else + err = mddev->pers->run(mddev); if (err) { printk(KERN_ERR "md: pers->run() failed ...\n"); module_put(mddev->pers->owner); mddev->pers = NULL; - return -EINVAL; + bitmap_destroy(mddev); + return err; } atomic_set(&mddev->writes_pending,0); mddev->safemode = 0; @@ -1742,6 +1758,14 @@ static int do_md_stop(mddev_t * mddev, int ro) if (ro) set_disk_ro(disk, 1); } + + bitmap_destroy(mddev); + if (mddev->bitmap_file) { + atomic_set(&mddev->bitmap_file->f_dentry->d_inode->i_writecount, 1); + fput(mddev->bitmap_file); + mddev->bitmap_file = NULL; + } + /* * Free resources if final stop */ @@ -2000,6 +2024,42 @@ static int get_array_info(mddev_t * mddev, void __user * arg) return 0; } +static int get_bitmap_file(mddev_t * mddev, void * arg) +{ + mdu_bitmap_file_t *file = NULL; /* too big for stack allocation */ + char *ptr, *buf = NULL; + int err = -ENOMEM; + + file = kmalloc(sizeof(*file), GFP_KERNEL); + if (!file) + goto out; + + /* bitmap disabled, zero the first byte and copy out */ + if (!mddev->bitmap || !mddev->bitmap->file) { + file->pathname[0] = '\0'; + goto copy_out; + } + + buf = kmalloc(sizeof(file->pathname), GFP_KERNEL); + if (!buf) + goto out; + + ptr = file_path(mddev->bitmap->file, buf, sizeof(file->pathname)); + if (!ptr) + goto out; + + strcpy(file->pathname, ptr); + +copy_out: + err = 0; + if (copy_to_user(arg, file, sizeof(*file))) + err = -EFAULT; +out: + kfree(buf); + kfree(file); + return err; +} + static int get_disk_info(mddev_t * mddev, void __user * arg) { mdu_disk_info_t info; @@ -2275,6 +2335,48 @@ abort_export: return err; } +/* similar to deny_write_access, but accounts for our holding a reference + * to the file ourselves */ +static int deny_bitmap_write_access(struct file * file) +{ + struct inode *inode = file->f_mapping->host; + + spin_lock(&inode->i_lock); + if (atomic_read(&inode->i_writecount) > 1) { + spin_unlock(&inode->i_lock); + return -ETXTBSY; + } + atomic_set(&inode->i_writecount, -1); + spin_unlock(&inode->i_lock); + + return 0; +} + +static int set_bitmap_file(mddev_t *mddev, int fd) +{ + int err; + + if (mddev->pers) + return -EBUSY; + + mddev->bitmap_file = fget(fd); + + if (mddev->bitmap_file == NULL) { + printk(KERN_ERR "%s: error: failed to get bitmap file\n", + mdname(mddev)); + return -EBADF; + } + + err = deny_bitmap_write_access(mddev->bitmap_file); + if (err) { + printk(KERN_ERR "%s: error: bitmap file is already in use\n", + mdname(mddev)); + fput(mddev->bitmap_file); + mddev->bitmap_file = NULL; + } + return err; +} + /* * set_array_info is used two different ways * The original usage is when creating a new array. @@ -2586,8 +2688,10 @@ static int md_ioctl(struct inode *inode, struct file *file, /* * Commands querying/configuring an existing array: */ - /* if we are initialised yet, only ADD_NEW_DISK or STOP_ARRAY is allowed */ - if (!mddev->raid_disks && cmd != ADD_NEW_DISK && cmd != STOP_ARRAY && cmd != RUN_ARRAY) { + /* if we are not initialised yet, only ADD_NEW_DISK, STOP_ARRAY, + * RUN_ARRAY, and SET_BITMAP_FILE are allowed */ + if (!mddev->raid_disks && cmd != ADD_NEW_DISK && cmd != STOP_ARRAY + && cmd != RUN_ARRAY && cmd != SET_BITMAP_FILE) { err = -ENODEV; goto abort_unlock; } @@ -2601,6 +2705,10 @@ static int md_ioctl(struct inode *inode, struct file *file, err = get_array_info(mddev, argp); goto done_unlock; + case GET_BITMAP_FILE: + err = get_bitmap_file(mddev, (void *)arg); + goto done_unlock; + case GET_DISK_INFO: err = get_disk_info(mddev, argp); goto done_unlock; @@ -2681,6 +2789,10 @@ static int md_ioctl(struct inode *inode, struct file *file, err = do_md_run (mddev); goto done_unlock; + case SET_BITMAP_FILE: + err = set_bitmap_file(mddev, (int)arg); + goto done_unlock; + default: if (_IOC_TYPE(cmd) == MD_MAJOR) printk(KERN_WARNING "md: %s(pid %d) used" @@ -2792,8 +2904,9 @@ static int md_thread(void * arg) while (thread->run) { void (*run)(mddev_t *); - wait_event_interruptible(thread->wqueue, - test_bit(THREAD_WAKEUP, &thread->flags)); + wait_event_interruptible_timeout(thread->wqueue, + test_bit(THREAD_WAKEUP, &thread->flags), + thread->timeout); if (current->flags & PF_FREEZE) refrigerator(PF_FREEZE); @@ -2839,6 +2952,7 @@ mdk_thread_t *md_register_thread(void (*run) (mddev_t *), mddev_t *mddev, thread->run = run; thread->mddev = mddev; thread->name = name; + thread->timeout = MAX_SCHEDULE_TIMEOUT; ret = kernel_thread(md_thread, thread, 0); if (ret < 0) { kfree(thread); @@ -2877,13 +2991,13 @@ void md_error(mddev_t *mddev, mdk_rdev_t *rdev) if (!rdev || rdev->faulty) return; - +/* dprintk("md_error dev:%s, rdev:(%d:%d), (caller: %p,%p,%p,%p).\n", mdname(mddev), MAJOR(rdev->bdev->bd_dev), MINOR(rdev->bdev->bd_dev), __builtin_return_address(0),__builtin_return_address(1), __builtin_return_address(2),__builtin_return_address(3)); - +*/ if (!mddev->pers->error_handler) return; mddev->pers->error_handler(mddev,rdev); @@ -3037,6 +3151,7 @@ static int md_seq_show(struct seq_file *seq, void *v) struct list_head *tmp2; mdk_rdev_t *rdev; int i; + struct bitmap *bitmap; if (v == (void*)1) { seq_printf(seq, "Personalities : "); @@ -3089,10 +3204,36 @@ static int md_seq_show(struct seq_file *seq, void *v) if (mddev->pers) { mddev->pers->status (seq, mddev); seq_printf(seq, "\n "); - if (mddev->curr_resync > 2) + if (mddev->curr_resync > 2) { status_resync (seq, mddev); - else if (mddev->curr_resync == 1 || mddev->curr_resync == 2) - seq_printf(seq, " resync=DELAYED"); + seq_printf(seq, "\n "); + } else if (mddev->curr_resync == 1 || mddev->curr_resync == 2) + seq_printf(seq, " resync=DELAYED\n "); + } else + seq_printf(seq, "\n "); + + if ((bitmap = mddev->bitmap)) { + char *buf, *path; + unsigned long chunk_kb; + unsigned long flags; + buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + spin_lock_irqsave(&bitmap->lock, flags); + chunk_kb = bitmap->chunksize >> 10; + seq_printf(seq, "bitmap: %lu/%lu pages [%luKB], " + "%lu%s chunk", + bitmap->pages - bitmap->missing_pages, + bitmap->pages, + (bitmap->pages - bitmap->missing_pages) + << (PAGE_SHIFT - 10), + chunk_kb ? chunk_kb : bitmap->chunksize, + chunk_kb ? "KB" : "B"); + if (bitmap->file && buf) { + path = file_path(bitmap->file, buf, PAGE_SIZE); + seq_printf(seq, ", file: %s", path ? path : ""); + } + seq_printf(seq, "\n"); + spin_unlock_irqrestore(&bitmap->lock, flags); + kfree(buf); } seq_printf(seq, "\n"); @@ -3328,7 +3469,8 @@ static void md_do_sync(mddev_t *mddev) sysctl_speed_limit_max); is_mddev_idle(mddev); /* this also initializes IO event counters */ - if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) + /* we don't use the checkpoint if there's a bitmap */ + if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) && !mddev->bitmap) j = mddev->recovery_cp; else j = 0; @@ -3673,6 +3815,8 @@ static int __init md_init(void) " MD_SB_DISKS=%d\n", MD_MAJOR_VERSION, MD_MINOR_VERSION, MD_PATCHLEVEL_VERSION, MAX_MD_DEVS, MD_SB_DISKS); + printk(KERN_INFO "md: bitmap version %d.%d\n", BITMAP_MAJOR, + BITMAP_MINOR); if (register_blkdev(MAJOR_NR, "md")) return -1; diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h new file mode 100644 index 000000000000..f785cf26cbad --- /dev/null +++ b/include/linux/raid/bitmap.h @@ -0,0 +1,280 @@ +/* + * bitmap.h: Copyright (C) Peter T. Breuer (ptb@ot.uc3m.es) 2003 + * + * additions: Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc. + */ +#ifndef BITMAP_H +#define BITMAP_H 1 + +#define BITMAP_MAJOR 3 +#define BITMAP_MINOR 38 + +/* + * in-memory bitmap: + * + * Use 16 bit block counters to track pending writes to each "chunk". + * The 2 high order bits are special-purpose, the first is a flag indicating + * whether a resync is needed. The second is a flag indicating whether a + * resync is active. + * This means that the counter is actually 14 bits: + * + * +--------+--------+------------------------------------------------+ + * | resync | resync | counter | + * | needed | active | | + * | (0-1) | (0-1) | (0-16383) | + * +--------+--------+------------------------------------------------+ + * + * The "resync needed" bit is set when: + * a '1' bit is read from storage at startup. + * a write request fails on some drives + * a resync is aborted on a chunk with 'resync active' set + * It is cleared (and resync-active set) when a resync starts across all drives + * of the chunk. + * + * + * The "resync active" bit is set when: + * a resync is started on all drives, and resync_needed is set. + * resync_needed will be cleared (as long as resync_active wasn't already set). + * It is cleared when a resync completes. + * + * The counter counts pending write requests, plus the on-disk bit. + * When the counter is '1' and the resync bits are clear, the on-disk + * bit can be cleared aswell, thus setting the counter to 0. + * When we set a bit, or in the counter (to start a write), if the fields is + * 0, we first set the disk bit and set the counter to 1. + * + * If the counter is 0, the on-disk bit is clear and the stipe is clean + * Anything that dirties the stipe pushes the counter to 2 (at least) + * and sets the on-disk bit (lazily). + * If a periodic sweep find the counter at 2, it is decremented to 1. + * If the sweep find the counter at 1, the on-disk bit is cleared and the + * counter goes to zero. + * + * Also, we'll hijack the "map" pointer itself and use it as two 16 bit block + * counters as a fallback when "page" memory cannot be allocated: + * + * Normal case (page memory allocated): + * + * page pointer (32-bit) + * + * [ ] ------+ + * | + * +-------> [ ][ ]..[ ] (4096 byte page == 2048 counters) + * c1 c2 c2048 + * + * Hijacked case (page memory allocation failed): + * + * hijacked page pointer (32-bit) + * + * [ ][ ] (no page memory allocated) + * counter #1 (16-bit) counter #2 (16-bit) + * + */ + +#ifdef __KERNEL__ + +#define PAGE_BITS (PAGE_SIZE << 3) +#define PAGE_BIT_SHIFT (PAGE_SHIFT + 3) + +typedef __u16 bitmap_counter_t; +#define COUNTER_BITS 16 +#define COUNTER_BIT_SHIFT 4 +#define COUNTER_BYTE_RATIO (COUNTER_BITS / 8) +#define COUNTER_BYTE_SHIFT (COUNTER_BIT_SHIFT - 3) + +#define NEEDED_MASK ((bitmap_counter_t) (1 << (COUNTER_BITS - 1))) +#define RESYNC_MASK ((bitmap_counter_t) (1 << (COUNTER_BITS - 2))) +#define COUNTER_MAX ((bitmap_counter_t) RESYNC_MASK - 1) +#define NEEDED(x) (((bitmap_counter_t) x) & NEEDED_MASK) +#define RESYNC(x) (((bitmap_counter_t) x) & RESYNC_MASK) +#define COUNTER(x) (((bitmap_counter_t) x) & COUNTER_MAX) + +/* how many counters per page? */ +#define PAGE_COUNTER_RATIO (PAGE_BITS / COUNTER_BITS) +/* same, except a shift value for more efficient bitops */ +#define PAGE_COUNTER_SHIFT (PAGE_BIT_SHIFT - COUNTER_BIT_SHIFT) +/* same, except a mask value for more efficient bitops */ +#define PAGE_COUNTER_MASK (PAGE_COUNTER_RATIO - 1) + +#define BITMAP_BLOCK_SIZE 512 +#define BITMAP_BLOCK_SHIFT 9 + +/* how many blocks per chunk? (this is variable) */ +#define CHUNK_BLOCK_RATIO(bitmap) ((bitmap)->chunksize >> BITMAP_BLOCK_SHIFT) +#define CHUNK_BLOCK_SHIFT(bitmap) ((bitmap)->chunkshift - BITMAP_BLOCK_SHIFT) +#define CHUNK_BLOCK_MASK(bitmap) (CHUNK_BLOCK_RATIO(bitmap) - 1) + +/* when hijacked, the counters and bits represent even larger "chunks" */ +/* there will be 1024 chunks represented by each counter in the page pointers */ +#define PAGEPTR_BLOCK_RATIO(bitmap) \ + (CHUNK_BLOCK_RATIO(bitmap) << PAGE_COUNTER_SHIFT >> 1) +#define PAGEPTR_BLOCK_SHIFT(bitmap) \ + (CHUNK_BLOCK_SHIFT(bitmap) + PAGE_COUNTER_SHIFT - 1) +#define PAGEPTR_BLOCK_MASK(bitmap) (PAGEPTR_BLOCK_RATIO(bitmap) - 1) + +/* + * on-disk bitmap: + * + * Use one bit per "chunk" (block set). We do the disk I/O on the bitmap + * file a page at a time. There's a superblock at the start of the file. + */ + +/* map chunks (bits) to file pages - offset by the size of the superblock */ +#define CHUNK_BIT_OFFSET(chunk) ((chunk) + (sizeof(bitmap_super_t) << 3)) + +#endif + +/* + * bitmap structures: + */ + +#define BITMAP_MAGIC 0x6d746962 + +/* use these for bitmap->flags and bitmap->sb->state bit-fields */ +enum bitmap_state { + BITMAP_ACTIVE = 0x001, /* the bitmap is in use */ + BITMAP_STALE = 0x002 /* the bitmap file is out of date or had -EIO */ +}; + +/* the superblock at the front of the bitmap file -- little endian */ +typedef struct bitmap_super_s { + __u32 magic; /* 0 BITMAP_MAGIC */ + __u32 version; /* 4 the bitmap major for now, could change... */ + __u8 uuid[16]; /* 8 128 bit uuid - must match md device uuid */ + __u64 events; /* 24 event counter for the bitmap (1)*/ + __u64 events_cleared;/*32 event counter when last bit cleared (2) */ + __u64 sync_size; /* 40 the size of the md device's sync range(3) */ + __u32 state; /* 48 bitmap state information */ + __u32 chunksize; /* 52 the bitmap chunk size in bytes */ + __u32 daemon_sleep; /* 56 seconds between disk flushes */ + + __u8 pad[256 - 60]; /* set to zero */ +} bitmap_super_t; + +/* notes: + * (1) This event counter is updated before the eventcounter in the md superblock + * When a bitmap is loaded, it is only accepted if this event counter is equal + * to, or one greater than, the event counter in the superblock. + * (2) This event counter is updated when the other one is *if*and*only*if* the + * array is not degraded. As bits are not cleared when the array is degraded, + * this represents the last time that any bits were cleared. + * If a device is being added that has an event count with this value or + * higher, it is accepted as conforming to the bitmap. + * (3)This is the number of sectors represented by the bitmap, and is the range that + * resync happens across. For raid1 and raid5/6 it is the size of individual + * devices. For raid10 it is the size of the array. + */ + +#ifdef __KERNEL__ + +/* the in-memory bitmap is represented by bitmap_pages */ +struct bitmap_page { + /* + * map points to the actual memory page + */ + char *map; + /* + * in emergencies (when map cannot be alloced), hijack the map + * pointer and use it as two counters itself + */ + unsigned int hijacked:1; + /* + * count of dirty bits on the page + */ + unsigned int count:31; +}; + +/* keep track of bitmap file pages that have pending writes on them */ +struct page_list { + struct list_head list; + struct page *page; +}; + +/* the main bitmap structure - one per mddev */ +struct bitmap { + struct bitmap_page *bp; + unsigned long pages; /* total number of pages in the bitmap */ + unsigned long missing_pages; /* number of pages not yet allocated */ + + mddev_t *mddev; /* the md device that the bitmap is for */ + + int counter_bits; /* how many bits per block counter */ + + /* bitmap chunksize -- how much data does each bit represent? */ + unsigned long chunksize; + unsigned long chunkshift; /* chunksize = 2^chunkshift (for bitops) */ + unsigned long chunks; /* total number of data chunks for the array */ + + /* We hold a count on the chunk currently being synced, and drop + * it when the last block is started. If the resync is aborted + * midway, we need to be able to drop that count, so we remember + * the counted chunk.. + */ + unsigned long syncchunk; + + __u64 events_cleared; + + /* bitmap spinlock */ + spinlock_t lock; + + struct file *file; /* backing disk file */ + struct page *sb_page; /* cached copy of the bitmap file superblock */ + struct page **filemap; /* list of cache pages for the file */ + unsigned long *filemap_attr; /* attributes associated w/ filemap pages */ + unsigned long file_pages; /* number of pages in the file */ + + unsigned long flags; + + /* + * the bitmap daemon - periodically wakes up and sweeps the bitmap + * file, cleaning up bits and flushing out pages to disk as necessary + */ + unsigned long daemon_lastrun; /* jiffies of last run */ + unsigned long daemon_sleep; /* how many seconds between updates? */ + + /* + * bitmap write daemon - this daemon performs writes to the bitmap file + * this thread is only needed because of a limitation in ext3 (jbd) + * that does not allow a task to have two journal transactions ongoing + * simultaneously (even if the transactions are for two different + * filesystems) -- in the case of bitmap, that would be the filesystem + * that the bitmap file resides on and the filesystem that is mounted + * on the md device -- see current->journal_info in jbd/transaction.c + */ + mdk_thread_t *writeback_daemon; + spinlock_t write_lock; + struct semaphore write_ready; + struct semaphore write_done; + unsigned long writes_pending; + wait_queue_head_t write_wait; + struct list_head write_pages; + struct list_head complete_pages; + mempool_t *write_pool; +}; + +/* the bitmap API */ + +/* these are used only by md/bitmap */ +int bitmap_create(mddev_t *mddev); +void bitmap_destroy(mddev_t *mddev); +int bitmap_active(struct bitmap *bitmap); + +char *file_path(struct file *file, char *buf, int count); +void bitmap_print_sb(struct bitmap *bitmap); +int bitmap_update_sb(struct bitmap *bitmap); + +int bitmap_setallbits(struct bitmap *bitmap); + +/* these are exported */ +int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors); +void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors, + int success); +int bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks); +void bitmap_end_sync(struct bitmap *bitmap, sector_t offset, int *blocks, int aborted); +void bitmap_close_sync(struct bitmap *bitmap); + +int bitmap_unplug(struct bitmap *bitmap); +int bitmap_daemon_work(struct bitmap *bitmap); +#endif + +#endif diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h index bce0032decff..16e94a9f0f8c 100644 --- a/include/linux/raid/md_k.h +++ b/include/linux/raid/md_k.h @@ -267,6 +267,9 @@ struct mddev_s atomic_t writes_pending; request_queue_t *queue; /* for plugging ... */ + struct bitmap *bitmap; /* the bitmap for the device */ + struct file *bitmap_file; /* the bitmap file */ + struct list_head all_mddevs; }; @@ -341,6 +344,7 @@ typedef struct mdk_thread_s { unsigned long flags; struct completion *event; struct task_struct *tsk; + unsigned long timeout; const char *name; } mdk_thread_t; diff --git a/include/linux/raid/md_u.h b/include/linux/raid/md_u.h index a2df5c2a42af..81da20ccec4d 100644 --- a/include/linux/raid/md_u.h +++ b/include/linux/raid/md_u.h @@ -23,6 +23,7 @@ #define GET_DISK_INFO _IOR (MD_MAJOR, 0x12, mdu_disk_info_t) #define PRINT_RAID_DEBUG _IO (MD_MAJOR, 0x13) #define RAID_AUTORUN _IO (MD_MAJOR, 0x14) +#define GET_BITMAP_FILE _IOR (MD_MAJOR, 0x15, mdu_bitmap_file_t) /* configuration */ #define CLEAR_ARRAY _IO (MD_MAJOR, 0x20) @@ -36,6 +37,7 @@ #define HOT_ADD_DISK _IO (MD_MAJOR, 0x28) #define SET_DISK_FAULTY _IO (MD_MAJOR, 0x29) #define HOT_GENERATE_ERROR _IO (MD_MAJOR, 0x2a) +#define SET_BITMAP_FILE _IOW (MD_MAJOR, 0x2b, int) /* usage */ #define RUN_ARRAY _IOW (MD_MAJOR, 0x30, mdu_param_t) @@ -106,6 +108,11 @@ typedef struct mdu_start_info_s { } mdu_start_info_t; +typedef struct mdu_bitmap_file_s +{ + char pathname[4096]; +} mdu_bitmap_file_t; + typedef struct mdu_param_s { int personality; /* 1,2,3,4 */ -- cgit v1.2.3 From 77ad4bc706fe6c52ab953f31c287a6af712d080c Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:21 -0700 Subject: [PATCH] md: enable the bitmap write-back daemon and wait for it. Currently we don't wait for updates to the bitmap to be flushed to disk properly. The infrastructure all there, but it isn't being used.... A separate kernel thread (bitmap_writeback_daemon) is needed to wait for each page as we cannot get callbacks when a page write completes. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/bitmap.c | 119 ++++++++++++++++++++------------------------ include/linux/raid/bitmap.h | 13 +---- 2 files changed, 55 insertions(+), 77 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 9462fdd517c0..86b6b037fa44 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -261,30 +261,33 @@ char *file_path(struct file *file, char *buf, int count) /* * write out a page */ -static int write_page(struct page *page, int wait) +static int write_page(struct bitmap *bitmap, struct page *page, int wait) { int ret = -ENOMEM; lock_page(page); - if (page->mapping == NULL) - goto unlock_out; - else if (i_size_read(page->mapping->host) < page->index << PAGE_SHIFT) { - ret = -ENOENT; - goto unlock_out; - } - ret = page->mapping->a_ops->prepare_write(NULL, page, 0, PAGE_SIZE); if (!ret) ret = page->mapping->a_ops->commit_write(NULL, page, 0, PAGE_SIZE); if (ret) { -unlock_out: unlock_page(page); return ret; } set_page_dirty(page); /* force it to be written out */ + + if (!wait) { + /* add to list to be waited for by daemon */ + struct page_list *item = mempool_alloc(bitmap->write_pool, GFP_NOIO); + item->page = page; + page_cache_get(page); + spin_lock(&bitmap->write_lock); + list_add(&item->list, &bitmap->complete_pages); + spin_unlock(&bitmap->write_lock); + md_wakeup_thread(bitmap->writeback_daemon); + } return write_one_page(page, wait); } @@ -343,14 +346,13 @@ int bitmap_update_sb(struct bitmap *bitmap) spin_unlock_irqrestore(&bitmap->lock, flags); return 0; } - page_cache_get(bitmap->sb_page); spin_unlock_irqrestore(&bitmap->lock, flags); sb = (bitmap_super_t *)kmap(bitmap->sb_page); sb->events = cpu_to_le64(bitmap->mddev->events); if (!bitmap->mddev->degraded) sb->events_cleared = cpu_to_le64(bitmap->mddev->events); kunmap(bitmap->sb_page); - return write_page(bitmap->sb_page, 0); + return write_page(bitmap, bitmap->sb_page, 0); } /* print out the bitmap file superblock */ @@ -556,10 +558,10 @@ static void bitmap_file_unmap(struct bitmap *bitmap) static void bitmap_stop_daemons(struct bitmap *bitmap); /* dequeue the next item in a page list -- don't call from irq context */ -static struct page_list *dequeue_page(struct bitmap *bitmap, - struct list_head *head) +static struct page_list *dequeue_page(struct bitmap *bitmap) { struct page_list *item = NULL; + struct list_head *head = &bitmap->complete_pages; spin_lock(&bitmap->write_lock); if (list_empty(head)) @@ -573,23 +575,15 @@ out: static void drain_write_queues(struct bitmap *bitmap) { - struct list_head *queues[] = { &bitmap->complete_pages, NULL }; - struct list_head *head; struct page_list *item; - int i; - for (i = 0; queues[i]; i++) { - head = queues[i]; - while ((item = dequeue_page(bitmap, head))) { - page_cache_release(item->page); - mempool_free(item, bitmap->write_pool); - } + while ((item = dequeue_page(bitmap))) { + /* don't bother to wait */ + page_cache_release(item->page); + mempool_free(item, bitmap->write_pool); } - spin_lock(&bitmap->write_lock); - bitmap->writes_pending = 0; /* make sure waiters continue */ wake_up(&bitmap->write_wait); - spin_unlock(&bitmap->write_lock); } static void bitmap_file_put(struct bitmap *bitmap) @@ -734,13 +728,13 @@ int bitmap_unplug(struct bitmap *bitmap) spin_unlock_irqrestore(&bitmap->lock, flags); if (attr & (BITMAP_PAGE_DIRTY | BITMAP_PAGE_NEEDWRITE)) - if (write_page(page, 0)) + if (write_page(bitmap, page, 0)) return 1; } if (wait) { /* if any writes were performed, we need to wait on them */ spin_lock_irq(&bitmap->write_lock); wait_event_lock_irq(bitmap->write_wait, - bitmap->writes_pending == 0, bitmap->write_lock, + list_empty(&bitmap->complete_pages), bitmap->write_lock, wake_up_process(bitmap->writeback_daemon->tsk)); spin_unlock_irq(&bitmap->write_lock); } @@ -841,7 +835,7 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, int in_sync) */ memset(page_address(page) + offset, 0xff, PAGE_SIZE - offset); - ret = write_page(page, 1); + ret = write_page(bitmap, page, 1); if (ret) { kunmap(page); /* release, page not in filemap yet */ @@ -934,7 +928,7 @@ int bitmap_daemon_work(struct bitmap *bitmap) } spin_unlock_irqrestore(&bitmap->lock, flags); if (attr & BITMAP_PAGE_NEEDWRITE) { - if (write_page(page, 0)) + if (write_page(bitmap, page, 0)) bitmap_file_kick(bitmap); page_cache_release(page); } @@ -950,7 +944,7 @@ int bitmap_daemon_work(struct bitmap *bitmap) if (get_page_attr(bitmap, lastpage) & BITMAP_PAGE_NEEDWRITE) { clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE); spin_unlock_irqrestore(&bitmap->lock, flags); - err = write_page(lastpage, 0); + err = write_page(bitmap, lastpage, 0); } else { set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE); spin_unlock_irqrestore(&bitmap->lock, flags); @@ -998,7 +992,7 @@ int bitmap_daemon_work(struct bitmap *bitmap) if (get_page_attr(bitmap, lastpage) &BITMAP_PAGE_NEEDWRITE) { clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE); spin_unlock_irqrestore(&bitmap->lock, flags); - err = write_page(lastpage, 0); + err = write_page(bitmap, lastpage, 0); } else { set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE); spin_unlock_irqrestore(&bitmap->lock, flags); @@ -1034,46 +1028,40 @@ static void bitmap_writeback_daemon(mddev_t *mddev) struct page_list *item; int err = 0; - while (1) { - PRINTK("%s: bitmap writeback daemon waiting...\n", bmname(bitmap)); - down_interruptible(&bitmap->write_done); - if (signal_pending(current)) { - printk(KERN_INFO - "%s: bitmap writeback daemon got signal, exiting...\n", - bmname(bitmap)); - break; - } + if (signal_pending(current)) { + printk(KERN_INFO + "%s: bitmap writeback daemon got signal, exiting...\n", + bmname(bitmap)); + err = -EINTR; + goto out; + } - PRINTK("%s: bitmap writeback daemon woke up...\n", bmname(bitmap)); - /* wait on bitmap page writebacks */ - while ((item = dequeue_page(bitmap, &bitmap->complete_pages))) { - page = item->page; - mempool_free(item, bitmap->write_pool); - PRINTK("wait on page writeback: %p %lu\n", page, bitmap->writes_pending); - wait_on_page_writeback(page); - PRINTK("finished page writeback: %p %lu\n", page, bitmap->writes_pending); - spin_lock(&bitmap->write_lock); - if (!--bitmap->writes_pending) - wake_up(&bitmap->write_wait); - spin_unlock(&bitmap->write_lock); - err = PageError(page); - page_cache_release(page); - if (err) { - printk(KERN_WARNING "%s: bitmap file writeback " - "failed (page %lu): %d\n", - bmname(bitmap), page->index, err); - bitmap_file_kick(bitmap); - goto out; - } + PRINTK("%s: bitmap writeback daemon woke up...\n", bmname(bitmap)); + /* wait on bitmap page writebacks */ + while ((item = dequeue_page(bitmap))) { + page = item->page; + mempool_free(item, bitmap->write_pool); + PRINTK("wait on page writeback: %p\n", page); + wait_on_page_writeback(page); + PRINTK("finished page writeback: %p\n", page); + + err = PageError(page); + page_cache_release(page); + if (err) { + printk(KERN_WARNING "%s: bitmap file writeback " + "failed (page %lu): %d\n", + bmname(bitmap), page->index, err); + bitmap_file_kick(bitmap); + goto out; } } -out: + out: + wake_up(&bitmap->write_wait); if (err) { printk(KERN_INFO "%s: bitmap writeback daemon exiting (%d)\n", - bmname(bitmap), err); + bmname(bitmap), err); daemon_exit(bitmap, &bitmap->writeback_daemon); } - return; } static int bitmap_start_daemon(struct bitmap *bitmap, mdk_thread_t **ptr, @@ -1375,7 +1363,7 @@ int bitmap_setallbits(struct bitmap *bitmap) spin_unlock_irqrestore(&bitmap->lock, flags); memset(kmap(page), 0xff, PAGE_SIZE); kunmap(page); - if (write_page(page, 0)) + if (write_page(bitmap, page, 0)) return 1; } @@ -1443,7 +1431,6 @@ int bitmap_create(mddev_t *mddev) mddev->bitmap = bitmap; spin_lock_init(&bitmap->write_lock); - init_MUTEX_LOCKED(&bitmap->write_done); INIT_LIST_HEAD(&bitmap->complete_pages); init_waitqueue_head(&bitmap->write_wait); bitmap->write_pool = mempool_create(WRITE_POOL_SIZE, write_pool_alloc, diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h index f785cf26cbad..cfe60cfc8f3d 100644 --- a/include/linux/raid/bitmap.h +++ b/include/linux/raid/bitmap.h @@ -233,21 +233,12 @@ struct bitmap { unsigned long daemon_sleep; /* how many seconds between updates? */ /* - * bitmap write daemon - this daemon performs writes to the bitmap file - * this thread is only needed because of a limitation in ext3 (jbd) - * that does not allow a task to have two journal transactions ongoing - * simultaneously (even if the transactions are for two different - * filesystems) -- in the case of bitmap, that would be the filesystem - * that the bitmap file resides on and the filesystem that is mounted - * on the md device -- see current->journal_info in jbd/transaction.c + * bitmap_writeback_daemon waits for file-pages that have been written, + * as there is no way to get a call-back when a page write completes. */ mdk_thread_t *writeback_daemon; spinlock_t write_lock; - struct semaphore write_ready; - struct semaphore write_done; - unsigned long writes_pending; wait_queue_head_t write_wait; - struct list_head write_pages; struct list_head complete_pages; mempool_t *write_pool; }; -- cgit v1.2.3 From 191ea9b2c7cc3ebbe0678834ab710d7d95ad3f9a Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:23 -0700 Subject: [PATCH] md: raid1 support for bitmap intent logging Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/raid1.c | 182 ++++++++++++++++++++++++++++++++++++++------- include/linux/raid/raid1.h | 16 +++- 2 files changed, 168 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 3c5c916cb09e..0fd4c3bfc851 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -12,6 +12,15 @@ * Fixes to reconstruction by Jakob Østergaard" * Various fixes by Neil Brown * + * Changes by Peter T. Breuer 31/1/2003 to support + * bitmapped intelligence in resync: + * + * - bitmap marked during normal i/o + * - bitmap used to skip nondirty blocks during sync + * + * Additions to bitmap code, (C) 2003-2004 Paul Clements, SteelEye Technology: + * - persistent bitmap code + * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2, or (at your option) @@ -22,7 +31,16 @@ * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ +#include "dm-bio-list.h" #include +#include + +#define DEBUG 0 +#if DEBUG +#define PRINTK(x...) printk(x) +#else +#define PRINTK(x...) +#endif /* * Number of guaranteed r1bios in case of extreme VM load: @@ -287,9 +305,11 @@ static int raid1_end_write_request(struct bio *bio, unsigned int bytes_done, int /* * this branch is our 'one mirror IO has finished' event handler: */ - if (!uptodate) + if (!uptodate) { md_error(r1_bio->mddev, conf->mirrors[mirror].rdev); - else + /* an I/O failed, we can't clear the bitmap */ + set_bit(R1BIO_Degraded, &r1_bio->state); + } else /* * Set R1BIO_Uptodate in our master bio, so that * we will return a good error code for to the higher @@ -309,6 +329,10 @@ static int raid1_end_write_request(struct bio *bio, unsigned int bytes_done, int * already. */ if (atomic_dec_and_test(&r1_bio->remaining)) { + /* clear the bitmap if all writes complete successfully */ + bitmap_endwrite(r1_bio->mddev->bitmap, r1_bio->sector, + r1_bio->sectors, + !test_bit(R1BIO_Degraded, &r1_bio->state)); md_write_end(r1_bio->mddev); raid_end_bio_io(r1_bio); } @@ -458,7 +482,10 @@ static void unplug_slaves(mddev_t *mddev) static void raid1_unplug(request_queue_t *q) { - unplug_slaves(q->queuedata); + mddev_t *mddev = q->queuedata; + + unplug_slaves(mddev); + md_wakeup_thread(mddev->thread); } static int raid1_issue_flush(request_queue_t *q, struct gendisk *disk, @@ -501,16 +528,16 @@ static void device_barrier(conf_t *conf, sector_t sect) { spin_lock_irq(&conf->resync_lock); wait_event_lock_irq(conf->wait_idle, !waitqueue_active(&conf->wait_resume), - conf->resync_lock, unplug_slaves(conf->mddev)); + conf->resync_lock, raid1_unplug(conf->mddev->queue)); if (!conf->barrier++) { wait_event_lock_irq(conf->wait_idle, !conf->nr_pending, - conf->resync_lock, unplug_slaves(conf->mddev)); + conf->resync_lock, raid1_unplug(conf->mddev->queue)); if (conf->nr_pending) BUG(); } wait_event_lock_irq(conf->wait_resume, conf->barrier < RESYNC_DEPTH, - conf->resync_lock, unplug_slaves(conf->mddev)); + conf->resync_lock, raid1_unplug(conf->mddev->queue)); conf->next_resync = sect; spin_unlock_irq(&conf->resync_lock); } @@ -522,8 +549,12 @@ static int make_request(request_queue_t *q, struct bio * bio) mirror_info_t *mirror; r1bio_t *r1_bio; struct bio *read_bio; - int i, disks; + int i, targets = 0, disks; mdk_rdev_t *rdev; + struct bitmap *bitmap = mddev->bitmap; + unsigned long flags; + struct bio_list bl; + /* * Register the new request and wait if the reconstruction @@ -554,7 +585,7 @@ static int make_request(request_queue_t *q, struct bio * bio) r1_bio->master_bio = bio; r1_bio->sectors = bio->bi_size >> 9; - + r1_bio->state = 0; r1_bio->mddev = mddev; r1_bio->sector = bio->bi_sector; @@ -597,6 +628,13 @@ static int make_request(request_queue_t *q, struct bio * bio) * bios[x] to bio */ disks = conf->raid_disks; +#if 0 + { static int first=1; + if (first) printk("First Write sector %llu disks %d\n", + (unsigned long long)r1_bio->sector, disks); + first = 0; + } +#endif rcu_read_lock(); for (i = 0; i < disks; i++) { if ((rdev=conf->mirrors[i].rdev) != NULL && @@ -607,13 +645,21 @@ static int make_request(request_queue_t *q, struct bio * bio) r1_bio->bios[i] = NULL; } else r1_bio->bios[i] = bio; + targets++; } else r1_bio->bios[i] = NULL; } rcu_read_unlock(); - atomic_set(&r1_bio->remaining, 1); + if (targets < conf->raid_disks) { + /* array is degraded, we will not clear the bitmap + * on I/O completion (see raid1_end_write_request) */ + set_bit(R1BIO_Degraded, &r1_bio->state); + } + + atomic_set(&r1_bio->remaining, 0); + bio_list_init(&bl); for (i = 0; i < disks; i++) { struct bio *mbio; if (!r1_bio->bios[i]) @@ -629,14 +675,23 @@ static int make_request(request_queue_t *q, struct bio * bio) mbio->bi_private = r1_bio; atomic_inc(&r1_bio->remaining); - generic_make_request(mbio); - } - if (atomic_dec_and_test(&r1_bio->remaining)) { - md_write_end(mddev); - raid_end_bio_io(r1_bio); + bio_list_add(&bl, mbio); } + bitmap_startwrite(bitmap, bio->bi_sector, r1_bio->sectors); + spin_lock_irqsave(&conf->device_lock, flags); + bio_list_merge(&conf->pending_bio_list, &bl); + bio_list_init(&bl); + + blk_plug_device(mddev->queue); + spin_unlock_irqrestore(&conf->device_lock, flags); + +#if 0 + while ((bio = bio_list_pop(&bl)) != NULL) + generic_make_request(bio); +#endif + return 0; } @@ -716,7 +771,7 @@ static void close_sync(conf_t *conf) { spin_lock_irq(&conf->resync_lock); wait_event_lock_irq(conf->wait_resume, !conf->barrier, - conf->resync_lock, unplug_slaves(conf->mddev)); + conf->resync_lock, raid1_unplug(conf->mddev->queue)); spin_unlock_irq(&conf->resync_lock); if (conf->barrier) BUG(); @@ -830,10 +885,11 @@ static int end_sync_read(struct bio *bio, unsigned int bytes_done, int error) * or re-read if the read failed. * We don't do much here, just schedule handling by raid1d */ - if (!uptodate) + if (!uptodate) { md_error(r1_bio->mddev, conf->mirrors[r1_bio->read_disk].rdev); - else + set_bit(R1BIO_Degraded, &r1_bio->state); + } else set_bit(R1BIO_Uptodate, &r1_bio->state); rdev_dec_pending(conf->mirrors[r1_bio->read_disk].rdev, conf->mddev); reschedule_retry(r1_bio); @@ -857,8 +913,10 @@ static int end_sync_write(struct bio *bio, unsigned int bytes_done, int error) mirror = i; break; } - if (!uptodate) + if (!uptodate) { md_error(mddev, conf->mirrors[mirror].rdev); + set_bit(R1BIO_Degraded, &r1_bio->state); + } update_head_pos(mirror, r1_bio); if (atomic_dec_and_test(&r1_bio->remaining)) { @@ -878,6 +936,9 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio) bio = r1_bio->bios[r1_bio->read_disk]; +/* + if (r1_bio->sector == 0) printk("First sync write startss\n"); +*/ /* * schedule writes */ @@ -905,10 +966,12 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio) atomic_inc(&conf->mirrors[i].rdev->nr_pending); atomic_inc(&r1_bio->remaining); md_sync_acct(conf->mirrors[i].rdev->bdev, wbio->bi_size >> 9); + generic_make_request(wbio); } if (atomic_dec_and_test(&r1_bio->remaining)) { + /* if we're here, all write(s) have completed, so clean up */ md_done_sync(mddev, r1_bio->sectors, 1); put_buf(r1_bio); } @@ -937,6 +1000,26 @@ static void raid1d(mddev_t *mddev) for (;;) { char b[BDEVNAME_SIZE]; spin_lock_irqsave(&conf->device_lock, flags); + + if (conf->pending_bio_list.head) { + bio = bio_list_get(&conf->pending_bio_list); + blk_remove_plug(mddev->queue); + spin_unlock_irqrestore(&conf->device_lock, flags); + /* flush any pending bitmap writes to disk before proceeding w/ I/O */ + if (bitmap_unplug(mddev->bitmap) != 0) + printk("%s: bitmap file write failed!\n", mdname(mddev)); + + while (bio) { /* submit pending writes */ + struct bio *next = bio->bi_next; + bio->bi_next = NULL; + generic_make_request(bio); + bio = next; + } + unplug = 1; + + continue; + } + if (list_empty(head)) break; r1_bio = list_entry(head->prev, r1bio_t, retry_list); @@ -1020,17 +1103,43 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i int disk; int i; int write_targets = 0; + int sync_blocks; if (!conf->r1buf_pool) + { +/* + printk("sync start - bitmap %p\n", mddev->bitmap); +*/ if (init_resync(conf)) return 0; + } max_sector = mddev->size << 1; if (sector_nr >= max_sector) { + /* If we aborted, we need to abort the + * sync on the 'current' bitmap chunk (there will + * only be one in raid1 resync. + * We can find the current addess in mddev->curr_resync + */ + if (!conf->fullsync) { + if (mddev->curr_resync < max_sector) + bitmap_end_sync(mddev->bitmap, + mddev->curr_resync, + &sync_blocks, 1); + bitmap_close_sync(mddev->bitmap); + } + if (mddev->curr_resync >= max_sector) + conf->fullsync = 0; close_sync(conf); return 0; } + if (!conf->fullsync && + !bitmap_start_sync(mddev->bitmap, sector_nr, &sync_blocks)) { + /* We can skip this block, and probably several more */ + *skipped = 1; + return sync_blocks; + } /* * If there is non-resync activity waiting for us then * put in a delay to throttle resync. @@ -1069,6 +1178,7 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i r1_bio->mddev = mddev; r1_bio->sector = sector_nr; + r1_bio->state = 0; set_bit(R1BIO_IsSync, &r1_bio->state); r1_bio->read_disk = disk; @@ -1103,6 +1213,11 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i bio->bi_bdev = conf->mirrors[i].rdev->bdev; bio->bi_private = r1_bio; } + + if (write_targets + 1 < conf->raid_disks) + /* array degraded, can't clear bitmap */ + set_bit(R1BIO_Degraded, &r1_bio->state); + if (write_targets == 0) { /* There is nowhere to write, so all non-sync * drives must be failed - so we are finished @@ -1122,6 +1237,14 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i len = (max_sector - sector_nr) << 9; if (len == 0) break; + if (!conf->fullsync && sync_blocks == 0) + if (!bitmap_start_sync(mddev->bitmap, + sector_nr, &sync_blocks)) + break; + if (sync_blocks < (PAGE_SIZE>>9)) + BUG(); + if (len > (sync_blocks<<9)) len = sync_blocks<<9; + for (i=0 ; i < conf->raid_disks; i++) { bio = r1_bio->bios[i]; if (bio->bi_end_io) { @@ -1144,6 +1267,7 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i } nr_sectors += len>>9; sector_nr += len>>9; + sync_blocks -= (len>>9); } while (r1_bio->bios[disk]->bi_vcnt < RESYNC_PAGES); bio_full: bio = r1_bio->bios[disk]; @@ -1232,6 +1356,9 @@ static int run(mddev_t *mddev) init_waitqueue_head(&conf->wait_idle); init_waitqueue_head(&conf->wait_resume); + bio_list_init(&conf->pending_bio_list); + bio_list_init(&conf->flushing_bio_list); + if (!conf->working_disks) { printk(KERN_ERR "raid1: no operational mirrors for %s\n", mdname(mddev)); @@ -1260,16 +1387,15 @@ static int run(mddev_t *mddev) conf->last_used = j; - - { - mddev->thread = md_register_thread(raid1d, mddev, "%s_raid1"); - if (!mddev->thread) { - printk(KERN_ERR - "raid1: couldn't allocate thread for %s\n", - mdname(mddev)); - goto out_free_conf; - } + mddev->thread = md_register_thread(raid1d, mddev, "%s_raid1"); + if (!mddev->thread) { + printk(KERN_ERR + "raid1: couldn't allocate thread for %s\n", + mdname(mddev)); + goto out_free_conf; } + if (mddev->bitmap) mddev->thread->timeout = mddev->bitmap->daemon_sleep * HZ; + printk(KERN_INFO "raid1: raid set %s active with %d out of %d mirrors\n", mdname(mddev), mddev->raid_disks - mddev->degraded, @@ -1394,7 +1520,7 @@ static int raid1_reshape(mddev_t *mddev, int raid_disks) spin_lock_irq(&conf->resync_lock); conf->barrier++; wait_event_lock_irq(conf->wait_idle, !conf->nr_pending, - conf->resync_lock, unplug_slaves(mddev)); + conf->resync_lock, raid1_unplug(mddev->queue)); spin_unlock_irq(&conf->resync_lock); /* ok, everything is stopped */ diff --git a/include/linux/raid/raid1.h b/include/linux/raid/raid1.h index abbfdd9afe1e..9d93cf12e890 100644 --- a/include/linux/raid/raid1.h +++ b/include/linux/raid/raid1.h @@ -36,12 +36,21 @@ struct r1_private_data_s { spinlock_t device_lock; struct list_head retry_list; + /* queue pending writes and submit them on unplug */ + struct bio_list pending_bio_list; + /* queue of writes that have been unplugged */ + struct bio_list flushing_bio_list; + /* for use when syncing mirrors: */ spinlock_t resync_lock; - int nr_pending; - int barrier; + int nr_pending; + int barrier; sector_t next_resync; + int fullsync; /* set to 1 if a full sync is needed, + * (fresh device added). + * Cleared when a sync completes. + */ wait_queue_head_t wait_idle; wait_queue_head_t wait_resume; @@ -85,14 +94,17 @@ struct r1bio_s { int read_disk; struct list_head retry_list; + struct bitmap_update *bitmap_update; /* * if the IO is in WRITE direction, then multiple bios are used. * We choose the number when they are allocated. */ struct bio *bios[0]; + /* DO NOT PUT ANY NEW FIELDS HERE - bios array is contiguously alloced*/ }; /* bits for r1bio.state */ #define R1BIO_Uptodate 0 #define R1BIO_IsSync 1 +#define R1BIO_Degraded 2 #endif -- cgit v1.2.3 From 41158c7eb22312cfaa256744e1553bb4042ff085 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:25 -0700 Subject: [PATCH] md: optimise reconstruction when re-adding a recently failed drive. When an array is degraded, bit in the intent-bitmap are never cleared. So if a recently failed drive is re-added, we only need to reconstruct the block that are still reflected in the bitmap. This patch adds support for this re-adding. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 71 ++++++++++++++++++++++++++++++++++++----------- drivers/md/raid1.c | 7 ++++- include/linux/raid/md_k.h | 4 +++ 3 files changed, 65 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/md.c b/drivers/md/md.c index b02f8d1d77e7..789b114f860a 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -577,6 +577,8 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev) mdp_disk_t *desc; mdp_super_t *sb = (mdp_super_t *)page_address(rdev->sb_page); + rdev->raid_disk = -1; + rdev->in_sync = 0; if (mddev->raid_disks == 0) { mddev->major_version = 0; mddev->minor_version = sb->minor_version; @@ -607,16 +609,24 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev) memcpy(mddev->uuid+12,&sb->set_uuid3, 4); mddev->max_disks = MD_SB_DISKS; - } else { - __u64 ev1; - ev1 = md_event(sb); + } else if (mddev->pers == NULL) { + /* Insist on good event counter while assembling */ + __u64 ev1 = md_event(sb); ++ev1; if (ev1 < mddev->events) return -EINVAL; - } + } else if (mddev->bitmap) { + /* if adding to array with a bitmap, then we can accept an + * older device ... but not too old. + */ + __u64 ev1 = md_event(sb); + if (ev1 < mddev->bitmap->events_cleared) + return 0; + } else /* just a hot-add of a new device, leave raid_disk at -1 */ + return 0; + if (mddev->level != LEVEL_MULTIPATH) { - rdev->raid_disk = -1; - rdev->in_sync = rdev->faulty = 0; + rdev->faulty = 0; desc = sb->disks + rdev->desc_nr; if (desc->state & (1<in_sync = 1; rdev->raid_disk = desc->raid_disk; } - } + } else /* MULTIPATH are always insync */ + rdev->in_sync = 1; return 0; } @@ -868,6 +879,8 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev) { struct mdp_superblock_1 *sb = (struct mdp_superblock_1*)page_address(rdev->sb_page); + rdev->raid_disk = -1; + rdev->in_sync = 0; if (mddev->raid_disks == 0) { mddev->major_version = 1; mddev->patch_version = 0; @@ -885,13 +898,21 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev) memcpy(mddev->uuid, sb->set_uuid, 16); mddev->max_disks = (4096-256)/2; - } else { - __u64 ev1; - ev1 = le64_to_cpu(sb->events); + } else if (mddev->pers == NULL) { + /* Insist of good event counter while assembling */ + __u64 ev1 = le64_to_cpu(sb->events); ++ev1; if (ev1 < mddev->events) return -EINVAL; - } + } else if (mddev->bitmap) { + /* If adding to array with a bitmap, then we can accept an + * older device, but not too old. + */ + __u64 ev1 = le64_to_cpu(sb->events); + if (ev1 < mddev->bitmap->events_cleared) + return 0; + } else /* just a hot-add of a new device, leave raid_disk at -1 */ + return 0; if (mddev->level != LEVEL_MULTIPATH) { int role; @@ -899,14 +920,10 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev) role = le16_to_cpu(sb->dev_roles[rdev->desc_nr]); switch(role) { case 0xffff: /* spare */ - rdev->in_sync = 0; rdev->faulty = 0; - rdev->raid_disk = -1; break; case 0xfffe: /* faulty */ - rdev->in_sync = 0; rdev->faulty = 1; - rdev->raid_disk = -1; break; default: rdev->in_sync = 1; @@ -914,7 +931,9 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev) rdev->raid_disk = role; break; } - } + } else /* MULTIPATH are always insync */ + rdev->in_sync = 1; + return 0; } @@ -2155,6 +2174,18 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info) PTR_ERR(rdev)); return PTR_ERR(rdev); } + /* set save_raid_disk if appropriate */ + if (!mddev->persistent) { + if (info->state & (1<raid_disk < mddev->raid_disks) + rdev->raid_disk = info->raid_disk; + else + rdev->raid_disk = -1; + } else + super_types[mddev->major_version]. + validate_super(mddev, rdev); + rdev->saved_raid_disk = rdev->raid_disk; + rdev->in_sync = 0; /* just to be sure */ rdev->raid_disk = -1; err = bind_rdev_to_array(rdev, mddev); @@ -3706,6 +3737,14 @@ void md_check_recovery(mddev_t *mddev) mddev->pers->spare_active(mddev); } md_update_sb(mddev); + + /* if array is no-longer degraded, then any saved_raid_disk + * information must be scrapped + */ + if (!mddev->degraded) + ITERATE_RDEV(mddev,rdev,rtmp) + rdev->saved_raid_disk = -1; + mddev->recovery = 0; /* flag recovery needed just to double check */ set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index c3b4772cfaea..3f5234fe3593 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -811,9 +811,12 @@ static int raid1_add_disk(mddev_t *mddev, mdk_rdev_t *rdev) { conf_t *conf = mddev->private; int found = 0; - int mirror; + int mirror = 0; mirror_info_t *p; + if (rdev->saved_raid_disk >= 0 && + conf->mirrors[rdev->saved_raid_disk].rdev == NULL) + mirror = rdev->saved_raid_disk; for (mirror=0; mirror < mddev->raid_disks; mirror++) if ( !(p=conf->mirrors+mirror)->rdev) { @@ -830,6 +833,8 @@ static int raid1_add_disk(mddev_t *mddev, mdk_rdev_t *rdev) p->head_position = 0; rdev->raid_disk = mirror; found = 1; + if (rdev->saved_raid_disk != mirror) + conf->fullsync = 1; p->rdev = rdev; break; } diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h index 16e94a9f0f8c..6cdcb4434c6c 100644 --- a/include/linux/raid/md_k.h +++ b/include/linux/raid/md_k.h @@ -183,6 +183,10 @@ struct mdk_rdev_s int desc_nr; /* descriptor index in the superblock */ int raid_disk; /* role of device in array */ + int saved_raid_disk; /* role that device used to have in the + * array and could again if we did a partial + * resync from the bitmap + */ atomic_t nr_pending; /* number of pending requests. * only maintained for arrays that -- cgit v1.2.3 From 3d310eb7b3df1252e8595d059d982b0a9825a137 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:26 -0700 Subject: [PATCH] md: fix deadlock due to md thread processing delayed requests. Before completing a 'write' the md superblock might need to be updated. This is best done by the md_thread. The current code schedules this up and queues the write request for later handling by the md_thread. However some personalities (Raid5/raid6) will deadlock if the md_thread tries to submit requests to its own array. So this patch changes things so the processes submitting the request waits for the superblock to be written and then submits the request itself. This fixes a recently-created deadlock in raid5/raid6 Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 45 ++++++++++++++++----------------------------- drivers/md/raid1.c | 4 ++-- drivers/md/raid10.c | 3 +-- drivers/md/raid5.c | 3 +-- drivers/md/raid6main.c | 3 +-- include/linux/raid/md.h | 2 +- include/linux/raid/md_k.h | 2 +- 7 files changed, 23 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/md.c b/drivers/md/md.c index 789b114f860a..7075bebb7f37 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -224,8 +224,8 @@ static mddev_t * mddev_find(dev_t unit) INIT_LIST_HEAD(&new->all_mddevs); init_timer(&new->safemode_timer); atomic_set(&new->active, 1); - bio_list_init(&new->write_list); spin_lock_init(&new->write_lock); + init_waitqueue_head(&new->sb_wait); new->queue = blk_alloc_queue(GFP_KERNEL); if (!new->queue) { @@ -1307,6 +1307,7 @@ repeat: if (!mddev->persistent) { mddev->sb_dirty = 0; spin_unlock(&mddev->write_lock); + wake_up(&mddev->sb_wait); return; } spin_unlock(&mddev->write_lock); @@ -1348,6 +1349,7 @@ repeat: } mddev->sb_dirty = 0; spin_unlock(&mddev->write_lock); + wake_up(&mddev->sb_wait); } @@ -3368,29 +3370,26 @@ void md_done_sync(mddev_t *mddev, int blocks, int ok) /* md_write_start(mddev, bi) * If we need to update some array metadata (e.g. 'active' flag - * in superblock) before writing, queue bi for later writing - * and return 0, else return 1 and it will be written now + * in superblock) before writing, schedule a superblock update + * and wait for it to complete. */ -int md_write_start(mddev_t *mddev, struct bio *bi) +void md_write_start(mddev_t *mddev, struct bio *bi) { + DEFINE_WAIT(w); if (bio_data_dir(bi) != WRITE) - return 1; + return; atomic_inc(&mddev->writes_pending); - spin_lock(&mddev->write_lock); - if (mddev->in_sync == 0 && mddev->sb_dirty == 0) { - spin_unlock(&mddev->write_lock); - return 1; - } - bio_list_add(&mddev->write_list, bi); - if (mddev->in_sync) { - mddev->in_sync = 0; - mddev->sb_dirty = 1; + spin_lock(&mddev->write_lock); + if (mddev->in_sync) { + mddev->in_sync = 0; + mddev->sb_dirty = 1; + md_wakeup_thread(mddev->thread); + } + spin_unlock(&mddev->write_lock); } - spin_unlock(&mddev->write_lock); - md_wakeup_thread(mddev->thread); - return 0; + wait_event(mddev->sb_wait, mddev->sb_dirty==0); } void md_write_end(mddev_t *mddev) @@ -3685,7 +3684,6 @@ void md_check_recovery(mddev_t *mddev) mddev->sb_dirty || test_bit(MD_RECOVERY_NEEDED, &mddev->recovery) || test_bit(MD_RECOVERY_DONE, &mddev->recovery) || - mddev->write_list.head || (mddev->safemode == 1) || (mddev->safemode == 2 && ! atomic_read(&mddev->writes_pending) && !mddev->in_sync && mddev->recovery_cp == MaxSector) @@ -3694,7 +3692,6 @@ void md_check_recovery(mddev_t *mddev) if (mddev_trylock(mddev)==0) { int spares =0; - struct bio *blist; spin_lock(&mddev->write_lock); if (mddev->safemode && !atomic_read(&mddev->writes_pending) && @@ -3704,21 +3701,11 @@ void md_check_recovery(mddev_t *mddev) } if (mddev->safemode == 1) mddev->safemode = 0; - blist = bio_list_get(&mddev->write_list); spin_unlock(&mddev->write_lock); if (mddev->sb_dirty) md_update_sb(mddev); - while (blist) { - struct bio *b = blist; - blist = blist->bi_next; - b->bi_next = NULL; - generic_make_request(b); - /* we already counted this, so need to un-count */ - md_write_end(mddev); - } - if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) && !test_bit(MD_RECOVERY_DONE, &mddev->recovery)) { diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 3f5234fe3593..98b09773e79e 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -561,8 +561,8 @@ static int make_request(request_queue_t *q, struct bio * bio) * thread has put up a bar for new requests. * Continue immediately if no resync is active currently. */ - if (md_write_start(mddev, bio)==0) - return 0; + md_write_start(mddev, bio); /* wait on superblock update early */ + spin_lock_irq(&conf->resync_lock); wait_event_lock_irq(conf->wait_resume, !conf->barrier, conf->resync_lock, ); conf->nr_pending++; diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 8476515bfdc7..fd7324a86d13 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -700,8 +700,7 @@ static int make_request(request_queue_t *q, struct bio * bio) return 0; } - if (md_write_start(mddev, bio) == 0) - return 0; + md_write_start(mddev, bio); /* * Register the new request and wait if the reconstruction diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 1ce3f5aaa984..93a9726cc2d6 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -1411,8 +1411,7 @@ static int make_request (request_queue_t *q, struct bio * bi) sector_t logical_sector, last_sector; struct stripe_head *sh; - if (md_write_start(mddev, bi)==0) - return 0; + md_write_start(mddev, bi); if (bio_data_dir(bi)==WRITE) { disk_stat_inc(mddev->gendisk, writes); diff --git a/drivers/md/raid6main.c b/drivers/md/raid6main.c index d9c385496dc5..f62ea1a73d0d 100644 --- a/drivers/md/raid6main.c +++ b/drivers/md/raid6main.c @@ -1570,8 +1570,7 @@ static int make_request (request_queue_t *q, struct bio * bi) sector_t logical_sector, last_sector; struct stripe_head *sh; - if (md_write_start(mddev, bi)==0) - return 0; + md_write_start(mddev, bi); if (bio_data_dir(bi)==WRITE) { disk_stat_inc(mddev->gendisk, writes); diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h index cfde8f497d6d..75f41d8faed2 100644 --- a/include/linux/raid/md.h +++ b/include/linux/raid/md.h @@ -69,7 +69,7 @@ extern mdk_thread_t * md_register_thread (void (*run) (mddev_t *mddev), extern void md_unregister_thread (mdk_thread_t *thread); extern void md_wakeup_thread(mdk_thread_t *thread); extern void md_check_recovery(mddev_t *mddev); -extern int md_write_start(mddev_t *mddev, struct bio *bi); +extern void md_write_start(mddev_t *mddev, struct bio *bi); extern void md_write_end(mddev_t *mddev); extern void md_handle_safemode(mddev_t *mddev); extern void md_done_sync(mddev_t *mddev, int blocks, int ok); diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h index 6cdcb4434c6c..3e977025cf43 100644 --- a/include/linux/raid/md_k.h +++ b/include/linux/raid/md_k.h @@ -261,7 +261,7 @@ struct mddev_s sector_t recovery_cp; spinlock_t write_lock; - struct bio_list write_list; + wait_queue_head_t sb_wait; /* for waiting on superblock updates */ unsigned int safemode; /* if set, update "clean" superblock * when no writes pending. -- cgit v1.2.3 From a654b9d8f851f4ca02649d5825cbe6c608adb10c Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:27 -0700 Subject: [PATCH] md: allow md intent bitmap to be stored near the superblock. This provides an alternate to storing the bitmap in a separate file. The bitmap can be stored at a given offset from the superblock. Obviously the creator of the array must make sure this doesn't intersect with data.... After is good for version-0.90 superblocks. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/bitmap.c | 132 +++++++++++++++++++++++++++++++++++--------- drivers/md/md.c | 40 +++++++++++++- include/linux/raid/bitmap.h | 2 + include/linux/raid/md.h | 15 ++++- include/linux/raid/md_k.h | 4 ++ include/linux/raid/md_p.h | 7 ++- 6 files changed, 170 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 204564dc6a0d..030d6861051a 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -116,7 +116,7 @@ static unsigned char *bitmap_alloc_page(struct bitmap *bitmap) if (!page) printk("%s: bitmap_alloc_page FAILED\n", bmname(bitmap)); else - printk("%s: bitmap_alloc_page: allocated page at %p\n", + PRINTK("%s: bitmap_alloc_page: allocated page at %p\n", bmname(bitmap), page); return page; } @@ -258,13 +258,61 @@ char *file_path(struct file *file, char *buf, int count) * basic page I/O operations */ +/* IO operations when bitmap is stored near all superblocks */ +static struct page *read_sb_page(mddev_t *mddev, long offset, unsigned long index) +{ + /* choose a good rdev and read the page from there */ + + mdk_rdev_t *rdev; + struct list_head *tmp; + struct page *page = alloc_page(GFP_KERNEL); + sector_t target; + + if (!page) + return ERR_PTR(-ENOMEM); + do { + ITERATE_RDEV(mddev, rdev, tmp) + if (rdev->in_sync && !rdev->faulty) + goto found; + return ERR_PTR(-EIO); + + found: + target = (rdev->sb_offset << 1) + offset + index * (PAGE_SIZE/512); + + } while (!sync_page_io(rdev->bdev, target, PAGE_SIZE, page, READ)); + + page->index = index; + return page; +} + +static int write_sb_page(mddev_t *mddev, long offset, struct page *page, int wait) +{ + mdk_rdev_t *rdev; + struct list_head *tmp; + + ITERATE_RDEV(mddev, rdev, tmp) + if (rdev->in_sync && !rdev->faulty) + md_super_write(mddev, rdev, + (rdev->sb_offset<<1) + offset + + page->index * (PAGE_SIZE/512), + PAGE_SIZE, + page); + + if (wait) + wait_event(mddev->sb_wait, atomic_read(&mddev->pending_writes)==0); + return 0; +} + /* - * write out a page + * write out a page to a file */ static int write_page(struct bitmap *bitmap, struct page *page, int wait) { int ret = -ENOMEM; + if (bitmap->file == NULL) + return write_sb_page(bitmap->mddev, bitmap->offset, page, wait); + lock_page(page); ret = page->mapping->a_ops->prepare_write(NULL, page, 0, PAGE_SIZE); @@ -394,7 +442,12 @@ static int bitmap_read_sb(struct bitmap *bitmap) int err = -EINVAL; /* page 0 is the superblock, read it... */ - bitmap->sb_page = read_page(bitmap->file, 0, &bytes_read); + if (bitmap->file) + bitmap->sb_page = read_page(bitmap->file, 0, &bytes_read); + else { + bitmap->sb_page = read_sb_page(bitmap->mddev, bitmap->offset, 0); + bytes_read = PAGE_SIZE; + } if (IS_ERR(bitmap->sb_page)) { err = PTR_ERR(bitmap->sb_page); bitmap->sb_page = NULL; @@ -625,14 +678,16 @@ static void bitmap_file_kick(struct bitmap *bitmap) bitmap_mask_state(bitmap, BITMAP_STALE, MASK_SET); bitmap_update_sb(bitmap); - path = kmalloc(PAGE_SIZE, GFP_KERNEL); - if (path) - ptr = file_path(bitmap->file, path, PAGE_SIZE); + if (bitmap->file) { + path = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (path) + ptr = file_path(bitmap->file, path, PAGE_SIZE); - printk(KERN_ALERT "%s: kicking failed bitmap file %s from array!\n", - bmname(bitmap), ptr ? ptr : ""); + printk(KERN_ALERT "%s: kicking failed bitmap file %s from array!\n", + bmname(bitmap), ptr ? ptr : ""); - kfree(path); + kfree(path); + } bitmap_file_put(bitmap); @@ -676,7 +731,7 @@ static void bitmap_file_set_bit(struct bitmap *bitmap, sector_t block) void *kaddr; unsigned long chunk = block >> CHUNK_BLOCK_SHIFT(bitmap); - if (!bitmap->file || !bitmap->filemap) { + if (!bitmap->filemap) { return; } @@ -715,7 +770,7 @@ int bitmap_unplug(struct bitmap *bitmap) * flushed out to disk */ for (i = 0; i < bitmap->file_pages; i++) { spin_lock_irqsave(&bitmap->lock, flags); - if (!bitmap->file || !bitmap->filemap) { + if (!bitmap->filemap) { spin_unlock_irqrestore(&bitmap->lock, flags); return 0; } @@ -732,11 +787,15 @@ int bitmap_unplug(struct bitmap *bitmap) return 1; } if (wait) { /* if any writes were performed, we need to wait on them */ - spin_lock_irq(&bitmap->write_lock); - wait_event_lock_irq(bitmap->write_wait, - list_empty(&bitmap->complete_pages), bitmap->write_lock, - wake_up_process(bitmap->writeback_daemon->tsk)); - spin_unlock_irq(&bitmap->write_lock); + if (bitmap->file) { + spin_lock_irq(&bitmap->write_lock); + wait_event_lock_irq(bitmap->write_wait, + list_empty(&bitmap->complete_pages), bitmap->write_lock, + wake_up_process(bitmap->writeback_daemon->tsk)); + spin_unlock_irq(&bitmap->write_lock); + } else + wait_event(bitmap->mddev->sb_wait, + atomic_read(&bitmap->mddev->pending_writes)==0); } return 0; } @@ -764,7 +823,7 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, int in_sync) chunks = bitmap->chunks; file = bitmap->file; - BUG_ON(!file); + BUG_ON(!file && !bitmap->offset); #if INJECT_FAULTS_3 outofdate = 1; @@ -779,7 +838,7 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, int in_sync) num_pages = (bytes + sizeof(bitmap_super_t) + PAGE_SIZE - 1) / PAGE_SIZE; - if (i_size_read(file->f_mapping->host) < bytes + sizeof(bitmap_super_t)) { + if (file && i_size_read(file->f_mapping->host) < bytes + sizeof(bitmap_super_t)) { printk(KERN_INFO "%s: bitmap file too short %lu < %lu\n", bmname(bitmap), (unsigned long) i_size_read(file->f_mapping->host), @@ -816,14 +875,18 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, int in_sync) */ page = bitmap->sb_page; offset = sizeof(bitmap_super_t); - } else { + } else if (file) { page = read_page(file, index, &dummy); - if (IS_ERR(page)) { /* read error */ - ret = PTR_ERR(page); - goto out; - } + offset = 0; + } else { + page = read_sb_page(bitmap->mddev, bitmap->offset, index); offset = 0; } + if (IS_ERR(page)) { /* read error */ + ret = PTR_ERR(page); + goto out; + } + oldindex = index; oldpage = page; kmap(page); @@ -874,6 +937,19 @@ out: return ret; } +void bitmap_write_all(struct bitmap *bitmap) +{ + /* We don't actually write all bitmap blocks here, + * just flag them as needing to be written + */ + + unsigned long chunks = bitmap->chunks; + unsigned long bytes = (chunks+7)/8 + sizeof(bitmap_super_t); + unsigned long num_pages = (bytes + PAGE_SIZE-1) / PAGE_SIZE; + while (num_pages--) + bitmap->filemap_attr[num_pages] |= BITMAP_PAGE_NEEDWRITE; +} + static void bitmap_count_page(struct bitmap *bitmap, sector_t offset, int inc) { @@ -913,7 +989,7 @@ int bitmap_daemon_work(struct bitmap *bitmap) for (j = 0; j < bitmap->chunks; j++) { bitmap_counter_t *bmc; spin_lock_irqsave(&bitmap->lock, flags); - if (!bitmap->file || !bitmap->filemap) { + if (!bitmap->filemap) { /* error or shutdown */ spin_unlock_irqrestore(&bitmap->lock, flags); break; @@ -1072,6 +1148,7 @@ static int bitmap_start_daemon(struct bitmap *bitmap, mdk_thread_t **ptr, spin_lock_irqsave(&bitmap->lock, flags); *ptr = NULL; + if (!bitmap->file) /* no need for daemon if there's no backing file */ goto out_unlock; @@ -1416,9 +1493,11 @@ int bitmap_create(mddev_t *mddev) BUG_ON(sizeof(bitmap_super_t) != 256); - if (!file) /* bitmap disabled, nothing to do */ + if (!file && !mddev->bitmap_offset) /* bitmap disabled, nothing to do */ return 0; + BUG_ON(file && mddev->bitmap_offset); + bitmap = kmalloc(sizeof(*bitmap), GFP_KERNEL); if (!bitmap) return -ENOMEM; @@ -1438,7 +1517,8 @@ int bitmap_create(mddev_t *mddev) return -ENOMEM; bitmap->file = file; - get_file(file); + bitmap->offset = mddev->bitmap_offset; + if (file) get_file(file); /* read superblock from bitmap file (this sets bitmap->chunksize) */ err = bitmap_read_sb(bitmap); if (err) diff --git a/drivers/md/md.c b/drivers/md/md.c index 7075bebb7f37..fde8acfac320 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -337,7 +337,7 @@ static int bi_complete(struct bio *bio, unsigned int bytes_done, int error) return 0; } -static int sync_page_io(struct block_device *bdev, sector_t sector, int size, +int sync_page_io(struct block_device *bdev, sector_t sector, int size, struct page *page, int rw) { struct bio *bio = bio_alloc(GFP_NOIO, 1); @@ -609,6 +609,17 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev) memcpy(mddev->uuid+12,&sb->set_uuid3, 4); mddev->max_disks = MD_SB_DISKS; + + if (sb->state & (1<bitmap_file == NULL) { + if (mddev->level != 1) { + /* FIXME use a better test */ + printk(KERN_WARNING "md: bitmaps only support for raid1\n"); + return -EINVAL; + } + mddev->bitmap_offset = (MD_SB_BYTES >> 9); + } + } else if (mddev->pers == NULL) { /* Insist on good event counter while assembling */ __u64 ev1 = md_event(sb); @@ -702,6 +713,9 @@ static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev) sb->layout = mddev->layout; sb->chunk_size = mddev->chunk_size; + if (mddev->bitmap && mddev->bitmap_file == NULL) + sb->state |= (1<disks[0].state = (1<uuid, sb->set_uuid, 16); mddev->max_disks = (4096-256)/2; + + if ((le32_to_cpu(sb->feature_map) & 1) && + mddev->bitmap_file == NULL ) { + if (mddev->level != 1) { + printk(KERN_WARNING "md: bitmaps only supported for raid1\n"); + return -EINVAL; + } + mddev->bitmap_offset = (__s32)le32_to_cpu(sb->bitmap_offset); + } } else if (mddev->pers == NULL) { /* Insist of good event counter while assembling */ __u64 ev1 = le64_to_cpu(sb->events); @@ -960,6 +983,11 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev) else sb->resync_offset = cpu_to_le64(0); + if (mddev->bitmap && mddev->bitmap_file == NULL) { + sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_offset); + sb->feature_map = cpu_to_le32(1); + } + max_dev = 0; ITERATE_RDEV(mddev,rdev2,tmp) if (rdev2->desc_nr+1 > max_dev) @@ -2406,7 +2434,8 @@ static int set_bitmap_file(mddev_t *mddev, int fd) mdname(mddev)); fput(mddev->bitmap_file); mddev->bitmap_file = NULL; - } + } else + mddev->bitmap_offset = 0; /* file overrides offset */ return err; } @@ -3774,6 +3803,13 @@ void md_check_recovery(mddev_t *mddev) set_bit(MD_RECOVERY_RUNNING, &mddev->recovery); if (!spares) set_bit(MD_RECOVERY_SYNC, &mddev->recovery); + if (spares && mddev->bitmap && ! mddev->bitmap->file) { + /* We are adding a device or devices to an array + * which has the bitmap stored on all devices. + * So make sure all bitmap pages get written + */ + bitmap_write_all(mddev->bitmap); + } mddev->sync_thread = md_register_thread(md_do_sync, mddev, "%s_resync"); diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h index cfe60cfc8f3d..e24b74b11150 100644 --- a/include/linux/raid/bitmap.h +++ b/include/linux/raid/bitmap.h @@ -217,6 +217,7 @@ struct bitmap { /* bitmap spinlock */ spinlock_t lock; + long offset; /* offset from superblock if file is NULL */ struct file *file; /* backing disk file */ struct page *sb_page; /* cached copy of the bitmap file superblock */ struct page **filemap; /* list of cache pages for the file */ @@ -255,6 +256,7 @@ void bitmap_print_sb(struct bitmap *bitmap); int bitmap_update_sb(struct bitmap *bitmap); int bitmap_setallbits(struct bitmap *bitmap); +void bitmap_write_all(struct bitmap *bitmap); /* these are exported */ int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors); diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h index 75f41d8faed2..ffa316ce4dc8 100644 --- a/include/linux/raid/md.h +++ b/include/linux/raid/md.h @@ -60,7 +60,14 @@ */ #define MD_MAJOR_VERSION 0 #define MD_MINOR_VERSION 90 -#define MD_PATCHLEVEL_VERSION 1 +/* + * MD_PATCHLEVEL_VERSION indicates kernel functionality. + * >=1 means different superblock formats are selectable using SET_ARRAY_INFO + * and major_version/minor_version accordingly + * >=2 means that Internal bitmaps are supported by setting MD_SB_BITMAP_PRESENT + * in the super status byte + */ +#define MD_PATCHLEVEL_VERSION 2 extern int register_md_personality (int p_num, mdk_personality_t *p); extern int unregister_md_personality (int p_num); @@ -78,6 +85,12 @@ extern void md_unplug_mddev(mddev_t *mddev); extern void md_print_devices (void); +extern void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev, + sector_t sector, int size, struct page *page); +extern int sync_page_io(struct block_device *bdev, sector_t sector, int size, + struct page *page, int rw); + + #define MD_BUG(x...) { printk("md: bug in file %s, line %d\n", __FILE__, __LINE__); md_print_devices(); } #endif diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h index 3e977025cf43..a3725b57fb7d 100644 --- a/include/linux/raid/md_k.h +++ b/include/linux/raid/md_k.h @@ -273,6 +273,10 @@ struct mddev_s struct bitmap *bitmap; /* the bitmap for the device */ struct file *bitmap_file; /* the bitmap file */ + long bitmap_offset; /* offset from superblock of + * start of bitmap. May be + * negative, but not '0' + */ struct list_head all_mddevs; }; diff --git a/include/linux/raid/md_p.h b/include/linux/raid/md_p.h index 8ba95d67329f..8e592a25a8b5 100644 --- a/include/linux/raid/md_p.h +++ b/include/linux/raid/md_p.h @@ -96,6 +96,7 @@ typedef struct mdp_device_descriptor_s { #define MD_SB_CLEAN 0 #define MD_SB_ERRORS 1 +#define MD_SB_BITMAP_PRESENT 8 /* bitmap may be present nearby */ typedef struct mdp_superblock_s { /* * Constant generic information @@ -184,7 +185,7 @@ struct mdp_superblock_1 { /* constant array information - 128 bytes */ __u32 magic; /* MD_SB_MAGIC: 0xa92b4efc - little endian */ __u32 major_version; /* 1 */ - __u32 feature_map; /* 0 for now */ + __u32 feature_map; /* bit 0 set if 'bitmap_offset' is meaningful */ __u32 pad0; /* always set to 0 when writing */ __u8 set_uuid[16]; /* user-space generated. */ @@ -197,6 +198,10 @@ struct mdp_superblock_1 { __u32 chunksize; /* in 512byte sectors */ __u32 raid_disks; + __u32 bitmap_offset; /* sectors after start of superblock that bitmap starts + * NOTE: signed, so bitmap can be before superblock + * only meaningful of feature_map[0] is set. + */ __u8 pad1[128-96]; /* set to 0 when written */ /* constant this-device information - 64 bytes */ -- cgit v1.2.3 From 7bfa19f2748000d646dbdf8f48258cfe1d257b52 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:28 -0700 Subject: [PATCH] md: allow md to update multiple superblocks in parallel. currently, md updates all superblocks (one on each device) in series. It waits for one write to complete before starting the next. This isn't a big problem as superblock updates don't happen that often. However it is neater to do it in parallel, and if the drives in the array have gone to "sleep" after a period of idleness, then waking them is parallel is faster (and someone else should be worrying about power drain). Futher, we will need parallel superblock updates for a future patch which keeps the intent-logging bitmap near the superblock. Also remove the silly code that retired superblock updates 100 times. This simply never made sense. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 85 ++++++++++++++++++++++++++--------------------- include/linux/raid/md_k.h | 1 + 2 files changed, 49 insertions(+), 37 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/md.c b/drivers/md/md.c index fde8acfac320..ef3ad99562c1 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -328,6 +328,40 @@ static void free_disk_sb(mdk_rdev_t * rdev) } +static int super_written(struct bio *bio, unsigned int bytes_done, int error) +{ + mdk_rdev_t *rdev = bio->bi_private; + if (bio->bi_size) + return 1; + + if (error || !test_bit(BIO_UPTODATE, &bio->bi_flags)) + md_error(rdev->mddev, rdev); + + if (atomic_dec_and_test(&rdev->mddev->pending_writes)) + wake_up(&rdev->mddev->sb_wait); + return 0; +} + +void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev, + sector_t sector, int size, struct page *page) +{ + /* write first size bytes of page to sector of rdev + * Increment mddev->pending_writes before returning + * and decrement it on completion, waking up sb_wait + * if zero is reached. + * If an error occurred, call md_error + */ + struct bio *bio = bio_alloc(GFP_NOIO, 1); + + bio->bi_bdev = rdev->bdev; + bio->bi_sector = sector; + bio_add_page(bio, page, size, 0); + bio->bi_private = rdev; + bio->bi_end_io = super_written; + atomic_inc(&mddev->pending_writes); + submit_bio((1<bi_size) @@ -1268,30 +1302,6 @@ void md_print_devices(void) } -static int write_disk_sb(mdk_rdev_t * rdev) -{ - char b[BDEVNAME_SIZE]; - if (!rdev->sb_loaded) { - MD_BUG(); - return 1; - } - if (rdev->faulty) { - MD_BUG(); - return 1; - } - - dprintk(KERN_INFO "(write) %s's sb offset: %llu\n", - bdevname(rdev->bdev,b), - (unsigned long long)rdev->sb_offset); - - if (sync_page_io(rdev->bdev, rdev->sb_offset<<1, MD_SB_BYTES, rdev->sb_page, WRITE)) - return 0; - - printk("md: write_disk_sb failed for device %s\n", - bdevname(rdev->bdev,b)); - return 1; -} - static void sync_sbs(mddev_t * mddev) { mdk_rdev_t *rdev; @@ -1306,7 +1316,7 @@ static void sync_sbs(mddev_t * mddev) static void md_update_sb(mddev_t * mddev) { - int err, count = 100; + int err; struct list_head *tmp; mdk_rdev_t *rdev; int sync_req; @@ -1326,6 +1336,7 @@ repeat: MD_BUG(); mddev->events --; } + mddev->sb_dirty = 2; sync_sbs(mddev); /* @@ -1353,24 +1364,24 @@ repeat: dprintk("%s ", bdevname(rdev->bdev,b)); if (!rdev->faulty) { - err += write_disk_sb(rdev); + md_super_write(mddev,rdev, + rdev->sb_offset<<1, MD_SB_BYTES, + rdev->sb_page); + dprintk(KERN_INFO "(write) %s's sb offset: %llu\n", + bdevname(rdev->bdev,b), + (unsigned long long)rdev->sb_offset); + } else dprintk(")\n"); - if (!err && mddev->level == LEVEL_MULTIPATH) + if (mddev->level == LEVEL_MULTIPATH) /* only need to write one superblock... */ break; } - if (err) { - if (--count) { - printk(KERN_ERR "md: errors occurred during superblock" - " update, repeating\n"); - goto repeat; - } - printk(KERN_ERR \ - "md: excessive errors occurred during superblock update, exiting\n"); - } + wait_event(mddev->sb_wait, atomic_read(&mddev->pending_writes)==0); + /* if there was a failure, sb_dirty was set to 1, and we re-write super */ + spin_lock(&mddev->write_lock); - if (mddev->in_sync != sync_req) { + if (mddev->in_sync != sync_req|| mddev->sb_dirty == 1) { /* have to write it out again */ spin_unlock(&mddev->write_lock); goto repeat; diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h index a3725b57fb7d..8c14ba565a45 100644 --- a/include/linux/raid/md_k.h +++ b/include/linux/raid/md_k.h @@ -262,6 +262,7 @@ struct mddev_s spinlock_t write_lock; wait_queue_head_t sb_wait; /* for waiting on superblock updates */ + atomic_t pending_writes; /* number of active superblock writes */ unsigned int safemode; /* if set, update "clean" superblock * when no writes pending. -- cgit v1.2.3 From 39730960d94306d7be414e8d54f4e5c071af1278 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Jun 2005 17:17:28 -0700 Subject: [PATCH] Two small fixes for md verion-1 superblocks. 1/ Must typecast int to (sector_t) before inverting or we might not invert enough bits. 2/ When "bitmap_offset" was added to mdp_superblock_1, we didn't increase the count of words-used (96 to 100). Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 2 +- include/linux/raid/md_p.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/md.c b/drivers/md/md.c index ef3ad99562c1..65fab74ad737 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -847,7 +847,7 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version) case 0: sb_offset = rdev->bdev->bd_inode->i_size >> 9; sb_offset -= 8*2; - sb_offset &= ~(4*2-1); + sb_offset &= ~(sector_t)(4*2-1); /* convert from sectors to K */ sb_offset /= 2; break; diff --git a/include/linux/raid/md_p.h b/include/linux/raid/md_p.h index 8e592a25a8b5..dc65cd435494 100644 --- a/include/linux/raid/md_p.h +++ b/include/linux/raid/md_p.h @@ -202,7 +202,7 @@ struct mdp_superblock_1 { * NOTE: signed, so bitmap can be before superblock * only meaningful of feature_map[0] is set. */ - __u8 pad1[128-96]; /* set to 0 when written */ + __u8 pad1[128-100]; /* set to 0 when written */ /* constant this-device information - 64 bytes */ __u64 data_offset; /* sector start of data, often 0 */ -- cgit v1.2.3 From b3d5496ea5915fa4848fe307af9f7097f312e932 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sat, 2 Apr 2005 20:31:02 +0200 Subject: [PATCH] I2C: Kill address ranges in non-sensors i2c chip drivers Some months ago, you killed the address ranges mechanism from all sensors i2c chip drivers (both the module parameters and the in-code address lists). I think it was a very good move, as the ranges can easily be replaced by individual addresses, and this allowed for significant cleanups in the i2c core (let alone the impressive size shrink for all these drivers). Unfortunately you did not do the same for non-sensors i2c chip drivers. These need the address ranges even less, so we could get rid of the ranges here as well for another significant i2c core cleanup. Here comes a patch which does just that. Since the process is exactly the same as what you did for the other drivers set already, I did not split this one in parts. A documentation update is included. The change saves 308 bytes in the i2c core, and an average 1382 bytes for chip drivers which use I2C_CLIENT_INSMOD, 126 bytes for those which do not. This change is required if we want to merge the sensors and non-sensors i2c code (and we want to do this). Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman Index: gregkh-2.6/Documentation/i2c/writing-clients =================================================================== --- Documentation/i2c/writing-clients | 62 +++++++-------------------------- drivers/acorn/char/pcf8583.c | 3 -- drivers/i2c/chips/isp1301_omap.c | 1 - drivers/i2c/chips/m41t00.c | 3 -- drivers/i2c/chips/rtc8564.c | 3 -- drivers/i2c/i2c-core.c | 35 ------------------- drivers/macintosh/therm_windtunnel.c | 6 ++-- drivers/media/video/adv7170.c | 6 ---- drivers/media/video/adv7175.c | 6 ---- drivers/media/video/bt819.c | 6 ---- drivers/media/video/bt832.c | 4 +-- drivers/media/video/bt856.c | 6 ---- drivers/media/video/msp3400.c | 1 - drivers/media/video/saa5246a.c | 1 - drivers/media/video/saa5249.c | 1 - drivers/media/video/saa7110.c | 6 ---- drivers/media/video/saa7111.c | 6 ---- drivers/media/video/saa7114.c | 6 ---- drivers/media/video/saa7134/saa6752hs.c | 1 - drivers/media/video/saa7185.c | 6 ---- drivers/media/video/tda7432.c | 1 - drivers/media/video/tda9840.c | 1 - drivers/media/video/tda9875.c | 1 - drivers/media/video/tda9887.c | 1 - drivers/media/video/tea6415c.c | 1 - drivers/media/video/tea6420.c | 1 - drivers/media/video/tuner-3036.c | 13 +++---- drivers/media/video/tuner-core.c | 11 +++--- drivers/media/video/tvaudio.c | 1 - drivers/media/video/tveeprom.c | 1 - drivers/media/video/vpx3220.c | 6 ---- drivers/video/matrox/matroxfb_maven.c | 1 - include/linux/i2c.h | 12 ------- 33 files changed, 27 insertions(+), 193 deletions(-) (limited to 'include/linux') diff --git a/Documentation/i2c/writing-clients b/Documentation/i2c/writing-clients index ad27511e3c7d..f482dae81de3 100644 --- a/Documentation/i2c/writing-clients +++ b/Documentation/i2c/writing-clients @@ -171,45 +171,31 @@ The following lists are used internally: normal_i2c: filled in by the module writer. A list of I2C addresses which should normally be examined. - normal_i2c_range: filled in by the module writer. - A list of pairs of I2C addresses, each pair being an inclusive range of - addresses which should normally be examined. probe: insmod parameter. A list of pairs. The first value is a bus number (-1 for any I2C bus), the second is the address. These addresses are also probed, as if they were in the 'normal' list. - probe_range: insmod parameter. - A list of triples. The first value is a bus number (-1 for any I2C bus), - the second and third are addresses. These form an inclusive range of - addresses that are also probed, as if they were in the 'normal' list. ignore: insmod parameter. A list of pairs. The first value is a bus number (-1 for any I2C bus), the second is the I2C address. These addresses are never probed. This parameter overrules 'normal' and 'probe', but not the 'force' lists. - ignore_range: insmod parameter. - A list of triples. The first value is a bus number (-1 for any I2C bus), - the second and third are addresses. These form an inclusive range of - I2C addresses that are never probed. - This parameter overrules 'normal' and 'probe', but not the 'force' lists. force: insmod parameter. A list of pairs. The first value is a bus number (-1 for any I2C bus), the second is the I2C address. A device is blindly assumed to be on the given address, no probing is done. -Fortunately, as a module writer, you just have to define the `normal' -and/or `normal_range' parameters. The complete declaration could look -like this: +Fortunately, as a module writer, you just have to define the `normal_i2c' +parameter. The complete declaration could look like this: - /* Scan 0x20 to 0x2f, 0x37, and 0x40 to 0x4f */ - static unsigned short normal_i2c[] = { 0x37,I2C_CLIENT_END }; - static unsigned short normal_i2c_range[] = { 0x20, 0x2f, 0x40, 0x4f, - I2C_CLIENT_END }; + /* Scan 0x37, and 0x48 to 0x4f */ + static unsigned short normal_i2c[] = { 0x37, 0x48, 0x49, 0x4a, 0x4b, 0x4c, + 0x4d, 0x4e, 0x4f, I2C_CLIENT_END }; /* Magic definition of all other variables and things */ I2C_CLIENT_INSMOD; -Note that you *have* to call the two defined variables `normal_i2c' and -`normal_i2c_range', without any prefix! +Note that you *have* to call the defined variable `normal_i2c', +without any prefix! Probing classes (sensors) @@ -223,39 +209,17 @@ The following lists are used internally. They are all lists of integers. normal_i2c: filled in by the module writer. Terminated by SENSORS_I2C_END. A list of I2C addresses which should normally be examined. - normal_i2c_range: filled in by the module writer. Terminated by - SENSORS_I2C_END - A list of pairs of I2C addresses, each pair being an inclusive range of - addresses which should normally be examined. normal_isa: filled in by the module writer. Terminated by SENSORS_ISA_END. A list of ISA addresses which should normally be examined. - normal_isa_range: filled in by the module writer. Terminated by - SENSORS_ISA_END - A list of triples. The first two elements are ISA addresses, being an - range of addresses which should normally be examined. The third is the - modulo parameter: only addresses which are 0 module this value relative - to the first address of the range are actually considered. probe: insmod parameter. Initialize this list with SENSORS_I2C_END values. A list of pairs. The first value is a bus number (SENSORS_ISA_BUS for the ISA bus, -1 for any I2C bus), the second is the address. These addresses are also probed, as if they were in the 'normal' list. - probe_range: insmod parameter. Initialize this list with SENSORS_I2C_END - values. - A list of triples. The first value is a bus number (SENSORS_ISA_BUS for - the ISA bus, -1 for any I2C bus), the second and third are addresses. - These form an inclusive range of addresses that are also probed, as - if they were in the 'normal' list. ignore: insmod parameter. Initialize this list with SENSORS_I2C_END values. A list of pairs. The first value is a bus number (SENSORS_ISA_BUS for the ISA bus, -1 for any I2C bus), the second is the I2C address. These addresses are never probed. This parameter overrules 'normal' and 'probe', but not the 'force' lists. - ignore_range: insmod parameter. Initialize this list with SENSORS_I2C_END - values. - A list of triples. The first value is a bus number (SENSORS_ISA_BUS for - the ISA bus, -1 for any I2C bus), the second and third are addresses. - These form an inclusive range of I2C addresses that are never probed. - This parameter overrules 'normal' and 'probe', but not the 'force' lists. Also used is a list of pointers to sensors_force_data structures: force_data: insmod parameters. A list, ending with an element of which @@ -269,16 +233,14 @@ Also used is a list of pointers to sensors_force_data structures: So we have a generic insmod variabled `force', and chip-specific variables `force_CHIPNAME'. -Fortunately, as a module writer, you just have to define the `normal' -and/or `normal_range' parameters, and define what chip names are used. +Fortunately, as a module writer, you just have to define the `normal_i2c' +and `normal_isa' parameters, and define what chip names are used. The complete declaration could look like this: - /* Scan i2c addresses 0x20 to 0x2f, 0x37, and 0x40 to 0x4f - static unsigned short normal_i2c[] = {0x37,SENSORS_I2C_END}; - static unsigned short normal_i2c_range[] = {0x20,0x2f,0x40,0x4f, - SENSORS_I2C_END}; + /* Scan i2c addresses 0x37, and 0x48 to 0x4f */ + static unsigned short normal_i2c[] = { 0x37, 0x48, 0x49, 0x4a, 0x4b, 0x4c, + 0x4d, 0x4e, 0x4f, I2C_CLIENT_END }; /* Scan ISA address 0x290 */ static unsigned int normal_isa[] = {0x0290,SENSORS_ISA_END}; - static unsigned int normal_isa_range[] = {SENSORS_ISA_END}; /* Define chips foo and bar, as well as all module parameters and things */ SENSORS_INSMOD_2(foo,bar); diff --git a/drivers/acorn/char/pcf8583.c b/drivers/acorn/char/pcf8583.c index ad7ae7ab8920..141b4c237a50 100644 --- a/drivers/acorn/char/pcf8583.c +++ b/drivers/acorn/char/pcf8583.c @@ -26,11 +26,8 @@ static unsigned short normal_addr[] = { 0x50, I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_addr, - .normal_i2c_range = ignore, .probe = ignore, - .probe_range = ignore, .ignore = ignore, - .ignore_range = ignore, .force = ignore, }; diff --git a/drivers/i2c/chips/isp1301_omap.c b/drivers/i2c/chips/isp1301_omap.c index 7f29a8aff165..354a26295672 100644 --- a/drivers/i2c/chips/isp1301_omap.c +++ b/drivers/i2c/chips/isp1301_omap.c @@ -145,7 +145,6 @@ static inline void notresponding(struct isp1301 *isp) static unsigned short normal_i2c[] = { ISP_BASE, ISP_BASE + 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; I2C_CLIENT_INSMOD; diff --git a/drivers/i2c/chips/m41t00.c b/drivers/i2c/chips/m41t00.c index e771566dffa8..5e463c47bfbc 100644 --- a/drivers/i2c/chips/m41t00.c +++ b/drivers/i2c/chips/m41t00.c @@ -40,11 +40,8 @@ static unsigned short normal_addr[] = { 0x68, I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_addr, - .normal_i2c_range = ignore, .probe = ignore, - .probe_range = ignore, .ignore = ignore, - .ignore_range = ignore, .force = ignore, }; diff --git a/drivers/i2c/chips/rtc8564.c b/drivers/i2c/chips/rtc8564.c index 5a9deddb626b..30f553e73700 100644 --- a/drivers/i2c/chips/rtc8564.c +++ b/drivers/i2c/chips/rtc8564.c @@ -66,11 +66,8 @@ static unsigned short normal_addr[] = { 0x51, I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_addr, - .normal_i2c_range = ignore, .probe = ignore, - .probe_range = ignore, .ignore = ignore, - .ignore_range = ignore, .force = ignore, }; diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c index a22e53badacb..4cc8c9f7211c 100644 --- a/drivers/i2c/i2c-core.c +++ b/drivers/i2c/i2c-core.c @@ -742,18 +742,6 @@ int i2c_probe(struct i2c_adapter *adapter, found = 1; } } - for (i = 0; - !found && (address_data->ignore_range[i] != I2C_CLIENT_END); - i += 3) { - if (((adap_id == address_data->ignore_range[i]) || - ((address_data->ignore_range[i]==ANY_I2C_BUS))) && - (addr >= address_data->ignore_range[i+1]) && - (addr <= address_data->ignore_range[i+2])) { - dev_dbg(&adapter->dev, "found ignore_range parameter for adapter %d, " - "addr %04x\n", adap_id,addr); - found = 1; - } - } if (found) continue; @@ -769,17 +757,6 @@ int i2c_probe(struct i2c_adapter *adapter, } } - for (i = 0; - !found && (address_data->normal_i2c_range[i] != I2C_CLIENT_END); - i += 2) { - if ((addr >= address_data->normal_i2c_range[i]) && - (addr <= address_data->normal_i2c_range[i+1])) { - found = 1; - dev_dbg(&adapter->dev, "found normal i2c_range entry for adapter %d, " - "addr %04x\n", adap_id,addr); - } - } - for (i = 0; !found && (address_data->probe[i] != I2C_CLIENT_END); i += 2) { @@ -791,18 +768,6 @@ int i2c_probe(struct i2c_adapter *adapter, "addr %04x\n", adap_id,addr); } } - for (i = 0; - !found && (address_data->probe_range[i] != I2C_CLIENT_END); - i += 3) { - if (((adap_id == address_data->probe_range[i]) || - (address_data->probe_range[i] == ANY_I2C_BUS)) && - (addr >= address_data->probe_range[i+1]) && - (addr <= address_data->probe_range[i+2])) { - found = 1; - dev_dbg(&adapter->dev, "found probe_range parameter for adapter %d, " - "addr %04x\n", adap_id,addr); - } - } if (!found) continue; diff --git a/drivers/macintosh/therm_windtunnel.c b/drivers/macintosh/therm_windtunnel.c index 0bdb47f08c2a..61400f04015e 100644 --- a/drivers/macintosh/therm_windtunnel.c +++ b/drivers/macintosh/therm_windtunnel.c @@ -51,8 +51,10 @@ static int do_probe( struct i2c_adapter *adapter, int addr, int kind); /* scan 0x48-0x4f (DS1775) and 0x2c-2x2f (ADM1030) */ -static unsigned short normal_i2c[] = { 0x49, 0x2c, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { 0x48, 0x4f, 0x2c, 0x2f, I2C_CLIENT_END }; +static unsigned short normal_i2c[] = { 0x48, 0x49, 0x4a, 0x4b, + 0x4c, 0x4d, 0x4e, 0x4f, + 0x2c, 0x2d, 0x2e, 0x2f, + I2C_CLIENT_END }; I2C_CLIENT_INSMOD; diff --git a/drivers/media/video/adv7170.c b/drivers/media/video/adv7170.c index 80254caa444c..e9bf3394296a 100644 --- a/drivers/media/video/adv7170.c +++ b/drivers/media/video/adv7170.c @@ -384,21 +384,15 @@ static unsigned short normal_i2c[] = I2C_ADV7171 >> 1, (I2C_ADV7171 >> 1) + 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, - .normal_i2c_range = normal_i2c_range, .probe = probe, - .probe_range = probe_range, .ignore = ignore, - .ignore_range = ignore_range, .force = force }; diff --git a/drivers/media/video/adv7175.c b/drivers/media/video/adv7175.c index 95d0974b0ab5..2d5fa44fcd4d 100644 --- a/drivers/media/video/adv7175.c +++ b/drivers/media/video/adv7175.c @@ -434,21 +434,15 @@ static unsigned short normal_i2c[] = I2C_ADV7176 >> 1, (I2C_ADV7176 >> 1) + 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, - .normal_i2c_range = normal_i2c_range, .probe = probe, - .probe_range = probe_range, .ignore = ignore, - .ignore_range = ignore_range, .force = force }; diff --git a/drivers/media/video/bt819.c b/drivers/media/video/bt819.c index cf0db2554a80..31d51851bb44 100644 --- a/drivers/media/video/bt819.c +++ b/drivers/media/video/bt819.c @@ -500,21 +500,15 @@ static unsigned short normal_i2c[] = { I2C_BT819 >> 1, I2C_CLIENT_END, }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, - .normal_i2c_range = normal_i2c_range, .probe = probe, - .probe_range = probe_range, .ignore = ignore, - .ignore_range = ignore_range, .force = force }; diff --git a/drivers/media/video/bt832.c b/drivers/media/video/bt832.c index efe605a113a1..07f72f64c5f7 100644 --- a/drivers/media/video/bt832.c +++ b/drivers/media/video/bt832.c @@ -39,8 +39,8 @@ MODULE_LICENSE("GPL"); /* Addresses to scan */ -static unsigned short normal_i2c[] = {I2C_CLIENT_END}; -static unsigned short normal_i2c_range[] = {I2C_BT832_ALT1>>1,I2C_BT832_ALT2>>1,I2C_CLIENT_END}; +static unsigned short normal_i2c[] = { I2C_BT832_ALT1>>1, I2C_BT832_ALT2>>1, + I2C_CLIENT_END }; I2C_CLIENT_INSMOD; /* ---------------------------------------------------------------------- */ diff --git a/drivers/media/video/bt856.c b/drivers/media/video/bt856.c index 72c7eb0f8c24..59121a0ec816 100644 --- a/drivers/media/video/bt856.c +++ b/drivers/media/video/bt856.c @@ -288,21 +288,15 @@ bt856_command (struct i2c_client *client, * concerning the addresses: i2c wants 7 bit (without the r/w bit), so '>>1' */ static unsigned short normal_i2c[] = { I2C_BT856 >> 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, - .normal_i2c_range = normal_i2c_range, .probe = probe, - .probe_range = probe_range, .ignore = ignore, - .ignore_range = ignore_range, .force = force }; diff --git a/drivers/media/video/msp3400.c b/drivers/media/video/msp3400.c index 7fbb8581a87d..09464d624a6b 100644 --- a/drivers/media/video/msp3400.c +++ b/drivers/media/video/msp3400.c @@ -147,7 +147,6 @@ static unsigned short normal_i2c[] = { I2C_MSP3400C_ALT >> 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = {I2C_CLIENT_END,I2C_CLIENT_END}; I2C_CLIENT_INSMOD; /* ----------------------------------------------------------------------- */ diff --git a/drivers/media/video/saa5246a.c b/drivers/media/video/saa5246a.c index ba69f09cbdd1..b8054da31ffd 100644 --- a/drivers/media/video/saa5246a.c +++ b/drivers/media/video/saa5246a.c @@ -64,7 +64,6 @@ static struct video_device saa_template; /* Declared near bottom */ /* Addresses to scan */ static unsigned short normal_i2c[] = { I2C_ADDRESS, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; I2C_CLIENT_INSMOD; static struct i2c_client client_template; diff --git a/drivers/media/video/saa5249.c b/drivers/media/video/saa5249.c index d74caa139f0a..7ffa2e9a9bf3 100644 --- a/drivers/media/video/saa5249.c +++ b/drivers/media/video/saa5249.c @@ -132,7 +132,6 @@ static struct video_device saa_template; /* Declared near bottom */ /* Addresses to scan */ static unsigned short normal_i2c[] = {34>>1,I2C_CLIENT_END}; -static unsigned short normal_i2c_range[] = {I2C_CLIENT_END}; I2C_CLIENT_INSMOD; static struct i2c_client client_template; diff --git a/drivers/media/video/saa7110.c b/drivers/media/video/saa7110.c index 64273b438530..90b0a0b34f38 100644 --- a/drivers/media/video/saa7110.c +++ b/drivers/media/video/saa7110.c @@ -463,21 +463,15 @@ static unsigned short normal_i2c[] = { (I2C_SAA7110 >> 1) + 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, - .normal_i2c_range = normal_i2c_range, .probe = probe, - .probe_range = probe_range, .ignore = ignore, - .ignore_range = ignore_range, .force = force }; diff --git a/drivers/media/video/saa7111.c b/drivers/media/video/saa7111.c index 0a873112ae23..e305a89f7cd7 100644 --- a/drivers/media/video/saa7111.c +++ b/drivers/media/video/saa7111.c @@ -482,21 +482,15 @@ saa7111_command (struct i2c_client *client, * concerning the addresses: i2c wants 7 bit (without the r/w bit), so '>>1' */ static unsigned short normal_i2c[] = { I2C_SAA7111 >> 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, - .normal_i2c_range = normal_i2c_range, .probe = probe, - .probe_range = probe_range, .ignore = ignore, - .ignore_range = ignore_range, .force = force }; diff --git a/drivers/media/video/saa7114.c b/drivers/media/video/saa7114.c index e73023695e58..1ca4e70fed76 100644 --- a/drivers/media/video/saa7114.c +++ b/drivers/media/video/saa7114.c @@ -820,21 +820,15 @@ saa7114_command (struct i2c_client *client, */ static unsigned short normal_i2c[] = { I2C_SAA7114 >> 1, I2C_SAA7114A >> 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, - .normal_i2c_range = normal_i2c_range, .probe = probe, - .probe_range = probe_range, .ignore = ignore, - .ignore_range = ignore_range, .force = force }; diff --git a/drivers/media/video/saa7134/saa6752hs.c b/drivers/media/video/saa7134/saa6752hs.c index 1db022682980..42c2b565c9fe 100644 --- a/drivers/media/video/saa7134/saa6752hs.c +++ b/drivers/media/video/saa7134/saa6752hs.c @@ -22,7 +22,6 @@ /* Addresses to scan */ static unsigned short normal_i2c[] = {0x20, I2C_CLIENT_END}; -static unsigned short normal_i2c_range[] = {I2C_CLIENT_END}; I2C_CLIENT_INSMOD; MODULE_DESCRIPTION("device driver for saa6752hs MPEG2 encoder"); diff --git a/drivers/media/video/saa7185.c b/drivers/media/video/saa7185.c index 5f0b224c3cb6..5c623fadc8fe 100644 --- a/drivers/media/video/saa7185.c +++ b/drivers/media/video/saa7185.c @@ -380,21 +380,15 @@ saa7185_command (struct i2c_client *client, * concerning the addresses: i2c wants 7 bit (without the r/w bit), so '>>1' */ static unsigned short normal_i2c[] = { I2C_SAA7185 >> 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, - .normal_i2c_range = normal_i2c_range, .probe = probe, - .probe_range = probe_range, .ignore = ignore, - .ignore_range = ignore_range, .force = force }; diff --git a/drivers/media/video/tda7432.c b/drivers/media/video/tda7432.c index 376a4a439e9b..07ba6d3ed08c 100644 --- a/drivers/media/video/tda7432.c +++ b/drivers/media/video/tda7432.c @@ -74,7 +74,6 @@ static unsigned short normal_i2c[] = { I2C_TDA7432 >> 1, I2C_CLIENT_END, }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END, I2C_CLIENT_END }; I2C_CLIENT_INSMOD; /* Structure of address and subaddresses for the tda7432 */ diff --git a/drivers/media/video/tda9840.c b/drivers/media/video/tda9840.c index b5177c6f54f6..c29bdfc3244e 100644 --- a/drivers/media/video/tda9840.c +++ b/drivers/media/video/tda9840.c @@ -43,7 +43,6 @@ MODULE_PARM_DESC(debug, "Turn on/off device debugging (default:off)."); /* addresses to scan, found only at 0x42 (7-Bit) */ static unsigned short normal_i2c[] = { I2C_TDA9840, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; /* magic definition of all other variables and things */ I2C_CLIENT_INSMOD; diff --git a/drivers/media/video/tda9875.c b/drivers/media/video/tda9875.c index 4f1114c033a1..97b113e070f3 100644 --- a/drivers/media/video/tda9875.c +++ b/drivers/media/video/tda9875.c @@ -44,7 +44,6 @@ static unsigned short normal_i2c[] = { I2C_TDA9875 >> 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = {I2C_CLIENT_END}; I2C_CLIENT_INSMOD; /* This is a superset of the TDA9875 */ diff --git a/drivers/media/video/tda9887.c b/drivers/media/video/tda9887.c index debef1910c37..7e6e6dd966a2 100644 --- a/drivers/media/video/tda9887.c +++ b/drivers/media/video/tda9887.c @@ -33,7 +33,6 @@ static unsigned short normal_i2c[] = { 0x96 >>1, I2C_CLIENT_END, }; -static unsigned short normal_i2c_range[] = {I2C_CLIENT_END,I2C_CLIENT_END}; I2C_CLIENT_INSMOD; /* insmod options */ diff --git a/drivers/media/video/tea6415c.c b/drivers/media/video/tea6415c.c index 3ec39550bf46..b44db8a7b94d 100644 --- a/drivers/media/video/tea6415c.c +++ b/drivers/media/video/tea6415c.c @@ -43,7 +43,6 @@ MODULE_PARM_DESC(debug, "Turn on/off device debugging (default:off)."); /* addresses to scan, found only at 0x03 and/or 0x43 (7-bit) */ static unsigned short normal_i2c[] = { I2C_TEA6415C_1, I2C_TEA6415C_2, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; /* magic definition of all other variables and things */ I2C_CLIENT_INSMOD; diff --git a/drivers/media/video/tea6420.c b/drivers/media/video/tea6420.c index bd10710fd909..48d4db7d507b 100644 --- a/drivers/media/video/tea6420.c +++ b/drivers/media/video/tea6420.c @@ -40,7 +40,6 @@ MODULE_PARM_DESC(debug, "Turn on/off device debugging (default:off)."); /* addresses to scan, found only at 0x4c and/or 0x4d (7-Bit) */ static unsigned short normal_i2c[] = { I2C_TEA6420_1, I2C_TEA6420_2, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; /* magic definition of all other variables and things */ I2C_CLIENT_INSMOD; diff --git a/drivers/media/video/tuner-3036.c b/drivers/media/video/tuner-3036.c index 6b20aa902a8f..bedb15e2f233 100644 --- a/drivers/media/video/tuner-3036.c +++ b/drivers/media/video/tuner-3036.c @@ -34,19 +34,16 @@ static int this_adap; static struct i2c_client client_template; /* Addresses to scan */ -static unsigned short normal_i2c[] = {I2C_CLIENT_END}; -static unsigned short normal_i2c_range[] = {0x60, 0x61, I2C_CLIENT_END}; +static unsigned short normal_i2c[] = { 0x60, 0x61, I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { - normal_i2c, normal_i2c_range, - probe, probe_range, - ignore, ignore_range, - force + .normal_i2c = normal_i2c, + .probe = probe, + .ignore = ignore, + .force = force, }; /* ---------------------------------------------------------------------- */ diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c index 6212388edb75..81882ddab859 100644 --- a/drivers/media/video/tuner-core.c +++ b/drivers/media/video/tuner-core.c @@ -28,10 +28,8 @@ /* standard i2c insmod options */ static unsigned short normal_i2c[] = { 0x4b, /* tda8290 */ - I2C_CLIENT_END -}; -static unsigned short normal_i2c_range[] = { - 0x60, 0x6f, + 0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, + 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, I2C_CLIENT_END }; I2C_CLIENT_INSMOD; @@ -225,9 +223,8 @@ static int tuner_attach(struct i2c_adapter *adap, int addr, int kind) static int tuner_probe(struct i2c_adapter *adap) { if (0 != addr) { - normal_i2c[0] = addr; - normal_i2c_range[0] = addr; - normal_i2c_range[1] = addr; + normal_i2c[0] = addr; + normal_i2c[1] = I2C_CLIENT_END; } this_adap = 0; diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c index 80dc34f18c2c..41b635e0d3c6 100644 --- a/drivers/media/video/tvaudio.c +++ b/drivers/media/video/tvaudio.c @@ -148,7 +148,6 @@ static unsigned short normal_i2c[] = { I2C_TDA9874 >> 1, I2C_PIC16C54 >> 1, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; I2C_CLIENT_INSMOD; static struct i2c_driver driver; diff --git a/drivers/media/video/tveeprom.c b/drivers/media/video/tveeprom.c index e1443a0937e3..3d216973798c 100644 --- a/drivers/media/video/tveeprom.c +++ b/drivers/media/video/tveeprom.c @@ -482,7 +482,6 @@ static unsigned short normal_i2c[] = { 0xa0 >> 1, I2C_CLIENT_END, }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; I2C_CLIENT_INSMOD; struct i2c_driver i2c_driver_tveeprom; diff --git a/drivers/media/video/vpx3220.c b/drivers/media/video/vpx3220.c index 0fd6c9a70917..b97036910fa9 100644 --- a/drivers/media/video/vpx3220.c +++ b/drivers/media/video/vpx3220.c @@ -569,21 +569,15 @@ static unsigned short normal_i2c[] = { I2C_VPX3220 >> 1, (I2C_VPX3220 >> 1) + 4, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { I2C_CLIENT_END }; static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; -static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END }; static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, - .normal_i2c_range = normal_i2c_range, .probe = probe, - .probe_range = probe_range, .ignore = ignore, - .ignore_range = ignore_range, .force = force }; diff --git a/drivers/video/matrox/matroxfb_maven.c b/drivers/video/matrox/matroxfb_maven.c index e529841cd83d..67f85344f0cc 100644 --- a/drivers/video/matrox/matroxfb_maven.c +++ b/drivers/video/matrox/matroxfb_maven.c @@ -1230,7 +1230,6 @@ static int maven_shutdown_client(struct i2c_client* clnt) { } static unsigned short normal_i2c[] = { MAVEN_I2CID, I2C_CLIENT_END }; -static unsigned short normal_i2c_range[] = { MAVEN_I2CID, MAVEN_I2CID, I2C_CLIENT_END }; I2C_CLIENT_INSMOD; static struct i2c_driver maven_driver; diff --git a/include/linux/i2c.h b/include/linux/i2c.h index ebcd745f4cd6..be837b13f297 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -290,11 +290,8 @@ static inline void i2c_set_adapdata (struct i2c_adapter *dev, void *data) */ struct i2c_client_address_data { unsigned short *normal_i2c; - unsigned short *normal_i2c_range; unsigned short *probe; - unsigned short *probe_range; unsigned short *ignore; - unsigned short *ignore_range; unsigned short *force; }; @@ -563,24 +560,15 @@ union i2c_smbus_data { #define I2C_CLIENT_INSMOD \ I2C_CLIENT_MODULE_PARM(probe, \ "List of adapter,address pairs to scan additionally"); \ - I2C_CLIENT_MODULE_PARM(probe_range, \ - "List of adapter,start-addr,end-addr triples to scan " \ - "additionally"); \ I2C_CLIENT_MODULE_PARM(ignore, \ "List of adapter,address pairs not to scan"); \ - I2C_CLIENT_MODULE_PARM(ignore_range, \ - "List of adapter,start-addr,end-addr triples not to " \ - "scan"); \ I2C_CLIENT_MODULE_PARM(force, \ "List of adapter,address pairs to boldly assume " \ "to be present"); \ static struct i2c_client_address_data addr_data = { \ .normal_i2c = normal_i2c, \ - .normal_i2c_range = normal_i2c_range, \ .probe = probe, \ - .probe_range = probe_range, \ .ignore = ignore, \ - .ignore_range = ignore_range, \ .force = force, \ } -- cgit v1.2.3 From 3886246a257e828248ce1e72ced00408a3557f0d Mon Sep 17 00:00:00 2001 From: Sebastian Witt Date: Wed, 13 Apr 2005 22:25:39 +0200 Subject: [PATCH] I2C: i2c-vid.h: Support for VID to reg conversion Adds conversion from VID (mV) to register value. Used by the atxp1 I2C module. Removed uneeded switch case. Signed-off-by: Sebastian Witt Signed-off-by: Greg Kroah-Hartman --- include/linux/i2c-vid.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'include/linux') diff --git a/include/linux/i2c-vid.h b/include/linux/i2c-vid.h index 974835e3530f..41d0635e0ba9 100644 --- a/include/linux/i2c-vid.h +++ b/include/linux/i2c-vid.h @@ -97,3 +97,15 @@ static inline int vid_from_reg(int val, int vrm) 2050 - (val) * 50); } } + +static inline int vid_to_reg(int val, int vrm) +{ + switch (vrm) { + case 91: /* VRM 9.1 */ + case 90: /* VRM 9.0 */ + return ((val >= 1100) && (val <= 1850) ? + ((18499 - val * 10) / 25 + 5) / 10 : -1); + default: + return -1; + } +} -- cgit v1.2.3 From 10c08f8100ee2c4d27b862635574cdf4ef439e67 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Mon, 6 Jun 2005 19:34:45 +0200 Subject: [PATCH] I2C: rename i2c-sysfs.h to hwmon-sysfs.h This patch renames the new linux/i2c-sysfs.h header file to linux/hwmon-sysfs.h. This names seems to be more appropriate since this file defines macros and structures not related to i2c but to hardware monitoring drivers. The patch also updates the five hardware monitoring driver which include that header file already. Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman --- drivers/i2c/chips/adm1026.c | 2 +- drivers/i2c/chips/it87.c | 2 +- drivers/i2c/chips/lm63.c | 2 +- drivers/i2c/chips/lm83.c | 2 +- drivers/i2c/chips/lm90.c | 2 +- include/linux/hwmon-sysfs.h | 36 ++++++++++++++++++++++++++++++++++++ include/linux/i2c-sysfs.h | 36 ------------------------------------ 7 files changed, 41 insertions(+), 41 deletions(-) create mode 100644 include/linux/hwmon-sysfs.h delete mode 100644 include/linux/i2c-sysfs.h (limited to 'include/linux') diff --git a/drivers/i2c/chips/adm1026.c b/drivers/i2c/chips/adm1026.c index ddbc01505ed3..3c85fe150cd7 100644 --- a/drivers/i2c/chips/adm1026.c +++ b/drivers/i2c/chips/adm1026.c @@ -29,8 +29,8 @@ #include #include #include -#include #include +#include /* Addresses to scan */ static unsigned short normal_i2c[] = { 0x2c, 0x2d, 0x2e, I2C_CLIENT_END }; diff --git a/drivers/i2c/chips/it87.c b/drivers/i2c/chips/it87.c index 6a9b65a10bbc..db20c9e47393 100644 --- a/drivers/i2c/chips/it87.c +++ b/drivers/i2c/chips/it87.c @@ -37,8 +37,8 @@ #include #include #include -#include #include +#include #include diff --git a/drivers/i2c/chips/lm63.c b/drivers/i2c/chips/lm63.c index a1fd12bd615f..7c6f9ea5a254 100644 --- a/drivers/i2c/chips/lm63.c +++ b/drivers/i2c/chips/lm63.c @@ -43,7 +43,7 @@ #include #include #include -#include +#include /* * Addresses to scan diff --git a/drivers/i2c/chips/lm83.c b/drivers/i2c/chips/lm83.c index 0e0eae4dceaa..a49008b444c8 100644 --- a/drivers/i2c/chips/lm83.c +++ b/drivers/i2c/chips/lm83.c @@ -33,7 +33,7 @@ #include #include #include -#include +#include /* * Addresses to scan diff --git a/drivers/i2c/chips/lm90.c b/drivers/i2c/chips/lm90.c index ebd99dfbf9c7..a67dcadf7cb0 100644 --- a/drivers/i2c/chips/lm90.c +++ b/drivers/i2c/chips/lm90.c @@ -76,7 +76,7 @@ #include #include #include -#include +#include /* * Addresses to scan diff --git a/include/linux/hwmon-sysfs.h b/include/linux/hwmon-sysfs.h new file mode 100644 index 000000000000..1b5018a965f5 --- /dev/null +++ b/include/linux/hwmon-sysfs.h @@ -0,0 +1,36 @@ +/* + * hwmon-sysfs.h - hardware monitoring chip driver sysfs defines + * + * Copyright (C) 2005 Yani Ioannou + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ +#ifndef _LINUX_HWMON_SYSFS_H +#define _LINUX_HWMON_SYSFS_H + +struct sensor_device_attribute{ + struct device_attribute dev_attr; + int index; +}; +#define to_sensor_dev_attr(_dev_attr) \ + container_of(_dev_attr, struct sensor_device_attribute, dev_attr) + +#define SENSOR_DEVICE_ATTR(_name,_mode,_show,_store,_index) \ +struct sensor_device_attribute sensor_dev_attr_##_name = { \ + .dev_attr = __ATTR(_name,_mode,_show,_store), \ + .index = _index, \ +} + +#endif /* _LINUX_HWMON_SYSFS_H */ diff --git a/include/linux/i2c-sysfs.h b/include/linux/i2c-sysfs.h deleted file mode 100644 index d7bf6ce11679..000000000000 --- a/include/linux/i2c-sysfs.h +++ /dev/null @@ -1,36 +0,0 @@ -/* - * i2c-sysfs.h - i2c chip driver sysfs defines - * - * Copyright (C) 2005 Yani Ioannou - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - */ -#ifndef _LINUX_I2C_SYSFS_H -#define _LINUX_I2C_SYSFS_H - -struct sensor_device_attribute{ - struct device_attribute dev_attr; - int index; -}; -#define to_sensor_dev_attr(_dev_attr) \ - container_of(_dev_attr, struct sensor_device_attribute, dev_attr) - -#define SENSOR_DEVICE_ATTR(_name,_mode,_show,_store,_index) \ -struct sensor_device_attribute sensor_dev_attr_##_name = { \ - .dev_attr = __ATTR(_name,_mode,_show,_store), \ - .index = _index, \ -} - -#endif /* _LINUX_I2C_SYSFS_H */ -- cgit v1.2.3 From c124a78d8c7475ecc43f385f34112b638c4228d9 Mon Sep 17 00:00:00 2001 From: Randy Vinson Date: Fri, 3 Jun 2005 14:36:06 -0700 Subject: [PATCH] I2C: Add support for Maxim/Dallas DS1374 Real-Time Clock Chip (1/2) Add support for Maxim/Dallas DS1374 Real-Time Clock Chip This change adds support for the Maxim/Dallas DS1374 RTC chip. This chip is an I2C-based RTC that maintains a simple 32-bit binary seconds count with battery backup support. Signed-off-by: Randy Vinson Signed-off-by: Greg Kroah-Hartman --- drivers/i2c/chips/Kconfig | 11 ++ drivers/i2c/chips/Makefile | 1 + drivers/i2c/chips/ds1374.c | 266 +++++++++++++++++++++++++++++++++++++++++++++ include/linux/i2c-id.h | 1 + 4 files changed, 279 insertions(+) create mode 100644 drivers/i2c/chips/ds1374.c (limited to 'include/linux') diff --git a/drivers/i2c/chips/Kconfig b/drivers/i2c/chips/Kconfig index 33de80afd6c6..a0982da09803 100644 --- a/drivers/i2c/chips/Kconfig +++ b/drivers/i2c/chips/Kconfig @@ -417,6 +417,17 @@ config SENSORS_DS1337 This driver can also be built as a module. If so, the module will be called ds1337. +config SENSORS_DS1374 + tristate "Maxim/Dallas Semiconductor DS1374 Real Time Clock" + depends on I2C && EXPERIMENTAL + select I2C_SENSOR + help + If you say yes here you get support for Dallas Semiconductor + DS1374 real-time clock chips. + + This driver can also be built as a module. If so, the module + will be called ds1374. + config SENSORS_EEPROM tristate "EEPROM reader" depends on I2C && EXPERIMENTAL diff --git a/drivers/i2c/chips/Makefile b/drivers/i2c/chips/Makefile index 6bebdc104166..b5e6d2f84f97 100644 --- a/drivers/i2c/chips/Makefile +++ b/drivers/i2c/chips/Makefile @@ -14,6 +14,7 @@ obj-$(CONFIG_SENSORS_ADM1031) += adm1031.o obj-$(CONFIG_SENSORS_ADM9240) += adm9240.o obj-$(CONFIG_SENSORS_ATXP1) += atxp1.o obj-$(CONFIG_SENSORS_DS1337) += ds1337.o +obj-$(CONFIG_SENSORS_DS1374) += ds1374.o obj-$(CONFIG_SENSORS_DS1621) += ds1621.o obj-$(CONFIG_SENSORS_EEPROM) += eeprom.o obj-$(CONFIG_SENSORS_FSCHER) += fscher.o diff --git a/drivers/i2c/chips/ds1374.c b/drivers/i2c/chips/ds1374.c new file mode 100644 index 000000000000..1278d979db2b --- /dev/null +++ b/drivers/i2c/chips/ds1374.c @@ -0,0 +1,266 @@ +/* + * drivers/i2c/chips/ds1374.c + * + * I2C client/driver for the Maxim/Dallas DS1374 Real-Time Clock + * + * Author: Randy Vinson + * + * Based on the m41t00.c by Mark Greer + * + * 2005 (c) MontaVista Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + */ +/* + * This i2c client/driver wedges between the drivers/char/genrtc.c RTC + * interface and the SMBus interface of the i2c subsystem. + * It would be more efficient to use i2c msgs/i2c_transfer directly but, as + * recommened in .../Documentation/i2c/writing-clients section + * "Sending and receiving", using SMBus level communication is preferred. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include + +#define DS1374_REG_TOD0 0x00 +#define DS1374_REG_TOD1 0x01 +#define DS1374_REG_TOD2 0x02 +#define DS1374_REG_TOD3 0x03 +#define DS1374_REG_WDALM0 0x04 +#define DS1374_REG_WDALM1 0x05 +#define DS1374_REG_WDALM2 0x06 +#define DS1374_REG_CR 0x07 +#define DS1374_REG_SR 0x08 +#define DS1374_REG_SR_OSF 0x80 +#define DS1374_REG_TCR 0x09 + +#define DS1374_DRV_NAME "ds1374" + +static DECLARE_MUTEX(ds1374_mutex); + +static struct i2c_driver ds1374_driver; +static struct i2c_client *save_client; + +static unsigned short ignore[] = { I2C_CLIENT_END }; +static unsigned short normal_addr[] = { 0x68, I2C_CLIENT_END }; + +static struct i2c_client_address_data addr_data = { + .normal_i2c = normal_addr, + .normal_i2c_range = ignore, + .probe = ignore, + .probe_range = ignore, + .ignore = ignore, + .ignore_range = ignore, + .force = ignore, +}; + +static ulong ds1374_read_rtc(void) +{ + ulong time = 0; + int reg = DS1374_REG_WDALM0; + + while (reg--) { + s32 tmp; + if ((tmp = i2c_smbus_read_byte_data(save_client, reg)) < 0) { + dev_warn(&save_client->dev, + "can't read from rtc chip\n"); + return 0; + } + time = (time << 8) | (tmp & 0xff); + } + return time; +} + +static void ds1374_write_rtc(ulong time) +{ + int reg; + + for (reg = DS1374_REG_TOD0; reg < DS1374_REG_WDALM0; reg++) { + if (i2c_smbus_write_byte_data(save_client, reg, time & 0xff) + < 0) { + dev_warn(&save_client->dev, + "can't write to rtc chip\n"); + break; + } + time = time >> 8; + } +} + +static void ds1374_check_rtc_status(void) +{ + s32 tmp; + + tmp = i2c_smbus_read_byte_data(save_client, DS1374_REG_SR); + if (tmp < 0) { + dev_warn(&save_client->dev, + "can't read status from rtc chip\n"); + return; + } + if (tmp & DS1374_REG_SR_OSF) { + dev_warn(&save_client->dev, + "oscillator discontinuity flagged, time unreliable\n"); + tmp &= ~DS1374_REG_SR_OSF; + tmp = i2c_smbus_write_byte_data(save_client, DS1374_REG_SR, + tmp & 0xff); + if (tmp < 0) + dev_warn(&save_client->dev, + "can't clear discontinuity notification\n"); + } +} + +ulong ds1374_get_rtc_time(void) +{ + ulong t1, t2; + int limit = 10; /* arbitrary retry limit */ + + down(&ds1374_mutex); + + /* + * Since the reads are being performed one byte at a time using + * the SMBus vs a 4-byte i2c transfer, there is a chance that a + * carry will occur during the read. To detect this, 2 reads are + * performed and compared. + */ + do { + t1 = ds1374_read_rtc(); + t2 = ds1374_read_rtc(); + } while (t1 != t2 && limit--); + + up(&ds1374_mutex); + + if (t1 != t2) { + dev_warn(&save_client->dev, + "can't get consistent time from rtc chip\n"); + t1 = 0; + } + + return t1; +} + +static void ds1374_set_tlet(ulong arg) +{ + ulong t1, t2; + int limit = 10; /* arbitrary retry limit */ + + t1 = *(ulong *) arg; + + down(&ds1374_mutex); + + /* + * Since the writes are being performed one byte at a time using + * the SMBus vs a 4-byte i2c transfer, there is a chance that a + * carry will occur during the write. To detect this, the write + * value is read back and compared. + */ + do { + ds1374_write_rtc(t1); + t2 = ds1374_read_rtc(); + } while (t1 != t2 && limit--); + + up(&ds1374_mutex); + + if (t1 != t2) + dev_warn(&save_client->dev, + "can't confirm time set from rtc chip\n"); +} + +ulong new_time; + +DECLARE_TASKLET_DISABLED(ds1374_tasklet, ds1374_set_tlet, (ulong) & new_time); + +int ds1374_set_rtc_time(ulong nowtime) +{ + new_time = nowtime; + + if (in_interrupt()) + tasklet_schedule(&ds1374_tasklet); + else + ds1374_set_tlet((ulong) & new_time); + + return 0; +} + +/* + ***************************************************************************** + * + * Driver Interface + * + ***************************************************************************** + */ +static int ds1374_probe(struct i2c_adapter *adap, int addr, int kind) +{ + struct i2c_client *client; + int rc; + + client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); + if (!client) + return -ENOMEM; + + memset(client, 0, sizeof(struct i2c_client)); + strncpy(client->name, DS1374_DRV_NAME, I2C_NAME_SIZE); + client->flags = I2C_DF_NOTIFY; + client->addr = addr; + client->adapter = adap; + client->driver = &ds1374_driver; + + if ((rc = i2c_attach_client(client)) != 0) { + kfree(client); + return rc; + } + + save_client = client; + + ds1374_check_rtc_status(); + + return 0; +} + +static int ds1374_attach(struct i2c_adapter *adap) +{ + return i2c_probe(adap, &addr_data, ds1374_probe); +} + +static int ds1374_detach(struct i2c_client *client) +{ + int rc; + + if ((rc = i2c_detach_client(client)) == 0) { + kfree(i2c_get_clientdata(client)); + tasklet_kill(&ds1374_tasklet); + } + return rc; +} + +static struct i2c_driver ds1374_driver = { + .owner = THIS_MODULE, + .name = DS1374_DRV_NAME, + .id = I2C_DRIVERID_DS1374, + .flags = I2C_DF_NOTIFY, + .attach_adapter = ds1374_attach, + .detach_client = ds1374_detach, +}; + +static int __init ds1374_init(void) +{ + return i2c_add_driver(&ds1374_driver); +} + +static void __exit ds1374_exit(void) +{ + i2c_del_driver(&ds1374_driver); +} + +module_init(ds1374_init); +module_exit(ds1374_exit); + +MODULE_AUTHOR("Randy Vinson "); +MODULE_DESCRIPTION("Maxim/Dallas DS1374 RTC I2C Client Driver"); +MODULE_LICENSE("GPL"); diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h index 89270ce51470..33f08258f22b 100644 --- a/include/linux/i2c-id.h +++ b/include/linux/i2c-id.h @@ -108,6 +108,7 @@ #define I2C_DRIVERID_TDA7313 62 /* TDA7313 audio processor */ #define I2C_DRIVERID_MAX6900 63 /* MAX6900 real-time clock */ #define I2C_DRIVERID_SAA7114H 64 /* video decoder */ +#define I2C_DRIVERID_DS1374 65 /* DS1374 real time clock */ #define I2C_DRIVERID_EXP0 0xF0 /* experimental use id's */ -- cgit v1.2.3 From dd7f0b80926befc8c70a873b5b0c0c7b5fd1e7b9 Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Wed, 22 Jun 2005 12:38:33 -0700 Subject: [NETFILTER]: Fix "iptables -D" rule deletion with ipt_CLUSTERIP target. The patch just changes the order of structure members. Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ipt_CLUSTERIP.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h b/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h index baa83e757156..d9bceedfb3dc 100644 --- a/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h +++ b/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h @@ -18,7 +18,6 @@ struct clusterip_config; struct ipt_clusterip_tgt_info { u_int32_t flags; - struct clusterip_config *config; /* only relevant for new ones */ u_int8_t clustermac[6]; @@ -27,6 +26,8 @@ struct ipt_clusterip_tgt_info { u_int16_t local_nodes[CLUSTERIP_MAX_NODES]; enum clusterip_hashmode hash_mode; u_int32_t hash_initval; + + struct clusterip_config *config; }; #endif /*_IPT_CLUSTERIP_H_target*/ -- cgit v1.2.3 From 5ee0ed7d3ab620a764740fb018f469d45f561931 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:20 +0000 Subject: [PATCH] RPC: Make rpc_create_client() probe server for RPC program+version support Ensure that we don't create an RPC client without checking that the server does indeed support the RPC program + version that we are trying to set up. This enables us to immediately return an error to "mount" if it turns out that the server is only supporting NFSv2, when we requested NFSv3 or NFSv4. Signed-off-by: Trond Myklebust --- fs/lockd/host.c | 4 +-- fs/lockd/mon.c | 2 +- include/linux/sunrpc/clnt.h | 4 +++ net/sunrpc/clnt.c | 59 ++++++++++++++++++++++++++++++++++++++++++++- net/sunrpc/pmap_clnt.c | 2 +- 5 files changed, 66 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/fs/lockd/host.c b/fs/lockd/host.c index 90a62f27914c..82c77df81c5f 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -189,6 +189,8 @@ nlm_bind_host(struct nlm_host *host) goto forgetit; xprt_set_timeout(&xprt->timeout, 5, nlmsvc_timeout); + xprt->nocong = 1; /* No congestion control for NLM */ + xprt->resvport = 1; /* NLM requires a reserved port */ /* Existing NLM servers accept AUTH_UNIX only */ clnt = rpc_create_client(xprt, host->h_name, &nlm_program, @@ -196,8 +198,6 @@ nlm_bind_host(struct nlm_host *host) if (IS_ERR(clnt)) goto forgetit; clnt->cl_autobind = 1; /* turn on pmap queries */ - xprt->nocong = 1; /* No congestion control for NLM */ - xprt->resvport = 1; /* NLM requires a reserved port */ host->h_rpcclnt = clnt; } diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index 81b5e7778d70..2d144abe84ad 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -115,6 +115,7 @@ nsm_create(void) xprt = xprt_create_proto(IPPROTO_UDP, &sin, NULL); if (IS_ERR(xprt)) return (struct rpc_clnt *)xprt; + xprt->resvport = 1; /* NSM requires a reserved port */ clnt = rpc_create_client(xprt, "localhost", &nsm_program, SM_VERSION, @@ -124,7 +125,6 @@ nsm_create(void) clnt->cl_softrtry = 1; clnt->cl_chatty = 1; clnt->cl_oneshot = 1; - xprt->resvport = 1; /* NSM requires a reserved port */ return clnt; out_err: diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index 2709caf4d128..d25e80f77ff5 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -111,6 +111,9 @@ struct rpc_procinfo { struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname, struct rpc_program *info, u32 version, rpc_authflavor_t authflavor); +struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname, + struct rpc_program *info, + u32 version, rpc_authflavor_t authflavor); struct rpc_clnt *rpc_clone_client(struct rpc_clnt *); int rpc_shutdown_client(struct rpc_clnt *); int rpc_destroy_client(struct rpc_clnt *); @@ -129,6 +132,7 @@ void rpc_clnt_sigmask(struct rpc_clnt *clnt, sigset_t *oldset); void rpc_clnt_sigunmask(struct rpc_clnt *clnt, sigset_t *oldset); void rpc_setbufsize(struct rpc_clnt *, unsigned int, unsigned int); size_t rpc_max_payload(struct rpc_clnt *); +int rpc_ping(struct rpc_clnt *clnt, int flags); static __inline__ int rpc_call(struct rpc_clnt *clnt, u32 proc, void *argp, void *resp, int flags) diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 99515d7727a6..b36797ad8083 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -97,7 +97,7 @@ rpc_setup_pipedir(struct rpc_clnt *clnt, char *dir_name) * made to sleep too long. */ struct rpc_clnt * -rpc_create_client(struct rpc_xprt *xprt, char *servname, +rpc_new_client(struct rpc_xprt *xprt, char *servname, struct rpc_program *program, u32 vers, rpc_authflavor_t flavor) { @@ -182,6 +182,36 @@ out_err: return ERR_PTR(err); } +/** + * Create an RPC client + * @xprt - pointer to xprt struct + * @servname - name of server + * @info - rpc_program + * @version - rpc_program version + * @authflavor - rpc_auth flavour to use + * + * Creates an RPC client structure, then pings the server in order to + * determine if it is up, and if it supports this program and version. + * + * This function should never be called by asynchronous tasks such as + * the portmapper. + */ +struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname, + struct rpc_program *info, u32 version, rpc_authflavor_t authflavor) +{ + struct rpc_clnt *clnt; + int err; + + clnt = rpc_new_client(xprt, servname, info, version, authflavor); + if (IS_ERR(clnt)) + return clnt; + err = rpc_ping(clnt, RPC_TASK_SOFT|RPC_TASK_NOINTR); + if (err == 0) + return clnt; + rpc_shutdown_client(clnt); + return ERR_PTR(err); +} + /* * This function clones the RPC client structure. It allows us to share the * same transport while varying parameters such as the authentication @@ -1086,3 +1116,30 @@ out_overflow: printk(KERN_WARNING "RPC %s: server reply was truncated.\n", __FUNCTION__); goto out_retry; } + +static int rpcproc_encode_null(void *rqstp, u32 *data, void *obj) +{ + return 0; +} + +static int rpcproc_decode_null(void *rqstp, u32 *data, void *obj) +{ + return 0; +} + +static struct rpc_procinfo rpcproc_null = { + .p_encode = rpcproc_encode_null, + .p_decode = rpcproc_decode_null, +}; + +int rpc_ping(struct rpc_clnt *clnt, int flags) +{ + struct rpc_message msg = { + .rpc_proc = &rpcproc_null, + }; + int err; + msg.rpc_cred = authnull_ops.lookup_cred(NULL, NULL, 0); + err = rpc_call_sync(clnt, &msg, flags); + put_rpccred(msg.rpc_cred); + return err; +} diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c index 97c420ff1ee0..df4d84c9020d 100644 --- a/net/sunrpc/pmap_clnt.c +++ b/net/sunrpc/pmap_clnt.c @@ -207,7 +207,7 @@ pmap_create(char *hostname, struct sockaddr_in *srvaddr, int proto) xprt->addr.sin_port = htons(RPC_PMAP_PORT); /* printk("pmap: create clnt\n"); */ - clnt = rpc_create_client(xprt, hostname, + clnt = rpc_new_client(xprt, hostname, &pmap_program, RPC_PMAP_VERSION, RPC_AUTH_UNIX); if (!IS_ERR(clnt)) { -- cgit v1.2.3 From 4ce79717ce32a9f88c1ddce4b9658556cb59d37a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:21 +0000 Subject: [PATCH] NFS: Header file cleanup... - Move NFSv4 state definitions into a private header file. - Clean up gunk in nfs_fs.h Signed-off-by: Trond Myklebust --- fs/nfs/callback.c | 1 + fs/nfs/callback_proc.c | 1 + fs/nfs/callback_xdr.c | 1 + fs/nfs/delegation.c | 1 + fs/nfs/dir.c | 1 + fs/nfs/idmap.c | 1 + fs/nfs/inode.c | 1 + fs/nfs/nfs4_fs.h | 250 +++++++++++++++++++++++++++++++++++++++++++++++++ fs/nfs/nfs4proc.c | 5 +- fs/nfs/nfs4renewd.c | 1 + fs/nfs/nfs4state.c | 12 +-- fs/nfs/nfs4xdr.c | 8 +- include/linux/nfs_fs.h | 241 ----------------------------------------------- 13 files changed, 264 insertions(+), 260 deletions(-) create mode 100644 fs/nfs/nfs4_fs.h (limited to 'include/linux') diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 560d6175dd58..f2ca782aba33 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -14,6 +14,7 @@ #include #include #include +#include "nfs4_fs.h" #include "callback.h" #define NFSDBG_FACILITY NFSDBG_CALLBACK diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index ece27e42b93b..65f1e19e4d19 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -8,6 +8,7 @@ #include #include #include +#include "nfs4_fs.h" #include "callback.h" #include "delegation.h" diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index d271df9df2b2..c99677ec58f8 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -10,6 +10,7 @@ #include #include #include +#include "nfs4_fs.h" #include "callback.h" #define CB_OP_TAGLEN_MAXSZ (512) diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 5b9c60f97791..d7f7eb669d03 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -16,6 +16,7 @@ #include #include +#include "nfs4_fs.h" #include "delegation.h" static struct nfs_delegation *nfs_alloc_delegation(void) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index ff6155f5e8d9..9ccb15e86967 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -32,6 +32,7 @@ #include #include +#include "nfs4_fs.h" #include "delegation.h" #define NFS_PARANOIA 1 diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c index 87f4f9aeac86..ffb8df91dc34 100644 --- a/fs/nfs/idmap.c +++ b/fs/nfs/idmap.c @@ -50,6 +50,7 @@ #include #include +#include "nfs4_fs.h" #define IDMAP_HASH_SZ 128 diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 32ddcf69e9ac..c80a81ff59c6 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -39,6 +39,7 @@ #include #include +#include "nfs4_fs.h" #include "delegation.h" #define NFSDBG_FACILITY NFSDBG_VFS diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h new file mode 100644 index 000000000000..85cf3bd36921 --- /dev/null +++ b/fs/nfs/nfs4_fs.h @@ -0,0 +1,250 @@ +/* + * linux/fs/nfs/nfs4_fs.h + * + * Copyright (C) 2005 Trond Myklebust + * + * NFSv4-specific filesystem definitions and declarations + */ + +#ifndef __LINUX_FS_NFS_NFS4_FS_H +#define __LINUX_FS_NFS_NFS4_FS_H + +#ifdef CONFIG_NFS_V4 + +struct idmap; + +/* + * In a seqid-mutating op, this macro controls which error return + * values trigger incrementation of the seqid. + * + * from rfc 3010: + * The client MUST monotonically increment the sequence number for the + * CLOSE, LOCK, LOCKU, OPEN, OPEN_CONFIRM, and OPEN_DOWNGRADE + * operations. This is true even in the event that the previous + * operation that used the sequence number received an error. The only + * exception to this rule is if the previous operation received one of + * the following errors: NFSERR_STALE_CLIENTID, NFSERR_STALE_STATEID, + * NFSERR_BAD_STATEID, NFSERR_BAD_SEQID, NFSERR_BADXDR, + * NFSERR_RESOURCE, NFSERR_NOFILEHANDLE. + * + */ +#define seqid_mutating_err(err) \ +(((err) != NFSERR_STALE_CLIENTID) && \ + ((err) != NFSERR_STALE_STATEID) && \ + ((err) != NFSERR_BAD_STATEID) && \ + ((err) != NFSERR_BAD_SEQID) && \ + ((err) != NFSERR_BAD_XDR) && \ + ((err) != NFSERR_RESOURCE) && \ + ((err) != NFSERR_NOFILEHANDLE)) + +enum nfs4_client_state { + NFS4CLNT_OK = 0, +}; + +/* + * The nfs4_client identifies our client state to the server. + */ +struct nfs4_client { + struct list_head cl_servers; /* Global list of servers */ + struct in_addr cl_addr; /* Server identifier */ + u64 cl_clientid; /* constant */ + nfs4_verifier cl_confirm; + unsigned long cl_state; + + u32 cl_lockowner_id; + + /* + * The following rwsem ensures exclusive access to the server + * while we recover the state following a lease expiration. + */ + struct rw_semaphore cl_sem; + + struct list_head cl_delegations; + struct list_head cl_state_owners; + struct list_head cl_unused; + int cl_nunused; + spinlock_t cl_lock; + atomic_t cl_count; + + struct rpc_clnt * cl_rpcclient; + struct rpc_cred * cl_cred; + + struct list_head cl_superblocks; /* List of nfs_server structs */ + + unsigned long cl_lease_time; + unsigned long cl_last_renewal; + struct work_struct cl_renewd; + struct work_struct cl_recoverd; + + wait_queue_head_t cl_waitq; + struct rpc_wait_queue cl_rpcwaitq; + + /* used for the setclientid verifier */ + struct timespec cl_boot_time; + + /* idmapper */ + struct idmap * cl_idmap; + + /* Our own IP address, as a null-terminated string. + * This is used to generate the clientid, and the callback address. + */ + char cl_ipaddr[16]; + unsigned char cl_id_uniquifier; +}; + +/* + * NFS4 state_owners and lock_owners are simply labels for ordered + * sequences of RPC calls. Their sole purpose is to provide once-only + * semantics by allowing the server to identify replayed requests. + * + * The ->so_sema is held during all state_owner seqid-mutating operations: + * OPEN, OPEN_DOWNGRADE, and CLOSE. Its purpose is to properly serialize + * so_seqid. + */ +struct nfs4_state_owner { + struct list_head so_list; /* per-clientid list of state_owners */ + struct nfs4_client *so_client; + u32 so_id; /* 32-bit identifier, unique */ + struct semaphore so_sema; + u32 so_seqid; /* protected by so_sema */ + atomic_t so_count; + + struct rpc_cred *so_cred; /* Associated cred */ + struct list_head so_states; + struct list_head so_delegations; +}; + +/* + * struct nfs4_state maintains the client-side state for a given + * (state_owner,inode) tuple (OPEN) or state_owner (LOCK). + * + * OPEN: + * In order to know when to OPEN_DOWNGRADE or CLOSE the state on the server, + * we need to know how many files are open for reading or writing on a + * given inode. This information too is stored here. + * + * LOCK: one nfs4_state (LOCK) to hold the lock stateid nfs4_state(OPEN) + */ + +struct nfs4_lock_state { + struct list_head ls_locks; /* Other lock stateids */ + fl_owner_t ls_owner; /* POSIX lock owner */ +#define NFS_LOCK_INITIALIZED 1 + int ls_flags; + u32 ls_seqid; + u32 ls_id; + nfs4_stateid ls_stateid; + atomic_t ls_count; +}; + +/* bits for nfs4_state->flags */ +enum { + LK_STATE_IN_USE, + NFS_DELEGATED_STATE, +}; + +struct nfs4_state { + struct list_head open_states; /* List of states for the same state_owner */ + struct list_head inode_states; /* List of states for the same inode */ + struct list_head lock_states; /* List of subservient lock stateids */ + + struct nfs4_state_owner *owner; /* Pointer to the open owner */ + struct inode *inode; /* Pointer to the inode */ + + unsigned long flags; /* Do we hold any locks? */ + struct semaphore lock_sema; /* Serializes file locking operations */ + rwlock_t state_lock; /* Protects the lock_states list */ + + nfs4_stateid stateid; + + unsigned int nreaders; + unsigned int nwriters; + int state; /* State on the server (R,W, or RW) */ + atomic_t count; +}; + + +struct nfs4_exception { + long timeout; + int retry; +}; + +struct nfs4_state_recovery_ops { + int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *); + int (*recover_lock)(struct nfs4_state *, struct file_lock *); +}; + +extern struct dentry_operations nfs4_dentry_operations; +extern struct inode_operations nfs4_dir_inode_operations; + +/* nfs4proc.c */ +extern int nfs4_map_errors(int err); +extern int nfs4_proc_setclientid(struct nfs4_client *, u32, unsigned short); +extern int nfs4_proc_setclientid_confirm(struct nfs4_client *); +extern int nfs4_proc_async_renew(struct nfs4_client *); +extern int nfs4_proc_renew(struct nfs4_client *); +extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode); +extern struct inode *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *); +extern int nfs4_open_revalidate(struct inode *, struct dentry *, int); + +extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops; +extern struct nfs4_state_recovery_ops nfs4_network_partition_recovery_ops; + +extern const u32 nfs4_fattr_bitmap[2]; +extern const u32 nfs4_statfs_bitmap[2]; +extern const u32 nfs4_pathconf_bitmap[2]; +extern const u32 nfs4_fsinfo_bitmap[2]; + +/* nfs4renewd.c */ +extern void nfs4_schedule_state_renewal(struct nfs4_client *); +extern void nfs4_renewd_prepare_shutdown(struct nfs_server *); +extern void nfs4_kill_renewd(struct nfs4_client *); +extern void nfs4_renew_state(void *); + +/* nfs4state.c */ +extern void init_nfsv4_state(struct nfs_server *); +extern void destroy_nfsv4_state(struct nfs_server *); +extern struct nfs4_client *nfs4_get_client(struct in_addr *); +extern void nfs4_put_client(struct nfs4_client *clp); +extern int nfs4_init_client(struct nfs4_client *clp); +extern struct nfs4_client *nfs4_find_client(struct in_addr *); +extern u32 nfs4_alloc_lockowner_id(struct nfs4_client *); + +extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *); +extern void nfs4_put_state_owner(struct nfs4_state_owner *); +extern void nfs4_drop_state_owner(struct nfs4_state_owner *); +extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *); +extern void nfs4_put_open_state(struct nfs4_state *); +extern void nfs4_close_state(struct nfs4_state *, mode_t); +extern struct nfs4_state *nfs4_find_state(struct inode *, struct rpc_cred *, mode_t mode); +extern void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp); +extern void nfs4_schedule_state_recovery(struct nfs4_client *); +extern struct nfs4_lock_state *nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t); +extern struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t); +extern void nfs4_put_lock_state(struct nfs4_lock_state *state); +extern void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *ls); +extern void nfs4_notify_setlk(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *); +extern void nfs4_notify_unlck(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *); +extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t); + +extern const nfs4_stateid zero_stateid; + +/* nfs4xdr.c */ +extern uint32_t *nfs4_decode_dirent(uint32_t *p, struct nfs_entry *entry, int plus); +extern struct rpc_procinfo nfs4_procedures[]; + +struct nfs4_mount_data; + +/* callback_xdr.c */ +extern struct svc_version nfs4_callback_version1; + +#else + +#define init_nfsv4_state(server) do { } while (0) +#define destroy_nfsv4_state(server) do { } while (0) +#define nfs4_put_state_owner(inode, owner) do { } while (0) +#define nfs4_put_open_state(state) do { } while (0) +#define nfs4_close_state(a, b) do { } while (0) + +#endif /* CONFIG_NFS_V4 */ +#endif /* __LINUX_FS_NFS_NFS4_FS.H */ diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 1d5cb3e80c3e..a69c02b206c1 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -48,6 +48,7 @@ #include #include +#include "nfs4_fs.h" #include "delegation.h" #define NFSDBG_FACILITY NFSDBG_PROC @@ -62,8 +63,6 @@ static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struc extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus); extern struct rpc_procinfo nfs4_procedures[]; -extern nfs4_stateid zero_stateid; - /* Prevent leaks of NFSv4 errors into userland */ int nfs4_map_errors(int err) { @@ -104,7 +103,7 @@ const u32 nfs4_statfs_bitmap[2] = { | FATTR4_WORD1_SPACE_TOTAL }; -u32 nfs4_pathconf_bitmap[2] = { +const u32 nfs4_pathconf_bitmap[2] = { FATTR4_WORD0_MAXLINK | FATTR4_WORD0_MAXNAME, 0 diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c index 667e06f1c647..a3001628ad32 100644 --- a/fs/nfs/nfs4renewd.c +++ b/fs/nfs/nfs4renewd.c @@ -53,6 +53,7 @@ #include #include #include +#include "nfs4_fs.h" #define NFSDBG_FACILITY NFSDBG_PROC diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 231cebce3c87..17b187f2d776 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -46,24 +46,18 @@ #include #include +#include "nfs4_fs.h" #include "callback.h" #include "delegation.h" #define OPENOWNER_POOL_SIZE 8 -static DEFINE_SPINLOCK(state_spinlock); - -nfs4_stateid zero_stateid; - -#if 0 -nfs4_stateid one_stateid = - { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }; -#endif +const nfs4_stateid zero_stateid; +static DEFINE_SPINLOCK(state_spinlock); static LIST_HEAD(nfs4_clientid_list); static void nfs4_recover_state(void *); -extern void nfs4_renew_state(void *); void init_nfsv4_state(struct nfs_server *server) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 5f4de05763c9..e86406eff0eb 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -51,6 +51,7 @@ #include #include #include +#include "nfs4_fs.h" #define NFSDBG_FACILITY NFSDBG_XDR @@ -660,8 +661,6 @@ static int encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm1 static int encode_getfattr(struct xdr_stream *xdr, const u32* bitmask) { - extern u32 nfs4_fattr_bitmap[]; - return encode_getattr_two(xdr, bitmask[0] & nfs4_fattr_bitmap[0], bitmask[1] & nfs4_fattr_bitmap[1]); @@ -669,8 +668,6 @@ static int encode_getfattr(struct xdr_stream *xdr, const u32* bitmask) static int encode_fsinfo(struct xdr_stream *xdr, const u32* bitmask) { - extern u32 nfs4_fsinfo_bitmap[]; - return encode_getattr_two(xdr, bitmask[0] & nfs4_fsinfo_bitmap[0], bitmask[1] & nfs4_fsinfo_bitmap[1]); } @@ -969,7 +966,6 @@ static int encode_putrootfh(struct xdr_stream *xdr) static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context *ctx) { - extern nfs4_stateid zero_stateid; nfs4_stateid stateid; uint32_t *p; @@ -1697,7 +1693,6 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs */ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, uint32_t *p, const struct nfs4_pathconf_arg *args) { - extern u32 nfs4_pathconf_bitmap[2]; struct xdr_stream xdr; struct compound_hdr hdr = { .nops = 2, @@ -1718,7 +1713,6 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, uint32_t *p, const struct */ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, uint32_t *p, const struct nfs4_statfs_arg *args) { - extern u32 nfs4_statfs_bitmap[]; struct xdr_stream xdr; struct compound_hdr hdr = { .nops = 2, diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index dbac7f363e5d..fb33e7655cfa 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -15,7 +15,6 @@ #include #include #include -#include #include @@ -29,7 +28,6 @@ #include #include #include -#include #include /* @@ -43,13 +41,6 @@ #define NFS_MAX_FILE_IO_BUFFER_SIZE 32768 #define NFS_DEF_FILE_IO_BUFFER_SIZE 4096 -/* - * The upper limit on timeouts for the exponential backoff algorithm. - */ -#define NFS_WRITEBACK_DELAY (5*HZ) -#define NFS_WRITEBACK_LOCKDELAY (60*HZ) -#define NFS_COMMIT_DELAY (5*HZ) - /* * superblock magic number for NFS */ @@ -60,9 +51,6 @@ */ #define NFS_RPC_SWAPFLAGS (RPC_TASK_SWAPPER|RPC_TASK_ROOTCREDS) -#define NFS_RW_SYNC 0x0001 /* O_SYNC handling */ -#define NFS_RW_SWAP 0x0002 /* This is a swap request */ - /* * When flushing a cluster of dirty pages, there can be different * strategies: @@ -434,11 +422,6 @@ static inline void nfs_writedata_free(struct nfs_write_data *p) mempool_free(p, nfs_wdata_mempool); } -/* Hack for future NFS swap support */ -#ifndef IS_SWAPFILE -# define IS_SWAPFILE(inode) (0) -#endif - /* * linux/fs/nfs/read.c */ @@ -515,230 +498,6 @@ extern void * nfs_root_data(void); #define NFS_JUKEBOX_RETRY_TIME (5 * HZ) -#ifdef CONFIG_NFS_V4 - -struct idmap; - -/* - * In a seqid-mutating op, this macro controls which error return - * values trigger incrementation of the seqid. - * - * from rfc 3010: - * The client MUST monotonically increment the sequence number for the - * CLOSE, LOCK, LOCKU, OPEN, OPEN_CONFIRM, and OPEN_DOWNGRADE - * operations. This is true even in the event that the previous - * operation that used the sequence number received an error. The only - * exception to this rule is if the previous operation received one of - * the following errors: NFSERR_STALE_CLIENTID, NFSERR_STALE_STATEID, - * NFSERR_BAD_STATEID, NFSERR_BAD_SEQID, NFSERR_BADXDR, - * NFSERR_RESOURCE, NFSERR_NOFILEHANDLE. - * - */ -#define seqid_mutating_err(err) \ -(((err) != NFSERR_STALE_CLIENTID) && \ - ((err) != NFSERR_STALE_STATEID) && \ - ((err) != NFSERR_BAD_STATEID) && \ - ((err) != NFSERR_BAD_SEQID) && \ - ((err) != NFSERR_BAD_XDR) && \ - ((err) != NFSERR_RESOURCE) && \ - ((err) != NFSERR_NOFILEHANDLE)) - -enum nfs4_client_state { - NFS4CLNT_OK = 0, -}; - -/* - * The nfs4_client identifies our client state to the server. - */ -struct nfs4_client { - struct list_head cl_servers; /* Global list of servers */ - struct in_addr cl_addr; /* Server identifier */ - u64 cl_clientid; /* constant */ - nfs4_verifier cl_confirm; - unsigned long cl_state; - - u32 cl_lockowner_id; - - /* - * The following rwsem ensures exclusive access to the server - * while we recover the state following a lease expiration. - */ - struct rw_semaphore cl_sem; - - struct list_head cl_delegations; - struct list_head cl_state_owners; - struct list_head cl_unused; - int cl_nunused; - spinlock_t cl_lock; - atomic_t cl_count; - - struct rpc_clnt * cl_rpcclient; - struct rpc_cred * cl_cred; - - struct list_head cl_superblocks; /* List of nfs_server structs */ - - unsigned long cl_lease_time; - unsigned long cl_last_renewal; - struct work_struct cl_renewd; - struct work_struct cl_recoverd; - - wait_queue_head_t cl_waitq; - struct rpc_wait_queue cl_rpcwaitq; - - /* used for the setclientid verifier */ - struct timespec cl_boot_time; - - /* idmapper */ - struct idmap * cl_idmap; - - /* Our own IP address, as a null-terminated string. - * This is used to generate the clientid, and the callback address. - */ - char cl_ipaddr[16]; - unsigned char cl_id_uniquifier; -}; - -/* - * NFS4 state_owners and lock_owners are simply labels for ordered - * sequences of RPC calls. Their sole purpose is to provide once-only - * semantics by allowing the server to identify replayed requests. - * - * The ->so_sema is held during all state_owner seqid-mutating operations: - * OPEN, OPEN_DOWNGRADE, and CLOSE. Its purpose is to properly serialize - * so_seqid. - */ -struct nfs4_state_owner { - struct list_head so_list; /* per-clientid list of state_owners */ - struct nfs4_client *so_client; - u32 so_id; /* 32-bit identifier, unique */ - struct semaphore so_sema; - u32 so_seqid; /* protected by so_sema */ - atomic_t so_count; - - struct rpc_cred *so_cred; /* Associated cred */ - struct list_head so_states; - struct list_head so_delegations; -}; - -/* - * struct nfs4_state maintains the client-side state for a given - * (state_owner,inode) tuple (OPEN) or state_owner (LOCK). - * - * OPEN: - * In order to know when to OPEN_DOWNGRADE or CLOSE the state on the server, - * we need to know how many files are open for reading or writing on a - * given inode. This information too is stored here. - * - * LOCK: one nfs4_state (LOCK) to hold the lock stateid nfs4_state(OPEN) - */ - -struct nfs4_lock_state { - struct list_head ls_locks; /* Other lock stateids */ - fl_owner_t ls_owner; /* POSIX lock owner */ -#define NFS_LOCK_INITIALIZED 1 - int ls_flags; - u32 ls_seqid; - u32 ls_id; - nfs4_stateid ls_stateid; - atomic_t ls_count; -}; - -/* bits for nfs4_state->flags */ -enum { - LK_STATE_IN_USE, - NFS_DELEGATED_STATE, -}; - -struct nfs4_state { - struct list_head open_states; /* List of states for the same state_owner */ - struct list_head inode_states; /* List of states for the same inode */ - struct list_head lock_states; /* List of subservient lock stateids */ - - struct nfs4_state_owner *owner; /* Pointer to the open owner */ - struct inode *inode; /* Pointer to the inode */ - - unsigned long flags; /* Do we hold any locks? */ - struct semaphore lock_sema; /* Serializes file locking operations */ - rwlock_t state_lock; /* Protects the lock_states list */ - - nfs4_stateid stateid; - - unsigned int nreaders; - unsigned int nwriters; - int state; /* State on the server (R,W, or RW) */ - atomic_t count; -}; - - -struct nfs4_exception { - long timeout; - int retry; -}; - -struct nfs4_state_recovery_ops { - int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *); - int (*recover_lock)(struct nfs4_state *, struct file_lock *); -}; - -extern struct dentry_operations nfs4_dentry_operations; -extern struct inode_operations nfs4_dir_inode_operations; - -/* nfs4proc.c */ -extern int nfs4_map_errors(int err); -extern int nfs4_proc_setclientid(struct nfs4_client *, u32, unsigned short); -extern int nfs4_proc_setclientid_confirm(struct nfs4_client *); -extern int nfs4_proc_async_renew(struct nfs4_client *); -extern int nfs4_proc_renew(struct nfs4_client *); -extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode); -extern struct inode *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *); -extern int nfs4_open_revalidate(struct inode *, struct dentry *, int); - -extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops; -extern struct nfs4_state_recovery_ops nfs4_network_partition_recovery_ops; - -/* nfs4renewd.c */ -extern void nfs4_schedule_state_renewal(struct nfs4_client *); -extern void nfs4_renewd_prepare_shutdown(struct nfs_server *); -extern void nfs4_kill_renewd(struct nfs4_client *); - -/* nfs4state.c */ -extern void init_nfsv4_state(struct nfs_server *); -extern void destroy_nfsv4_state(struct nfs_server *); -extern struct nfs4_client *nfs4_get_client(struct in_addr *); -extern void nfs4_put_client(struct nfs4_client *clp); -extern int nfs4_init_client(struct nfs4_client *clp); -extern struct nfs4_client *nfs4_find_client(struct in_addr *); -extern u32 nfs4_alloc_lockowner_id(struct nfs4_client *); - -extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *); -extern void nfs4_put_state_owner(struct nfs4_state_owner *); -extern void nfs4_drop_state_owner(struct nfs4_state_owner *); -extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *); -extern void nfs4_put_open_state(struct nfs4_state *); -extern void nfs4_close_state(struct nfs4_state *, mode_t); -extern struct nfs4_state *nfs4_find_state(struct inode *, struct rpc_cred *, mode_t mode); -extern void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp); -extern void nfs4_schedule_state_recovery(struct nfs4_client *); -extern struct nfs4_lock_state *nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t); -extern struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t); -extern void nfs4_put_lock_state(struct nfs4_lock_state *state); -extern void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *ls); -extern void nfs4_notify_setlk(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *); -extern void nfs4_notify_unlck(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *); -extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t); - - - -struct nfs4_mount_data; -#else -#define init_nfsv4_state(server) do { } while (0) -#define destroy_nfsv4_state(server) do { } while (0) -#define nfs4_put_state_owner(inode, owner) do { } while (0) -#define nfs4_put_open_state(state) do { } while (0) -#define nfs4_close_state(a, b) do { } while (0) -#define nfs4_renewd_prepare_shutdown(server) do { } while (0) -#endif - #endif /* __KERNEL__ */ /* -- cgit v1.2.3 From a656db998785324a818005bcf71bae6dcbbb3cf5 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:21 +0000 Subject: [PATCH] NFS: Remove unused NFS inode field readdir_timestamp. Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 8 +++----- include/linux/nfs_fs.h | 1 - 2 files changed, 3 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 9ccb15e86967..dffa21abd3ea 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -165,12 +165,10 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) NFS_FLAGS(inode) |= NFS_INO_INVALID_ATIME; /* Ensure consistent page alignment of the data. * Note: assumes we have exclusive access to this mapping either - * throught inode->i_sem or some other mechanism. + * through inode->i_sem or some other mechanism. */ - if (page->index == 0) { - invalidate_inode_pages(inode->i_mapping); - NFS_I(inode)->readdir_timestamp = timestamp; - } + if (page->index == 0) + invalidate_inode_pages2_range(inode->i_mapping, PAGE_CACHE_SIZE, -1); unlock_page(page); return 0; error: diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index fb33e7655cfa..68d5aae89972 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -128,7 +128,6 @@ struct nfs_inode { * * mtime != read_cache_mtime */ - unsigned long readdir_timestamp; unsigned long read_cache_jiffies; unsigned long attrtimeo; unsigned long attrtimeo_timestamp; -- cgit v1.2.3 From 96651ab341cde0fee940ec837f323d711cbfa7d5 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:21 +0000 Subject: [PATCH] RPC: Shrink struct rpc_task by switching to wait_on_bit() Signed-off-by: Trond Myklebust --- include/linux/sunrpc/sched.h | 1 - net/sunrpc/sched.c | 31 ++++++++++++++++++------------- 2 files changed, 18 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index 99d17ed7cebb..4d77e90d0b30 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -31,7 +31,6 @@ struct rpc_wait_queue; struct rpc_wait { struct list_head list; /* wait queue links */ struct list_head links; /* Links to related tasks */ - wait_queue_head_t waitq; /* sync: sleep on this q */ struct rpc_wait_queue * rpc_waitq; /* RPC wait queue we're on */ }; diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index cc298fa4b81d..2d9eb7fbd521 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -290,7 +290,7 @@ static void rpc_make_runnable(struct rpc_task *task) return; } } else - wake_up(&task->u.tk_wait.waitq); + wake_up_bit(&task->tk_runstate, RPC_TASK_QUEUED); } /* @@ -578,6 +578,14 @@ static inline int __rpc_do_exit(struct rpc_task *task) return 1; } +static int rpc_wait_bit_interruptible(void *word) +{ + if (signal_pending(current)) + return -ERESTARTSYS; + schedule(); + return 0; +} + /* * This is the RPC `scheduler' (or rather, the finite state machine). */ @@ -648,22 +656,21 @@ static int __rpc_execute(struct rpc_task *task) /* sync task: sleep here */ dprintk("RPC: %4d sync task going to sleep\n", task->tk_pid); - if (RPC_TASK_UNINTERRUPTIBLE(task)) { - __wait_event(task->u.tk_wait.waitq, !RPC_IS_QUEUED(task)); - } else { - __wait_event_interruptible(task->u.tk_wait.waitq, !RPC_IS_QUEUED(task), status); + /* Note: Caller should be using rpc_clnt_sigmask() */ + status = out_of_line_wait_on_bit(&task->tk_runstate, + RPC_TASK_QUEUED, rpc_wait_bit_interruptible, + TASK_INTERRUPTIBLE); + if (status == -ERESTARTSYS) { /* * When a sync task receives a signal, it exits with * -ERESTARTSYS. In order to catch any callbacks that * clean up after sleeping on some queue, we don't * break the loop here, but go around once more. */ - if (status == -ERESTARTSYS) { - dprintk("RPC: %4d got signal\n", task->tk_pid); - task->tk_flags |= RPC_TASK_KILLED; - rpc_exit(task, -ERESTARTSYS); - rpc_wake_up_task(task); - } + dprintk("RPC: %4d got signal\n", task->tk_pid); + task->tk_flags |= RPC_TASK_KILLED; + rpc_exit(task, -ERESTARTSYS); + rpc_wake_up_task(task); } rpc_set_running(task); dprintk("RPC: %4d sync task resuming\n", task->tk_pid); @@ -766,8 +773,6 @@ void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, rpc_action call /* Initialize workqueue for async tasks */ task->tk_workqueue = rpciod_workqueue; - if (!RPC_IS_ASYNC(task)) - init_waitqueue_head(&task->u.tk_wait.waitq); if (clnt) { atomic_inc(&clnt->cl_users); -- cgit v1.2.3 From 464a98bd70bae8c559cfc82af799faf44824ce64 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:21 +0000 Subject: [PATCH] NFS: cleanup: shrink struct nfs_open_context Remove the wait queue, and replace the functions that depended on it with wait_on_bit(). Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 1 - fs/nfs/pagelist.c | 35 ++++++++++++++++++++++++++++------- include/linux/nfs_fs.h | 1 - 3 files changed, 28 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index c80a81ff59c6..a38d4b22d1f8 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -848,7 +848,6 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rp ctx->state = NULL; ctx->lockowner = current->files; ctx->error = 0; - init_waitqueue_head(&ctx->waitq); } return ctx; } diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 4f1ba723848d..80777f99a58a 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -107,7 +107,7 @@ void nfs_unlock_request(struct nfs_page *req) smp_mb__before_clear_bit(); clear_bit(PG_BUSY, &req->wb_flags); smp_mb__after_clear_bit(); - wake_up_all(&req->wb_context->waitq); + wake_up_bit(&req->wb_flags, PG_BUSY); nfs_release_request(req); } @@ -180,6 +180,17 @@ nfs_list_add_request(struct nfs_page *req, struct list_head *head) req->wb_list_head = head; } +static int nfs_wait_bit_interruptible(void *word) +{ + int ret = 0; + + if (signal_pending(current)) + ret = -ERESTARTSYS; + else + schedule(); + return ret; +} + /** * nfs_wait_on_request - Wait for a request to complete. * @req: request to wait upon. @@ -190,12 +201,22 @@ nfs_list_add_request(struct nfs_page *req, struct list_head *head) int nfs_wait_on_request(struct nfs_page *req) { - struct inode *inode = req->wb_context->dentry->d_inode; - struct rpc_clnt *clnt = NFS_CLIENT(inode); - - if (!NFS_WBACK_BUSY(req)) - return 0; - return nfs_wait_event(clnt, req->wb_context->waitq, !NFS_WBACK_BUSY(req)); + struct rpc_clnt *clnt = NFS_CLIENT(req->wb_context->dentry->d_inode); + sigset_t oldmask; + int ret = 0; + + if (!test_bit(PG_BUSY, &req->wb_flags)) + goto out; + /* + * Note: the call to rpc_clnt_sigmask() suffices to ensure that we + * are not interrupted if intr flag is not set + */ + rpc_clnt_sigmask(clnt, &oldmask); + ret = out_of_line_wait_on_bit(&req->wb_flags, PG_BUSY, + nfs_wait_bit_interruptible, TASK_INTERRUPTIBLE); + rpc_clnt_sigunmask(clnt, &oldmask); +out: + return ret; } /** diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 68d5aae89972..0b01b96337f8 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -84,7 +84,6 @@ struct nfs_open_context { int error; struct list_head list; - wait_queue_head_t waitq; }; /* -- cgit v1.2.3 From 92cfc62cb8412c9563860b1bf70cd4701f03092e Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 22 Jun 2005 17:16:22 +0000 Subject: [PATCH] NFS: Allow NFS versions to support different sets of inode operations. ACL support will require supporting additional inode operations in v4 (getxattr, setxattr, listxattr). This patch allows different protocol versions to support different inode operations by adding a file_inode_ops to the nfs_rpc_ops (to match the existing dir_inode_ops). Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 2 +- fs/nfs/nfs3proc.c | 1 + fs/nfs/nfs4proc.c | 1 + fs/nfs/proc.c | 1 + include/linux/nfs_xdr.h | 1 + 5 files changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index a38d4b22d1f8..a82f0340744f 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -686,7 +686,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) /* Why so? Because we want revalidate for devices/FIFOs, and * that's precisely what we have in nfs_file_inode_operations. */ - inode->i_op = &nfs_file_inode_operations; + inode->i_op = NFS_SB(sb)->rpc_ops->file_inode_ops; if (S_ISREG(inode->i_mode)) { inode->i_fop = &nfs_file_operations; inode->i_data.a_ops = &nfs_file_aops; diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 3878494dfc2c..53953a775714 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -826,6 +826,7 @@ struct nfs_rpc_ops nfs_v3_clientops = { .version = 3, /* protocol version */ .dentry_ops = &nfs_dentry_operations, .dir_inode_ops = &nfs_dir_inode_operations, + .file_inode_ops = &nfs_file_inode_operations, .getroot = nfs3_proc_get_root, .getattr = nfs3_proc_getattr, .setattr = nfs3_proc_setattr, diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index a69c02b206c1..a5a8cb3159a0 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2746,6 +2746,7 @@ struct nfs_rpc_ops nfs_v4_clientops = { .version = 4, /* protocol version */ .dentry_ops = &nfs4_dentry_operations, .dir_inode_ops = &nfs4_dir_inode_operations, + .file_inode_ops = &nfs_file_inode_operations, .getroot = nfs4_proc_get_root, .getattr = nfs4_proc_getattr, .setattr = nfs4_proc_setattr, diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index d31b4d6e5a5e..cedf636bcf3c 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -622,6 +622,7 @@ struct nfs_rpc_ops nfs_v2_clientops = { .version = 2, /* protocol version */ .dentry_ops = &nfs_dentry_operations, .dir_inode_ops = &nfs_dir_inode_operations, + .file_inode_ops = &nfs_file_inode_operations, .getroot = nfs_proc_get_root, .getattr = nfs_proc_getattr, .setattr = nfs_proc_setattr, diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 47037d9521cb..5b45bafd9db5 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -667,6 +667,7 @@ struct nfs_rpc_ops { int version; /* Protocol version */ struct dentry_operations *dentry_ops; struct inode_operations *dir_inode_ops; + struct inode_operations *file_inode_ops; int (*getroot) (struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); -- cgit v1.2.3 From ada70d9425bcc5e376fef8591e4e76e204c0834c Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:22 +0000 Subject: [PATCH] NFS: Add hooks to allow common NFS attribute code to clear cached acls Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 33 ++++++++++++++++++++++++++------- include/linux/nfs_fs.h | 1 + include/linux/nfs_xdr.h | 1 + 3 files changed, 28 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index a82f0340744f..c45bd52cc1d7 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -64,6 +64,7 @@ static void nfs_clear_inode(struct inode *); static void nfs_umount_begin(struct super_block *); static int nfs_statfs(struct super_block *, struct kstatfs *); static int nfs_show_options(struct seq_file *, struct vfsmount *); +static void nfs_zap_acl_cache(struct inode *); static struct rpc_program nfs_program; @@ -153,6 +154,7 @@ nfs_clear_inode(struct inode *inode) nfs_wb_all(inode); BUG_ON (!list_empty(&nfsi->open_files)); + nfs_zap_acl_cache(inode); cred = nfsi->cache_access.cred; if (cred) put_rpccred(cred); @@ -587,9 +589,19 @@ nfs_zap_caches(struct inode *inode) memset(NFS_COOKIEVERF(inode), 0, sizeof(NFS_COOKIEVERF(inode))); if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)) - nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS; + nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; else - nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS; + nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; +} + +static void nfs_zap_acl_cache(struct inode *inode) +{ + void (*clear_acl_cache)(struct inode *); + + clear_acl_cache = NFS_PROTO(inode)->clear_acl_cache; + if (clear_acl_cache != NULL) + clear_acl_cache(inode); + NFS_I(inode)->flags &= ~NFS_INO_INVALID_ACL; } /* @@ -789,7 +801,7 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr) } } if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) - NFS_FLAGS(inode) |= NFS_INO_INVALID_ACCESS; + NFS_FLAGS(inode) |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; nfs_end_data_update(inode); unlock_kernel(); return error; @@ -1033,6 +1045,8 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) /* This ensures we revalidate dentries */ nfsi->cache_change_attribute++; } + if (flags & NFS_INO_INVALID_ACL) + nfs_zap_acl_cache(inode); dfprintk(PAGECACHE, "NFS: (%s/%Ld) revalidation complete\n", inode->i_sb->s_id, (long long)NFS_FILEID(inode)); @@ -1183,7 +1197,7 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO) || inode->i_uid != fattr->uid || inode->i_gid != fattr->gid) - nfsi->flags |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS; + nfsi->flags |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL; /* Has the link count changed? */ if (inode->i_nlink != fattr->nlink) @@ -1292,16 +1306,21 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign #endif nfsi->change_attr = fattr->change_attr; if (!data_unstable) - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS; + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; } - memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime)); + /* If ctime has changed we should definitely clear access+acl caches */ + if (!timespec_equal(&inode->i_ctime, &fattr->ctime)) { + if (!data_unstable) + invalid |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime)); + } memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime)); if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO) || inode->i_uid != fattr->uid || inode->i_gid != fattr->gid) - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS; + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; inode->i_mode = fattr->mode; inode->i_nlink = fattr->nlink; diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 0b01b96337f8..140bdf489f71 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -189,6 +189,7 @@ struct nfs_inode { #define NFS_INO_INVALID_DATA 0x0010 /* cached data is invalid */ #define NFS_INO_INVALID_ATIME 0x0020 /* cached atime is invalid */ #define NFS_INO_INVALID_ACCESS 0x0040 /* cached access cred invalid */ +#define NFS_INO_INVALID_ACL 0x0080 /* cached acls are invalid */ static inline struct nfs_inode *NFS_I(struct inode *inode) { diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 5b45bafd9db5..cf38db59f347 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -714,6 +714,7 @@ struct nfs_rpc_ops { int (*file_open) (struct inode *, struct file *); int (*file_release) (struct inode *, struct file *); int (*lock)(struct file *, int, struct file_lock *); + void (*clear_acl_cache)(struct inode *); }; /* -- cgit v1.2.3 From 029d105e66e5a90850d5a09dad76815d0bcfcaa3 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 22 Jun 2005 17:16:22 +0000 Subject: [PATCH] NFSv4: Client-side xdr for reading NFSv4 acls Client-side support for NFSv4 acls: xdr encoding and decoding routines for reading acls Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 100 ++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/nfs4.h | 1 + include/linux/nfs_xdr.h | 7 ++++ 3 files changed, 108 insertions(+) (limited to 'include/linux') diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 8204926bb467..6f1c003ee33a 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -365,6 +365,13 @@ static int nfs_stat_to_errno(int); encode_delegreturn_maxsz) #define NFS4_dec_delegreturn_sz (compound_decode_hdr_maxsz + \ decode_delegreturn_maxsz) +#define NFS4_enc_getacl_sz (compound_encode_hdr_maxsz + \ + encode_putfh_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_getacl_sz (compound_decode_hdr_maxsz + \ + decode_putfh_maxsz + \ + op_decode_hdr_maxsz + \ + nfs4_fattr_bitmap_maxsz + 1) static struct { unsigned int mode; @@ -1631,6 +1638,34 @@ out: return status; } +/* + * Encode a GETACL request + */ +static int +nfs4_xdr_enc_getacl(struct rpc_rqst *req, uint32_t *p, + struct nfs_getaclargs *args) +{ + struct xdr_stream xdr; + struct rpc_auth *auth = req->rq_task->tk_auth; + struct compound_hdr hdr = { + .nops = 2, + }; + int replen, status; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + encode_compound_hdr(&xdr, &hdr); + status = encode_putfh(&xdr, args->fh); + if (status) + goto out; + status = encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0); + /* set up reply buffer: */ + replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_getacl_sz) << 2; + xdr_inline_pages(&req->rq_rcv_buf, replen, + args->acl_pages, args->acl_pgbase, args->acl_len); +out: + return status; +} + /* * Encode a WRITE request */ @@ -3125,6 +3160,47 @@ static int decode_renew(struct xdr_stream *xdr) return decode_op_hdr(xdr, OP_RENEW); } +static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, + size_t *acl_len) +{ + uint32_t *savep; + uint32_t attrlen, + bitmap[2] = {0}; + struct kvec *iov = req->rq_rcv_buf.head; + int status; + + *acl_len = 0; + if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) + goto out; + if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) + goto out; + if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) + goto out; + + if (unlikely(bitmap[0] & (FATTR4_WORD0_ACL - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_ACL)) { + int hdrlen, recvd; + + /* We ignore &savep and don't do consistency checks on + * the attr length. Let userspace figure it out.... */ + hdrlen = (u8 *)xdr->p - (u8 *)iov->iov_base; + recvd = req->rq_rcv_buf.len - hdrlen; + if (attrlen > recvd) { + printk(KERN_WARNING "NFS: server cheating in getattr" + " acl reply: attrlen %u > recvd %u\n", + attrlen, recvd); + return -EINVAL; + } + if (attrlen <= *acl_len) + xdr_read_pages(xdr, attrlen); + *acl_len = attrlen; + } + +out: + return status; +} + static int decode_savefh(struct xdr_stream *xdr) { @@ -3417,6 +3493,29 @@ out: } +/* + * Decode GETACL response + */ +static int +nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, uint32_t *p, size_t *acl_len) +{ + struct xdr_stream xdr; + struct compound_hdr hdr; + int status; + + xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_putfh(&xdr); + if (status) + goto out; + status = decode_getacl(&xdr, rqstp, acl_len); + +out: + return status; +} + /* * Decode CLOSE response */ @@ -4017,6 +4116,7 @@ struct rpc_procinfo nfs4_procedures[] = { PROC(READDIR, enc_readdir, dec_readdir), PROC(SERVER_CAPS, enc_server_caps, dec_server_caps), PROC(DELEGRETURN, enc_delegreturn, dec_delegreturn), + PROC(GETACL, enc_getacl, dec_getacl), }; struct rpc_version nfs_version4 = { diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index 5ca8a8d8ccdf..6ee7e2585af5 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -382,6 +382,7 @@ enum { NFSPROC4_CLNT_READDIR, NFSPROC4_CLNT_SERVER_CAPS, NFSPROC4_CLNT_DELEGRETURN, + NFSPROC4_CLNT_GETACL, }; #endif diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index cf38db59f347..9f5e1d407c7b 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -326,6 +326,13 @@ struct nfs_setattrargs { const u32 * bitmask; }; +struct nfs_getaclargs { + struct nfs_fh * fh; + size_t acl_len; + unsigned int acl_pgbase; + struct page ** acl_pages; +}; + struct nfs_setattrres { struct nfs_fattr * fattr; const struct nfs_server * server; -- cgit v1.2.3 From 23ec6965c20db96bc8ea7af0ec178f074dd31c40 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 22 Jun 2005 17:16:22 +0000 Subject: [PATCH] NFSv4: Client-side xdr for writing NFSv4 acls Client-side support for NFSv4 acls: xdr encoding and decoding routines for writing acls Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 71 ++++++++++++++++++++++++++++++++++++++++++++++++- include/linux/nfs4.h | 1 + include/linux/nfs_xdr.h | 7 +++++ 3 files changed, 78 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 6f1c003ee33a..325cd6d4f23a 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -372,6 +372,13 @@ static int nfs_stat_to_errno(int); decode_putfh_maxsz + \ op_decode_hdr_maxsz + \ nfs4_fattr_bitmap_maxsz + 1) +#define NFS4_enc_setacl_sz (compound_encode_hdr_maxsz + \ + encode_putfh_maxsz + \ + op_encode_hdr_maxsz + 4 + \ + nfs4_fattr_bitmap_maxsz + 1) +#define NFS4_dec_setacl_sz (compound_decode_hdr_maxsz + \ + decode_putfh_maxsz + \ + op_decode_hdr_maxsz + nfs4_fattr_bitmap_maxsz) static struct { unsigned int mode; @@ -471,7 +478,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s * In the worst-case, this would be * 12(bitmap) + 4(attrlen) + 8(size) + 4(mode) + 4(atime) + 4(mtime) * = 36 bytes, plus any contribution from variable-length fields - * such as owner/group/acl's. + * such as owner/group. */ len = 16; @@ -1095,6 +1102,25 @@ static int encode_renew(struct xdr_stream *xdr, const struct nfs4_client *client return 0; } +static int +encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg) +{ + uint32_t *p; + + RESERVE_SPACE(4+sizeof(zero_stateid.data)); + WRITE32(OP_SETATTR); + WRITEMEM(zero_stateid.data, sizeof(zero_stateid.data)); + RESERVE_SPACE(2*4); + WRITE32(1); + WRITE32(FATTR4_WORD0_ACL); + if (arg->acl_len % 4) + return -EINVAL; + RESERVE_SPACE(4); + WRITE32(arg->acl_len); + xdr_write_pages(xdr, arg->acl_pages, arg->acl_pgbase, arg->acl_len); + return 0; +} + static int encode_savefh(struct xdr_stream *xdr) { @@ -3492,6 +3518,48 @@ out: } +/* + * Encode an SETACL request + */ +static int +nfs4_xdr_enc_setacl(struct rpc_rqst *req, uint32_t *p, struct nfs_setaclargs *args) +{ + struct xdr_stream xdr; + struct compound_hdr hdr = { + .nops = 2, + }; + int status; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + encode_compound_hdr(&xdr, &hdr); + status = encode_putfh(&xdr, args->fh); + if (status) + goto out; + status = encode_setacl(&xdr, args); +out: + return status; +} +/* + * Decode SETACL response + */ +static int +nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, uint32_t *p, void *res) +{ + struct xdr_stream xdr; + struct compound_hdr hdr; + int status; + + xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_putfh(&xdr); + if (status) + goto out; + status = decode_setattr(&xdr, res); +out: + return status; +} /* * Decode GETACL response @@ -4117,6 +4185,7 @@ struct rpc_procinfo nfs4_procedures[] = { PROC(SERVER_CAPS, enc_server_caps, dec_server_caps), PROC(DELEGRETURN, enc_delegreturn, dec_delegreturn), PROC(GETACL, enc_getacl, dec_getacl), + PROC(SETACL, enc_setacl, dec_setacl), }; struct rpc_version nfs_version4 = { diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index 6ee7e2585af5..5bb5b2fd7ba2 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -383,6 +383,7 @@ enum { NFSPROC4_CLNT_SERVER_CAPS, NFSPROC4_CLNT_DELEGRETURN, NFSPROC4_CLNT_GETACL, + NFSPROC4_CLNT_SETACL, }; #endif diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 9f5e1d407c7b..46b206b460c0 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -326,6 +326,13 @@ struct nfs_setattrargs { const u32 * bitmask; }; +struct nfs_setaclargs { + struct nfs_fh * fh; + size_t acl_len; + unsigned int acl_pgbase; + struct page ** acl_pages; +}; + struct nfs_getaclargs { struct nfs_fh * fh; size_t acl_len; -- cgit v1.2.3 From e50a1c2e1f816c81eed6a589019052cb44189267 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 22 Jun 2005 17:16:23 +0000 Subject: [PATCH] NFSv4: client-side caching NFSv4 ACLs Add nfs4_acl field to the nfs_inode, and use it to cache acls. Only cache acls of size up to a page. Also prepare for up to a page of acl data even when the user doesn't pass in a buffer, as when they want to get the acl length to decide what size buffer to allocate. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 7 ++- fs/nfs/nfs4proc.c | 129 +++++++++++++++++++++++++++++++++++++++++++++---- include/linux/nfs_fs.h | 2 +- 3 files changed, 124 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index c45bd52cc1d7..350c48c12639 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -142,10 +142,6 @@ nfs_delete_inode(struct inode * inode) clear_inode(inode); } -/* - * For the moment, the only task for the NFS clear_inode method is to - * release the mmap credential - */ static void nfs_clear_inode(struct inode *inode) { @@ -1923,6 +1919,9 @@ static struct inode *nfs_alloc_inode(struct super_block *sb) if (!nfsi) return NULL; nfsi->flags = 0; +#ifdef CONFIG_NFS_V4 + nfsi->nfs4_acl = NULL; +#endif /* CONFIG_NFS_V4 */ return &nfsi->vfs_inode; } diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index d969dd13e7db..128d01cfea19 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2188,9 +2188,75 @@ static void buf_to_pages(const void *buf, size_t buflen, } } -static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen) +struct nfs4_cached_acl { + int cached; + size_t len; + char data[]; +}; + +static void nfs4_set_cached_acl(struct inode *inode, struct nfs4_cached_acl *acl) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + spin_lock(&inode->i_lock); + kfree(nfsi->nfs4_acl); + nfsi->nfs4_acl = acl; + spin_unlock(&inode->i_lock); +} + +static void nfs4_zap_acl_attr(struct inode *inode) +{ + nfs4_set_cached_acl(inode, NULL); +} + +static inline ssize_t nfs4_read_cached_acl(struct inode *inode, char *buf, size_t buflen) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs4_cached_acl *acl; + int ret = -ENOENT; + + spin_lock(&inode->i_lock); + acl = nfsi->nfs4_acl; + if (acl == NULL) + goto out; + if (buf == NULL) /* user is just asking for length */ + goto out_len; + if (acl->cached == 0) + goto out; + ret = -ERANGE; /* see getxattr(2) man page */ + if (acl->len > buflen) + goto out; + memcpy(buf, acl->data, acl->len); +out_len: + ret = acl->len; +out: + spin_unlock(&inode->i_lock); + return ret; +} + +static void nfs4_write_cached_acl(struct inode *inode, const char *buf, size_t acl_len) +{ + struct nfs4_cached_acl *acl; + + if (buf && acl_len <= PAGE_SIZE) { + acl = kmalloc(sizeof(*acl) + acl_len, GFP_KERNEL); + if (acl == NULL) + goto out; + acl->cached = 1; + memcpy(acl->data, buf, acl_len); + } else { + acl = kmalloc(sizeof(*acl), GFP_KERNEL); + if (acl == NULL) + goto out; + acl->cached = 0; + } + acl->len = acl_len; +out: + nfs4_set_cached_acl(inode, acl); +} + +static inline ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) { - struct nfs_server *server = NFS_SERVER(inode); struct page *pages[NFS4ACL_MAXPAGES]; struct nfs_getaclargs args = { .fh = NFS_FH(inode), @@ -2198,24 +2264,66 @@ static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen) .acl_len = buflen, }; size_t resp_len = buflen; + void *resp_buf; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETACL], .rpc_argp = &args, .rpc_resp = &resp_len, }; + struct page *localpage = NULL; int ret; - if (!nfs4_server_supports_acls(server)) - return -EOPNOTSUPP; - buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase); + if (buflen < PAGE_SIZE) { + /* As long as we're doing a round trip to the server anyway, + * let's be prepared for a page of acl data. */ + localpage = alloc_page(GFP_KERNEL); + resp_buf = page_address(localpage); + if (localpage == NULL) + return -ENOMEM; + args.acl_pages[0] = localpage; + args.acl_pgbase = 0; + args.acl_len = PAGE_SIZE; + } else { + resp_buf = buf; + buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase); + } ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); - if (buflen && resp_len > buflen) - return -ERANGE; - if (ret == 0) - ret = resp_len; + if (ret) + goto out_free; + if (resp_len > args.acl_len) + nfs4_write_cached_acl(inode, NULL, resp_len); + else + nfs4_write_cached_acl(inode, resp_buf, resp_len); + if (buf) { + ret = -ERANGE; + if (resp_len > buflen) + goto out_free; + if (localpage) + memcpy(buf, resp_buf, resp_len); + } + ret = resp_len; +out_free: + if (localpage) + __free_page(localpage); return ret; } +static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen) +{ + struct nfs_server *server = NFS_SERVER(inode); + int ret; + + if (!nfs4_server_supports_acls(server)) + return -EOPNOTSUPP; + ret = nfs_revalidate_inode(server, inode); + if (ret < 0) + return ret; + ret = nfs4_read_cached_acl(inode, buf, buflen); + if (ret != -ENOENT) + return ret; + return nfs4_get_acl_uncached(inode, buf, buflen); +} + static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen) { struct nfs_server *server = NFS_SERVER(inode); @@ -2236,6 +2344,8 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen return -EOPNOTSUPP; buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase); ret = rpc_call_sync(NFS_SERVER(inode)->client, &msg, 0); + if (ret == 0) + nfs4_write_cached_acl(inode, buf, buflen); return ret; } @@ -2907,6 +3017,7 @@ struct nfs_rpc_ops nfs_v4_clientops = { .file_open = nfs4_proc_file_open, .file_release = nfs4_proc_file_release, .lock = nfs4_proc_lock, + .clear_acl_cache = nfs4_zap_acl_attr, }; /* diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 140bdf489f71..d2b5d7e0e85a 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -169,13 +169,13 @@ struct nfs_inode { wait_queue_head_t nfs_i_wait; #ifdef CONFIG_NFS_V4 + struct nfs4_cached_acl *nfs4_acl; /* NFSv4 state */ struct list_head open_states; struct nfs_delegation *delegation; int delegation_state; struct rw_semaphore rwsem; #endif /* CONFIG_NFS_V4*/ - struct inode vfs_inode; }; -- cgit v1.2.3 From 007e251f2b2760f738c92adc8c80cbae0bed3ce5 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:23 +0000 Subject: [PATCH] RPC: Allow multiple RPC client programs to share the same transport Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 2 ++ net/sunrpc/clnt.c | 40 ++++++++++++++++++++++++++++++++++++++++ net/sunrpc/pmap_clnt.c | 3 +++ net/sunrpc/sunrpc_syms.c | 1 + 4 files changed, 46 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index d25e80f77ff5..ab151bbb66df 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -114,6 +114,8 @@ struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname, struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname, struct rpc_program *info, u32 version, rpc_authflavor_t authflavor); +struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *, + struct rpc_program *, int); struct rpc_clnt *rpc_clone_client(struct rpc_clnt *); int rpc_shutdown_client(struct rpc_clnt *); int rpc_destroy_client(struct rpc_clnt *); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 33f12b84e265..c979fcf88798 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -241,6 +241,8 @@ rpc_clone_client(struct rpc_clnt *clnt) rpc_init_rtt(&new->cl_rtt_default, clnt->cl_xprt->timeout.to_initval); if (new->cl_auth) atomic_inc(&new->cl_auth->au_count); + new->cl_pmap = &new->cl_pmap_default; + rpc_init_wait_queue(&new->cl_pmap_default.pm_bindwait, "bindwait"); return new; out_no_clnt: printk(KERN_INFO "RPC: out of memory in %s\n", __FUNCTION__); @@ -329,6 +331,44 @@ rpc_release_client(struct rpc_clnt *clnt) rpc_destroy_client(clnt); } +/** + * rpc_bind_new_program - bind a new RPC program to an existing client + * @old - old rpc_client + * @program - rpc program to set + * @vers - rpc program version + * + * Clones the rpc client and sets up a new RPC program. This is mainly + * of use for enabling different RPC programs to share the same transport. + * The Sun NFSv2/v3 ACL protocol can do this. + */ +struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *old, + struct rpc_program *program, + int vers) +{ + struct rpc_clnt *clnt; + struct rpc_version *version; + int err; + + BUG_ON(vers >= program->nrvers || !program->version[vers]); + version = program->version[vers]; + clnt = rpc_clone_client(old); + if (IS_ERR(clnt)) + goto out; + clnt->cl_procinfo = version->procs; + clnt->cl_maxproc = version->nrprocs; + clnt->cl_protname = program->name; + clnt->cl_prog = program->number; + clnt->cl_vers = version->number; + clnt->cl_stats = program->stats; + err = rpc_ping(clnt, RPC_TASK_SOFT|RPC_TASK_NOINTR); + if (err != 0) { + rpc_shutdown_client(clnt); + clnt = ERR_PTR(err); + } +out: + return clnt; +} + /* * Default callback for async RPC calls */ diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c index df4d84c9020d..4e81f2766923 100644 --- a/net/sunrpc/pmap_clnt.c +++ b/net/sunrpc/pmap_clnt.c @@ -53,6 +53,9 @@ rpc_getport(struct rpc_task *task, struct rpc_clnt *clnt) task->tk_pid, clnt->cl_server, map->pm_prog, map->pm_vers, map->pm_prot); + /* Autobind on cloned rpc clients is discouraged */ + BUG_ON(clnt->cl_parent != clnt); + spin_lock(&pmap_lock); if (map->pm_binding) { rpc_sleep_on(&map->pm_bindwait, task, NULL, NULL); diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c index 1b0ff7e0e869..d8673f66acc3 100644 --- a/net/sunrpc/sunrpc_syms.c +++ b/net/sunrpc/sunrpc_syms.c @@ -42,6 +42,7 @@ EXPORT_SYMBOL(rpc_release_task); /* RPC client functions */ EXPORT_SYMBOL(rpc_create_client); EXPORT_SYMBOL(rpc_clone_client); +EXPORT_SYMBOL(rpc_bind_new_program); EXPORT_SYMBOL(rpc_destroy_client); EXPORT_SYMBOL(rpc_shutdown_client); EXPORT_SYMBOL(rpc_release_client); -- cgit v1.2.3 From e053d1ab62c8ef0eff3dd4c95448cad3c6d2fbf4 Mon Sep 17 00:00:00 2001 From: Olaf Kirch Date: Wed, 22 Jun 2005 17:16:24 +0000 Subject: [PATCH] RPC: Lazy RPC receive buffer allocation Signed-off-by: Olaf Kirch Signed-off-by: Andreas Gruenbacher Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 2 +- net/sunrpc/xdr.c | 16 +++++++++++++--- net/sunrpc/xprt.c | 26 ++++++++++++++++++++++---- 3 files changed, 36 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 541dcf838abf..0f5b7a5a7432 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -160,7 +160,7 @@ typedef struct { typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len); -extern void xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, +extern int xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, skb_reader_t *, skb_read_actor_t); struct socket; diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c index f86d1baa6302..65b268d39782 100644 --- a/net/sunrpc/xdr.c +++ b/net/sunrpc/xdr.c @@ -176,7 +176,7 @@ xdr_inline_pages(struct xdr_buf *xdr, unsigned int offset, xdr->buflen += len; } -void +int xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, skb_reader_t *desc, skb_read_actor_t copy_actor) @@ -190,7 +190,7 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, len -= base; ret = copy_actor(desc, (char *)xdr->head[0].iov_base + base, len); if (ret != len || !desc->count) - return; + return 0; base = 0; } else base -= len; @@ -210,6 +210,14 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, do { char *kaddr; + /* ACL likes to be lazy in allocating pages - ACLs + * are small by default but can get huge. */ + if (unlikely(*ppage == NULL)) { + *ppage = alloc_page(GFP_ATOMIC); + if (unlikely(*ppage == NULL)) + return -ENOMEM; + } + len = PAGE_CACHE_SIZE; kaddr = kmap_atomic(*ppage, KM_SKB_SUNRPC_DATA); if (base) { @@ -226,13 +234,15 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, flush_dcache_page(*ppage); kunmap_atomic(kaddr, KM_SKB_SUNRPC_DATA); if (ret != len || !desc->count) - return; + return 0; ppage++; } while ((pglen -= len) != 0); copy_tail: len = xdr->tail[0].iov_len; if (base < len) copy_actor(desc, (char *)xdr->tail[0].iov_base + base, len - base); + + return 0; } diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index c74a6bb94074..a180ed4952d6 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -725,7 +725,8 @@ csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb) goto no_checksum; desc.csum = csum_partial(skb->data, desc.offset, skb->csum); - xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_and_csum_bits); + if (xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_and_csum_bits) < 0) + return -1; if (desc.offset != skb->len) { unsigned int csum2; csum2 = skb_checksum(skb, desc.offset, skb->len - desc.offset, 0); @@ -737,7 +738,8 @@ csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb) return -1; return 0; no_checksum: - xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_bits); + if (xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_bits) < 0) + return -1; if (desc.count) return -1; return 0; @@ -907,6 +909,7 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc) struct rpc_rqst *req; struct xdr_buf *rcvbuf; size_t len; + int r; /* Find and lock the request corresponding to this xid */ spin_lock(&xprt->sock_lock); @@ -927,16 +930,30 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc) len = xprt->tcp_reclen - xprt->tcp_offset; memcpy(&my_desc, desc, sizeof(my_desc)); my_desc.count = len; - xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied, + r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied, &my_desc, tcp_copy_data); desc->count -= len; desc->offset += len; } else - xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied, + r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied, desc, tcp_copy_data); xprt->tcp_copied += len; xprt->tcp_offset += len; + if (r < 0) { + /* Error when copying to the receive buffer, + * usually because we weren't able to allocate + * additional buffer pages. All we can do now + * is turn off XPRT_COPY_DATA, so the request + * will not receive any additional updates, + * and time out. + * Any remaining data from this record will + * be discarded. + */ + xprt->tcp_flags &= ~XPRT_COPY_DATA; + goto out; + } + if (xprt->tcp_copied == req->rq_private_buf.buflen) xprt->tcp_flags &= ~XPRT_COPY_DATA; else if (xprt->tcp_offset == xprt->tcp_reclen) { @@ -949,6 +966,7 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc) req->rq_task->tk_pid); xprt_complete_rqst(xprt, req, xprt->tcp_copied); } +out: spin_unlock(&xprt->sock_lock); tcp_check_recm(xprt); } -- cgit v1.2.3 From 7e06b53d796a3740307b54aa2799077f8a0c84e7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:24 +0000 Subject: [PATCH] RPC: fix accounting bug in the case of a truncated RPC message Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 2 +- net/sunrpc/xdr.c | 22 ++++++++++++++-------- net/sunrpc/xprt.c | 35 +++++++++++++++++++++++++++-------- 3 files changed, 42 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 0f5b7a5a7432..5d1eed2b58a1 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -160,7 +160,7 @@ typedef struct { typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len); -extern int xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, +extern ssize_t xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, skb_reader_t *, skb_read_actor_t); struct socket; diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c index 65b268d39782..b3ac3f72bf9c 100644 --- a/net/sunrpc/xdr.c +++ b/net/sunrpc/xdr.c @@ -176,21 +176,23 @@ xdr_inline_pages(struct xdr_buf *xdr, unsigned int offset, xdr->buflen += len; } -int +ssize_t xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, skb_reader_t *desc, skb_read_actor_t copy_actor) { struct page **ppage = xdr->pages; unsigned int len, pglen = xdr->page_len; + ssize_t copied = 0; int ret; len = xdr->head[0].iov_len; if (base < len) { len -= base; ret = copy_actor(desc, (char *)xdr->head[0].iov_base + base, len); + copied += ret; if (ret != len || !desc->count) - return 0; + goto out; base = 0; } else base -= len; @@ -214,8 +216,11 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, * are small by default but can get huge. */ if (unlikely(*ppage == NULL)) { *ppage = alloc_page(GFP_ATOMIC); - if (unlikely(*ppage == NULL)) - return -ENOMEM; + if (unlikely(*ppage == NULL)) { + if (copied == 0) + copied = -ENOMEM; + goto out; + } } len = PAGE_CACHE_SIZE; @@ -233,16 +238,17 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, } flush_dcache_page(*ppage); kunmap_atomic(kaddr, KM_SKB_SUNRPC_DATA); + copied += ret; if (ret != len || !desc->count) - return 0; + goto out; ppage++; } while ((pglen -= len) != 0); copy_tail: len = xdr->tail[0].iov_len; if (base < len) - copy_actor(desc, (char *)xdr->tail[0].iov_base + base, len - base); - - return 0; + copied += copy_actor(desc, (char *)xdr->tail[0].iov_base + base, len - base); +out: + return copied; } diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index a180ed4952d6..ef941e7de8bf 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -823,10 +823,15 @@ tcp_copy_data(skb_reader_t *desc, void *p, size_t len) { if (len > desc->count) len = desc->count; - if (skb_copy_bits(desc->skb, desc->offset, p, len)) + if (skb_copy_bits(desc->skb, desc->offset, p, len)) { + dprintk("RPC: failed to copy %zu bytes from skb. %zu bytes remain\n", + len, desc->count); return 0; + } desc->offset += len; desc->count -= len; + dprintk("RPC: copied %zu bytes from skb. %zu bytes remain\n", + len, desc->count); return len; } @@ -865,6 +870,8 @@ tcp_read_fraghdr(struct rpc_xprt *xprt, skb_reader_t *desc) static void tcp_check_recm(struct rpc_xprt *xprt) { + dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u, tcp_flags = %lx\n", + xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen, xprt->tcp_flags); if (xprt->tcp_offset == xprt->tcp_reclen) { xprt->tcp_flags |= XPRT_COPY_RECM; xprt->tcp_offset = 0; @@ -909,7 +916,7 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc) struct rpc_rqst *req; struct xdr_buf *rcvbuf; size_t len; - int r; + ssize_t r; /* Find and lock the request corresponding to this xid */ spin_lock(&xprt->sock_lock); @@ -932,15 +939,17 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc) my_desc.count = len; r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied, &my_desc, tcp_copy_data); - desc->count -= len; - desc->offset += len; + desc->count -= r; + desc->offset += r; } else r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied, desc, tcp_copy_data); - xprt->tcp_copied += len; - xprt->tcp_offset += len; - if (r < 0) { + if (r > 0) { + xprt->tcp_copied += r; + xprt->tcp_offset += r; + } + if (r != len) { /* Error when copying to the receive buffer, * usually because we weren't able to allocate * additional buffer pages. All we can do now @@ -951,9 +960,18 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc) * be discarded. */ xprt->tcp_flags &= ~XPRT_COPY_DATA; + dprintk("RPC: XID %08x truncated request\n", + ntohl(xprt->tcp_xid)); + dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u\n", + xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen); goto out; } + dprintk("RPC: XID %08x read %u bytes\n", + ntohl(xprt->tcp_xid), r); + dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u\n", + xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen); + if (xprt->tcp_copied == req->rq_private_buf.buflen) xprt->tcp_flags &= ~XPRT_COPY_DATA; else if (xprt->tcp_offset == xprt->tcp_reclen) { @@ -961,12 +979,12 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc) xprt->tcp_flags &= ~XPRT_COPY_DATA; } +out: if (!(xprt->tcp_flags & XPRT_COPY_DATA)) { dprintk("RPC: %4d received reply complete\n", req->rq_task->tk_pid); xprt_complete_rqst(xprt, req, xprt->tcp_copied); } -out: spin_unlock(&xprt->sock_lock); tcp_check_recm(xprt); } @@ -985,6 +1003,7 @@ tcp_read_discard(struct rpc_xprt *xprt, skb_reader_t *desc) desc->count -= len; desc->offset += len; xprt->tcp_offset += len; + dprintk("RPC: discarded %u bytes\n", len); tcp_check_recm(xprt); } -- cgit v1.2.3 From bd8100e7eda87507649c6ba4cb32173b34e49986 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:24 +0000 Subject: [PATCH] RPC: Encode and decode arbitrary XDR arrays Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 19 +++- net/sunrpc/sunrpc_syms.c | 4 + net/sunrpc/xdr.c | 256 ++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 275 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 5d1eed2b58a1..34ec3e8d99b3 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -146,7 +146,8 @@ extern void xdr_shift_buf(struct xdr_buf *, size_t); extern void xdr_buf_from_iov(struct kvec *, struct xdr_buf *); extern int xdr_buf_subsegment(struct xdr_buf *, struct xdr_buf *, int, int); extern int xdr_buf_read_netobj(struct xdr_buf *, struct xdr_netobj *, int); -extern int read_bytes_from_xdr_buf(struct xdr_buf *buf, int base, void *obj, int len); +extern int read_bytes_from_xdr_buf(struct xdr_buf *, int, void *, int); +extern int write_bytes_to_xdr_buf(struct xdr_buf *, int, void *, int); /* * Helper structure for copying from an sk_buff. @@ -168,6 +169,22 @@ struct sockaddr; extern int xdr_sendpages(struct socket *, struct sockaddr *, int, struct xdr_buf *, unsigned int, int); +extern int xdr_encode_word(struct xdr_buf *, int, u32); +extern int xdr_decode_word(struct xdr_buf *, int, u32 *); + +struct xdr_array2_desc; +typedef int (*xdr_xcode_elem_t)(struct xdr_array2_desc *desc, void *elem); +struct xdr_array2_desc { + unsigned int elem_size; + unsigned int array_len; + xdr_xcode_elem_t xcode; +}; + +extern int xdr_decode_array2(struct xdr_buf *buf, unsigned int base, + struct xdr_array2_desc *desc); +extern int xdr_encode_array2(struct xdr_buf *buf, unsigned int base, + struct xdr_array2_desc *desc); + /* * Provide some simple tools for XDR buffer overflow-checking etc. */ diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c index d8673f66acc3..32e8acbc60fe 100644 --- a/net/sunrpc/sunrpc_syms.c +++ b/net/sunrpc/sunrpc_syms.c @@ -129,6 +129,10 @@ EXPORT_SYMBOL(xdr_encode_netobj); EXPORT_SYMBOL(xdr_encode_pages); EXPORT_SYMBOL(xdr_inline_pages); EXPORT_SYMBOL(xdr_shift_buf); +EXPORT_SYMBOL(xdr_encode_word); +EXPORT_SYMBOL(xdr_decode_word); +EXPORT_SYMBOL(xdr_encode_array2); +EXPORT_SYMBOL(xdr_decode_array2); EXPORT_SYMBOL(xdr_buf_from_iov); EXPORT_SYMBOL(xdr_buf_subsegment); EXPORT_SYMBOL(xdr_buf_read_netobj); diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c index b3ac3f72bf9c..8a4d9c106af1 100644 --- a/net/sunrpc/xdr.c +++ b/net/sunrpc/xdr.c @@ -887,8 +887,34 @@ out: return status; } -static int -read_u32_from_xdr_buf(struct xdr_buf *buf, int base, u32 *obj) +/* obj is assumed to point to allocated memory of size at least len: */ +int +write_bytes_to_xdr_buf(struct xdr_buf *buf, int base, void *obj, int len) +{ + struct xdr_buf subbuf; + int this_len; + int status; + + status = xdr_buf_subsegment(buf, &subbuf, base, len); + if (status) + goto out; + this_len = min(len, (int)subbuf.head[0].iov_len); + memcpy(subbuf.head[0].iov_base, obj, this_len); + len -= this_len; + obj += this_len; + this_len = min(len, (int)subbuf.page_len); + if (this_len) + _copy_to_pages(subbuf.pages, subbuf.page_base, obj, this_len); + len -= this_len; + obj += this_len; + this_len = min(len, (int)subbuf.tail[0].iov_len); + memcpy(subbuf.tail[0].iov_base, obj, this_len); +out: + return status; +} + +int +xdr_decode_word(struct xdr_buf *buf, int base, u32 *obj) { u32 raw; int status; @@ -900,6 +926,14 @@ read_u32_from_xdr_buf(struct xdr_buf *buf, int base, u32 *obj) return 0; } +int +xdr_encode_word(struct xdr_buf *buf, int base, u32 obj) +{ + u32 raw = htonl(obj); + + return write_bytes_to_xdr_buf(buf, base, &raw, sizeof(obj)); +} + /* If the netobj starting offset bytes from the start of xdr_buf is contained * entirely in the head or the tail, set object to point to it; otherwise * try to find space for it at the end of the tail, copy it there, and @@ -910,7 +944,7 @@ xdr_buf_read_netobj(struct xdr_buf *buf, struct xdr_netobj *obj, int offset) u32 tail_offset = buf->head[0].iov_len + buf->page_len; u32 obj_end_offset; - if (read_u32_from_xdr_buf(buf, offset, &obj->len)) + if (xdr_decode_word(buf, offset, &obj->len)) goto out; obj_end_offset = offset + 4 + obj->len; @@ -943,3 +977,219 @@ xdr_buf_read_netobj(struct xdr_buf *buf, struct xdr_netobj *obj, int offset) out: return -1; } + +/* Returns 0 on success, or else a negative error code. */ +static int +xdr_xcode_array2(struct xdr_buf *buf, unsigned int base, + struct xdr_array2_desc *desc, int encode) +{ + char *elem = NULL, *c; + unsigned int copied = 0, todo, avail_here; + struct page **ppages = NULL; + int err; + + if (encode) { + if (xdr_encode_word(buf, base, desc->array_len) != 0) + return -EINVAL; + } else { + if (xdr_decode_word(buf, base, &desc->array_len) != 0 || + (unsigned long) base + 4 + desc->array_len * + desc->elem_size > buf->len) + return -EINVAL; + } + base += 4; + + if (!desc->xcode) + return 0; + + todo = desc->array_len * desc->elem_size; + + /* process head */ + if (todo && base < buf->head->iov_len) { + c = buf->head->iov_base + base; + avail_here = min_t(unsigned int, todo, + buf->head->iov_len - base); + todo -= avail_here; + + while (avail_here >= desc->elem_size) { + err = desc->xcode(desc, c); + if (err) + goto out; + c += desc->elem_size; + avail_here -= desc->elem_size; + } + if (avail_here) { + if (!elem) { + elem = kmalloc(desc->elem_size, GFP_KERNEL); + err = -ENOMEM; + if (!elem) + goto out; + } + if (encode) { + err = desc->xcode(desc, elem); + if (err) + goto out; + memcpy(c, elem, avail_here); + } else + memcpy(elem, c, avail_here); + copied = avail_here; + } + base = buf->head->iov_len; /* align to start of pages */ + } + + /* process pages array */ + base -= buf->head->iov_len; + if (todo && base < buf->page_len) { + unsigned int avail_page; + + avail_here = min(todo, buf->page_len - base); + todo -= avail_here; + + base += buf->page_base; + ppages = buf->pages + (base >> PAGE_CACHE_SHIFT); + base &= ~PAGE_CACHE_MASK; + avail_page = min_t(unsigned int, PAGE_CACHE_SIZE - base, + avail_here); + c = kmap(*ppages) + base; + + while (avail_here) { + avail_here -= avail_page; + if (copied || avail_page < desc->elem_size) { + unsigned int l = min(avail_page, + desc->elem_size - copied); + if (!elem) { + elem = kmalloc(desc->elem_size, + GFP_KERNEL); + err = -ENOMEM; + if (!elem) + goto out; + } + if (encode) { + if (!copied) { + err = desc->xcode(desc, elem); + if (err) + goto out; + } + memcpy(c, elem + copied, l); + copied += l; + if (copied == desc->elem_size) + copied = 0; + } else { + memcpy(elem + copied, c, l); + copied += l; + if (copied == desc->elem_size) { + err = desc->xcode(desc, elem); + if (err) + goto out; + copied = 0; + } + } + avail_page -= l; + c += l; + } + while (avail_page >= desc->elem_size) { + err = desc->xcode(desc, c); + if (err) + goto out; + c += desc->elem_size; + avail_page -= desc->elem_size; + } + if (avail_page) { + unsigned int l = min(avail_page, + desc->elem_size - copied); + if (!elem) { + elem = kmalloc(desc->elem_size, + GFP_KERNEL); + err = -ENOMEM; + if (!elem) + goto out; + } + if (encode) { + if (!copied) { + err = desc->xcode(desc, elem); + if (err) + goto out; + } + memcpy(c, elem + copied, l); + copied += l; + if (copied == desc->elem_size) + copied = 0; + } else { + memcpy(elem + copied, c, l); + copied += l; + if (copied == desc->elem_size) { + err = desc->xcode(desc, elem); + if (err) + goto out; + copied = 0; + } + } + } + if (avail_here) { + kunmap(*ppages); + ppages++; + c = kmap(*ppages); + } + + avail_page = min(avail_here, + (unsigned int) PAGE_CACHE_SIZE); + } + base = buf->page_len; /* align to start of tail */ + } + + /* process tail */ + base -= buf->page_len; + if (todo) { + c = buf->tail->iov_base + base; + if (copied) { + unsigned int l = desc->elem_size - copied; + + if (encode) + memcpy(c, elem + copied, l); + else { + memcpy(elem + copied, c, l); + err = desc->xcode(desc, elem); + if (err) + goto out; + } + todo -= l; + c += l; + } + while (todo) { + err = desc->xcode(desc, c); + if (err) + goto out; + c += desc->elem_size; + todo -= desc->elem_size; + } + } + err = 0; + +out: + if (elem) + kfree(elem); + if (ppages) + kunmap(*ppages); + return err; +} + +int +xdr_decode_array2(struct xdr_buf *buf, unsigned int base, + struct xdr_array2_desc *desc) +{ + if (base >= buf->len) + return -EINVAL; + + return xdr_xcode_array2(buf, base, desc, 0); +} + +int +xdr_encode_array2(struct xdr_buf *buf, unsigned int base, + struct xdr_array2_desc *desc) +{ + if ((unsigned long) base + 4 + desc->array_len * desc->elem_size > + buf->head->iov_len + buf->page_len + buf->tail->iov_len) + return -EINVAL; + + return xdr_xcode_array2(buf, base, desc, 1); +} -- cgit v1.2.3 From 9ba02638e4be28dd4ff724202a640264427c62d1 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:24 +0000 Subject: [PATCH] RPC: Allow the sunrpc server to multiplex serveral programs on a single port The NFS and NFSACL programs run on the same RPC transport. This patch adds support for this by converting svc_program into a chained list of programs (server-side). Signed-off-by: Andreas Gruenbacher Signed-off-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- include/linux/sunrpc/svc.h | 3 ++- net/sunrpc/svc.c | 35 ++++++++++++++++++----------------- 2 files changed, 20 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 37003970cf2e..facb94488bb1 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -240,9 +240,10 @@ struct svc_deferred_req { }; /* - * RPC program + * List of RPC programs on the same transport endpoint */ struct svc_program { + struct svc_program * pg_next; /* other programs (same xprt) */ u32 pg_prog; /* program number */ unsigned int pg_lovers; /* lowest version */ unsigned int pg_hivers; /* lowest version */ diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index a02d424a7409..e9bd91265f70 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -35,20 +35,24 @@ svc_create(struct svc_program *prog, unsigned int bufsize) if (!(serv = (struct svc_serv *) kmalloc(sizeof(*serv), GFP_KERNEL))) return NULL; memset(serv, 0, sizeof(*serv)); + serv->sv_name = prog->pg_name; serv->sv_program = prog; serv->sv_nrthreads = 1; serv->sv_stats = prog->pg_stats; serv->sv_bufsz = bufsize? bufsize : 4096; - prog->pg_lovers = prog->pg_nvers-1; xdrsize = 0; - for (vers=0; verspg_nvers ; vers++) - if (prog->pg_vers[vers]) { - prog->pg_hivers = vers; - if (prog->pg_lovers > vers) - prog->pg_lovers = vers; - if (prog->pg_vers[vers]->vs_xdrsize > xdrsize) - xdrsize = prog->pg_vers[vers]->vs_xdrsize; - } + while (prog) { + prog->pg_lovers = prog->pg_nvers-1; + for (vers=0; verspg_nvers ; vers++) + if (prog->pg_vers[vers]) { + prog->pg_hivers = vers; + if (prog->pg_lovers > vers) + prog->pg_lovers = vers; + if (prog->pg_vers[vers]->vs_xdrsize > xdrsize) + xdrsize = prog->pg_vers[vers]->vs_xdrsize; + } + prog = prog->pg_next; + } serv->sv_xdrsize = xdrsize; INIT_LIST_HEAD(&serv->sv_threads); INIT_LIST_HEAD(&serv->sv_sockets); @@ -56,8 +60,6 @@ svc_create(struct svc_program *prog, unsigned int bufsize) INIT_LIST_HEAD(&serv->sv_permsocks); spin_lock_init(&serv->sv_lock); - serv->sv_name = prog->pg_name; - /* Remove any stale portmap registrations */ svc_register(serv, 0, 0); @@ -339,7 +341,10 @@ svc_process(struct svc_serv *serv, struct svc_rqst *rqstp) goto sendit; } - if (prog != progp->pg_prog) + for (progp = serv->sv_program; progp; progp = progp->pg_next) + if (prog == progp->pg_prog) + break; + if (progp == NULL) goto err_bad_prog; if (vers >= progp->pg_nvers || @@ -452,11 +457,7 @@ err_bad_auth: goto sendit; err_bad_prog: -#ifdef RPC_PARANOIA - if (prog != 100227 || progp->pg_prog != 100003) - printk("svc: unknown program %d (me %d)\n", prog, progp->pg_prog); - /* else it is just a Solaris client seeing if ACLs are supported */ -#endif + dprintk("svc: unknown program %d\n", prog); serv->sv_stats->rpcbadfmt++; svc_putu32(resv, rpc_prog_unavail); goto sendit; -- cgit v1.2.3 From a257cdd0e2179630d3201c32ba14d7fcb3c3a055 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:26 +0000 Subject: [PATCH] NFSD: Add server support for NFSv3 ACLs. This adds functions for encoding and decoding POSIX ACLs for the NFSACL protocol extension, and the GETACL and SETACL RPCs. The implementation is compatible with NFSACL in Solaris. Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- fs/Kconfig | 24 ++++ fs/Makefile | 1 + fs/nfs_common/Makefile | 7 + fs/nfs_common/nfsacl.c | 257 ++++++++++++++++++++++++++++++++++ fs/nfsd/Makefile | 2 + fs/nfsd/nfs2acl.c | 336 +++++++++++++++++++++++++++++++++++++++++++++ fs/nfsd/nfs3acl.c | 267 +++++++++++++++++++++++++++++++++++ fs/nfsd/nfs3xdr.c | 13 ++ fs/nfsd/nfssvc.c | 27 ++++ fs/nfsd/nfsxdr.c | 11 ++ fs/nfsd/vfs.c | 107 ++++++++++++++- include/linux/nfsacl.h | 58 ++++++++ include/linux/nfsd/nfsd.h | 16 +++ include/linux/nfsd/xdr.h | 4 + include/linux/nfsd/xdr3.h | 26 ++++ include/linux/sunrpc/svc.h | 11 ++ 16 files changed, 1166 insertions(+), 1 deletion(-) create mode 100644 fs/nfs_common/Makefile create mode 100644 fs/nfs_common/nfsacl.c create mode 100644 fs/nfsd/nfs2acl.c create mode 100644 fs/nfsd/nfs3acl.c create mode 100644 include/linux/nfsacl.h (limited to 'include/linux') diff --git a/fs/Kconfig b/fs/Kconfig index 178e27494b74..d44b04d9b0a9 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1353,6 +1353,7 @@ config NFSD select LOCKD select SUNRPC select EXPORTFS + select NFS_ACL_SUPPORT if NFSD_V3_ACL || NFSD_V2_ACL help If you want your Linux box to act as an NFS *server*, so that other computers on your local network which support NFS can access certain @@ -1376,6 +1377,10 @@ config NFSD To compile the NFS server support as a module, choose M here: the module will be called nfsd. If unsure, say N. +config NFSD_V2_ACL + bool + depends on NFSD + config NFSD_V3 bool "Provide NFSv3 server support" depends on NFSD @@ -1383,6 +1388,16 @@ config NFSD_V3 If you would like to include the NFSv3 server as well as the NFSv2 server, say Y here. If unsure, say Y. +config NFSD_V3_ACL + bool "Provide server support for the NFSv3 ACL protocol extension" + depends on NFSD_V3 + select NFSD_V2_ACL + help + Implement the NFSv3 ACL protocol extension for manipulating POSIX + Access Control Lists on exported file systems. NFS clients should + be compiled with the NFSv3 ACL protocol extension; see the + CONFIG_NFS_V3_ACL option. If unsure, say N. + config NFSD_V4 bool "Provide NFSv4 server support (EXPERIMENTAL)" depends on NFSD_V3 && EXPERIMENTAL @@ -1427,6 +1442,15 @@ config LOCKD_V4 config EXPORTFS tristate +config NFS_ACL_SUPPORT + tristate + select FS_POSIX_ACL + +config NFS_COMMON + bool + depends on NFSD || NFS_FS + default y + config SUNRPC tristate diff --git a/fs/Makefile b/fs/Makefile index 443f2bc56ccf..fc92e59e9faf 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -31,6 +31,7 @@ obj-$(CONFIG_BINFMT_FLAT) += binfmt_flat.o obj-$(CONFIG_FS_MBCACHE) += mbcache.o obj-$(CONFIG_FS_POSIX_ACL) += posix_acl.o xattr_acl.o +obj-$(CONFIG_NFS_COMMON) += nfs_common/ obj-$(CONFIG_QUOTA) += dquot.o obj-$(CONFIG_QFMT_V1) += quota_v1.o diff --git a/fs/nfs_common/Makefile b/fs/nfs_common/Makefile new file mode 100644 index 000000000000..f689ed82af3a --- /dev/null +++ b/fs/nfs_common/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for Linux filesystem routines that are shared by client and server. +# + +obj-$(CONFIG_NFS_ACL_SUPPORT) += nfs_acl.o + +nfs_acl-objs := nfsacl.o diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c new file mode 100644 index 000000000000..18c58c32e326 --- /dev/null +++ b/fs/nfs_common/nfsacl.c @@ -0,0 +1,257 @@ +/* + * fs/nfs_common/nfsacl.c + * + * Copyright (C) 2002-2003 Andreas Gruenbacher + */ + +/* + * The Solaris nfsacl protocol represents some ACLs slightly differently + * than POSIX 1003.1e draft 17 does (and we do): + * + * - Minimal ACLs always have an ACL_MASK entry, so they have + * four instead of three entries. + * - The ACL_MASK entry in such minimal ACLs always has the same + * permissions as the ACL_GROUP_OBJ entry. (In extended ACLs + * the ACL_MASK and ACL_GROUP_OBJ entries may differ.) + * - The identifier fields of the ACL_USER_OBJ and ACL_GROUP_OBJ + * entries contain the identifiers of the owner and owning group. + * (In POSIX ACLs we always set them to ACL_UNDEFINED_ID). + * - ACL entries in the kernel are kept sorted in ascending order + * of (e_tag, e_id). Solaris ACLs are unsorted. + */ + +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); + +EXPORT_SYMBOL(nfsacl_encode); +EXPORT_SYMBOL(nfsacl_decode); + +struct nfsacl_encode_desc { + struct xdr_array2_desc desc; + unsigned int count; + struct posix_acl *acl; + int typeflag; + uid_t uid; + gid_t gid; +}; + +static int +xdr_nfsace_encode(struct xdr_array2_desc *desc, void *elem) +{ + struct nfsacl_encode_desc *nfsacl_desc = + (struct nfsacl_encode_desc *) desc; + u32 *p = (u32 *) elem; + + if (nfsacl_desc->count < nfsacl_desc->acl->a_count) { + struct posix_acl_entry *entry = + &nfsacl_desc->acl->a_entries[nfsacl_desc->count++]; + + *p++ = htonl(entry->e_tag | nfsacl_desc->typeflag); + switch(entry->e_tag) { + case ACL_USER_OBJ: + *p++ = htonl(nfsacl_desc->uid); + break; + case ACL_GROUP_OBJ: + *p++ = htonl(nfsacl_desc->gid); + break; + case ACL_USER: + case ACL_GROUP: + *p++ = htonl(entry->e_id); + break; + default: /* Solaris depends on that! */ + *p++ = 0; + break; + } + *p++ = htonl(entry->e_perm & S_IRWXO); + } else { + const struct posix_acl_entry *pa, *pe; + int group_obj_perm = ACL_READ|ACL_WRITE|ACL_EXECUTE; + + FOREACH_ACL_ENTRY(pa, nfsacl_desc->acl, pe) { + if (pa->e_tag == ACL_GROUP_OBJ) { + group_obj_perm = pa->e_perm & S_IRWXO; + break; + } + } + /* fake up ACL_MASK entry */ + *p++ = htonl(ACL_MASK | nfsacl_desc->typeflag); + *p++ = htonl(0); + *p++ = htonl(group_obj_perm); + } + + return 0; +} + +unsigned int +nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode, + struct posix_acl *acl, int encode_entries, int typeflag) +{ + int entries = (acl && acl->a_count) ? max_t(int, acl->a_count, 4) : 0; + struct nfsacl_encode_desc nfsacl_desc = { + .desc = { + .elem_size = 12, + .array_len = encode_entries ? entries : 0, + .xcode = xdr_nfsace_encode, + }, + .acl = acl, + .typeflag = typeflag, + .uid = inode->i_uid, + .gid = inode->i_gid, + }; + int err; + + if (entries > NFS_ACL_MAX_ENTRIES || + xdr_encode_word(buf, base, entries)) + return -EINVAL; + err = xdr_encode_array2(buf, base + 4, &nfsacl_desc.desc); + if (!err) + err = 8 + nfsacl_desc.desc.elem_size * + nfsacl_desc.desc.array_len; + return err; +} + +struct nfsacl_decode_desc { + struct xdr_array2_desc desc; + unsigned int count; + struct posix_acl *acl; +}; + +static int +xdr_nfsace_decode(struct xdr_array2_desc *desc, void *elem) +{ + struct nfsacl_decode_desc *nfsacl_desc = + (struct nfsacl_decode_desc *) desc; + u32 *p = (u32 *) elem; + struct posix_acl_entry *entry; + + if (!nfsacl_desc->acl) { + if (desc->array_len > NFS_ACL_MAX_ENTRIES) + return -EINVAL; + nfsacl_desc->acl = posix_acl_alloc(desc->array_len, GFP_KERNEL); + if (!nfsacl_desc->acl) + return -ENOMEM; + nfsacl_desc->count = 0; + } + + entry = &nfsacl_desc->acl->a_entries[nfsacl_desc->count++]; + entry->e_tag = ntohl(*p++) & ~NFS_ACL_DEFAULT; + entry->e_id = ntohl(*p++); + entry->e_perm = ntohl(*p++); + + switch(entry->e_tag) { + case ACL_USER_OBJ: + case ACL_USER: + case ACL_GROUP_OBJ: + case ACL_GROUP: + case ACL_OTHER: + if (entry->e_perm & ~S_IRWXO) + return -EINVAL; + break; + case ACL_MASK: + /* Solaris sometimes sets additonal bits in the mask */ + entry->e_perm &= S_IRWXO; + break; + default: + return -EINVAL; + } + + return 0; +} + +static int +cmp_acl_entry(const void *x, const void *y) +{ + const struct posix_acl_entry *a = x, *b = y; + + if (a->e_tag != b->e_tag) + return a->e_tag - b->e_tag; + else if (a->e_id > b->e_id) + return 1; + else if (a->e_id < b->e_id) + return -1; + else + return 0; +} + +/* + * Convert from a Solaris ACL to a POSIX 1003.1e draft 17 ACL. + */ +static int +posix_acl_from_nfsacl(struct posix_acl *acl) +{ + struct posix_acl_entry *pa, *pe, + *group_obj = NULL, *mask = NULL; + + if (!acl) + return 0; + + sort(acl->a_entries, acl->a_count, sizeof(struct posix_acl_entry), + cmp_acl_entry, NULL); + + /* Clear undefined identifier fields and find the ACL_GROUP_OBJ + and ACL_MASK entries. */ + FOREACH_ACL_ENTRY(pa, acl, pe) { + switch(pa->e_tag) { + case ACL_USER_OBJ: + pa->e_id = ACL_UNDEFINED_ID; + break; + case ACL_GROUP_OBJ: + pa->e_id = ACL_UNDEFINED_ID; + group_obj = pa; + break; + case ACL_MASK: + mask = pa; + /* fall through */ + case ACL_OTHER: + pa->e_id = ACL_UNDEFINED_ID; + break; + } + } + if (acl->a_count == 4 && group_obj && mask && + mask->e_perm == group_obj->e_perm) { + /* remove bogus ACL_MASK entry */ + memmove(mask, mask+1, (3 - (mask - acl->a_entries)) * + sizeof(struct posix_acl_entry)); + acl->a_count = 3; + } + return 0; +} + +unsigned int +nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt, + struct posix_acl **pacl) +{ + struct nfsacl_decode_desc nfsacl_desc = { + .desc = { + .elem_size = 12, + .xcode = pacl ? xdr_nfsace_decode : NULL, + }, + }; + u32 entries; + int err; + + if (xdr_decode_word(buf, base, &entries) || + entries > NFS_ACL_MAX_ENTRIES) + return -EINVAL; + err = xdr_decode_array2(buf, base + 4, &nfsacl_desc.desc); + if (err) + return err; + if (pacl) { + if (entries != nfsacl_desc.desc.array_len || + posix_acl_from_nfsacl(nfsacl_desc.acl) != 0) { + posix_acl_release(nfsacl_desc.acl); + return -EINVAL; + } + *pacl = nfsacl_desc.acl; + } + if (aclcnt) + *aclcnt = entries; + return 8 + nfsacl_desc.desc.elem_size * + nfsacl_desc.desc.array_len; +} diff --git a/fs/nfsd/Makefile b/fs/nfsd/Makefile index b8680a247f8b..9f043f44c92f 100644 --- a/fs/nfsd/Makefile +++ b/fs/nfsd/Makefile @@ -6,7 +6,9 @@ obj-$(CONFIG_NFSD) += nfsd.o nfsd-y := nfssvc.o nfsctl.o nfsproc.o nfsfh.o vfs.o \ export.o auth.o lockd.o nfscache.o nfsxdr.o stats.o +nfsd-$(CONFIG_NFSD_V2_ACL) += nfs2acl.o nfsd-$(CONFIG_NFSD_V3) += nfs3proc.o nfs3xdr.o +nfsd-$(CONFIG_NFSD_V3_ACL) += nfs3acl.o nfsd-$(CONFIG_NFSD_V4) += nfs4proc.o nfs4xdr.o nfs4state.o nfs4idmap.o \ nfs4acl.o nfs4callback.o nfsd-objs := $(nfsd-y) diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c new file mode 100644 index 000000000000..7cbf0682b2f0 --- /dev/null +++ b/fs/nfsd/nfs2acl.c @@ -0,0 +1,336 @@ +/* + * linux/fs/nfsd/nfsacl.c + * + * Process version 2 NFSACL requests. + * + * Copyright (C) 2002-2003 Andreas Gruenbacher + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#define NFSDDBG_FACILITY NFSDDBG_PROC +#define RETURN_STATUS(st) { resp->status = (st); return (st); } + +/* + * NULL call. + */ +static int +nfsacld_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) +{ + return nfs_ok; +} + +/* + * Get the Access and/or Default ACL of a file. + */ +static int nfsacld_proc_getacl(struct svc_rqst * rqstp, + struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp) +{ + svc_fh *fh; + struct posix_acl *acl; + int nfserr = 0; + + dprintk("nfsd: GETACL(2acl) %s\n", SVCFH_fmt(&argp->fh)); + + fh = fh_copy(&resp->fh, &argp->fh); + if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP))) + RETURN_STATUS(nfserr_inval); + + if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT)) + RETURN_STATUS(nfserr_inval); + resp->mask = argp->mask; + + if (resp->mask & (NFS_ACL|NFS_ACLCNT)) { + acl = nfsd_get_posix_acl(fh, ACL_TYPE_ACCESS); + if (IS_ERR(acl)) { + int err = PTR_ERR(acl); + + if (err == -ENODATA || err == -EOPNOTSUPP) + acl = NULL; + else { + nfserr = nfserrno(err); + goto fail; + } + } + if (acl == NULL) { + /* Solaris returns the inode's minimum ACL. */ + + struct inode *inode = fh->fh_dentry->d_inode; + acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL); + } + resp->acl_access = acl; + } + if (resp->mask & (NFS_DFACL|NFS_DFACLCNT)) { + /* Check how Solaris handles requests for the Default ACL + of a non-directory! */ + + acl = nfsd_get_posix_acl(fh, ACL_TYPE_DEFAULT); + if (IS_ERR(acl)) { + int err = PTR_ERR(acl); + + if (err == -ENODATA || err == -EOPNOTSUPP) + acl = NULL; + else { + nfserr = nfserrno(err); + goto fail; + } + } + resp->acl_default = acl; + } + + /* resp->acl_{access,default} are released in nfssvc_release_getacl. */ + RETURN_STATUS(0); + +fail: + posix_acl_release(resp->acl_access); + posix_acl_release(resp->acl_default); + RETURN_STATUS(nfserr); +} + +/* + * Set the Access and/or Default ACL of a file. + */ +static int nfsacld_proc_setacl(struct svc_rqst * rqstp, + struct nfsd3_setaclargs *argp, + struct nfsd_attrstat *resp) +{ + svc_fh *fh; + int nfserr = 0; + + dprintk("nfsd: SETACL(2acl) %s\n", SVCFH_fmt(&argp->fh)); + + fh = fh_copy(&resp->fh, &argp->fh); + nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP); + + if (!nfserr) { + nfserr = nfserrno( nfsd_set_posix_acl( + fh, ACL_TYPE_ACCESS, argp->acl_access) ); + } + if (!nfserr) { + nfserr = nfserrno( nfsd_set_posix_acl( + fh, ACL_TYPE_DEFAULT, argp->acl_default) ); + } + + /* argp->acl_{access,default} may have been allocated in + nfssvc_decode_setaclargs. */ + posix_acl_release(argp->acl_access); + posix_acl_release(argp->acl_default); + return nfserr; +} + +/* + * Check file attributes + */ +static int nfsacld_proc_getattr(struct svc_rqst * rqstp, + struct nfsd_fhandle *argp, struct nfsd_attrstat *resp) +{ + dprintk("nfsd: GETATTR %s\n", SVCFH_fmt(&argp->fh)); + + fh_copy(&resp->fh, &argp->fh); + return fh_verify(rqstp, &resp->fh, 0, MAY_NOP); +} + +/* + * Check file access + */ +static int nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, + struct nfsd3_accessres *resp) +{ + int nfserr; + + dprintk("nfsd: ACCESS(2acl) %s 0x%x\n", + SVCFH_fmt(&argp->fh), + argp->access); + + fh_copy(&resp->fh, &argp->fh); + resp->access = argp->access; + nfserr = nfsd_access(rqstp, &resp->fh, &resp->access, NULL); + return nfserr; +} + +/* + * XDR decode functions + */ +static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_getaclargs *argp) +{ + if (!(p = nfs2svc_decode_fh(p, &argp->fh))) + return 0; + argp->mask = ntohl(*p); p++; + + return xdr_argsize_check(rqstp, p); +} + + +static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_setaclargs *argp) +{ + struct kvec *head = rqstp->rq_arg.head; + unsigned int base; + int n; + + if (!(p = nfs2svc_decode_fh(p, &argp->fh))) + return 0; + argp->mask = ntohl(*p++); + if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT) || + !xdr_argsize_check(rqstp, p)) + return 0; + + base = (char *)p - (char *)head->iov_base; + n = nfsacl_decode(&rqstp->rq_arg, base, NULL, + (argp->mask & NFS_ACL) ? + &argp->acl_access : NULL); + if (n > 0) + n = nfsacl_decode(&rqstp->rq_arg, base + n, NULL, + (argp->mask & NFS_DFACL) ? + &argp->acl_default : NULL); + return (n > 0); +} + +static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, u32 *p, + struct nfsd_fhandle *argp) +{ + if (!(p = nfs2svc_decode_fh(p, &argp->fh))) + return 0; + return xdr_argsize_check(rqstp, p); +} + +static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_accessargs *argp) +{ + if (!(p = nfs2svc_decode_fh(p, &argp->fh))) + return 0; + argp->access = ntohl(*p++); + + return xdr_argsize_check(rqstp, p); +} + +/* + * XDR encode functions + */ + +/* GETACL */ +static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_getaclres *resp) +{ + struct dentry *dentry = resp->fh.fh_dentry; + struct inode *inode = dentry->d_inode; + int w = nfsacl_size( + (resp->mask & NFS_ACL) ? resp->acl_access : NULL, + (resp->mask & NFS_DFACL) ? resp->acl_default : NULL); + struct kvec *head = rqstp->rq_res.head; + unsigned int base; + int n; + + if (dentry == NULL || dentry->d_inode == NULL) + return 0; + inode = dentry->d_inode; + + p = nfs2svc_encode_fattr(rqstp, p, &resp->fh); + *p++ = htonl(resp->mask); + if (!xdr_ressize_check(rqstp, p)) + return 0; + base = (char *)p - (char *)head->iov_base; + + rqstp->rq_res.page_len = w; + while (w > 0) { + if (!svc_take_res_page(rqstp)) + return 0; + w -= PAGE_SIZE; + } + + n = nfsacl_encode(&rqstp->rq_res, base, inode, + resp->acl_access, + resp->mask & NFS_ACL, 0); + if (n > 0) + n = nfsacl_encode(&rqstp->rq_res, base + n, inode, + resp->acl_default, + resp->mask & NFS_DFACL, + NFS_ACL_DEFAULT); + if (n <= 0) + return 0; + return 1; +} + +static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, u32 *p, + struct nfsd_attrstat *resp) +{ + p = nfs2svc_encode_fattr(rqstp, p, &resp->fh); + return xdr_ressize_check(rqstp, p); +} + +/* ACCESS */ +static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_accessres *resp) +{ + p = nfs2svc_encode_fattr(rqstp, p, &resp->fh); + *p++ = htonl(resp->access); + return xdr_ressize_check(rqstp, p); +} + +/* + * XDR release functions + */ +static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_getaclres *resp) +{ + fh_put(&resp->fh); + posix_acl_release(resp->acl_access); + posix_acl_release(resp->acl_default); + return 1; +} + +static int nfsaclsvc_release_fhandle(struct svc_rqst *rqstp, u32 *p, + struct nfsd_fhandle *resp) +{ + fh_put(&resp->fh); + return 1; +} + +#define nfsaclsvc_decode_voidargs NULL +#define nfsaclsvc_encode_voidres NULL +#define nfsaclsvc_release_void NULL +#define nfsd3_fhandleargs nfsd_fhandle +#define nfsd3_attrstatres nfsd_attrstat +#define nfsd3_voidres nfsd3_voidargs +struct nfsd3_voidargs { int dummy; }; + +#define PROC(name, argt, rest, relt, cache, respsize) \ + { (svc_procfunc) nfsacld_proc_##name, \ + (kxdrproc_t) nfsaclsvc_decode_##argt##args, \ + (kxdrproc_t) nfsaclsvc_encode_##rest##res, \ + (kxdrproc_t) nfsaclsvc_release_##relt, \ + sizeof(struct nfsd3_##argt##args), \ + sizeof(struct nfsd3_##rest##res), \ + 0, \ + cache, \ + respsize, \ + } + +#define ST 1 /* status*/ +#define AT 21 /* attributes */ +#define pAT (1+AT) /* post attributes - conditional */ +#define ACL (1+NFS_ACL_MAX_ENTRIES*3) /* Access Control List */ + +static struct svc_procedure nfsd_acl_procedures2[] = { + PROC(null, void, void, void, RC_NOCACHE, ST), + PROC(getacl, getacl, getacl, getacl, RC_NOCACHE, ST+1+2*(1+ACL)), + PROC(setacl, setacl, attrstat, fhandle, RC_NOCACHE, ST+AT), + PROC(getattr, fhandle, attrstat, fhandle, RC_NOCACHE, ST+AT), + PROC(access, access, access, fhandle, RC_NOCACHE, ST+AT+1), +}; + +struct svc_version nfsd_acl_version2 = { + .vs_vers = 2, + .vs_nproc = 5, + .vs_proc = nfsd_acl_procedures2, + .vs_dispatch = nfsd_dispatch, + .vs_xdrsize = NFS3_SVC_XDRSIZE, +}; diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c new file mode 100644 index 000000000000..64ba40572fea --- /dev/null +++ b/fs/nfsd/nfs3acl.c @@ -0,0 +1,267 @@ +/* + * linux/fs/nfsd/nfs3acl.c + * + * Process version 3 NFSACL requests. + * + * Copyright (C) 2002-2003 Andreas Gruenbacher + */ + +#include +#include +#include +#include +#include +#include +#include + +#define RETURN_STATUS(st) { resp->status = (st); return (st); } + +/* + * NULL call. + */ +static int +nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) +{ + return nfs_ok; +} + +/* + * Get the Access and/or Default ACL of a file. + */ +static int nfsd3_proc_getacl(struct svc_rqst * rqstp, + struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp) +{ + svc_fh *fh; + struct posix_acl *acl; + int nfserr = 0; + + fh = fh_copy(&resp->fh, &argp->fh); + if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP))) + RETURN_STATUS(nfserr_inval); + + if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT)) + RETURN_STATUS(nfserr_inval); + resp->mask = argp->mask; + + if (resp->mask & (NFS_ACL|NFS_ACLCNT)) { + acl = nfsd_get_posix_acl(fh, ACL_TYPE_ACCESS); + if (IS_ERR(acl)) { + int err = PTR_ERR(acl); + + if (err == -ENODATA || err == -EOPNOTSUPP) + acl = NULL; + else { + nfserr = nfserrno(err); + goto fail; + } + } + if (acl == NULL) { + /* Solaris returns the inode's minimum ACL. */ + + struct inode *inode = fh->fh_dentry->d_inode; + acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL); + } + resp->acl_access = acl; + } + if (resp->mask & (NFS_DFACL|NFS_DFACLCNT)) { + /* Check how Solaris handles requests for the Default ACL + of a non-directory! */ + + acl = nfsd_get_posix_acl(fh, ACL_TYPE_DEFAULT); + if (IS_ERR(acl)) { + int err = PTR_ERR(acl); + + if (err == -ENODATA || err == -EOPNOTSUPP) + acl = NULL; + else { + nfserr = nfserrno(err); + goto fail; + } + } + resp->acl_default = acl; + } + + /* resp->acl_{access,default} are released in nfs3svc_release_getacl. */ + RETURN_STATUS(0); + +fail: + posix_acl_release(resp->acl_access); + posix_acl_release(resp->acl_default); + RETURN_STATUS(nfserr); +} + +/* + * Set the Access and/or Default ACL of a file. + */ +static int nfsd3_proc_setacl(struct svc_rqst * rqstp, + struct nfsd3_setaclargs *argp, + struct nfsd3_attrstat *resp) +{ + svc_fh *fh; + int nfserr = 0; + + fh = fh_copy(&resp->fh, &argp->fh); + nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP); + + if (!nfserr) { + nfserr = nfserrno( nfsd_set_posix_acl( + fh, ACL_TYPE_ACCESS, argp->acl_access) ); + } + if (!nfserr) { + nfserr = nfserrno( nfsd_set_posix_acl( + fh, ACL_TYPE_DEFAULT, argp->acl_default) ); + } + + /* argp->acl_{access,default} may have been allocated in + nfs3svc_decode_setaclargs. */ + posix_acl_release(argp->acl_access); + posix_acl_release(argp->acl_default); + RETURN_STATUS(nfserr); +} + +/* + * XDR decode functions + */ +static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_getaclargs *args) +{ + if (!(p = nfs3svc_decode_fh(p, &args->fh))) + return 0; + args->mask = ntohl(*p); p++; + + return xdr_argsize_check(rqstp, p); +} + + +static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_setaclargs *args) +{ + struct kvec *head = rqstp->rq_arg.head; + unsigned int base; + int n; + + if (!(p = nfs3svc_decode_fh(p, &args->fh))) + return 0; + args->mask = ntohl(*p++); + if (args->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT) || + !xdr_argsize_check(rqstp, p)) + return 0; + + base = (char *)p - (char *)head->iov_base; + n = nfsacl_decode(&rqstp->rq_arg, base, NULL, + (args->mask & NFS_ACL) ? + &args->acl_access : NULL); + if (n > 0) + n = nfsacl_decode(&rqstp->rq_arg, base + n, NULL, + (args->mask & NFS_DFACL) ? + &args->acl_default : NULL); + return (n > 0); +} + +/* + * XDR encode functions + */ + +/* GETACL */ +static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_getaclres *resp) +{ + struct dentry *dentry = resp->fh.fh_dentry; + + p = nfs3svc_encode_post_op_attr(rqstp, p, &resp->fh); + if (resp->status == 0 && dentry && dentry->d_inode) { + struct inode *inode = dentry->d_inode; + int w = nfsacl_size( + (resp->mask & NFS_ACL) ? resp->acl_access : NULL, + (resp->mask & NFS_DFACL) ? resp->acl_default : NULL); + struct kvec *head = rqstp->rq_res.head; + unsigned int base; + int n; + + *p++ = htonl(resp->mask); + if (!xdr_ressize_check(rqstp, p)) + return 0; + base = (char *)p - (char *)head->iov_base; + + rqstp->rq_res.page_len = w; + while (w > 0) { + if (!svc_take_res_page(rqstp)) + return 0; + w -= PAGE_SIZE; + } + + n = nfsacl_encode(&rqstp->rq_res, base, inode, + resp->acl_access, + resp->mask & NFS_ACL, 0); + if (n > 0) + n = nfsacl_encode(&rqstp->rq_res, base + n, inode, + resp->acl_default, + resp->mask & NFS_DFACL, + NFS_ACL_DEFAULT); + if (n <= 0) + return 0; + } else + if (!xdr_ressize_check(rqstp, p)) + return 0; + + return 1; +} + +/* SETACL */ +static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_attrstat *resp) +{ + p = nfs3svc_encode_post_op_attr(rqstp, p, &resp->fh); + + return xdr_ressize_check(rqstp, p); +} + +/* + * XDR release functions + */ +static int nfs3svc_release_getacl(struct svc_rqst *rqstp, u32 *p, + struct nfsd3_getaclres *resp) +{ + fh_put(&resp->fh); + posix_acl_release(resp->acl_access); + posix_acl_release(resp->acl_default); + return 1; +} + +#define nfs3svc_decode_voidargs NULL +#define nfs3svc_release_void NULL +#define nfsd3_setaclres nfsd3_attrstat +#define nfsd3_voidres nfsd3_voidargs +struct nfsd3_voidargs { int dummy; }; + +#define PROC(name, argt, rest, relt, cache, respsize) \ + { (svc_procfunc) nfsd3_proc_##name, \ + (kxdrproc_t) nfs3svc_decode_##argt##args, \ + (kxdrproc_t) nfs3svc_encode_##rest##res, \ + (kxdrproc_t) nfs3svc_release_##relt, \ + sizeof(struct nfsd3_##argt##args), \ + sizeof(struct nfsd3_##rest##res), \ + 0, \ + cache, \ + respsize, \ + } + +#define ST 1 /* status*/ +#define AT 21 /* attributes */ +#define pAT (1+AT) /* post attributes - conditional */ +#define ACL (1+NFS_ACL_MAX_ENTRIES*3) /* Access Control List */ + +static struct svc_procedure nfsd_acl_procedures3[] = { + PROC(null, void, void, void, RC_NOCACHE, ST), + PROC(getacl, getacl, getacl, getacl, RC_NOCACHE, ST+1+2*(1+ACL)), + PROC(setacl, setacl, setacl, fhandle, RC_NOCACHE, ST+pAT), +}; + +struct svc_version nfsd_acl_version3 = { + .vs_vers = 3, + .vs_nproc = 3, + .vs_proc = nfsd_acl_procedures3, + .vs_dispatch = nfsd_dispatch, + .vs_xdrsize = NFS3_SVC_XDRSIZE, +}; + diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 11f806835c5a..e0e134d6baba 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -71,6 +71,12 @@ decode_fh(u32 *p, struct svc_fh *fhp) return p + XDR_QUADLEN(size); } +/* Helper function for NFSv3 ACL code */ +u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp) +{ + return decode_fh(p, fhp); +} + static inline u32 * encode_fh(u32 *p, struct svc_fh *fhp) { @@ -233,6 +239,13 @@ encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) return p; } +/* Helper for NFSv3 ACLs */ +u32 * +nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +{ + return encode_post_op_attr(rqstp, p, fhp); +} + /* * Enocde weak cache consistency data */ diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 02ded7cfbdcf..79b25b19fec8 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -31,6 +31,7 @@ #include #include #include +#include #define NFSDDBG_FACILITY NFSDDBG_SVC @@ -362,6 +363,31 @@ nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) return 1; } +#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) +static struct svc_stat nfsd_acl_svcstats; +static struct svc_version * nfsd_acl_version[] = { + [2] = &nfsd_acl_version2, + [3] = &nfsd_acl_version3, +}; + +#define NFSD_ACL_NRVERS (sizeof(nfsd_acl_version)/sizeof(nfsd_acl_version[0])) +static struct svc_program nfsd_acl_program = { + .pg_prog = NFS_ACL_PROGRAM, + .pg_nvers = NFSD_ACL_NRVERS, + .pg_vers = nfsd_acl_version, + .pg_name = "nfsd", + .pg_stats = &nfsd_acl_svcstats, +}; + +static struct svc_stat nfsd_acl_svcstats = { + .program = &nfsd_acl_program, +}; + +#define nfsd_acl_program_p &nfsd_acl_program +#else +#define nfsd_acl_program_p NULL +#endif /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */ + extern struct svc_version nfsd_version2, nfsd_version3, nfsd_version4; static struct svc_version * nfsd_version[] = { @@ -376,6 +402,7 @@ static struct svc_version * nfsd_version[] = { #define NFSD_NRVERS (sizeof(nfsd_version)/sizeof(nfsd_version[0])) struct svc_program nfsd_program = { + .pg_next = nfsd_acl_program_p, .pg_prog = NFS_PROGRAM, /* program number */ .pg_nvers = NFSD_NRVERS, /* nr of entries in nfsd_version */ .pg_vers = nfsd_version, /* version table */ diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index 948b08287c99..b45999ff33e6 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -49,6 +49,12 @@ decode_fh(u32 *p, struct svc_fh *fhp) return p + (NFS_FHSIZE >> 2); } +/* Helper function for NFSv2 ACL code */ +u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp) +{ + return decode_fh(p, fhp); +} + static inline u32 * encode_fh(u32 *p, struct svc_fh *fhp) { @@ -190,6 +196,11 @@ encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) return p; } +/* Helper function for NFSv2 ACL code */ +u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +{ + return encode_fattr(rqstp, p, fhp); +} /* * XDR decode functions diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index e3e9d217236e..ae3940dc85cc 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -46,8 +46,9 @@ #include #include #include -#ifdef CONFIG_NFSD_V4 +#include #include +#ifdef CONFIG_NFSD_V4 #include #include #include @@ -1857,3 +1858,107 @@ nfsd_racache_init(int cache_size) nfsdstats.ra_size = cache_size; return 0; } + +#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) +struct posix_acl * +nfsd_get_posix_acl(struct svc_fh *fhp, int type) +{ + struct inode *inode = fhp->fh_dentry->d_inode; + char *name; + void *value = NULL; + ssize_t size; + struct posix_acl *acl; + + if (!IS_POSIXACL(inode) || !inode->i_op || !inode->i_op->getxattr) + return ERR_PTR(-EOPNOTSUPP); + switch(type) { + case ACL_TYPE_ACCESS: + name = XATTR_NAME_ACL_ACCESS; + break; + case ACL_TYPE_DEFAULT: + name = XATTR_NAME_ACL_DEFAULT; + break; + default: + return ERR_PTR(-EOPNOTSUPP); + } + + size = inode->i_op->getxattr(fhp->fh_dentry, name, NULL, 0); + + if (size < 0) { + acl = ERR_PTR(size); + goto getout; + } else if (size > 0) { + value = kmalloc(size, GFP_KERNEL); + if (!value) { + acl = ERR_PTR(-ENOMEM); + goto getout; + } + size = inode->i_op->getxattr(fhp->fh_dentry, name, value, size); + if (size < 0) { + acl = ERR_PTR(size); + goto getout; + } + } + acl = posix_acl_from_xattr(value, size); + +getout: + kfree(value); + return acl; +} + +int +nfsd_set_posix_acl(struct svc_fh *fhp, int type, struct posix_acl *acl) +{ + struct inode *inode = fhp->fh_dentry->d_inode; + char *name; + void *value = NULL; + size_t size; + int error; + + if (!IS_POSIXACL(inode) || !inode->i_op || + !inode->i_op->setxattr || !inode->i_op->removexattr) + return -EOPNOTSUPP; + switch(type) { + case ACL_TYPE_ACCESS: + name = XATTR_NAME_ACL_ACCESS; + break; + case ACL_TYPE_DEFAULT: + name = XATTR_NAME_ACL_DEFAULT; + break; + default: + return -EOPNOTSUPP; + } + + if (acl && acl->a_count) { + size = xattr_acl_size(acl->a_count); + value = kmalloc(size, GFP_KERNEL); + if (!value) + return -ENOMEM; + size = posix_acl_to_xattr(acl, value, size); + if (size < 0) { + error = size; + goto getout; + } + } else + size = 0; + + if (!fhp->fh_locked) + fh_lock(fhp); /* unlocking is done automatically */ + if (size) + error = inode->i_op->setxattr(fhp->fh_dentry, name, + value, size, 0); + else { + if (!S_ISDIR(inode->i_mode) && type == ACL_TYPE_DEFAULT) + error = 0; + else { + error = inode->i_op->removexattr(fhp->fh_dentry, name); + if (error == -ENODATA) + error = 0; + } + } + +getout: + kfree(value); + return error; +} +#endif /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */ diff --git a/include/linux/nfsacl.h b/include/linux/nfsacl.h new file mode 100644 index 000000000000..54487a99beb8 --- /dev/null +++ b/include/linux/nfsacl.h @@ -0,0 +1,58 @@ +/* + * File: linux/nfsacl.h + * + * (C) 2003 Andreas Gruenbacher + */ +#ifndef __LINUX_NFSACL_H +#define __LINUX_NFSACL_H + +#define NFS_ACL_PROGRAM 100227 + +#define ACLPROC2_GETACL 1 +#define ACLPROC2_SETACL 2 +#define ACLPROC2_GETATTR 3 +#define ACLPROC2_ACCESS 4 + +#define ACLPROC3_GETACL 1 +#define ACLPROC3_SETACL 2 + + +/* Flags for the getacl/setacl mode */ +#define NFS_ACL 0x0001 +#define NFS_ACLCNT 0x0002 +#define NFS_DFACL 0x0004 +#define NFS_DFACLCNT 0x0008 + +/* Flag for Default ACL entries */ +#define NFS_ACL_DEFAULT 0x1000 + +#ifdef __KERNEL__ + +#include + +/* Maximum number of ACL entries over NFS */ +#define NFS_ACL_MAX_ENTRIES 1024 + +#define NFSACL_MAXWORDS (2*(2+3*NFS_ACL_MAX_ENTRIES)) +#define NFSACL_MAXPAGES ((2*(8+12*NFS_ACL_MAX_ENTRIES) + PAGE_SIZE-1) \ + >> PAGE_SHIFT) + +static inline unsigned int +nfsacl_size(struct posix_acl *acl_access, struct posix_acl *acl_default) +{ + unsigned int w = 16; + w += max(acl_access ? (int)acl_access->a_count : 3, 4) * 12; + if (acl_default) + w += max((int)acl_default->a_count, 4) * 12; + return w; +} + +extern unsigned int +nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode, + struct posix_acl *acl, int encode_entries, int typeflag); +extern unsigned int +nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt, + struct posix_acl **pacl); + +#endif /* __KERNEL__ */ +#endif /* __LINUX_NFSACL_H */ diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 8f85d9a59607..4bf931d5ff56 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -124,6 +125,21 @@ int nfsd_statfs(struct svc_rqst *, struct svc_fh *, int nfsd_notify_change(struct inode *, struct iattr *); int nfsd_permission(struct svc_export *, struct dentry *, int); +#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) +#ifdef CONFIG_NFSD_V2_ACL +extern struct svc_version nfsd_acl_version2; +#else +#define nfsd_acl_version2 NULL +#endif +#ifdef CONFIG_NFSD_V3_ACL +extern struct svc_version nfsd_acl_version3; +#else +#define nfsd_acl_version3 NULL +#endif +struct posix_acl *nfsd_get_posix_acl(struct svc_fh *, int); +int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *); +#endif + /* * NFSv4 State diff --git a/include/linux/nfsd/xdr.h b/include/linux/nfsd/xdr.h index ecccef777dae..130d4f588a37 100644 --- a/include/linux/nfsd/xdr.h +++ b/include/linux/nfsd/xdr.h @@ -169,4 +169,8 @@ int nfssvc_encode_entry(struct readdir_cd *, const char *name, int nfssvc_release_fhandle(struct svc_rqst *, u32 *, struct nfsd_fhandle *); +/* Helper functions for NFSv2 ACL code */ +u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp); +u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp); + #endif /* LINUX_NFSD_H */ diff --git a/include/linux/nfsd/xdr3.h b/include/linux/nfsd/xdr3.h index 0ae9e0ef5f68..21e18ce7ca63 100644 --- a/include/linux/nfsd/xdr3.h +++ b/include/linux/nfsd/xdr3.h @@ -110,6 +110,19 @@ struct nfsd3_commitargs { __u32 count; }; +struct nfsd3_getaclargs { + struct svc_fh fh; + int mask; +}; + +struct posix_acl; +struct nfsd3_setaclargs { + struct svc_fh fh; + int mask; + struct posix_acl *acl_access; + struct posix_acl *acl_default; +}; + struct nfsd3_attrstat { __u32 status; struct svc_fh fh; @@ -209,6 +222,14 @@ struct nfsd3_commitres { struct svc_fh fh; }; +struct nfsd3_getaclres { + __u32 status; + struct svc_fh fh; + int mask; + struct posix_acl *acl_access; + struct posix_acl *acl_default; +}; + /* dummy type for release */ struct nfsd3_fhandle_pair { __u32 dummy; @@ -241,6 +262,7 @@ union nfsd3_xdrstore { struct nfsd3_fsinfores fsinfores; struct nfsd3_pathconfres pathconfres; struct nfsd3_commitres commitres; + struct nfsd3_getaclres getaclres; }; #define NFS3_SVC_XDRSIZE sizeof(union nfsd3_xdrstore) @@ -316,6 +338,10 @@ int nfs3svc_encode_entry(struct readdir_cd *, const char *name, int nfs3svc_encode_entry_plus(struct readdir_cd *, const char *name, int namlen, loff_t offset, ino_t ino, unsigned int); +/* Helper functions for NFSv3 ACL code */ +u32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, + struct svc_fh *fhp); +u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp); #endif /* _LINUX_NFSD_XDR3_H */ diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index facb94488bb1..5af8800e0ce3 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -185,6 +185,17 @@ xdr_ressize_check(struct svc_rqst *rqstp, u32 *p) return vec->iov_len <= PAGE_SIZE; } +static inline struct page * +svc_take_res_page(struct svc_rqst *rqstp) +{ + if (rqstp->rq_arghi <= rqstp->rq_argused) + return NULL; + rqstp->rq_arghi--; + rqstp->rq_respages[rqstp->rq_resused] = + rqstp->rq_argpages[rqstp->rq_arghi]; + return rqstp->rq_respages[rqstp->rq_resused++]; +} + static inline int svc_take_page(struct svc_rqst *rqstp) { if (rqstp->rq_arghi <= rqstp->rq_argused) -- cgit v1.2.3 From b7fa0554cf1ba6d6895cd0a5b02989a26e0bc704 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:27 +0000 Subject: [PATCH] NFS: Add support for NFSv3 ACLs This adds acl support fo nfs clients via the NFSACL protocol extension, by implementing the getxattr, listxattr, setxattr, and removexattr iops for the system.posix_acl_access and system.posix_acl_default attributes. This patch implements a dumb version that uses no caching (and thus adds some overhead). (Another patch in this patchset adds caching as well.) Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- fs/Kconfig | 11 ++ fs/nfs/Makefile | 1 + fs/nfs/dir.c | 21 ++++ fs/nfs/file.c | 12 ++ fs/nfs/inode.c | 36 +++++- fs/nfs/nfs3acl.c | 303 ++++++++++++++++++++++++++++++++++++++++++++++ fs/nfs/nfs3proc.c | 7 +- fs/nfs/nfs3xdr.c | 147 ++++++++++++++++++++++ fs/nfs/nfsroot.c | 9 ++ include/linux/nfs_fs.h | 31 +++++ include/linux/nfs_fs_sb.h | 1 + include/linux/nfs_mount.h | 1 + include/linux/nfs_xdr.h | 27 +++++ 13 files changed, 601 insertions(+), 6 deletions(-) create mode 100644 fs/nfs/nfs3acl.c (limited to 'include/linux') diff --git a/fs/Kconfig b/fs/Kconfig index d44b04d9b0a9..a7c0cc3203cb 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1268,6 +1268,7 @@ config NFS_FS depends on INET select LOCKD select SUNRPC + select NFS_ACL_SUPPORT if NFS_V3_ACL help If you are connected to some other (usually local) Unix computer (using SLIP, PLIP, PPP or Ethernet) and want to mount files residing @@ -1310,6 +1311,16 @@ config NFS_V3 If unsure, say Y. +config NFS_V3_ACL + bool "Provide client support for the NFSv3 ACL protocol extension" + depends on NFS_V3 + help + Implement the NFSv3 ACL protocol extension for manipulating POSIX + Access Control Lists. The server should also be compiled with + the NFSv3 ACL protocol extension; see the CONFIG_NFSD_V3_ACL option. + + If unsure, say N. + config NFS_V4 bool "Provide NFSv4 client support (EXPERIMENTAL)" depends on NFS_FS && EXPERIMENTAL diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile index b4baa031edf4..8b3bb715d177 100644 --- a/fs/nfs/Makefile +++ b/fs/nfs/Makefile @@ -8,6 +8,7 @@ nfs-y := dir.o file.o inode.o nfs2xdr.o pagelist.o \ proc.o read.o symlink.o unlink.o write.o nfs-$(CONFIG_ROOT_NFS) += nfsroot.o mount_clnt.o nfs-$(CONFIG_NFS_V3) += nfs3proc.o nfs3xdr.o +nfs-$(CONFIG_NFS_V3_ACL) += nfs3acl.o nfs-$(CONFIG_NFS_V4) += nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o \ delegation.o idmap.o \ callback.o callback_xdr.o callback_proc.o diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 5720537bffdd..2c6a95945684 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -75,6 +75,27 @@ struct inode_operations nfs_dir_inode_operations = { .setattr = nfs_setattr, }; +#ifdef CONFIG_NFS_V3 +struct inode_operations nfs3_dir_inode_operations = { + .create = nfs_create, + .lookup = nfs_lookup, + .link = nfs_link, + .unlink = nfs_unlink, + .symlink = nfs_symlink, + .mkdir = nfs_mkdir, + .rmdir = nfs_rmdir, + .mknod = nfs_mknod, + .rename = nfs_rename, + .permission = nfs_permission, + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .listxattr = nfs3_listxattr, + .getxattr = nfs3_getxattr, + .setxattr = nfs3_setxattr, + .removexattr = nfs3_removexattr, +}; +#endif /* CONFIG_NFS_V3 */ + #ifdef CONFIG_NFS_V4 static struct dentry *nfs_atomic_lookup(struct inode *, struct dentry *, struct nameidata *); diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 55c907592490..a606708264ed 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -71,6 +71,18 @@ struct inode_operations nfs_file_inode_operations = { .setattr = nfs_setattr, }; +#ifdef CONFIG_NFS_V3 +struct inode_operations nfs3_file_inode_operations = { + .permission = nfs_permission, + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .listxattr = nfs3_listxattr, + .getxattr = nfs3_getxattr, + .setxattr = nfs3_setxattr, + .removexattr = nfs3_removexattr, +}; +#endif /* CONFIG_NFS_v3 */ + /* Hack for future NFS swap support */ #ifndef IS_SWAPFILE # define IS_SWAPFILE(inode) (0) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 97b3fe7ece63..440b9cbb6f81 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -108,6 +108,21 @@ static struct rpc_program nfs_program = { .pipe_dir_name = "/nfs", }; +#ifdef CONFIG_NFS_V3_ACL +static struct rpc_stat nfsacl_rpcstat = { &nfsacl_program }; +static struct rpc_version * nfsacl_version[] = { + [3] = &nfsacl_version3, +}; + +struct rpc_program nfsacl_program = { + .name = "nfsacl", + .number = NFS_ACL_PROGRAM, + .nrvers = sizeof(nfsacl_version) / sizeof(nfsacl_version[0]), + .version = nfsacl_version, + .stats = &nfsacl_rpcstat, +}; +#endif /* CONFIG_NFS_V3_ACL */ + static inline unsigned long nfs_fattr_to_ino_t(struct nfs_fattr *fattr) { @@ -165,6 +180,9 @@ nfs_umount_begin(struct super_block *sb) /* -EIO all pending I/O */ if (!IS_ERR(rpc)) rpc_killall_tasks(rpc); + rpc = NFS_SB(sb)->client_acl; + if (!IS_ERR(rpc)) + rpc_killall_tasks(rpc); } @@ -461,8 +479,17 @@ nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent) atomic_inc(&server->client->cl_count); server->client_sys = server->client; } - if (server->flags & NFS_MOUNT_VER3) { +#ifdef CONFIG_NFS_V3_ACL + if (!(server->flags & NFS_MOUNT_NOACL)) { + server->client_acl = rpc_bind_new_program(server->client, &nfsacl_program, 3); + /* No errors! Assume that Sun nfsacls are supported */ + if (!IS_ERR(server->client_acl)) + server->caps |= NFS_CAP_ACLS; + } +#else + server->flags &= ~NFS_MOUNT_NOACL; +#endif /* CONFIG_NFS_V3_ACL */ if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN) server->namelen = NFS3_MAXNAMLEN; sb->s_time_gran = 1; @@ -546,6 +573,7 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) { NFS_MOUNT_NOCTO, ",nocto", "" }, { NFS_MOUNT_NOAC, ",noac", "" }, { NFS_MOUNT_NONLM, ",nolock", ",lock" }, + { NFS_MOUNT_NOACL, ",noacl", "" }, { 0, NULL, NULL } }; struct proc_nfs_info *nfs_infop; @@ -1452,7 +1480,7 @@ static struct super_block *nfs_get_sb(struct file_system_type *fs_type, memset(server, 0, sizeof(struct nfs_server)); /* Zero out the NFS state stuff */ init_nfsv4_state(server); - server->client = server->client_sys = ERR_PTR(-EINVAL); + server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL); root = &server->fh; if (data->flags & NFS_MOUNT_VER3) @@ -1513,6 +1541,8 @@ static void nfs_kill_super(struct super_block *s) rpc_shutdown_client(server->client); if (!IS_ERR(server->client_sys)) rpc_shutdown_client(server->client_sys); + if (!IS_ERR(server->client_acl)) + rpc_shutdown_client(server->client_acl); if (!(server->flags & NFS_MOUNT_NONLM)) lockd_down(); /* release rpc.lockd */ @@ -1794,7 +1824,7 @@ static struct super_block *nfs4_get_sb(struct file_system_type *fs_type, memset(server, 0, sizeof(struct nfs_server)); /* Zero out the NFS state stuff */ init_nfsv4_state(server); - server->client = server->client_sys = ERR_PTR(-EINVAL); + server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL); p = nfs_copy_user_string(NULL, &data->hostname, 256); if (IS_ERR(p)) diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c new file mode 100644 index 000000000000..393ba79fc14f --- /dev/null +++ b/fs/nfs/nfs3acl.c @@ -0,0 +1,303 @@ +#include +#include +#include +#include +#include +#include + +#define NFSDBG_FACILITY NFSDBG_PROC + +ssize_t nfs3_listxattr(struct dentry *dentry, char *buffer, size_t size) +{ + struct inode *inode = dentry->d_inode; + struct posix_acl *acl; + int pos=0, len=0; + +# define output(s) do { \ + if (pos + sizeof(s) <= size) { \ + memcpy(buffer + pos, s, sizeof(s)); \ + pos += sizeof(s); \ + } \ + len += sizeof(s); \ + } while(0) + + acl = nfs3_proc_getacl(inode, ACL_TYPE_ACCESS); + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl) { + output("system.posix_acl_access"); + posix_acl_release(acl); + } + + if (S_ISDIR(inode->i_mode)) { + acl = nfs3_proc_getacl(inode, ACL_TYPE_DEFAULT); + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl) { + output("system.posix_acl_default"); + posix_acl_release(acl); + } + } + +# undef output + + if (!buffer || len <= size) + return len; + return -ERANGE; +} + +ssize_t nfs3_getxattr(struct dentry *dentry, const char *name, + void *buffer, size_t size) +{ + struct inode *inode = dentry->d_inode; + struct posix_acl *acl; + int type, error = 0; + + if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0) + type = ACL_TYPE_ACCESS; + else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0) + type = ACL_TYPE_DEFAULT; + else + return -EOPNOTSUPP; + + acl = nfs3_proc_getacl(inode, type); + if (IS_ERR(acl)) + return PTR_ERR(acl); + else if (acl) { + if (type == ACL_TYPE_ACCESS && acl->a_count == 0) + error = -ENODATA; + else + error = posix_acl_to_xattr(acl, buffer, size); + posix_acl_release(acl); + } else + error = -ENODATA; + + return error; +} + +int nfs3_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) +{ + struct inode *inode = dentry->d_inode; + struct posix_acl *acl; + int type, error; + + if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0) + type = ACL_TYPE_ACCESS; + else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0) + type = ACL_TYPE_DEFAULT; + else + return -EOPNOTSUPP; + + acl = posix_acl_from_xattr(value, size); + if (IS_ERR(acl)) + return PTR_ERR(acl); + error = nfs3_proc_setacl(inode, type, acl); + posix_acl_release(acl); + + return error; +} + +int nfs3_removexattr(struct dentry *dentry, const char *name) +{ + struct inode *inode = dentry->d_inode; + int type; + + if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0) + type = ACL_TYPE_ACCESS; + else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0) + type = ACL_TYPE_DEFAULT; + else + return -EOPNOTSUPP; + + return nfs3_proc_setacl(inode, type, NULL); +} + +struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_fattr fattr; + struct page *pages[NFSACL_MAXPAGES] = { }; + struct nfs3_getaclargs args = { + .fh = NFS_FH(inode), + /* The xdr layer may allocate pages here. */ + .pages = pages, + }; + struct nfs3_getaclres res = { + .fattr = &fattr, + }; + struct posix_acl *acl = NULL; + int status, count; + + if (!nfs_server_capable(inode, NFS_CAP_ACLS)) + return ERR_PTR(-EOPNOTSUPP); + + switch (type) { + case ACL_TYPE_ACCESS: + args.mask = NFS_ACLCNT|NFS_ACL; + break; + + case ACL_TYPE_DEFAULT: + if (!S_ISDIR(inode->i_mode)) + return NULL; + args.mask = NFS_DFACLCNT|NFS_DFACL; + break; + + default: + return ERR_PTR(-EINVAL); + } + + dprintk("NFS call getacl\n"); + status = rpc_call(server->client_acl, ACLPROC3_GETACL, + &args, &res, 0); + dprintk("NFS reply getacl: %d\n", status); + + /* pages may have been allocated at the xdr layer. */ + for (count = 0; count < NFSACL_MAXPAGES && args.pages[count]; count++) + __free_page(args.pages[count]); + + switch (status) { + case 0: + status = nfs_refresh_inode(inode, &fattr); + break; + case -EPFNOSUPPORT: + case -EPROTONOSUPPORT: + dprintk("NFS_V3_ACL extension not supported; disabling\n"); + server->caps &= ~NFS_CAP_ACLS; + case -ENOTSUPP: + status = -EOPNOTSUPP; + default: + goto getout; + } + if ((args.mask & res.mask) != args.mask) { + status = -EIO; + goto getout; + } + + if (res.acl_access != NULL) { + if (posix_acl_equiv_mode(res.acl_access, NULL) == 0) { + posix_acl_release(res.acl_access); + res.acl_access = NULL; + } + } + + switch(type) { + case ACL_TYPE_ACCESS: + acl = res.acl_access; + res.acl_access = NULL; + break; + + case ACL_TYPE_DEFAULT: + acl = res.acl_default; + res.acl_default = NULL; + } + +getout: + posix_acl_release(res.acl_access); + posix_acl_release(res.acl_default); + + if (status != 0) { + posix_acl_release(acl); + acl = ERR_PTR(status); + } + return acl; +} + +static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, + struct posix_acl *dfacl) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_fattr fattr; + struct page *pages[NFSACL_MAXPAGES] = { }; + struct nfs3_setaclargs args = { + .inode = inode, + .mask = NFS_ACL, + .acl_access = acl, + .pages = pages, + }; + int status, count; + + status = -EOPNOTSUPP; + if (!nfs_server_capable(inode, NFS_CAP_ACLS)) + goto out; + + /* We are doing this here, because XDR marshalling can only + return -ENOMEM. */ + status = -ENOSPC; + if (acl != NULL && acl->a_count > NFS_ACL_MAX_ENTRIES) + goto out; + if (dfacl != NULL && dfacl->a_count > NFS_ACL_MAX_ENTRIES) + goto out; + if (S_ISDIR(inode->i_mode)) { + args.mask |= NFS_DFACL; + args.acl_default = dfacl; + } + + dprintk("NFS call setacl\n"); + nfs_begin_data_update(inode); + status = rpc_call(server->client_acl, ACLPROC3_SETACL, + &args, &fattr, 0); + NFS_FLAGS(inode) |= NFS_INO_INVALID_ACCESS; + nfs_end_data_update(inode); + dprintk("NFS reply setacl: %d\n", status); + + /* pages may have been allocated at the xdr layer. */ + for (count = 0; count < NFSACL_MAXPAGES && args.pages[count]; count++) + __free_page(args.pages[count]); + + switch (status) { + case 0: + status = nfs_refresh_inode(inode, &fattr); + break; + case -EPFNOSUPPORT: + case -EPROTONOSUPPORT: + dprintk("NFS_V3_ACL SETACL RPC not supported" + "(will not retry)\n"); + server->caps &= ~NFS_CAP_ACLS; + case -ENOTSUPP: + status = -EOPNOTSUPP; + } +out: + return status; +} + +int nfs3_proc_setacl(struct inode *inode, int type, struct posix_acl *acl) +{ + struct posix_acl *alloc = NULL, *dfacl = NULL; + int status; + + if (S_ISDIR(inode->i_mode)) { + switch(type) { + case ACL_TYPE_ACCESS: + alloc = dfacl = nfs3_proc_getacl(inode, + ACL_TYPE_DEFAULT); + if (IS_ERR(alloc)) + goto fail; + break; + + case ACL_TYPE_DEFAULT: + dfacl = acl; + alloc = acl = nfs3_proc_getacl(inode, + ACL_TYPE_ACCESS); + if (IS_ERR(alloc)) + goto fail; + break; + + default: + return -EINVAL; + } + } else if (type != ACL_TYPE_ACCESS) + return -EINVAL; + + if (acl == NULL) { + alloc = acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL); + if (IS_ERR(alloc)) + goto fail; + } + status = nfs3_proc_setacls(inode, acl, dfacl); + posix_acl_release(alloc); + return status; + +fail: + return PTR_ERR(alloc); +} diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 53953a775714..d03bac0cc42f 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -17,6 +17,7 @@ #include #include #include +#include #define NFSDBG_FACILITY NFSDBG_PROC @@ -45,7 +46,7 @@ static inline int nfs3_rpc_call_wrapper(struct rpc_clnt *clnt, u32 proc, void *argp, void *resp, int flags) { struct rpc_message msg = { - .rpc_proc = &nfs3_procedures[proc], + .rpc_proc = &clnt->cl_procinfo[proc], .rpc_argp = argp, .rpc_resp = resp, }; @@ -825,8 +826,8 @@ nfs3_proc_lock(struct file *filp, int cmd, struct file_lock *fl) struct nfs_rpc_ops nfs_v3_clientops = { .version = 3, /* protocol version */ .dentry_ops = &nfs_dentry_operations, - .dir_inode_ops = &nfs_dir_inode_operations, - .file_inode_ops = &nfs_file_inode_operations, + .dir_inode_ops = &nfs3_dir_inode_operations, + .file_inode_ops = &nfs3_file_inode_operations, .getroot = nfs3_proc_get_root, .getattr = nfs3_proc_getattr, .setattr = nfs3_proc_setattr, diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index a3593d47e5ab..a4437fb177f0 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -21,6 +21,7 @@ #include #include #include +#include #define NFSDBG_FACILITY NFSDBG_XDR @@ -79,6 +80,11 @@ extern int nfs_stat_to_errno(int); #define NFS3_pathconfres_sz (1+NFS3_post_op_attr_sz+6) #define NFS3_commitres_sz (1+NFS3_wcc_data_sz+2) +#define ACL3_getaclargs_sz (NFS3_fh_sz+1) +#define ACL3_setaclargs_sz (NFS3_fh_sz+1+2*(2+5*3)) +#define ACL3_getaclres_sz (1+NFS3_post_op_attr_sz+1+2*(2+5*3)) +#define ACL3_setaclres_sz (1+NFS3_post_op_attr_sz) + /* * Map file type to S_IFMT bits */ @@ -627,6 +633,74 @@ nfs3_xdr_commitargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) return 0; } +#ifdef CONFIG_NFS_V3_ACL +/* + * Encode GETACL arguments + */ +static int +nfs3_xdr_getaclargs(struct rpc_rqst *req, u32 *p, + struct nfs3_getaclargs *args) +{ + struct rpc_auth *auth = req->rq_task->tk_auth; + unsigned int replen; + + p = xdr_encode_fhandle(p, args->fh); + *p++ = htonl(args->mask); + req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); + + if (args->mask & (NFS_ACL | NFS_DFACL)) { + /* Inline the page array */ + replen = (RPC_REPHDRSIZE + auth->au_rslack + + ACL3_getaclres_sz) << 2; + xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages, 0, + NFSACL_MAXPAGES << PAGE_SHIFT); + } + return 0; +} + +/* + * Encode SETACL arguments + */ +static int +nfs3_xdr_setaclargs(struct rpc_rqst *req, u32 *p, + struct nfs3_setaclargs *args) +{ + struct xdr_buf *buf = &req->rq_snd_buf; + unsigned int base, len_in_head, len = nfsacl_size( + (args->mask & NFS_ACL) ? args->acl_access : NULL, + (args->mask & NFS_DFACL) ? args->acl_default : NULL); + int count, err; + + p = xdr_encode_fhandle(p, NFS_FH(args->inode)); + *p++ = htonl(args->mask); + base = (char *)p - (char *)buf->head->iov_base; + /* put as much of the acls into head as possible. */ + len_in_head = min_t(unsigned int, buf->head->iov_len - base, len); + len -= len_in_head; + req->rq_slen = xdr_adjust_iovec(req->rq_svec, p + len_in_head); + + for (count = 0; (count << PAGE_SHIFT) < len; count++) { + args->pages[count] = alloc_page(GFP_KERNEL); + if (!args->pages[count]) { + while (count) + __free_page(args->pages[--count]); + return -ENOMEM; + } + } + xdr_encode_pages(buf, args->pages, 0, len); + + err = nfsacl_encode(buf, base, args->inode, + (args->mask & NFS_ACL) ? + args->acl_access : NULL, 1, 0); + if (err > 0) + err = nfsacl_encode(buf, base + err, args->inode, + (args->mask & NFS_DFACL) ? + args->acl_default : NULL, 1, + NFS_ACL_DEFAULT); + return (err > 0) ? 0 : err; +} +#endif /* CONFIG_NFS_V3_ACL */ + /* * NFS XDR decode functions */ @@ -978,6 +1052,54 @@ nfs3_xdr_commitres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) return 0; } +#ifdef CONFIG_NFS_V3_ACL +/* + * Decode GETACL reply + */ +static int +nfs3_xdr_getaclres(struct rpc_rqst *req, u32 *p, + struct nfs3_getaclres *res) +{ + struct xdr_buf *buf = &req->rq_rcv_buf; + int status = ntohl(*p++); + struct posix_acl **acl; + unsigned int *aclcnt; + int err, base; + + if (status != 0) + return -nfs_stat_to_errno(status); + p = xdr_decode_post_op_attr(p, res->fattr); + res->mask = ntohl(*p++); + if (res->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT)) + return -EINVAL; + base = (char *)p - (char *)req->rq_rcv_buf.head->iov_base; + + acl = (res->mask & NFS_ACL) ? &res->acl_access : NULL; + aclcnt = (res->mask & NFS_ACLCNT) ? &res->acl_access_count : NULL; + err = nfsacl_decode(buf, base, aclcnt, acl); + + acl = (res->mask & NFS_DFACL) ? &res->acl_default : NULL; + aclcnt = (res->mask & NFS_DFACLCNT) ? &res->acl_default_count : NULL; + if (err > 0) + err = nfsacl_decode(buf, base + err, aclcnt, acl); + return (err > 0) ? 0 : err; +} + +/* + * Decode setacl reply. + */ +static int +nfs3_xdr_setaclres(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +{ + int status = ntohl(*p++); + + if (status) + return -nfs_stat_to_errno(status); + xdr_decode_post_op_attr(p, fattr); + return 0; +} +#endif /* CONFIG_NFS_V3_ACL */ + #ifndef MAX # define MAX(a, b) (((a) > (b))? (a) : (b)) #endif @@ -1021,3 +1143,28 @@ struct rpc_version nfs_version3 = { .procs = nfs3_procedures }; +#ifdef CONFIG_NFS_V3_ACL +static struct rpc_procinfo nfs3_acl_procedures[] = { + [ACLPROC3_GETACL] = { + .p_proc = ACLPROC3_GETACL, + .p_encode = (kxdrproc_t) nfs3_xdr_getaclargs, + .p_decode = (kxdrproc_t) nfs3_xdr_getaclres, + .p_bufsiz = MAX(ACL3_getaclargs_sz, ACL3_getaclres_sz) << 2, + .p_timer = 1, + }, + [ACLPROC3_SETACL] = { + .p_proc = ACLPROC3_SETACL, + .p_encode = (kxdrproc_t) nfs3_xdr_setaclargs, + .p_decode = (kxdrproc_t) nfs3_xdr_setaclres, + .p_bufsiz = MAX(ACL3_setaclargs_sz, ACL3_setaclres_sz) << 2, + .p_timer = 0, + }, +}; + +struct rpc_version nfsacl_version3 = { + .number = 3, + .nrprocs = sizeof(nfs3_acl_procedures)/ + sizeof(nfs3_acl_procedures[0]), + .procs = nfs3_acl_procedures, +}; +#endif /* CONFIG_NFS_V3_ACL */ diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c index fd5bc596fe8a..1b272a135a31 100644 --- a/fs/nfs/nfsroot.c +++ b/fs/nfs/nfsroot.c @@ -124,6 +124,7 @@ enum { Opt_soft, Opt_hard, Opt_intr, Opt_nointr, Opt_posix, Opt_noposix, Opt_cto, Opt_nocto, Opt_ac, Opt_noac, Opt_lock, Opt_nolock, Opt_v2, Opt_v3, Opt_udp, Opt_tcp, + Opt_acl, Opt_noacl, /* Error token */ Opt_err }; @@ -158,6 +159,8 @@ static match_table_t __initdata tokens = { {Opt_udp, "udp"}, {Opt_tcp, "proto=tcp"}, {Opt_tcp, "tcp"}, + {Opt_acl, "acl"}, + {Opt_noacl, "noacl"}, {Opt_err, NULL} }; @@ -266,6 +269,12 @@ static int __init root_nfs_parse(char *name, char *buf) case Opt_tcp: nfs_data.flags |= NFS_MOUNT_TCP; break; + case Opt_acl: + nfs_data.flags &= ~NFS_MOUNT_NOACL; + break; + case Opt_noacl: + nfs_data.flags |= NFS_MOUNT_NOACL; + break; default : return 0; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index d2b5d7e0e85a..3a5e442ac776 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -301,6 +301,9 @@ extern u32 root_nfs_parse_addr(char *name); /*__init*/ * linux/fs/nfs/file.c */ extern struct inode_operations nfs_file_inode_operations; +#ifdef CONFIG_NFS_V3 +extern struct inode_operations nfs3_file_inode_operations; +#endif /* CONFIG_NFS_V3 */ extern struct file_operations nfs_file_operations; extern struct address_space_operations nfs_file_aops; @@ -315,6 +318,22 @@ static inline struct rpc_cred *nfs_file_cred(struct file *file) return NULL; } +/* + * linux/fs/nfs/xattr.c + */ +#ifdef CONFIG_NFS_V3_ACL +extern ssize_t nfs3_listxattr(struct dentry *, char *, size_t); +extern ssize_t nfs3_getxattr(struct dentry *, const char *, void *, size_t); +extern int nfs3_setxattr(struct dentry *, const char *, + const void *, size_t, int); +extern int nfs3_removexattr (struct dentry *, const char *name); +#else +# define nfs3_listxattr NULL +# define nfs3_getxattr NULL +# define nfs3_setxattr NULL +# define nfs3_removexattr NULL +#endif + /* * linux/fs/nfs/direct.c */ @@ -329,6 +348,9 @@ extern ssize_t nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, * linux/fs/nfs/dir.c */ extern struct inode_operations nfs_dir_inode_operations; +#ifdef CONFIG_NFS_V3 +extern struct inode_operations nfs3_dir_inode_operations; +#endif /* CONFIG_NFS_V3 */ extern struct file_operations nfs_dir_operations; extern struct dentry_operations nfs_dentry_operations; @@ -449,6 +471,15 @@ static inline void nfs_readdata_free(struct nfs_read_data *p) extern void nfs_readdata_release(struct rpc_task *task); +/* + * linux/fs/nfs3proc.c + */ +#ifdef CONFIG_NFS_V3_ACL +extern struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type); +extern int nfs3_proc_setacl(struct inode *inode, int type, + struct posix_acl *acl); +#endif /* CONFIG_NFS_V3_ACL */ + /* * linux/fs/mount_clnt.c * (Used only by nfsroot module) diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index fc51645d61ee..3d3a305488cf 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -10,6 +10,7 @@ struct nfs_server { struct rpc_clnt * client; /* RPC client handle */ struct rpc_clnt * client_sys; /* 2nd handle for FSINFO */ + struct rpc_clnt * client_acl; /* ACL RPC client handle */ struct nfs_rpc_ops * rpc_ops; /* NFS protocol vector */ struct backing_dev_info backing_dev_info; int flags; /* various flags */ diff --git a/include/linux/nfs_mount.h b/include/linux/nfs_mount.h index 0071428231f9..659c75438454 100644 --- a/include/linux/nfs_mount.h +++ b/include/linux/nfs_mount.h @@ -58,6 +58,7 @@ struct nfs_mount_data { #define NFS_MOUNT_KERBEROS 0x0100 /* 3 */ #define NFS_MOUNT_NONLM 0x0200 /* 3 */ #define NFS_MOUNT_BROKEN_SUID 0x0400 /* 4 */ +#define NFS_MOUNT_NOACL 0x0800 /* 4 */ #define NFS_MOUNT_STRICTLOCK 0x1000 /* reserved for NFSv4 */ #define NFS_MOUNT_SECFLAVOUR 0x2000 /* 5 */ #define NFS_MOUNT_FLAGMASK 0xFFFF diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 46b206b460c0..a2bf6914ff1b 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -2,6 +2,7 @@ #define _LINUX_NFS_XDR_H #include +#include struct nfs4_fsid { __u64 major; @@ -368,6 +369,20 @@ struct nfs_readdirargs { struct page ** pages; }; +struct nfs3_getaclargs { + struct nfs_fh * fh; + int mask; + struct page ** pages; +}; + +struct nfs3_setaclargs { + struct inode * inode; + int mask; + struct posix_acl * acl_access; + struct posix_acl * acl_default; + struct page ** pages; +}; + struct nfs_diropok { struct nfs_fh * fh; struct nfs_fattr * fattr; @@ -491,6 +506,15 @@ struct nfs3_readdirres { int plus; }; +struct nfs3_getaclres { + struct nfs_fattr * fattr; + int mask; + unsigned int acl_access_count; + unsigned int acl_default_count; + struct posix_acl * acl_access; + struct posix_acl * acl_default; +}; + #ifdef CONFIG_NFS_V4 typedef u64 clientid4; @@ -748,4 +772,7 @@ extern struct rpc_version nfs_version2; extern struct rpc_version nfs_version3; extern struct rpc_version nfs_version4; +extern struct rpc_version nfsacl_version3; +extern struct rpc_program nfsacl_program; + #endif -- cgit v1.2.3 From 055ffbea0596942579b0dae71d5dab78de8135f6 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:27 +0000 Subject: [PATCH] NFS: Fix handling of the umask when an NFSv3 default acl is present. NFSv3 has no concept of a umask on the server side: The client applies the umask locally, and sends the effective permissions to the server. This behavior is wrong when files are created in a directory that has a default ACL. In this case, the umask is supposed to be ignored, and only the default ACL determines the file's effective permissions. Usually its the server's task to conditionally apply the umask. But since the server knows nothing about the umask, we have to do it on the client side. This patch tries to fetch the parent directory's default ACL before creating a new file, computes the appropriate create mode to send to the server, and finally sets the new file's access and default acl appropriately. Many thanks to Buck Huppmann for sending the initial version of this patch, as well as for arguing why we need this change. Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 5 +++++ fs/nfs/nfs3acl.c | 29 +++++++++++++++++++++++++++++ fs/nfs/nfs3proc.c | 36 ++++++++++++++++++++++++++++++------ include/linux/nfs_fs.h | 9 +++++++++ 4 files changed, 73 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 440b9cbb6f81..50a03f1504a1 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -490,6 +490,11 @@ nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent) #else server->flags &= ~NFS_MOUNT_NOACL; #endif /* CONFIG_NFS_V3_ACL */ + /* + * The VFS shouldn't apply the umask to mode bits. We will + * do so ourselves when necessary. + */ + sb->s_flags |= MS_POSIXACL; if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN) server->namelen = NFS3_MAXNAMLEN; sb->s_time_gran = 1; diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index 393ba79fc14f..89b6468700e7 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -301,3 +301,32 @@ int nfs3_proc_setacl(struct inode *inode, int type, struct posix_acl *acl) fail: return PTR_ERR(alloc); } + +int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode, + mode_t mode) +{ + struct posix_acl *dfacl, *acl; + int error = 0; + + dfacl = nfs3_proc_getacl(dir, ACL_TYPE_DEFAULT); + if (IS_ERR(dfacl)) { + error = PTR_ERR(dfacl); + return (error == -EOPNOTSUPP) ? 0 : error; + } + if (!dfacl) + return 0; + acl = posix_acl_clone(dfacl, GFP_KERNEL); + error = -ENOMEM; + if (!acl) + goto out_release_dfacl; + error = posix_acl_create_masq(acl, &mode); + if (error < 0) + goto out_release_acl; + error = nfs3_proc_setacls(inode, acl, S_ISDIR(inode->i_mode) ? + dfacl : NULL); +out_release_acl: + posix_acl_release(acl); +out_release_dfacl: + posix_acl_release(dfacl); + return error; +} diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index d03bac0cc42f..a9ddc196224d 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -314,7 +314,8 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, .fh = &fhandle, .fattr = &fattr }; - int status; + mode_t mode = sattr->ia_mode; + int status; dprintk("NFS call create %s\n", dentry->d_name.name); arg.createmode = NFS3_CREATE_UNCHECKED; @@ -324,6 +325,8 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, arg.verifier[1] = current->pid; } + sattr->ia_mode &= ~current->fs->umask; + again: dir_attr.valid = 0; fattr.valid = 0; @@ -370,6 +373,9 @@ again: nfs_refresh_inode(dentry->d_inode, &fattr); dprintk("NFS reply setattr (post-create): %d\n", status); } + if (status != 0) + goto out; + status = nfs3_proc_set_default_acl(dir, dentry->d_inode, mode); out: dprintk("NFS reply create: %d\n", status); return status; @@ -539,15 +545,24 @@ nfs3_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr) .fh = &fhandle, .fattr = &fattr }; - int status; + int mode = sattr->ia_mode; + int status; dprintk("NFS call mkdir %s\n", dentry->d_name.name); dir_attr.valid = 0; fattr.valid = 0; + + sattr->ia_mode &= ~current->fs->umask; + status = rpc_call(NFS_CLIENT(dir), NFS3PROC_MKDIR, &arg, &res, 0); nfs_refresh_inode(dir, &dir_attr); - if (status == 0) - status = nfs_instantiate(dentry, &fhandle, &fattr); + if (status != 0) + goto out; + status = nfs_instantiate(dentry, &fhandle, &fattr); + if (status != 0) + goto out; + status = nfs3_proc_set_default_acl(dir, dentry->d_inode, mode); +out: dprintk("NFS reply mkdir: %d\n", status); return status; } @@ -642,6 +657,7 @@ nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, .fh = &fh, .fattr = &fattr }; + mode_t mode = sattr->ia_mode; int status; switch (sattr->ia_mode & S_IFMT) { @@ -654,12 +670,20 @@ nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, dprintk("NFS call mknod %s %u:%u\n", dentry->d_name.name, MAJOR(rdev), MINOR(rdev)); + + sattr->ia_mode &= ~current->fs->umask; + dir_attr.valid = 0; fattr.valid = 0; status = rpc_call(NFS_CLIENT(dir), NFS3PROC_MKNOD, &arg, &res, 0); nfs_refresh_inode(dir, &dir_attr); - if (status == 0) - status = nfs_instantiate(dentry, &fh, &fattr); + if (status != 0) + goto out; + status = nfs_instantiate(dentry, &fh, &fattr); + if (status != 0) + goto out; + status = nfs3_proc_set_default_acl(dir, dentry->d_inode, mode); +out: dprintk("NFS reply mknod: %d\n", status); return status; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 3a5e442ac776..7662c5131b47 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -478,6 +478,15 @@ extern void nfs_readdata_release(struct rpc_task *task); extern struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type); extern int nfs3_proc_setacl(struct inode *inode, int type, struct posix_acl *acl); +extern int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode, + mode_t mode); +#else +static inline int nfs3_proc_set_default_acl(struct inode *dir, + struct inode *inode, + mode_t mode) +{ + return 0; +} #endif /* CONFIG_NFS_V3_ACL */ /* -- cgit v1.2.3 From 5c6a9f7d92291c832d47e792ed1fafa44acb066e Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:27 +0000 Subject: [PATCH] NFS: Cache the NFSv3 acls. Attach acls to inodes in the icache to avoid unnecessary GETACL RPC round-trips. As long as the client doesn't retrieve any acls itself, only the default acls of exiting directories and the default and access acls of new directories will end up in the cache, which preserves some memory compared to always caching the access and default acl of all files. Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- fs/nfs/nfs3acl.c | 100 +++++++++++++++++++++++++++++++++++++++++-------- fs/nfs/nfs3proc.c | 1 + include/linux/nfs_fs.h | 11 ++++++ 3 files changed, 97 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index 89b6468700e7..451112ff9aa4 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -113,6 +113,69 @@ int nfs3_removexattr(struct dentry *dentry, const char *name) return nfs3_proc_setacl(inode, type, NULL); } +static void __nfs3_forget_cached_acls(struct nfs_inode *nfsi) +{ + if (nfsi->acl_access != ERR_PTR(-EAGAIN)) { + posix_acl_release(nfsi->acl_access); + nfsi->acl_access = ERR_PTR(-EAGAIN); + } + if (nfsi->acl_default != ERR_PTR(-EAGAIN)) { + posix_acl_release(nfsi->acl_default); + nfsi->acl_default = ERR_PTR(-EAGAIN); + } +} + +void nfs3_forget_cached_acls(struct inode *inode) +{ + dprintk("NFS: nfs3_forget_cached_acls(%s/%ld)\n", inode->i_sb->s_id, + inode->i_ino); + spin_lock(&inode->i_lock); + __nfs3_forget_cached_acls(NFS_I(inode)); + spin_unlock(&inode->i_lock); +} + +static struct posix_acl *nfs3_get_cached_acl(struct inode *inode, int type) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct posix_acl *acl = ERR_PTR(-EAGAIN); + + spin_lock(&inode->i_lock); + switch(type) { + case ACL_TYPE_ACCESS: + acl = nfsi->acl_access; + break; + + case ACL_TYPE_DEFAULT: + acl = nfsi->acl_default; + break; + + default: + return ERR_PTR(-EINVAL); + } + if (acl == ERR_PTR(-EAGAIN)) + acl = ERR_PTR(-EAGAIN); + else + acl = posix_acl_dup(acl); + spin_unlock(&inode->i_lock); + dprintk("NFS: nfs3_get_cached_acl(%s/%ld, %d) = %p\n", inode->i_sb->s_id, + inode->i_ino, type, acl); + return acl; +} + +static void nfs3_cache_acls(struct inode *inode, struct posix_acl *acl, + struct posix_acl *dfacl) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + dprintk("nfs3_cache_acls(%s/%ld, %p, %p)\n", inode->i_sb->s_id, + inode->i_ino, acl, dfacl); + spin_lock(&inode->i_lock); + __nfs3_forget_cached_acls(NFS_I(inode)); + nfsi->acl_access = posix_acl_dup(acl); + nfsi->acl_default = posix_acl_dup(dfacl); + spin_unlock(&inode->i_lock); +} + struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type) { struct nfs_server *server = NFS_SERVER(inode); @@ -126,26 +189,32 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type) struct nfs3_getaclres res = { .fattr = &fattr, }; - struct posix_acl *acl = NULL; + struct posix_acl *acl; int status, count; if (!nfs_server_capable(inode, NFS_CAP_ACLS)) return ERR_PTR(-EOPNOTSUPP); - switch (type) { - case ACL_TYPE_ACCESS: - args.mask = NFS_ACLCNT|NFS_ACL; - break; - - case ACL_TYPE_DEFAULT: - if (!S_ISDIR(inode->i_mode)) - return NULL; - args.mask = NFS_DFACLCNT|NFS_DFACL; - break; - - default: - return ERR_PTR(-EINVAL); - } + status = nfs_revalidate_inode(server, inode); + if (status < 0) + return ERR_PTR(status); + acl = nfs3_get_cached_acl(inode, type); + if (acl != ERR_PTR(-EAGAIN)) + return acl; + acl = NULL; + + /* + * Only get the access acl when explicitly requested: We don't + * need it for access decisions, and only some applications use + * it. Applications which request the access acl first are not + * penalized from this optimization. + */ + if (type == ACL_TYPE_ACCESS) + args.mask |= NFS_ACLCNT|NFS_ACL; + if (S_ISDIR(inode->i_mode)) + args.mask |= NFS_DFACLCNT|NFS_DFACL; + if (args.mask == 0) + return NULL; dprintk("NFS call getacl\n"); status = rpc_call(server->client_acl, ACLPROC3_GETACL, @@ -180,6 +249,7 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type) res.acl_access = NULL; } } + nfs3_cache_acls(inode, res.acl_access, res.acl_default); switch(type) { case ACL_TYPE_ACCESS: diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index a9ddc196224d..7851569b31c6 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -882,4 +882,5 @@ struct nfs_rpc_ops nfs_v3_clientops = { .file_open = nfs_open, .file_release = nfs_release, .lock = nfs3_proc_lock, + .clear_acl_cache = nfs3_forget_cached_acls, }; diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 7662c5131b47..4ceac9ddac93 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -91,6 +91,8 @@ struct nfs_open_context { */ struct nfs_delegation; +struct posix_acl; + /* * nfs fs inode data in memory */ @@ -144,6 +146,10 @@ struct nfs_inode { atomic_t data_updates; struct nfs_access_entry cache_access; +#ifdef CONFIG_NFS_V3_ACL + struct posix_acl *acl_access; + struct posix_acl *acl_default; +#endif /* * This is the cookie verifier used for NFSv3 readdir @@ -480,6 +486,7 @@ extern int nfs3_proc_setacl(struct inode *inode, int type, struct posix_acl *acl); extern int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode, mode_t mode); +extern void nfs3_forget_cached_acls(struct inode *inode); #else static inline int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode, @@ -487,6 +494,10 @@ static inline int nfs3_proc_set_default_acl(struct inode *dir, { return 0; } + +static inline void nfs3_forget_cached_acls(struct inode *inode) +{ +} #endif /* CONFIG_NFS_V3_ACL */ /* -- cgit v1.2.3 From 00a926422765064cb28e218d4837411c88bf6a3e Mon Sep 17 00:00:00 2001 From: Olivier Galibert Date: Wed, 22 Jun 2005 17:16:29 +0000 Subject: [PATCH] NFS: Hide NFS server-generated readdir cookies from userland NFSv3 currently returns the unsigned 64-bit cookie directly to userspace. The following patch causes the kernel to generate loff_t offsets for the benefit of userland. The current server-generated READDIR cookie is cached in the nfs_open_context instead of in filp->f_pos, so we still end up work correctly under directory insertions/deletion. Signed-off-by: Olivier Galibert Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 114 ++++++++++++++++++++++++++++++++++++++----------- fs/nfs/inode.c | 2 + include/linux/nfs_fs.h | 3 ++ 3 files changed, 95 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 2c6a95945684..fceef29c65a3 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -141,7 +141,9 @@ typedef struct { struct page *page; unsigned long page_index; u32 *ptr; - u64 target; + u64 target_cookie; + int target_index; + int current_index; struct nfs_entry *entry; decode_dirent_t decode; int plus; @@ -225,14 +227,14 @@ void dir_page_release(nfs_readdir_descriptor_t *desc) /* * Given a pointer to a buffer that has already been filled by a call - * to readdir, find the next entry. + * to readdir, find the next entry with cookie 'desc->target_cookie'. * * If the end of the buffer has been reached, return -EAGAIN, if not, * return the offset within the buffer of the next entry to be * read. */ static inline -int find_dirent(nfs_readdir_descriptor_t *desc, struct page *page) +int find_dirent(nfs_readdir_descriptor_t *desc) { struct nfs_entry *entry = desc->entry; int loop_count = 0, @@ -240,7 +242,7 @@ int find_dirent(nfs_readdir_descriptor_t *desc, struct page *page) while((status = dir_decode(desc)) == 0) { dfprintk(VFS, "NFS: found cookie %Lu\n", (long long)entry->cookie); - if (entry->prev_cookie == desc->target) + if (entry->prev_cookie == desc->target_cookie) break; if (loop_count++ > 200) { loop_count = 0; @@ -252,8 +254,44 @@ int find_dirent(nfs_readdir_descriptor_t *desc, struct page *page) } /* - * Find the given page, and call find_dirent() in order to try to - * return the next entry. + * Given a pointer to a buffer that has already been filled by a call + * to readdir, find the entry at offset 'desc->target_index'. + * + * If the end of the buffer has been reached, return -EAGAIN, if not, + * return the offset within the buffer of the next entry to be + * read. + */ +static inline +int find_dirent_index(nfs_readdir_descriptor_t *desc) +{ + struct nfs_entry *entry = desc->entry; + int loop_count = 0, + status; + + for(;;) { + status = dir_decode(desc); + if (status) + break; + + dfprintk(VFS, "NFS: found cookie %Lu at index %d\n", (long long)entry->cookie, desc->current_index); + + if (desc->target_index == desc->current_index) { + desc->target_cookie = entry->cookie; + break; + } + desc->current_index++; + if (loop_count++ > 200) { + loop_count = 0; + schedule(); + } + } + dfprintk(VFS, "NFS: find_dirent_index() returns %d\n", status); + return status; +} + +/* + * Find the given page, and call find_dirent() or find_dirent_index in + * order to try to return the next entry. */ static inline int find_dirent_page(nfs_readdir_descriptor_t *desc) @@ -276,7 +314,10 @@ int find_dirent_page(nfs_readdir_descriptor_t *desc) /* NOTE: Someone else may have changed the READDIRPLUS flag */ desc->page = page; desc->ptr = kmap(page); /* matching kunmap in nfs_do_filldir */ - status = find_dirent(desc, page); + if (desc->target_cookie) + status = find_dirent(desc); + else + status = find_dirent_index(desc); if (status < 0) dir_page_release(desc); out: @@ -291,7 +332,8 @@ int find_dirent_page(nfs_readdir_descriptor_t *desc) * Recurse through the page cache pages, and return a * filled nfs_entry structure of the next directory entry if possible. * - * The target for the search is 'desc->target'. + * The target for the search is 'desc->target_cookie' if non-0, + * 'desc->target_index' otherwise */ static inline int readdir_search_pagecache(nfs_readdir_descriptor_t *desc) @@ -299,7 +341,19 @@ int readdir_search_pagecache(nfs_readdir_descriptor_t *desc) int loop_count = 0; int res; - dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie %Lu\n", (long long)desc->target); + if (desc->target_cookie) + dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie %Lu\n", (long long)desc->target_cookie); + else + dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie number %d\n", desc->target_index); + + /* Always search-by-index from the beginning of the cache */ + if (!(desc->target_cookie)) { + desc->page_index = 0; + desc->entry->cookie = desc->entry->prev_cookie = 0; + desc->entry->eof = 0; + desc->current_index = 0; + } + for (;;) { res = find_dirent_page(desc); if (res != -EAGAIN) @@ -332,11 +386,12 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent, struct file *file = desc->file; struct nfs_entry *entry = desc->entry; struct dentry *dentry = NULL; + struct nfs_open_context *ctx = file->private_data; unsigned long fileid; int loop_count = 0, res; - dfprintk(VFS, "NFS: nfs_do_filldir() filling starting @ cookie %Lu\n", (long long)desc->target); + dfprintk(VFS, "NFS: nfs_do_filldir() filling starting @ cookie %Lu\n", (long long)entry->cookie); for(;;) { unsigned d_type = DT_UNKNOWN; @@ -356,10 +411,11 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent, } res = filldir(dirent, entry->name, entry->len, - entry->prev_cookie, fileid, d_type); + file->f_pos, fileid, d_type); if (res < 0) break; - file->f_pos = desc->target = entry->cookie; + file->f_pos++; + desc->target_cookie = entry->cookie; if (dir_decode(desc) != 0) { desc->page_index ++; break; @@ -369,10 +425,12 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent, schedule(); } } + ctx->dir_pos = file->f_pos; + ctx->dir_cookie = desc->target_cookie; dir_page_release(desc); if (dentry != NULL) dput(dentry); - dfprintk(VFS, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", (long long)desc->target, res); + dfprintk(VFS, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", (long long)desc->target_cookie, res); return res; } @@ -398,14 +456,14 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, struct page *page = NULL; int status; - dfprintk(VFS, "NFS: uncached_readdir() searching for cookie %Lu\n", (long long)desc->target); + dfprintk(VFS, "NFS: uncached_readdir() searching for cookie %Lu\n", (long long)desc->target_cookie); page = alloc_page(GFP_HIGHUSER); if (!page) { status = -ENOMEM; goto out; } - desc->error = NFS_PROTO(inode)->readdir(file->f_dentry, cred, desc->target, + desc->error = NFS_PROTO(inode)->readdir(file->f_dentry, cred, desc->target_cookie, page, NFS_SERVER(inode)->dtsize, desc->plus); @@ -414,7 +472,7 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, desc->ptr = kmap(page); /* matching kunmap in nfs_do_filldir */ if (desc->error >= 0) { if ((status = dir_decode(desc)) == 0) - desc->entry->prev_cookie = desc->target; + desc->entry->prev_cookie = desc->target_cookie; } else status = -EIO; if (status < 0) @@ -435,13 +493,15 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, goto out; } -/* The file offset position is now represented as a true offset into the - * page cache as is the case in most of the other filesystems. +/* The file offset position represents the dirent entry number. A + last cookie cache takes care of the common case of reading the + whole directory. */ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) { struct dentry *dentry = filp->f_dentry; struct inode *inode = dentry->d_inode; + struct nfs_open_context *ctx = filp->private_data; nfs_readdir_descriptor_t my_desc, *desc = &my_desc; struct nfs_entry my_entry; @@ -458,17 +518,22 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) } /* - * filp->f_pos points to the file offset in the page cache. - * but if the cache has meanwhile been zapped, we need to - * read from the last dirent to revalidate f_pos - * itself. + * filp->f_pos points to the dirent entry number. + * ctx->dir_pos has the number of the cached cookie. We have + * to either find the entry with the appropriate number or + * revalidate the cookie. */ memset(desc, 0, sizeof(*desc)); desc->file = filp; - desc->target = filp->f_pos; desc->decode = NFS_PROTO(inode)->decode_dirent; desc->plus = NFS_USE_READDIRPLUS(inode); + desc->target_index = filp->f_pos; + + if (filp->f_pos == ctx->dir_pos) + desc->target_cookie = ctx->dir_cookie; + else + desc->target_cookie = 0; my_entry.cookie = my_entry.prev_cookie = 0; my_entry.eof = 0; @@ -478,9 +543,10 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) while(!desc->entry->eof) { res = readdir_search_pagecache(desc); + if (res == -EBADCOOKIE) { /* This means either end of directory */ - if (desc->entry->cookie != desc->target) { + if (desc->target_cookie && desc->entry->cookie != desc->target_cookie) { /* Or that the server has 'lost' a cookie */ res = uncached_readdir(desc, dirent, filldir); if (res >= 0) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 8a8d57d9d660..9fa02e7984ac 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -891,6 +891,8 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rp ctx->state = NULL; ctx->lockowner = current->files; ctx->error = 0; + ctx->dir_pos = 0; + ctx->dir_cookie = 0; } return ctx; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 4ceac9ddac93..f810195ef7ad 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -84,6 +84,9 @@ struct nfs_open_context { int error; struct list_head list; + + int dir_pos; /* Directory cookie cache */ + __u64 dir_cookie; }; /* -- cgit v1.2.3 From f0dd2136da6d2070e12bfa6d199b136318e666c7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:29 +0000 Subject: [PATCH] NFS: Clean up readdir changes. Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 85 ++++++++++++++++++++++++++++---------------------- fs/nfs/inode.c | 1 - include/linux/nfs_fs.h | 1 - 3 files changed, 48 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index fceef29c65a3..b38a57e78a63 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -51,8 +51,10 @@ static int nfs_mknod(struct inode *, struct dentry *, int, dev_t); static int nfs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *); static int nfs_fsync_dir(struct file *, struct dentry *, int); +static loff_t nfs_llseek_dir(struct file *, loff_t, int); struct file_operations nfs_dir_operations = { + .llseek = nfs_llseek_dir, .read = generic_read_dir, .readdir = nfs_readdir, .open = nfs_opendir, @@ -141,9 +143,8 @@ typedef struct { struct page *page; unsigned long page_index; u32 *ptr; - u64 target_cookie; - int target_index; - int current_index; + u64 *dir_cookie; + loff_t current_index; struct nfs_entry *entry; decode_dirent_t decode; int plus; @@ -227,7 +228,7 @@ void dir_page_release(nfs_readdir_descriptor_t *desc) /* * Given a pointer to a buffer that has already been filled by a call - * to readdir, find the next entry with cookie 'desc->target_cookie'. + * to readdir, find the next entry with cookie '*desc->dir_cookie'. * * If the end of the buffer has been reached, return -EAGAIN, if not, * return the offset within the buffer of the next entry to be @@ -241,8 +242,8 @@ int find_dirent(nfs_readdir_descriptor_t *desc) status; while((status = dir_decode(desc)) == 0) { - dfprintk(VFS, "NFS: found cookie %Lu\n", (long long)entry->cookie); - if (entry->prev_cookie == desc->target_cookie) + dfprintk(VFS, "NFS: found cookie %Lu\n", (unsigned long long)entry->cookie); + if (entry->prev_cookie == *desc->dir_cookie) break; if (loop_count++ > 200) { loop_count = 0; @@ -255,7 +256,7 @@ int find_dirent(nfs_readdir_descriptor_t *desc) /* * Given a pointer to a buffer that has already been filled by a call - * to readdir, find the entry at offset 'desc->target_index'. + * to readdir, find the entry at offset 'desc->file->f_pos'. * * If the end of the buffer has been reached, return -EAGAIN, if not, * return the offset within the buffer of the next entry to be @@ -273,10 +274,10 @@ int find_dirent_index(nfs_readdir_descriptor_t *desc) if (status) break; - dfprintk(VFS, "NFS: found cookie %Lu at index %d\n", (long long)entry->cookie, desc->current_index); + dfprintk(VFS, "NFS: found cookie %Lu at index %Ld\n", (unsigned long long)entry->cookie, desc->current_index); - if (desc->target_index == desc->current_index) { - desc->target_cookie = entry->cookie; + if (desc->file->f_pos == desc->current_index) { + *desc->dir_cookie = entry->cookie; break; } desc->current_index++; @@ -314,7 +315,7 @@ int find_dirent_page(nfs_readdir_descriptor_t *desc) /* NOTE: Someone else may have changed the READDIRPLUS flag */ desc->page = page; desc->ptr = kmap(page); /* matching kunmap in nfs_do_filldir */ - if (desc->target_cookie) + if (*desc->dir_cookie != 0) status = find_dirent(desc); else status = find_dirent_index(desc); @@ -332,8 +333,8 @@ int find_dirent_page(nfs_readdir_descriptor_t *desc) * Recurse through the page cache pages, and return a * filled nfs_entry structure of the next directory entry if possible. * - * The target for the search is 'desc->target_cookie' if non-0, - * 'desc->target_index' otherwise + * The target for the search is '*desc->dir_cookie' if non-0, + * 'desc->file->f_pos' otherwise */ static inline int readdir_search_pagecache(nfs_readdir_descriptor_t *desc) @@ -341,18 +342,15 @@ int readdir_search_pagecache(nfs_readdir_descriptor_t *desc) int loop_count = 0; int res; - if (desc->target_cookie) - dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie %Lu\n", (long long)desc->target_cookie); - else - dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie number %d\n", desc->target_index); - /* Always search-by-index from the beginning of the cache */ - if (!(desc->target_cookie)) { + if (*desc->dir_cookie == 0) { + dfprintk(VFS, "NFS: readdir_search_pagecache() searching for offset %Ld\n", (long long)desc->file->f_pos); desc->page_index = 0; desc->entry->cookie = desc->entry->prev_cookie = 0; desc->entry->eof = 0; desc->current_index = 0; - } + } else + dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie %Lu\n", (unsigned long long)*desc->dir_cookie); for (;;) { res = find_dirent_page(desc); @@ -386,7 +384,6 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent, struct file *file = desc->file; struct nfs_entry *entry = desc->entry; struct dentry *dentry = NULL; - struct nfs_open_context *ctx = file->private_data; unsigned long fileid; int loop_count = 0, res; @@ -415,7 +412,7 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent, if (res < 0) break; file->f_pos++; - desc->target_cookie = entry->cookie; + *desc->dir_cookie = entry->cookie; if (dir_decode(desc) != 0) { desc->page_index ++; break; @@ -425,12 +422,10 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent, schedule(); } } - ctx->dir_pos = file->f_pos; - ctx->dir_cookie = desc->target_cookie; dir_page_release(desc); if (dentry != NULL) dput(dentry); - dfprintk(VFS, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", (long long)desc->target_cookie, res); + dfprintk(VFS, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", (unsigned long long)*desc->dir_cookie, res); return res; } @@ -456,14 +451,14 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, struct page *page = NULL; int status; - dfprintk(VFS, "NFS: uncached_readdir() searching for cookie %Lu\n", (long long)desc->target_cookie); + dfprintk(VFS, "NFS: uncached_readdir() searching for cookie %Lu\n", (unsigned long long)*desc->dir_cookie); page = alloc_page(GFP_HIGHUSER); if (!page) { status = -ENOMEM; goto out; } - desc->error = NFS_PROTO(inode)->readdir(file->f_dentry, cred, desc->target_cookie, + desc->error = NFS_PROTO(inode)->readdir(file->f_dentry, cred, *desc->dir_cookie, page, NFS_SERVER(inode)->dtsize, desc->plus); @@ -472,7 +467,7 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, desc->ptr = kmap(page); /* matching kunmap in nfs_do_filldir */ if (desc->error >= 0) { if ((status = dir_decode(desc)) == 0) - desc->entry->prev_cookie = desc->target_cookie; + desc->entry->prev_cookie = *desc->dir_cookie; } else status = -EIO; if (status < 0) @@ -501,7 +496,6 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) { struct dentry *dentry = filp->f_dentry; struct inode *inode = dentry->d_inode; - struct nfs_open_context *ctx = filp->private_data; nfs_readdir_descriptor_t my_desc, *desc = &my_desc; struct nfs_entry my_entry; @@ -519,21 +513,16 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) /* * filp->f_pos points to the dirent entry number. - * ctx->dir_pos has the number of the cached cookie. We have + * *desc->dir_cookie has the cookie for the next entry. We have * to either find the entry with the appropriate number or * revalidate the cookie. */ memset(desc, 0, sizeof(*desc)); desc->file = filp; + desc->dir_cookie = &((struct nfs_open_context *)filp->private_data)->dir_cookie; desc->decode = NFS_PROTO(inode)->decode_dirent; desc->plus = NFS_USE_READDIRPLUS(inode); - desc->target_index = filp->f_pos; - - if (filp->f_pos == ctx->dir_pos) - desc->target_cookie = ctx->dir_cookie; - else - desc->target_cookie = 0; my_entry.cookie = my_entry.prev_cookie = 0; my_entry.eof = 0; @@ -546,7 +535,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (res == -EBADCOOKIE) { /* This means either end of directory */ - if (desc->target_cookie && desc->entry->cookie != desc->target_cookie) { + if (*desc->dir_cookie && desc->entry->cookie != *desc->dir_cookie) { /* Or that the server has 'lost' a cookie */ res = uncached_readdir(desc, dirent, filldir); if (res >= 0) @@ -579,6 +568,28 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) return 0; } +loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int origin) +{ + down(&filp->f_dentry->d_inode->i_sem); + switch (origin) { + case 1: + offset += filp->f_pos; + case 0: + if (offset >= 0) + break; + default: + offset = -EINVAL; + goto out; + } + if (offset != filp->f_pos) { + filp->f_pos = offset; + ((struct nfs_open_context *)filp->private_data)->dir_cookie = 0; + } +out: + up(&filp->f_dentry->d_inode->i_sem); + return offset; +} + /* * All directory operations under NFS are synchronous, so fsync() * is a dummy operation. diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 9fa02e7984ac..6300e05e9463 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -891,7 +891,6 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rp ctx->state = NULL; ctx->lockowner = current->files; ctx->error = 0; - ctx->dir_pos = 0; ctx->dir_cookie = 0; } return ctx; diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index f810195ef7ad..c90313bfa435 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -85,7 +85,6 @@ struct nfs_open_context { struct list_head list; - int dir_pos; /* Directory cookie cache */ __u64 dir_cookie; }; -- cgit v1.2.3 From 951a143b3fcf15cfa9d38250b7462f821db241db Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:30 +0000 Subject: [PATCH] NFS: Fix the file size revalidation Instead of looking at whether or not the file is open for writes before we accept to update the length using the server value, we should rather be looking at whether or not we are currently caching any writes. Failure to do so means in particular that we're not updating the file length correctly after obtaining a POSIX or BSD lock. Signed-off-by: Trond Myklebust --- fs/nfs/direct.c | 2 +- fs/nfs/inode.c | 69 +++++++++++++------------------------------------- fs/nfs/write.c | 4 +-- include/linux/nfs_fs.h | 1 - 4 files changed, 21 insertions(+), 55 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 68df803f27ca..d6a30c844de3 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -517,7 +517,7 @@ retry: result = tot_bytes; out: - nfs_end_data_update_defer(inode); + nfs_end_data_update(inode); nfs_writedata_free(wdata); return result; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 6300e05e9463..b2d16758ced8 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1147,27 +1147,6 @@ void nfs_end_data_update(struct inode *inode) atomic_dec(&nfsi->data_updates); } -/** - * nfs_end_data_update_defer - * @inode - pointer to inode - * Declare end of the operations that will update file data - * This will defer marking the inode as needing revalidation - * unless there are no other pending updates. - */ -void nfs_end_data_update_defer(struct inode *inode) -{ - struct nfs_inode *nfsi = NFS_I(inode); - - if (atomic_dec_and_test(&nfsi->data_updates)) { - /* Mark the attribute cache for revalidation */ - nfsi->flags |= NFS_INO_INVALID_ATTR; - /* Directories and symlinks: invalidate page cache too */ - if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) - nfsi->flags |= NFS_INO_INVALID_DATA; - nfsi->cache_change_attribute ++; - } -} - /** * nfs_refresh_inode - verify consistency of the inode attribute cache * @inode - pointer to inode @@ -1222,8 +1201,8 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) if (!timespec_equal(&inode->i_mtime, &fattr->mtime) || cur_size != new_isize) nfsi->flags |= NFS_INO_INVALID_ATTR; - } else if (S_ISREG(inode->i_mode) && new_isize > cur_size) - nfsi->flags |= NFS_INO_INVALID_ATTR; + } else if (new_isize != cur_size && nfsi->npages == 0) + nfsi->flags |= NFS_INO_INVALID_ATTR; /* Have any file permissions changed? */ if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO) @@ -1257,10 +1236,8 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsigned long verifier) { struct nfs_inode *nfsi = NFS_I(inode); - __u64 new_size; - loff_t new_isize; + loff_t cur_isize, new_isize; unsigned int invalid = 0; - loff_t cur_isize; int data_unstable; dfprintk(VFS, "NFS: %s(%s/%ld ct=%d info=0x%x)\n", @@ -1293,49 +1270,39 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign /* Are we racing with known updates of the metadata on the server? */ data_unstable = ! nfs_verify_change_attribute(inode, verifier); - /* Check if the file size agrees */ - new_size = fattr->size; + /* Check if our cached file size is stale */ new_isize = nfs_size_to_loff_t(fattr->size); cur_isize = i_size_read(inode); - if (cur_isize != new_size) { -#ifdef NFS_DEBUG_VERBOSE - printk(KERN_DEBUG "NFS: isize change on %s/%ld\n", inode->i_sb->s_id, inode->i_ino); -#endif - /* - * If we have pending writebacks, things can get - * messy. - */ - if (S_ISREG(inode->i_mode) && data_unstable) { - if (new_isize > cur_isize) { + if (new_isize != cur_isize) { + /* Do we perhaps have any outstanding writes? */ + if (nfsi->npages == 0) { + /* No, but did we race with nfs_end_data_update()? */ + if (verifier == nfsi->cache_change_attribute) { inode->i_size = new_isize; - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; + invalid |= NFS_INO_INVALID_DATA; } - } else { + invalid |= NFS_INO_INVALID_ATTR; + } else if (new_isize > cur_isize) { inode->i_size = new_isize; invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; } + dprintk("NFS: isize change on server for file %s/%ld\n", + inode->i_sb->s_id, inode->i_ino); } - /* - * Note: we don't check inode->i_mtime since pipes etc. - * can change this value in VFS without requiring a - * cache revalidation. - */ + /* Check if the mtime agrees */ if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) { memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime)); -#ifdef NFS_DEBUG_VERBOSE - printk(KERN_DEBUG "NFS: mtime change on %s/%ld\n", inode->i_sb->s_id, inode->i_ino); -#endif + dprintk("NFS: mtime change on server for file %s/%ld\n", + inode->i_sb->s_id, inode->i_ino); if (!data_unstable) invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; } if ((fattr->valid & NFS_ATTR_FATTR_V4) && nfsi->change_attr != fattr->change_attr) { -#ifdef NFS_DEBUG_VERBOSE - printk(KERN_DEBUG "NFS: change_attr change on %s/%ld\n", + dprintk("NFS: change_attr change on server for file %s/%ld\n", inode->i_sb->s_id, inode->i_ino); -#endif nfsi->change_attr = fattr->change_attr; if (!data_unstable) invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 6f7a4af3bc46..c574d551f029 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -220,7 +220,7 @@ static int nfs_writepage_sync(struct nfs_open_context *ctx, struct inode *inode, ClearPageError(page); io_error: - nfs_end_data_update_defer(inode); + nfs_end_data_update(inode); nfs_writedata_free(wdata); return written ? written : result; } @@ -401,7 +401,7 @@ static void nfs_inode_remove_request(struct nfs_page *req) nfsi->npages--; if (!nfsi->npages) { spin_unlock(&nfsi->req_lock); - nfs_end_data_update_defer(inode); + nfs_end_data_update(inode); iput(inode); } else spin_unlock(&nfsi->req_lock); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index c90313bfa435..211266c56ce5 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -294,7 +294,6 @@ extern void nfs_begin_attr_update(struct inode *); extern void nfs_end_attr_update(struct inode *); extern void nfs_begin_data_update(struct inode *); extern void nfs_end_data_update(struct inode *); -extern void nfs_end_data_update_defer(struct inode *); extern struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rpc_cred *cred); extern struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx); extern void put_nfs_open_context(struct nfs_open_context *ctx); -- cgit v1.2.3 From 7d52e86274e09fce8ac8f963e3605a84d0a305a7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:30 +0000 Subject: [PATCH] NFS: Cleanup of caching code, and slight optimization of writes. Unless we're doing O_APPEND writes, we really don't care about revalidating the file length. Just make sure that we catch any page cache invalidations. Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 12 +++++++++--- fs/nfs/inode.c | 44 +++++++++++++++++++++++++++++--------------- include/linux/nfs_fs.h | 1 + 3 files changed, 39 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/file.c b/fs/nfs/file.c index a606708264ed..40436857ed42 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -333,9 +333,15 @@ nfs_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t result = -EBUSY; if (IS_SWAPFILE(inode)) goto out_swapfile; - result = nfs_revalidate_inode(NFS_SERVER(inode), inode); - if (result) - goto out; + /* + * O_APPEND implies that we must revalidate the file length. + */ + if (iocb->ki_filp->f_flags & O_APPEND) { + result = nfs_revalidate_file_size(inode, iocb->ki_filp); + if (result) + goto out; + } else + nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping); result = count; if (!count) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index b2d16758ced8..a3922f4cc0a8 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1062,21 +1062,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) if (verifier == nfsi->cache_change_attribute) nfsi->flags &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME); /* Do the page cache invalidation */ - if (flags & NFS_INO_INVALID_DATA) { - if (S_ISREG(inode->i_mode)) { - if (filemap_fdatawrite(inode->i_mapping) == 0) - filemap_fdatawait(inode->i_mapping); - nfs_wb_all(inode); - } - nfsi->flags &= ~NFS_INO_INVALID_DATA; - invalidate_inode_pages2(inode->i_mapping); - memset(NFS_COOKIEVERF(inode), 0, sizeof(NFS_COOKIEVERF(inode))); - dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", - inode->i_sb->s_id, - (long long)NFS_FILEID(inode)); - /* This ensures we revalidate dentries */ - nfsi->cache_change_attribute++; - } + nfs_revalidate_mapping(inode, inode->i_mapping); if (flags & NFS_INO_INVALID_ACL) nfs_zap_acl_cache(inode); dfprintk(PAGECACHE, "NFS: (%s/%Ld) revalidation complete\n", @@ -1115,6 +1101,34 @@ int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) return __nfs_revalidate_inode(server, inode); } +/** + * nfs_revalidate_mapping - Revalidate the pagecache + * @inode - pointer to host inode + * @mapping - pointer to mapping + */ +void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + if (nfsi->flags & NFS_INO_INVALID_DATA) { + if (S_ISREG(inode->i_mode)) { + if (filemap_fdatawrite(mapping) == 0) + filemap_fdatawait(mapping); + nfs_wb_all(inode); + } + invalidate_inode_pages2(mapping); + nfsi->flags &= ~NFS_INO_INVALID_DATA; + if (S_ISDIR(inode->i_mode)) { + memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); + /* This ensures we revalidate child dentries */ + nfsi->cache_change_attribute++; + } + dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", + inode->i_sb->s_id, + (long long)NFS_FILEID(inode)); + } +} + /** * nfs_begin_data_update * @inode - pointer to inode diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 211266c56ce5..443103c13e53 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -289,6 +289,7 @@ extern int nfs_release(struct inode *, struct file *); extern int nfs_attribute_timeout(struct inode *inode); extern int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode); extern int __nfs_revalidate_inode(struct nfs_server *, struct inode *); +extern void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping); extern int nfs_setattr(struct dentry *, struct iattr *); extern void nfs_begin_attr_update(struct inode *); extern void nfs_end_attr_update(struct inode *); -- cgit v1.2.3 From fe51beecc55d0b0dce289e4758e7c529a642f63e Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:30 +0000 Subject: [PATCH] NFS: Ensure that fstat() always returns the correct mtime Even if the file is open for writes. Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 28 ++++++++++++++++++++++------ fs/nfs/inode.c | 24 ++++++++++++++++-------- include/linux/nfs_fs.h | 1 + 3 files changed, 39 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 40436857ed42..5621ba9885f4 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -127,6 +127,21 @@ nfs_file_release(struct inode *inode, struct file *filp) return NFS_PROTO(inode)->file_release(inode, filp); } +/** + * nfs_revalidate_file - Revalidate the page cache & related metadata + * @inode - pointer to inode struct + * @file - pointer to file + */ +static int nfs_revalidate_file(struct inode *inode, struct file *filp) +{ + int retval = 0; + + if ((NFS_FLAGS(inode) & NFS_INO_REVAL_PAGECACHE) || nfs_attribute_timeout(inode)) + retval = __nfs_revalidate_inode(NFS_SERVER(inode), inode); + nfs_revalidate_mapping(inode, filp->f_mapping); + return 0; +} + /** * nfs_revalidate_size - Revalidate the file size * @inode - pointer to inode struct @@ -149,7 +164,8 @@ static int nfs_revalidate_file_size(struct inode *inode, struct file *filp) goto force_reval; if (nfsi->npages != 0) return 0; - return nfs_revalidate_inode(server, inode); + if (!(NFS_FLAGS(inode) & NFS_INO_REVAL_PAGECACHE) && !nfs_attribute_timeout(inode)) + return 0; force_reval: return __nfs_revalidate_inode(server, inode); } @@ -210,7 +226,7 @@ nfs_file_read(struct kiocb *iocb, char __user * buf, size_t count, loff_t pos) dentry->d_parent->d_name.name, dentry->d_name.name, (unsigned long) count, (unsigned long) pos); - result = nfs_revalidate_inode(NFS_SERVER(inode), inode); + result = nfs_revalidate_file(inode, iocb->ki_filp); if (!result) result = generic_file_aio_read(iocb, buf, count, pos); return result; @@ -228,7 +244,7 @@ nfs_file_sendfile(struct file *filp, loff_t *ppos, size_t count, dentry->d_parent->d_name.name, dentry->d_name.name, (unsigned long) count, (unsigned long long) *ppos); - res = nfs_revalidate_inode(NFS_SERVER(inode), inode); + res = nfs_revalidate_file(inode, filp); if (!res) res = generic_file_sendfile(filp, ppos, count, actor, target); return res; @@ -244,7 +260,7 @@ nfs_file_mmap(struct file * file, struct vm_area_struct * vma) dfprintk(VFS, "nfs: mmap(%s/%s)\n", dentry->d_parent->d_name.name, dentry->d_name.name); - status = nfs_revalidate_inode(NFS_SERVER(inode), inode); + status = nfs_revalidate_file(inode, file); if (!status) status = generic_file_mmap(file, vma); return status; @@ -340,8 +356,8 @@ nfs_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t result = nfs_revalidate_file_size(inode, iocb->ki_filp); if (result) goto out; - } else - nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping); + } + nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping); result = count; if (!count) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index a3922f4cc0a8..4f545f382ba6 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -620,9 +620,9 @@ nfs_zap_caches(struct inode *inode) memset(NFS_COOKIEVERF(inode), 0, sizeof(NFS_COOKIEVERF(inode))); if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)) - nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE; else - nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE; } static void nfs_zap_acl_cache(struct inode *inode) @@ -1055,6 +1055,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) goto out; } flags = nfsi->flags; + nfsi->flags &= ~NFS_INO_REVAL_PAGECACHE; /* * We may need to keep the attributes marked as invalid if * we raced with nfs_end_attr_update(). @@ -1187,8 +1188,11 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) if ((fattr->valid & NFS_ATTR_PRE_CHANGE) != 0 && nfsi->change_attr == fattr->pre_change_attr) nfsi->change_attr = fattr->change_attr; - if (!data_unstable && nfsi->change_attr != fattr->change_attr) + if (nfsi->change_attr != fattr->change_attr) { nfsi->flags |= NFS_INO_INVALID_ATTR; + if (!data_unstable) + nfsi->flags |= NFS_INO_REVAL_PAGECACHE; + } } if ((fattr->valid & NFS_ATTR_FATTR) == 0) @@ -1211,12 +1215,16 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) } /* Verify a few of the more important attributes */ - if (!data_unstable) { - if (!timespec_equal(&inode->i_mtime, &fattr->mtime) - || cur_size != new_isize) - nfsi->flags |= NFS_INO_INVALID_ATTR; - } else if (new_isize != cur_size && nfsi->npages == 0) + if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) { nfsi->flags |= NFS_INO_INVALID_ATTR; + if (!data_unstable) + nfsi->flags |= NFS_INO_REVAL_PAGECACHE; + } + if (cur_size != new_isize) { + nfsi->flags |= NFS_INO_INVALID_ATTR; + if (nfsi->npages == 0) + nfsi->flags |= NFS_INO_REVAL_PAGECACHE; + } /* Have any file permissions changed? */ if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO) diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 443103c13e53..2954e44ed498 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -198,6 +198,7 @@ struct nfs_inode { #define NFS_INO_INVALID_ATIME 0x0020 /* cached atime is invalid */ #define NFS_INO_INVALID_ACCESS 0x0040 /* cached access cred invalid */ #define NFS_INO_INVALID_ACL 0x0080 /* cached acls are invalid */ +#define NFS_INO_REVAL_PAGECACHE 0x1000 /* must revalidate pagecache */ static inline struct nfs_inode *NFS_I(struct inode *inode) { -- cgit v1.2.3 From c6a556b88adfacd2af90be84357c8165d716c27d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:30 +0000 Subject: [PATCH] NFS: Make searching and waiting on busy writeback requests more efficient. Basically copies the VFS's method for tracking writebacks and applies it to the struct nfs_page. Signed-off-by: Trond Myklebust --- fs/nfs/pagelist.c | 29 ++++++++++++++++++++++++++++- fs/nfs/read.c | 3 --- fs/nfs/write.c | 19 +++++++++---------- include/linux/nfs_page.h | 12 ++++++++---- 4 files changed, 45 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 80777f99a58a..356a33bb38a6 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -111,6 +111,33 @@ void nfs_unlock_request(struct nfs_page *req) nfs_release_request(req); } +/** + * nfs_set_page_writeback_locked - Lock a request for writeback + * @req: + */ +int nfs_set_page_writeback_locked(struct nfs_page *req) +{ + struct nfs_inode *nfsi = NFS_I(req->wb_context->dentry->d_inode); + + if (!nfs_lock_request(req)) + return 0; + radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_WRITEBACK); + return 1; +} + +/** + * nfs_clear_page_writeback - Unlock request and wake up sleepers + */ +void nfs_clear_page_writeback(struct nfs_page *req) +{ + struct nfs_inode *nfsi = NFS_I(req->wb_context->dentry->d_inode); + + spin_lock(&nfsi->req_lock); + radix_tree_tag_clear(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_WRITEBACK); + spin_unlock(&nfsi->req_lock); + nfs_unlock_request(req); +} + /** * nfs_clear_request - Free up all resources allocated to the request * @req: @@ -301,7 +328,7 @@ nfs_scan_list(struct list_head *head, struct list_head *dst, if (req->wb_index > idx_end) break; - if (!nfs_lock_request(req)) + if (!nfs_set_page_writeback_locked(req)) continue; nfs_list_remove_request(req); nfs_list_add_request(req, dst); diff --git a/fs/nfs/read.c b/fs/nfs/read.c index a0042fb58634..6f866b8aa2d5 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -173,7 +173,6 @@ static int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode, if (len < PAGE_CACHE_SIZE) memclear_highpage_flush(page, len, PAGE_CACHE_SIZE - len); - nfs_lock_request(new); nfs_list_add_request(new, &one_request); nfs_pagein_one(&one_request, inode); return 0; @@ -185,7 +184,6 @@ static void nfs_readpage_release(struct nfs_page *req) nfs_clear_request(req); nfs_release_request(req); - nfs_unlock_request(req); dprintk("NFS: read done (%s/%Ld %d@%Ld)\n", req->wb_context->dentry->d_inode->i_sb->s_id, @@ -553,7 +551,6 @@ readpage_async_filler(void *data, struct page *page) } if (len < PAGE_CACHE_SIZE) memclear_highpage_flush(page, len, PAGE_CACHE_SIZE - len); - nfs_lock_request(new); nfs_list_add_request(new, desc->head); return 0; } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 79b621a545b2..58a39b0486a7 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -503,13 +503,12 @@ nfs_wait_on_requests(struct inode *inode, unsigned long idx_start, unsigned int spin_lock(&nfsi->req_lock); next = idx_start; - while (radix_tree_gang_lookup(&nfsi->nfs_page_tree, (void **)&req, next, 1)) { + while (radix_tree_gang_lookup_tag(&nfsi->nfs_page_tree, (void **)&req, next, 1, NFS_PAGE_TAG_WRITEBACK)) { if (req->wb_index > idx_end) break; next = req->wb_index + 1; - if (!NFS_WBACK_BUSY(req)) - continue; + BUG_ON(!NFS_WBACK_BUSY(req)); atomic_inc(&req->wb_count); spin_unlock(&nfsi->req_lock); @@ -821,7 +820,7 @@ out: #else nfs_inode_remove_request(req); #endif - nfs_unlock_request(req); + nfs_clear_page_writeback(req); } static inline int flush_task_priority(int how) @@ -952,7 +951,7 @@ out_bad: nfs_writedata_free(data); } nfs_mark_request_dirty(req); - nfs_unlock_request(req); + nfs_clear_page_writeback(req); return -ENOMEM; } @@ -1002,7 +1001,7 @@ static int nfs_flush_one(struct list_head *head, struct inode *inode, int how) struct nfs_page *req = nfs_list_entry(head->next); nfs_list_remove_request(req); nfs_mark_request_dirty(req); - nfs_unlock_request(req); + nfs_clear_page_writeback(req); } return -ENOMEM; } @@ -1029,7 +1028,7 @@ nfs_flush_list(struct list_head *head, int wpages, int how) req = nfs_list_entry(head->next); nfs_list_remove_request(req); nfs_mark_request_dirty(req); - nfs_unlock_request(req); + nfs_clear_page_writeback(req); } return error; } @@ -1121,7 +1120,7 @@ static void nfs_writeback_done_full(struct nfs_write_data *data, int status) nfs_inode_remove_request(req); #endif next: - nfs_unlock_request(req); + nfs_clear_page_writeback(req); } } @@ -1278,7 +1277,7 @@ nfs_commit_list(struct list_head *head, int how) req = nfs_list_entry(head->next); nfs_list_remove_request(req); nfs_mark_request_commit(req); - nfs_unlock_request(req); + nfs_clear_page_writeback(req); } return -ENOMEM; } @@ -1324,7 +1323,7 @@ nfs_commit_done(struct rpc_task *task) dprintk(" mismatch\n"); nfs_mark_request_dirty(req); next: - nfs_unlock_request(req); + nfs_clear_page_writeback(req); res++; } sub_page_state(nr_unstable,res); diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index 39e4895bcdb4..db40e4590ba2 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -19,6 +19,11 @@ #include +/* + * Valid flags for the radix tree + */ +#define NFS_PAGE_TAG_WRITEBACK 1 + /* * Valid flags for a dirty buffer */ @@ -62,6 +67,9 @@ extern int nfs_coalesce_requests(struct list_head *, struct list_head *, unsigned int); extern int nfs_wait_on_request(struct nfs_page *); extern void nfs_unlock_request(struct nfs_page *req); +extern int nfs_set_page_writeback_locked(struct nfs_page *req); +extern void nfs_clear_page_writeback(struct nfs_page *req); + /* * Lock the page of an asynchronous request without incrementing the wb_count @@ -96,10 +104,6 @@ nfs_list_remove_request(struct nfs_page *req) { if (list_empty(&req->wb_list)) return; - if (!NFS_WBACK_BUSY(req)) { - printk(KERN_ERR "NFS: unlocked request attempted removed from list!\n"); - BUG(); - } list_del_init(&req->wb_list); req->wb_list_head = NULL; } -- cgit v1.2.3 From 3da28eb1c6545fe73263a24eba0996217490e1eb Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:31 +0000 Subject: [PATCH] NFS: Replace nfs_page insertion sort with a radix sort Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 2 +- fs/nfs/pagelist.c | 86 +++++++++++++++++++++++++++++++----------------- fs/nfs/write.c | 71 ++++++++++++++++++--------------------- include/linux/nfs_fs.h | 4 +-- include/linux/nfs_page.h | 18 ++++++++-- 5 files changed, 107 insertions(+), 74 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 4f545f382ba6..4845911f1c63 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -135,7 +135,7 @@ nfs_write_inode(struct inode *inode, int sync) int flags = sync ? FLUSH_WAIT : 0; int ret; - ret = nfs_commit_inode(inode, 0, 0, flags); + ret = nfs_commit_inode(inode, flags); if (ret < 0) return ret; return 0; diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 356a33bb38a6..d53857b148e2 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -177,36 +177,6 @@ nfs_release_request(struct nfs_page *req) nfs_page_free(req); } -/** - * nfs_list_add_request - Insert a request into a sorted list - * @req: request - * @head: head of list into which to insert the request. - * - * Note that the wb_list is sorted by page index in order to facilitate - * coalescing of requests. - * We use an insertion sort that is optimized for the case of appended - * writes. - */ -void -nfs_list_add_request(struct nfs_page *req, struct list_head *head) -{ - struct list_head *pos; - -#ifdef NFS_PARANOIA - if (!list_empty(&req->wb_list)) { - printk(KERN_ERR "NFS: Add to list failed!\n"); - BUG(); - } -#endif - list_for_each_prev(pos, head) { - struct nfs_page *p = nfs_list_entry(pos); - if (p->wb_index < req->wb_index) - break; - } - list_add(&req->wb_list, pos); - req->wb_list_head = head; -} - static int nfs_wait_bit_interruptible(void *word) { int ret = 0; @@ -291,6 +261,62 @@ nfs_coalesce_requests(struct list_head *head, struct list_head *dst, return npages; } +#define NFS_SCAN_MAXENTRIES 16 +/** + * nfs_scan_lock_dirty - Scan the radix tree for dirty requests + * @nfsi: NFS inode + * @dst: Destination list + * @idx_start: lower bound of page->index to scan + * @npages: idx_start + npages sets the upper bound to scan. + * + * Moves elements from one of the inode request lists. + * If the number of requests is set to 0, the entire address_space + * starting at index idx_start, is scanned. + * The requests are *not* checked to ensure that they form a contiguous set. + * You must be holding the inode's req_lock when calling this function + */ +int +nfs_scan_lock_dirty(struct nfs_inode *nfsi, struct list_head *dst, + unsigned long idx_start, unsigned int npages) +{ + struct nfs_page *pgvec[NFS_SCAN_MAXENTRIES]; + struct nfs_page *req; + unsigned long idx_end; + int found, i; + int res; + + res = 0; + if (npages == 0) + idx_end = ~0; + else + idx_end = idx_start + npages - 1; + + for (;;) { + found = radix_tree_gang_lookup_tag(&nfsi->nfs_page_tree, + (void **)&pgvec[0], idx_start, NFS_SCAN_MAXENTRIES, + NFS_PAGE_TAG_DIRTY); + if (found <= 0) + break; + for (i = 0; i < found; i++) { + req = pgvec[i]; + if (req->wb_index > idx_end) + goto out; + + idx_start = req->wb_index + 1; + + if (nfs_set_page_writeback_locked(req)) { + radix_tree_tag_clear(&nfsi->nfs_page_tree, + req->wb_index, NFS_PAGE_TAG_DIRTY); + nfs_list_remove_request(req); + nfs_list_add_request(req, dst); + res++; + } + } + } +out: + return res; +} + /** * nfs_scan_list - Scan a list for matching requests * @head: One of the NFS inode request lists diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 58a39b0486a7..5130eda231d7 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -352,7 +352,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) if (err < 0) goto out; } - err = nfs_commit_inode(inode, 0, 0, wb_priority(wbc)); + err = nfs_commit_inode(inode, wb_priority(wbc)); if (err > 0) { wbc->nr_to_write -= err; err = 0; @@ -446,6 +446,8 @@ nfs_mark_request_dirty(struct nfs_page *req) struct nfs_inode *nfsi = NFS_I(inode); spin_lock(&nfsi->req_lock); + radix_tree_tag_set(&nfsi->nfs_page_tree, + req->wb_index, NFS_PAGE_TAG_DIRTY); nfs_list_add_request(req, &nfsi->dirty); nfsi->ndirty++; spin_unlock(&nfsi->req_lock); @@ -537,12 +539,15 @@ static int nfs_scan_dirty(struct inode *inode, struct list_head *dst, unsigned long idx_start, unsigned int npages) { struct nfs_inode *nfsi = NFS_I(inode); - int res; - res = nfs_scan_list(&nfsi->dirty, dst, idx_start, npages); - nfsi->ndirty -= res; - sub_page_state(nr_dirty,res); - if ((nfsi->ndirty == 0) != list_empty(&nfsi->dirty)) - printk(KERN_ERR "NFS: desynchronized value of nfs_i.ndirty.\n"); + int res = 0; + + if (nfsi->ndirty != 0) { + res = nfs_scan_lock_dirty(nfsi, dst, idx_start, npages); + nfsi->ndirty -= res; + sub_page_state(nr_dirty,res); + if ((nfsi->ndirty == 0) != list_empty(&nfsi->dirty)) + printk(KERN_ERR "NFS: desynchronized value of nfs_i.ndirty.\n"); + } return res; } @@ -561,11 +566,14 @@ static int nfs_scan_commit(struct inode *inode, struct list_head *dst, unsigned long idx_start, unsigned int npages) { struct nfs_inode *nfsi = NFS_I(inode); - int res; - res = nfs_scan_list(&nfsi->commit, dst, idx_start, npages); - nfsi->ncommit -= res; - if ((nfsi->ncommit == 0) != list_empty(&nfsi->commit)) - printk(KERN_ERR "NFS: desynchronized value of nfs_i.ncommit.\n"); + int res = 0; + + if (nfsi->ncommit != 0) { + res = nfs_scan_list(&nfsi->commit, dst, idx_start, npages); + nfsi->ncommit -= res; + if ((nfsi->ncommit == 0) != list_empty(&nfsi->commit)) + printk(KERN_ERR "NFS: desynchronized value of nfs_i.ncommit.\n"); + } return res; } #endif @@ -1209,36 +1217,24 @@ static void nfs_commit_rpcsetup(struct list_head *head, struct nfs_write_data *data, int how) { struct rpc_task *task = &data->task; - struct nfs_page *first, *last; + struct nfs_page *first; struct inode *inode; - loff_t start, end, len; /* Set up the RPC argument and reply structs * NB: take care not to mess about with data->commit et al. */ list_splice_init(head, &data->pages); first = nfs_list_entry(data->pages.next); - last = nfs_list_entry(data->pages.prev); inode = first->wb_context->dentry->d_inode; - /* - * Determine the offset range of requests in the COMMIT call. - * We rely on the fact that data->pages is an ordered list... - */ - start = req_offset(first); - end = req_offset(last) + last->wb_bytes; - len = end - start; - /* If 'len' is not a 32-bit quantity, pass '0' in the COMMIT call */ - if (end >= i_size_read(inode) || len < 0 || len > (~((u32)0) >> 1)) - len = 0; - data->inode = inode; data->cred = first->wb_context->cred; data->args.fh = NFS_FH(data->inode); - data->args.offset = start; - data->args.count = len; - data->res.count = len; + /* Note: we always request a commit of the entire inode */ + data->args.offset = 0; + data->args.count = 0; + data->res.count = 0; data->res.fattr = &data->fattr; data->res.verf = &data->verf; @@ -1357,8 +1353,7 @@ static int nfs_flush_inode(struct inode *inode, unsigned long idx_start, } #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) -int nfs_commit_inode(struct inode *inode, unsigned long idx_start, - unsigned int npages, int how) +int nfs_commit_inode(struct inode *inode, int how) { struct nfs_inode *nfsi = NFS_I(inode); LIST_HEAD(head); @@ -1366,15 +1361,13 @@ int nfs_commit_inode(struct inode *inode, unsigned long idx_start, error = 0; spin_lock(&nfsi->req_lock); - res = nfs_scan_commit(inode, &head, idx_start, npages); + res = nfs_scan_commit(inode, &head, 0, 0); + spin_unlock(&nfsi->req_lock); if (res) { - res += nfs_scan_commit(inode, &head, 0, 0); - spin_unlock(&nfsi->req_lock); error = nfs_commit_list(&head, how); - } else - spin_unlock(&nfsi->req_lock); - if (error < 0) - return error; + if (error < 0) + return error; + } return res; } #endif @@ -1396,7 +1389,7 @@ int nfs_sync_inode(struct inode *inode, unsigned long idx_start, error = nfs_flush_inode(inode, idx_start, npages, how); #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) if (error == 0) - error = nfs_commit_inode(inode, idx_start, npages, how); + error = nfs_commit_inode(inode, how); #endif } while (error > 0); return error; diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 2954e44ed498..8ea249110fb0 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -395,10 +395,10 @@ extern void nfs_commit_done(struct rpc_task *); */ extern int nfs_sync_inode(struct inode *, unsigned long, unsigned int, int); #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) -extern int nfs_commit_inode(struct inode *, unsigned long, unsigned int, int); +extern int nfs_commit_inode(struct inode *, int); #else static inline int -nfs_commit_inode(struct inode *inode, unsigned long idx_start, unsigned int npages, int how) +nfs_commit_inode(struct inode *inode, int how) { return 0; } diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index db40e4590ba2..da2e077b65e2 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -22,6 +22,7 @@ /* * Valid flags for the radix tree */ +#define NFS_PAGE_TAG_DIRTY 0 #define NFS_PAGE_TAG_WRITEBACK 1 /* @@ -31,6 +32,7 @@ #define PG_NEED_COMMIT 1 #define PG_NEED_RESCHED 2 +struct nfs_inode; struct nfs_page { struct list_head wb_list, /* Defines state of page: */ *wb_list_head; /* read/write/commit */ @@ -59,8 +61,8 @@ extern void nfs_clear_request(struct nfs_page *req); extern void nfs_release_request(struct nfs_page *req); -extern void nfs_list_add_request(struct nfs_page *, struct list_head *); - +extern int nfs_scan_lock_dirty(struct nfs_inode *nfsi, struct list_head *dst, + unsigned long idx_start, unsigned int npages); extern int nfs_scan_list(struct list_head *, struct list_head *, unsigned long, unsigned int); extern int nfs_coalesce_requests(struct list_head *, struct list_head *, @@ -94,6 +96,18 @@ nfs_lock_request(struct nfs_page *req) return 1; } +/** + * nfs_list_add_request - Insert a request into a list + * @req: request + * @head: head of list into which to insert the request. + */ +static inline void +nfs_list_add_request(struct nfs_page *req, struct list_head *head) +{ + list_add_tail(&req->wb_list, head); + req->wb_list_head = head; +} + /** * nfs_list_remove_request - Remove a request from its wb_list -- cgit v1.2.3 From ecdbf769b2cb8903e07cd482334c714d89fd1146 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:31 +0000 Subject: [PATCH] NLM: fix a client-side race on blocking locks. If the lock blocks, the server may send us a GRANTED message that races with the reply to our LOCK request. Make sure that we catch the GRANTED by queueing up our request on the nlm_blocked list before we send off the first LOCK rpc call. Signed-off-by: Trond Myklebust --- fs/lockd/clntlock.c | 99 ++++++++++++++++++++++++++------------------- fs/lockd/clntproc.c | 40 ++++++++++++++---- include/linux/lockd/lockd.h | 7 +++- 3 files changed, 96 insertions(+), 50 deletions(-) (limited to 'include/linux') diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index 44adb84183b6..006bb9e14579 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -42,23 +42,51 @@ struct nlm_wait { static LIST_HEAD(nlm_blocked); /* - * Block on a lock + * Queue up a lock for blocking so that the GRANTED request can see it */ -int -nlmclnt_block(struct nlm_host *host, struct file_lock *fl, u32 *statp) +int nlmclnt_prepare_block(struct nlm_rqst *req, struct nlm_host *host, struct file_lock *fl) +{ + struct nlm_wait *block; + + BUG_ON(req->a_block != NULL); + block = kmalloc(sizeof(*block), GFP_KERNEL); + if (block == NULL) + return -ENOMEM; + block->b_host = host; + block->b_lock = fl; + init_waitqueue_head(&block->b_wait); + block->b_status = NLM_LCK_BLOCKED; + + list_add(&block->b_list, &nlm_blocked); + req->a_block = block; + + return 0; +} + +void nlmclnt_finish_block(struct nlm_rqst *req) { - struct nlm_wait block, **head; - int err; - u32 pstate; + struct nlm_wait *block = req->a_block; - block.b_host = host; - block.b_lock = fl; - init_waitqueue_head(&block.b_wait); - block.b_status = NLM_LCK_BLOCKED; - list_add(&block.b_list, &nlm_blocked); + if (block == NULL) + return; + req->a_block = NULL; + list_del(&block->b_list); + kfree(block); +} - /* Remember pseudo nsm state */ - pstate = host->h_state; +/* + * Block on a lock + */ +long nlmclnt_block(struct nlm_rqst *req, long timeout) +{ + struct nlm_wait *block = req->a_block; + long ret; + + /* A borken server might ask us to block even if we didn't + * request it. Just say no! + */ + if (!req->a_args.block) + return -EAGAIN; /* Go to sleep waiting for GRANT callback. Some servers seem * to lose callbacks, however, so we're going to poll from @@ -68,23 +96,16 @@ nlmclnt_block(struct nlm_host *host, struct file_lock *fl, u32 *statp) * a 1 minute timeout would do. See the comment before * nlmclnt_lock for an explanation. */ - sleep_on_timeout(&block.b_wait, 30*HZ); + ret = wait_event_interruptible_timeout(block->b_wait, + block->b_status != NLM_LCK_BLOCKED, + timeout); - list_del(&block.b_list); - - if (!signalled()) { - *statp = block.b_status; - return 0; + if (block->b_status != NLM_LCK_BLOCKED) { + req->a_res.status = block->b_status; + block->b_status = NLM_LCK_BLOCKED; } - /* Okay, we were interrupted. Cancel the pending request - * unless the server has rebooted. - */ - if (pstate == host->h_state && (err = nlmclnt_cancel(host, fl)) < 0) - printk(KERN_NOTICE - "lockd: CANCEL call failed (errno %d)\n", -err); - - return -ERESTARTSYS; + return ret; } /* @@ -94,27 +115,23 @@ u32 nlmclnt_grant(struct nlm_lock *lock) { struct nlm_wait *block; + u32 res = nlm_lck_denied; /* * Look up blocked request based on arguments. * Warning: must not use cookie to match it! */ list_for_each_entry(block, &nlm_blocked, b_list) { - if (nlm_compare_locks(block->b_lock, &lock->fl)) - break; + if (nlm_compare_locks(block->b_lock, &lock->fl)) { + /* Alright, we found a lock. Set the return status + * and wake up the caller + */ + block->b_status = NLM_LCK_GRANTED; + wake_up(&block->b_wait); + res = nlm_granted; + } } - - /* Ooops, no blocked request found. */ - if (block == NULL) - return nlm_lck_denied; - - /* Alright, we found the lock. Set the return status and - * wake up the caller. - */ - block->b_status = NLM_LCK_GRANTED; - wake_up(&block->b_wait); - - return nlm_granted; + return res; } /* diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index a4407619b1f1..fd77ed1d710d 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -21,6 +21,7 @@ #define NLMDBG_FACILITY NLMDBG_CLIENT #define NLMCLNT_GRACE_WAIT (5*HZ) +#define NLMCLNT_POLL_TIMEOUT (30*HZ) static int nlmclnt_test(struct nlm_rqst *, struct file_lock *); static int nlmclnt_lock(struct nlm_rqst *, struct file_lock *); @@ -553,7 +554,8 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl) { struct nlm_host *host = req->a_host; struct nlm_res *resp = &req->a_res; - int status; + long timeout; + int status; if (!host->h_monitored && nsm_monitor(host) < 0) { printk(KERN_NOTICE "lockd: failed to monitor %s\n", @@ -562,15 +564,32 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl) goto out; } - do { - if ((status = nlmclnt_call(req, NLMPROC_LOCK)) >= 0) { - if (resp->status != NLM_LCK_BLOCKED) - break; - status = nlmclnt_block(host, fl, &resp->status); - } + if (req->a_args.block) { + status = nlmclnt_prepare_block(req, host, fl); if (status < 0) goto out; - } while (resp->status == NLM_LCK_BLOCKED && req->a_args.block); + } + for(;;) { + status = nlmclnt_call(req, NLMPROC_LOCK); + if (status < 0) + goto out_unblock; + if (resp->status != NLM_LCK_BLOCKED) + break; + /* Wait on an NLM blocking lock */ + timeout = nlmclnt_block(req, NLMCLNT_POLL_TIMEOUT); + /* Did a reclaimer thread notify us of a server reboot? */ + if (resp->status == NLM_LCK_DENIED_GRACE_PERIOD) + continue; + if (resp->status != NLM_LCK_BLOCKED) + break; + if (timeout >= 0) + continue; + /* We were interrupted. Send a CANCEL request to the server + * and exit + */ + status = (int)timeout; + goto out_unblock; + } if (resp->status == NLM_LCK_GRANTED) { fl->fl_u.nfs_fl.state = host->h_state; @@ -579,6 +598,11 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl) do_vfs_lock(fl); } status = nlm_stat_to_errno(resp->status); +out_unblock: + nlmclnt_finish_block(req); + /* Cancel the blocked request if it is still pending */ + if (resp->status == NLM_LCK_BLOCKED) + nlmclnt_cancel(host, fl); out: nlmclnt_release_lockargs(req); return status; diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h index 0d9d22578212..16d4e5a08e1d 100644 --- a/include/linux/lockd/lockd.h +++ b/include/linux/lockd/lockd.h @@ -72,6 +72,8 @@ struct nlm_lockowner { uint32_t pid; }; +struct nlm_wait; + /* * Memory chunk for NLM client RPC request. */ @@ -81,6 +83,7 @@ struct nlm_rqst { struct nlm_host * a_host; /* host handle */ struct nlm_args a_args; /* arguments */ struct nlm_res a_res; /* result */ + struct nlm_wait * a_block; char a_owner[NLMCLNT_OHSIZE]; }; @@ -142,7 +145,9 @@ extern unsigned long nlmsvc_timeout; * Lockd client functions */ struct nlm_rqst * nlmclnt_alloc_call(void); -int nlmclnt_block(struct nlm_host *, struct file_lock *, u32 *); +int nlmclnt_prepare_block(struct nlm_rqst *req, struct nlm_host *host, struct file_lock *fl); +void nlmclnt_finish_block(struct nlm_rqst *req); +long nlmclnt_block(struct nlm_rqst *req, long timeout); int nlmclnt_cancel(struct nlm_host *, struct file_lock *); u32 nlmclnt_grant(struct nlm_lock *); void nlmclnt_recovery(struct nlm_host *, u32); -- cgit v1.2.3 From 8d0a8a9d0ec790086c64d210af413ac351d89e35 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:32 +0000 Subject: [PATCH] NFSv4: Clean up nfs4 lock state accounting Ensure that lock owner structures are not released prematurely. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4_fs.h | 9 +-- fs/nfs/nfs4proc.c | 69 ++++++++---------- fs/nfs/nfs4state.c | 178 +++++++++++++++++++++-------------------------- include/linux/fs.h | 1 + include/linux/nfs_fs_i.h | 5 ++ 5 files changed, 118 insertions(+), 144 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 7c6f1d668fbd..ec1a22d7b876 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -128,6 +128,7 @@ struct nfs4_state_owner { struct nfs4_lock_state { struct list_head ls_locks; /* Other lock stateids */ + struct nfs4_state * ls_state; /* Pointer to open state */ fl_owner_t ls_owner; /* POSIX lock owner */ #define NFS_LOCK_INITIALIZED 1 int ls_flags; @@ -153,7 +154,7 @@ struct nfs4_state { unsigned long flags; /* Do we hold any locks? */ struct semaphore lock_sema; /* Serializes file locking operations */ - rwlock_t state_lock; /* Protects the lock_states list */ + spinlock_t state_lock; /* Protects the lock_states list */ nfs4_stateid stateid; @@ -225,12 +226,8 @@ extern void nfs4_close_state(struct nfs4_state *, mode_t); extern struct nfs4_state *nfs4_find_state(struct inode *, struct rpc_cred *, mode_t mode); extern void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp); extern void nfs4_schedule_state_recovery(struct nfs4_client *); -extern struct nfs4_lock_state *nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t); -extern struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t); -extern void nfs4_put_lock_state(struct nfs4_lock_state *state); +extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl); extern void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *ls); -extern void nfs4_notify_setlk(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *); -extern void nfs4_notify_unlck(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *); extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t); extern const nfs4_stateid zero_stateid; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index af80b5981486..0ddc20102d46 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2626,14 +2626,11 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock down_read(&clp->cl_sem); nlo.clientid = clp->cl_clientid; down(&state->lock_sema); - lsp = nfs4_find_lock_state(state, request->fl_owner); - if (lsp) - nlo.id = lsp->ls_id; - else { - spin_lock(&clp->cl_lock); - nlo.id = nfs4_alloc_lockowner_id(clp); - spin_unlock(&clp->cl_lock); - } + status = nfs4_set_lock_state(state, request); + if (status != 0) + goto out; + lsp = request->fl_u.nfs4_fl.owner; + nlo.id = lsp->ls_id; arg.u.lockt = &nlo; status = rpc_call_sync(server->client, &msg, 0); if (!status) { @@ -2654,8 +2651,7 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock request->fl_pid = 0; status = 0; } - if (lsp) - nfs4_put_lock_state(lsp); +out: up(&state->lock_sema); up_read(&clp->cl_sem); return status; @@ -2715,28 +2711,26 @@ static int _nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock }; struct nfs4_lock_state *lsp; struct nfs_locku_opargs luargs; - int status = 0; + int status; down_read(&clp->cl_sem); down(&state->lock_sema); - lsp = nfs4_find_lock_state(state, request->fl_owner); - if (!lsp) + status = nfs4_set_lock_state(state, request); + if (status != 0) goto out; + lsp = request->fl_u.nfs4_fl.owner; /* We might have lost the locks! */ - if ((lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0) { - luargs.seqid = lsp->ls_seqid; - memcpy(&luargs.stateid, &lsp->ls_stateid, sizeof(luargs.stateid)); - arg.u.locku = &luargs; - status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR); - nfs4_increment_lock_seqid(status, lsp); - } + if ((lsp->ls_flags & NFS_LOCK_INITIALIZED) == 0) + goto out; + luargs.seqid = lsp->ls_seqid; + memcpy(&luargs.stateid, &lsp->ls_stateid, sizeof(luargs.stateid)); + arg.u.locku = &luargs; + status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR); + nfs4_increment_lock_seqid(status, lsp); - if (status == 0) { + if (status == 0) memcpy(&lsp->ls_stateid, &res.u.stateid, sizeof(lsp->ls_stateid)); - nfs4_notify_unlck(state, request, lsp); - } - nfs4_put_lock_state(lsp); out: up(&state->lock_sema); if (status == 0) @@ -2762,7 +2756,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *r { struct inode *inode = state->inode; struct nfs_server *server = NFS_SERVER(inode); - struct nfs4_lock_state *lsp; + struct nfs4_lock_state *lsp = request->fl_u.nfs4_fl.owner; struct nfs_lockargs arg = { .fh = NFS_FH(inode), .type = nfs4_lck_type(cmd, request), @@ -2784,9 +2778,6 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *r }; int status; - lsp = nfs4_get_lock_state(state, request->fl_owner); - if (lsp == NULL) - return -ENOMEM; if (!(lsp->ls_flags & NFS_LOCK_INITIALIZED)) { struct nfs4_state_owner *owner = state->owner; struct nfs_open_to_lock otl = { @@ -2808,27 +2799,26 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *r * seqid mutating errors */ nfs4_increment_seqid(status, owner); up(&owner->so_sema); + if (status == 0) { + lsp->ls_flags |= NFS_LOCK_INITIALIZED; + lsp->ls_seqid++; + } } else { struct nfs_exist_lock el = { .seqid = lsp->ls_seqid, }; memcpy(&el.stateid, &lsp->ls_stateid, sizeof(el.stateid)); largs.u.exist_lock = ⪙ - largs.new_lock_owner = 0; arg.u.lock = &largs; status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR); + /* increment seqid on success, and * seqid mutating errors*/ + nfs4_increment_lock_seqid(status, lsp); } - /* increment seqid on success, and * seqid mutating errors*/ - nfs4_increment_lock_seqid(status, lsp); /* save the returned stateid. */ - if (status == 0) { + if (status == 0) memcpy(&lsp->ls_stateid, &res.u.stateid, sizeof(nfs4_stateid)); - lsp->ls_flags |= NFS_LOCK_INITIALIZED; - if (!reclaim) - nfs4_notify_setlk(state, request, lsp); - } else if (status == -NFS4ERR_DENIED) + else if (status == -NFS4ERR_DENIED) status = -EAGAIN; - nfs4_put_lock_state(lsp); return status; } @@ -2869,7 +2859,9 @@ static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock down_read(&clp->cl_sem); down(&state->lock_sema); - status = _nfs4_do_setlk(state, cmd, request, 0); + status = nfs4_set_lock_state(state, request); + if (status == 0) + status = _nfs4_do_setlk(state, cmd, request, 0); up(&state->lock_sema); if (status == 0) { /* Note: we always want to sleep here! */ @@ -2927,7 +2919,6 @@ nfs4_proc_lock(struct file *filp, int cmd, struct file_lock *request) if (signalled()) break; } while(status < 0); - return status; } diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 591ad1d51880..afe587d82f1e 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -360,7 +360,7 @@ nfs4_alloc_open_state(void) atomic_set(&state->count, 1); INIT_LIST_HEAD(&state->lock_states); init_MUTEX(&state->lock_sema); - rwlock_init(&state->state_lock); + spin_lock_init(&state->state_lock); return state; } @@ -542,16 +542,6 @@ __nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner) return NULL; } -struct nfs4_lock_state * -nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner) -{ - struct nfs4_lock_state *lsp; - read_lock(&state->state_lock); - lsp = __nfs4_find_lock_state(state, fl_owner); - read_unlock(&state->state_lock); - return lsp; -} - /* * Return a compatible lock_state. If no initialized lock_state structure * exists, return an uninitialized one. @@ -568,14 +558,13 @@ static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, f return NULL; lsp->ls_flags = 0; lsp->ls_seqid = 0; /* arbitrary */ - lsp->ls_id = -1; memset(lsp->ls_stateid.data, 0, sizeof(lsp->ls_stateid.data)); atomic_set(&lsp->ls_count, 1); lsp->ls_owner = fl_owner; - INIT_LIST_HEAD(&lsp->ls_locks); spin_lock(&clp->cl_lock); lsp->ls_id = nfs4_alloc_lockowner_id(clp); spin_unlock(&clp->cl_lock); + INIT_LIST_HEAD(&lsp->ls_locks); return lsp; } @@ -585,121 +574,112 @@ static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, f * * The caller must be holding state->lock_sema and clp->cl_sem */ -struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner) +static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner) { - struct nfs4_lock_state * lsp; + struct nfs4_lock_state *lsp, *new = NULL; - lsp = nfs4_find_lock_state(state, owner); - if (lsp == NULL) - lsp = nfs4_alloc_lock_state(state, owner); + for(;;) { + spin_lock(&state->state_lock); + lsp = __nfs4_find_lock_state(state, owner); + if (lsp != NULL) + break; + if (new != NULL) { + new->ls_state = state; + list_add(&new->ls_locks, &state->lock_states); + set_bit(LK_STATE_IN_USE, &state->flags); + lsp = new; + new = NULL; + break; + } + spin_unlock(&state->state_lock); + new = nfs4_alloc_lock_state(state, owner); + if (new == NULL) + return NULL; + } + spin_unlock(&state->state_lock); + kfree(new); return lsp; } /* - * Byte-range lock aware utility to initialize the stateid of read/write - * requests. + * Release reference to lock_state, and free it if we see that + * it is no longer in use */ -void -nfs4_copy_stateid(nfs4_stateid *dst, struct nfs4_state *state, fl_owner_t fl_owner) +static void nfs4_put_lock_state(struct nfs4_lock_state *lsp) { - if (test_bit(LK_STATE_IN_USE, &state->flags)) { - struct nfs4_lock_state *lsp; + struct nfs4_state *state; - lsp = nfs4_find_lock_state(state, fl_owner); - if (lsp) { - memcpy(dst, &lsp->ls_stateid, sizeof(*dst)); - nfs4_put_lock_state(lsp); - return; - } - } - memcpy(dst, &state->stateid, sizeof(*dst)); + if (lsp == NULL) + return; + state = lsp->ls_state; + if (!atomic_dec_and_lock(&lsp->ls_count, &state->state_lock)) + return; + list_del(&lsp->ls_locks); + if (list_empty(&state->lock_states)) + clear_bit(LK_STATE_IN_USE, &state->flags); + spin_unlock(&state->state_lock); + kfree(lsp); } -/* -* Called with state->lock_sema and clp->cl_sem held. -*/ -void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *lsp) +static void nfs4_fl_copy_lock(struct file_lock *dst, struct file_lock *src) { - if (status == NFS_OK || seqid_mutating_err(-status)) - lsp->ls_seqid++; -} + struct nfs4_lock_state *lsp = src->fl_u.nfs4_fl.owner; -/* -* Check to see if the request lock (type FL_UNLK) effects the fl lock. -* -* fl and request must have the same posix owner -* -* return: -* 0 -> fl not effected by request -* 1 -> fl consumed by request -*/ + dst->fl_u.nfs4_fl.owner = lsp; + atomic_inc(&lsp->ls_count); +} -static int -nfs4_check_unlock(struct file_lock *fl, struct file_lock *request) +static void nfs4_fl_release_lock(struct file_lock *fl) { - if (fl->fl_start >= request->fl_start && fl->fl_end <= request->fl_end) - return 1; - return 0; + nfs4_put_lock_state(fl->fl_u.nfs4_fl.owner); } -/* - * Post an initialized lock_state on the state->lock_states list. - */ -void nfs4_notify_setlk(struct nfs4_state *state, struct file_lock *request, struct nfs4_lock_state *lsp) +static struct file_lock_operations nfs4_fl_lock_ops = { + .fl_copy_lock = nfs4_fl_copy_lock, + .fl_release_private = nfs4_fl_release_lock, +}; + +int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl) { - if (!list_empty(&lsp->ls_locks)) - return; - atomic_inc(&lsp->ls_count); - write_lock(&state->state_lock); - list_add(&lsp->ls_locks, &state->lock_states); - set_bit(LK_STATE_IN_USE, &state->flags); - write_unlock(&state->state_lock); + struct nfs4_lock_state *lsp; + + if (fl->fl_ops != NULL) + return 0; + lsp = nfs4_get_lock_state(state, fl->fl_owner); + if (lsp == NULL) + return -ENOMEM; + fl->fl_u.nfs4_fl.owner = lsp; + fl->fl_ops = &nfs4_fl_lock_ops; + return 0; } -/* - * to decide to 'reap' lock state: - * 1) search i_flock for file_locks with fl.lock_state = to ls. - * 2) determine if unlock will consume found lock. - * if so, reap - * - * else, don't reap. - * +/* + * Byte-range lock aware utility to initialize the stateid of read/write + * requests. */ -void -nfs4_notify_unlck(struct nfs4_state *state, struct file_lock *request, struct nfs4_lock_state *lsp) +void nfs4_copy_stateid(nfs4_stateid *dst, struct nfs4_state *state, fl_owner_t fl_owner) { - struct inode *inode = state->inode; - struct file_lock *fl; + struct nfs4_lock_state *lsp; - for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) { - if (!(fl->fl_flags & FL_POSIX)) - continue; - if (fl->fl_owner != lsp->ls_owner) - continue; - /* Exit if we find at least one lock which is not consumed */ - if (nfs4_check_unlock(fl,request) == 0) - return; - } + memcpy(dst, &state->stateid, sizeof(*dst)); + if (test_bit(LK_STATE_IN_USE, &state->flags) == 0) + return; - write_lock(&state->state_lock); - list_del_init(&lsp->ls_locks); - if (list_empty(&state->lock_states)) - clear_bit(LK_STATE_IN_USE, &state->flags); - write_unlock(&state->state_lock); + spin_lock(&state->state_lock); + lsp = __nfs4_find_lock_state(state, fl_owner); + if (lsp != NULL && (lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0) + memcpy(dst, &lsp->ls_stateid, sizeof(*dst)); + spin_unlock(&state->state_lock); nfs4_put_lock_state(lsp); } /* - * Release reference to lock_state, and free it if we see that - * it is no longer in use - */ -void -nfs4_put_lock_state(struct nfs4_lock_state *lsp) +* Called with state->lock_sema and clp->cl_sem held. +*/ +void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *lsp) { - if (!atomic_dec_and_test(&lsp->ls_count)) - return; - BUG_ON (!list_empty(&lsp->ls_locks)); - kfree(lsp); + if (status == NFS_OK || seqid_mutating_err(-status)) + lsp->ls_seqid++; } /* diff --git a/include/linux/fs.h b/include/linux/fs.h index 9b8b696d4f15..e5a8db00df29 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -674,6 +674,7 @@ struct file_lock { struct lock_manager_operations *fl_lmops; /* Callbacks for lockmanagers */ union { struct nfs_lock_info nfs_fl; + struct nfs4_lock_info nfs4_fl; } fl_u; }; diff --git a/include/linux/nfs_fs_i.h b/include/linux/nfs_fs_i.h index e9a749588a7b..e2c18dabff86 100644 --- a/include/linux/nfs_fs_i.h +++ b/include/linux/nfs_fs_i.h @@ -16,6 +16,11 @@ struct nfs_lock_info { struct nlm_lockowner *owner; }; +struct nfs4_lock_state; +struct nfs4_lock_info { + struct nfs4_lock_state *owner; +}; + /* * Lock flag values */ -- cgit v1.2.3 From 6ca4f65e6b390d09e1de7280cf9fd4f5d8e4b48b Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Wed, 22 Jun 2005 22:04:55 -0700 Subject: [NETPOLL]: Set poll_owner to -1 before unlocking in netpoll_poll_unlock() This trivial patch moves the assignment of poll_owner to -1 inside of the lock. This fixes a potential SMP race in the code. Signed-off-by: Jeff Moyer Signed-off-by: David S. Miller --- include/linux/netpoll.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index c0d8b90c5202..449a4fde6587 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -53,8 +53,8 @@ static inline void netpoll_poll_lock(struct net_device *dev) static inline void netpoll_poll_unlock(struct net_device *dev) { if (dev->np) { - spin_unlock(&dev->np->poll_lock); dev->np->poll_owner = -1; + spin_unlock(&dev->np->poll_lock); } } -- cgit v1.2.3 From 115c1d6e61b70851d9a363328c3b8d4c2559a1d3 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Wed, 22 Jun 2005 22:05:31 -0700 Subject: [NETPOLL]: Introduce a netpoll_info struct This patch introduces a netpoll_info structure, which the struct net_device will now point to instead of pointing to a struct netpoll. The reason for this is two-fold: 1) fields such as the rx_flags, poll_owner, and poll_lock should be maintained per net_device, not per netpoll; and 2) this is a first step in providing support for multiple netpoll clients to register against the same net_device. The struct netpoll is now pointed to by the netpoll_info structure. As such, the previous behaviour of the code is preserved. Signed-off-by: Jeff Moyer Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 ++-- include/linux/netpoll.h | 25 ++++++++++++++------- net/core/netpoll.c | 57 +++++++++++++++++++++++++++++++---------------- 3 files changed, 57 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ba5d1236aa17..d6afd440cf7b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -41,7 +41,7 @@ struct divert_blk; struct vlan_group; struct ethtool_ops; -struct netpoll; +struct netpoll_info; /* source back-compat hooks */ #define SET_ETHTOOL_OPS(netdev,ops) \ ( (netdev)->ethtool_ops = (ops) ) @@ -468,7 +468,7 @@ struct net_device unsigned char *haddr); int (*neigh_setup)(struct net_device *dev, struct neigh_parms *); #ifdef CONFIG_NETPOLL - struct netpoll *np; + struct netpoll_info *npinfo; #endif #ifdef CONFIG_NET_POLL_CONTROLLER void (*poll_controller)(struct net_device *dev); diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 449a4fde6587..388cd91bc7a6 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -16,14 +16,18 @@ struct netpoll; struct netpoll { struct net_device *dev; char dev_name[16], *name; - int rx_flags; void (*rx_hook)(struct netpoll *, int, char *, int); void (*drop)(struct sk_buff *skb); u32 local_ip, remote_ip; u16 local_port, remote_port; unsigned char local_mac[6], remote_mac[6]; +}; + +struct netpoll_info { spinlock_t poll_lock; int poll_owner; + int rx_flags; + struct netpoll *np; }; void netpoll_poll(struct netpoll *np); @@ -39,22 +43,27 @@ void netpoll_queue(struct sk_buff *skb); #ifdef CONFIG_NETPOLL static inline int netpoll_rx(struct sk_buff *skb) { - return skb->dev->np && skb->dev->np->rx_flags && __netpoll_rx(skb); + struct netpoll_info *npinfo = skb->dev->npinfo; + + if (!npinfo || !npinfo->rx_flags) + return 0; + + return npinfo->np && __netpoll_rx(skb); } static inline void netpoll_poll_lock(struct net_device *dev) { - if (dev->np) { - spin_lock(&dev->np->poll_lock); - dev->np->poll_owner = smp_processor_id(); + if (dev->npinfo) { + spin_lock(&dev->npinfo->poll_lock); + dev->npinfo->poll_owner = smp_processor_id(); } } static inline void netpoll_poll_unlock(struct net_device *dev) { - if (dev->np) { - dev->np->poll_owner = -1; - spin_unlock(&dev->np->poll_lock); + if (dev->npinfo) { + dev->npinfo->poll_owner = -1; + spin_unlock(&dev->npinfo->poll_lock); } } diff --git a/net/core/netpoll.c b/net/core/netpoll.c index a119696d5521..ab3c0c9713b0 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -130,19 +130,20 @@ static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, */ static void poll_napi(struct netpoll *np) { + struct netpoll_info *npinfo = np->dev->npinfo; int budget = 16; if (test_bit(__LINK_STATE_RX_SCHED, &np->dev->state) && - np->poll_owner != smp_processor_id() && - spin_trylock(&np->poll_lock)) { - np->rx_flags |= NETPOLL_RX_DROP; + npinfo->poll_owner != smp_processor_id() && + spin_trylock(&npinfo->poll_lock)) { + npinfo->rx_flags |= NETPOLL_RX_DROP; atomic_inc(&trapped); np->dev->poll(np->dev, &budget); atomic_dec(&trapped); - np->rx_flags &= ~NETPOLL_RX_DROP; - spin_unlock(&np->poll_lock); + npinfo->rx_flags &= ~NETPOLL_RX_DROP; + spin_unlock(&npinfo->poll_lock); } } @@ -245,6 +246,7 @@ repeat: static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) { int status; + struct netpoll_info *npinfo; repeat: if(!np || !np->dev || !netif_running(np->dev)) { @@ -253,8 +255,9 @@ repeat: } /* avoid recursion */ - if(np->poll_owner == smp_processor_id() || - np->dev->xmit_lock_owner == smp_processor_id()) { + npinfo = np->dev->npinfo; + if (npinfo->poll_owner == smp_processor_id() || + np->dev->xmit_lock_owner == smp_processor_id()) { if (np->drop) np->drop(skb); else @@ -341,14 +344,18 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len) static void arp_reply(struct sk_buff *skb) { + struct netpoll_info *npinfo = skb->dev->npinfo; struct arphdr *arp; unsigned char *arp_ptr; int size, type = ARPOP_REPLY, ptype = ETH_P_ARP; u32 sip, tip; struct sk_buff *send_skb; - struct netpoll *np = skb->dev->np; + struct netpoll *np = NULL; - if (!np) return; + if (npinfo) + np = npinfo->np; + if (!np) + return; /* No arp on this interface */ if (skb->dev->flags & IFF_NOARP) @@ -429,7 +436,7 @@ int __netpoll_rx(struct sk_buff *skb) int proto, len, ulen; struct iphdr *iph; struct udphdr *uh; - struct netpoll *np = skb->dev->np; + struct netpoll *np = skb->dev->npinfo->np; if (!np->rx_hook) goto out; @@ -611,9 +618,7 @@ int netpoll_setup(struct netpoll *np) { struct net_device *ndev = NULL; struct in_device *in_dev; - - np->poll_lock = SPIN_LOCK_UNLOCKED; - np->poll_owner = -1; + struct netpoll_info *npinfo; if (np->dev_name) ndev = dev_get_by_name(np->dev_name); @@ -624,7 +629,16 @@ int netpoll_setup(struct netpoll *np) } np->dev = ndev; - ndev->np = np; + if (!ndev->npinfo) { + npinfo = kmalloc(sizeof(*npinfo), GFP_KERNEL); + if (!npinfo) + goto release; + + npinfo->np = NULL; + npinfo->poll_lock = SPIN_LOCK_UNLOCKED; + npinfo->poll_owner = -1; + } else + npinfo = ndev->npinfo; if (!ndev->poll_controller) { printk(KERN_ERR "%s: %s doesn't support polling, aborting.\n", @@ -693,12 +707,15 @@ int netpoll_setup(struct netpoll *np) } if(np->rx_hook) - np->rx_flags = NETPOLL_RX_ENABLED; + npinfo->rx_flags = NETPOLL_RX_ENABLED; + npinfo->np = np; + ndev->npinfo = npinfo; return 0; release: - ndev->np = NULL; + if (!ndev->npinfo) + kfree(npinfo); np->dev = NULL; dev_put(ndev); return -1; @@ -706,9 +723,11 @@ int netpoll_setup(struct netpoll *np) void netpoll_cleanup(struct netpoll *np) { - if (np->dev) - np->dev->np = NULL; - dev_put(np->dev); + if (np->dev) { + if (np->dev->npinfo) + np->dev->npinfo->np = NULL; + dev_put(np->dev); + } np->dev = NULL; } -- cgit v1.2.3 From fbeec2e1552949002065435c9829dc244ad85407 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Wed, 22 Jun 2005 22:05:59 -0700 Subject: [NETPOLL]: allow multiple netpoll_clients to register against one interface This patch provides support for registering multiple netpoll clients to the same network device. Only one of these clients may register an rx_hook, however. In practice, this restriction has not been problematic. It is worth mentioning, though, that the current design can be easily extended to allow for the registration of multiple rx_hooks. The basic idea of the patch is that the rx_np pointer in the netpoll_info structure points to the struct netpoll that has rx_hook filled in. Aside from this one case, there is no need for a pointer from the struct net_device to an individual struct netpoll. A lock is introduced to protect the setting and clearing of the np_rx pointer. The pointer will only be cleared upon netpoll client module removal, and the lock should be uncontested. Signed-off-by: Jeff Moyer Signed-off-by: David S. Miller --- include/linux/netpoll.h | 15 ++++++++++++--- net/core/netpoll.c | 39 +++++++++++++++++++++++++++++---------- 2 files changed, 41 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 388cd91bc7a6..bcd0ac33f592 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -27,7 +27,8 @@ struct netpoll_info { spinlock_t poll_lock; int poll_owner; int rx_flags; - struct netpoll *np; + spinlock_t rx_lock; + struct netpoll *rx_np; /* netpoll that registered an rx_hook */ }; void netpoll_poll(struct netpoll *np); @@ -44,11 +45,19 @@ void netpoll_queue(struct sk_buff *skb); static inline int netpoll_rx(struct sk_buff *skb) { struct netpoll_info *npinfo = skb->dev->npinfo; + unsigned long flags; + int ret = 0; - if (!npinfo || !npinfo->rx_flags) + if (!npinfo || (!npinfo->rx_np && !npinfo->rx_flags)) return 0; - return npinfo->np && __netpoll_rx(skb); + spin_lock_irqsave(&npinfo->rx_lock, flags); + /* check rx_flags again with the lock held */ + if (npinfo->rx_flags && __netpoll_rx(skb)) + ret = 1; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + + return ret; } static inline void netpoll_poll_lock(struct net_device *dev) diff --git a/net/core/netpoll.c b/net/core/netpoll.c index ab3c0c9713b0..c327c9edadc5 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -349,11 +349,15 @@ static void arp_reply(struct sk_buff *skb) unsigned char *arp_ptr; int size, type = ARPOP_REPLY, ptype = ETH_P_ARP; u32 sip, tip; + unsigned long flags; struct sk_buff *send_skb; struct netpoll *np = NULL; - if (npinfo) - np = npinfo->np; + spin_lock_irqsave(&npinfo->rx_lock, flags); + if (npinfo->rx_np && npinfo->rx_np->dev == skb->dev) + np = npinfo->rx_np; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + if (!np) return; @@ -436,9 +440,9 @@ int __netpoll_rx(struct sk_buff *skb) int proto, len, ulen; struct iphdr *iph; struct udphdr *uh; - struct netpoll *np = skb->dev->npinfo->np; + struct netpoll *np = skb->dev->npinfo->rx_np; - if (!np->rx_hook) + if (!np) goto out; if (skb->dev->type != ARPHRD_ETHER) goto out; @@ -619,6 +623,7 @@ int netpoll_setup(struct netpoll *np) struct net_device *ndev = NULL; struct in_device *in_dev; struct netpoll_info *npinfo; + unsigned long flags; if (np->dev_name) ndev = dev_get_by_name(np->dev_name); @@ -634,9 +639,10 @@ int netpoll_setup(struct netpoll *np) if (!npinfo) goto release; - npinfo->np = NULL; + npinfo->rx_np = NULL; npinfo->poll_lock = SPIN_LOCK_UNLOCKED; npinfo->poll_owner = -1; + npinfo->rx_lock = SPIN_LOCK_UNLOCKED; } else npinfo = ndev->npinfo; @@ -706,9 +712,13 @@ int netpoll_setup(struct netpoll *np) np->name, HIPQUAD(np->local_ip)); } - if(np->rx_hook) - npinfo->rx_flags = NETPOLL_RX_ENABLED; - npinfo->np = np; + if (np->rx_hook) { + spin_lock_irqsave(&npinfo->rx_lock, flags); + npinfo->rx_flags |= NETPOLL_RX_ENABLED; + npinfo->rx_np = np; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + } + /* last thing to do is link it to the net device structure */ ndev->npinfo = npinfo; return 0; @@ -723,11 +733,20 @@ int netpoll_setup(struct netpoll *np) void netpoll_cleanup(struct netpoll *np) { + struct netpoll_info *npinfo; + unsigned long flags; + if (np->dev) { - if (np->dev->npinfo) - np->dev->npinfo->np = NULL; + npinfo = np->dev->npinfo; + if (npinfo && npinfo->rx_np == np) { + spin_lock_irqsave(&npinfo->rx_lock, flags); + npinfo->rx_np = NULL; + npinfo->rx_flags &= ~NETPOLL_RX_ENABLED; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + } dev_put(np->dev); } + np->dev = NULL; } -- cgit v1.2.3 From cb65d506c34c86df5bcef939ce5a8666a451bd8b Mon Sep 17 00:00:00 2001 From: Shaun Pereira Date: Wed, 22 Jun 2005 22:15:01 -0700 Subject: [X25]: Selective sub-address matching with call user data. From: Shaun Pereira This is the first (independent of the second) patch of two that I am working on with x25 on linux (tested with xot on a cisco router). Details are as follows. Current state of module: A server using the current implementation (2.6.11.7) of the x25 module will accept a call request/ incoming call packet at the listening x.25 address, from all callers to that address, as long as NO call user data is present in the packet header. If the server needs to choose to accept a particular call request/ incoming call packet arriving at its listening x25 address, then the kernel has to allow a match of call user data present in the call request packet with its own. This is required when multiple servers listen at the same x25 address and device interface. The kernel currently matches ALL call user data, if present. Current Changes: This patch is a follow up to the patch submitted previously by Andrew Hendry, and allows the user to selectively control the number of octets of call user data in the call request packet, that the kernel will match. By default no call user data is matched, even if call user data is present. To allow call user data matching, a cudmatchlength > 0 has to be passed into the kernel after which the passed number of octets will be matched. Otherwise the kernel behavior is exactly as the original implementation. This patch also ensures that as is normally the case, no call user data will be present in the Call accepted / call connected packet sent back to the caller Future Changes on next patch: There are cases however when call user data may be present in the call accepted packet. According to the X.25 recommendation (ITU-T 10/96) section 5.2.3.2 call user data may be present in the call accepted packet provided the fast select facility is used. My next patch will include this fast select utility and the ability to send up to 128 octets call user data in the call accepted packet provided the fast select facility is used. I am currently testing this, again with xot on linux and cisco. Signed-off-by: Shaun Pereira (With a fix from Alexey Dobriyan ) Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- include/linux/x25.h | 10 ++++++++ include/net/x25.h | 3 +-- net/x25/af_x25.c | 73 +++++++++++++++++++++++++++++++++++------------------ net/x25/x25_subr.c | 18 ------------- 4 files changed, 59 insertions(+), 45 deletions(-) (limited to 'include/linux') diff --git a/include/linux/x25.h b/include/linux/x25.h index 7531cfed5885..6f43b3d20248 100644 --- a/include/linux/x25.h +++ b/include/linux/x25.h @@ -4,6 +4,8 @@ * History * mar/20/00 Daniela Squassoni Disabling/enabling of facilities * negotiation. + * apr/02/05 Shaun Pereira Selective sub address matching with + * call user data */ #ifndef X25_KERNEL_H @@ -16,6 +18,7 @@ #define SIOCX25GCALLUSERDATA (SIOCPROTOPRIVATE + 4) #define SIOCX25SCALLUSERDATA (SIOCPROTOPRIVATE + 5) #define SIOCX25GCAUSEDIAG (SIOCPROTOPRIVATE + 6) +#define SIOCX25SCUDMATCHLEN (SIOCPROTOPRIVATE + 7) /* * Values for {get,set}sockopt. @@ -109,4 +112,11 @@ struct x25_causediag { unsigned char diagnostic; }; +/* + * Further optional call user data match length selection + */ +struct x25_subaddr { + unsigned int cudmatchlength; +}; + #endif diff --git a/include/net/x25.h b/include/net/x25.h index 7a1ba5bbb868..9dd70dd4a9b7 100644 --- a/include/net/x25.h +++ b/include/net/x25.h @@ -134,7 +134,7 @@ struct x25_sock { struct sock sk; struct x25_address source_addr, dest_addr; struct x25_neigh *neighbour; - unsigned int lci; + unsigned int lci, cudmatchlength; unsigned char state, condition, qbitincl, intflag; unsigned short vs, vr, va, vl; unsigned long t2, t21, t22, t23; @@ -242,7 +242,6 @@ extern int x25_validate_nr(struct sock *, unsigned short); extern void x25_write_internal(struct sock *, int); extern int x25_decode(struct sock *, struct sk_buff *, int *, int *, int *, int *, int *); extern void x25_disconnect(struct sock *, int, unsigned char, unsigned char); -extern int x25_check_calluserdata(struct x25_calluserdata *,struct x25_calluserdata *); /* x25_timer.c */ extern void x25_start_heartbeat(struct sock *); diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index 2a24b243b841..e17d84a55d5e 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -29,6 +29,8 @@ * 2000-11-14 Henner Eisen Closing datalink from NETDEV_GOING_DOWN * 2002-10-06 Arnaldo C. Melo Get rid of cli/sti, move proc stuff to * x25_proc.c, using seq_file + * 2005-04-02 Shaun Pereira Selective sub address matching + * with call user data */ #include @@ -219,7 +221,8 @@ static void x25_insert_socket(struct sock *sk) * Note: if a listening socket has cud set it must only get calls * with matching cud. */ -static struct sock *x25_find_listener(struct x25_address *addr, struct x25_calluserdata *calluserdata) +static struct sock *x25_find_listener(struct x25_address *addr, + struct sk_buff *skb) { struct sock *s; struct sock *next_best; @@ -230,22 +233,23 @@ static struct sock *x25_find_listener(struct x25_address *addr, struct x25_callu sk_for_each(s, node, &x25_list) if ((!strcmp(addr->x25_addr, - x25_sk(s)->source_addr.x25_addr) || - !strcmp(addr->x25_addr, - null_x25_address.x25_addr)) && - s->sk_state == TCP_LISTEN) { - + x25_sk(s)->source_addr.x25_addr) || + !strcmp(addr->x25_addr, + null_x25_address.x25_addr)) && + s->sk_state == TCP_LISTEN) { /* * Found a listening socket, now check the incoming * call user data vs this sockets call user data */ - if (x25_check_calluserdata(&x25_sk(s)->calluserdata, calluserdata)) { - sock_hold(s); - goto found; - } - if (x25_sk(s)->calluserdata.cudlength == 0) { + if(skb->len > 0 && x25_sk(s)->cudmatchlength > 0) { + if((memcmp(x25_sk(s)->calluserdata.cuddata, + skb->data, + x25_sk(s)->cudmatchlength)) == 0) { + sock_hold(s); + goto found; + } + } else next_best = s; - } } if (next_best) { s = next_best; @@ -497,6 +501,7 @@ static int x25_create(struct socket *sock, int protocol) x25->t23 = sysctl_x25_clear_request_timeout; x25->t2 = sysctl_x25_ack_holdback_timeout; x25->state = X25_STATE_0; + x25->cudmatchlength = 0; x25->facilities.winsize_in = X25_DEFAULT_WINDOW_SIZE; x25->facilities.winsize_out = X25_DEFAULT_WINDOW_SIZE; @@ -545,6 +550,7 @@ static struct sock *x25_make_new(struct sock *osk) x25->t2 = ox25->t2; x25->facilities = ox25->facilities; x25->qbitincl = ox25->qbitincl; + x25->cudmatchlength = ox25->cudmatchlength; x25_init_timers(sk); out: @@ -822,7 +828,6 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb, struct x25_sock *makex25; struct x25_address source_addr, dest_addr; struct x25_facilities facilities; - struct x25_calluserdata calluserdata; int len, rc; /* @@ -844,20 +849,11 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb, len = skb->data[0] + 1; skb_pull(skb,len); - /* - * Incoming Call User Data. - */ - if (skb->len >= 0) { - memcpy(calluserdata.cuddata, skb->data, skb->len); - calluserdata.cudlength = skb->len; - } - - skb_push(skb,len); - /* * Find a listener for the particular address/cud pair. */ - sk = x25_find_listener(&source_addr,&calluserdata); + sk = x25_find_listener(&source_addr,skb); + skb_push(skb,len); /* * We can't accept the Call Request. @@ -900,12 +896,22 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb, makex25->neighbour = nb; makex25->facilities = facilities; makex25->vc_facil_mask = x25_sk(sk)->vc_facil_mask; - makex25->calluserdata = calluserdata; + /* ensure no reverse facil on accept */ + makex25->vc_facil_mask &= ~X25_MASK_REVERSE; + makex25->cudmatchlength = x25_sk(sk)->cudmatchlength; x25_write_internal(make, X25_CALL_ACCEPTED); makex25->state = X25_STATE_3; + /* + * Incoming Call User Data. + */ + if (skb->len >= 0) { + memcpy(makex25->calluserdata.cuddata, skb->data, skb->len); + makex25->calluserdata.cudlength = skb->len; + } + sk->sk_ack_backlog++; x25_insert_socket(make); @@ -1325,6 +1331,23 @@ static int x25_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) break; } + case SIOCX25SCUDMATCHLEN: { + struct x25_subaddr sub_addr; + rc = -EINVAL; + if(sk->sk_state != TCP_CLOSE) + break; + rc = -EFAULT; + if (copy_from_user(&sub_addr, argp, + sizeof(sub_addr))) + break; + rc = -EINVAL; + if(sub_addr.cudmatchlength > X25_MAX_CUD_LEN) + break; + x25->cudmatchlength = sub_addr.cudmatchlength; + rc = 0; + break; + } + default: rc = dev_ioctl(cmd, argp); break; diff --git a/net/x25/x25_subr.c b/net/x25/x25_subr.c index 183fea3bba67..c349bbd61684 100644 --- a/net/x25/x25_subr.c +++ b/net/x25/x25_subr.c @@ -354,21 +354,3 @@ void x25_check_rbuf(struct sock *sk) } } -/* - * Compare 2 calluserdata structures, used to find correct listening sockets - * when call user data is used. - */ -int x25_check_calluserdata(struct x25_calluserdata *ours, struct x25_calluserdata *theirs) -{ - int i; - if (ours->cudlength != theirs->cudlength) - return 0; - - for (i=0;icudlength;i++) { - if (ours->cuddata[i] != theirs->cuddata[i]) { - return 0; - } - } - return 1; -} - -- cgit v1.2.3 From ebc3f64b864fc16a594c2e63bf55a55c7d42084b Mon Sep 17 00:00:00 2001 From: Shaun Pereira Date: Wed, 22 Jun 2005 22:16:17 -0700 Subject: [X25]: Fast select with no restriction on response This patch is a follow up to patch 1 regarding "Selective Sub Address matching with call user data". It allows use of the Fast-Select-Acceptance optional user facility for X.25. This patch just implements fast select with no restriction on response (NRR). What this means (according to ITU-T Recomendation 10/96 section 6.16) is that if in an incoming call packet, the relevant facility bits are set for fast-select-NRR, then the called DTE can issue a direct response to the incoming packet using a call-accepted packet that contains call-user-data. This patch allows such a response. The called DTE can also respond with a clear-request packet that contains call-user-data. However, this feature is currently not implemented by the patch. How is Fast Select Acceptance used? By default, the system does not allow fast select acceptance (as before). To enable a response to fast select acceptance, After a listen socket in created and bound as follows socket(AF_X25, SOCK_SEQPACKET, 0); bind(call_soc, (struct sockaddr *)&locl_addr, sizeof(locl_addr)); but before a listen system call is made, the following ioctl should be used. ioctl(call_soc,SIOCX25CALLACCPTAPPRV); Now the listen system call can be made listen(call_soc, 4); After this, an incoming-call packet will be accepted, but no call-accepted packet will be sent back until the following system call is made on the socket that accepts the call ioctl(vc_soc,SIOCX25SENDCALLACCPT); The network (or cisco xot router used for testing here) will allow the application server's call-user-data in the call-accepted packet, provided the call-request was made with Fast-select NRR. Signed-off-by: Shaun Pereira Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- include/linux/x25.h | 2 ++ include/net/x25.h | 6 ++++-- net/x25/af_x25.c | 37 +++++++++++++++++++++++++++++++++---- net/x25/x25_facilities.c | 34 +++++++++++++++++++++++++++++----- net/x25/x25_subr.c | 23 ++++++++++++++++++----- 5 files changed, 86 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/x25.h b/include/linux/x25.h index 6f43b3d20248..16d44931afa0 100644 --- a/include/linux/x25.h +++ b/include/linux/x25.h @@ -19,6 +19,8 @@ #define SIOCX25SCALLUSERDATA (SIOCPROTOPRIVATE + 5) #define SIOCX25GCAUSEDIAG (SIOCPROTOPRIVATE + 6) #define SIOCX25SCUDMATCHLEN (SIOCPROTOPRIVATE + 7) +#define SIOCX25CALLACCPTAPPRV (SIOCPROTOPRIVATE + 8) +#define SIOCX25SENDCALLACCPT (SIOCPROTOPRIVATE + 9) /* * Values for {get,set}sockopt. diff --git a/include/net/x25.h b/include/net/x25.h index 9dd70dd4a9b7..8b39b98876e8 100644 --- a/include/net/x25.h +++ b/include/net/x25.h @@ -79,6 +79,8 @@ enum { #define X25_DEFAULT_PACKET_SIZE X25_PS128 /* Default Packet Size */ #define X25_DEFAULT_THROUGHPUT 0x0A /* Deafult Throughput */ #define X25_DEFAULT_REVERSE 0x00 /* Default Reverse Charging */ +#define X25_DENY_ACCPT_APPRV 0x01 /* Default value */ +#define X25_ALLOW_ACCPT_APPRV 0x00 /* Control enabled */ #define X25_SMODULUS 8 #define X25_EMODULUS 128 @@ -94,7 +96,7 @@ enum { #define X25_FAC_CLASS_C 0x80 #define X25_FAC_CLASS_D 0xC0 -#define X25_FAC_REVERSE 0x01 +#define X25_FAC_REVERSE 0x01 /* also fast select */ #define X25_FAC_THROUGHPUT 0x02 #define X25_FAC_PACKET_SIZE 0x42 #define X25_FAC_WINDOW_SIZE 0x43 @@ -135,7 +137,7 @@ struct x25_sock { struct x25_address source_addr, dest_addr; struct x25_neigh *neighbour; unsigned int lci, cudmatchlength; - unsigned char state, condition, qbitincl, intflag; + unsigned char state, condition, qbitincl, intflag, accptapprv; unsigned short vs, vr, va, vl; unsigned long t2, t21, t22, t23; unsigned short fraglen; diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index e17d84a55d5e..04bec047fa9a 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -31,6 +31,8 @@ * x25_proc.c, using seq_file * 2005-04-02 Shaun Pereira Selective sub address matching * with call user data + * 2005-04-15 Shaun Pereira Fast select with no restriction on + * response */ #include @@ -502,6 +504,8 @@ static int x25_create(struct socket *sock, int protocol) x25->t2 = sysctl_x25_ack_holdback_timeout; x25->state = X25_STATE_0; x25->cudmatchlength = 0; + x25->accptapprv = X25_DENY_ACCPT_APPRV; /* normally no cud */ + /* on call accept */ x25->facilities.winsize_in = X25_DEFAULT_WINDOW_SIZE; x25->facilities.winsize_out = X25_DEFAULT_WINDOW_SIZE; @@ -551,6 +555,7 @@ static struct sock *x25_make_new(struct sock *osk) x25->facilities = ox25->facilities; x25->qbitincl = ox25->qbitincl; x25->cudmatchlength = ox25->cudmatchlength; + x25->accptapprv = ox25->accptapprv; x25_init_timers(sk); out: @@ -900,9 +905,11 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb, makex25->vc_facil_mask &= ~X25_MASK_REVERSE; makex25->cudmatchlength = x25_sk(sk)->cudmatchlength; - x25_write_internal(make, X25_CALL_ACCEPTED); - - makex25->state = X25_STATE_3; + /* Normally all calls are accepted immediatly */ + if(makex25->accptapprv & X25_DENY_ACCPT_APPRV) { + x25_write_internal(make, X25_CALL_ACCEPTED); + makex25->state = X25_STATE_3; + } /* * Incoming Call User Data. @@ -1294,7 +1301,8 @@ static int x25_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) if (facilities.throughput < 0x03 || facilities.throughput > 0xDD) break; - if (facilities.reverse && facilities.reverse != 1) + if (facilities.reverse && + (facilities.reverse | 0x81)!= 0x81) break; x25->facilities = facilities; rc = 0; @@ -1348,6 +1356,27 @@ static int x25_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) break; } + case SIOCX25CALLACCPTAPPRV: { + rc = -EINVAL; + if (sk->sk_state != TCP_CLOSE) + break; + x25->accptapprv = X25_ALLOW_ACCPT_APPRV; + rc = 0; + break; + } + + case SIOCX25SENDCALLACCPT: { + rc = -EINVAL; + if (sk->sk_state != TCP_ESTABLISHED) + break; + if (x25->accptapprv) /* must call accptapprv above */ + break; + x25_write_internal(sk, X25_CALL_ACCEPTED); + x25->state = X25_STATE_3; + rc = 0; + break; + } + default: rc = dev_ioctl(cmd, argp); break; diff --git a/net/x25/x25_facilities.c b/net/x25/x25_facilities.c index a21bdb95f9a8..54278b962f4c 100644 --- a/net/x25/x25_facilities.c +++ b/net/x25/x25_facilities.c @@ -17,6 +17,8 @@ * X.25 001 Split from x25_subr.c * mar/20/00 Daniela Squassoni Disabling/enabling of facilities * negotiation. + * apr/14/05 Shaun Pereira - Allow fast select with no restriction + * on response. */ #include @@ -43,9 +45,31 @@ int x25_parse_facilities(struct sk_buff *skb, case X25_FAC_CLASS_A: switch (*p) { case X25_FAC_REVERSE: - facilities->reverse = p[1] & 0x01; - *vc_fac_mask |= X25_MASK_REVERSE; - break; + if((p[1] & 0x81) == 0x81) { + facilities->reverse = p[1] & 0x81; + *vc_fac_mask |= X25_MASK_REVERSE; + break; + } + + if((p[1] & 0x01) == 0x01) { + facilities->reverse = p[1] & 0x01; + *vc_fac_mask |= X25_MASK_REVERSE; + break; + } + + if((p[1] & 0x80) == 0x80) { + facilities->reverse = p[1] & 0x80; + *vc_fac_mask |= X25_MASK_REVERSE; + break; + } + + if(p[1] == 0x00) { + facilities->reverse + = X25_DEFAULT_REVERSE; + *vc_fac_mask |= X25_MASK_REVERSE; + break; + } + case X25_FAC_THROUGHPUT: facilities->throughput = p[1]; *vc_fac_mask |= X25_MASK_THROUGHPUT; @@ -122,7 +146,7 @@ int x25_create_facilities(unsigned char *buffer, if (facilities->reverse && (facil_mask & X25_MASK_REVERSE)) { *p++ = X25_FAC_REVERSE; - *p++ = !!facilities->reverse; + *p++ = facilities->reverse; } if (facilities->throughput && (facil_mask & X25_MASK_THROUGHPUT)) { @@ -171,7 +195,7 @@ int x25_negotiate_facilities(struct sk_buff *skb, struct sock *sk, /* * They want reverse charging, we won't accept it. */ - if (theirs.reverse && ours->reverse) { + if ((theirs.reverse & 0x01 ) && (ours->reverse & 0x01)) { SOCK_DEBUG(sk, "X.25: rejecting reverse charging request"); return -1; } diff --git a/net/x25/x25_subr.c b/net/x25/x25_subr.c index c349bbd61684..7fd872ad0c20 100644 --- a/net/x25/x25_subr.c +++ b/net/x25/x25_subr.c @@ -19,6 +19,8 @@ * mar/20/00 Daniela Squassoni Disabling/enabling of facilities * negotiation. * jun/24/01 Arnaldo C. Melo use skb_queue_purge, cleanups + * apr/04/15 Shaun Pereira Fast select with no + * restriction on response. */ #include @@ -127,8 +129,12 @@ void x25_write_internal(struct sock *sk, int frametype) len += 1 + X25_ADDR_LEN + X25_MAX_FAC_LEN + X25_MAX_CUD_LEN; break; - case X25_CALL_ACCEPTED: - len += 1 + X25_MAX_FAC_LEN + X25_MAX_CUD_LEN; + case X25_CALL_ACCEPTED: /* fast sel with no restr on resp */ + if(x25->facilities.reverse & 0x80) { + len += 1 + X25_MAX_FAC_LEN + X25_MAX_CUD_LEN; + } else { + len += 1 + X25_MAX_FAC_LEN; + } break; case X25_CLEAR_REQUEST: case X25_RESET_REQUEST: @@ -203,9 +209,16 @@ void x25_write_internal(struct sock *sk, int frametype) x25->vc_facil_mask); dptr = skb_put(skb, len); memcpy(dptr, facilities, len); - dptr = skb_put(skb, x25->calluserdata.cudlength); - memcpy(dptr, x25->calluserdata.cuddata, - x25->calluserdata.cudlength); + + /* fast select with no restriction on response + allows call user data. Userland must + ensure it is ours and not theirs */ + if(x25->facilities.reverse & 0x80) { + dptr = skb_put(skb, + x25->calluserdata.cudlength); + memcpy(dptr, x25->calluserdata.cuddata, + x25->calluserdata.cudlength); + } x25->calluserdata.cudlength = 0; break; -- cgit v1.2.3 From 408fde81c1bff15c875a3618481e93a01dcc79ea Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jun 2005 00:07:37 -0700 Subject: [PATCH] remove non-DISCONTIG use of pgdat->node_mem_map This patch effectively eliminates direct use of pgdat->node_mem_map outside of the DISCONTIG code. On a flat memory system, these fields aren't currently used, neither are they on a sparsemem system. There was also a node_mem_map(nid) macro on many architectures. Its use along with the use of ->node_mem_map itself was not consistent. It has been removed in favor of two new, more explicit, arch-independent macros: pgdat_page_nr(pgdat, pagenr) nid_page_nr(nid, pagenr) I called them "pgdat" and "nid" because we overload the term "node" to mean "NUMA node", "DISCONTIG node" or "pg_data_t" in very confusing ways. I believe the newer names are much clearer. These macros can be overridden in the sparsemem case with a theoretically slower operation using node_start_pfn and pfn_to_page(), instead. We could make this the only behavior if people want, but I don't want to change too much at once. One thing at a time. This patch removes more code than it adds. Compile tested on alpha, alpha discontig, arm, arm-discontig, i386, i386 generic, NUMAQ, Summit, ppc64, ppc64 discontig, and x86_64. Full list here: http://sr71.net/patches/2.6.12/2.6.12-rc1-mhp2/configs/ Boot tested on NUMAQ, x86 SMP and ppc64 power4/5 LPARs. Signed-off-by: Dave Hansen Signed-off-by: Martin J. Bligh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/mm/numa.c | 16 +++++++--------- arch/i386/mm/pgtable.c | 2 +- arch/ia64/mm/discontig.c | 9 +++++---- arch/m32r/mm/init.c | 4 ++-- arch/mips/sgi-ip27/ip27-memory.c | 5 ++--- arch/parisc/mm/init.c | 2 +- arch/ppc64/mm/init.c | 4 ++-- include/asm-alpha/mmzone.h | 3 +-- include/asm-i386/mmzone.h | 3 +-- include/asm-m32r/mmzone.h | 3 +-- include/asm-parisc/mmzone.h | 3 +-- include/asm-ppc64/mmzone.h | 3 +-- include/asm-x86_64/mmzone.h | 5 +---- include/linux/mmzone.h | 2 ++ 14 files changed, 28 insertions(+), 36 deletions(-) (limited to 'include/linux') diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c index ba81c4422aaf..c7481d59b6df 100644 --- a/arch/alpha/mm/numa.c +++ b/arch/alpha/mm/numa.c @@ -327,8 +327,6 @@ void __init mem_init(void) extern char _text, _etext, _data, _edata; extern char __init_begin, __init_end; unsigned long nid, i; - struct page * lmem_map; - high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT); reservedpages = 0; @@ -338,10 +336,10 @@ void __init mem_init(void) */ totalram_pages += free_all_bootmem_node(NODE_DATA(nid)); - lmem_map = node_mem_map(nid); pfn = NODE_DATA(nid)->node_start_pfn; for (i = 0; i < node_spanned_pages(nid); i++, pfn++) - if (page_is_ram(pfn) && PageReserved(lmem_map+i)) + if (page_is_ram(pfn) && + PageReserved(nid_page_nr(nid, i))) reservedpages++; } @@ -373,18 +371,18 @@ show_mem(void) show_free_areas(); printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); for_each_online_node(nid) { - struct page * lmem_map = node_mem_map(nid); i = node_spanned_pages(nid); while (i-- > 0) { + struct page *page = nid_page_nr(nid, i); total++; - if (PageReserved(lmem_map+i)) + if (PageReserved(page)) reserved++; - else if (PageSwapCache(lmem_map+i)) + else if (PageSwapCache(page)) cached++; - else if (!page_count(lmem_map+i)) + else if (!page_count(page)) free++; else - shared += page_count(lmem_map + i) - 1; + shared += page_count(page) - 1; } } printk("%ld pages of RAM\n",total); diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c index dd81479ff88a..80c84cdf22ef 100644 --- a/arch/i386/mm/pgtable.c +++ b/arch/i386/mm/pgtable.c @@ -36,7 +36,7 @@ void show_mem(void) printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); for_each_pgdat(pgdat) { for (i = 0; i < pgdat->node_spanned_pages; ++i) { - page = pgdat->node_mem_map + i; + page = pgdat_page_nr(pgdat, i); total++; if (PageHighMem(page)) highmem++; diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index c00710929390..f3fd528ead3b 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -560,14 +560,15 @@ void show_mem(void) int shared = 0, cached = 0, reserved = 0; printk("Node ID: %d\n", pgdat->node_id); for(i = 0; i < pgdat->node_spanned_pages; i++) { + struct page *page = pgdat_page_nr(pgdat, i); if (!ia64_pfn_valid(pgdat->node_start_pfn+i)) continue; - if (PageReserved(pgdat->node_mem_map+i)) + if (PageReserved(page)) reserved++; - else if (PageSwapCache(pgdat->node_mem_map+i)) + else if (PageSwapCache(page)) cached++; - else if (page_count(pgdat->node_mem_map+i)) - shared += page_count(pgdat->node_mem_map+i)-1; + else if (page_count(page)) + shared += page_count(page)-1; } total_present += present; total_reserved += reserved; diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c index bc423d838fb8..d9a40b1fe8ba 100644 --- a/arch/m32r/mm/init.c +++ b/arch/m32r/mm/init.c @@ -49,7 +49,7 @@ void show_mem(void) printk("Free swap: %6ldkB\n",nr_swap_pages<<(PAGE_SHIFT-10)); for_each_pgdat(pgdat) { for (i = 0; i < pgdat->node_spanned_pages; ++i) { - page = pgdat->node_mem_map + i; + page = pgdat_page_nr(pgdat, i); total++; if (PageHighMem(page)) highmem++; @@ -152,7 +152,7 @@ int __init reservedpages_count(void) reservedpages = 0; for_each_online_node(nid) for (i = 0 ; i < MAX_LOW_PFN(nid) - START_PFN(nid) ; i++) - if (PageReserved(NODE_DATA(nid)->node_mem_map + i)) + if (PageReserved(nid_page_nr(nid, i))) reservedpages++; return reservedpages; diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c index 0a44a98d7adc..a160d04f7dbe 100644 --- a/arch/mips/sgi-ip27/ip27-memory.c +++ b/arch/mips/sgi-ip27/ip27-memory.c @@ -549,9 +549,8 @@ void __init mem_init(void) */ numslots = node_getlastslot(node); for (slot = 1; slot <= numslots; slot++) { - p = NODE_DATA(node)->node_mem_map + - (slot_getbasepfn(node, slot) - - slot_getbasepfn(node, 0)); + p = nid_page_nr(node, slot_getbasepfn(node, slot) - + slot_getbasepfn(node, 0)); /* * Free valid memory in current slot. diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c index cac37589e35c..2886ad70db48 100644 --- a/arch/parisc/mm/init.c +++ b/arch/parisc/mm/init.c @@ -506,7 +506,7 @@ void show_mem(void) for (j = node_start_pfn(i); j < node_end_pfn(i); j++) { struct page *p; - p = node_mem_map(i) + j - node_start_pfn(i); + p = nid_page_nr(i, j) - node_start_pfn(i); total++; if (PageReserved(p)) diff --git a/arch/ppc64/mm/init.c b/arch/ppc64/mm/init.c index 6fa1e6490b57..29dbe084c21f 100644 --- a/arch/ppc64/mm/init.c +++ b/arch/ppc64/mm/init.c @@ -98,7 +98,7 @@ void show_mem(void) printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); for_each_pgdat(pgdat) { for (i = 0; i < pgdat->node_spanned_pages; i++) { - page = pgdat->node_mem_map + i; + page = pgdat_page_nr(pgdat, i); total++; if (PageReserved(page)) reserved++; @@ -654,7 +654,7 @@ void __init mem_init(void) for_each_pgdat(pgdat) { for (i = 0; i < pgdat->node_spanned_pages; i++) { - page = pgdat->node_mem_map + i; + page = pgdat_page_nr(pgdat, i); if (PageReserved(page)) reservedpages++; } diff --git a/include/asm-alpha/mmzone.h b/include/asm-alpha/mmzone.h index 726c150dcbe4..a011ef4cf3d3 100644 --- a/include/asm-alpha/mmzone.h +++ b/include/asm-alpha/mmzone.h @@ -57,7 +57,6 @@ PLAT_NODE_DATA_LOCALNR(unsigned long p, int n) * Given a kernel address, find the home node of the underlying memory. */ #define kvaddr_to_nid(kaddr) pa_to_nid(__pa(kaddr)) -#define node_mem_map(nid) (NODE_DATA(nid)->node_mem_map) #define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) #define local_mapnr(kvaddr) \ @@ -108,7 +107,7 @@ PLAT_NODE_DATA_LOCALNR(unsigned long p, int n) #define pfn_to_page(pfn) \ ({ \ unsigned long kaddr = (unsigned long)__va((pfn) << PAGE_SHIFT); \ - (node_mem_map(kvaddr_to_nid(kaddr)) + local_mapnr(kaddr)); \ + (NODE_DATA(kvaddr_to_nid(kaddr))->node_mem_map + local_mapnr(kaddr)); \ }) #define page_to_pfn(page) \ diff --git a/include/asm-i386/mmzone.h b/include/asm-i386/mmzone.h index 13830ae67cac..9cec191f462c 100644 --- a/include/asm-i386/mmzone.h +++ b/include/asm-i386/mmzone.h @@ -79,7 +79,6 @@ static inline int pfn_to_nid(unsigned long pfn) */ #define kvaddr_to_nid(kaddr) pfn_to_nid(__pa(kaddr) >> PAGE_SHIFT) -#define node_mem_map(nid) (NODE_DATA(nid)->node_mem_map) #define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) #define node_end_pfn(nid) \ ({ \ @@ -100,7 +99,7 @@ static inline int pfn_to_nid(unsigned long pfn) ({ \ unsigned long __pfn = pfn; \ int __node = pfn_to_nid(__pfn); \ - &node_mem_map(__node)[node_localnr(__pfn,__node)]; \ + &NODE_DATA(__node)->node_mem_map[node_localnr(__pfn,__node)]; \ }) #define page_to_pfn(pg) \ diff --git a/include/asm-m32r/mmzone.h b/include/asm-m32r/mmzone.h index ebf0228fec42..d58878ec899e 100644 --- a/include/asm-m32r/mmzone.h +++ b/include/asm-m32r/mmzone.h @@ -14,7 +14,6 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) #define node_localnr(pfn, nid) ((pfn) - NODE_DATA(nid)->node_start_pfn) -#define node_mem_map(nid) (NODE_DATA(nid)->node_mem_map) #define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) #define node_end_pfn(nid) \ ({ \ @@ -32,7 +31,7 @@ extern struct pglist_data *node_data[]; ({ \ unsigned long __pfn = pfn; \ int __node = pfn_to_nid(__pfn); \ - &node_mem_map(__node)[node_localnr(__pfn,__node)]; \ + &NODE_DATA(__node)->node_mem_map[node_localnr(__pfn,__node)]; \ }) #define page_to_pfn(pg) \ diff --git a/include/asm-parisc/mmzone.h b/include/asm-parisc/mmzone.h index 928bf50c4693..595d3dce120a 100644 --- a/include/asm-parisc/mmzone.h +++ b/include/asm-parisc/mmzone.h @@ -19,7 +19,6 @@ extern struct node_map_data node_data[]; */ #define kvaddr_to_nid(kaddr) pfn_to_nid(__pa(kaddr) >> PAGE_SHIFT) -#define node_mem_map(nid) (NODE_DATA(nid)->node_mem_map) #define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) #define node_end_pfn(nid) \ ({ \ @@ -38,7 +37,7 @@ extern struct node_map_data node_data[]; ({ \ unsigned long __pfn = (pfn); \ int __node = pfn_to_nid(__pfn); \ - &node_mem_map(__node)[node_localnr(__pfn,__node)]; \ + &NODE_DATA(__node)->node_mem_map[node_localnr(__pfn,__node)]; \ }) #define page_to_pfn(pg) \ diff --git a/include/asm-ppc64/mmzone.h b/include/asm-ppc64/mmzone.h index 0619a41a3c9d..cbfc5ecfe875 100644 --- a/include/asm-ppc64/mmzone.h +++ b/include/asm-ppc64/mmzone.h @@ -65,7 +65,6 @@ static inline int pa_to_nid(unsigned long pa) */ #define kvaddr_to_nid(kaddr) pa_to_nid(__pa(kaddr)) -#define node_mem_map(nid) (NODE_DATA(nid)->node_mem_map) #define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) #define node_end_pfn(nid) (NODE_DATA(nid)->node_end_pfn) @@ -76,7 +75,7 @@ static inline int pa_to_nid(unsigned long pa) #define discontigmem_pfn_to_page(pfn) \ ({ \ unsigned long __tmp = pfn; \ - (node_mem_map(pfn_to_nid(__tmp)) + \ + (NODE_DATA(pfn_to_nid(__tmp))->node_mem_map + \ node_localnr(__tmp, pfn_to_nid(__tmp))); \ }) diff --git a/include/asm-x86_64/mmzone.h b/include/asm-x86_64/mmzone.h index d95b7c240831..ca4fc3fe0dee 100644 --- a/include/asm-x86_64/mmzone.h +++ b/include/asm-x86_64/mmzone.h @@ -35,9 +35,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) #define kvaddr_to_nid(kaddr) phys_to_nid(__pa(kaddr)) #define NODE_DATA(nid) (node_data[nid]) -#define node_mem_map(nid) (NODE_DATA(nid)->node_mem_map) - -#define node_mem_map(nid) (NODE_DATA(nid)->node_mem_map) #define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) #define node_end_pfn(nid) (NODE_DATA(nid)->node_start_pfn + \ NODE_DATA(nid)->node_spanned_pages) @@ -50,7 +47,7 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) (2.4 used to). */ #define pfn_to_page(pfn) ({ \ int nid = phys_to_nid(((unsigned long)(pfn)) << PAGE_SHIFT); \ - ((pfn) - node_start_pfn(nid)) + node_mem_map(nid); \ + ((pfn) - node_start_pfn(nid)) + NODE_DATA(nid)->node_mem_map; \ }) #define page_to_pfn(page) \ diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 4733d35d8223..b79633d3a97b 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -284,6 +284,8 @@ typedef struct pglist_data { #define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages) #define node_spanned_pages(nid) (NODE_DATA(nid)->node_spanned_pages) +#define pgdat_page_nr(pgdat, pagenr) ((pgdat)->node_mem_map + (pagenr)) +#define nid_page_nr(nid, pagenr) pgdat_page_nr(NODE_DATA(nid),(pagenr)) extern struct pglist_data *pgdat_list; -- cgit v1.2.3 From 6f167ec721108c9282d54424516a12c805e3c306 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jun 2005 00:07:39 -0700 Subject: [PATCH] sparsemem base: simple NUMA remap space allocator Introduce a simple allocator for the NUMA remap space. This space is very scarce, used for structures which are best allocated node local. This mechanism is also used on non-NUMA ia64 systems with a vmem_map to keep the pgdat->node_mem_map initialized in a consistent place for all architectures. Issues: o alloc_remap takes a node_id where we might expect a pgdat which was intended to allow us to allocate the pgdat's using this mechanism; which we do not yet do. Could have alloc_remap_node() and alloc_remap_nid() for this purpose. Signed-off-by: Andy Whitcroft Signed-off-by: Dave Hansen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/Kconfig | 5 ++++ arch/i386/mm/discontig.c | 59 +++++++++++++++++++++++++----------------------- include/linux/bootmem.h | 9 ++++++++ mm/page_alloc.c | 6 ++++- 4 files changed, 50 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index dfd904f6883b..35ca3a17ed20 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -803,6 +803,11 @@ config NEED_NODE_MEMMAP_SIZE depends on DISCONTIGMEM default y +config HAVE_ARCH_ALLOC_REMAP + bool + depends on NUMA + default y + config HIGHPTE bool "Allocate 3rd-level pagetables from highmem" depends on HIGHMEM4G || HIGHMEM64G diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c index 85d2fcbe1079..dcc71f969b01 100644 --- a/arch/i386/mm/discontig.c +++ b/arch/i386/mm/discontig.c @@ -108,6 +108,9 @@ unsigned long node_remap_offset[MAX_NUMNODES]; void *node_remap_start_vaddr[MAX_NUMNODES]; void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags); +void *node_remap_end_vaddr[MAX_NUMNODES]; +void *node_remap_alloc_vaddr[MAX_NUMNODES]; + /* * FLAT - support for basic PC memory model with discontig enabled, essentially * a single node with all available processors in it with a flat @@ -178,6 +181,21 @@ static void __init allocate_pgdat(int nid) } } +void *alloc_remap(int nid, unsigned long size) +{ + void *allocation = node_remap_alloc_vaddr[nid]; + + size = ALIGN(size, L1_CACHE_BYTES); + + if (!allocation || (allocation + size) >= node_remap_end_vaddr[nid]) + return 0; + + node_remap_alloc_vaddr[nid] += size; + memset(allocation, 0, size); + + return allocation; +} + void __init remap_numa_kva(void) { void *vaddr; @@ -185,8 +203,6 @@ void __init remap_numa_kva(void) int node; for_each_online_node(node) { - if (node == 0) - continue; for (pfn=0; pfn < node_remap_size[node]; pfn += PTRS_PER_PTE) { vaddr = node_remap_start_vaddr[node]+(pfn<node_mem_map = (struct page *)lmem_map; - free_area_init_node(nid, NODE_DATA(nid), zones_size, - start, zholes_size); - } + + free_area_init_node(nid, NODE_DATA(nid), zones_size, start, + zholes_size); } return; } diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index 0dd8ca1a3d5a..500f451ce0c0 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -67,6 +67,15 @@ extern void * __init __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, __alloc_bootmem_node((pgdat), (x), PAGE_SIZE, 0) #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */ +#ifdef CONFIG_HAVE_ARCH_ALLOC_REMAP +extern void *alloc_remap(int nid, unsigned long size); +#else +static inline void *alloc_remap(int nid, unsigned long size) +{ + return NULL; +} +#endif + extern unsigned long __initdata nr_kernel_pages; extern unsigned long __initdata nr_all_pages; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 559336de9687..bf1dd8819097 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1936,6 +1936,7 @@ static void __init free_area_init_core(struct pglist_data *pgdat, static void __init alloc_node_mem_map(struct pglist_data *pgdat) { unsigned long size; + struct page *map; /* Skip empty nodes */ if (!pgdat->node_spanned_pages) @@ -1944,7 +1945,10 @@ static void __init alloc_node_mem_map(struct pglist_data *pgdat) /* ia64 gets its own node_mem_map, before this, without bootmem */ if (!pgdat->node_mem_map) { size = (pgdat->node_spanned_pages + 1) * sizeof(struct page); - pgdat->node_mem_map = alloc_bootmem_node(pgdat, size); + map = alloc_remap(pgdat->node_id, size); + if (!map) + map = alloc_bootmem_node(pgdat, size); + pgdat->node_mem_map = map; } #ifndef CONFIG_DISCONTIGMEM /* -- cgit v1.2.3 From 348f8b6c4837a07304d2f72b11ce8d96588065e0 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jun 2005 00:07:40 -0700 Subject: [PATCH] sparsemem base: reorganize page->flags bit operations Generify the value fields in the page_flags. The aim is to allow the location and size of these fields to be varied. Additionally we want to move away from fixed allocations per field whilst still enforcing the overall bit utilisation limits. We rely on the compiler to spot and optimise the accessor functions. Signed-off-by: Andy Whitcroft Signed-off-by: Dave Hansen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 53 +++++++++++++++++++++++++++++++++++++++++--------- include/linux/mmzone.h | 19 +++++++----------- mm/page_alloc.c | 2 +- 3 files changed, 52 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 1813b162b0a8..57b2ead51dba 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -395,19 +395,41 @@ static inline void put_page(struct page *page) /* * The zone field is never updated after free_area_init_core() * sets it, so none of the operations on it need to be atomic. - * We'll have up to (MAX_NUMNODES * MAX_NR_ZONES) zones total, - * so we use (MAX_NODES_SHIFT + MAX_ZONES_SHIFT) here to get enough bits. */ -#define NODEZONE_SHIFT (sizeof(page_flags_t)*8 - MAX_NODES_SHIFT - MAX_ZONES_SHIFT) + +/* Page flags: | NODE | ZONE | ... | FLAGS | */ +#define NODES_PGOFF ((sizeof(page_flags_t)*8) - NODES_SHIFT) +#define ZONES_PGOFF (NODES_PGOFF - ZONES_SHIFT) + +/* + * Define the bit shifts to access each section. For non-existant + * sections we define the shift as 0; that plus a 0 mask ensures + * the compiler will optimise away reference to them. + */ +#define NODES_PGSHIFT (NODES_PGOFF * (NODES_SHIFT != 0)) +#define ZONES_PGSHIFT (ZONES_PGOFF * (ZONES_SHIFT != 0)) + +/* NODE:ZONE is used to lookup the zone from a page. */ +#define ZONETABLE_SHIFT (NODES_SHIFT + ZONES_SHIFT) +#define ZONETABLE_PGSHIFT ZONES_PGSHIFT + +#if NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED +#error NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED +#endif + #define NODEZONE(node, zone) ((node << ZONES_SHIFT) | zone) +#define ZONES_MASK ((1UL << ZONES_SHIFT) - 1) +#define NODES_MASK ((1UL << NODES_SHIFT) - 1) +#define ZONETABLE_MASK ((1UL << ZONETABLE_SHIFT) - 1) + static inline unsigned long page_zonenum(struct page *page) { - return (page->flags >> NODEZONE_SHIFT) & (~(~0UL << ZONES_SHIFT)); + return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK; } static inline unsigned long page_to_nid(struct page *page) { - return (page->flags >> (NODEZONE_SHIFT + ZONES_SHIFT)); + return (page->flags >> NODES_PGSHIFT) & NODES_MASK; } struct zone; @@ -415,13 +437,26 @@ extern struct zone *zone_table[]; static inline struct zone *page_zone(struct page *page) { - return zone_table[page->flags >> NODEZONE_SHIFT]; + return zone_table[(page->flags >> ZONETABLE_PGSHIFT) & + ZONETABLE_MASK]; +} + +static inline void set_page_zone(struct page *page, unsigned long zone) +{ + page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT); + page->flags |= (zone & ZONES_MASK) << ZONES_PGSHIFT; +} +static inline void set_page_node(struct page *page, unsigned long node) +{ + page->flags &= ~(NODES_MASK << NODES_PGSHIFT); + page->flags |= (node & NODES_MASK) << NODES_PGSHIFT; } -static inline void set_page_zone(struct page *page, unsigned long nodezone_num) +static inline void set_page_links(struct page *page, unsigned long zone, + unsigned long node) { - page->flags &= ~(~0UL << NODEZONE_SHIFT); - page->flags |= nodezone_num << NODEZONE_SHIFT; + set_page_zone(page, zone); + set_page_node(page, node); } #ifndef CONFIG_DISCONTIGMEM diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index b79633d3a97b..39e912708e2a 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -414,30 +414,25 @@ extern struct pglist_data contig_page_data; #include +#endif /* !CONFIG_DISCONTIGMEM */ + #if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED) /* * with 32 bit page->flags field, we reserve 8 bits for node/zone info. * there are 3 zones (2 bits) and this leaves 8-2=6 bits for nodes. */ -#define MAX_NODES_SHIFT 6 +#define FLAGS_RESERVED 8 + #elif BITS_PER_LONG == 64 /* * with 64 bit flags field, there's plenty of room. */ -#define MAX_NODES_SHIFT 10 -#endif +#define FLAGS_RESERVED 32 -#endif /* !CONFIG_DISCONTIGMEM */ - -#if NODES_SHIFT > MAX_NODES_SHIFT -#error NODES_SHIFT > MAX_NODES_SHIFT -#endif +#else -/* There are currently 3 zones: DMA, Normal & Highmem, thus we need 2 bits */ -#define MAX_ZONES_SHIFT 2 +#error BITS_PER_LONG not defined -#if ZONES_SHIFT > MAX_ZONES_SHIFT -#error ZONES_SHIFT > MAX_ZONES_SHIFT #endif #endif /* !__ASSEMBLY__ */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index bf1dd8819097..1958358e29b0 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1653,7 +1653,7 @@ void __init memmap_init_zone(unsigned long size, int nid, unsigned long zone, struct page *page; for (page = start; page < (start + size); page++) { - set_page_zone(page, NODEZONE(nid, zone)); + set_page_links(page, zone, nid); set_page_count(page, 0); reset_page_mapcount(page); SetPageReserved(page); -- cgit v1.2.3 From 93b7504e3e6c1d98586854806e51bea329ea3aa9 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jun 2005 00:07:47 -0700 Subject: [PATCH] Introduce new Kconfig option for NUMA or DISCONTIG There is some confusion that arose when working on SPARSEMEM patch between what is needed for DISCONTIG vs. NUMA. Multiple pg_data_t's are needed for DISCONTIGMEM or NUMA, independently. All of the current NUMA implementations require an implementation of DISCONTIG. Because of this, quite a lot of code which is really needed for NUMA is actually under DISCONTIG #ifdefs. For SPARSEMEM, we changed some of these #ifdefs to CONFIG_NUMA, but that broke the DISCONTIG=y and NUMA=n case. Introducing this new NEED_MULTIPLE_NODES config option allows code that is needed for both NUMA or DISCONTIG to be separated out from code that is specific to DISCONTIG. One great advantage of this approach is that it doesn't require every architecture to be converted over. All of the current implementations should "just work", only the ones implementing SPARSEMEM will have to be fixed up. The change to free_area_init() makes it work inside, or out of the new config option. Signed-off-by: Dave Hansen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 6 +++--- mm/Kconfig | 8 ++++++++ mm/page_alloc.c | 6 +++--- 3 files changed, 14 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 39e912708e2a..95f4a780ea66 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -402,7 +402,7 @@ int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int, struct file *, /* Returns the number of the current Node. */ #define numa_node_id() (cpu_to_node(raw_smp_processor_id())) -#ifndef CONFIG_DISCONTIGMEM +#ifndef CONFIG_NEED_MULTIPLE_NODES extern struct pglist_data contig_page_data; #define NODE_DATA(nid) (&contig_page_data) @@ -410,11 +410,11 @@ extern struct pglist_data contig_page_data; #define MAX_NODES_SHIFT 1 #define pfn_to_nid(pfn) (0) -#else /* CONFIG_DISCONTIGMEM */ +#else /* CONFIG_NEED_MULTIPLE_NODES */ #include -#endif /* !CONFIG_DISCONTIGMEM */ +#endif /* !CONFIG_NEED_MULTIPLE_NODES */ #if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED) /* diff --git a/mm/Kconfig b/mm/Kconfig index 69caa9d8674e..15c131393639 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -23,3 +23,11 @@ config DISCONTIGMEM endchoice +# +# Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's +# to represent different areas of memory. This variable allows +# those dependencies to exist individually. +# +config NEED_MULTIPLE_NODES + def_bool y + depends on DISCONTIGMEM || NUMA diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 1958358e29b0..20e239599db0 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1972,18 +1972,18 @@ void __init free_area_init_node(int nid, struct pglist_data *pgdat, free_area_init_core(pgdat, zones_size, zholes_size); } -#ifndef CONFIG_DISCONTIGMEM +#ifndef CONFIG_NEED_MULTIPLE_NODES static bootmem_data_t contig_bootmem_data; struct pglist_data contig_page_data = { .bdata = &contig_bootmem_data }; EXPORT_SYMBOL(contig_page_data); +#endif void __init free_area_init(unsigned long *zones_size) { - free_area_init_node(0, &contig_page_data, zones_size, + free_area_init_node(0, NODE_DATA(0), zones_size, __pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL); } -#endif #ifdef CONFIG_PROC_FS -- cgit v1.2.3 From b159d43fbf7eaaac6ecc647f51cf4257332db47b Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Thu, 23 Jun 2005 00:07:52 -0700 Subject: [PATCH] generify early_pfn_to_nid Provide a default implementation for early_pfn_to_nid returning node 0. Allow architectures to override this with their own implementation out of asm/mmzone.h. Signed-off-by: Andy Whitcroft Signed-off-by: Dave Hansen Signed-off-by: Martin Bligh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/Kconfig | 4 ++++ include/asm-i386/mmzone.h | 3 +++ include/linux/mmzone.h | 4 ++++ 3 files changed, 11 insertions(+) (limited to 'include/linux') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 8e5242c8e09d..a8128f997339 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -810,6 +810,10 @@ config HAVE_ARCH_ALLOC_REMAP source "mm/Kconfig" +config HAVE_ARCH_EARLY_PFN_TO_NID + bool + default y + config HIGHPTE bool "Allocate 3rd-level pagetables from highmem" depends on HIGHMEM4G || HIGHMEM64G diff --git a/include/asm-i386/mmzone.h b/include/asm-i386/mmzone.h index 9cec191f462c..48e46d403aa6 100644 --- a/include/asm-i386/mmzone.h +++ b/include/asm-i386/mmzone.h @@ -143,4 +143,7 @@ static inline void get_memcfg_numa(void) } #endif /* CONFIG_DISCONTIGMEM */ + +extern int early_pfn_to_nid(unsigned long pfn); + #endif /* _ASM_MMZONE_H_ */ diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 95f4a780ea66..6ef07de98d69 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -435,6 +435,10 @@ extern struct pglist_data contig_page_data; #endif +#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID +#define early_pfn_to_nid(nid) (0UL) +#endif + #endif /* !__ASSEMBLY__ */ #endif /* __KERNEL__ */ #endif /* _LINUX_MMZONE_H */ -- cgit v1.2.3 From d41dee369bff3b9dcb6328d4d822926c28cc2594 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Thu, 23 Jun 2005 00:07:54 -0700 Subject: [PATCH] sparsemem memory model Sparsemem abstracts the use of discontiguous mem_maps[]. This kind of mem_map[] is needed by discontiguous memory machines (like in the old CONFIG_DISCONTIGMEM case) as well as memory hotplug systems. Sparsemem replaces DISCONTIGMEM when enabled, and it is hoped that it can eventually become a complete replacement. A significant advantage over DISCONTIGMEM is that it's completely separated from CONFIG_NUMA. When producing this patch, it became apparent in that NUMA and DISCONTIG are often confused. Another advantage is that sparse doesn't require each NUMA node's ranges to be contiguous. It can handle overlapping ranges between nodes with no problems, where DISCONTIGMEM currently throws away that memory. Sparsemem uses an array to provide different pfn_to_page() translations for each SECTION_SIZE area of physical memory. This is what allows the mem_map[] to be chopped up. In order to do quick pfn_to_page() operations, the section number of the page is encoded in page->flags. Part of the sparsemem infrastructure enables sharing of these bits more dynamically (at compile-time) between the page_zone() and sparsemem operations. However, on 32-bit architectures, the number of bits is quite limited, and may require growing the size of the page->flags type in certain conditions. Several things might force this to occur: a decrease in the SECTION_SIZE (if you want to hotplug smaller areas of memory), an increase in the physical address space, or an increase in the number of used page->flags. One thing to note is that, once sparsemem is present, the NUMA node information no longer needs to be stored in the page->flags. It might provide speed increases on certain platforms and will be stored there if there is room. But, if out of room, an alternate (theoretically slower) mechanism is used. This patch introduces CONFIG_FLATMEM. It is used in almost all cases where there used to be an #ifndef DISCONTIG, because SPARSEMEM and DISCONTIGMEM often have to compile out the same areas of code. Signed-off-by: Andy Whitcroft Signed-off-by: Dave Hansen Signed-off-by: Martin Bligh Signed-off-by: Adrian Bunk Signed-off-by: Yasunori Goto Signed-off-by: Bob Picco Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/Kconfig | 1 + include/linux/mm.h | 92 ++++++++++++++++++++++++++++++++++++++--------- include/linux/mmzone.h | 96 ++++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/numa.h | 2 +- mm/Kconfig | 38 ++++++++++++++++++-- mm/Makefile | 1 + mm/bootmem.c | 9 +++-- mm/memory.c | 2 +- mm/page_alloc.c | 39 +++++++++++++++----- mm/sparse.c | 85 ++++++++++++++++++++++++++++++++++++++++++++ 10 files changed, 332 insertions(+), 33 deletions(-) create mode 100644 mm/sparse.c (limited to 'include/linux') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 3b7248126d29..f0064b5e3702 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -813,6 +813,7 @@ source "mm/Kconfig" config HAVE_ARCH_EARLY_PFN_TO_NID bool default y + depends on NUMA config HIGHPTE bool "Allocate 3rd-level pagetables from highmem" diff --git a/include/linux/mm.h b/include/linux/mm.h index 57b2ead51dba..6eb7f48317f8 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -397,40 +397,80 @@ static inline void put_page(struct page *page) * sets it, so none of the operations on it need to be atomic. */ -/* Page flags: | NODE | ZONE | ... | FLAGS | */ -#define NODES_PGOFF ((sizeof(page_flags_t)*8) - NODES_SHIFT) -#define ZONES_PGOFF (NODES_PGOFF - ZONES_SHIFT) + +/* + * page->flags layout: + * + * There are three possibilities for how page->flags get + * laid out. The first is for the normal case, without + * sparsemem. The second is for sparsemem when there is + * plenty of space for node and section. The last is when + * we have run out of space and have to fall back to an + * alternate (slower) way of determining the node. + * + * No sparsemem: | NODE | ZONE | ... | FLAGS | + * with space for node: | SECTION | NODE | ZONE | ... | FLAGS | + * no space for node: | SECTION | ZONE | ... | FLAGS | + */ +#ifdef CONFIG_SPARSEMEM +#define SECTIONS_WIDTH SECTIONS_SHIFT +#else +#define SECTIONS_WIDTH 0 +#endif + +#define ZONES_WIDTH ZONES_SHIFT + +#if SECTIONS_WIDTH+ZONES_WIDTH+NODES_SHIFT <= FLAGS_RESERVED +#define NODES_WIDTH NODES_SHIFT +#else +#define NODES_WIDTH 0 +#endif + +/* Page flags: | [SECTION] | [NODE] | ZONE | ... | FLAGS | */ +#define SECTIONS_PGOFF ((sizeof(page_flags_t)*8) - SECTIONS_WIDTH) +#define NODES_PGOFF (SECTIONS_PGOFF - NODES_WIDTH) +#define ZONES_PGOFF (NODES_PGOFF - ZONES_WIDTH) + +/* + * We are going to use the flags for the page to node mapping if its in + * there. This includes the case where there is no node, so it is implicit. + */ +#define FLAGS_HAS_NODE (NODES_WIDTH > 0 || NODES_SHIFT == 0) + +#ifndef PFN_SECTION_SHIFT +#define PFN_SECTION_SHIFT 0 +#endif /* * Define the bit shifts to access each section. For non-existant * sections we define the shift as 0; that plus a 0 mask ensures * the compiler will optimise away reference to them. */ -#define NODES_PGSHIFT (NODES_PGOFF * (NODES_SHIFT != 0)) -#define ZONES_PGSHIFT (ZONES_PGOFF * (ZONES_SHIFT != 0)) +#define SECTIONS_PGSHIFT (SECTIONS_PGOFF * (SECTIONS_WIDTH != 0)) +#define NODES_PGSHIFT (NODES_PGOFF * (NODES_WIDTH != 0)) +#define ZONES_PGSHIFT (ZONES_PGOFF * (ZONES_WIDTH != 0)) -/* NODE:ZONE is used to lookup the zone from a page. */ +/* NODE:ZONE or SECTION:ZONE is used to lookup the zone from a page. */ +#if FLAGS_HAS_NODE #define ZONETABLE_SHIFT (NODES_SHIFT + ZONES_SHIFT) +#else +#define ZONETABLE_SHIFT (SECTIONS_SHIFT + ZONES_SHIFT) +#endif #define ZONETABLE_PGSHIFT ZONES_PGSHIFT -#if NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED -#error NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED +#if SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > FLAGS_RESERVED +#error SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > FLAGS_RESERVED #endif -#define NODEZONE(node, zone) ((node << ZONES_SHIFT) | zone) - -#define ZONES_MASK ((1UL << ZONES_SHIFT) - 1) -#define NODES_MASK ((1UL << NODES_SHIFT) - 1) +#define ZONES_MASK ((1UL << ZONES_WIDTH) - 1) +#define NODES_MASK ((1UL << NODES_WIDTH) - 1) +#define SECTIONS_MASK ((1UL << SECTIONS_WIDTH) - 1) #define ZONETABLE_MASK ((1UL << ZONETABLE_SHIFT) - 1) static inline unsigned long page_zonenum(struct page *page) { return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK; } -static inline unsigned long page_to_nid(struct page *page) -{ - return (page->flags >> NODES_PGSHIFT) & NODES_MASK; -} struct zone; extern struct zone *zone_table[]; @@ -441,6 +481,18 @@ static inline struct zone *page_zone(struct page *page) ZONETABLE_MASK]; } +static inline unsigned long page_to_nid(struct page *page) +{ + if (FLAGS_HAS_NODE) + return (page->flags >> NODES_PGSHIFT) & NODES_MASK; + else + return page_zone(page)->zone_pgdat->node_id; +} +static inline unsigned long page_to_section(struct page *page) +{ + return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK; +} + static inline void set_page_zone(struct page *page, unsigned long zone) { page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT); @@ -451,12 +503,18 @@ static inline void set_page_node(struct page *page, unsigned long node) page->flags &= ~(NODES_MASK << NODES_PGSHIFT); page->flags |= (node & NODES_MASK) << NODES_PGSHIFT; } +static inline void set_page_section(struct page *page, unsigned long section) +{ + page->flags &= ~(SECTIONS_MASK << SECTIONS_PGSHIFT); + page->flags |= (section & SECTIONS_MASK) << SECTIONS_PGSHIFT; +} static inline void set_page_links(struct page *page, unsigned long zone, - unsigned long node) + unsigned long node, unsigned long pfn) { set_page_zone(page, zone); set_page_node(page, node); + set_page_section(page, pfn_to_section_nr(pfn)); } #ifndef CONFIG_DISCONTIGMEM diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 6ef07de98d69..19860d317ec2 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -269,7 +269,9 @@ typedef struct pglist_data { struct zone node_zones[MAX_NR_ZONES]; struct zonelist node_zonelists[GFP_ZONETYPES]; int nr_zones; +#ifdef CONFIG_FLAT_NODE_MEM_MAP struct page *node_mem_map; +#endif struct bootmem_data *bdata; unsigned long node_start_pfn; unsigned long node_present_pages; /* total number of physical pages */ @@ -284,7 +286,11 @@ typedef struct pglist_data { #define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages) #define node_spanned_pages(nid) (NODE_DATA(nid)->node_spanned_pages) +#ifdef CONFIG_FLAT_NODE_MEM_MAP #define pgdat_page_nr(pgdat, pagenr) ((pgdat)->node_mem_map + (pagenr)) +#else +#define pgdat_page_nr(pgdat, pagenr) pfn_to_page((pgdat)->node_start_pfn + (pagenr)) +#endif #define nid_page_nr(nid, pagenr) pgdat_page_nr(NODE_DATA(nid),(pagenr)) extern struct pglist_data *pgdat_list; @@ -416,6 +422,10 @@ extern struct pglist_data contig_page_data; #endif /* !CONFIG_NEED_MULTIPLE_NODES */ +#ifdef CONFIG_SPARSEMEM +#include +#endif + #if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED) /* * with 32 bit page->flags field, we reserve 8 bits for node/zone info. @@ -439,6 +449,92 @@ extern struct pglist_data contig_page_data; #define early_pfn_to_nid(nid) (0UL) #endif +#define pfn_to_section_nr(pfn) ((pfn) >> PFN_SECTION_SHIFT) +#define section_nr_to_pfn(sec) ((sec) << PFN_SECTION_SHIFT) + +#ifdef CONFIG_SPARSEMEM + +/* + * SECTION_SHIFT #bits space required to store a section # + * + * PA_SECTION_SHIFT physical address to/from section number + * PFN_SECTION_SHIFT pfn to/from section number + */ +#define SECTIONS_SHIFT (MAX_PHYSMEM_BITS - SECTION_SIZE_BITS) + +#define PA_SECTION_SHIFT (SECTION_SIZE_BITS) +#define PFN_SECTION_SHIFT (SECTION_SIZE_BITS - PAGE_SHIFT) + +#define NR_MEM_SECTIONS (1UL << SECTIONS_SHIFT) + +#define PAGES_PER_SECTION (1UL << PFN_SECTION_SHIFT) +#define PAGE_SECTION_MASK (~(PAGES_PER_SECTION-1)) + +#if (MAX_ORDER - 1 + PAGE_SHIFT) > SECTION_SIZE_BITS +#error Allocator MAX_ORDER exceeds SECTION_SIZE +#endif + +struct page; +struct mem_section { + struct page *section_mem_map; +}; + +extern struct mem_section mem_section[NR_MEM_SECTIONS]; + +/* + * Given a kernel address, find the home node of the underlying memory. + */ +#define kvaddr_to_nid(kaddr) pfn_to_nid(__pa(kaddr) >> PAGE_SHIFT) + +static inline struct mem_section *__pfn_to_section(unsigned long pfn) +{ + return &mem_section[pfn_to_section_nr(pfn)]; +} + +#define pfn_to_page(pfn) \ +({ \ + unsigned long __pfn = (pfn); \ + __pfn_to_section(__pfn)->section_mem_map + __pfn; \ +}) +#define page_to_pfn(page) \ +({ \ + page - mem_section[page_to_section(page)].section_mem_map; \ +}) + +static inline int pfn_valid(unsigned long pfn) +{ + if (pfn_to_section_nr(pfn) >= NR_MEM_SECTIONS) + return 0; + return mem_section[pfn_to_section_nr(pfn)].section_mem_map != 0; +} + +/* + * These are _only_ used during initialisation, therefore they + * can use __initdata ... They could have names to indicate + * this restriction. + */ +#ifdef CONFIG_NUMA +#define pfn_to_nid early_pfn_to_nid +#endif + +#define pfn_to_pgdat(pfn) \ +({ \ + NODE_DATA(pfn_to_nid(pfn)); \ +}) + +#define early_pfn_valid(pfn) pfn_valid(pfn) +void sparse_init(void); +#else +#define sparse_init() do {} while (0) +#endif /* CONFIG_SPARSEMEM */ + +#ifndef early_pfn_valid +#define early_pfn_valid(pfn) (1) +#endif + +void memory_present(int nid, unsigned long start, unsigned long end); +unsigned long __init node_memmap_size_bytes(int, unsigned long, unsigned long); + #endif /* !__ASSEMBLY__ */ #endif /* __KERNEL__ */ #endif /* _LINUX_MMZONE_H */ diff --git a/include/linux/numa.h b/include/linux/numa.h index bd0c8c4e9a95..f0c539bd3cfc 100644 --- a/include/linux/numa.h +++ b/include/linux/numa.h @@ -3,7 +3,7 @@ #include -#ifdef CONFIG_DISCONTIGMEM +#ifndef CONFIG_FLATMEM #include #endif diff --git a/mm/Kconfig b/mm/Kconfig index 5127441561b4..cd379936cac6 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -6,6 +6,7 @@ choice prompt "Memory model" depends on SELECT_MEMORY_MODEL default DISCONTIGMEM_MANUAL if ARCH_DISCONTIGMEM_DEFAULT + default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT default FLATMEM_MANUAL config FLATMEM_MANUAL @@ -17,7 +18,15 @@ config FLATMEM_MANUAL only have one option here: FLATMEM. This is normal and a correct option. - If unsure, choose this option over any other. + Some users of more advanced features like NUMA and + memory hotplug may have different options here. + DISCONTIGMEM is an more mature, better tested system, + but is incompatible with memory hotplug and may suffer + decreased performance over SPARSEMEM. If unsure between + "Sparse Memory" and "Discontiguous Memory", choose + "Discontiguous Memory". + + If unsure, choose this option (Flat Memory) over any other. config DISCONTIGMEM_MANUAL bool "Discontigious Memory" @@ -35,15 +44,38 @@ config DISCONTIGMEM_MANUAL If unsure, choose "Flat Memory" over this option. +config SPARSEMEM_MANUAL + bool "Sparse Memory" + depends on ARCH_SPARSEMEM_ENABLE + help + This will be the only option for some systems, including + memory hotplug systems. This is normal. + + For many other systems, this will be an alternative to + "Discontigious Memory". This option provides some potential + performance benefits, along with decreased code complexity, + but it is newer, and more experimental. + + If unsure, choose "Discontiguous Memory" or "Flat Memory" + over this option. + endchoice config DISCONTIGMEM def_bool y depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL +config SPARSEMEM + def_bool y + depends on SPARSEMEM_MANUAL + config FLATMEM def_bool y - depends on !DISCONTIGMEM || FLATMEM_MANUAL + depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL + +config FLAT_NODE_MEM_MAP + def_bool y + depends on !SPARSEMEM # # Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's @@ -56,4 +88,4 @@ config NEED_MULTIPLE_NODES config HAVE_MEMORY_PRESENT def_bool y - depends on ARCH_HAVE_MEMORY_PRESENT + depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM diff --git a/mm/Makefile b/mm/Makefile index 097408064f6a..8f70ffd763c8 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -15,6 +15,7 @@ obj-y := bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \ obj-$(CONFIG_SWAP) += page_io.o swap_state.o swapfile.o thrash.o obj-$(CONFIG_HUGETLBFS) += hugetlb.o obj-$(CONFIG_NUMA) += mempolicy.o +obj-$(CONFIG_SPARSEMEM) += sparse.o obj-$(CONFIG_SHMEM) += shmem.o obj-$(CONFIG_TINY_SHMEM) += tiny-shmem.o diff --git a/mm/bootmem.c b/mm/bootmem.c index 260e703850d8..f82f7aebbee3 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -256,6 +256,7 @@ found: static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) { struct page *page; + unsigned long pfn; bootmem_data_t *bdata = pgdat->bdata; unsigned long i, count, total = 0; unsigned long idx; @@ -266,7 +267,7 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) count = 0; /* first extant page of the node */ - page = virt_to_page(phys_to_virt(bdata->node_boot_start)); + pfn = bdata->node_boot_start >> PAGE_SHIFT; idx = bdata->node_low_pfn - (bdata->node_boot_start >> PAGE_SHIFT); map = bdata->node_bootmem_map; /* Check physaddr is O(LOG2(BITS_PER_LONG)) page aligned */ @@ -275,9 +276,11 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) gofast = 1; for (i = 0; i < idx; ) { unsigned long v = ~map[i / BITS_PER_LONG]; + if (gofast && v == ~0UL) { int j, order; + page = pfn_to_page(pfn); count += BITS_PER_LONG; __ClearPageReserved(page); order = ffs(BITS_PER_LONG) - 1; @@ -292,6 +295,8 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) page += BITS_PER_LONG; } else if (v) { unsigned long m; + + page = pfn_to_page(pfn); for (m = 1; m && i < idx; m<<=1, page++, i++) { if (v & m) { count++; @@ -302,8 +307,8 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) } } else { i+=BITS_PER_LONG; - page += BITS_PER_LONG; } + pfn += BITS_PER_LONG; } total += count; diff --git a/mm/memory.c b/mm/memory.c index da91b7bf9986..30975ef48722 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -58,7 +58,7 @@ #include #include -#ifndef CONFIG_DISCONTIGMEM +#ifndef CONFIG_NEED_MULTIPLE_NODES /* use the per-pgdat data instead for discontigmem - mbligh */ unsigned long max_mapnr; struct page *mem_map; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 20e239599db0..5c1b8982a6da 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -68,7 +68,7 @@ EXPORT_SYMBOL(nr_swap_pages); * Used by page_zone() to look up the address of the struct zone whose * id is encoded in the upper bits of page->flags */ -struct zone *zone_table[1 << (ZONES_SHIFT + NODES_SHIFT)]; +struct zone *zone_table[1 << ZONETABLE_SHIFT]; EXPORT_SYMBOL(zone_table); static char *zone_names[MAX_NR_ZONES] = { "DMA", "Normal", "HighMem" }; @@ -1649,11 +1649,15 @@ static void __init calculate_zone_totalpages(struct pglist_data *pgdat, void __init memmap_init_zone(unsigned long size, int nid, unsigned long zone, unsigned long start_pfn) { - struct page *start = pfn_to_page(start_pfn); struct page *page; + int end_pfn = start_pfn + size; + int pfn; - for (page = start; page < (start + size); page++) { - set_page_links(page, zone, nid); + for (pfn = start_pfn; pfn < end_pfn; pfn++, page++) { + if (!early_pfn_valid(pfn)) + continue; + page = pfn_to_page(pfn); + set_page_links(page, zone, nid, pfn); set_page_count(page, 0); reset_page_mapcount(page); SetPageReserved(page); @@ -1677,6 +1681,20 @@ void zone_init_free_lists(struct pglist_data *pgdat, struct zone *zone, } } +#define ZONETABLE_INDEX(x, zone_nr) ((x << ZONES_SHIFT) | zone_nr) +void zonetable_add(struct zone *zone, int nid, int zid, unsigned long pfn, + unsigned long size) +{ + unsigned long snum = pfn_to_section_nr(pfn); + unsigned long end = pfn_to_section_nr(pfn + size); + + if (FLAGS_HAS_NODE) + zone_table[ZONETABLE_INDEX(nid, zid)] = zone; + else + for (; snum <= end; snum++) + zone_table[ZONETABLE_INDEX(snum, zid)] = zone; +} + #ifndef __HAVE_ARCH_MEMMAP_INIT #define memmap_init(size, nid, zone, start_pfn) \ memmap_init_zone((size), (nid), (zone), (start_pfn)) @@ -1861,7 +1879,6 @@ static void __init free_area_init_core(struct pglist_data *pgdat, unsigned long size, realsize; unsigned long batch; - zone_table[NODEZONE(nid, j)] = zone; realsize = size = zones_size[j]; if (zholes_size) realsize -= zholes_size[j]; @@ -1927,6 +1944,8 @@ static void __init free_area_init_core(struct pglist_data *pgdat, memmap_init(size, nid, j, zone_start_pfn); + zonetable_add(zone, nid, j, zone_start_pfn, size); + zone_start_pfn += size; zone_init_free_lists(pgdat, zone, zone->spanned_pages); @@ -1935,28 +1954,30 @@ static void __init free_area_init_core(struct pglist_data *pgdat, static void __init alloc_node_mem_map(struct pglist_data *pgdat) { - unsigned long size; - struct page *map; - /* Skip empty nodes */ if (!pgdat->node_spanned_pages) return; +#ifdef CONFIG_FLAT_NODE_MEM_MAP /* ia64 gets its own node_mem_map, before this, without bootmem */ if (!pgdat->node_mem_map) { + unsigned long size; + struct page *map; + size = (pgdat->node_spanned_pages + 1) * sizeof(struct page); map = alloc_remap(pgdat->node_id, size); if (!map) map = alloc_bootmem_node(pgdat, size); pgdat->node_mem_map = map; } -#ifndef CONFIG_DISCONTIGMEM +#ifdef CONFIG_FLATMEM /* * With no DISCONTIG, the global mem_map is just set as node 0's */ if (pgdat == NODE_DATA(0)) mem_map = NODE_DATA(0)->node_mem_map; #endif +#endif /* CONFIG_FLAT_NODE_MEM_MAP */ } void __init free_area_init_node(int nid, struct pglist_data *pgdat, diff --git a/mm/sparse.c b/mm/sparse.c new file mode 100644 index 000000000000..f888385b9e14 --- /dev/null +++ b/mm/sparse.c @@ -0,0 +1,85 @@ +/* + * sparse memory mappings. + */ +#include +#include +#include +#include +#include +#include + +/* + * Permanent SPARSEMEM data: + * + * 1) mem_section - memory sections, mem_map's for valid memory + */ +struct mem_section mem_section[NR_MEM_SECTIONS]; +EXPORT_SYMBOL(mem_section); + +/* Record a memory area against a node. */ +void memory_present(int nid, unsigned long start, unsigned long end) +{ + unsigned long pfn; + + start &= PAGE_SECTION_MASK; + for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION) { + unsigned long section = pfn_to_section_nr(pfn); + if (!mem_section[section].section_mem_map) + mem_section[section].section_mem_map = (void *) -1; + } +} + +/* + * Only used by the i386 NUMA architecures, but relatively + * generic code. + */ +unsigned long __init node_memmap_size_bytes(int nid, unsigned long start_pfn, + unsigned long end_pfn) +{ + unsigned long pfn; + unsigned long nr_pages = 0; + + for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) { + if (nid != early_pfn_to_nid(pfn)) + continue; + + if (pfn_valid(pfn)) + nr_pages += PAGES_PER_SECTION; + } + + return nr_pages * sizeof(struct page); +} + +/* + * Allocate the accumulated non-linear sections, allocate a mem_map + * for each and record the physical to section mapping. + */ +void sparse_init(void) +{ + unsigned long pnum; + struct page *map; + int nid; + + for (pnum = 0; pnum < NR_MEM_SECTIONS; pnum++) { + if (!mem_section[pnum].section_mem_map) + continue; + + nid = early_pfn_to_nid(section_nr_to_pfn(pnum)); + map = alloc_remap(nid, sizeof(struct page) * PAGES_PER_SECTION); + if (!map) + map = alloc_bootmem_node(NODE_DATA(nid), + sizeof(struct page) * PAGES_PER_SECTION); + if (!map) { + mem_section[pnum].section_mem_map = 0; + continue; + } + + /* + * Subtle, we encode the real pfn into the mem_map such that + * the identity pfn - section_mem_map will return the actual + * physical page frame number. + */ + mem_section[pnum].section_mem_map = map - + section_nr_to_pfn(pnum); + } +} -- cgit v1.2.3 From 641c767389b19859a45e6de46d8e18cd935bdb60 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Thu, 23 Jun 2005 00:07:59 -0700 Subject: [PATCH] sparsemem swiss cheese numa layouts The part of the sparsemem patch which modifies memmap_init_zone() has recently become a problem. It changes behavior so that there is a call to pfn_to_page() for each individual page inside of a node's range: node_start_pfn through node_end_pfn. It used to simply do this once, at the beginning of the node, but having sparsemem's non-contiguous mem_map[]s inside of a node made it necessary to change. Mike Kravetz recently wrote a patch which made the NUMA code accept some new kinds of layouts. The system's memory was laid out like this, with node 0's memory in two pieces: one before and one after node 1's memory: Node 0: +++++ +++++ Node 1: +++++ Previous behavior before Mike's patch was to assign nodes like this: Node 0: 00000 XXXXX Node 1: 11111 Where the 'X' areas were simply thrown away. The new behavior was to make the pg_data_t span node 0 across all of its areas, including areas that are really node 1's: Node 0: 000000000000000 Node 1: 11111 This wastes a little bit of mem_map space, but ends up being OK, and more fully utilizes the system's memory. memmap_init_zone() initializes all of the "struct page"s for node 0, even for the "hole", but those never get used, because there is no pfn_to_page() that resolves to those pages. However, only calling pfn_to_page() once, memmap_init_zone() always uses the pages that were allocated for node0->node_mem_map because: struct page *start = pfn_to_page(start_pfn); // effectively start = &node->node_mem_map[0] for (page = start; page < (start + size); page++) { init_page_here();... page++; } Slow, and wasteful, but generally harmless. But, modify that to call pfn_to_page() for each loop iteration (like sparsemem does): for (pfn = start_pfn; pfn < < (start_pfn + size); pfn++++) { page = pfn_to_page(pfn); } And you end up trying to initialize node 1's pages too early, along with bogus data from node 0. This patch checks for those weird layouts and declines to touch the pages, making the more frequent pfn_to_page() calls OK to do. Signed-off-by: Dave Hansen Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc64/Kconfig | 12 ++++++++++++ include/linux/mmzone.h | 6 ++++++ mm/page_alloc.c | 2 ++ 3 files changed, 20 insertions(+) (limited to 'include/linux') diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig index 011b5c0bf1d0..85f8fcf44b6c 100644 --- a/arch/ppc64/Kconfig +++ b/arch/ppc64/Kconfig @@ -211,6 +211,18 @@ config ARCH_FLATMEM_ENABLE source "mm/Kconfig" +# Some NUMA nodes have memory ranges that span +# other nodes. Even though a pfn is valid and +# between a node's start and end pfns, it may not +# reside on that node. +# +# This is a relatively temporary hack that should +# be able to go away when sparsemem is fully in +# place +config NODES_SPAN_OTHER_NODES + def_bool y + depends on NEED_MULTIPLE_NODES + config NUMA bool "NUMA support" depends on DISCONTIGMEM diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 19860d317ec2..746b57e3d370 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -528,6 +528,12 @@ void sparse_init(void); #define sparse_init() do {} while (0) #endif /* CONFIG_SPARSEMEM */ +#ifdef CONFIG_NODES_SPAN_OTHER_NODES +#define early_pfn_in_nid(pfn, nid) (early_pfn_to_nid(pfn) == (nid)) +#else +#define early_pfn_in_nid(pfn, nid) (1) +#endif + #ifndef early_pfn_valid #define early_pfn_valid(pfn) (1) #endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 5c1b8982a6da..1eb683f9b3af 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1656,6 +1656,8 @@ void __init memmap_init_zone(unsigned long size, int nid, unsigned long zone, for (pfn = start_pfn; pfn < end_pfn; pfn++, page++) { if (!early_pfn_valid(pfn)) continue; + if (!early_pfn_in_nid(pfn, nid)) + continue; page = pfn_to_page(pfn); set_page_links(page, zone, nid, pfn); set_page_count(page, 0); -- cgit v1.2.3 From 29751f6991e845f7d002a6ae520bf996b38c8dcd Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Thu, 23 Jun 2005 00:08:00 -0700 Subject: [PATCH] sparsemem hotplug base Make sparse's initalization be accessible at runtime. This allows sparse mappings to be created after boot in a hotplug situation. This patch is separated from the previous one just to give an indication how much of the sparse infrastructure is *just* for hotplug memory. The section_mem_map doesn't really store a pointer. It stores something that is convenient to do some math against to get a pointer. It isn't valid to just do *section_mem_map, so I don't think it should be stored as a pointer. There are a couple of things I'd like to store about a section. First of all, the fact that it is !NULL does not mean that it is present. There could be such a combination where section_mem_map *is* NULL, but the math gets you properly to a real mem_map. So, I don't think that check is safe. Since we're storing 32-bit-aligned structures, we have a few bits in the bottom of the pointer to play with. Use one bit to encode whether there's really a mem_map there, and the other one to tell whether there's a valid section there. We need to distinguish between the two because sometimes there's a gap between when a section is discovered to be present and when we can get the mem_map for it. Signed-off-by: Dave Hansen Signed-off-by: Andy Whitcroft Signed-off-by: Jack Steiner Signed-off-by: Bob Picco Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 56 +++++++++++++++++++++++++++--- mm/page_alloc.c | 4 +-- mm/sparse.c | 92 +++++++++++++++++++++++++++++++++++++++----------- 3 files changed, 125 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 746b57e3d370..6c90461ed99f 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -476,11 +476,56 @@ extern struct pglist_data contig_page_data; struct page; struct mem_section { - struct page *section_mem_map; + /* + * This is, logically, a pointer to an array of struct + * pages. However, it is stored with some other magic. + * (see sparse.c::sparse_init_one_section()) + * + * Making it a UL at least makes someone do a cast + * before using it wrong. + */ + unsigned long section_mem_map; }; extern struct mem_section mem_section[NR_MEM_SECTIONS]; +static inline struct mem_section *__nr_to_section(unsigned long nr) +{ + return &mem_section[nr]; +} + +/* + * We use the lower bits of the mem_map pointer to store + * a little bit of information. There should be at least + * 3 bits here due to 32-bit alignment. + */ +#define SECTION_MARKED_PRESENT (1UL<<0) +#define SECTION_HAS_MEM_MAP (1UL<<1) +#define SECTION_MAP_LAST_BIT (1UL<<2) +#define SECTION_MAP_MASK (~(SECTION_MAP_LAST_BIT-1)) + +static inline struct page *__section_mem_map_addr(struct mem_section *section) +{ + unsigned long map = section->section_mem_map; + map &= SECTION_MAP_MASK; + return (struct page *)map; +} + +static inline int valid_section(struct mem_section *section) +{ + return (section->section_mem_map & SECTION_MARKED_PRESENT); +} + +static inline int section_has_mem_map(struct mem_section *section) +{ + return (section->section_mem_map & SECTION_HAS_MEM_MAP); +} + +static inline int valid_section_nr(unsigned long nr) +{ + return valid_section(__nr_to_section(nr)); +} + /* * Given a kernel address, find the home node of the underlying memory. */ @@ -488,24 +533,25 @@ extern struct mem_section mem_section[NR_MEM_SECTIONS]; static inline struct mem_section *__pfn_to_section(unsigned long pfn) { - return &mem_section[pfn_to_section_nr(pfn)]; + return __nr_to_section(pfn_to_section_nr(pfn)); } #define pfn_to_page(pfn) \ ({ \ unsigned long __pfn = (pfn); \ - __pfn_to_section(__pfn)->section_mem_map + __pfn; \ + __section_mem_map_addr(__pfn_to_section(__pfn)) + __pfn; \ }) #define page_to_pfn(page) \ ({ \ - page - mem_section[page_to_section(page)].section_mem_map; \ + page - __section_mem_map_addr(__nr_to_section( \ + page_to_section(page))); \ }) static inline int pfn_valid(unsigned long pfn) { if (pfn_to_section_nr(pfn) >= NR_MEM_SECTIONS) return 0; - return mem_section[pfn_to_section_nr(pfn)].section_mem_map != 0; + return valid_section(__nr_to_section(pfn_to_section_nr(pfn))); } /* diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 1eb683f9b3af..7ee675ad101e 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1650,8 +1650,8 @@ void __init memmap_init_zone(unsigned long size, int nid, unsigned long zone, unsigned long start_pfn) { struct page *page; - int end_pfn = start_pfn + size; - int pfn; + unsigned long end_pfn = start_pfn + size; + unsigned long pfn; for (pfn = start_pfn; pfn < end_pfn; pfn++, page++) { if (!early_pfn_valid(pfn)) diff --git a/mm/sparse.c b/mm/sparse.c index f888385b9e14..b54e304df4a7 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -25,7 +25,7 @@ void memory_present(int nid, unsigned long start, unsigned long end) for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION) { unsigned long section = pfn_to_section_nr(pfn); if (!mem_section[section].section_mem_map) - mem_section[section].section_mem_map = (void *) -1; + mem_section[section].section_mem_map = SECTION_MARKED_PRESENT; } } @@ -50,6 +50,56 @@ unsigned long __init node_memmap_size_bytes(int nid, unsigned long start_pfn, return nr_pages * sizeof(struct page); } +/* + * Subtle, we encode the real pfn into the mem_map such that + * the identity pfn - section_mem_map will return the actual + * physical page frame number. + */ +static unsigned long sparse_encode_mem_map(struct page *mem_map, unsigned long pnum) +{ + return (unsigned long)(mem_map - (section_nr_to_pfn(pnum))); +} + +/* + * We need this if we ever free the mem_maps. While not implemented yet, + * this function is included for parity with its sibling. + */ +static __attribute((unused)) +struct page *sparse_decode_mem_map(unsigned long coded_mem_map, unsigned long pnum) +{ + return ((struct page *)coded_mem_map) + section_nr_to_pfn(pnum); +} + +static int sparse_init_one_section(struct mem_section *ms, + unsigned long pnum, struct page *mem_map) +{ + if (!valid_section(ms)) + return -EINVAL; + + ms->section_mem_map |= sparse_encode_mem_map(mem_map, pnum); + + return 1; +} + +static struct page *sparse_early_mem_map_alloc(unsigned long pnum) +{ + struct page *map; + int nid = early_pfn_to_nid(section_nr_to_pfn(pnum)); + + map = alloc_remap(nid, sizeof(struct page) * PAGES_PER_SECTION); + if (map) + return map; + + map = alloc_bootmem_node(NODE_DATA(nid), + sizeof(struct page) * PAGES_PER_SECTION); + if (map) + return map; + + printk(KERN_WARNING "%s: allocation failed\n", __FUNCTION__); + mem_section[pnum].section_mem_map = 0; + return NULL; +} + /* * Allocate the accumulated non-linear sections, allocate a mem_map * for each and record the physical to section mapping. @@ -58,28 +108,30 @@ void sparse_init(void) { unsigned long pnum; struct page *map; - int nid; for (pnum = 0; pnum < NR_MEM_SECTIONS; pnum++) { - if (!mem_section[pnum].section_mem_map) + if (!valid_section_nr(pnum)) continue; - nid = early_pfn_to_nid(section_nr_to_pfn(pnum)); - map = alloc_remap(nid, sizeof(struct page) * PAGES_PER_SECTION); - if (!map) - map = alloc_bootmem_node(NODE_DATA(nid), - sizeof(struct page) * PAGES_PER_SECTION); - if (!map) { - mem_section[pnum].section_mem_map = 0; - continue; - } - - /* - * Subtle, we encode the real pfn into the mem_map such that - * the identity pfn - section_mem_map will return the actual - * physical page frame number. - */ - mem_section[pnum].section_mem_map = map - - section_nr_to_pfn(pnum); + map = sparse_early_mem_map_alloc(pnum); + if (map) + sparse_init_one_section(&mem_section[pnum], pnum, map); } } + +/* + * returns the number of sections whose mem_maps were properly + * set. If this is <=0, then that means that the passed-in + * map was not consumed and must be freed. + */ +int sparse_add_one_section(unsigned long start_pfn, int nr_pages, struct page *map) +{ + struct mem_section *ms = __pfn_to_section(start_pfn); + + if (ms->section_mem_map & SECTION_MARKED_PRESENT) + return -EEXIST; + + ms->section_mem_map |= SECTION_MARKED_PRESENT; + + return sparse_init_one_section(ms, pfn_to_section_nr(start_pfn), map); +} -- cgit v1.2.3 From 1946089a109251655c5438d92c539bd2930e71ea Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 23 Jun 2005 00:08:19 -0700 Subject: [PATCH] NUMA aware block device control structure allocation Patch to allocate the control structures for for ide devices on the node of the device itself (for NUMA systems). The patch depends on the Slab API change patch by Manfred and me (in mm) and the pcidev_to_node patch that I posted today. Does some realignment too. Signed-off-by: Justin M. Forbes Signed-off-by: Christoph Lameter Signed-off-by: Pravin Shelar Signed-off-by: Shobhit Dayal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/as-iosched.c | 8 +++++--- drivers/block/deadline-iosched.c | 8 +++++--- drivers/block/genhd.c | 13 ++++++++++--- drivers/block/ll_rw_blk.c | 30 +++++++++++++++++++++++------- drivers/ide/ide-disk.c | 3 ++- drivers/ide/ide-probe.c | 8 +++++--- include/linux/blkdev.h | 6 +++++- include/linux/genhd.h | 1 + include/linux/ide.h | 2 +- include/linux/mempool.h | 11 ++++++++--- mm/mempool.c | 17 ++++++++++++----- 11 files changed, 77 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/as-iosched.c b/drivers/block/as-iosched.c index 638db06de2be..3410b4d294b9 100644 --- a/drivers/block/as-iosched.c +++ b/drivers/block/as-iosched.c @@ -1871,20 +1871,22 @@ static int as_init_queue(request_queue_t *q, elevator_t *e) if (!arq_pool) return -ENOMEM; - ad = kmalloc(sizeof(*ad), GFP_KERNEL); + ad = kmalloc_node(sizeof(*ad), GFP_KERNEL, q->node); if (!ad) return -ENOMEM; memset(ad, 0, sizeof(*ad)); ad->q = q; /* Identify what queue the data belongs to */ - ad->hash = kmalloc(sizeof(struct list_head)*AS_HASH_ENTRIES,GFP_KERNEL); + ad->hash = kmalloc_node(sizeof(struct list_head)*AS_HASH_ENTRIES, + GFP_KERNEL, q->node); if (!ad->hash) { kfree(ad); return -ENOMEM; } - ad->arq_pool = mempool_create(BLKDEV_MIN_RQ, mempool_alloc_slab, mempool_free_slab, arq_pool); + ad->arq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab, + mempool_free_slab, arq_pool, q->node); if (!ad->arq_pool) { kfree(ad->hash); kfree(ad); diff --git a/drivers/block/deadline-iosched.c b/drivers/block/deadline-iosched.c index 7f79f3dd0165..4bc2fea73273 100644 --- a/drivers/block/deadline-iosched.c +++ b/drivers/block/deadline-iosched.c @@ -711,18 +711,20 @@ static int deadline_init_queue(request_queue_t *q, elevator_t *e) if (!drq_pool) return -ENOMEM; - dd = kmalloc(sizeof(*dd), GFP_KERNEL); + dd = kmalloc_node(sizeof(*dd), GFP_KERNEL, q->node); if (!dd) return -ENOMEM; memset(dd, 0, sizeof(*dd)); - dd->hash = kmalloc(sizeof(struct list_head)*DL_HASH_ENTRIES,GFP_KERNEL); + dd->hash = kmalloc_node(sizeof(struct list_head)*DL_HASH_ENTRIES, + GFP_KERNEL, q->node); if (!dd->hash) { kfree(dd); return -ENOMEM; } - dd->drq_pool = mempool_create(BLKDEV_MIN_RQ, mempool_alloc_slab, mempool_free_slab, drq_pool); + dd->drq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab, + mempool_free_slab, drq_pool, q->node); if (!dd->drq_pool) { kfree(dd->hash); kfree(dd); diff --git a/drivers/block/genhd.c b/drivers/block/genhd.c index 53f7d846b747..43805e4d31e9 100644 --- a/drivers/block/genhd.c +++ b/drivers/block/genhd.c @@ -582,10 +582,16 @@ struct seq_operations diskstats_op = { .show = diskstats_show }; - struct gendisk *alloc_disk(int minors) { - struct gendisk *disk = kmalloc(sizeof(struct gendisk), GFP_KERNEL); + return alloc_disk_node(minors, -1); +} + +struct gendisk *alloc_disk_node(int minors, int node_id) +{ + struct gendisk *disk; + + disk = kmalloc_node(sizeof(struct gendisk), GFP_KERNEL, node_id); if (disk) { memset(disk, 0, sizeof(struct gendisk)); if (!init_disk_stats(disk)) { @@ -594,7 +600,7 @@ struct gendisk *alloc_disk(int minors) } if (minors > 1) { int size = (minors - 1) * sizeof(struct hd_struct *); - disk->part = kmalloc(size, GFP_KERNEL); + disk->part = kmalloc_node(size, GFP_KERNEL, node_id); if (!disk->part) { kfree(disk); return NULL; @@ -610,6 +616,7 @@ struct gendisk *alloc_disk(int minors) } EXPORT_SYMBOL(alloc_disk); +EXPORT_SYMBOL(alloc_disk_node); struct kobject *get_disk(struct gendisk *disk) { diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c index 81fe3a0c1fe7..cd8cf302068c 100644 --- a/drivers/block/ll_rw_blk.c +++ b/drivers/block/ll_rw_blk.c @@ -28,6 +28,7 @@ #include #include #include +#include /* * for max sense size @@ -1645,7 +1646,8 @@ static int blk_init_free_list(request_queue_t *q) init_waitqueue_head(&rl->wait[WRITE]); init_waitqueue_head(&rl->drain); - rl->rq_pool = mempool_create(BLKDEV_MIN_RQ, mempool_alloc_slab, mempool_free_slab, request_cachep); + rl->rq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab, + mempool_free_slab, request_cachep, q->node); if (!rl->rq_pool) return -ENOMEM; @@ -1657,8 +1659,15 @@ static int __make_request(request_queue_t *, struct bio *); request_queue_t *blk_alloc_queue(int gfp_mask) { - request_queue_t *q = kmem_cache_alloc(requestq_cachep, gfp_mask); + return blk_alloc_queue_node(gfp_mask, -1); +} +EXPORT_SYMBOL(blk_alloc_queue); + +request_queue_t *blk_alloc_queue_node(int gfp_mask, int node_id) +{ + request_queue_t *q; + q = kmem_cache_alloc_node(requestq_cachep, gfp_mask, node_id); if (!q) return NULL; @@ -1671,8 +1680,7 @@ request_queue_t *blk_alloc_queue(int gfp_mask) return q; } - -EXPORT_SYMBOL(blk_alloc_queue); +EXPORT_SYMBOL(blk_alloc_queue_node); /** * blk_init_queue - prepare a request queue for use with a block device @@ -1705,13 +1713,22 @@ EXPORT_SYMBOL(blk_alloc_queue); * blk_init_queue() must be paired with a blk_cleanup_queue() call * when the block device is deactivated (such as at module unload). **/ + request_queue_t *blk_init_queue(request_fn_proc *rfn, spinlock_t *lock) { - request_queue_t *q = blk_alloc_queue(GFP_KERNEL); + return blk_init_queue_node(rfn, lock, -1); +} +EXPORT_SYMBOL(blk_init_queue); + +request_queue_t * +blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id) +{ + request_queue_t *q = blk_alloc_queue_node(GFP_KERNEL, node_id); if (!q) return NULL; + q->node = node_id; if (blk_init_free_list(q)) goto out_init; @@ -1754,8 +1771,7 @@ out_init: kmem_cache_free(requestq_cachep, q); return NULL; } - -EXPORT_SYMBOL(blk_init_queue); +EXPORT_SYMBOL(blk_init_queue_node); int blk_get_queue(request_queue_t *q) { diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c index 3302cd8eab4c..d6f934886b04 100644 --- a/drivers/ide/ide-disk.c +++ b/drivers/ide/ide-disk.c @@ -1215,7 +1215,8 @@ static int ide_disk_probe(struct device *dev) if (!idkp) goto failed; - g = alloc_disk(1 << PARTN_BITS); + g = alloc_disk_node(1 << PARTN_BITS, + pcibus_to_node(drive->hwif->pci_dev->bus)); if (!g) goto out_free_idkp; diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c index 5d876f53c697..7df85af75371 100644 --- a/drivers/ide/ide-probe.c +++ b/drivers/ide/ide-probe.c @@ -977,8 +977,9 @@ static int ide_init_queue(ide_drive_t *drive) * limits and LBA48 we could raise it but as yet * do not. */ - - q = blk_init_queue(do_ide_request, &ide_lock); + + q = blk_init_queue_node(do_ide_request, &ide_lock, + pcibus_to_node(drive->hwif->pci_dev->bus)); if (!q) return 1; @@ -1095,7 +1096,8 @@ static int init_irq (ide_hwif_t *hwif) hwgroup->hwif->next = hwif; spin_unlock_irq(&ide_lock); } else { - hwgroup = kmalloc(sizeof(ide_hwgroup_t),GFP_KERNEL); + hwgroup = kmalloc_node(sizeof(ide_hwgroup_t), GFP_KERNEL, + pcibus_to_node(hwif->drives[0].hwif->pci_dev->bus)); if (!hwgroup) goto out_up; diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 4a99b76c5a33..235c3414d268 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -396,6 +396,7 @@ struct request_queue */ unsigned int sg_timeout; unsigned int sg_reserved_size; + int node; struct list_head drain_list; @@ -615,6 +616,8 @@ static inline void blkdev_dequeue_request(struct request *req) /* * Access functions for manipulating queue properties */ +extern request_queue_t *blk_init_queue_node(request_fn_proc *rfn, + spinlock_t *lock, int node_id); extern request_queue_t *blk_init_queue(request_fn_proc *, spinlock_t *); extern void blk_cleanup_queue(request_queue_t *); extern void blk_queue_make_request(request_queue_t *, make_request_fn *); @@ -646,7 +649,8 @@ extern void blk_wait_queue_drained(request_queue_t *, int); extern void blk_finish_queue_drain(request_queue_t *); int blk_get_queue(request_queue_t *); -request_queue_t *blk_alloc_queue(int); +request_queue_t *blk_alloc_queue(int gfp_mask); +request_queue_t *blk_alloc_queue_node(int,int); #define blk_put_queue(q) blk_cleanup_queue((q)) /* diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 47dedaf971d6..af26dc718ef6 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -403,6 +403,7 @@ extern int rescan_partitions(struct gendisk *disk, struct block_device *bdev); extern void add_partition(struct gendisk *, int, sector_t, sector_t); extern void delete_partition(struct gendisk *, int); +extern struct gendisk *alloc_disk_node(int minors, int node_id); extern struct gendisk *alloc_disk(int minors); extern struct kobject *get_disk(struct gendisk *disk); extern void put_disk(struct gendisk *disk); diff --git a/include/linux/ide.h b/include/linux/ide.h index 336d6e509f59..92129078d4f3 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -917,7 +917,7 @@ typedef struct hwif_s { unsigned dma; void (*led_act)(void *data, int rw); -} ide_hwif_t; +} ____cacheline_maxaligned_in_smp ide_hwif_t; /* * internal ide interrupt handler type diff --git a/include/linux/mempool.h b/include/linux/mempool.h index 4a36edf1c974..796220ce47cc 100644 --- a/include/linux/mempool.h +++ b/include/linux/mempool.h @@ -20,9 +20,14 @@ typedef struct mempool_s { mempool_free_t *free; wait_queue_head_t wait; } mempool_t; -extern mempool_t * mempool_create(int min_nr, mempool_alloc_t *alloc_fn, - mempool_free_t *free_fn, void *pool_data); -extern int mempool_resize(mempool_t *pool, int new_min_nr, unsigned int __nocast gfp_mask); + +extern mempool_t *mempool_create(int min_nr, mempool_alloc_t *alloc_fn, + mempool_free_t *free_fn, void *pool_data); +extern mempool_t *mempool_create_node(int min_nr, mempool_alloc_t *alloc_fn, + mempool_free_t *free_fn, void *pool_data, int nid); + +extern int mempool_resize(mempool_t *pool, int new_min_nr, + unsigned int __nocast gfp_mask); extern void mempool_destroy(mempool_t *pool); extern void * mempool_alloc(mempool_t *pool, unsigned int __nocast gfp_mask); extern void mempool_free(void *element, mempool_t *pool); diff --git a/mm/mempool.c b/mm/mempool.c index c9f3d4620428..920c8c3ab1b8 100644 --- a/mm/mempool.c +++ b/mm/mempool.c @@ -51,16 +51,23 @@ static void free_pool(mempool_t *pool) * functions might sleep - as long as the mempool_alloc function is not called * from IRQ contexts. */ -mempool_t * mempool_create(int min_nr, mempool_alloc_t *alloc_fn, +mempool_t *mempool_create(int min_nr, mempool_alloc_t *alloc_fn, mempool_free_t *free_fn, void *pool_data) { - mempool_t *pool; + return mempool_create_node(min_nr,alloc_fn,free_fn, pool_data,-1); +} +EXPORT_SYMBOL(mempool_create); - pool = kmalloc(sizeof(*pool), GFP_KERNEL); +mempool_t *mempool_create_node(int min_nr, mempool_alloc_t *alloc_fn, + mempool_free_t *free_fn, void *pool_data, int node_id) +{ + mempool_t *pool; + pool = kmalloc_node(sizeof(*pool), GFP_KERNEL, node_id); if (!pool) return NULL; memset(pool, 0, sizeof(*pool)); - pool->elements = kmalloc(min_nr * sizeof(void *), GFP_KERNEL); + pool->elements = kmalloc_node(min_nr * sizeof(void *), + GFP_KERNEL, node_id); if (!pool->elements) { kfree(pool); return NULL; @@ -87,7 +94,7 @@ mempool_t * mempool_create(int min_nr, mempool_alloc_t *alloc_fn, } return pool; } -EXPORT_SYMBOL(mempool_create); +EXPORT_SYMBOL(mempool_create_node); /** * mempool_resize - resize an existing memory pool -- cgit v1.2.3 From fa72b903f75e4f0f0b2c2feed093005167da4023 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 23 Jun 2005 00:08:49 -0700 Subject: [PATCH] blk: remove blk_queue_tag->real_max_depth optimization blk_queue_tag->real_max_depth was used to optimize out unnecessary allocations/frees on tag resize. However, the whole thing was very broken - tag_map was never allocated to real_max_depth resulting in access beyond the end of the map, bits in [max_depth..real_max_depth] were set when initializing a map and copied when resizing resulting in pre-occupied tags. As the gain of the optimization is very small, well, almost nill, remove the whole thing. Signed-off-by: Tejun Heo Acked-by: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/ll_rw_blk.c | 35 ++++++++++------------------------- include/linux/blkdev.h | 1 - 2 files changed, 10 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c index 808390c74200..896d17c28f42 100644 --- a/drivers/block/ll_rw_blk.c +++ b/drivers/block/ll_rw_blk.c @@ -717,7 +717,7 @@ struct request *blk_queue_find_tag(request_queue_t *q, int tag) { struct blk_queue_tag *bqt = q->queue_tags; - if (unlikely(bqt == NULL || tag >= bqt->real_max_depth)) + if (unlikely(bqt == NULL || tag >= bqt->max_depth)) return NULL; return bqt->tag_index[tag]; @@ -775,9 +775,9 @@ EXPORT_SYMBOL(blk_queue_free_tags); static int init_tag_map(request_queue_t *q, struct blk_queue_tag *tags, int depth) { - int bits, i; struct request **tag_index; unsigned long *tag_map; + int nr_ulongs; if (depth > q->nr_requests * 2) { depth = q->nr_requests * 2; @@ -789,24 +789,17 @@ init_tag_map(request_queue_t *q, struct blk_queue_tag *tags, int depth) if (!tag_index) goto fail; - bits = (depth / BLK_TAGS_PER_LONG) + 1; - tag_map = kmalloc(bits * sizeof(unsigned long), GFP_ATOMIC); + nr_ulongs = ALIGN(depth, BLK_TAGS_PER_LONG) / BLK_TAGS_PER_LONG; + tag_map = kmalloc(nr_ulongs * sizeof(unsigned long), GFP_ATOMIC); if (!tag_map) goto fail; memset(tag_index, 0, depth * sizeof(struct request *)); - memset(tag_map, 0, bits * sizeof(unsigned long)); + memset(tag_map, 0, nr_ulongs * sizeof(unsigned long)); tags->max_depth = depth; - tags->real_max_depth = bits * BITS_PER_LONG; tags->tag_index = tag_index; tags->tag_map = tag_map; - /* - * set the upper bits if the depth isn't a multiple of the word size - */ - for (i = depth; i < bits * BLK_TAGS_PER_LONG; i++) - __set_bit(i, tag_map); - return 0; fail: kfree(tag_index); @@ -871,32 +864,24 @@ int blk_queue_resize_tags(request_queue_t *q, int new_depth) struct blk_queue_tag *bqt = q->queue_tags; struct request **tag_index; unsigned long *tag_map; - int bits, max_depth; + int max_depth, nr_ulongs; if (!bqt) return -ENXIO; - /* - * don't bother sizing down - */ - if (new_depth <= bqt->real_max_depth) { - bqt->max_depth = new_depth; - return 0; - } - /* * save the old state info, so we can copy it back */ tag_index = bqt->tag_index; tag_map = bqt->tag_map; - max_depth = bqt->real_max_depth; + max_depth = bqt->max_depth; if (init_tag_map(q, bqt, new_depth)) return -ENOMEM; memcpy(bqt->tag_index, tag_index, max_depth * sizeof(struct request *)); - bits = max_depth / BLK_TAGS_PER_LONG; - memcpy(bqt->tag_map, tag_map, bits * sizeof(unsigned long)); + nr_ulongs = ALIGN(max_depth, BLK_TAGS_PER_LONG) / BLK_TAGS_PER_LONG; + memcpy(bqt->tag_map, tag_map, nr_ulongs * sizeof(unsigned long)); kfree(tag_index); kfree(tag_map); @@ -926,7 +911,7 @@ void blk_queue_end_tag(request_queue_t *q, struct request *rq) BUG_ON(tag == -1); - if (unlikely(tag >= bqt->real_max_depth)) + if (unlikely(tag >= bqt->max_depth)) return; if (unlikely(!__test_and_clear_bit(tag, bqt->tag_map))) { diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 235c3414d268..8d7e2f4151d0 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -294,7 +294,6 @@ struct blk_queue_tag { struct list_head busy_list; /* fifo list of busy tags */ int busy; /* current depth */ int max_depth; /* what we will send to device */ - int real_max_depth; /* what the array can hold */ atomic_t refcnt; /* map can be shared */ }; -- cgit v1.2.3 From f7d37d028dfba90b1b747f8ac685bf0959aeda8b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 23 Jun 2005 00:08:50 -0700 Subject: [PATCH] blk: remove BLK_TAGS_{PER_LONG|MASK} Replace BLK_TAGS_PER_LONG with BITS_PER_LONG and remove unused BLK_TAGS_MASK. Signed-off-by: Tejun Heo Acked-by: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/ll_rw_blk.c | 4 ++-- include/linux/blkdev.h | 3 --- 2 files changed, 2 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c index 896d17c28f42..99afeec1031f 100644 --- a/drivers/block/ll_rw_blk.c +++ b/drivers/block/ll_rw_blk.c @@ -789,7 +789,7 @@ init_tag_map(request_queue_t *q, struct blk_queue_tag *tags, int depth) if (!tag_index) goto fail; - nr_ulongs = ALIGN(depth, BLK_TAGS_PER_LONG) / BLK_TAGS_PER_LONG; + nr_ulongs = ALIGN(depth, BITS_PER_LONG) / BITS_PER_LONG; tag_map = kmalloc(nr_ulongs * sizeof(unsigned long), GFP_ATOMIC); if (!tag_map) goto fail; @@ -880,7 +880,7 @@ int blk_queue_resize_tags(request_queue_t *q, int new_depth) return -ENOMEM; memcpy(bqt->tag_index, tag_index, max_depth * sizeof(struct request *)); - nr_ulongs = ALIGN(max_depth, BLK_TAGS_PER_LONG) / BLK_TAGS_PER_LONG; + nr_ulongs = ALIGN(max_depth, BITS_PER_LONG) / BITS_PER_LONG; memcpy(bqt->tag_map, tag_map, nr_ulongs * sizeof(unsigned long)); kfree(tag_index); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 8d7e2f4151d0..60272141ff19 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -285,9 +285,6 @@ enum blk_queue_state { Queue_up, }; -#define BLK_TAGS_PER_LONG (sizeof(unsigned long) * 8) -#define BLK_TAGS_MASK (BLK_TAGS_PER_LONG - 1) - struct blk_queue_tag { struct request **tag_index; /* map of busy tags */ unsigned long *tag_map; /* bit map of free/busy tags */ -- cgit v1.2.3 From 55c888d6d09a0df236adfaf8ccf06ff5d0646775 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 23 Jun 2005 00:08:56 -0700 Subject: [PATCH] timers fixes/improvements This patch tries to solve following problems: 1. del_timer_sync() is racy. The timer can be fired again after del_timer_sync have checked all cpus and before it will recheck timer_pending(). 2. It has scalability problems. All cpus are scanned to determine if the timer is running on that cpu. With this patch del_timer_sync is O(1) and no slower than plain del_timer(pending_timer), unless it has to actually wait for completion of the currently running timer. The only restriction is that the recurring timer should not use add_timer_on(). 3. The timers are not serialized wrt to itself. If CPU_0 does mod_timer(jiffies+1) while the timer is currently running on CPU 1, it is quite possible that local interrupt on CPU_0 will start that timer before it finished on CPU_1. 4. The timers locking is suboptimal. __mod_timer() takes 3 locks at once and still requires wmb() in del_timer/run_timers. The new implementation takes 2 locks sequentially and does not need memory barriers. Currently ->base != NULL means that the timer is pending. In that case ->base.lock is used to lock the timer. __mod_timer also takes timer->lock because ->base can be == NULL. This patch uses timer->entry.next != NULL as indication that the timer is pending. So it does __list_del(), entry->next = NULL instead of list_del() when the timer is deleted. The ->base field is used for hashed locking only, it is initialized in init_timer() which sets ->base = per_cpu(tvec_bases). When the tvec_bases.lock is locked, it means that all timers which are tied to this base via timer->base are locked, and the base itself is locked too. So __run_timers/migrate_timers can safely modify all timers which could be found on ->tvX lists (pending timers). When the timer's base is locked, and the timer removed from ->entry list (which means that _run_timers/migrate_timers can't see this timer), it is possible to set timer->base = NULL and drop the lock: the timer remains locked. This patch adds lock_timer_base() helper, which waits for ->base != NULL, locks the ->base, and checks it is still the same. __mod_timer() schedules the timer on the local CPU and changes it's base. However, it does not lock both old and new bases at once. It locks the timer via lock_timer_base(), deletes the timer, sets ->base = NULL, and unlocks old base. Then __mod_timer() locks new_base, sets ->base = new_base, and adds this timer. This simplifies the code, because AB-BA deadlock is not possible. __mod_timer() also ensures that the timer's base is not changed while the timer's handler is running on the old base. __run_timers(), del_timer() do not change ->base anymore, they only clear pending flag. So del_timer_sync() can test timer->base->running_timer == timer to detect whether it is running or not. We don't need timer_list->lock anymore, this patch kills it. We also don't need barriers. del_timer() and __run_timers() used smp_wmb() before clearing timer's pending flag. It was needed because __mod_timer() did not lock old_base if the timer is not pending, so __mod_timer()->list_add() could race with del_timer()->list_del(). With this patch these functions are serialized through base->lock. One problem. TIMER_INITIALIZER can't use per_cpu(tvec_bases). So this patch adds global struct timer_base_s { spinlock_t lock; struct timer_list *running_timer; } __init_timer_base; which is used by TIMER_INITIALIZER. The corresponding fields in tvec_t_base_s struct are replaced by struct timer_base_s t_base. It is indeed ugly. But this can't have scalability problems. The global __init_timer_base.lock is used only when __mod_timer() is called for the first time AND the timer was compile time initialized. After that the timer migrates to the local CPU. Signed-off-by: Oleg Nesterov Acked-by: Ingo Molnar Signed-off-by: Renaud Lienhart Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/timer.h | 30 ++--- kernel/timer.c | 328 ++++++++++++++++++++++++-------------------------- 2 files changed, 166 insertions(+), 192 deletions(-) (limited to 'include/linux') diff --git a/include/linux/timer.h b/include/linux/timer.h index 90db1cc62ddd..2e78fedfc069 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -6,45 +6,33 @@ #include #include -struct tvec_t_base_s; +struct timer_base_s; struct timer_list { struct list_head entry; unsigned long expires; - spinlock_t lock; unsigned long magic; void (*function)(unsigned long); unsigned long data; - struct tvec_t_base_s *base; + struct timer_base_s *base; }; #define TIMER_MAGIC 0x4b87ad6e +extern struct timer_base_s __init_timer_base; + #define TIMER_INITIALIZER(_function, _expires, _data) { \ .function = (_function), \ .expires = (_expires), \ .data = (_data), \ - .base = NULL, \ + .base = &__init_timer_base, \ .magic = TIMER_MAGIC, \ - .lock = SPIN_LOCK_UNLOCKED, \ } -/*** - * init_timer - initialize a timer. - * @timer: the timer to be initialized - * - * init_timer() must be done to a timer prior calling *any* of the - * other timer functions. - */ -static inline void init_timer(struct timer_list * timer) -{ - timer->base = NULL; - timer->magic = TIMER_MAGIC; - spin_lock_init(&timer->lock); -} +void fastcall init_timer(struct timer_list * timer); /*** * timer_pending - is a timer pending? @@ -58,7 +46,7 @@ static inline void init_timer(struct timer_list * timer) */ static inline int timer_pending(const struct timer_list * timer) { - return timer->base != NULL; + return timer->entry.next != NULL; } extern void add_timer_on(struct timer_list *timer, int cpu); @@ -89,12 +77,12 @@ static inline void add_timer(struct timer_list * timer) #ifdef CONFIG_SMP extern int del_timer_sync(struct timer_list *timer); - extern int del_singleshot_timer_sync(struct timer_list *timer); #else # define del_timer_sync(t) del_timer(t) -# define del_singleshot_timer_sync(t) del_timer(t) #endif +#define del_singleshot_timer_sync(t) del_timer_sync(t) + extern void init_timers(void); extern void run_local_timers(void); extern void it_real_fn(unsigned long); diff --git a/kernel/timer.c b/kernel/timer.c index 207aa4f0aa10..8aadc62efd65 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -57,6 +57,11 @@ static void time_interpolator_update(long delta_nsec); #define TVN_MASK (TVN_SIZE - 1) #define TVR_MASK (TVR_SIZE - 1) +struct timer_base_s { + spinlock_t lock; + struct timer_list *running_timer; +}; + typedef struct tvec_s { struct list_head vec[TVN_SIZE]; } tvec_t; @@ -66,9 +71,8 @@ typedef struct tvec_root_s { } tvec_root_t; struct tvec_t_base_s { - spinlock_t lock; + struct timer_base_s t_base; unsigned long timer_jiffies; - struct timer_list *running_timer; tvec_root_t tv1; tvec_t tv2; tvec_t tv3; @@ -77,18 +81,16 @@ struct tvec_t_base_s { } ____cacheline_aligned_in_smp; typedef struct tvec_t_base_s tvec_base_t; +static DEFINE_PER_CPU(tvec_base_t, tvec_bases); static inline void set_running_timer(tvec_base_t *base, struct timer_list *timer) { #ifdef CONFIG_SMP - base->running_timer = timer; + base->t_base.running_timer = timer; #endif } -/* Fake initialization */ -static DEFINE_PER_CPU(tvec_base_t, tvec_bases) = { SPIN_LOCK_UNLOCKED }; - static void check_timer_failed(struct timer_list *timer) { static int whine_count; @@ -103,7 +105,6 @@ static void check_timer_failed(struct timer_list *timer) /* * Now fix it up */ - spin_lock_init(&timer->lock); timer->magic = TIMER_MAGIC; } @@ -156,65 +157,113 @@ static void internal_add_timer(tvec_base_t *base, struct timer_list *timer) list_add_tail(&timer->entry, vec); } +typedef struct timer_base_s timer_base_t; +/* + * Used by TIMER_INITIALIZER, we can't use per_cpu(tvec_bases) + * at compile time, and we need timer->base to lock the timer. + */ +timer_base_t __init_timer_base + ____cacheline_aligned_in_smp = { .lock = SPIN_LOCK_UNLOCKED }; +EXPORT_SYMBOL(__init_timer_base); + +/*** + * init_timer - initialize a timer. + * @timer: the timer to be initialized + * + * init_timer() must be done to a timer prior calling *any* of the + * other timer functions. + */ +void fastcall init_timer(struct timer_list *timer) +{ + timer->entry.next = NULL; + timer->base = &per_cpu(tvec_bases, raw_smp_processor_id()).t_base; + timer->magic = TIMER_MAGIC; +} +EXPORT_SYMBOL(init_timer); + +static inline void detach_timer(struct timer_list *timer, + int clear_pending) +{ + struct list_head *entry = &timer->entry; + + __list_del(entry->prev, entry->next); + if (clear_pending) + entry->next = NULL; + entry->prev = LIST_POISON2; +} + +/* + * We are using hashed locking: holding per_cpu(tvec_bases).t_base.lock + * means that all timers which are tied to this base via timer->base are + * locked, and the base itself is locked too. + * + * So __run_timers/migrate_timers can safely modify all timers which could + * be found on ->tvX lists. + * + * When the timer's base is locked, and the timer removed from list, it is + * possible to set timer->base = NULL and drop the lock: the timer remains + * locked. + */ +static timer_base_t *lock_timer_base(struct timer_list *timer, + unsigned long *flags) +{ + timer_base_t *base; + + for (;;) { + base = timer->base; + if (likely(base != NULL)) { + spin_lock_irqsave(&base->lock, *flags); + if (likely(base == timer->base)) + return base; + /* The timer has migrated to another CPU */ + spin_unlock_irqrestore(&base->lock, *flags); + } + cpu_relax(); + } +} + int __mod_timer(struct timer_list *timer, unsigned long expires) { - tvec_base_t *old_base, *new_base; + timer_base_t *base; + tvec_base_t *new_base; unsigned long flags; int ret = 0; BUG_ON(!timer->function); - check_timer(timer); - spin_lock_irqsave(&timer->lock, flags); + base = lock_timer_base(timer, &flags); + + if (timer_pending(timer)) { + detach_timer(timer, 0); + ret = 1; + } + new_base = &__get_cpu_var(tvec_bases); -repeat: - old_base = timer->base; - /* - * Prevent deadlocks via ordering by old_base < new_base. - */ - if (old_base && (new_base != old_base)) { - if (old_base < new_base) { - spin_lock(&new_base->lock); - spin_lock(&old_base->lock); - } else { - spin_lock(&old_base->lock); - spin_lock(&new_base->lock); - } + if (base != &new_base->t_base) { /* - * The timer base might have been cancelled while we were - * trying to take the lock(s): + * We are trying to schedule the timer on the local CPU. + * However we can't change timer's base while it is running, + * otherwise del_timer_sync() can't detect that the timer's + * handler yet has not finished. This also guarantees that + * the timer is serialized wrt itself. */ - if (timer->base != old_base) { - spin_unlock(&new_base->lock); - spin_unlock(&old_base->lock); - goto repeat; - } - } else { - spin_lock(&new_base->lock); - if (timer->base != old_base) { - spin_unlock(&new_base->lock); - goto repeat; + if (unlikely(base->running_timer == timer)) { + /* The timer remains on a former base */ + new_base = container_of(base, tvec_base_t, t_base); + } else { + /* See the comment in lock_timer_base() */ + timer->base = NULL; + spin_unlock(&base->lock); + spin_lock(&new_base->t_base.lock); + timer->base = &new_base->t_base; } } - /* - * Delete the previous timeout (if there was any), and install - * the new one: - */ - if (old_base) { - list_del(&timer->entry); - ret = 1; - } timer->expires = expires; internal_add_timer(new_base, timer); - timer->base = new_base; - - if (old_base && (new_base != old_base)) - spin_unlock(&old_base->lock); - spin_unlock(&new_base->lock); - spin_unlock_irqrestore(&timer->lock, flags); + spin_unlock_irqrestore(&new_base->t_base.lock, flags); return ret; } @@ -232,15 +281,15 @@ void add_timer_on(struct timer_list *timer, int cpu) { tvec_base_t *base = &per_cpu(tvec_bases, cpu); unsigned long flags; - + BUG_ON(timer_pending(timer) || !timer->function); check_timer(timer); - spin_lock_irqsave(&base->lock, flags); + spin_lock_irqsave(&base->t_base.lock, flags); + timer->base = &base->t_base; internal_add_timer(base, timer); - timer->base = base; - spin_unlock_irqrestore(&base->lock, flags); + spin_unlock_irqrestore(&base->t_base.lock, flags); } @@ -295,27 +344,22 @@ EXPORT_SYMBOL(mod_timer); */ int del_timer(struct timer_list *timer) { + timer_base_t *base; unsigned long flags; - tvec_base_t *base; + int ret = 0; check_timer(timer); -repeat: - base = timer->base; - if (!base) - return 0; - spin_lock_irqsave(&base->lock, flags); - if (base != timer->base) { + if (timer_pending(timer)) { + base = lock_timer_base(timer, &flags); + if (timer_pending(timer)) { + detach_timer(timer, 1); + ret = 1; + } spin_unlock_irqrestore(&base->lock, flags); - goto repeat; } - list_del(&timer->entry); - /* Need to make sure that anybody who sees a NULL base also sees the list ops */ - smp_wmb(); - timer->base = NULL; - spin_unlock_irqrestore(&base->lock, flags); - return 1; + return ret; } EXPORT_SYMBOL(del_timer); @@ -332,72 +376,39 @@ EXPORT_SYMBOL(del_timer); * Synchronization rules: callers must prevent restarting of the timer, * otherwise this function is meaningless. It must not be called from * interrupt contexts. The caller must not hold locks which would prevent - * completion of the timer's handler. Upon exit the timer is not queued and - * the handler is not running on any CPU. + * completion of the timer's handler. The timer's handler must not call + * add_timer_on(). Upon exit the timer is not queued and the handler is + * not running on any CPU. * * The function returns whether it has deactivated a pending timer or not. - * - * del_timer_sync() is slow and complicated because it copes with timer - * handlers which re-arm the timer (periodic timers). If the timer handler - * is known to not do this (a single shot timer) then use - * del_singleshot_timer_sync() instead. */ int del_timer_sync(struct timer_list *timer) { - tvec_base_t *base; - int i, ret = 0; + timer_base_t *base; + unsigned long flags; + int ret = -1; check_timer(timer); -del_again: - ret += del_timer(timer); + do { + base = lock_timer_base(timer, &flags); - for_each_online_cpu(i) { - base = &per_cpu(tvec_bases, i); - if (base->running_timer == timer) { - while (base->running_timer == timer) { - cpu_relax(); - preempt_check_resched(); - } - break; + if (base->running_timer == timer) + goto unlock; + + ret = 0; + if (timer_pending(timer)) { + detach_timer(timer, 1); + ret = 1; } - } - smp_rmb(); - if (timer_pending(timer)) - goto del_again; +unlock: + spin_unlock_irqrestore(&base->lock, flags); + } while (ret < 0); return ret; } -EXPORT_SYMBOL(del_timer_sync); -/*** - * del_singleshot_timer_sync - deactivate a non-recursive timer - * @timer: the timer to be deactivated - * - * This function is an optimization of del_timer_sync for the case where the - * caller can guarantee the timer does not reschedule itself in its timer - * function. - * - * Synchronization rules: callers must prevent restarting of the timer, - * otherwise this function is meaningless. It must not be called from - * interrupt contexts. The caller must not hold locks which wold prevent - * completion of the timer's handler. Upon exit the timer is not queued and - * the handler is not running on any CPU. - * - * The function returns whether it has deactivated a pending timer or not. - */ -int del_singleshot_timer_sync(struct timer_list *timer) -{ - int ret = del_timer(timer); - - if (!ret) { - ret = del_timer_sync(timer); - BUG_ON(ret); - } - - return ret; -} -EXPORT_SYMBOL(del_singleshot_timer_sync); +EXPORT_SYMBOL(del_timer_sync); #endif static int cascade(tvec_base_t *base, tvec_t *tv, int index) @@ -415,7 +426,7 @@ static int cascade(tvec_base_t *base, tvec_t *tv, int index) struct timer_list *tmp; tmp = list_entry(curr, struct timer_list, entry); - BUG_ON(tmp->base != base); + BUG_ON(tmp->base != &base->t_base); curr = curr->next; internal_add_timer(base, tmp); } @@ -437,7 +448,7 @@ static inline void __run_timers(tvec_base_t *base) { struct timer_list *timer; - spin_lock_irq(&base->lock); + spin_lock_irq(&base->t_base.lock); while (time_after_eq(jiffies, base->timer_jiffies)) { struct list_head work_list = LIST_HEAD_INIT(work_list); struct list_head *head = &work_list; @@ -453,8 +464,7 @@ static inline void __run_timers(tvec_base_t *base) cascade(base, &base->tv5, INDEX(3)); ++base->timer_jiffies; list_splice_init(base->tv1.vec + index, &work_list); -repeat: - if (!list_empty(head)) { + while (!list_empty(head)) { void (*fn)(unsigned long); unsigned long data; @@ -462,11 +472,9 @@ repeat: fn = timer->function; data = timer->data; - list_del(&timer->entry); set_running_timer(base, timer); - smp_wmb(); - timer->base = NULL; - spin_unlock_irq(&base->lock); + detach_timer(timer, 1); + spin_unlock_irq(&base->t_base.lock); { u32 preempt_count = preempt_count(); fn(data); @@ -475,12 +483,11 @@ repeat: BUG(); } } - spin_lock_irq(&base->lock); - goto repeat; + spin_lock_irq(&base->t_base.lock); } } set_running_timer(base, NULL); - spin_unlock_irq(&base->lock); + spin_unlock_irq(&base->t_base.lock); } #ifdef CONFIG_NO_IDLE_HZ @@ -499,7 +506,7 @@ unsigned long next_timer_interrupt(void) int i, j; base = &__get_cpu_var(tvec_bases); - spin_lock(&base->lock); + spin_lock(&base->t_base.lock); expires = base->timer_jiffies + (LONG_MAX >> 1); list = 0; @@ -547,7 +554,7 @@ found: expires = nte->expires; } } - spin_unlock(&base->lock); + spin_unlock(&base->t_base.lock); return expires; } #endif @@ -1286,9 +1293,9 @@ static void __devinit init_timers_cpu(int cpu) { int j; tvec_base_t *base; - + base = &per_cpu(tvec_bases, cpu); - spin_lock_init(&base->lock); + spin_lock_init(&base->t_base.lock); for (j = 0; j < TVN_SIZE; j++) { INIT_LIST_HEAD(base->tv5.vec + j); INIT_LIST_HEAD(base->tv4.vec + j); @@ -1302,22 +1309,16 @@ static void __devinit init_timers_cpu(int cpu) } #ifdef CONFIG_HOTPLUG_CPU -static int migrate_timer_list(tvec_base_t *new_base, struct list_head *head) +static void migrate_timer_list(tvec_base_t *new_base, struct list_head *head) { struct timer_list *timer; while (!list_empty(head)) { timer = list_entry(head->next, struct timer_list, entry); - /* We're locking backwards from __mod_timer order here, - beware deadlock. */ - if (!spin_trylock(&timer->lock)) - return 0; - list_del(&timer->entry); + detach_timer(timer, 0); + timer->base = &new_base->t_base; internal_add_timer(new_base, timer); - timer->base = new_base; - spin_unlock(&timer->lock); } - return 1; } static void __devinit migrate_timers(int cpu) @@ -1331,39 +1332,24 @@ static void __devinit migrate_timers(int cpu) new_base = &get_cpu_var(tvec_bases); local_irq_disable(); -again: - /* Prevent deadlocks via ordering by old_base < new_base. */ - if (old_base < new_base) { - spin_lock(&new_base->lock); - spin_lock(&old_base->lock); - } else { - spin_lock(&old_base->lock); - spin_lock(&new_base->lock); - } + spin_lock(&new_base->t_base.lock); + spin_lock(&old_base->t_base.lock); - if (old_base->running_timer) + if (old_base->t_base.running_timer) BUG(); for (i = 0; i < TVR_SIZE; i++) - if (!migrate_timer_list(new_base, old_base->tv1.vec + i)) - goto unlock_again; - for (i = 0; i < TVN_SIZE; i++) - if (!migrate_timer_list(new_base, old_base->tv2.vec + i) - || !migrate_timer_list(new_base, old_base->tv3.vec + i) - || !migrate_timer_list(new_base, old_base->tv4.vec + i) - || !migrate_timer_list(new_base, old_base->tv5.vec + i)) - goto unlock_again; - spin_unlock(&old_base->lock); - spin_unlock(&new_base->lock); + migrate_timer_list(new_base, old_base->tv1.vec + i); + for (i = 0; i < TVN_SIZE; i++) { + migrate_timer_list(new_base, old_base->tv2.vec + i); + migrate_timer_list(new_base, old_base->tv3.vec + i); + migrate_timer_list(new_base, old_base->tv4.vec + i); + migrate_timer_list(new_base, old_base->tv5.vec + i); + } + + spin_unlock(&old_base->t_base.lock); + spin_unlock(&new_base->t_base.lock); local_irq_enable(); put_cpu_var(tvec_bases); - return; - -unlock_again: - /* Avoid deadlock with __mod_timer, by backing off. */ - spin_unlock(&old_base->lock); - spin_unlock(&new_base->lock); - cpu_relax(); - goto again; } #endif /* CONFIG_HOTPLUG_CPU */ -- cgit v1.2.3 From fd450b7318b75343fd76b3d95416853e34e72c95 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 23 Jun 2005 00:08:59 -0700 Subject: [PATCH] timers: introduce try_to_del_timer_sync() This patch splits del_timer_sync() into 2 functions. The new one, try_to_del_timer_sync(), returns -1 when it hits executing timer. It can be used in interrupt context, or when the caller hold locks which can prevent completion of the timer's handler. NOTE. Currently it can't be used in interrupt context in UP case, because ->running_timer is used only with CONFIG_SMP. Should the need arise, it is possible to kill #ifdef CONFIG_SMP in set_running_timer(), it is cheap. Signed-off-by: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/timer.h | 4 +++- kernel/timer.c | 53 ++++++++++++++++++++++++++++++++------------------- 2 files changed, 36 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/timer.h b/include/linux/timer.h index 2e78fedfc069..221f81ac2002 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -76,9 +76,11 @@ static inline void add_timer(struct timer_list * timer) } #ifdef CONFIG_SMP + extern int try_to_del_timer_sync(struct timer_list *timer); extern int del_timer_sync(struct timer_list *timer); #else -# define del_timer_sync(t) del_timer(t) +# define try_to_del_timer_sync(t) del_timer(t) +# define del_timer_sync(t) del_timer(t) #endif #define del_singleshot_timer_sync(t) del_timer_sync(t) diff --git a/kernel/timer.c b/kernel/timer.c index 8aadc62efd65..1f986c16d89f 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -365,6 +365,34 @@ int del_timer(struct timer_list *timer) EXPORT_SYMBOL(del_timer); #ifdef CONFIG_SMP +/* + * This function tries to deactivate a timer. Upon successful (ret >= 0) + * exit the timer is not queued and the handler is not running on any CPU. + * + * It must not be called from interrupt contexts. + */ +int try_to_del_timer_sync(struct timer_list *timer) +{ + timer_base_t *base; + unsigned long flags; + int ret = -1; + + base = lock_timer_base(timer, &flags); + + if (base->running_timer == timer) + goto out; + + ret = 0; + if (timer_pending(timer)) { + detach_timer(timer, 1); + ret = 1; + } +out: + spin_unlock_irqrestore(&base->lock, flags); + + return ret; +} + /*** * del_timer_sync - deactivate a timer and wait for the handler to finish. * @timer: the timer to be deactivated @@ -384,28 +412,13 @@ EXPORT_SYMBOL(del_timer); */ int del_timer_sync(struct timer_list *timer) { - timer_base_t *base; - unsigned long flags; - int ret = -1; - check_timer(timer); - do { - base = lock_timer_base(timer, &flags); - - if (base->running_timer == timer) - goto unlock; - - ret = 0; - if (timer_pending(timer)) { - detach_timer(timer, 1); - ret = 1; - } -unlock: - spin_unlock_irqrestore(&base->lock, flags); - } while (ret < 0); - - return ret; + for (;;) { + int ret = try_to_del_timer_sync(timer); + if (ret >= 0) + return ret; + } } EXPORT_SYMBOL(del_timer_sync); -- cgit v1.2.3 From 991114c6fa6a21d1fa4d544abe78592352860c82 Mon Sep 17 00:00:00 2001 From: Alexander Viro Date: Thu, 23 Jun 2005 00:09:01 -0700 Subject: [PATCH] fix for prune_icache()/forced final iput() races Based on analysis and a patch from Russ Weight There is a race condition that can occur if an inode is allocated and then released (using iput) during the ->fill_super functions. The race condition is between kswapd and mount. For most filesystems this can only happen in an error path when kswapd is running concurrently. For isofs, however, the error can occur in a more common code path (which is how the bug was found). The logic here is "we want final iput() to free inode *now* instead of letting it sit in cache if fs is going down or had not quite come up". The problem is with kswapd seeing such inodes in the middle of being killed and happily taking over. The clean solution would be to tell kswapd to leave those inodes alone and let our final iput deal with them. I.e. add a new flag (I_FORCED_FREEING), set it before write_inode_now() there and make prune_icache() leave those alone. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/inode.c | 16 ++++++++++------ include/linux/fs.h | 1 + 2 files changed, 11 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/fs/inode.c b/fs/inode.c index 801fe7f36280..1f9a3a2b89bc 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -500,7 +500,7 @@ repeat: continue; if (!test(inode, data)) continue; - if (inode->i_state & (I_FREEING|I_CLEAR)) { + if (inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) { __wait_on_freeing_inode(inode); goto repeat; } @@ -525,7 +525,7 @@ repeat: continue; if (inode->i_sb != sb) continue; - if (inode->i_state & (I_FREEING|I_CLEAR)) { + if (inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) { __wait_on_freeing_inode(inode); goto repeat; } @@ -727,7 +727,7 @@ EXPORT_SYMBOL(iunique); struct inode *igrab(struct inode *inode) { spin_lock(&inode_lock); - if (!(inode->i_state & I_FREEING)) + if (!(inode->i_state & (I_FREEING|I_WILL_FREE))) __iget(inode); else /* @@ -1024,17 +1024,21 @@ static void generic_forget_inode(struct inode *inode) if (!(inode->i_state & (I_DIRTY|I_LOCK))) list_move(&inode->i_list, &inode_unused); inodes_stat.nr_unused++; - spin_unlock(&inode_lock); - if (!sb || (sb->s_flags & MS_ACTIVE)) + if (!sb || (sb->s_flags & MS_ACTIVE)) { + spin_unlock(&inode_lock); return; + } + inode->i_state |= I_WILL_FREE; + spin_unlock(&inode_lock); write_inode_now(inode, 1); spin_lock(&inode_lock); + inode->i_state &= ~I_WILL_FREE; inodes_stat.nr_unused--; hlist_del_init(&inode->i_hash); } list_del_init(&inode->i_list); list_del_init(&inode->i_sb_list); - inode->i_state|=I_FREEING; + inode->i_state |= I_FREEING; inodes_stat.nr_inodes--; spin_unlock(&inode_lock); if (inode->i_data.nrpages) diff --git a/include/linux/fs.h b/include/linux/fs.h index e5a8db00df29..3622e952e98c 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1025,6 +1025,7 @@ struct super_operations { #define I_FREEING 16 #define I_CLEAR 32 #define I_NEW 64 +#define I_WILL_FREE 128 #define I_DIRTY (I_DIRTY_SYNC | I_DIRTY_DATASYNC | I_DIRTY_PAGES) -- cgit v1.2.3 From 543537bd922692bc978e2e356fcd8bfc9c2ee7d5 Mon Sep 17 00:00:00 2001 From: Paulo Marques Date: Thu, 23 Jun 2005 00:09:02 -0700 Subject: [PATCH] create a kstrdup library function This patch creates a new kstrdup library function and changes the "local" implementations in several places to use this function. Most of the changes come from the sound and net subsystems. The sound part had already been acknowledged by Takashi Iwai and the net part by David S. Miller. I left UML alone for now because I would need more time to read the code carefully before making changes there. Signed-off-by: Paulo Marques Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/dm-ioctl.c | 14 +++----------- drivers/parport/probe.c | 18 +++++------------- include/linux/netdevice.h | 4 ---- include/linux/string.h | 2 ++ include/sound/core.h | 3 ++- mm/slab.c | 24 ++++++++++++++++++++++++ net/core/neighbour.c | 3 ++- net/core/sysctl_net_core.c | 15 --------------- net/ipv4/devinet.c | 2 +- net/ipv6/addrconf.c | 3 ++- net/sunrpc/svcauth_unix.c | 11 ++--------- sound/core/info.c | 3 ++- sound/core/info_oss.c | 3 ++- sound/core/memory.c | 41 ++++++++++++++--------------------------- sound/core/oss/mixer_oss.c | 3 ++- sound/core/oss/pcm_oss.c | 3 ++- sound/core/sound.c | 2 +- sound/core/timer.c | 3 ++- sound/isa/gus/gus_mem.c | 7 ++++--- sound/pci/hda/patch_realtek.c | 2 +- sound/synth/emux/emux.c | 3 ++- 21 files changed, 75 insertions(+), 94 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index ee3c869d9701..200a0688f717 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -122,14 +122,6 @@ static struct hash_cell *__get_uuid_cell(const char *str) /*----------------------------------------------------------------- * Inserting, removing and renaming a device. *---------------------------------------------------------------*/ -static inline char *kstrdup(const char *str) -{ - char *r = kmalloc(strlen(str) + 1, GFP_KERNEL); - if (r) - strcpy(r, str); - return r; -} - static struct hash_cell *alloc_cell(const char *name, const char *uuid, struct mapped_device *md) { @@ -139,7 +131,7 @@ static struct hash_cell *alloc_cell(const char *name, const char *uuid, if (!hc) return NULL; - hc->name = kstrdup(name); + hc->name = kstrdup(name, GFP_KERNEL); if (!hc->name) { kfree(hc); return NULL; @@ -149,7 +141,7 @@ static struct hash_cell *alloc_cell(const char *name, const char *uuid, hc->uuid = NULL; else { - hc->uuid = kstrdup(uuid); + hc->uuid = kstrdup(uuid, GFP_KERNEL); if (!hc->uuid) { kfree(hc->name); kfree(hc); @@ -273,7 +265,7 @@ static int dm_hash_rename(const char *old, const char *new) /* * duplicate new. */ - new_name = kstrdup(new); + new_name = kstrdup(new, GFP_KERNEL); if (!new_name) return -ENOMEM; diff --git a/drivers/parport/probe.c b/drivers/parport/probe.c index c94963145e17..6e6f42d01e64 100644 --- a/drivers/parport/probe.c +++ b/drivers/parport/probe.c @@ -48,14 +48,6 @@ static void pretty_print(struct parport *port, int device) printk("\n"); } -static char *strdup(char *str) -{ - int n = strlen(str)+1; - char *s = kmalloc(n, GFP_KERNEL); - if (!s) return NULL; - return strcpy(s, str); -} - static void parse_data(struct parport *port, int device, char *str) { char *txt = kmalloc(strlen(str)+1, GFP_KERNEL); @@ -88,16 +80,16 @@ static void parse_data(struct parport *port, int device, char *str) if (!strcmp(p, "MFG") || !strcmp(p, "MANUFACTURER")) { if (info->mfr) kfree (info->mfr); - info->mfr = strdup(sep); + info->mfr = kstrdup(sep, GFP_KERNEL); } else if (!strcmp(p, "MDL") || !strcmp(p, "MODEL")) { if (info->model) kfree (info->model); - info->model = strdup(sep); + info->model = kstrdup(sep, GFP_KERNEL); } else if (!strcmp(p, "CLS") || !strcmp(p, "CLASS")) { int i; if (info->class_name) kfree (info->class_name); - info->class_name = strdup(sep); + info->class_name = kstrdup(sep, GFP_KERNEL); for (u = sep; *u; u++) *u = toupper(*u); for (i = 0; classes[i].token; i++) { @@ -112,7 +104,7 @@ static void parse_data(struct parport *port, int device, char *str) !strcmp(p, "COMMAND SET")) { if (info->cmdset) kfree (info->cmdset); - info->cmdset = strdup(sep); + info->cmdset = kstrdup(sep, GFP_KERNEL); /* if it speaks printer language, it's probably a printer */ if (strstr(sep, "PJL") || strstr(sep, "PCL")) @@ -120,7 +112,7 @@ static void parse_data(struct parport *port, int device, char *str) } else if (!strcmp(p, "DES") || !strcmp(p, "DESCRIPTION")) { if (info->description) kfree (info->description); - info->description = strdup(sep); + info->description = kstrdup(sep, GFP_KERNEL); } } rock_on: diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d6afd440cf7b..d89816ad642f 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -925,10 +925,6 @@ extern int skb_checksum_help(struct sk_buff *skb, int inward); extern void net_enable_timestamp(void); extern void net_disable_timestamp(void); -#ifdef CONFIG_SYSCTL -extern char *net_sysctl_strdup(const char *s); -#endif - #endif /* __KERNEL__ */ #endif /* _LINUX_DEV_H */ diff --git a/include/linux/string.h b/include/linux/string.h index b9fc59469956..93994c613095 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -88,6 +88,8 @@ extern int memcmp(const void *,const void *,__kernel_size_t); extern void * memchr(const void *,int,__kernel_size_t); #endif +extern char *kstrdup(const char *s, int gfp); + #ifdef __cplusplus } #endif diff --git a/include/sound/core.h b/include/sound/core.h index 9117c23e3a01..f8c4ef0aa352 100644 --- a/include/sound/core.h +++ b/include/sound/core.h @@ -292,6 +292,7 @@ void *snd_hidden_kcalloc(size_t n, size_t size, int flags); void snd_hidden_kfree(const void *obj); void *snd_hidden_vmalloc(unsigned long size); void snd_hidden_vfree(void *obj); +char *snd_hidden_kstrdup(const char *s, int flags); #define kmalloc(size, flags) snd_hidden_kmalloc(size, flags) #define kcalloc(n, size, flags) snd_hidden_kcalloc(n, size, flags) #define kfree(obj) snd_hidden_kfree(obj) @@ -301,6 +302,7 @@ void snd_hidden_vfree(void *obj); #define vmalloc_nocheck(size) snd_wrapper_vmalloc(size) #define kfree_nocheck(obj) snd_wrapper_kfree(obj) #define vfree_nocheck(obj) snd_wrapper_vfree(obj) +#define kstrdup(s, flags) snd_hidden_kstrdup(s, flags) #else #define snd_memory_init() /*NOP*/ #define snd_memory_done() /*NOP*/ @@ -311,7 +313,6 @@ void snd_hidden_vfree(void *obj); #define kfree_nocheck(obj) kfree(obj) #define vfree_nocheck(obj) vfree(obj) #endif -char *snd_kmalloc_strdup(const char *string, int flags); int copy_to_user_fromio(void __user *dst, const volatile void __iomem *src, size_t count); int copy_from_user_toio(volatile void __iomem *dst, const void __user *src, size_t count); diff --git a/mm/slab.c b/mm/slab.c index 93cbbbb39f42..122d031baab2 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -92,6 +92,7 @@ #include #include #include +#include #include #include @@ -3082,3 +3083,26 @@ unsigned int ksize(const void *objp) return size; } + + +/* + * kstrdup - allocate space for and copy an existing string + * + * @s: the string to duplicate + * @gfp: the GFP mask used in the kmalloc() call when allocating memory + */ +char *kstrdup(const char *s, int gfp) +{ + size_t len; + char *buf; + + if (!s) + return NULL; + + len = strlen(s) + 1; + buf = kmalloc(len, gfp); + if (buf) + memcpy(buf, s, len); + return buf; +} +EXPORT_SYMBOL(kstrdup); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index f6bdcad47da6..851eb927ed97 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -32,6 +32,7 @@ #include #include #include +#include #define NEIGH_DEBUG 1 @@ -2592,7 +2593,7 @@ int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p, t->neigh_vars[17].extra1 = dev; } - dev_name = net_sysctl_strdup(dev_name_source); + dev_name = kstrdup(dev_name_source, GFP_KERNEL); if (!dev_name) { err = -ENOBUFS; goto free; diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index c8be646cb191..880a88815211 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -35,19 +35,6 @@ extern int sysctl_somaxconn; extern char sysctl_divert_version[]; #endif /* CONFIG_NET_DIVERT */ -/* - * This strdup() is used for creating copies of network - * device names to be handed over to sysctl. - */ - -char *net_sysctl_strdup(const char *s) -{ - char *rv = kmalloc(strlen(s)+1, GFP_KERNEL); - if (rv) - strcpy(rv, s); - return rv; -} - ctl_table core_table[] = { #ifdef CONFIG_NET { @@ -177,6 +164,4 @@ ctl_table core_table[] = { { .ctl_name = 0 } }; -EXPORT_SYMBOL(net_sysctl_strdup); - #endif diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 650dcb12d9a1..d8a10e3dd77d 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1471,7 +1471,7 @@ static void devinet_sysctl_register(struct in_device *in_dev, * by sysctl and we wouldn't want anyone to change it under our feet * (see SIOCSIFNAME). */ - dev_name = net_sysctl_strdup(dev_name); + dev_name = kstrdup(dev_name, GFP_KERNEL); if (!dev_name) goto free; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 14f5c53235fe..a54d4ef3fd35 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -57,6 +57,7 @@ #endif #include #include +#include #include #include @@ -3437,7 +3438,7 @@ static void addrconf_sysctl_register(struct inet6_dev *idev, struct ipv6_devconf * by sysctl and we wouldn't want anyone to change it under our feet * (see SIOCSIFNAME). */ - dev_name = net_sysctl_strdup(dev_name); + dev_name = kstrdup(dev_name, GFP_KERNEL); if (!dev_name) goto free; diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c index 2b99b4028d31..d6baf6fdf8a9 100644 --- a/net/sunrpc/svcauth_unix.c +++ b/net/sunrpc/svcauth_unix.c @@ -8,6 +8,7 @@ #include #include #include +#include #define RPCDBG_FACILITY RPCDBG_AUTH @@ -20,14 +21,6 @@ */ -static char *strdup(char *s) -{ - char *rv = kmalloc(strlen(s)+1, GFP_KERNEL); - if (rv) - strcpy(rv, s); - return rv; -} - struct unix_domain { struct auth_domain h; int addr_changes; @@ -55,7 +48,7 @@ struct auth_domain *unix_domain_find(char *name) if (new == NULL) return NULL; cache_init(&new->h.h); - new->h.name = strdup(name); + new->h.name = kstrdup(name, GFP_KERNEL); new->h.flavour = RPC_AUTH_UNIX; new->addr_changes = 0; new->h.h.expiry_time = NEVER; diff --git a/sound/core/info.c b/sound/core/info.c index 31faffe01cb0..5e122bbe7c92 100644 --- a/sound/core/info.c +++ b/sound/core/info.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -754,7 +755,7 @@ static snd_info_entry_t *snd_info_create_entry(const char *name) entry = kcalloc(1, sizeof(*entry), GFP_KERNEL); if (entry == NULL) return NULL; - entry->name = snd_kmalloc_strdup(name, GFP_KERNEL); + entry->name = kstrdup(name, GFP_KERNEL); if (entry->name == NULL) { kfree(entry); return NULL; diff --git a/sound/core/info_oss.c b/sound/core/info_oss.c index f9e4ce443454..12107968d402 100644 --- a/sound/core/info_oss.c +++ b/sound/core/info_oss.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -51,7 +52,7 @@ int snd_oss_info_register(int dev, int num, char *string) x = NULL; } } else { - x = snd_kmalloc_strdup(string, GFP_KERNEL); + x = kstrdup(string, GFP_KERNEL); if (x == NULL) { up(&strings); return -ENOMEM; diff --git a/sound/core/memory.c b/sound/core/memory.c index 20860fec9364..c1fb28e84330 100644 --- a/sound/core/memory.c +++ b/sound/core/memory.c @@ -184,6 +184,20 @@ void snd_hidden_vfree(void *obj) snd_wrapper_vfree(obj); } +char *snd_hidden_kstrdup(const char *s, int flags) +{ + int len; + char *buf; + + if (!s) return NULL; + + len = strlen(s) + 1; + buf = _snd_kmalloc(len, flags); + if (buf) + memcpy(buf, s, len); + return buf; +} + static void snd_memory_info_read(snd_info_entry_t *entry, snd_info_buffer_t * buffer) { snd_iprintf(buffer, "kmalloc: %li bytes\n", snd_alloc_kmalloc); @@ -214,35 +228,8 @@ int __exit snd_memory_info_done(void) return 0; } -#else - -#define _snd_kmalloc kmalloc - #endif /* CONFIG_SND_DEBUG_MEMORY */ -/** - * snd_kmalloc_strdup - copy the string - * @string: the original string - * @flags: allocation conditions, GFP_XXX - * - * Allocates a memory chunk via kmalloc() and copies the string to it. - * - * Returns the pointer, or NULL if no enoguh memory. - */ -char *snd_kmalloc_strdup(const char *string, int flags) -{ - size_t len; - char *ptr; - - if (!string) - return NULL; - len = strlen(string) + 1; - ptr = _snd_kmalloc(len, flags); - if (ptr) - memcpy(ptr, string, len); - return ptr; -} - /** * copy_to_user_fromio - copy data from mmio-space to user-space * @dst: the destination pointer on user-space diff --git a/sound/core/oss/mixer_oss.c b/sound/core/oss/mixer_oss.c index 98ed9a9f0da6..98fc0766f885 100644 --- a/sound/core/oss/mixer_oss.c +++ b/sound/core/oss/mixer_oss.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -1137,7 +1138,7 @@ static void snd_mixer_oss_proc_write(snd_info_entry_t *entry, goto __unlock; } tbl->oss_id = ch; - tbl->name = snd_kmalloc_strdup(str, GFP_KERNEL); + tbl->name = kstrdup(str, GFP_KERNEL); if (! tbl->name) { kfree(tbl); goto __unlock; diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c index cab30977e7c0..de7444c586f9 100644 --- a/sound/core/oss/pcm_oss.c +++ b/sound/core/oss/pcm_oss.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include @@ -2360,7 +2361,7 @@ static void snd_pcm_oss_proc_write(snd_info_entry_t *entry, for (setup1 = pstr->oss.setup_list; setup1->next; setup1 = setup1->next); setup1->next = setup; } - template.task_name = snd_kmalloc_strdup(task_name, GFP_KERNEL); + template.task_name = kstrdup(task_name, GFP_KERNEL); } else { buffer->error = -ENOMEM; } diff --git a/sound/core/sound.c b/sound/core/sound.c index 0815fadeb3ec..7612884f530b 100644 --- a/sound/core/sound.c +++ b/sound/core/sound.c @@ -399,8 +399,8 @@ EXPORT_SYMBOL(snd_hidden_kcalloc); EXPORT_SYMBOL(snd_hidden_kfree); EXPORT_SYMBOL(snd_hidden_vmalloc); EXPORT_SYMBOL(snd_hidden_vfree); +EXPORT_SYMBOL(snd_hidden_kstrdup); #endif -EXPORT_SYMBOL(snd_kmalloc_strdup); EXPORT_SYMBOL(copy_to_user_fromio); EXPORT_SYMBOL(copy_from_user_toio); /* init.c */ diff --git a/sound/core/timer.c b/sound/core/timer.c index b498e5482d77..cfaccd415b3b 100644 --- a/sound/core/timer.c +++ b/sound/core/timer.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -100,7 +101,7 @@ static snd_timer_instance_t *snd_timer_instance_new(char *owner, snd_timer_t *ti timeri = kcalloc(1, sizeof(*timeri), GFP_KERNEL); if (timeri == NULL) return NULL; - timeri->owner = snd_kmalloc_strdup(owner, GFP_KERNEL); + timeri->owner = kstrdup(owner, GFP_KERNEL); if (! timeri->owner) { kfree(timeri); return NULL; diff --git a/sound/isa/gus/gus_mem.c b/sound/isa/gus/gus_mem.c index 609838e8ef67..5eb766dd564b 100644 --- a/sound/isa/gus/gus_mem.c +++ b/sound/isa/gus/gus_mem.c @@ -21,6 +21,7 @@ #include #include +#include #include #include #include @@ -213,7 +214,7 @@ snd_gf1_mem_block_t *snd_gf1_mem_alloc(snd_gf1_mem_t * alloc, int owner, if (share_id != NULL) memcpy(&block.share_id, share_id, sizeof(block.share_id)); block.owner = owner; - block.name = snd_kmalloc_strdup(name, GFP_KERNEL); + block.name = kstrdup(name, GFP_KERNEL); nblock = snd_gf1_mem_xalloc(alloc, &block); snd_gf1_mem_lock(alloc, 1); return nblock; @@ -253,13 +254,13 @@ int snd_gf1_mem_init(snd_gus_card_t * gus) if (gus->gf1.enh_mode) { block.ptr = 0; block.size = 1024; - block.name = snd_kmalloc_strdup("InterWave LFOs", GFP_KERNEL); + block.name = kstrdup("InterWave LFOs", GFP_KERNEL); if (snd_gf1_mem_xalloc(alloc, &block) == NULL) return -ENOMEM; } block.ptr = gus->gf1.default_voice_address; block.size = 4; - block.name = snd_kmalloc_strdup("Voice default (NULL's)", GFP_KERNEL); + block.name = kstrdup("Voice default (NULL's)", GFP_KERNEL); if (snd_gf1_mem_xalloc(alloc, &block) == NULL) return -ENOMEM; #ifdef CONFIG_SND_DEBUG diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 9edd558d6bd3..bab89843d850 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -1781,7 +1781,7 @@ static int add_control(struct alc_spec *spec, int type, const char *name, unsign knew = &spec->kctl_alloc[spec->num_kctl_used]; *knew = alc880_control_templates[type]; - knew->name = snd_kmalloc_strdup(name, GFP_KERNEL); + knew->name = kstrdup(name, GFP_KERNEL); if (! knew->name) return -ENOMEM; knew->private_value = val; diff --git a/sound/synth/emux/emux.c b/sound/synth/emux/emux.c index 16f3b461627a..60d0b2c66698 100644 --- a/sound/synth/emux/emux.c +++ b/sound/synth/emux/emux.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -76,7 +77,7 @@ int snd_emux_register(snd_emux_t *emu, snd_card_t *card, int index, char *name) snd_assert(name != NULL, return -EINVAL); emu->card = card; - emu->name = snd_kmalloc_strdup(name, GFP_KERNEL); + emu->name = kstrdup(name, GFP_KERNEL); emu->voices = kcalloc(emu->max_voices, sizeof(snd_emux_voice_t), GFP_KERNEL); if (emu->voices == NULL) return -ENOMEM; -- cgit v1.2.3 From 35a82d1a53e1a9ad54efafcc940f9335beaed5c3 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Thu, 23 Jun 2005 00:09:06 -0700 Subject: [PATCH] optimise loop driver a bit Looks like locking can be optimised quite a lot. Increase lock widths slightly so lo_lock is taken fewer times per request. Also it was quite trivial to cover lo_pending with that lock, and remove the atomic requirement. This also makes memory ordering explicitly correct, which is nice (not that I particularly saw any mem ordering bugs). Test was reading 4 250MB files in parallel on ext2-on-tmpfs filesystem (1K block size, 4K page size). System is 2 socket Xeon with HT (4 thread). intel:/home/npiggin# umount /dev/loop0 ; mount /dev/loop0 /mnt/loop ; /usr/bin/time ./mtloop.sh Before: 0.24user 5.51system 0:02.84elapsed 202%CPU (0avgtext+0avgdata 0maxresident)k 0.19user 5.52system 0:02.88elapsed 198%CPU (0avgtext+0avgdata 0maxresident)k 0.19user 5.57system 0:02.89elapsed 198%CPU (0avgtext+0avgdata 0maxresident)k 0.22user 5.51system 0:02.90elapsed 197%CPU (0avgtext+0avgdata 0maxresident)k 0.19user 5.44system 0:02.91elapsed 193%CPU (0avgtext+0avgdata 0maxresident)k After: 0.07user 2.34system 0:01.68elapsed 143%CPU (0avgtext+0avgdata 0maxresident)k 0.06user 2.37system 0:01.68elapsed 144%CPU (0avgtext+0avgdata 0maxresident)k 0.06user 2.39system 0:01.68elapsed 145%CPU (0avgtext+0avgdata 0maxresident)k 0.06user 2.36system 0:01.68elapsed 144%CPU (0avgtext+0avgdata 0maxresident)k 0.06user 2.42system 0:01.68elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/loop.c | 81 ++++++++++++++++++++++++---------------------------- include/linux/loop.h | 2 +- 2 files changed, 39 insertions(+), 44 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 6f011d0d8e97..b35e08876dd4 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -472,17 +472,11 @@ static int do_bio_filebacked(struct loop_device *lo, struct bio *bio) */ static void loop_add_bio(struct loop_device *lo, struct bio *bio) { - unsigned long flags; - - spin_lock_irqsave(&lo->lo_lock, flags); if (lo->lo_biotail) { lo->lo_biotail->bi_next = bio; lo->lo_biotail = bio; } else lo->lo_bio = lo->lo_biotail = bio; - spin_unlock_irqrestore(&lo->lo_lock, flags); - - up(&lo->lo_bh_mutex); } /* @@ -492,14 +486,12 @@ static struct bio *loop_get_bio(struct loop_device *lo) { struct bio *bio; - spin_lock_irq(&lo->lo_lock); if ((bio = lo->lo_bio)) { if (bio == lo->lo_biotail) lo->lo_biotail = NULL; lo->lo_bio = bio->bi_next; bio->bi_next = NULL; } - spin_unlock_irq(&lo->lo_lock); return bio; } @@ -509,35 +501,28 @@ static int loop_make_request(request_queue_t *q, struct bio *old_bio) struct loop_device *lo = q->queuedata; int rw = bio_rw(old_bio); - if (!lo) - goto out; + if (rw == READA) + rw = READ; + + BUG_ON(!lo || (rw != READ && rw != WRITE)); spin_lock_irq(&lo->lo_lock); if (lo->lo_state != Lo_bound) - goto inactive; - atomic_inc(&lo->lo_pending); - spin_unlock_irq(&lo->lo_lock); - - if (rw == WRITE) { - if (lo->lo_flags & LO_FLAGS_READ_ONLY) - goto err; - } else if (rw == READA) { - rw = READ; - } else if (rw != READ) { - printk(KERN_ERR "loop: unknown command (%x)\n", rw); - goto err; - } + goto out; + if (unlikely(rw == WRITE && (lo->lo_flags & LO_FLAGS_READ_ONLY))) + goto out; + lo->lo_pending++; loop_add_bio(lo, old_bio); + spin_unlock_irq(&lo->lo_lock); + up(&lo->lo_bh_mutex); return 0; -err: - if (atomic_dec_and_test(&lo->lo_pending)) - up(&lo->lo_bh_mutex); + out: + if (lo->lo_pending == 0) + up(&lo->lo_bh_mutex); + spin_unlock_irq(&lo->lo_lock); bio_io_error(old_bio, old_bio->bi_size); return 0; -inactive: - spin_unlock_irq(&lo->lo_lock); - goto out; } /* @@ -560,13 +545,11 @@ static void do_loop_switch(struct loop_device *, struct switch_request *); static inline void loop_handle_bio(struct loop_device *lo, struct bio *bio) { - int ret; - if (unlikely(!bio->bi_bdev)) { do_loop_switch(lo, bio->bi_private); bio_put(bio); } else { - ret = do_bio_filebacked(lo, bio); + int ret = do_bio_filebacked(lo, bio); bio_endio(bio, bio->bi_size, ret); } } @@ -594,7 +577,7 @@ static int loop_thread(void *data) set_user_nice(current, -20); lo->lo_state = Lo_bound; - atomic_inc(&lo->lo_pending); + lo->lo_pending = 1; /* * up sem, we are running @@ -602,26 +585,37 @@ static int loop_thread(void *data) up(&lo->lo_sem); for (;;) { - down_interruptible(&lo->lo_bh_mutex); + int pending; + /* - * could be upped because of tear-down, not because of - * pending work + * interruptible just to not contribute to load avg */ - if (!atomic_read(&lo->lo_pending)) + if (down_interruptible(&lo->lo_bh_mutex)) + continue; + + spin_lock_irq(&lo->lo_lock); + + /* + * could be upped because of tear-down, not pending work + */ + if (unlikely(!lo->lo_pending)) { + spin_unlock_irq(&lo->lo_lock); break; + } bio = loop_get_bio(lo); - if (!bio) { - printk("loop: missing bio\n"); - continue; - } + lo->lo_pending--; + pending = lo->lo_pending; + spin_unlock_irq(&lo->lo_lock); + + BUG_ON(!bio); loop_handle_bio(lo, bio); /* * upped both for pending work and tear-down, lo_pending * will hit zero then */ - if (atomic_dec_and_test(&lo->lo_pending)) + if (unlikely(!pending)) break; } @@ -900,7 +894,8 @@ static int loop_clr_fd(struct loop_device *lo, struct block_device *bdev) spin_lock_irq(&lo->lo_lock); lo->lo_state = Lo_rundown; - if (atomic_dec_and_test(&lo->lo_pending)) + lo->lo_pending--; + if (!lo->lo_pending) up(&lo->lo_bh_mutex); spin_unlock_irq(&lo->lo_lock); diff --git a/include/linux/loop.h b/include/linux/loop.h index 8220d9c9da00..53fa51595443 100644 --- a/include/linux/loop.h +++ b/include/linux/loop.h @@ -61,7 +61,7 @@ struct loop_device { struct semaphore lo_sem; struct semaphore lo_ctl_mutex; struct semaphore lo_bh_mutex; - atomic_t lo_pending; + int lo_pending; request_queue_t *lo_queue; }; -- cgit v1.2.3 From ac20427ef6aa63da663bdc88b71d16f7394f5e23 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Thu, 23 Jun 2005 00:09:11 -0700 Subject: [PATCH] add check to /proc/devices read routines Patch to add check to get_chrdev_list and get_blkdev_list to prevent reads of /proc/devices from spilling over the provided page if more than 4096 bytes of string data are generated from all the registered character and block devices in a system Signed-off-by: Neil Horman Cc: Christoph Hellwig Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/genhd.c | 12 ++++++++++-- fs/char_dev.c | 13 ++++++++++++- fs/proc/proc_misc.c | 2 +- include/linux/genhd.h | 2 +- 4 files changed, 24 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/genhd.c b/drivers/block/genhd.c index 43805e4d31e9..47fd3659a061 100644 --- a/drivers/block/genhd.c +++ b/drivers/block/genhd.c @@ -40,7 +40,7 @@ static inline int major_to_index(int major) #ifdef CONFIG_PROC_FS /* get block device names in somewhat random order */ -int get_blkdev_list(char *p) +int get_blkdev_list(char *p, int used) { struct blk_major_name *n; int i, len; @@ -49,10 +49,18 @@ int get_blkdev_list(char *p) down(&block_subsys_sem); for (i = 0; i < ARRAY_SIZE(major_names); i++) { - for (n = major_names[i]; n; n = n->next) + for (n = major_names[i]; n; n = n->next) { + /* + * If the curent string plus the 5 extra characters + * in the line would run us off the page, then we're done + */ + if ((len + used + strlen(n->name) + 5) >= PAGE_SIZE) + goto page_full; len += sprintf(p+len, "%3d %s\n", n->major, n->name); + } } +page_full: up(&block_subsys_sem); return len; diff --git a/fs/char_dev.c b/fs/char_dev.c index c1e3537909fc..e82aac9cc2f5 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -56,10 +56,21 @@ int get_chrdev_list(char *page) down(&chrdevs_lock); for (i = 0; i < ARRAY_SIZE(chrdevs) ; i++) { - for (cd = chrdevs[i]; cd; cd = cd->next) + for (cd = chrdevs[i]; cd; cd = cd->next) { + /* + * if the current name, plus the 5 extra characters + * in the device line for this entry + * would run us off the page, we're done + */ + if ((len+strlen(cd->name) + 5) >= PAGE_SIZE) + goto page_full; + + len += sprintf(page+len, "%3d %s\n", cd->major, cd->name); + } } +page_full: up(&chrdevs_lock); return len; diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index 63a9fbf1ac51..94b570ad037d 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -451,7 +451,7 @@ static int devices_read_proc(char *page, char **start, off_t off, int count, int *eof, void *data) { int len = get_chrdev_list(page); - len += get_blkdev_list(page+len); + len += get_blkdev_list(page+len, len); return proc_calc_metrics(page, start, off, count, eof, len); } diff --git a/include/linux/genhd.h b/include/linux/genhd.h index af26dc718ef6..01796c41c951 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -224,7 +224,7 @@ static inline void free_disk_stats(struct gendisk *disk) extern void disk_round_stats(struct gendisk *disk); /* drivers/block/genhd.c */ -extern int get_blkdev_list(char *); +extern int get_blkdev_list(char *, int); extern void add_disk(struct gendisk *disk); extern void del_gendisk(struct gendisk *gp); extern void unlink_gendisk(struct gendisk *gp); -- cgit v1.2.3 From 84de856ed30c568c2bb7b9ac0679772bd2737d9b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Jun 2005 00:09:16 -0700 Subject: [PATCH] quota: consolidate code surrounding vfs_quota_on_mount Move some code duplicated in both callers into vfs_quota_on_mount Signed-off-by: Christoph Hellwig Acked-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dquot.c | 23 +++++++++++++++++++---- fs/ext3/super.c | 18 ++---------------- fs/reiserfs/super.c | 21 +++------------------ include/linux/quotaops.h | 3 ++- 4 files changed, 26 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/fs/dquot.c b/fs/dquot.c index 3995ce7907cc..343c03655619 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -1519,14 +1519,29 @@ out_path: * This function is used when filesystem needs to initialize quotas * during mount time. */ -int vfs_quota_on_mount(int type, int format_id, struct dentry *dentry) +int vfs_quota_on_mount(struct super_block *sb, char *qf_name, + int format_id, int type) { + struct qstr name = {.name = qf_name, .len = 0, .len = strlen(qf_name)}; + struct dentry *dentry; int error; + dentry = lookup_hash(&name, sb->s_root); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + error = security_quota_on(dentry); - if (error) - return error; - return vfs_quota_on_inode(dentry->d_inode, type, format_id); + if (!error) + error = vfs_quota_on_inode(dentry->d_inode, type, format_id); + + /* + * Now invalidate and put the dentry - quota got its own reference + * to inode and dentry has at least wrong hash so we had better + * throw it away. + */ + d_invalidate(dentry); + dput(dentry); + return error; } /* Generic routine for getting common part of quota structure */ diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 981ccb233ef5..9630fbfdc24a 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -2348,22 +2348,8 @@ static int ext3_write_info(struct super_block *sb, int type) */ static int ext3_quota_on_mount(struct super_block *sb, int type) { - int err; - struct dentry *dentry; - struct qstr name = { .name = EXT3_SB(sb)->s_qf_names[type], - .hash = 0, - .len = strlen(EXT3_SB(sb)->s_qf_names[type])}; - - dentry = lookup_hash(&name, sb->s_root); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - err = vfs_quota_on_mount(type, EXT3_SB(sb)->s_jquota_fmt, dentry); - /* Now invalidate and put the dentry - quota got its own reference - * to inode and dentry has at least wrong hash so we had better - * throw it away */ - d_invalidate(dentry); - dput(dentry); - return err; + return vfs_quota_on_mount(sb, EXT3_SB(sb)->s_qf_names[type], + EXT3_SB(sb)->s_jquota_fmt, type); } /* diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index b35b87744983..aae0779ed5b4 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -1932,27 +1932,12 @@ static int reiserfs_write_info(struct super_block *sb, int type) } /* - * Turn on quotas during mount time - we need to find - * the quota file and such... + * Turn on quotas during mount time - we need to find the quota file and such... */ static int reiserfs_quota_on_mount(struct super_block *sb, int type) { - int err; - struct dentry *dentry; - struct qstr name = { .name = REISERFS_SB(sb)->s_qf_names[type], - .hash = 0, - .len = strlen(REISERFS_SB(sb)->s_qf_names[type])}; - - dentry = lookup_hash(&name, sb->s_root); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - err = vfs_quota_on_mount(type, REISERFS_SB(sb)->s_jquota_fmt, dentry); - /* Now invalidate and put the dentry - quota got its own reference - * to inode and dentry has at least wrong hash so we had better - * throw it away */ - d_invalidate(dentry); - dput(dentry); - return err; + return vfs_quota_on_mount(sb, REISERFS_SB(sb)->s_qf_names[type], + REISERFS_SB(sb)->s_jquota_fmt, type); } /* diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h index e57baa85e744..d211507ab246 100644 --- a/include/linux/quotaops.h +++ b/include/linux/quotaops.h @@ -39,7 +39,8 @@ extern int dquot_commit_info(struct super_block *sb, int type); extern int dquot_mark_dquot_dirty(struct dquot *dquot); extern int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path); -extern int vfs_quota_on_mount(int type, int format_id, struct dentry *dentry); +extern int vfs_quota_on_mount(struct super_block *sb, char *qf_name, + int format_id, int type); extern int vfs_quota_off(struct super_block *sb, int type); #define vfs_quota_off_mount(sb, type) vfs_quota_off(sb, type) extern int vfs_quota_sync(struct super_block *sb, int type); -- cgit v1.2.3 From b94cce926b2b902b79380ccba370d6f9f2980de0 Mon Sep 17 00:00:00 2001 From: Hien Nguyen Date: Thu, 23 Jun 2005 00:09:19 -0700 Subject: [PATCH] kprobes: function-return probes This patch adds function-return probes to kprobes for the i386 architecture. This enables you to establish a handler to be run when a function returns. 1. API Two new functions are added to kprobes: int register_kretprobe(struct kretprobe *rp); void unregister_kretprobe(struct kretprobe *rp); 2. Registration and unregistration 2.1 Register To register a function-return probe, the user populates the following fields in a kretprobe object and calls register_kretprobe() with the kretprobe address as an argument: kp.addr - the function's address handler - this function is run after the ret instruction executes, but before control returns to the return address in the caller. maxactive - The maximum number of instances of the probed function that can be active concurrently. For example, if the function is non- recursive and is called with a spinlock or mutex held, maxactive = 1 should be enough. If the function is non-recursive and can never relinquish the CPU (e.g., via a semaphore or preemption), NR_CPUS should be enough. maxactive is used to determine how many kretprobe_instance objects to allocate for this particular probed function. If maxactive <= 0, it is set to a default value (if CONFIG_PREEMPT maxactive=max(10, 2 * NR_CPUS) else maxactive=NR_CPUS) For example: struct kretprobe rp; rp.kp.addr = /* entrypoint address */ rp.handler = /*return probe handler */ rp.maxactive = /* e.g., 1 or NR_CPUS or 0, see the above explanation */ register_kretprobe(&rp); The following field may also be of interest: nmissed - Initialized to zero when the function-return probe is registered, and incremented every time the probed function is entered but there is no kretprobe_instance object available for establishing the function-return probe (i.e., because maxactive was set too low). 2.2 Unregister To unregiter a function-return probe, the user calls unregister_kretprobe() with the same kretprobe object as registered previously. If a probed function is running when the return probe is unregistered, the function will return as expected, but the handler won't be run. 3. Limitations 3.1 This patch supports only the i386 architecture, but patches for x86_64 and ppc64 are anticipated soon. 3.2 Return probes operates by replacing the return address in the stack (or in a known register, such as the lr register for ppc). This may cause __builtin_return_address(0), when invoked from the return-probed function, to return the address of the return-probes trampoline. 3.3 This implementation uses the "Multiprobes at an address" feature in 2.6.12-rc3-mm3. 3.4 Due to a limitation in multi-probes, you cannot currently establish a return probe and a jprobe on the same function. A patch to remove this limitation is being tested. This feature is required by SystemTap (http://sourceware.org/systemtap), and reflects ideas contributed by several SystemTap developers, including Will Cohen and Ananth Mavinakayanahalli. Signed-off-by: Hien Nguyen Signed-off-by: Prasanna S Panchamukhi Signed-off-by: Frederik Deweerdt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/kprobes.c | 102 +++++++++++++++++++++- arch/i386/kernel/process.c | 15 ++++ include/asm-i386/kprobes.h | 3 + include/linux/kprobes.h | 90 ++++++++++++++++++- kernel/kprobes.c | 213 +++++++++++++++++++++++++++++++++++++++++++-- 5 files changed, 415 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index 59ff9b455069..048f754bbe23 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -23,6 +23,9 @@ * Rusty Russell). * 2004-July Suparna Bhattacharya added jumper probes * interface to access function arguments. + * 2005-May Hien Nguyen , Jim Keniston + * and Prasanna S Panchamukhi + * added function-return probes. */ #include @@ -91,6 +94,53 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs) regs->eip = (unsigned long)&p->ainsn.insn; } +struct task_struct *arch_get_kprobe_task(void *ptr) +{ + return ((struct thread_info *) (((unsigned long) ptr) & + (~(THREAD_SIZE -1))))->task; +} + +void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs) +{ + unsigned long *sara = (unsigned long *)®s->esp; + struct kretprobe_instance *ri; + static void *orig_ret_addr; + + /* + * Save the return address when the return probe hits + * the first time, and use it to populate the (krprobe + * instance)->ret_addr for subsequent return probes at + * the same addrress since stack address would have + * the kretprobe_trampoline by then. + */ + if (((void*) *sara) != kretprobe_trampoline) + orig_ret_addr = (void*) *sara; + + if ((ri = get_free_rp_inst(rp)) != NULL) { + ri->rp = rp; + ri->stack_addr = sara; + ri->ret_addr = orig_ret_addr; + add_rp_inst(ri); + /* Replace the return addr with trampoline addr */ + *sara = (unsigned long) &kretprobe_trampoline; + } else { + rp->nmissed++; + } +} + +void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock) +{ + unsigned long flags = 0; + struct kretprobe_instance *ri; + spin_lock_irqsave(kp_lock, flags); + while ((ri = get_rp_inst_tsk(tk)) != NULL) { + *((unsigned long *)(ri->stack_addr)) = + (unsigned long) ri->ret_addr; + recycle_rp_inst(ri); + } + spin_unlock_irqrestore(kp_lock, flags); +} + /* * Interrupts are disabled on entry as trap3 is an interrupt gate and they * remain disabled thorough out this function. @@ -183,6 +233,55 @@ no_kprobe: return ret; } +/* + * For function-return probes, init_kprobes() establishes a probepoint + * here. When a retprobed function returns, this probe is hit and + * trampoline_probe_handler() runs, calling the kretprobe's handler. + */ + void kretprobe_trampoline_holder(void) + { + asm volatile ( ".global kretprobe_trampoline\n" + "kretprobe_trampoline: \n" + "nop\n"); + } + +/* + * Called when we hit the probe point at kretprobe_trampoline + */ +int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) +{ + struct task_struct *tsk; + struct kretprobe_instance *ri; + struct hlist_head *head; + struct hlist_node *node; + unsigned long *sara = ((unsigned long *) ®s->esp) - 1; + + tsk = arch_get_kprobe_task(sara); + head = kretprobe_inst_table_head(tsk); + + hlist_for_each_entry(ri, node, head, hlist) { + if (ri->stack_addr == sara && ri->rp) { + if (ri->rp->handler) + ri->rp->handler(ri, regs); + } + } + return 0; +} + +void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs, + unsigned long flags) +{ + struct kretprobe_instance *ri; + /* RA already popped */ + unsigned long *sara = ((unsigned long *)®s->esp) - 1; + + while ((ri = get_rp_inst(sara))) { + regs->eip = (unsigned long)ri->ret_addr; + recycle_rp_inst(ri); + } + regs->eflags &= ~TF_MASK; +} + /* * Called after single-stepping. p->addr is the address of the * instruction whose first byte has been replaced by the "int 3" @@ -266,7 +365,8 @@ static inline int post_kprobe_handler(struct pt_regs *regs) if (current_kprobe->post_handler) current_kprobe->post_handler(current_kprobe, regs, 0); - resume_execution(current_kprobe, regs); + if (current_kprobe->post_handler != trampoline_post_handler) + resume_execution(current_kprobe, regs); regs->eflags |= kprobe_saved_eflags; unlock_kprobes(); diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index be3efba7caf7..aea2ce1145df 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include @@ -339,6 +340,13 @@ void exit_thread(void) struct task_struct *tsk = current; struct thread_struct *t = &tsk->thread; + /* + * Remove function-return probe instances associated with this task + * and put them back on the free list. Do not insert an exit probe for + * this function, it will be disabled by kprobe_flush_task if you do. + */ + kprobe_flush_task(tsk); + /* The process may have allocated an io port bitmap... nuke it. */ if (unlikely(NULL != t->io_bitmap_ptr)) { int cpu = get_cpu(); @@ -362,6 +370,13 @@ void flush_thread(void) { struct task_struct *tsk = current; + /* + * Remove function-return probe instances associated with this task + * and put them back on the free list. Do not insert an exit probe for + * this function, it will be disabled by kprobe_flush_task if you do. + */ + kprobe_flush_task(tsk); + memset(tsk->thread.debugreg, 0, sizeof(unsigned long)*8); memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array)); /* diff --git a/include/asm-i386/kprobes.h b/include/asm-i386/kprobes.h index 4092f68d123a..8b6d3a90cd78 100644 --- a/include/asm-i386/kprobes.h +++ b/include/asm-i386/kprobes.h @@ -39,6 +39,9 @@ typedef u8 kprobe_opcode_t; : (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) #define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)pentry +#define ARCH_SUPPORTS_KRETPROBES + +void kretprobe_trampoline(void); /* Architecture specific copy of original instruction*/ struct arch_specific_insn { diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index 99ddba5a4e00..fba39f87efec 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -25,21 +25,31 @@ * Rusty Russell). * 2004-July Suparna Bhattacharya added jumper probes * interface to access function arguments. + * 2005-May Hien Nguyen and Jim Keniston + * and Prasanna S Panchamukhi + * added function-return probes. */ #include #include #include #include +#include + #include struct kprobe; struct pt_regs; +struct kretprobe; +struct kretprobe_instance; typedef int (*kprobe_pre_handler_t) (struct kprobe *, struct pt_regs *); typedef int (*kprobe_break_handler_t) (struct kprobe *, struct pt_regs *); typedef void (*kprobe_post_handler_t) (struct kprobe *, struct pt_regs *, unsigned long flags); typedef int (*kprobe_fault_handler_t) (struct kprobe *, struct pt_regs *, int trapnr); +typedef int (*kretprobe_handler_t) (struct kretprobe_instance *, + struct pt_regs *); + struct kprobe { struct hlist_node hlist; @@ -85,6 +95,62 @@ struct jprobe { kprobe_opcode_t *entry; /* probe handling code to jump to */ }; +#ifdef ARCH_SUPPORTS_KRETPROBES +extern int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs); +extern void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs, + unsigned long flags); +extern struct task_struct *arch_get_kprobe_task(void *ptr); +extern void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs); +extern void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock); +#else /* ARCH_SUPPORTS_KRETPROBES */ +static inline void kretprobe_trampoline(void) +{ +} +static inline int trampoline_probe_handler(struct kprobe *p, + struct pt_regs *regs) +{ + return 0; +} +static inline void trampoline_post_handler(struct kprobe *p, + struct pt_regs *regs, unsigned long flags) +{ +} +static inline void arch_prepare_kretprobe(struct kretprobe *rp, + struct pt_regs *regs) +{ +} +static inline void arch_kprobe_flush_task(struct task_struct *tk) +{ +} +#define arch_get_kprobe_task(ptr) ((struct task_struct *)NULL) +#endif /* ARCH_SUPPORTS_KRETPROBES */ +/* + * Function-return probe - + * Note: + * User needs to provide a handler function, and initialize maxactive. + * maxactive - The maximum number of instances of the probed function that + * can be active concurrently. + * nmissed - tracks the number of times the probed function's return was + * ignored, due to maxactive being too low. + * + */ +struct kretprobe { + struct kprobe kp; + kretprobe_handler_t handler; + int maxactive; + int nmissed; + struct hlist_head free_instances; + struct hlist_head used_instances; +}; + +struct kretprobe_instance { + struct hlist_node uflist; /* either on free list or used list */ + struct hlist_node hlist; + struct kretprobe *rp; + void *ret_addr; + void *stack_addr; +}; + #ifdef CONFIG_KPROBES /* Locks kprobe: irq must be disabled */ void lock_kprobes(void); @@ -104,6 +170,7 @@ extern void show_registers(struct pt_regs *regs); /* Get the kprobe at this addr (if any). Must have called lock_kprobes */ struct kprobe *get_kprobe(void *addr); +struct hlist_head * kretprobe_inst_table_head(struct task_struct *tsk); int register_kprobe(struct kprobe *p); void unregister_kprobe(struct kprobe *p); @@ -113,7 +180,16 @@ int register_jprobe(struct jprobe *p); void unregister_jprobe(struct jprobe *p); void jprobe_return(void); -#else +int register_kretprobe(struct kretprobe *rp); +void unregister_kretprobe(struct kretprobe *rp); + +struct kretprobe_instance *get_free_rp_inst(struct kretprobe *rp); +struct kretprobe_instance *get_rp_inst(void *sara); +struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk); +void add_rp_inst(struct kretprobe_instance *ri); +void kprobe_flush_task(struct task_struct *tk); +void recycle_rp_inst(struct kretprobe_instance *ri); +#else /* CONFIG_KPROBES */ static inline int kprobe_running(void) { return 0; @@ -135,5 +211,15 @@ static inline void unregister_jprobe(struct jprobe *p) static inline void jprobe_return(void) { } -#endif +static inline int register_kretprobe(struct kretprobe *rp) +{ + return -ENOSYS; +} +static inline void unregister_kretprobe(struct kretprobe *rp) +{ +} +static inline void kprobe_flush_task(struct task_struct *tk) +{ +} +#endif /* CONFIG_KPROBES */ #endif /* _LINUX_KPROBES_H */ diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 037142b72a49..692fbf75ab49 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -27,6 +27,9 @@ * interface to access function arguments. * 2004-Sep Prasanna S Panchamukhi Changed Kprobes * exceptions notifier to be first on the priority list. + * 2005-May Hien Nguyen , Jim Keniston + * and Prasanna S Panchamukhi + * added function-return probes. */ #include #include @@ -41,6 +44,7 @@ #define KPROBE_TABLE_SIZE (1 << KPROBE_HASH_BITS) static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE]; +static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE]; unsigned int kprobe_cpu = NR_CPUS; static DEFINE_SPINLOCK(kprobe_lock); @@ -78,7 +82,7 @@ struct kprobe *get_kprobe(void *addr) * Aggregate handlers for multiple kprobes support - these handlers * take care of invoking the individual kprobe handlers on p->list */ -int aggr_pre_handler(struct kprobe *p, struct pt_regs *regs) +static int aggr_pre_handler(struct kprobe *p, struct pt_regs *regs) { struct kprobe *kp; @@ -92,8 +96,8 @@ int aggr_pre_handler(struct kprobe *p, struct pt_regs *regs) return 0; } -void aggr_post_handler(struct kprobe *p, struct pt_regs *regs, - unsigned long flags) +static void aggr_post_handler(struct kprobe *p, struct pt_regs *regs, + unsigned long flags) { struct kprobe *kp; @@ -107,7 +111,8 @@ void aggr_post_handler(struct kprobe *p, struct pt_regs *regs, return; } -int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr) +static int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, + int trapnr) { /* * if we faulted "during" the execution of a user specified @@ -120,6 +125,135 @@ int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr) return 0; } +struct kprobe trampoline_p = { + .addr = (kprobe_opcode_t *) &kretprobe_trampoline, + .pre_handler = trampoline_probe_handler, + .post_handler = trampoline_post_handler +}; + +struct kretprobe_instance *get_free_rp_inst(struct kretprobe *rp) +{ + struct hlist_node *node; + struct kretprobe_instance *ri; + hlist_for_each_entry(ri, node, &rp->free_instances, uflist) + return ri; + return NULL; +} + +static struct kretprobe_instance *get_used_rp_inst(struct kretprobe *rp) +{ + struct hlist_node *node; + struct kretprobe_instance *ri; + hlist_for_each_entry(ri, node, &rp->used_instances, uflist) + return ri; + return NULL; +} + +struct kretprobe_instance *get_rp_inst(void *sara) +{ + struct hlist_head *head; + struct hlist_node *node; + struct task_struct *tsk; + struct kretprobe_instance *ri; + + tsk = arch_get_kprobe_task(sara); + head = &kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)]; + hlist_for_each_entry(ri, node, head, hlist) { + if (ri->stack_addr == sara) + return ri; + } + return NULL; +} + +void add_rp_inst(struct kretprobe_instance *ri) +{ + struct task_struct *tsk; + /* + * Remove rp inst off the free list - + * Add it back when probed function returns + */ + hlist_del(&ri->uflist); + tsk = arch_get_kprobe_task(ri->stack_addr); + /* Add rp inst onto table */ + INIT_HLIST_NODE(&ri->hlist); + hlist_add_head(&ri->hlist, + &kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)]); + + /* Also add this rp inst to the used list. */ + INIT_HLIST_NODE(&ri->uflist); + hlist_add_head(&ri->uflist, &ri->rp->used_instances); +} + +void recycle_rp_inst(struct kretprobe_instance *ri) +{ + /* remove rp inst off the rprobe_inst_table */ + hlist_del(&ri->hlist); + if (ri->rp) { + /* remove rp inst off the used list */ + hlist_del(&ri->uflist); + /* put rp inst back onto the free list */ + INIT_HLIST_NODE(&ri->uflist); + hlist_add_head(&ri->uflist, &ri->rp->free_instances); + } else + /* Unregistering */ + kfree(ri); +} + +struct hlist_head * kretprobe_inst_table_head(struct task_struct *tsk) +{ + return &kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)]; +} + +struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk) +{ + struct task_struct *tsk; + struct hlist_head *head; + struct hlist_node *node; + struct kretprobe_instance *ri; + + head = &kretprobe_inst_table[hash_ptr(tk, KPROBE_HASH_BITS)]; + + hlist_for_each_entry(ri, node, head, hlist) { + tsk = arch_get_kprobe_task(ri->stack_addr); + if (tsk == tk) + return ri; + } + return NULL; +} + +/* + * This function is called from do_exit or do_execv when task tk's stack is + * about to be recycled. Recycle any function-return probe instances + * associated with this task. These represent probed functions that have + * been called but may never return. + */ +void kprobe_flush_task(struct task_struct *tk) +{ + arch_kprobe_flush_task(tk, &kprobe_lock); +} + +/* + * This kprobe pre_handler is registered with every kretprobe. When probe + * hits it will set up the return probe. + */ +static int pre_handler_kretprobe(struct kprobe *p, struct pt_regs *regs) +{ + struct kretprobe *rp = container_of(p, struct kretprobe, kp); + + /*TODO: consider to only swap the RA after the last pre_handler fired */ + arch_prepare_kretprobe(rp, regs); + return 0; +} + +static inline void free_rp_inst(struct kretprobe *rp) +{ + struct kretprobe_instance *ri; + while ((ri = get_free_rp_inst(rp)) != NULL) { + hlist_del(&ri->uflist); + kfree(ri); + } +} + /* * Fill in the required fields of the "manager kprobe". Replace the * earlier kprobe in the hlist with the manager kprobe @@ -257,16 +391,82 @@ void unregister_jprobe(struct jprobe *jp) unregister_kprobe(&jp->kp); } +#ifdef ARCH_SUPPORTS_KRETPROBES + +int register_kretprobe(struct kretprobe *rp) +{ + int ret = 0; + struct kretprobe_instance *inst; + int i; + + rp->kp.pre_handler = pre_handler_kretprobe; + + /* Pre-allocate memory for max kretprobe instances */ + if (rp->maxactive <= 0) { +#ifdef CONFIG_PREEMPT + rp->maxactive = max(10, 2 * NR_CPUS); +#else + rp->maxactive = NR_CPUS; +#endif + } + INIT_HLIST_HEAD(&rp->used_instances); + INIT_HLIST_HEAD(&rp->free_instances); + for (i = 0; i < rp->maxactive; i++) { + inst = kmalloc(sizeof(struct kretprobe_instance), GFP_KERNEL); + if (inst == NULL) { + free_rp_inst(rp); + return -ENOMEM; + } + INIT_HLIST_NODE(&inst->uflist); + hlist_add_head(&inst->uflist, &rp->free_instances); + } + + rp->nmissed = 0; + /* Establish function entry probe point */ + if ((ret = register_kprobe(&rp->kp)) != 0) + free_rp_inst(rp); + return ret; +} + +#else /* ARCH_SUPPORTS_KRETPROBES */ + +int register_kretprobe(struct kretprobe *rp) +{ + return -ENOSYS; +} + +#endif /* ARCH_SUPPORTS_KRETPROBES */ + +void unregister_kretprobe(struct kretprobe *rp) +{ + unsigned long flags; + struct kretprobe_instance *ri; + + unregister_kprobe(&rp->kp); + /* No race here */ + spin_lock_irqsave(&kprobe_lock, flags); + free_rp_inst(rp); + while ((ri = get_used_rp_inst(rp)) != NULL) { + ri->rp = NULL; + hlist_del(&ri->uflist); + } + spin_unlock_irqrestore(&kprobe_lock, flags); +} + static int __init init_kprobes(void) { int i, err = 0; /* FIXME allocate the probe table, currently defined statically */ /* initialize all list heads */ - for (i = 0; i < KPROBE_TABLE_SIZE; i++) + for (i = 0; i < KPROBE_TABLE_SIZE; i++) { INIT_HLIST_HEAD(&kprobe_table[i]); + INIT_HLIST_HEAD(&kretprobe_inst_table[i]); + } err = register_die_notifier(&kprobe_exceptions_nb); + /* Register the trampoline probe for return probe */ + register_kprobe(&trampoline_p); return err; } @@ -277,3 +477,6 @@ EXPORT_SYMBOL_GPL(unregister_kprobe); EXPORT_SYMBOL_GPL(register_jprobe); EXPORT_SYMBOL_GPL(unregister_jprobe); EXPORT_SYMBOL_GPL(jprobe_return); +EXPORT_SYMBOL_GPL(register_kretprobe); +EXPORT_SYMBOL_GPL(unregister_kretprobe); + -- cgit v1.2.3 From 7e1048b11c5afe79aac46a42e3ccec86b8365c6d Mon Sep 17 00:00:00 2001 From: Rusty Lynch Date: Thu, 23 Jun 2005 00:09:25 -0700 Subject: [PATCH] Move kprobe [dis]arming into arch specific code The architecture independent code of the current kprobes implementation is arming and disarming kprobes at registration time. The problem is that the code is assuming that arming and disarming is a just done by a simple write of some magic value to an address. This is problematic for ia64 where our instructions look more like structures, and we can not insert break points by just doing something like: *p->addr = BREAKPOINT_INSTRUCTION; The following patch to 2.6.12-rc4-mm2 adds two new architecture dependent functions: * void arch_arm_kprobe(struct kprobe *p) * void arch_disarm_kprobe(struct kprobe *p) and then adds the new functions for each of the architectures that already implement kprobes (spar64/ppc64/i386/x86_64). I thought arch_[dis]arm_kprobe was the most descriptive of what was really happening, but each of the architectures already had a disarm_kprobe() function that was really a "disarm and do some other clean-up items as needed when you stumble across a recursive kprobe." So... I took the liberty of changing the code that was calling disarm_kprobe() to call arch_disarm_kprobe(), and then do the cleanup in the block of code dealing with the recursive kprobe case. So far this patch as been tested on i386, x86_64, and ppc64, but still needs to be tested in sparc64. Signed-off-by: Rusty Lynch Signed-off-by: Anil S Keshavamurthy Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/kprobes.c | 19 +++++++++++++++---- arch/ppc64/kernel/kprobes.c | 19 +++++++++++++++---- arch/sparc64/kernel/kprobes.c | 31 ++++++++++++++++++------------- arch/x86_64/kernel/kprobes.c | 26 ++++++++++++++++++-------- include/linux/kprobes.h | 2 ++ kernel/kprobes.c | 12 ++++-------- 6 files changed, 72 insertions(+), 37 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index 048f754bbe23..2314d8d306fd 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include @@ -71,16 +72,25 @@ int arch_prepare_kprobe(struct kprobe *p) void arch_copy_kprobe(struct kprobe *p) { memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); + p->opcode = *p->addr; } -void arch_remove_kprobe(struct kprobe *p) +void arch_arm_kprobe(struct kprobe *p) { + *p->addr = BREAKPOINT_INSTRUCTION; + flush_icache_range((unsigned long) p->addr, + (unsigned long) p->addr + sizeof(kprobe_opcode_t)); } -static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs) +void arch_disarm_kprobe(struct kprobe *p) { *p->addr = p->opcode; - regs->eip = (unsigned long)p->addr; + flush_icache_range((unsigned long) p->addr, + (unsigned long) p->addr + sizeof(kprobe_opcode_t)); +} + +void arch_remove_kprobe(struct kprobe *p) +{ } static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs) @@ -177,7 +187,8 @@ static int kprobe_handler(struct pt_regs *regs) unlock_kprobes(); goto no_kprobe; } - disarm_kprobe(p, regs); + arch_disarm_kprobe(p); + regs->eip = (unsigned long)p->addr; ret = 1; } else { p = current_kprobe; diff --git a/arch/ppc64/kernel/kprobes.c b/arch/ppc64/kernel/kprobes.c index e950a2058a19..8c0920a6d03e 100644 --- a/arch/ppc64/kernel/kprobes.c +++ b/arch/ppc64/kernel/kprobes.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include @@ -61,16 +62,25 @@ int arch_prepare_kprobe(struct kprobe *p) void arch_copy_kprobe(struct kprobe *p) { memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); + p->opcode = *p->addr; } -void arch_remove_kprobe(struct kprobe *p) +void arch_arm_kprobe(struct kprobe *p) { + *p->addr = BREAKPOINT_INSTRUCTION; + flush_icache_range((unsigned long) p->addr, + (unsigned long) p->addr + sizeof(kprobe_opcode_t)); } -static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs) +void arch_disarm_kprobe(struct kprobe *p) { *p->addr = p->opcode; - regs->nip = (unsigned long)p->addr; + flush_icache_range((unsigned long) p->addr, + (unsigned long) p->addr + sizeof(kprobe_opcode_t)); +} + +void arch_remove_kprobe(struct kprobe *p) +{ } static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs) @@ -101,7 +111,8 @@ static inline int kprobe_handler(struct pt_regs *regs) unlock_kprobes(); goto no_kprobe; } - disarm_kprobe(p, regs); + arch_disarm_kprobe(p); + regs->nip = (unsigned long)p->addr; ret = 1; } else { p = current_kprobe; diff --git a/arch/sparc64/kernel/kprobes.c b/arch/sparc64/kernel/kprobes.c index 7066d7ba667a..d67195ba3fa2 100644 --- a/arch/sparc64/kernel/kprobes.c +++ b/arch/sparc64/kernel/kprobes.c @@ -6,7 +6,6 @@ #include #include #include - #include #include @@ -47,6 +46,19 @@ void arch_copy_kprobe(struct kprobe *p) { p->ainsn.insn[0] = *p->addr; p->ainsn.insn[1] = BREAKPOINT_INSTRUCTION_2; + p->opcode = *p->addr; +} + +void arch_arm_kprobe(struct kprobe *p) +{ + *p->addr = BREAKPOINT_INSTRUCTION; + flushi(p->addr); +} + +void arch_disarm_kprobe(struct kprobe *p) +{ + *p->addr = p->opcode; + flushi(p->addr); } void arch_remove_kprobe(struct kprobe *p) @@ -78,17 +90,6 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs) } } -static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs) -{ - *p->addr = p->opcode; - flushi(p->addr); - - regs->tpc = (unsigned long) p->addr; - regs->tnpc = current_kprobe_orig_tnpc; - regs->tstate = ((regs->tstate & ~TSTATE_PIL) | - current_kprobe_orig_tstate_pil); -} - static int kprobe_handler(struct pt_regs *regs) { struct kprobe *p; @@ -109,7 +110,11 @@ static int kprobe_handler(struct pt_regs *regs) unlock_kprobes(); goto no_kprobe; } - disarm_kprobe(p, regs); + arch_disarm_kprobe(p); + regs->tpc = (unsigned long) p->addr; + regs->tnpc = current_kprobe_orig_tnpc; + regs->tstate = ((regs->tstate & ~TSTATE_PIL) | + current_kprobe_orig_tstate_pil); ret = 1; } else { p = current_kprobe; diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c index 203672ca7401..324bf57925a9 100644 --- a/arch/x86_64/kernel/kprobes.c +++ b/arch/x86_64/kernel/kprobes.c @@ -39,7 +39,7 @@ #include #include #include - +#include #include #include @@ -216,19 +216,28 @@ void arch_copy_kprobe(struct kprobe *p) BUG_ON((s64) (s32) disp != disp); /* Sanity check. */ *ripdisp = disp; } + p->opcode = *p->addr; } -void arch_remove_kprobe(struct kprobe *p) +void arch_arm_kprobe(struct kprobe *p) { - up(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); - down(&kprobe_mutex); + *p->addr = BREAKPOINT_INSTRUCTION; + flush_icache_range((unsigned long) p->addr, + (unsigned long) p->addr + sizeof(kprobe_opcode_t)); } -static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs) +void arch_disarm_kprobe(struct kprobe *p) { *p->addr = p->opcode; - regs->rip = (unsigned long)p->addr; + flush_icache_range((unsigned long) p->addr, + (unsigned long) p->addr + sizeof(kprobe_opcode_t)); +} + +void arch_remove_kprobe(struct kprobe *p) +{ + up(&kprobe_mutex); + free_insn_slot(p->ainsn.insn); + down(&kprobe_mutex); } static void prepare_singlestep(struct kprobe *p, struct pt_regs *regs) @@ -311,7 +320,8 @@ int kprobe_handler(struct pt_regs *regs) unlock_kprobes(); goto no_kprobe; } - disarm_kprobe(p, regs); + arch_disarm_kprobe(p); + regs->rip = (unsigned long)p->addr; ret = 1; } else { p = current_kprobe; diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index fba39f87efec..0f90466fb8b0 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -165,6 +165,8 @@ static inline int kprobe_running(void) extern int arch_prepare_kprobe(struct kprobe *p); extern void arch_copy_kprobe(struct kprobe *p); +extern void arch_arm_kprobe(struct kprobe *p); +extern void arch_disarm_kprobe(struct kprobe *p); extern void arch_remove_kprobe(struct kprobe *p); extern void show_registers(struct pt_regs *regs); diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 692fbf75ab49..e8e0ae8a6e14 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -261,7 +261,7 @@ static inline void free_rp_inst(struct kretprobe *rp) static inline void add_aggr_kprobe(struct kprobe *ap, struct kprobe *p) { ap->addr = p->addr; - ap->opcode = p->opcode; + memcpy(&ap->opcode, &p->opcode, sizeof(kprobe_opcode_t)); memcpy(&ap->ainsn, &p->ainsn, sizeof(struct arch_specific_insn)); ap->pre_handler = aggr_pre_handler; @@ -304,10 +304,8 @@ static int register_aggr_kprobe(struct kprobe *old_p, struct kprobe *p) /* kprobe removal house-keeping routines */ static inline void cleanup_kprobe(struct kprobe *p, unsigned long flags) { - *p->addr = p->opcode; + arch_disarm_kprobe(p); hlist_del(&p->hlist); - flush_icache_range((unsigned long) p->addr, - (unsigned long) p->addr + sizeof(kprobe_opcode_t)); spin_unlock_irqrestore(&kprobe_lock, flags); arch_remove_kprobe(p); } @@ -344,10 +342,8 @@ int register_kprobe(struct kprobe *p) hlist_add_head(&p->hlist, &kprobe_table[hash_ptr(p->addr, KPROBE_HASH_BITS)]); - p->opcode = *p->addr; - *p->addr = BREAKPOINT_INSTRUCTION; - flush_icache_range((unsigned long) p->addr, - (unsigned long) p->addr + sizeof(kprobe_opcode_t)); + arch_arm_kprobe(p); + out: spin_unlock_irqrestore(&kprobe_lock, flags); rm_kprobe: -- cgit v1.2.3 From 0aa55e4d7db822059fe8132fe9f2b7773c48216c Mon Sep 17 00:00:00 2001 From: Hien Nguyen Date: Thu, 23 Jun 2005 00:09:26 -0700 Subject: [PATCH] kprobes: moves lock-unlock to non-arch kprobe_flush_task This patch moves the lock/unlock of the arch specific kprobe_flush_task() to the non-arch specific kprobe_flusk_task(). Signed-off-by: Hien Nguyen Acked-by: Prasanna S Panchamukhi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/kprobes.c | 5 +---- include/linux/kprobes.h | 3 +-- kernel/kprobes.c | 5 ++++- 3 files changed, 6 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index 2314d8d306fd..b8e2bae0ab4f 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -138,17 +138,14 @@ void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs) } } -void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock) +void arch_kprobe_flush_task(struct task_struct *tk) { - unsigned long flags = 0; struct kretprobe_instance *ri; - spin_lock_irqsave(kp_lock, flags); while ((ri = get_rp_inst_tsk(tk)) != NULL) { *((unsigned long *)(ri->stack_addr)) = (unsigned long) ri->ret_addr; recycle_rp_inst(ri); } - spin_unlock_irqrestore(kp_lock, flags); } /* diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index 0f90466fb8b0..461391decc46 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -33,7 +33,6 @@ #include #include #include -#include #include @@ -101,7 +100,7 @@ extern void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs, unsigned long flags); extern struct task_struct *arch_get_kprobe_task(void *ptr); extern void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs); -extern void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock); +extern void arch_kprobe_flush_task(struct task_struct *tk); #else /* ARCH_SUPPORTS_KRETPROBES */ static inline void kretprobe_trampoline(void) { diff --git a/kernel/kprobes.c b/kernel/kprobes.c index e8e0ae8a6e14..dd42e717dd35 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -229,7 +229,10 @@ struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk) */ void kprobe_flush_task(struct task_struct *tk) { - arch_kprobe_flush_task(tk, &kprobe_lock); + unsigned long flags = 0; + spin_lock_irqsave(&kprobe_lock, flags); + arch_kprobe_flush_task(tk); + spin_unlock_irqrestore(&kprobe_lock, flags); } /* -- cgit v1.2.3 From ea32c65cc2d2294c04e9f81d0578a6f51febfdbf Mon Sep 17 00:00:00 2001 From: Prasanna S Panchamukhi Date: Thu, 23 Jun 2005 00:09:36 -0700 Subject: [PATCH] kprobes: Temporary disarming of reentrant probe In situations where a kprobes handler calls a routine which has a probe on it, then kprobes_handler() disarms the new probe forever. This patch removes the above limitation by temporarily disarming the new probe. When the another probe hits while handling the old probe, the kprobes_handler() saves previous kprobes state and handles the new probe without calling the new kprobes registered handlers. kprobe_post_handler() restores back the previous kprobes state and the normal execution continues. However on x86_64 architecture, re-rentrancy is provided only through pre_handler(). If a routine having probe is referenced through post_handler(), then the probes on that routine are disarmed forever, since the exception stack is gets changed after the processor single steps the instruction of the new probe. This patch includes generic changes to support temporary disarming on reentrancy of probes. Signed-of-by: Prasanna S Panchamukhi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kprobes.h | 9 +++++++++ kernel/kprobes.c | 1 + 2 files changed, 10 insertions(+) (limited to 'include/linux') diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index 461391decc46..5e1a7b0d7b3f 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -36,6 +36,12 @@ #include +/* kprobe_status settings */ +#define KPROBE_HIT_ACTIVE 0x00000001 +#define KPROBE_HIT_SS 0x00000002 +#define KPROBE_REENTER 0x00000004 +#define KPROBE_HIT_SSDONE 0x00000008 + struct kprobe; struct pt_regs; struct kretprobe; @@ -55,6 +61,9 @@ struct kprobe { /* list of kprobes for multi-handler support */ struct list_head list; + /*count the number of times this probe was temporarily disarmed */ + unsigned long nmissed; + /* location of the probe point */ kprobe_opcode_t *addr; diff --git a/kernel/kprobes.c b/kernel/kprobes.c index dd42e717dd35..456ecedff2d4 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -335,6 +335,7 @@ int register_kprobe(struct kprobe *p) } spin_lock_irqsave(&kprobe_lock, flags); old_p = get_kprobe(p->addr); + p->nmissed = 0; if (old_p) { ret = register_aggr_kprobe(old_p, p); goto out; -- cgit v1.2.3 From d6e711448137ca3301512cec41a2c2ce852b3d0a Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Thu, 23 Jun 2005 00:09:43 -0700 Subject: [PATCH] setuid core dump Add a new `suid_dumpable' sysctl: This value can be used to query and set the core dump mode for setuid or otherwise protected/tainted binaries. The modes are 0 - (default) - traditional behaviour. Any process which has changed privilege levels or is execute only will not be dumped 1 - (debug) - all processes dump core when possible. The core dump is owned by the current user and no security is applied. This is intended for system debugging situations only. Ptrace is unchecked. 2 - (suidsafe) - any binary which normally would not be dumped is dumped readable by root only. This allows the end user to remove such a dump but not access it directly. For security reasons core dumps in this mode will not overwrite one another or other files. This mode is appropriate when adminstrators are attempting to debug problems in a normal environment. (akpm: > > +EXPORT_SYMBOL(suid_dumpable); > > EXPORT_SYMBOL_GPL? No problem to me. > > if (current->euid == current->uid && current->egid == current->gid) > > current->mm->dumpable = 1; > > Should this be SUID_DUMP_USER? Actually the feedback I had from last time was that the SUID_ defines should go because its clearer to follow the numbers. They can go everywhere (and there are lots of places where dumpable is tested/used as a bool in untouched code) > Maybe this should be renamed to `dump_policy' or something. Doing that > would help us catch any code which isn't using the #defines, too. Fair comment. The patch was designed to be easy to maintain for Red Hat rather than for merging. Changing that field would create a gigantic diff because it is used all over the place. ) Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/sysctl/kernel.txt | 20 ++++++++++++++++++++ fs/exec.c | 23 +++++++++++++++++++++-- fs/proc/base.c | 6 ++++-- include/linux/binfmts.h | 5 +++++ include/linux/sched.h | 2 +- include/linux/sysctl.h | 1 + kernel/sys.c | 22 +++++++++++----------- kernel/sysctl.c | 9 +++++++++ security/commoncap.c | 2 +- security/dummy.c | 2 +- 10 files changed, 74 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt index 35159176997b..9f11d36a8c10 100644 --- a/Documentation/sysctl/kernel.txt +++ b/Documentation/sysctl/kernel.txt @@ -49,6 +49,7 @@ show up in /proc/sys/kernel: - shmmax [ sysv ipc ] - shmmni - stop-a [ SPARC only ] +- suid_dumpable - sysrq ==> Documentation/sysrq.txt - tainted - threads-max @@ -300,6 +301,25 @@ kernel. This value defaults to SHMMAX. ============================================================== +suid_dumpable: + +This value can be used to query and set the core dump mode for setuid +or otherwise protected/tainted binaries. The modes are + +0 - (default) - traditional behaviour. Any process which has changed + privilege levels or is execute only will not be dumped +1 - (debug) - all processes dump core when possible. The core dump is + owned by the current user and no security is applied. This is + intended for system debugging situations only. Ptrace is unchecked. +2 - (suidsafe) - any binary which normally would not be dumped is dumped + readable by root only. This allows the end user to remove + such a dump but not access it directly. For security reasons + core dumps in this mode will not overwrite one another or + other files. This mode is appropriate when adminstrators are + attempting to debug problems in a normal environment. + +============================================================== + tainted: Non-zero if the kernel has been tainted. Numeric values, which diff --git a/fs/exec.c b/fs/exec.c index 3a4b35a14c0d..48871917d363 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -58,6 +58,9 @@ int core_uses_pid; char core_pattern[65] = "core"; +int suid_dumpable = 0; + +EXPORT_SYMBOL(suid_dumpable); /* The maximal length of core_pattern is also specified in sysctl.c */ static struct linux_binfmt *formats; @@ -864,6 +867,9 @@ int flush_old_exec(struct linux_binprm * bprm) if (current->euid == current->uid && current->egid == current->gid) current->mm->dumpable = 1; + else + current->mm->dumpable = suid_dumpable; + name = bprm->filename; /* Copies the binary name from after last slash */ @@ -884,7 +890,7 @@ int flush_old_exec(struct linux_binprm * bprm) permission(bprm->file->f_dentry->d_inode,MAY_READ, NULL) || (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP)) { suid_keys(current); - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; } /* An exec changes our domain. We are no longer part of the thread @@ -1432,6 +1438,8 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) struct inode * inode; struct file * file; int retval = 0; + int fsuid = current->fsuid; + int flag = 0; binfmt = current->binfmt; if (!binfmt || !binfmt->core_dump) @@ -1441,6 +1449,16 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) up_write(&mm->mmap_sem); goto fail; } + + /* + * We cannot trust fsuid as being the "true" uid of the + * process nor do we know its entire history. We only know it + * was tainted so we dump it as root in mode 2. + */ + if (mm->dumpable == 2) { /* Setuid core dump mode */ + flag = O_EXCL; /* Stop rewrite attacks */ + current->fsuid = 0; /* Dump root private */ + } mm->dumpable = 0; init_completion(&mm->core_done); spin_lock_irq(¤t->sighand->siglock); @@ -1466,7 +1484,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) lock_kernel(); format_corename(corename, core_pattern, signr); unlock_kernel(); - file = filp_open(corename, O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE, 0600); + file = filp_open(corename, O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag, 0600); if (IS_ERR(file)) goto fail_unlock; inode = file->f_dentry->d_inode; @@ -1491,6 +1509,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) close_fail: filp_close(file, NULL); fail_unlock: + current->fsuid = fsuid; complete_all(&mm->core_done); fail: return retval; diff --git a/fs/proc/base.c b/fs/proc/base.c index e31903aadd96..ace151fa4878 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -314,7 +314,7 @@ static int may_ptrace_attach(struct task_struct *task) (current->gid != task->gid)) && !capable(CAP_SYS_PTRACE)) goto out; rmb(); - if (!task->mm->dumpable && !capable(CAP_SYS_PTRACE)) + if (task->mm->dumpable != 1 && !capable(CAP_SYS_PTRACE)) goto out; if (security_ptrace(current, task)) goto out; @@ -1113,7 +1113,9 @@ static int task_dumpable(struct task_struct *task) if (mm) dumpable = mm->dumpable; task_unlock(task); - return dumpable; + if(dumpable == 1) + return 1; + return 0; } diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index 7e736e201c46..c1e82c514443 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -69,6 +69,11 @@ extern void remove_arg_zero(struct linux_binprm *); extern int search_binary_handler(struct linux_binprm *,struct pt_regs *); extern int flush_old_exec(struct linux_binprm * bprm); +extern int suid_dumpable; +#define SUID_DUMP_DISABLE 0 /* No setuid dumping */ +#define SUID_DUMP_USER 1 /* Dump as user of process */ +#define SUID_DUMP_ROOT 2 /* Dump as root */ + /* Stack area protections */ #define EXSTACK_DEFAULT 0 /* Whatever the arch defaults to */ #define EXSTACK_DISABLE_X 1 /* Disable executable stacks */ diff --git a/include/linux/sched.h b/include/linux/sched.h index b58afd97a180..901742f92389 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -246,7 +246,7 @@ struct mm_struct { unsigned long saved_auxv[42]; /* for /proc/PID/auxv */ - unsigned dumpable:1; + unsigned dumpable:2; cpumask_t cpu_vm_mask; /* Architecture-specific MM context */ diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index a17745c80a91..614e939c78a4 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -136,6 +136,7 @@ enum KERN_UNKNOWN_NMI_PANIC=66, /* int: unknown nmi panic flag */ KERN_BOOTLOADER_TYPE=67, /* int: boot loader type */ KERN_RANDOMIZE=68, /* int: randomize virtual address space */ + KERN_SETUID_DUMPABLE=69, /* int: behaviour of dumps for setuid core */ }; diff --git a/kernel/sys.c b/kernel/sys.c index f006632c2ba7..0a2c8cda9638 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -525,7 +525,7 @@ asmlinkage long sys_setregid(gid_t rgid, gid_t egid) } if (new_egid != old_egid) { - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } if (rgid != (gid_t) -1 || @@ -556,7 +556,7 @@ asmlinkage long sys_setgid(gid_t gid) { if(old_egid != gid) { - current->mm->dumpable=0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } current->gid = current->egid = current->sgid = current->fsgid = gid; @@ -565,7 +565,7 @@ asmlinkage long sys_setgid(gid_t gid) { if(old_egid != gid) { - current->mm->dumpable=0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } current->egid = current->fsgid = gid; @@ -596,7 +596,7 @@ static int set_user(uid_t new_ruid, int dumpclear) if(dumpclear) { - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } current->uid = new_ruid; @@ -653,7 +653,7 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) if (new_euid != old_euid) { - current->mm->dumpable=0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } current->fsuid = current->euid = new_euid; @@ -703,7 +703,7 @@ asmlinkage long sys_setuid(uid_t uid) if (old_euid != uid) { - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } current->fsuid = current->euid = uid; @@ -748,7 +748,7 @@ asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid) if (euid != (uid_t) -1) { if (euid != current->euid) { - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } current->euid = euid; @@ -798,7 +798,7 @@ asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid) if (egid != (gid_t) -1) { if (egid != current->egid) { - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } current->egid = egid; @@ -845,7 +845,7 @@ asmlinkage long sys_setfsuid(uid_t uid) { if (uid != old_fsuid) { - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } current->fsuid = uid; @@ -875,7 +875,7 @@ asmlinkage long sys_setfsgid(gid_t gid) { if (gid != old_fsgid) { - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; smp_wmb(); } current->fsgid = gid; @@ -1652,7 +1652,7 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, error = 1; break; case PR_SET_DUMPABLE: - if (arg2 != 0 && arg2 != 1) { + if (arg2 < 0 || arg2 > 2) { error = -EINVAL; break; } diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 701d12c63068..24a4d12d5aa9 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -58,6 +58,7 @@ extern int sysctl_overcommit_ratio; extern int max_threads; extern int sysrq_enabled; extern int core_uses_pid; +extern int suid_dumpable; extern char core_pattern[]; extern int cad_pid; extern int pid_max; @@ -950,6 +951,14 @@ static ctl_table fs_table[] = { .proc_handler = &proc_dointvec, }, #endif + { + .ctl_name = KERN_SETUID_DUMPABLE, + .procname = "suid_dumpable", + .data = &suid_dumpable, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, { .ctl_name = 0 } }; diff --git a/security/commoncap.c b/security/commoncap.c index 849b8c338ee8..04c12f58d656 100644 --- a/security/commoncap.c +++ b/security/commoncap.c @@ -149,7 +149,7 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) if (bprm->e_uid != current->uid || bprm->e_gid != current->gid || !cap_issubset (new_permitted, current->cap_permitted)) { - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; if (unsafe & ~LSM_UNSAFE_PTRACE_CAP) { if (!capable(CAP_SETUID)) { diff --git a/security/dummy.c b/security/dummy.c index b32eff146547..6ff887586479 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -130,7 +130,7 @@ static void dummy_bprm_free_security (struct linux_binprm *bprm) static void dummy_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) { if (bprm->e_uid != current->uid || bprm->e_gid != current->gid) { - current->mm->dumpable = 0; + current->mm->dumpable = suid_dumpable; if ((unsafe & ~LSM_UNSAFE_PTRACE_CAP) && !capable(CAP_SETUID)) { bprm->e_uid = current->uid; -- cgit v1.2.3 From ef3daeda7b58f046f94b26637d500354038d39f4 Mon Sep 17 00:00:00 2001 From: Yoav Zach Date: Thu, 23 Jun 2005 00:09:58 -0700 Subject: [PATCH] Don't force O_LARGEFILE for 32 bit processes on ia64 In ia64 kernel, the O_LARGEFILE flag is forced when opening a file. This is problematic for execution of 32 bit processes, which are not largefile aware, either by SW emulation or by HW execution. For such processes, the problem is two-fold: 1) When trying to open a file that is larger than 4G the operation should fail, but it's not 2) Writing to offset larger than 4G should fail, but it's not The proposed patch takes advantage of the way 32 bit processes are identified in ia64 systems. Such processes have PER_LINUX32 for their personality. With the patch, the ia64 kernel will not enforce the O_LARGEFILE flag if the current process has PER_LINUX32 set. The behavior for all other architectures remains unchanged. Signed-off-by: Yoav Zach Acked-by: Tony Luck Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/open.c | 7 ++++--- include/asm-ia64/fcntl.h | 2 ++ include/linux/fcntl.h | 4 ++++ 3 files changed, 10 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/fs/open.c b/fs/open.c index 963bd81a44c8..2ebb72c1a876 100644 --- a/fs/open.c +++ b/fs/open.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include @@ -935,9 +936,9 @@ asmlinkage long sys_open(const char __user * filename, int flags, int mode) char * tmp; int fd, error; -#if BITS_PER_LONG != 32 - flags |= O_LARGEFILE; -#endif + if (force_o_largefile()) + flags |= O_LARGEFILE; + tmp = getname(filename); fd = PTR_ERR(tmp); if (!IS_ERR(tmp)) { diff --git a/include/asm-ia64/fcntl.h b/include/asm-ia64/fcntl.h index d193981bb1d8..c9f8d835d0cc 100644 --- a/include/asm-ia64/fcntl.h +++ b/include/asm-ia64/fcntl.h @@ -81,4 +81,6 @@ struct flock { #define F_LINUX_SPECIFIC_BASE 1024 +#define force_o_largefile() ( ! (current->personality & PER_LINUX32) ) + #endif /* _ASM_IA64_FCNTL_H */ diff --git a/include/linux/fcntl.h b/include/linux/fcntl.h index 704fb76b6334..8a7c82151de9 100644 --- a/include/linux/fcntl.h +++ b/include/linux/fcntl.h @@ -25,6 +25,10 @@ #ifdef __KERNEL__ +#ifndef force_o_largefile +#define force_o_largefile() (BITS_PER_LONG != 32) +#endif + #if BITS_PER_LONG == 32 #define IS_GETLK32(cmd) ((cmd) == F_GETLK) #define IS_SETLK32(cmd) ((cmd) == F_SETLK) -- cgit v1.2.3 From 46c271bedd2c8444b1d05bc44928beec0c07debc Mon Sep 17 00:00:00 2001 From: Peter Osterlund Date: Thu, 23 Jun 2005 00:10:02 -0700 Subject: [PATCH] Improve CD/DVD packet driver write performance This patch improves write performance for the CD/DVD packet writing driver. The logic for switching between reading and writing has been changed so that streaming writes are no longer interrupted by read requests. Signed-off-by: Peter Osterlund Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/pktcdvd.c | 36 ++++++++++++++++++++---------------- include/linux/pktcdvd.h | 2 +- 2 files changed, 21 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index bc56770bcc90..7f3d78de265c 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -467,14 +467,12 @@ static int pkt_set_speed(struct pktcdvd_device *pd, unsigned write_speed, unsign * Queue a bio for processing by the low-level CD device. Must be called * from process context. */ -static void pkt_queue_bio(struct pktcdvd_device *pd, struct bio *bio, int high_prio_read) +static void pkt_queue_bio(struct pktcdvd_device *pd, struct bio *bio) { spin_lock(&pd->iosched.lock); if (bio_data_dir(bio) == READ) { pkt_add_list_last(bio, &pd->iosched.read_queue, &pd->iosched.read_queue_tail); - if (high_prio_read) - pd->iosched.high_prio_read = 1; } else { pkt_add_list_last(bio, &pd->iosched.write_queue, &pd->iosched.write_queue_tail); @@ -490,15 +488,16 @@ static void pkt_queue_bio(struct pktcdvd_device *pd, struct bio *bio, int high_p * requirements for CDRW drives: * - A cache flush command must be inserted before a read request if the * previous request was a write. - * - Switching between reading and writing is slow, so don't it more often + * - Switching between reading and writing is slow, so don't do it more often * than necessary. + * - Optimize for throughput at the expense of latency. This means that streaming + * writes will never be interrupted by a read, but if the drive has to seek + * before the next write, switch to reading instead if there are any pending + * read requests. * - Set the read speed according to current usage pattern. When only reading * from the device, it's best to use the highest possible read speed, but * when switching often between reading and writing, it's better to have the * same read and write speeds. - * - Reads originating from user space should have higher priority than reads - * originating from pkt_gather_data, because some process is usually waiting - * on reads of the first kind. */ static void pkt_iosched_process_queue(struct pktcdvd_device *pd) { @@ -512,21 +511,24 @@ static void pkt_iosched_process_queue(struct pktcdvd_device *pd) for (;;) { struct bio *bio; - int reads_queued, writes_queued, high_prio_read; + int reads_queued, writes_queued; spin_lock(&pd->iosched.lock); reads_queued = (pd->iosched.read_queue != NULL); writes_queued = (pd->iosched.write_queue != NULL); - if (!reads_queued) - pd->iosched.high_prio_read = 0; - high_prio_read = pd->iosched.high_prio_read; spin_unlock(&pd->iosched.lock); if (!reads_queued && !writes_queued) break; if (pd->iosched.writing) { - if (high_prio_read || (!writes_queued && reads_queued)) { + int need_write_seek = 1; + spin_lock(&pd->iosched.lock); + bio = pd->iosched.write_queue; + spin_unlock(&pd->iosched.lock); + if (bio && (bio->bi_sector == pd->iosched.last_write)) + need_write_seek = 0; + if (need_write_seek && reads_queued) { if (atomic_read(&pd->cdrw.pending_bios) > 0) { VPRINTK("pktcdvd: write, waiting\n"); break; @@ -559,8 +561,10 @@ static void pkt_iosched_process_queue(struct pktcdvd_device *pd) if (bio_data_dir(bio) == READ) pd->iosched.successive_reads += bio->bi_size >> 10; - else + else { pd->iosched.successive_reads = 0; + pd->iosched.last_write = bio->bi_sector + bio_sectors(bio); + } if (pd->iosched.successive_reads >= HI_SPEED_SWITCH) { if (pd->read_speed == pd->write_speed) { pd->read_speed = MAX_SPEED; @@ -765,7 +769,7 @@ static void pkt_gather_data(struct pktcdvd_device *pd, struct packet_data *pkt) atomic_inc(&pkt->io_wait); bio->bi_rw = READ; - pkt_queue_bio(pd, bio, 0); + pkt_queue_bio(pd, bio); frames_read++; } @@ -1062,7 +1066,7 @@ static void pkt_start_write(struct pktcdvd_device *pd, struct packet_data *pkt) atomic_set(&pkt->io_wait, 1); pkt->w_bio->bi_rw = WRITE; - pkt_queue_bio(pd, pkt->w_bio, 0); + pkt_queue_bio(pd, pkt->w_bio); } static void pkt_finish_packet(struct packet_data *pkt, int uptodate) @@ -2120,7 +2124,7 @@ static int pkt_make_request(request_queue_t *q, struct bio *bio) cloned_bio->bi_private = psd; cloned_bio->bi_end_io = pkt_end_io_read_cloned; pd->stats.secs_r += bio->bi_size >> 9; - pkt_queue_bio(pd, cloned_bio, 1); + pkt_queue_bio(pd, cloned_bio); return 0; } diff --git a/include/linux/pktcdvd.h b/include/linux/pktcdvd.h index 4e2d2a942ecb..4b32bce9a289 100644 --- a/include/linux/pktcdvd.h +++ b/include/linux/pktcdvd.h @@ -159,7 +159,7 @@ struct packet_iosched struct bio *read_queue_tail; struct bio *write_queue; struct bio *write_queue_tail; - int high_prio_read; /* An important read request has been queued */ + sector_t last_write; /* The sector where the last write ended */ int successive_reads; }; -- cgit v1.2.3 From bb93e3a52f8db7210258a1a2134cced0b78a46e1 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 23 Jun 2005 00:10:15 -0700 Subject: [PATCH] block: add unlocked_ioctl support for block devices This patch allows block device drivers to convert their ioctl functions to unlocked_ioctl() like character devices and other subsystems. All functions that were called with the BKL held before are still used that way, but I would not be surprised if it could be removed from the ioctl functions in drivers/block/ioctl.c themselves. As a side note, I found that compat_blkdev_ioctl() acquires the BKL as well, which looks like a bug. I have checked that every user of disk->fops->compat_ioctl() in the current git tree gets the BKL itself, so it could easily be removed from compat_blkdev_ioctl(). Signed-off-by: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/ioctl.c | 74 +++++++++++++++++++++++++++++++++++++-------------- fs/block_dev.c | 5 ++-- include/linux/fs.h | 1 + 3 files changed, 57 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/ioctl.c b/drivers/block/ioctl.c index 6d7bcc9da9e7..6e278474f9a8 100644 --- a/drivers/block/ioctl.c +++ b/drivers/block/ioctl.c @@ -133,11 +133,9 @@ static int put_u64(unsigned long arg, u64 val) return put_user(val, (u64 __user *)arg); } -int blkdev_ioctl(struct inode *inode, struct file *file, unsigned cmd, - unsigned long arg) +static int blkdev_locked_ioctl(struct file *file, struct block_device *bdev, + unsigned cmd, unsigned long arg) { - struct block_device *bdev = inode->i_bdev; - struct gendisk *disk = bdev->bd_disk; struct backing_dev_info *bdi; int ret, n; @@ -190,36 +188,72 @@ int blkdev_ioctl(struct inode *inode, struct file *file, unsigned cmd, return put_ulong(arg, bdev->bd_inode->i_size >> 9); case BLKGETSIZE64: return put_u64(arg, bdev->bd_inode->i_size); + } + return -ENOIOCTLCMD; +} + +static int blkdev_driver_ioctl(struct inode *inode, struct file *file, + struct gendisk *disk, unsigned cmd, unsigned long arg) +{ + int ret; + if (disk->fops->unlocked_ioctl) + return disk->fops->unlocked_ioctl(file, cmd, arg); + + if (disk->fops->ioctl) { + lock_kernel(); + ret = disk->fops->ioctl(inode, file, cmd, arg); + unlock_kernel(); + return ret; + } + + return -ENOTTY; +} + +int blkdev_ioctl(struct inode *inode, struct file *file, unsigned cmd, + unsigned long arg) +{ + struct block_device *bdev = inode->i_bdev; + struct gendisk *disk = bdev->bd_disk; + int ret, n; + + switch(cmd) { case BLKFLSBUF: if (!capable(CAP_SYS_ADMIN)) return -EACCES; - if (disk->fops->ioctl) { - ret = disk->fops->ioctl(inode, file, cmd, arg); - /* -EINVAL to handle old uncorrected drivers */ - if (ret != -EINVAL && ret != -ENOTTY) - return ret; - } + + ret = blkdev_driver_ioctl(inode, file, disk, cmd, arg); + /* -EINVAL to handle old uncorrected drivers */ + if (ret != -EINVAL && ret != -ENOTTY) + return ret; + + lock_kernel(); fsync_bdev(bdev); invalidate_bdev(bdev, 0); + unlock_kernel(); return 0; + case BLKROSET: - if (disk->fops->ioctl) { - ret = disk->fops->ioctl(inode, file, cmd, arg); - /* -EINVAL to handle old uncorrected drivers */ - if (ret != -EINVAL && ret != -ENOTTY) - return ret; - } + ret = blkdev_driver_ioctl(inode, file, disk, cmd, arg); + /* -EINVAL to handle old uncorrected drivers */ + if (ret != -EINVAL && ret != -ENOTTY) + return ret; if (!capable(CAP_SYS_ADMIN)) return -EACCES; if (get_user(n, (int __user *)(arg))) return -EFAULT; + lock_kernel(); set_device_ro(bdev, n); + unlock_kernel(); return 0; - default: - if (disk->fops->ioctl) - return disk->fops->ioctl(inode, file, cmd, arg); } - return -ENOTTY; + + lock_kernel(); + ret = blkdev_locked_ioctl(file, bdev, cmd, arg); + unlock_kernel(); + if (ret != -ENOIOCTLCMD) + return ret; + + return blkdev_driver_ioctl(inode, file, disk, cmd, arg); } /* Most of the generic ioctls are handled in the normal fallback path. diff --git a/fs/block_dev.c b/fs/block_dev.c index c0cbd1bc1a02..e0df94c37b7e 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -777,8 +777,7 @@ static ssize_t blkdev_file_aio_write(struct kiocb *iocb, const char __user *buf, return generic_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos); } -static int block_ioctl(struct inode *inode, struct file *file, unsigned cmd, - unsigned long arg) +static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) { return blkdev_ioctl(file->f_mapping->host, file, cmd, arg); } @@ -803,7 +802,7 @@ struct file_operations def_blk_fops = { .aio_write = blkdev_file_aio_write, .mmap = generic_file_mmap, .fsync = block_fsync, - .ioctl = block_ioctl, + .unlocked_ioctl = block_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = compat_blkdev_ioctl, #endif diff --git a/include/linux/fs.h b/include/linux/fs.h index 3622e952e98c..9b1278e21279 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -884,6 +884,7 @@ struct block_device_operations { int (*open) (struct inode *, struct file *); int (*release) (struct inode *, struct file *); int (*ioctl) (struct inode *, struct file *, unsigned, unsigned long); + long (*unlocked_ioctl) (struct file *, unsigned, unsigned long); long (*compat_ioctl) (struct file *, unsigned, unsigned long); int (*media_changed) (struct gendisk *); int (*revalidate_disk) (struct gendisk *); -- cgit v1.2.3 From 45778ca819accab1a4a3378b3566cab0f189164f Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 23 Jun 2005 00:10:17 -0700 Subject: [PATCH] Remove f_error field from struct file The following patch removes the f_error field and all checks of f_error. Trond said: f_error was introduced for NFS, and made sense when we were guaranteed always to have a file pointer around when write errors occurred. Since then, we have (for various reasons) had to introduce the nfs_open_context in order to track the file read/write state, and it made sense to move our f_error tracking there too. Signed-off-by: Christoph Lameter Acked-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfs/direct.c | 5 ----- fs/open.c | 16 ++++------------ include/linux/fs.h | 1 - mm/filemap.c | 6 ------ 4 files changed, 4 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index d6a30c844de3..6537f2c4ae44 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -751,11 +751,6 @@ nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, size_t count, retval = -EFAULT; if (!access_ok(VERIFY_READ, iov.iov_base, iov.iov_len)) goto out; - if (file->f_error) { - retval = file->f_error; - file->f_error = 0; - goto out; - } retval = -EFBIG; if (limit != RLIM_INFINITY) { if (pos >= limit) { diff --git a/fs/open.c b/fs/open.c index 2ebb72c1a876..5dd411b084bf 100644 --- a/fs/open.c +++ b/fs/open.c @@ -981,23 +981,15 @@ asmlinkage long sys_creat(const char __user * pathname, int mode) */ int filp_close(struct file *filp, fl_owner_t id) { - int retval; - - /* Report and clear outstanding errors */ - retval = filp->f_error; - if (retval) - filp->f_error = 0; + int retval = 0; if (!file_count(filp)) { printk(KERN_ERR "VFS: Close: file count is 0\n"); - return retval; + return 0; } - if (filp->f_op && filp->f_op->flush) { - int err = filp->f_op->flush(filp); - if (!retval) - retval = err; - } + if (filp->f_op && filp->f_op->flush) + retval = filp->f_op->flush(filp); dnotify_flush(filp, id); locks_remove_posix(filp, id); diff --git a/include/linux/fs.h b/include/linux/fs.h index 9b1278e21279..517bf4966bf5 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -581,7 +581,6 @@ struct file { atomic_t f_count; unsigned int f_flags; mode_t f_mode; - int f_error; loff_t f_pos; struct fown_struct f_owner; unsigned int f_uid, f_gid; diff --git a/mm/filemap.c b/mm/filemap.c index 4a2fee2cb62b..a3598b542a31 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1827,12 +1827,6 @@ inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, i if (unlikely(*pos < 0)) return -EINVAL; - if (unlikely(file->f_error)) { - int err = file->f_error; - file->f_error = 0; - return err; - } - if (!isblk) { /* FIXME: this is for backwards compatibility with 2.4 */ if (file->f_flags & O_APPEND) -- cgit v1.2.3 From f9fd27a253d5e0b23531d12ce7ad15b6535d4486 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Jun 2005 00:10:19 -0700 Subject: [PATCH] acl endianess annotations Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/posix_acl_xattr.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/posix_acl_xattr.h b/include/linux/posix_acl_xattr.h index 5efd0a6dad94..fe271c1947b2 100644 --- a/include/linux/posix_acl_xattr.h +++ b/include/linux/posix_acl_xattr.h @@ -23,13 +23,13 @@ #define ACL_UNDEFINED_ID (-1) typedef struct { - __u16 e_tag; - __u16 e_perm; - __u32 e_id; + __le16 e_tag; + __le16 e_perm; + __le32 e_id; } posix_acl_xattr_entry; typedef struct { - __u32 a_version; + __le32 a_version; posix_acl_xattr_entry a_entries[0]; } posix_acl_xattr_header; -- cgit v1.2.3 From 9a59f452abe11f569e13ec16c51e6d61c54b9838 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Jun 2005 00:10:19 -0700 Subject: [PATCH] remove This file duplicates , using slightly different names. Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc/boot/simple/misc.c | 2 +- arch/ppc/boot/simple/mpc10x_memory.c | 2 +- fs/ext2/acl.c | 12 ++++++------ fs/ext2/acl.h | 2 +- fs/ext3/acl.c | 12 ++++++------ fs/ext3/acl.h | 2 +- fs/jfs/acl.c | 11 ++++++----- fs/jfs/jfs_acl.h | 2 -- fs/jfs/super.c | 1 + fs/jfs/xattr.c | 7 ++++--- fs/nfsd/vfs.c | 9 ++++----- fs/reiserfs/xattr_acl.c | 26 +++++++++++++------------- include/linux/posix_acl_xattr.h | 3 +++ include/linux/reiserfs_acl.h | 1 - 14 files changed, 47 insertions(+), 45 deletions(-) (limited to 'include/linux') diff --git a/arch/ppc/boot/simple/misc.c b/arch/ppc/boot/simple/misc.c index ab0f9902cb67..e02de5b467a4 100644 --- a/arch/ppc/boot/simple/misc.c +++ b/arch/ppc/boot/simple/misc.c @@ -222,7 +222,7 @@ decompress_kernel(unsigned long load_addr, int num_words, unsigned long cksum) puts("\n"); puts("Uncompressing Linux..."); - gunzip(0x0, 0x400000, zimage_start, &zimage_size); + gunzip(NULL, 0x400000, zimage_start, &zimage_size); puts("done.\n"); /* get the bi_rec address */ diff --git a/arch/ppc/boot/simple/mpc10x_memory.c b/arch/ppc/boot/simple/mpc10x_memory.c index 977daedc14c0..20d92a34ceb8 100644 --- a/arch/ppc/boot/simple/mpc10x_memory.c +++ b/arch/ppc/boot/simple/mpc10x_memory.c @@ -33,7 +33,7 @@ #define MPC10X_PCI_OP(rw, size, type, op, mask) \ static void \ -mpc10x_##rw##_config_##size(unsigned int *cfg_addr, \ +mpc10x_##rw##_config_##size(unsigned int __iomem *cfg_addr, \ unsigned int *cfg_data, int devfn, int offset, \ type val) \ { \ diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c index 25f4a64fd6bc..213148c36ebe 100644 --- a/fs/ext2/acl.c +++ b/fs/ext2/acl.c @@ -396,12 +396,12 @@ static size_t ext2_xattr_list_acl_access(struct inode *inode, char *list, size_t list_size, const char *name, size_t name_len) { - const size_t size = sizeof(XATTR_NAME_ACL_ACCESS); + const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS); if (!test_opt(inode->i_sb, POSIX_ACL)) return 0; if (list && size <= list_size) - memcpy(list, XATTR_NAME_ACL_ACCESS, size); + memcpy(list, POSIX_ACL_XATTR_ACCESS, size); return size; } @@ -409,12 +409,12 @@ static size_t ext2_xattr_list_acl_default(struct inode *inode, char *list, size_t list_size, const char *name, size_t name_len) { - const size_t size = sizeof(XATTR_NAME_ACL_DEFAULT); + const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT); if (!test_opt(inode->i_sb, POSIX_ACL)) return 0; if (list && size <= list_size) - memcpy(list, XATTR_NAME_ACL_DEFAULT, size); + memcpy(list, POSIX_ACL_XATTR_DEFAULT, size); return size; } @@ -506,14 +506,14 @@ ext2_xattr_set_acl_default(struct inode *inode, const char *name, } struct xattr_handler ext2_xattr_acl_access_handler = { - .prefix = XATTR_NAME_ACL_ACCESS, + .prefix = POSIX_ACL_XATTR_ACCESS, .list = ext2_xattr_list_acl_access, .get = ext2_xattr_get_acl_access, .set = ext2_xattr_set_acl_access, }; struct xattr_handler ext2_xattr_acl_default_handler = { - .prefix = XATTR_NAME_ACL_DEFAULT, + .prefix = POSIX_ACL_XATTR_DEFAULT, .list = ext2_xattr_list_acl_default, .get = ext2_xattr_get_acl_default, .set = ext2_xattr_set_acl_default, diff --git a/fs/ext2/acl.h b/fs/ext2/acl.h index fed96ae81a7d..0bde85bafe38 100644 --- a/fs/ext2/acl.h +++ b/fs/ext2/acl.h @@ -4,7 +4,7 @@ (C) 2001 Andreas Gruenbacher, */ -#include +#include #define EXT2_ACL_VERSION 0x0001 diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c index 638c13a26c03..133f5aa581bb 100644 --- a/fs/ext3/acl.c +++ b/fs/ext3/acl.c @@ -417,12 +417,12 @@ static size_t ext3_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len, const char *name, size_t name_len) { - const size_t size = sizeof(XATTR_NAME_ACL_ACCESS); + const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS); if (!test_opt(inode->i_sb, POSIX_ACL)) return 0; if (list && size <= list_len) - memcpy(list, XATTR_NAME_ACL_ACCESS, size); + memcpy(list, POSIX_ACL_XATTR_ACCESS, size); return size; } @@ -430,12 +430,12 @@ static size_t ext3_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len, const char *name, size_t name_len) { - const size_t size = sizeof(XATTR_NAME_ACL_DEFAULT); + const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT); if (!test_opt(inode->i_sb, POSIX_ACL)) return 0; if (list && size <= list_len) - memcpy(list, XATTR_NAME_ACL_DEFAULT, size); + memcpy(list, POSIX_ACL_XATTR_DEFAULT, size); return size; } @@ -535,14 +535,14 @@ ext3_xattr_set_acl_default(struct inode *inode, const char *name, } struct xattr_handler ext3_xattr_acl_access_handler = { - .prefix = XATTR_NAME_ACL_ACCESS, + .prefix = POSIX_ACL_XATTR_ACCESS, .list = ext3_xattr_list_acl_access, .get = ext3_xattr_get_acl_access, .set = ext3_xattr_set_acl_access, }; struct xattr_handler ext3_xattr_acl_default_handler = { - .prefix = XATTR_NAME_ACL_DEFAULT, + .prefix = POSIX_ACL_XATTR_DEFAULT, .list = ext3_xattr_list_acl_default, .get = ext3_xattr_get_acl_default, .set = ext3_xattr_set_acl_default, diff --git a/fs/ext3/acl.h b/fs/ext3/acl.h index 98af0c0d0ba9..92d50b53a933 100644 --- a/fs/ext3/acl.h +++ b/fs/ext3/acl.h @@ -4,7 +4,7 @@ (C) 2001 Andreas Gruenbacher, */ -#include +#include #define EXT3_ACL_VERSION 0x0001 diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c index 30a2bf9eeda5..e892dab40c26 100644 --- a/fs/jfs/acl.c +++ b/fs/jfs/acl.c @@ -21,6 +21,7 @@ #include #include #include +#include #include "jfs_incore.h" #include "jfs_xattr.h" #include "jfs_acl.h" @@ -36,11 +37,11 @@ static struct posix_acl *jfs_get_acl(struct inode *inode, int type) switch(type) { case ACL_TYPE_ACCESS: - ea_name = XATTR_NAME_ACL_ACCESS; + ea_name = POSIX_ACL_XATTR_ACCESS; p_acl = &ji->i_acl; break; case ACL_TYPE_DEFAULT: - ea_name = XATTR_NAME_ACL_DEFAULT; + ea_name = POSIX_ACL_XATTR_DEFAULT; p_acl = &ji->i_default_acl; break; default: @@ -88,11 +89,11 @@ static int jfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) switch(type) { case ACL_TYPE_ACCESS: - ea_name = XATTR_NAME_ACL_ACCESS; + ea_name = POSIX_ACL_XATTR_ACCESS; p_acl = &ji->i_acl; break; case ACL_TYPE_DEFAULT: - ea_name = XATTR_NAME_ACL_DEFAULT; + ea_name = POSIX_ACL_XATTR_DEFAULT; p_acl = &ji->i_default_acl; if (!S_ISDIR(inode->i_mode)) return acl ? -EACCES : 0; @@ -101,7 +102,7 @@ static int jfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) return -EINVAL; } if (acl) { - size = xattr_acl_size(acl->a_count); + size = posix_acl_xattr_size(acl->a_count); value = kmalloc(size, GFP_KERNEL); if (!value) return -ENOMEM; diff --git a/fs/jfs/jfs_acl.h b/fs/jfs/jfs_acl.h index d2ae430adecf..a3acd3eec059 100644 --- a/fs/jfs/jfs_acl.h +++ b/fs/jfs/jfs_acl.h @@ -20,8 +20,6 @@ #ifdef CONFIG_JFS_POSIX_ACL -#include - int jfs_permission(struct inode *, int, struct nameidata *); int jfs_init_acl(struct inode *, struct inode *); int jfs_setattr(struct dentry *, struct iattr *); diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 810a3653d8b3..ee32211288ce 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include "jfs_incore.h" diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c index 6016373701a3..ee438d429d45 100644 --- a/fs/jfs/xattr.c +++ b/fs/jfs/xattr.c @@ -19,6 +19,7 @@ #include #include +#include #include #include "jfs_incore.h" #include "jfs_superblock.h" @@ -718,9 +719,9 @@ static int can_set_system_xattr(struct inode *inode, const char *name, return -EPERM; /* - * XATTR_NAME_ACL_ACCESS is tied to i_mode + * POSIX_ACL_XATTR_ACCESS is tied to i_mode */ - if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0) { + if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0) { acl = posix_acl_from_xattr(value, value_len); if (IS_ERR(acl)) { rc = PTR_ERR(acl); @@ -750,7 +751,7 @@ static int can_set_system_xattr(struct inode *inode, const char *name, JFS_IP(inode)->i_acl = JFS_ACL_NOT_CACHED; return 0; - } else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0) { + } else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) { acl = posix_acl_from_xattr(value, value_len); if (IS_ERR(acl)) { rc = PTR_ERR(acl); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index ae3940dc85cc..de340ffd33c3 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -50,7 +50,6 @@ #include #ifdef CONFIG_NFSD_V4 #include -#include #include #include #include @@ -425,13 +424,13 @@ nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out_nfserr; if (pacl) { - error = set_nfsv4_acl_one(dentry, pacl, XATTR_NAME_ACL_ACCESS); + error = set_nfsv4_acl_one(dentry, pacl, POSIX_ACL_XATTR_ACCESS); if (error < 0) goto out_nfserr; } if (dpacl) { - error = set_nfsv4_acl_one(dentry, dpacl, XATTR_NAME_ACL_DEFAULT); + error = set_nfsv4_acl_one(dentry, dpacl, POSIX_ACL_XATTR_DEFAULT); if (error < 0) goto out_nfserr; } @@ -498,7 +497,7 @@ nfsd4_get_nfs4_acl(struct svc_rqst *rqstp, struct dentry *dentry, struct nfs4_ac struct posix_acl *pacl = NULL, *dpacl = NULL; unsigned int flags = 0; - pacl = _get_posix_acl(dentry, XATTR_NAME_ACL_ACCESS); + pacl = _get_posix_acl(dentry, POSIX_ACL_XATTR_ACCESS); if (IS_ERR(pacl) && PTR_ERR(pacl) == -ENODATA) pacl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL); if (IS_ERR(pacl)) { @@ -508,7 +507,7 @@ nfsd4_get_nfs4_acl(struct svc_rqst *rqstp, struct dentry *dentry, struct nfs4_ac } if (S_ISDIR(inode->i_mode)) { - dpacl = _get_posix_acl(dentry, XATTR_NAME_ACL_DEFAULT); + dpacl = _get_posix_acl(dentry, POSIX_ACL_XATTR_DEFAULT); if (IS_ERR(dpacl) && PTR_ERR(dpacl) == -ENODATA) dpacl = NULL; else if (IS_ERR(dpacl)) { diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c index e302071903a1..c312881c5f53 100644 --- a/fs/reiserfs/xattr_acl.c +++ b/fs/reiserfs/xattr_acl.c @@ -4,7 +4,7 @@ #include #include #include -#include +#include #include #include #include @@ -192,11 +192,11 @@ reiserfs_get_acl(struct inode *inode, int type) switch (type) { case ACL_TYPE_ACCESS: - name = XATTR_NAME_ACL_ACCESS; + name = POSIX_ACL_XATTR_ACCESS; p_acl = &reiserfs_i->i_acl_access; break; case ACL_TYPE_DEFAULT: - name = XATTR_NAME_ACL_DEFAULT; + name = POSIX_ACL_XATTR_DEFAULT; p_acl = &reiserfs_i->i_acl_default; break; default: @@ -260,7 +260,7 @@ reiserfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) switch (type) { case ACL_TYPE_ACCESS: - name = XATTR_NAME_ACL_ACCESS; + name = POSIX_ACL_XATTR_ACCESS; p_acl = &reiserfs_i->i_acl_access; if (acl) { mode_t mode = inode->i_mode; @@ -275,7 +275,7 @@ reiserfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) } break; case ACL_TYPE_DEFAULT: - name = XATTR_NAME_ACL_DEFAULT; + name = POSIX_ACL_XATTR_DEFAULT; p_acl = &reiserfs_i->i_acl_default; if (!S_ISDIR (inode->i_mode)) return acl ? -EACCES : 0; @@ -468,7 +468,7 @@ static int posix_acl_access_get(struct inode *inode, const char *name, void *buffer, size_t size) { - if (strlen(name) != sizeof(XATTR_NAME_ACL_ACCESS)-1) + if (strlen(name) != sizeof(POSIX_ACL_XATTR_ACCESS)-1) return -EINVAL; return xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size); } @@ -477,7 +477,7 @@ static int posix_acl_access_set(struct inode *inode, const char *name, const void *value, size_t size, int flags) { - if (strlen(name) != sizeof(XATTR_NAME_ACL_ACCESS)-1) + if (strlen(name) != sizeof(POSIX_ACL_XATTR_ACCESS)-1) return -EINVAL; return xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size); } @@ -487,7 +487,7 @@ posix_acl_access_del (struct inode *inode, const char *name) { struct reiserfs_inode_info *reiserfs_i = REISERFS_I(inode); struct posix_acl **acl = &reiserfs_i->i_acl_access; - if (strlen(name) != sizeof(XATTR_NAME_ACL_ACCESS)-1) + if (strlen(name) != sizeof(POSIX_ACL_XATTR_ACCESS)-1) return -EINVAL; if (!IS_ERR (*acl) && *acl) { posix_acl_release (*acl); @@ -510,7 +510,7 @@ posix_acl_access_list (struct inode *inode, const char *name, int namelen, char } struct reiserfs_xattr_handler posix_acl_access_handler = { - .prefix = XATTR_NAME_ACL_ACCESS, + .prefix = POSIX_ACL_XATTR_ACCESS, .get = posix_acl_access_get, .set = posix_acl_access_set, .del = posix_acl_access_del, @@ -521,7 +521,7 @@ static int posix_acl_default_get (struct inode *inode, const char *name, void *buffer, size_t size) { - if (strlen(name) != sizeof(XATTR_NAME_ACL_DEFAULT)-1) + if (strlen(name) != sizeof(POSIX_ACL_XATTR_DEFAULT)-1) return -EINVAL; return xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size); } @@ -530,7 +530,7 @@ static int posix_acl_default_set(struct inode *inode, const char *name, const void *value, size_t size, int flags) { - if (strlen(name) != sizeof(XATTR_NAME_ACL_DEFAULT)-1) + if (strlen(name) != sizeof(POSIX_ACL_XATTR_DEFAULT)-1) return -EINVAL; return xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size); } @@ -540,7 +540,7 @@ posix_acl_default_del (struct inode *inode, const char *name) { struct reiserfs_inode_info *reiserfs_i = REISERFS_I(inode); struct posix_acl **acl = &reiserfs_i->i_acl_default; - if (strlen(name) != sizeof(XATTR_NAME_ACL_DEFAULT)-1) + if (strlen(name) != sizeof(POSIX_ACL_XATTR_DEFAULT)-1) return -EINVAL; if (!IS_ERR (*acl) && *acl) { posix_acl_release (*acl); @@ -563,7 +563,7 @@ posix_acl_default_list (struct inode *inode, const char *name, int namelen, char } struct reiserfs_xattr_handler posix_acl_default_handler = { - .prefix = XATTR_NAME_ACL_DEFAULT, + .prefix = POSIX_ACL_XATTR_DEFAULT, .get = posix_acl_default_get, .set = posix_acl_default_set, .del = posix_acl_default_del, diff --git a/include/linux/posix_acl_xattr.h b/include/linux/posix_acl_xattr.h index fe271c1947b2..6e53c34035cd 100644 --- a/include/linux/posix_acl_xattr.h +++ b/include/linux/posix_acl_xattr.h @@ -52,4 +52,7 @@ posix_acl_xattr_count(size_t size) return size / sizeof(posix_acl_xattr_entry); } +struct posix_acl *posix_acl_from_xattr(const void *value, size_t size); +int posix_acl_to_xattr(const struct posix_acl *acl, void *buffer, size_t size); + #endif /* _POSIX_ACL_XATTR_H */ diff --git a/include/linux/reiserfs_acl.h b/include/linux/reiserfs_acl.h index 2aef9c3f5ce8..0760507a545b 100644 --- a/include/linux/reiserfs_acl.h +++ b/include/linux/reiserfs_acl.h @@ -1,6 +1,5 @@ #include #include -#include #define REISERFS_ACL_VERSION 0x0001 -- cgit v1.2.3 From c43dc2fd885b5658cfd7cedb7bcca20910c517a4 Mon Sep 17 00:00:00 2001 From: Benjamin LaHaise Date: Thu, 23 Jun 2005 00:10:27 -0700 Subject: [PATCH] aio: make wait_queue ->task ->private In the upcoming aio_down patch, it is useful to store a private data pointer in the kiocb's wait_queue. Since we provide our own wake up function and do not require the task_struct pointer, it makes sense to convert the task pointer into a generic private pointer. Signed-off-by: Benjamin LaHaise Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/wait.h | 16 ++++++++-------- kernel/sched.c | 2 +- 2 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/wait.h b/include/linux/wait.h index c9486c3efb4a..d38c9fecdc36 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -33,7 +33,7 @@ int default_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key struct __wait_queue { unsigned int flags; #define WQ_FLAG_EXCLUSIVE 0x01 - struct task_struct * task; + void *private; wait_queue_func_t func; struct list_head task_list; }; @@ -60,7 +60,7 @@ typedef struct __wait_queue_head wait_queue_head_t; */ #define __WAITQUEUE_INITIALIZER(name, tsk) { \ - .task = tsk, \ + .private = tsk, \ .func = default_wake_function, \ .task_list = { NULL, NULL } } @@ -86,7 +86,7 @@ static inline void init_waitqueue_head(wait_queue_head_t *q) static inline void init_waitqueue_entry(wait_queue_t *q, struct task_struct *p) { q->flags = 0; - q->task = p; + q->private = p; q->func = default_wake_function; } @@ -94,7 +94,7 @@ static inline void init_waitqueue_func_entry(wait_queue_t *q, wait_queue_func_t func) { q->flags = 0; - q->task = NULL; + q->private = NULL; q->func = func; } @@ -110,7 +110,7 @@ static inline int waitqueue_active(wait_queue_head_t *q) * aio specifies a wait queue entry with an async notification * callback routine, not associated with any task. */ -#define is_sync_wait(wait) (!(wait) || ((wait)->task)) +#define is_sync_wait(wait) (!(wait) || ((wait)->private)) extern void FASTCALL(add_wait_queue(wait_queue_head_t *q, wait_queue_t * wait)); extern void FASTCALL(add_wait_queue_exclusive(wait_queue_head_t *q, wait_queue_t * wait)); @@ -384,7 +384,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key); #define DEFINE_WAIT(name) \ wait_queue_t name = { \ - .task = current, \ + .private = current, \ .func = autoremove_wake_function, \ .task_list = LIST_HEAD_INIT((name).task_list), \ } @@ -393,7 +393,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key); struct wait_bit_queue name = { \ .key = __WAIT_BIT_KEY_INITIALIZER(word, bit), \ .wait = { \ - .task = current, \ + .private = current, \ .func = wake_bit_function, \ .task_list = \ LIST_HEAD_INIT((name).wait.task_list), \ @@ -402,7 +402,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key); #define init_wait(wait) \ do { \ - (wait)->task = current; \ + (wait)->private = current; \ (wait)->func = autoremove_wake_function; \ INIT_LIST_HEAD(&(wait)->task_list); \ } while (0) diff --git a/kernel/sched.c b/kernel/sched.c index 6ee4515d5a20..76080d142e3d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2869,7 +2869,7 @@ need_resched: int default_wake_function(wait_queue_t *curr, unsigned mode, int sync, void *key) { - task_t *p = curr->task; + task_t *p = curr->private; return try_to_wake_up(p, mode, sync); } -- cgit v1.2.3 From bfb07599da289881d3bcbb601a110e997fc7444b Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 23 Jun 2005 00:10:32 -0700 Subject: [PATCH] Introduce tty_unregister_ldisc() It's a bit strange to see tty_register_ldisc call in modules' exit functions. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/tty.txt | 2 +- drivers/char/tty_io.c | 37 ++++++++++++++++++++++++------------- include/linux/tty.h | 1 + 3 files changed, 26 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/Documentation/tty.txt b/Documentation/tty.txt index 3958cf746dde..8ff7bc2a0811 100644 --- a/Documentation/tty.txt +++ b/Documentation/tty.txt @@ -22,7 +22,7 @@ copy of the structure. You must not re-register over the top of the line discipline even with the same data or your computer again will be eaten by demons. -In order to remove a line discipline call tty_register_ldisc passing NULL. +In order to remove a line discipline call tty_unregister_ldisc(). In ancient times this always worked. In modern times the function will return -EBUSY if the ldisc is currently in use. Since the ldisc referencing code manages the module counts this should not usually be a concern. diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 31831030f73f..cc4b43bad703 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -251,7 +251,7 @@ static void tty_set_termios_ldisc(struct tty_struct *tty, int num) static DEFINE_SPINLOCK(tty_ldisc_lock); static DECLARE_WAIT_QUEUE_HEAD(tty_ldisc_wait); -static struct tty_ldisc tty_ldiscs[NR_LDISCS]; /* line disc dispatch table */ +static struct tty_ldisc tty_ldiscs[NR_LDISCS]; /* line disc dispatch table */ int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc) { @@ -262,24 +262,35 @@ int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc) return -EINVAL; spin_lock_irqsave(&tty_ldisc_lock, flags); - if (new_ldisc) { - tty_ldiscs[disc] = *new_ldisc; - tty_ldiscs[disc].num = disc; - tty_ldiscs[disc].flags |= LDISC_FLAG_DEFINED; - tty_ldiscs[disc].refcount = 0; - } else { - if(tty_ldiscs[disc].refcount) - ret = -EBUSY; - else - tty_ldiscs[disc].flags &= ~LDISC_FLAG_DEFINED; - } + tty_ldiscs[disc] = *new_ldisc; + tty_ldiscs[disc].num = disc; + tty_ldiscs[disc].flags |= LDISC_FLAG_DEFINED; + tty_ldiscs[disc].refcount = 0; spin_unlock_irqrestore(&tty_ldisc_lock, flags); return ret; } - EXPORT_SYMBOL(tty_register_ldisc); +int tty_unregister_ldisc(int disc) +{ + unsigned long flags; + int ret = 0; + + if (disc < N_TTY || disc >= NR_LDISCS) + return -EINVAL; + + spin_lock_irqsave(&tty_ldisc_lock, flags); + if (tty_ldiscs[disc].refcount) + ret = -EBUSY; + else + tty_ldiscs[disc].flags &= ~LDISC_FLAG_DEFINED; + spin_unlock_irqrestore(&tty_ldisc_lock, flags); + + return ret; +} +EXPORT_SYMBOL(tty_unregister_ldisc); + struct tty_ldisc *tty_ldisc_get(int disc) { unsigned long flags; diff --git a/include/linux/tty.h b/include/linux/tty.h index 1b76106272d3..59ff42c629ec 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -345,6 +345,7 @@ extern int tty_check_change(struct tty_struct * tty); extern void stop_tty(struct tty_struct * tty); extern void start_tty(struct tty_struct * tty); extern int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc); +extern int tty_unregister_ldisc(int disc); extern int tty_register_driver(struct tty_driver *driver); extern int tty_unregister_driver(struct tty_driver *driver); extern void tty_register_device(struct tty_driver *driver, unsigned index, struct device *dev); -- cgit v1.2.3 From 4749f32da939d4e4160541b2cadc22492bb507ec Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 23 Jun 2005 11:36:56 +0200 Subject: [PATCH] better USB_MON dependencies This makes the USB_MON less confusing. Signed-off-by: Adrian Bunk Signed-off-by: Linus Torvalds --- drivers/usb/core/hcd.c | 2 +- drivers/usb/core/hcd.h | 2 +- drivers/usb/mon/Kconfig | 13 ++++--------- drivers/usb/mon/Makefile | 2 +- include/linux/usb.h | 2 +- 5 files changed, 8 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c index d041782e0c8b..0da23732e807 100644 --- a/drivers/usb/core/hcd.c +++ b/drivers/usb/core/hcd.c @@ -1794,7 +1794,7 @@ EXPORT_SYMBOL (usb_remove_hcd); /*-------------------------------------------------------------------------*/ -#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE) +#if defined(CONFIG_USB_MON) struct usb_mon_operations *mon_ops; diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h index f67cf1e634fc..325a51656c3f 100644 --- a/drivers/usb/core/hcd.h +++ b/drivers/usb/core/hcd.h @@ -399,7 +399,7 @@ static inline void usbfs_cleanup(void) { } /*-------------------------------------------------------------------------*/ -#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE) +#if defined(CONFIG_USB_MON) struct usb_mon_operations { void (*urb_submit)(struct usb_bus *bus, struct urb *urb); diff --git a/drivers/usb/mon/Kconfig b/drivers/usb/mon/Kconfig index 4e6152aa5f19..777642e26b9a 100644 --- a/drivers/usb/mon/Kconfig +++ b/drivers/usb/mon/Kconfig @@ -2,13 +2,9 @@ # USB Monitor configuration # -# In normal life, it makes little sense to have usbmon as a module, and in fact -# it is harmful, because there is no way to autoload the module. -# The 'm' option is allowed for hackers who debug the usbmon itself, -# and for those who have usbcore as a module. config USB_MON - tristate "USB Monitor" - depends on USB + bool "USB Monitor" + depends on USB!=n default y help If you say Y here, a component which captures the USB traffic @@ -17,6 +13,5 @@ config USB_MON Harding's USBMon. This is somewhat experimental at this time, but it should be safe, - as long as you aren't building this as a module and then removing it. - - If unsure, say Y. Do not say M. + as long as you aren't using modular USB and try to remove this + module. diff --git a/drivers/usb/mon/Makefile b/drivers/usb/mon/Makefile index 3cff8d444bb1..f18d10ce91f9 100644 --- a/drivers/usb/mon/Makefile +++ b/drivers/usb/mon/Makefile @@ -4,4 +4,4 @@ usbmon-objs := mon_main.o mon_stat.o mon_text.o -obj-$(CONFIG_USB_MON) += usbmon.o +obj-$(CONFIG_USB) += usbmon.o diff --git a/include/linux/usb.h b/include/linux/usb.h index 3d508bf08402..eb282b581546 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -290,7 +290,7 @@ struct usb_bus { struct class_device *class_dev; /* class device for this bus */ struct kref kref; /* handles reference counting this bus */ void (*release)(struct usb_bus *bus); /* function to destroy this bus's memory */ -#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE) +#if defined(CONFIG_USB_MON) struct mon_bus *mon_bus; /* non-null when associated */ int monitored; /* non-zero when monitored */ #endif -- cgit v1.2.3 From 317a76f9a44b437d6301718f4e5d08bd93f98da7 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 12:19:55 -0700 Subject: [TCP]: Add pluggable congestion control algorithm infrastructure. Allow TCP to have multiple pluggable congestion control algorithms. Algorithms are defined by a set of operations and can be built in or modules. The legacy "new RENO" algorithm is used as a starting point and fallback. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/sysctl.h | 9 +- include/linux/tcp.h | 49 +-- include/net/tcp.h | 237 +++++---------- net/ipv4/Makefile | 3 +- net/ipv4/sysctl_net_ipv4.c | 114 +++---- net/ipv4/tcp.c | 2 + net/ipv4/tcp_cong.c | 195 ++++++++++++ net/ipv4/tcp_diag.c | 20 +- net/ipv4/tcp_input.c | 737 ++++----------------------------------------- net/ipv4/tcp_ipv4.c | 3 + net/ipv4/tcp_minisocks.c | 4 +- net/ipv4/tcp_output.c | 23 +- net/ipv6/tcp_ipv6.c | 2 +- 13 files changed, 399 insertions(+), 999 deletions(-) create mode 100644 net/ipv4/tcp_cong.c (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 614e939c78a4..72965bfe6cfb 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -333,21 +333,14 @@ enum NET_TCP_FRTO=92, NET_TCP_LOW_LATENCY=93, NET_IPV4_IPFRAG_SECRET_INTERVAL=94, - NET_TCP_WESTWOOD=95, NET_IPV4_IGMP_MAX_MSF=96, NET_TCP_NO_METRICS_SAVE=97, - NET_TCP_VEGAS=98, - NET_TCP_VEGAS_ALPHA=99, - NET_TCP_VEGAS_BETA=100, - NET_TCP_VEGAS_GAMMA=101, - NET_TCP_BIC=102, - NET_TCP_BIC_FAST_CONVERGENCE=103, - NET_TCP_BIC_LOW_WINDOW=104, NET_TCP_DEFAULT_WIN_SCALE=105, NET_TCP_MODERATE_RCVBUF=106, NET_TCP_TSO_WIN_DIVISOR=107, NET_TCP_BIC_BETA=108, NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR=109, + NET_TCP_CONG_CONTROL=110, }; enum { diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 97a7c9e03df5..3ea75dd6640a 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -203,13 +203,6 @@ struct tcp_sack_block { __u32 end_seq; }; -enum tcp_congestion_algo { - TCP_RENO=0, - TCP_VEGAS, - TCP_WESTWOOD, - TCP_BIC, -}; - struct tcp_options_received { /* PAWS/RTTM data */ long ts_recent_stamp;/* Time we stored ts_recent (for aging) */ @@ -305,7 +298,7 @@ struct tcp_sock { __u8 reordering; /* Packet reordering metric. */ __u8 frto_counter; /* Number of new acks after RTO */ - __u8 adv_cong; /* Using Vegas, Westwood, or BIC */ + __u8 unused; __u8 defer_accept; /* User waits for some data after accept() */ /* RTT measurement */ @@ -401,37 +394,10 @@ struct tcp_sock { __u32 time; } rcvq_space; -/* TCP Westwood structure */ - struct { - __u32 bw_ns_est; /* first bandwidth estimation..not too smoothed 8) */ - __u32 bw_est; /* bandwidth estimate */ - __u32 rtt_win_sx; /* here starts a new evaluation... */ - __u32 bk; - __u32 snd_una; /* used for evaluating the number of acked bytes */ - __u32 cumul_ack; - __u32 accounted; - __u32 rtt; - __u32 rtt_min; /* minimum observed RTT */ - } westwood; - -/* Vegas variables */ - struct { - __u32 beg_snd_nxt; /* right edge during last RTT */ - __u32 beg_snd_una; /* left edge during last RTT */ - __u32 beg_snd_cwnd; /* saves the size of the cwnd */ - __u8 doing_vegas_now;/* if true, do vegas for this RTT */ - __u16 cntRTT; /* # of RTTs measured within last RTT */ - __u32 minRTT; /* min of RTTs measured within last RTT (in usec) */ - __u32 baseRTT; /* the min of all Vegas RTT measurements seen (in usec) */ - } vegas; - - /* BI TCP Parameters */ - struct { - __u32 cnt; /* increase cwnd by 1 after this number of ACKs */ - __u32 last_max_cwnd; /* last maximium snd_cwnd */ - __u32 last_cwnd; /* the last snd_cwnd */ - __u32 last_stamp; /* time when updated last_cwnd */ - } bictcp; + /* Pluggable TCP congestion control hook */ + struct tcp_congestion_ops *ca_ops; + u32 ca_priv[16]; +#define TCP_CA_PRIV_SIZE (16*sizeof(u32)) }; static inline struct tcp_sock *tcp_sk(const struct sock *sk) @@ -439,6 +405,11 @@ static inline struct tcp_sock *tcp_sk(const struct sock *sk) return (struct tcp_sock *)sk; } +static inline void *tcp_ca(const struct tcp_sock *tp) +{ + return (void *) tp->ca_priv; +} + #endif #endif /* _LINUX_TCP_H */ diff --git a/include/net/tcp.h b/include/net/tcp.h index f730935b824a..e427cf35915c 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -505,25 +505,6 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) #else # define TCP_TW_RECYCLE_TICK (12+2-TCP_TW_RECYCLE_SLOTS_LOG) #endif - -#define BICTCP_BETA_SCALE 1024 /* Scale factor beta calculation - * max_cwnd = snd_cwnd * beta - */ -#define BICTCP_MAX_INCREMENT 32 /* - * Limit on the amount of - * increment allowed during - * binary search. - */ -#define BICTCP_FUNC_OF_MIN_INCR 11 /* - * log(B/Smin)/log(B/(B-1))+1, - * Smin:min increment - * B:log factor - */ -#define BICTCP_B 4 /* - * In binary search, - * go to point (max+min)/N - */ - /* * TCP option */ @@ -596,16 +577,7 @@ extern int sysctl_tcp_adv_win_scale; extern int sysctl_tcp_tw_reuse; extern int sysctl_tcp_frto; extern int sysctl_tcp_low_latency; -extern int sysctl_tcp_westwood; -extern int sysctl_tcp_vegas_cong_avoid; -extern int sysctl_tcp_vegas_alpha; -extern int sysctl_tcp_vegas_beta; -extern int sysctl_tcp_vegas_gamma; extern int sysctl_tcp_nometrics_save; -extern int sysctl_tcp_bic; -extern int sysctl_tcp_bic_fast_convergence; -extern int sysctl_tcp_bic_low_window; -extern int sysctl_tcp_bic_beta; extern int sysctl_tcp_moderate_rcvbuf; extern int sysctl_tcp_tso_win_divisor; @@ -1136,6 +1108,80 @@ static inline void tcp_packets_out_dec(struct tcp_sock *tp, tp->packets_out -= tcp_skb_pcount(skb); } +/* Events passed to congestion control interface */ +enum tcp_ca_event { + CA_EVENT_TX_START, /* first transmit when no packets in flight */ + CA_EVENT_CWND_RESTART, /* congestion window restart */ + CA_EVENT_COMPLETE_CWR, /* end of congestion recovery */ + CA_EVENT_FRTO, /* fast recovery timeout */ + CA_EVENT_LOSS, /* loss timeout */ + CA_EVENT_FAST_ACK, /* in sequence ack */ + CA_EVENT_SLOW_ACK, /* other ack */ +}; + +/* + * Interface for adding new TCP congestion control handlers + */ +#define TCP_CA_NAME_MAX 16 +struct tcp_congestion_ops { + struct list_head list; + + /* initialize private data (optional) */ + void (*init)(struct tcp_sock *tp); + /* cleanup private data (optional) */ + void (*release)(struct tcp_sock *tp); + + /* return slow start threshold (required) */ + u32 (*ssthresh)(struct tcp_sock *tp); + /* lower bound for congestion window (optional) */ + u32 (*min_cwnd)(struct tcp_sock *tp); + /* do new cwnd calculation (required) */ + void (*cong_avoid)(struct tcp_sock *tp, u32 ack, + u32 rtt, u32 in_flight, int good_ack); + /* round trip time sample per acked packet (optional) */ + void (*rtt_sample)(struct tcp_sock *tp, u32 usrtt); + /* call before changing ca_state (optional) */ + void (*set_state)(struct tcp_sock *tp, u8 new_state); + /* call when cwnd event occurs (optional) */ + void (*cwnd_event)(struct tcp_sock *tp, enum tcp_ca_event ev); + /* new value of cwnd after loss (optional) */ + u32 (*undo_cwnd)(struct tcp_sock *tp); + /* hook for packet ack accounting (optional) */ + void (*pkts_acked)(struct tcp_sock *tp, u32 num_acked); + /* get info for tcp_diag (optional) */ + void (*get_info)(struct tcp_sock *tp, u32 ext, struct sk_buff *skb); + + char name[TCP_CA_NAME_MAX]; + struct module *owner; +}; + +extern int tcp_register_congestion_control(struct tcp_congestion_ops *type); +extern void tcp_unregister_congestion_control(struct tcp_congestion_ops *type); + +extern void tcp_init_congestion_control(struct tcp_sock *tp); +extern void tcp_cleanup_congestion_control(struct tcp_sock *tp); +extern int tcp_set_default_congestion_control(const char *name); +extern void tcp_get_default_congestion_control(char *name); + +extern struct tcp_congestion_ops tcp_reno; +extern u32 tcp_reno_ssthresh(struct tcp_sock *tp); +extern void tcp_reno_cong_avoid(struct tcp_sock *tp, u32 ack, + u32 rtt, u32 in_flight, int flag); +extern u32 tcp_reno_min_cwnd(struct tcp_sock *tp); + +static inline void tcp_set_ca_state(struct tcp_sock *tp, u8 ca_state) +{ + if (tp->ca_ops->set_state) + tp->ca_ops->set_state(tp, ca_state); + tp->ca_state = ca_state; +} + +static inline void tcp_ca_event(struct tcp_sock *tp, enum tcp_ca_event event) +{ + if (tp->ca_ops->cwnd_event) + tp->ca_ops->cwnd_event(tp, event); +} + /* This determines how many packets are "in the network" to the best * of our knowledge. In many cases it is conservative, but where * detailed information is available from the receiver (via SACK @@ -1155,91 +1201,6 @@ static __inline__ unsigned int tcp_packets_in_flight(const struct tcp_sock *tp) return (tp->packets_out - tp->left_out + tp->retrans_out); } -/* - * Which congestion algorithim is in use on the connection. - */ -#define tcp_is_vegas(__tp) ((__tp)->adv_cong == TCP_VEGAS) -#define tcp_is_westwood(__tp) ((__tp)->adv_cong == TCP_WESTWOOD) -#define tcp_is_bic(__tp) ((__tp)->adv_cong == TCP_BIC) - -/* Recalculate snd_ssthresh, we want to set it to: - * - * Reno: - * one half the current congestion window, but no - * less than two segments - * - * BIC: - * behave like Reno until low_window is reached, - * then increase congestion window slowly - */ -static inline __u32 tcp_recalc_ssthresh(struct tcp_sock *tp) -{ - if (tcp_is_bic(tp)) { - if (sysctl_tcp_bic_fast_convergence && - tp->snd_cwnd < tp->bictcp.last_max_cwnd) - tp->bictcp.last_max_cwnd = (tp->snd_cwnd * - (BICTCP_BETA_SCALE - + sysctl_tcp_bic_beta)) - / (2 * BICTCP_BETA_SCALE); - else - tp->bictcp.last_max_cwnd = tp->snd_cwnd; - - if (tp->snd_cwnd > sysctl_tcp_bic_low_window) - return max((tp->snd_cwnd * sysctl_tcp_bic_beta) - / BICTCP_BETA_SCALE, 2U); - } - - return max(tp->snd_cwnd >> 1U, 2U); -} - -/* Stop taking Vegas samples for now. */ -#define tcp_vegas_disable(__tp) ((__tp)->vegas.doing_vegas_now = 0) - -static inline void tcp_vegas_enable(struct tcp_sock *tp) -{ - /* There are several situations when we must "re-start" Vegas: - * - * o when a connection is established - * o after an RTO - * o after fast recovery - * o when we send a packet and there is no outstanding - * unacknowledged data (restarting an idle connection) - * - * In these circumstances we cannot do a Vegas calculation at the - * end of the first RTT, because any calculation we do is using - * stale info -- both the saved cwnd and congestion feedback are - * stale. - * - * Instead we must wait until the completion of an RTT during - * which we actually receive ACKs. - */ - - /* Begin taking Vegas samples next time we send something. */ - tp->vegas.doing_vegas_now = 1; - - /* Set the beginning of the next send window. */ - tp->vegas.beg_snd_nxt = tp->snd_nxt; - - tp->vegas.cntRTT = 0; - tp->vegas.minRTT = 0x7fffffff; -} - -/* Should we be taking Vegas samples right now? */ -#define tcp_vegas_enabled(__tp) ((__tp)->vegas.doing_vegas_now) - -extern void tcp_ca_init(struct tcp_sock *tp); - -static inline void tcp_set_ca_state(struct tcp_sock *tp, u8 ca_state) -{ - if (tcp_is_vegas(tp)) { - if (ca_state == TCP_CA_Open) - tcp_vegas_enable(tp); - else - tcp_vegas_disable(tp); - } - tp->ca_state = ca_state; -} - /* If cwnd > ssthresh, we may raise ssthresh to be half-way to cwnd. * The exception is rate halving phase, when cwnd is decreasing towards * ssthresh. @@ -1288,7 +1249,7 @@ static inline void tcp_cwnd_validate(struct sock *sk, struct tcp_sock *tp) static inline void __tcp_enter_cwr(struct tcp_sock *tp) { tp->undo_marker = 0; - tp->snd_ssthresh = tcp_recalc_ssthresh(tp); + tp->snd_ssthresh = tp->ca_ops->ssthresh(tp); tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp) + 1U); tp->snd_cwnd_cnt = 0; @@ -1876,52 +1837,4 @@ struct tcp_iter_state { extern int tcp_proc_register(struct tcp_seq_afinfo *afinfo); extern void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo); -/* TCP Westwood functions and constants */ - -#define TCP_WESTWOOD_INIT_RTT (20*HZ) /* maybe too conservative?! */ -#define TCP_WESTWOOD_RTT_MIN (HZ/20) /* 50ms */ - -static inline void tcp_westwood_update_rtt(struct tcp_sock *tp, __u32 rtt_seq) -{ - if (tcp_is_westwood(tp)) - tp->westwood.rtt = rtt_seq; -} - -static inline __u32 __tcp_westwood_bw_rttmin(const struct tcp_sock *tp) -{ - return max((tp->westwood.bw_est) * (tp->westwood.rtt_min) / - (__u32) (tp->mss_cache_std), - 2U); -} - -static inline __u32 tcp_westwood_bw_rttmin(const struct tcp_sock *tp) -{ - return tcp_is_westwood(tp) ? __tcp_westwood_bw_rttmin(tp) : 0; -} - -static inline int tcp_westwood_ssthresh(struct tcp_sock *tp) -{ - __u32 ssthresh = 0; - - if (tcp_is_westwood(tp)) { - ssthresh = __tcp_westwood_bw_rttmin(tp); - if (ssthresh) - tp->snd_ssthresh = ssthresh; - } - - return (ssthresh != 0); -} - -static inline int tcp_westwood_cwnd(struct tcp_sock *tp) -{ - __u32 cwnd = 0; - - if (tcp_is_westwood(tp)) { - cwnd = __tcp_westwood_bw_rttmin(tp); - if (cwnd) - tp->snd_cwnd = cwnd; - } - - return (cwnd != 0); -} #endif /* _TCP_H */ diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 65d57d8e1add..89c0b4cb470e 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -5,7 +5,8 @@ obj-y := utils.o route.o inetpeer.o protocol.o \ ip_input.o ip_fragment.o ip_forward.o ip_options.o \ ip_output.o ip_sockglue.o \ - tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o tcp_minisocks.o \ + tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \ + tcp_minisocks.o tcp_cong.o \ datagram.o raw.o udp.o arp.o icmp.o devinet.o af_inet.o igmp.o \ sysctl_net_ipv4.o fib_frontend.o fib_semantics.o diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 23068bddbf0b..e32894532416 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -118,6 +118,45 @@ static int ipv4_sysctl_forward_strategy(ctl_table *table, return 1; } +static int proc_tcp_congestion_control(ctl_table *ctl, int write, struct file * filp, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + char val[TCP_CA_NAME_MAX]; + ctl_table tbl = { + .data = val, + .maxlen = TCP_CA_NAME_MAX, + }; + int ret; + + tcp_get_default_congestion_control(val); + + ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos); + if (write && ret == 0) + ret = tcp_set_default_congestion_control(val); + return ret; +} + +int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, int nlen, + void __user *oldval, size_t __user *oldlenp, + void __user *newval, size_t newlen, + void **context) +{ + char val[TCP_CA_NAME_MAX]; + ctl_table tbl = { + .data = val, + .maxlen = TCP_CA_NAME_MAX, + }; + int ret; + + tcp_get_default_congestion_control(val); + ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen, + context); + if (ret == 0 && newval && newlen) + ret = tcp_set_default_congestion_control(val); + return ret; +} + + ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_TCP_TIMESTAMPS, @@ -611,70 +650,6 @@ ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, - { - .ctl_name = NET_TCP_WESTWOOD, - .procname = "tcp_westwood", - .data = &sysctl_tcp_westwood, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_VEGAS, - .procname = "tcp_vegas_cong_avoid", - .data = &sysctl_tcp_vegas_cong_avoid, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_VEGAS_ALPHA, - .procname = "tcp_vegas_alpha", - .data = &sysctl_tcp_vegas_alpha, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_VEGAS_BETA, - .procname = "tcp_vegas_beta", - .data = &sysctl_tcp_vegas_beta, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_VEGAS_GAMMA, - .procname = "tcp_vegas_gamma", - .data = &sysctl_tcp_vegas_gamma, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_BIC, - .procname = "tcp_bic", - .data = &sysctl_tcp_bic, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_BIC_FAST_CONVERGENCE, - .procname = "tcp_bic_fast_convergence", - .data = &sysctl_tcp_bic_fast_convergence, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_BIC_LOW_WINDOW, - .procname = "tcp_bic_low_window", - .data = &sysctl_tcp_bic_low_window, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, { .ctl_name = NET_TCP_MODERATE_RCVBUF, .procname = "tcp_moderate_rcvbuf", @@ -692,13 +667,14 @@ ctl_table ipv4_table[] = { .proc_handler = &proc_dointvec, }, { - .ctl_name = NET_TCP_BIC_BETA, - .procname = "tcp_bic_beta", - .data = &sysctl_tcp_bic_beta, - .maxlen = sizeof(int), + .ctl_name = NET_TCP_CONG_CONTROL, + .procname = "tcp_congestion_control", .mode = 0644, - .proc_handler = &proc_dointvec, + .maxlen = TCP_CA_NAME_MAX, + .proc_handler = &proc_tcp_congestion_control, + .strategy = &sysctl_tcp_congestion_control, }, + { .ctl_name = 0 } }; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 674bbd8cfd36..f3dbc8dc1263 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2333,6 +2333,8 @@ void __init tcp_init(void) printk(KERN_INFO "TCP: Hash tables configured " "(established %d bind %d)\n", tcp_ehash_size << 1, tcp_bhash_size); + + tcp_register_congestion_control(&tcp_reno); } EXPORT_SYMBOL(tcp_accept); diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c new file mode 100644 index 000000000000..665394a63ae4 --- /dev/null +++ b/net/ipv4/tcp_cong.c @@ -0,0 +1,195 @@ +/* + * Plugable TCP congestion control support and newReno + * congestion control. + * Based on ideas from I/O scheduler suport and Web100. + * + * Copyright (C) 2005 Stephen Hemminger + */ + +#include +#include +#include +#include +#include +#include + +static DEFINE_SPINLOCK(tcp_cong_list_lock); +static LIST_HEAD(tcp_cong_list); + +/* Simple linear search, don't expect many entries! */ +static struct tcp_congestion_ops *tcp_ca_find(const char *name) +{ + struct tcp_congestion_ops *e; + + list_for_each_entry(e, &tcp_cong_list, list) { + if (strcmp(e->name, name) == 0) + return e; + } + + return NULL; +} + +/* + * Attach new congestion control algorthim to the list + * of available options. + */ +int tcp_register_congestion_control(struct tcp_congestion_ops *ca) +{ + int ret = 0; + + /* all algorithms must implement ssthresh and cong_avoid ops */ + if (!ca->ssthresh || !ca->cong_avoid || !ca->min_cwnd) { + printk(KERN_ERR "TCP %s does not implement required ops\n", + ca->name); + return -EINVAL; + } + + spin_lock(&tcp_cong_list_lock); + if (tcp_ca_find(ca->name)) { + printk(KERN_NOTICE "TCP %s already registered\n", ca->name); + ret = -EEXIST; + } else { + list_add_rcu(&ca->list, &tcp_cong_list); + printk(KERN_INFO "TCP %s registered\n", ca->name); + } + spin_unlock(&tcp_cong_list_lock); + + return ret; +} +EXPORT_SYMBOL_GPL(tcp_register_congestion_control); + +/* + * Remove congestion control algorithm, called from + * the module's remove function. Module ref counts are used + * to ensure that this can't be done till all sockets using + * that method are closed. + */ +void tcp_unregister_congestion_control(struct tcp_congestion_ops *ca) +{ + spin_lock(&tcp_cong_list_lock); + list_del_rcu(&ca->list); + spin_unlock(&tcp_cong_list_lock); +} +EXPORT_SYMBOL_GPL(tcp_unregister_congestion_control); + +/* Assign choice of congestion control. */ +void tcp_init_congestion_control(struct tcp_sock *tp) +{ + struct tcp_congestion_ops *ca; + + rcu_read_lock(); + list_for_each_entry_rcu(ca, &tcp_cong_list, list) { + if (try_module_get(ca->owner)) { + tp->ca_ops = ca; + break; + } + + } + rcu_read_unlock(); + + if (tp->ca_ops->init) + tp->ca_ops->init(tp); +} + +/* Manage refcounts on socket close. */ +void tcp_cleanup_congestion_control(struct tcp_sock *tp) +{ + if (tp->ca_ops->release) + tp->ca_ops->release(tp); + module_put(tp->ca_ops->owner); +} + +/* Used by sysctl to change default congestion control */ +int tcp_set_default_congestion_control(const char *name) +{ + struct tcp_congestion_ops *ca; + int ret = -ENOENT; + + spin_lock(&tcp_cong_list_lock); + ca = tcp_ca_find(name); +#ifdef CONFIG_KMOD + if (!ca) { + spin_unlock(&tcp_cong_list_lock); + + request_module("tcp_%s", name); + spin_lock(&tcp_cong_list_lock); + ca = tcp_ca_find(name); + } +#endif + + if (ca) { + list_move(&ca->list, &tcp_cong_list); + ret = 0; + } + spin_unlock(&tcp_cong_list_lock); + + return ret; +} + +/* Get current default congestion control */ +void tcp_get_default_congestion_control(char *name) +{ + struct tcp_congestion_ops *ca; + /* We will always have reno... */ + BUG_ON(list_empty(&tcp_cong_list)); + + rcu_read_lock(); + ca = list_entry(tcp_cong_list.next, struct tcp_congestion_ops, list); + strncpy(name, ca->name, TCP_CA_NAME_MAX); + rcu_read_unlock(); +} + +/* + * TCP Reno congestion control + * This is special case used for fallback as well. + */ +/* This is Jacobson's slow start and congestion avoidance. + * SIGCOMM '88, p. 328. + */ +void tcp_reno_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt, u32 in_flight, + int flag) +{ + if (in_flight < tp->snd_cwnd) + return; + + if (tp->snd_cwnd <= tp->snd_ssthresh) { + /* In "safe" area, increase. */ + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + } else { + /* In dangerous area, increase slowly. + * In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd + */ + if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + tp->snd_cwnd_cnt = 0; + } else + tp->snd_cwnd_cnt++; + } +} +EXPORT_SYMBOL_GPL(tcp_reno_cong_avoid); + +/* Slow start threshold is half the congestion window (min 2) */ +u32 tcp_reno_ssthresh(struct tcp_sock *tp) +{ + return max(tp->snd_cwnd >> 1U, 2U); +} +EXPORT_SYMBOL_GPL(tcp_reno_ssthresh); + +/* Lower bound on congestion window. */ +u32 tcp_reno_min_cwnd(struct tcp_sock *tp) +{ + return tp->snd_ssthresh/2; +} +EXPORT_SYMBOL_GPL(tcp_reno_min_cwnd); + +struct tcp_congestion_ops tcp_reno = { + .name = "reno", + .owner = THIS_MODULE, + .ssthresh = tcp_reno_ssthresh, + .cong_avoid = tcp_reno_cong_avoid, + .min_cwnd = tcp_reno_min_cwnd, +}; + +EXPORT_SYMBOL_GPL(tcp_reno); diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 634befc07921..867acc0f79d8 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -42,7 +42,6 @@ struct tcpdiag_entry static struct sock *tcpnl; - #define TCPDIAG_PUT(skb, attrtype, attrlen) \ ({ int rtalen = RTA_LENGTH(attrlen); \ struct rtattr *rta; \ @@ -61,7 +60,6 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk, struct nlmsghdr *nlh; struct tcp_info *info = NULL; struct tcpdiag_meminfo *minfo = NULL; - struct tcpvegas_info *vinfo = NULL; unsigned char *b = skb->tail; nlh = NLMSG_PUT(skb, pid, seq, TCPDIAG_GETSOCK, sizeof(*r)); @@ -73,9 +71,6 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk, if (ext & (1<<(TCPDIAG_INFO-1))) info = TCPDIAG_PUT(skb, TCPDIAG_INFO, sizeof(*info)); - if ((tcp_is_westwood(tp) || tcp_is_vegas(tp)) - && (ext & (1<<(TCPDIAG_VEGASINFO-1)))) - vinfo = TCPDIAG_PUT(skb, TCPDIAG_VEGASINFO, sizeof(*vinfo)); } r->tcpdiag_family = sk->sk_family; r->tcpdiag_state = sk->sk_state; @@ -166,19 +161,8 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk, if (info) tcp_get_info(sk, info); - if (vinfo) { - if (tcp_is_vegas(tp)) { - vinfo->tcpv_enabled = tp->vegas.doing_vegas_now; - vinfo->tcpv_rttcnt = tp->vegas.cntRTT; - vinfo->tcpv_rtt = jiffies_to_usecs(tp->vegas.baseRTT); - vinfo->tcpv_minrtt = jiffies_to_usecs(tp->vegas.minRTT); - } else { - vinfo->tcpv_enabled = 0; - vinfo->tcpv_rttcnt = 0; - vinfo->tcpv_rtt = jiffies_to_usecs(tp->westwood.rtt); - vinfo->tcpv_minrtt = jiffies_to_usecs(tp->westwood.rtt_min); - } - } + if (sk->sk_state < TCP_TIME_WAIT && tp->ca_ops->get_info) + tp->ca_ops->get_info(tp, ext, skb); nlh->nlmsg_len = skb->tail - b; return skb->len; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5bad504630a3..7bbbbc33eb4b 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -61,7 +61,6 @@ * Panu Kuhlberg: Experimental audit of TCP (re)transmission * engine. Lots of bugs are found. * Pasi Sarolahti: F-RTO for dealing with spurious RTOs - * Angelo Dell'Aera: TCP Westwood+ support */ #include @@ -88,23 +87,9 @@ int sysctl_tcp_rfc1337; int sysctl_tcp_max_orphans = NR_FILE; int sysctl_tcp_frto; int sysctl_tcp_nometrics_save; -int sysctl_tcp_westwood; -int sysctl_tcp_vegas_cong_avoid; int sysctl_tcp_moderate_rcvbuf = 1; -/* Default values of the Vegas variables, in fixed-point representation - * with V_PARAM_SHIFT bits to the right of the binary point. - */ -#define V_PARAM_SHIFT 1 -int sysctl_tcp_vegas_alpha = 1<snd_cwnd_stamp = tcp_time_stamp; } -static void init_bictcp(struct tcp_sock *tp) -{ - tp->bictcp.cnt = 0; - - tp->bictcp.last_max_cwnd = 0; - tp->bictcp.last_cwnd = 0; - tp->bictcp.last_stamp = 0; -} - /* 5. Recalculate window clamp after socket hit its memory bounds. */ static void tcp_clamp_window(struct sock *sk, struct tcp_sock *tp) { @@ -558,45 +534,6 @@ static void tcp_event_data_recv(struct sock *sk, struct tcp_sock *tp, struct sk_ tcp_grow_window(sk, tp, skb); } -/* When starting a new connection, pin down the current choice of - * congestion algorithm. - */ -void tcp_ca_init(struct tcp_sock *tp) -{ - if (sysctl_tcp_westwood) - tp->adv_cong = TCP_WESTWOOD; - else if (sysctl_tcp_bic) - tp->adv_cong = TCP_BIC; - else if (sysctl_tcp_vegas_cong_avoid) { - tp->adv_cong = TCP_VEGAS; - tp->vegas.baseRTT = 0x7fffffff; - tcp_vegas_enable(tp); - } -} - -/* Do RTT sampling needed for Vegas. - * Basically we: - * o min-filter RTT samples from within an RTT to get the current - * propagation delay + queuing delay (we are min-filtering to try to - * avoid the effects of delayed ACKs) - * o min-filter RTT samples from a much longer window (forever for now) - * to find the propagation delay (baseRTT) - */ -static inline void vegas_rtt_calc(struct tcp_sock *tp, __u32 rtt) -{ - __u32 vrtt = rtt + 1; /* Never allow zero rtt or baseRTT */ - - /* Filter to find propagation delay: */ - if (vrtt < tp->vegas.baseRTT) - tp->vegas.baseRTT = vrtt; - - /* Find the min RTT during the last RTT to find - * the current prop. delay + queuing delay: - */ - tp->vegas.minRTT = min(tp->vegas.minRTT, vrtt); - tp->vegas.cntRTT++; -} - /* Called to compute a smoothed rtt estimate. The data fed to this * routine either comes from timestamps, or from segments that were * known _not_ to have been retransmitted [see Karn/Partridge @@ -606,13 +543,10 @@ static inline void vegas_rtt_calc(struct tcp_sock *tp, __u32 rtt) * To save cycles in the RFC 1323 implementation it was better to break * it up into three procedures. -- erics */ -static void tcp_rtt_estimator(struct tcp_sock *tp, __u32 mrtt) +static void tcp_rtt_estimator(struct tcp_sock *tp, __u32 mrtt, u32 *usrtt) { long m = mrtt; /* RTT */ - if (tcp_vegas_enabled(tp)) - vegas_rtt_calc(tp, mrtt); - /* The following amusing code comes from Jacobson's * article in SIGCOMM '88. Note that rtt and mdev * are scaled versions of rtt and mean deviation. @@ -670,7 +604,8 @@ static void tcp_rtt_estimator(struct tcp_sock *tp, __u32 mrtt) tp->rtt_seq = tp->snd_nxt; } - tcp_westwood_update_rtt(tp, tp->srtt >> 3); + if (tp->ca_ops->rtt_sample) + tp->ca_ops->rtt_sample(tp, *usrtt); } /* Calculate rto without backoff. This is the second half of Van Jacobson's @@ -1185,8 +1120,8 @@ void tcp_enter_frto(struct sock *sk) tp->snd_una == tp->high_seq || (tp->ca_state == TCP_CA_Loss && !tp->retransmits)) { tp->prior_ssthresh = tcp_current_ssthresh(tp); - if (!tcp_westwood_ssthresh(tp)) - tp->snd_ssthresh = tcp_recalc_ssthresh(tp); + tp->snd_ssthresh = tp->ca_ops->ssthresh(tp); + tcp_ca_event(tp, CA_EVENT_FRTO); } /* Have to clear retransmission markers here to keep the bookkeeping @@ -1252,8 +1187,6 @@ static void tcp_enter_frto_loss(struct sock *sk) tcp_set_ca_state(tp, TCP_CA_Loss); tp->high_seq = tp->frto_highmark; TCP_ECN_queue_cwr(tp); - - init_bictcp(tp); } void tcp_clear_retrans(struct tcp_sock *tp) @@ -1283,7 +1216,8 @@ void tcp_enter_loss(struct sock *sk, int how) if (tp->ca_state <= TCP_CA_Disorder || tp->snd_una == tp->high_seq || (tp->ca_state == TCP_CA_Loss && !tp->retransmits)) { tp->prior_ssthresh = tcp_current_ssthresh(tp); - tp->snd_ssthresh = tcp_recalc_ssthresh(tp); + tp->snd_ssthresh = tp->ca_ops->ssthresh(tp); + tcp_ca_event(tp, CA_EVENT_LOSS); } tp->snd_cwnd = 1; tp->snd_cwnd_cnt = 0; @@ -1596,28 +1530,14 @@ static inline void tcp_moderate_cwnd(struct tcp_sock *tp) } /* Decrease cwnd each second ack. */ - static void tcp_cwnd_down(struct tcp_sock *tp) { int decr = tp->snd_cwnd_cnt + 1; - __u32 limit; - - /* - * TCP Westwood - * Here limit is evaluated as BWestimation*RTTmin (for obtaining it - * in packets we use mss_cache). If sysctl_tcp_westwood is off - * tcp_westwood_bw_rttmin() returns 0. In such case snd_ssthresh is - * still used as usual. It prevents other strange cases in which - * BWE*RTTmin could assume value 0. It should not happen but... - */ - - if (!(limit = tcp_westwood_bw_rttmin(tp))) - limit = tp->snd_ssthresh/2; tp->snd_cwnd_cnt = decr&1; decr >>= 1; - if (decr && tp->snd_cwnd > limit) + if (decr && tp->snd_cwnd > tp->ca_ops->min_cwnd(tp)) tp->snd_cwnd -= decr; tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp)+1); @@ -1654,8 +1574,8 @@ static void DBGUNDO(struct sock *sk, struct tcp_sock *tp, const char *msg) static void tcp_undo_cwr(struct tcp_sock *tp, int undo) { if (tp->prior_ssthresh) { - if (tcp_is_bic(tp)) - tp->snd_cwnd = max(tp->snd_cwnd, tp->bictcp.last_max_cwnd); + if (tp->ca_ops->undo_cwnd) + tp->snd_cwnd = tp->ca_ops->undo_cwnd(tp); else tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh<<1); @@ -1767,11 +1687,9 @@ static int tcp_try_undo_loss(struct sock *sk, struct tcp_sock *tp) static inline void tcp_complete_cwr(struct tcp_sock *tp) { - if (tcp_westwood_cwnd(tp)) - tp->snd_ssthresh = tp->snd_cwnd; - else - tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh); + tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh); tp->snd_cwnd_stamp = tcp_time_stamp; + tcp_ca_event(tp, CA_EVENT_COMPLETE_CWR); } static void tcp_try_to_open(struct sock *sk, struct tcp_sock *tp, int flag) @@ -1946,7 +1864,7 @@ tcp_fastretrans_alert(struct sock *sk, u32 prior_snd_una, if (tp->ca_state < TCP_CA_CWR) { if (!(flag&FLAG_ECE)) tp->prior_ssthresh = tcp_current_ssthresh(tp); - tp->snd_ssthresh = tcp_recalc_ssthresh(tp); + tp->snd_ssthresh = tp->ca_ops->ssthresh(tp); TCP_ECN_queue_cwr(tp); } @@ -1963,7 +1881,7 @@ tcp_fastretrans_alert(struct sock *sk, u32 prior_snd_una, /* Read draft-ietf-tcplw-high-performance before mucking * with this code. (Superceeds RFC1323) */ -static void tcp_ack_saw_tstamp(struct tcp_sock *tp, int flag) +static void tcp_ack_saw_tstamp(struct tcp_sock *tp, u32 *usrtt, int flag) { __u32 seq_rtt; @@ -1983,13 +1901,13 @@ static void tcp_ack_saw_tstamp(struct tcp_sock *tp, int flag) * in window is lost... Voila. --ANK (010210) */ seq_rtt = tcp_time_stamp - tp->rx_opt.rcv_tsecr; - tcp_rtt_estimator(tp, seq_rtt); + tcp_rtt_estimator(tp, seq_rtt, usrtt); tcp_set_rto(tp); tp->backoff = 0; tcp_bound_rto(tp); } -static void tcp_ack_no_tstamp(struct tcp_sock *tp, u32 seq_rtt, int flag) +static void tcp_ack_no_tstamp(struct tcp_sock *tp, u32 seq_rtt, u32 *usrtt, int flag) { /* We don't have a timestamp. Can only use * packets that are not retransmitted to determine @@ -2003,338 +1921,29 @@ static void tcp_ack_no_tstamp(struct tcp_sock *tp, u32 seq_rtt, int flag) if (flag & FLAG_RETRANS_DATA_ACKED) return; - tcp_rtt_estimator(tp, seq_rtt); + tcp_rtt_estimator(tp, seq_rtt, usrtt); tcp_set_rto(tp); tp->backoff = 0; tcp_bound_rto(tp); } static inline void tcp_ack_update_rtt(struct tcp_sock *tp, - int flag, s32 seq_rtt) + int flag, s32 seq_rtt, u32 *usrtt) { /* Note that peer MAY send zero echo. In this case it is ignored. (rfc1323) */ if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr) - tcp_ack_saw_tstamp(tp, flag); + tcp_ack_saw_tstamp(tp, usrtt, flag); else if (seq_rtt >= 0) - tcp_ack_no_tstamp(tp, seq_rtt, flag); + tcp_ack_no_tstamp(tp, seq_rtt, usrtt, flag); } -/* - * Compute congestion window to use. - * - * This is from the implementation of BICTCP in - * Lison-Xu, Kahaled Harfoush, and Injog Rhee. - * "Binary Increase Congestion Control for Fast, Long Distance - * Networks" in InfoComm 2004 - * Available from: - * http://www.csc.ncsu.edu/faculty/rhee/export/bitcp.pdf - * - * Unless BIC is enabled and congestion window is large - * this behaves the same as the original Reno. - */ -static inline __u32 bictcp_cwnd(struct tcp_sock *tp) -{ - /* orignal Reno behaviour */ - if (!tcp_is_bic(tp)) - return tp->snd_cwnd; - - if (tp->bictcp.last_cwnd == tp->snd_cwnd && - (s32)(tcp_time_stamp - tp->bictcp.last_stamp) <= (HZ>>5)) - return tp->bictcp.cnt; - - tp->bictcp.last_cwnd = tp->snd_cwnd; - tp->bictcp.last_stamp = tcp_time_stamp; - - /* start off normal */ - if (tp->snd_cwnd <= sysctl_tcp_bic_low_window) - tp->bictcp.cnt = tp->snd_cwnd; - - /* binary increase */ - else if (tp->snd_cwnd < tp->bictcp.last_max_cwnd) { - __u32 dist = (tp->bictcp.last_max_cwnd - tp->snd_cwnd) - / BICTCP_B; - - if (dist > BICTCP_MAX_INCREMENT) - /* linear increase */ - tp->bictcp.cnt = tp->snd_cwnd / BICTCP_MAX_INCREMENT; - else if (dist <= 1U) - /* binary search increase */ - tp->bictcp.cnt = tp->snd_cwnd * BICTCP_FUNC_OF_MIN_INCR - / BICTCP_B; - else - /* binary search increase */ - tp->bictcp.cnt = tp->snd_cwnd / dist; - } else { - /* slow start amd linear increase */ - if (tp->snd_cwnd < tp->bictcp.last_max_cwnd + BICTCP_B) - /* slow start */ - tp->bictcp.cnt = tp->snd_cwnd * BICTCP_FUNC_OF_MIN_INCR - / BICTCP_B; - else if (tp->snd_cwnd < tp->bictcp.last_max_cwnd - + BICTCP_MAX_INCREMENT*(BICTCP_B-1)) - /* slow start */ - tp->bictcp.cnt = tp->snd_cwnd * (BICTCP_B-1) - / (tp->snd_cwnd-tp->bictcp.last_max_cwnd); - else - /* linear increase */ - tp->bictcp.cnt = tp->snd_cwnd / BICTCP_MAX_INCREMENT; - } - return tp->bictcp.cnt; -} - -/* This is Jacobson's slow start and congestion avoidance. - * SIGCOMM '88, p. 328. - */ -static inline void reno_cong_avoid(struct tcp_sock *tp) +static inline void tcp_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt, + u32 in_flight, int good) { - if (tp->snd_cwnd <= tp->snd_ssthresh) { - /* In "safe" area, increase. */ - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - } else { - /* In dangerous area, increase slowly. - * In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd - */ - if (tp->snd_cwnd_cnt >= bictcp_cwnd(tp)) { - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - tp->snd_cwnd_cnt=0; - } else - tp->snd_cwnd_cnt++; - } + tp->ca_ops->cong_avoid(tp, ack, rtt, in_flight, good); tp->snd_cwnd_stamp = tcp_time_stamp; } -/* This is based on the congestion detection/avoidance scheme described in - * Lawrence S. Brakmo and Larry L. Peterson. - * "TCP Vegas: End to end congestion avoidance on a global internet." - * IEEE Journal on Selected Areas in Communication, 13(8):1465--1480, - * October 1995. Available from: - * ftp://ftp.cs.arizona.edu/xkernel/Papers/jsac.ps - * - * See http://www.cs.arizona.edu/xkernel/ for their implementation. - * The main aspects that distinguish this implementation from the - * Arizona Vegas implementation are: - * o We do not change the loss detection or recovery mechanisms of - * Linux in any way. Linux already recovers from losses quite well, - * using fine-grained timers, NewReno, and FACK. - * o To avoid the performance penalty imposed by increasing cwnd - * only every-other RTT during slow start, we increase during - * every RTT during slow start, just like Reno. - * o Largely to allow continuous cwnd growth during slow start, - * we use the rate at which ACKs come back as the "actual" - * rate, rather than the rate at which data is sent. - * o To speed convergence to the right rate, we set the cwnd - * to achieve the right ("actual") rate when we exit slow start. - * o To filter out the noise caused by delayed ACKs, we use the - * minimum RTT sample observed during the last RTT to calculate - * the actual rate. - * o When the sender re-starts from idle, it waits until it has - * received ACKs for an entire flight of new data before making - * a cwnd adjustment decision. The original Vegas implementation - * assumed senders never went idle. - */ -static void vegas_cong_avoid(struct tcp_sock *tp, u32 ack, u32 seq_rtt) -{ - /* The key players are v_beg_snd_una and v_beg_snd_nxt. - * - * These are so named because they represent the approximate values - * of snd_una and snd_nxt at the beginning of the current RTT. More - * precisely, they represent the amount of data sent during the RTT. - * At the end of the RTT, when we receive an ACK for v_beg_snd_nxt, - * we will calculate that (v_beg_snd_nxt - v_beg_snd_una) outstanding - * bytes of data have been ACKed during the course of the RTT, giving - * an "actual" rate of: - * - * (v_beg_snd_nxt - v_beg_snd_una) / (rtt duration) - * - * Unfortunately, v_beg_snd_una is not exactly equal to snd_una, - * because delayed ACKs can cover more than one segment, so they - * don't line up nicely with the boundaries of RTTs. - * - * Another unfortunate fact of life is that delayed ACKs delay the - * advance of the left edge of our send window, so that the number - * of bytes we send in an RTT is often less than our cwnd will allow. - * So we keep track of our cwnd separately, in v_beg_snd_cwnd. - */ - - if (after(ack, tp->vegas.beg_snd_nxt)) { - /* Do the Vegas once-per-RTT cwnd adjustment. */ - u32 old_wnd, old_snd_cwnd; - - - /* Here old_wnd is essentially the window of data that was - * sent during the previous RTT, and has all - * been acknowledged in the course of the RTT that ended - * with the ACK we just received. Likewise, old_snd_cwnd - * is the cwnd during the previous RTT. - */ - old_wnd = (tp->vegas.beg_snd_nxt - tp->vegas.beg_snd_una) / - tp->mss_cache_std; - old_snd_cwnd = tp->vegas.beg_snd_cwnd; - - /* Save the extent of the current window so we can use this - * at the end of the next RTT. - */ - tp->vegas.beg_snd_una = tp->vegas.beg_snd_nxt; - tp->vegas.beg_snd_nxt = tp->snd_nxt; - tp->vegas.beg_snd_cwnd = tp->snd_cwnd; - - /* Take into account the current RTT sample too, to - * decrease the impact of delayed acks. This double counts - * this sample since we count it for the next window as well, - * but that's not too awful, since we're taking the min, - * rather than averaging. - */ - vegas_rtt_calc(tp, seq_rtt); - - /* We do the Vegas calculations only if we got enough RTT - * samples that we can be reasonably sure that we got - * at least one RTT sample that wasn't from a delayed ACK. - * If we only had 2 samples total, - * then that means we're getting only 1 ACK per RTT, which - * means they're almost certainly delayed ACKs. - * If we have 3 samples, we should be OK. - */ - - if (tp->vegas.cntRTT <= 2) { - /* We don't have enough RTT samples to do the Vegas - * calculation, so we'll behave like Reno. - */ - if (tp->snd_cwnd > tp->snd_ssthresh) - tp->snd_cwnd++; - } else { - u32 rtt, target_cwnd, diff; - - /* We have enough RTT samples, so, using the Vegas - * algorithm, we determine if we should increase or - * decrease cwnd, and by how much. - */ - - /* Pluck out the RTT we are using for the Vegas - * calculations. This is the min RTT seen during the - * last RTT. Taking the min filters out the effects - * of delayed ACKs, at the cost of noticing congestion - * a bit later. - */ - rtt = tp->vegas.minRTT; - - /* Calculate the cwnd we should have, if we weren't - * going too fast. - * - * This is: - * (actual rate in segments) * baseRTT - * We keep it as a fixed point number with - * V_PARAM_SHIFT bits to the right of the binary point. - */ - target_cwnd = ((old_wnd * tp->vegas.baseRTT) - << V_PARAM_SHIFT) / rtt; - - /* Calculate the difference between the window we had, - * and the window we would like to have. This quantity - * is the "Diff" from the Arizona Vegas papers. - * - * Again, this is a fixed point number with - * V_PARAM_SHIFT bits to the right of the binary - * point. - */ - diff = (old_wnd << V_PARAM_SHIFT) - target_cwnd; - - if (tp->snd_cwnd < tp->snd_ssthresh) { - /* Slow start. */ - if (diff > sysctl_tcp_vegas_gamma) { - /* Going too fast. Time to slow down - * and switch to congestion avoidance. - */ - tp->snd_ssthresh = 2; - - /* Set cwnd to match the actual rate - * exactly: - * cwnd = (actual rate) * baseRTT - * Then we add 1 because the integer - * truncation robs us of full link - * utilization. - */ - tp->snd_cwnd = min(tp->snd_cwnd, - (target_cwnd >> - V_PARAM_SHIFT)+1); - - } - } else { - /* Congestion avoidance. */ - u32 next_snd_cwnd; - - /* Figure out where we would like cwnd - * to be. - */ - if (diff > sysctl_tcp_vegas_beta) { - /* The old window was too fast, so - * we slow down. - */ - next_snd_cwnd = old_snd_cwnd - 1; - } else if (diff < sysctl_tcp_vegas_alpha) { - /* We don't have enough extra packets - * in the network, so speed up. - */ - next_snd_cwnd = old_snd_cwnd + 1; - } else { - /* Sending just as fast as we - * should be. - */ - next_snd_cwnd = old_snd_cwnd; - } - - /* Adjust cwnd upward or downward, toward the - * desired value. - */ - if (next_snd_cwnd > tp->snd_cwnd) - tp->snd_cwnd++; - else if (next_snd_cwnd < tp->snd_cwnd) - tp->snd_cwnd--; - } - } - - /* Wipe the slate clean for the next RTT. */ - tp->vegas.cntRTT = 0; - tp->vegas.minRTT = 0x7fffffff; - } - - /* The following code is executed for every ack we receive, - * except for conditions checked in should_advance_cwnd() - * before the call to tcp_cong_avoid(). Mainly this means that - * we only execute this code if the ack actually acked some - * data. - */ - - /* If we are in slow start, increase our cwnd in response to this ACK. - * (If we are not in slow start then we are in congestion avoidance, - * and adjust our congestion window only once per RTT. See the code - * above.) - */ - if (tp->snd_cwnd <= tp->snd_ssthresh) - tp->snd_cwnd++; - - /* to keep cwnd from growing without bound */ - tp->snd_cwnd = min_t(u32, tp->snd_cwnd, tp->snd_cwnd_clamp); - - /* Make sure that we are never so timid as to reduce our cwnd below - * 2 MSS. - * - * Going below 2 MSS would risk huge delayed ACKs from our receiver. - */ - tp->snd_cwnd = max(tp->snd_cwnd, 2U); - - tp->snd_cwnd_stamp = tcp_time_stamp; -} - -static inline void tcp_cong_avoid(struct tcp_sock *tp, u32 ack, u32 seq_rtt) -{ - if (tcp_vegas_enabled(tp)) - vegas_cong_avoid(tp, ack, seq_rtt); - else - reno_cong_avoid(tp); -} - /* Restart timer after forward progress on connection. * RFC2988 recommends to restart timer to now+rto. */ @@ -2415,13 +2024,18 @@ static int tcp_tso_acked(struct sock *sk, struct sk_buff *skb, /* Remove acknowledged frames from the retransmission queue. */ -static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) +static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p, s32 *seq_usrtt) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; __u32 now = tcp_time_stamp; int acked = 0; __s32 seq_rtt = -1; + struct timeval usnow; + u32 pkts_acked = 0; + + if (seq_usrtt) + do_gettimeofday(&usnow); while ((skb = skb_peek(&sk->sk_write_queue)) && skb != sk->sk_send_head) { @@ -2448,6 +2062,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) */ if (!(scb->flags & TCPCB_FLAG_SYN)) { acked |= FLAG_DATA_ACKED; + ++pkts_acked; } else { acked |= FLAG_SYN_ACKED; tp->retrans_stamp = 0; @@ -2461,6 +2076,10 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) seq_rtt = -1; } else if (seq_rtt < 0) seq_rtt = now - scb->when; + if (seq_usrtt) + *seq_usrtt = (usnow.tv_sec - skb->stamp.tv_sec) * 1000000 + + (usnow.tv_usec - skb->stamp.tv_usec); + if (sacked & TCPCB_SACKED_ACKED) tp->sacked_out -= tcp_skb_pcount(skb); if (sacked & TCPCB_LOST) @@ -2479,8 +2098,11 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) } if (acked&FLAG_ACKED) { - tcp_ack_update_rtt(tp, acked, seq_rtt); + tcp_ack_update_rtt(tp, acked, seq_rtt, seq_usrtt); tcp_ack_packets_out(sk, tp); + + if (tp->ca_ops->pkts_acked) + tp->ca_ops->pkts_acked(tp, pkts_acked); } #if FASTRETRANS_DEBUG > 0 @@ -2624,257 +2246,6 @@ static void tcp_process_frto(struct sock *sk, u32 prior_snd_una) tp->frto_counter = (tp->frto_counter + 1) % 3; } -/* - * TCP Westwood+ - */ - -/* - * @init_westwood - * This function initializes fields used in TCP Westwood+. We can't - * get no information about RTTmin at this time so we simply set it to - * TCP_WESTWOOD_INIT_RTT. This value was chosen to be too conservative - * since in this way we're sure it will be updated in a consistent - * way as soon as possible. It will reasonably happen within the first - * RTT period of the connection lifetime. - */ - -static void init_westwood(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - - tp->westwood.bw_ns_est = 0; - tp->westwood.bw_est = 0; - tp->westwood.accounted = 0; - tp->westwood.cumul_ack = 0; - tp->westwood.rtt_win_sx = tcp_time_stamp; - tp->westwood.rtt = TCP_WESTWOOD_INIT_RTT; - tp->westwood.rtt_min = TCP_WESTWOOD_INIT_RTT; - tp->westwood.snd_una = tp->snd_una; -} - -/* - * @westwood_do_filter - * Low-pass filter. Implemented using constant coeffients. - */ - -static inline __u32 westwood_do_filter(__u32 a, __u32 b) -{ - return (((7 * a) + b) >> 3); -} - -static void westwood_filter(struct sock *sk, __u32 delta) -{ - struct tcp_sock *tp = tcp_sk(sk); - - tp->westwood.bw_ns_est = - westwood_do_filter(tp->westwood.bw_ns_est, - tp->westwood.bk / delta); - tp->westwood.bw_est = - westwood_do_filter(tp->westwood.bw_est, - tp->westwood.bw_ns_est); -} - -/* - * @westwood_update_rttmin - * It is used to update RTTmin. In this case we MUST NOT use - * WESTWOOD_RTT_MIN minimum bound since we could be on a LAN! - */ - -static inline __u32 westwood_update_rttmin(const struct sock *sk) -{ - const struct tcp_sock *tp = tcp_sk(sk); - __u32 rttmin = tp->westwood.rtt_min; - - if (tp->westwood.rtt != 0 && - (tp->westwood.rtt < tp->westwood.rtt_min || !rttmin)) - rttmin = tp->westwood.rtt; - - return rttmin; -} - -/* - * @westwood_acked - * Evaluate increases for dk. - */ - -static inline __u32 westwood_acked(const struct sock *sk) -{ - const struct tcp_sock *tp = tcp_sk(sk); - - return tp->snd_una - tp->westwood.snd_una; -} - -/* - * @westwood_new_window - * It evaluates if we are receiving data inside the same RTT window as - * when we started. - * Return value: - * It returns 0 if we are still evaluating samples in the same RTT - * window, 1 if the sample has to be considered in the next window. - */ - -static int westwood_new_window(const struct sock *sk) -{ - const struct tcp_sock *tp = tcp_sk(sk); - __u32 left_bound; - __u32 rtt; - int ret = 0; - - left_bound = tp->westwood.rtt_win_sx; - rtt = max(tp->westwood.rtt, (u32) TCP_WESTWOOD_RTT_MIN); - - /* - * A RTT-window has passed. Be careful since if RTT is less than - * 50ms we don't filter but we continue 'building the sample'. - * This minimum limit was choosen since an estimation on small - * time intervals is better to avoid... - * Obvioulsy on a LAN we reasonably will always have - * right_bound = left_bound + WESTWOOD_RTT_MIN - */ - - if ((left_bound + rtt) < tcp_time_stamp) - ret = 1; - - return ret; -} - -/* - * @westwood_update_window - * It updates RTT evaluation window if it is the right moment to do - * it. If so it calls filter for evaluating bandwidth. - */ - -static void __westwood_update_window(struct sock *sk, __u32 now) -{ - struct tcp_sock *tp = tcp_sk(sk); - __u32 delta = now - tp->westwood.rtt_win_sx; - - if (delta) { - if (tp->westwood.rtt) - westwood_filter(sk, delta); - - tp->westwood.bk = 0; - tp->westwood.rtt_win_sx = tcp_time_stamp; - } -} - - -static void westwood_update_window(struct sock *sk, __u32 now) -{ - if (westwood_new_window(sk)) - __westwood_update_window(sk, now); -} - -/* - * @__tcp_westwood_fast_bw - * It is called when we are in fast path. In particular it is called when - * header prediction is successfull. In such case infact update is - * straight forward and doesn't need any particular care. - */ - -static void __tcp_westwood_fast_bw(struct sock *sk, struct sk_buff *skb) -{ - struct tcp_sock *tp = tcp_sk(sk); - - westwood_update_window(sk, tcp_time_stamp); - - tp->westwood.bk += westwood_acked(sk); - tp->westwood.snd_una = tp->snd_una; - tp->westwood.rtt_min = westwood_update_rttmin(sk); -} - -static inline void tcp_westwood_fast_bw(struct sock *sk, struct sk_buff *skb) -{ - if (tcp_is_westwood(tcp_sk(sk))) - __tcp_westwood_fast_bw(sk, skb); -} - - -/* - * @westwood_dupack_update - * It updates accounted and cumul_ack when receiving a dupack. - */ - -static void westwood_dupack_update(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - - tp->westwood.accounted += tp->mss_cache_std; - tp->westwood.cumul_ack = tp->mss_cache_std; -} - -static inline int westwood_may_change_cumul(struct tcp_sock *tp) -{ - return (tp->westwood.cumul_ack > tp->mss_cache_std); -} - -static inline void westwood_partial_update(struct tcp_sock *tp) -{ - tp->westwood.accounted -= tp->westwood.cumul_ack; - tp->westwood.cumul_ack = tp->mss_cache_std; -} - -static inline void westwood_complete_update(struct tcp_sock *tp) -{ - tp->westwood.cumul_ack -= tp->westwood.accounted; - tp->westwood.accounted = 0; -} - -/* - * @westwood_acked_count - * This function evaluates cumul_ack for evaluating dk in case of - * delayed or partial acks. - */ - -static inline __u32 westwood_acked_count(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - - tp->westwood.cumul_ack = westwood_acked(sk); - - /* If cumul_ack is 0 this is a dupack since it's not moving - * tp->snd_una. - */ - if (!(tp->westwood.cumul_ack)) - westwood_dupack_update(sk); - - if (westwood_may_change_cumul(tp)) { - /* Partial or delayed ack */ - if (tp->westwood.accounted >= tp->westwood.cumul_ack) - westwood_partial_update(tp); - else - westwood_complete_update(tp); - } - - tp->westwood.snd_una = tp->snd_una; - - return tp->westwood.cumul_ack; -} - - -/* - * @__tcp_westwood_slow_bw - * It is called when something is going wrong..even if there could - * be no problems! Infact a simple delayed packet may trigger a - * dupack. But we need to be careful in such case. - */ - -static void __tcp_westwood_slow_bw(struct sock *sk, struct sk_buff *skb) -{ - struct tcp_sock *tp = tcp_sk(sk); - - westwood_update_window(sk, tcp_time_stamp); - - tp->westwood.bk += westwood_acked_count(sk); - tp->westwood.rtt_min = westwood_update_rttmin(sk); -} - -static inline void tcp_westwood_slow_bw(struct sock *sk, struct sk_buff *skb) -{ - if (tcp_is_westwood(tcp_sk(sk))) - __tcp_westwood_slow_bw(sk, skb); -} - /* This routine deals with incoming acks, but not outgoing ones. */ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) { @@ -2884,6 +2255,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) u32 ack = TCP_SKB_CB(skb)->ack_seq; u32 prior_in_flight; s32 seq_rtt; + s32 seq_usrtt = 0; int prior_packets; /* If the ack is newer than sent or older than previous acks @@ -2902,9 +2274,10 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) */ tcp_update_wl(tp, ack, ack_seq); tp->snd_una = ack; - tcp_westwood_fast_bw(sk, skb); flag |= FLAG_WIN_UPDATE; + tcp_ca_event(tp, CA_EVENT_FAST_ACK); + NET_INC_STATS_BH(LINUX_MIB_TCPHPACKS); } else { if (ack_seq != TCP_SKB_CB(skb)->end_seq) @@ -2920,7 +2293,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) if (TCP_ECN_rcv_ecn_echo(tp, skb->h.th)) flag |= FLAG_ECE; - tcp_westwood_slow_bw(sk,skb); + tcp_ca_event(tp, CA_EVENT_SLOW_ACK); } /* We passed data and got it acked, remove any soft error @@ -2935,22 +2308,20 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) prior_in_flight = tcp_packets_in_flight(tp); /* See if we can take anything off of the retransmit queue. */ - flag |= tcp_clean_rtx_queue(sk, &seq_rtt); + flag |= tcp_clean_rtx_queue(sk, &seq_rtt, + tp->ca_ops->rtt_sample ? &seq_usrtt : NULL); if (tp->frto_counter) tcp_process_frto(sk, prior_snd_una); if (tcp_ack_is_dubious(tp, flag)) { /* Advanve CWND, if state allows this. */ - if ((flag & FLAG_DATA_ACKED) && - (tcp_vegas_enabled(tp) || prior_in_flight >= tp->snd_cwnd) && - tcp_may_raise_cwnd(tp, flag)) - tcp_cong_avoid(tp, ack, seq_rtt); + if ((flag & FLAG_DATA_ACKED) && tcp_may_raise_cwnd(tp, flag)) + tcp_cong_avoid(tp, ack, seq_rtt, prior_in_flight, 0); tcp_fastretrans_alert(sk, prior_snd_una, prior_packets, flag); } else { - if ((flag & FLAG_DATA_ACKED) && - (tcp_vegas_enabled(tp) || prior_in_flight >= tp->snd_cwnd)) - tcp_cong_avoid(tp, ack, seq_rtt); + if ((flag & FLAG_DATA_ACKED)) + tcp_cong_avoid(tp, ack, seq_rtt, prior_in_flight, 1); } if ((flag & FLAG_FORWARD_PROGRESS) || !(flag&FLAG_NOT_DUP)) @@ -4552,6 +3923,8 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, tcp_init_metrics(sk); + tcp_init_congestion_control(tp); + /* Prevent spurious tcp_cwnd_restart() on first data * packet. */ @@ -4708,9 +4081,6 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, if(tp->af_specific->conn_request(sk, skb) < 0) return 1; - init_westwood(sk); - init_bictcp(tp); - /* Now we have several options: In theory there is * nothing else in the frame. KA9Q has an option to * send data with the syn, BSD accepts data with the @@ -4732,9 +4102,6 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, goto discard; case TCP_SYN_SENT: - init_westwood(sk); - init_bictcp(tp); - queued = tcp_rcv_synsent_state_process(sk, skb, th, len); if (queued >= 0) return queued; @@ -4816,7 +4183,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, */ if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr && !tp->srtt) - tcp_ack_saw_tstamp(tp, 0); + tcp_ack_saw_tstamp(tp, 0, 0); if (tp->rx_opt.tstamp_ok) tp->advmss -= TCPOLEN_TSTAMP_ALIGNED; @@ -4828,6 +4195,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, tcp_init_metrics(sk); + tcp_init_congestion_control(tp); + /* Prevent spurious tcp_cwnd_restart() on * first data packet. */ diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 2d41d5d6ad19..9122814c13ad 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2048,6 +2048,7 @@ static int tcp_v4_init_sock(struct sock *sk) tp->mss_cache_std = tp->mss_cache = 536; tp->reordering = sysctl_tcp_reordering; + tp->ca_ops = &tcp_reno; sk->sk_state = TCP_CLOSE; @@ -2070,6 +2071,8 @@ int tcp_v4_destroy_sock(struct sock *sk) tcp_clear_xmit_timers(sk); + tcp_cleanup_congestion_control(tp); + /* Cleanup up the write buffer. */ sk_stream_writequeue_purge(sk); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index b3943e7562f3..f42a284164b7 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -774,6 +774,8 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, newtp->frto_counter = 0; newtp->frto_highmark = 0; + newtp->ca_ops = &tcp_reno; + tcp_set_ca_state(newtp, TCP_CA_Open); tcp_init_xmit_timers(newsk); skb_queue_head_init(&newtp->out_of_order_queue); @@ -842,8 +844,6 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, if (newtp->ecn_flags&TCP_ECN_OK) sock_set_flag(newsk, SOCK_NO_LARGESEND); - tcp_ca_init(newtp); - TCP_INC_STATS_BH(TCP_MIB_PASSIVEOPENS); } return newsk; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f17c6577e337..0e17c244875c 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -111,8 +111,7 @@ static void tcp_cwnd_restart(struct tcp_sock *tp, struct dst_entry *dst) u32 restart_cwnd = tcp_init_cwnd(tp, dst); u32 cwnd = tp->snd_cwnd; - if (tcp_is_vegas(tp)) - tcp_vegas_enable(tp); + tcp_ca_event(tp, CA_EVENT_CWND_RESTART); tp->snd_ssthresh = tcp_current_ssthresh(tp); restart_cwnd = min(restart_cwnd, cwnd); @@ -280,6 +279,10 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb) #define SYSCTL_FLAG_WSCALE 0x2 #define SYSCTL_FLAG_SACK 0x4 + /* If congestion control is doing timestamping */ + if (tp->ca_ops->rtt_sample) + do_gettimeofday(&skb->stamp); + sysctl_flags = 0; if (tcb->flags & TCPCB_FLAG_SYN) { tcp_header_size = sizeof(struct tcphdr) + TCPOLEN_MSS; @@ -304,17 +307,8 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb) (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK)); } - /* - * If the connection is idle and we are restarting, - * then we don't want to do any Vegas calculations - * until we get fresh RTT samples. So when we - * restart, we reset our Vegas state to a clean - * slate. After we get acks for this flight of - * packets, _then_ we can make Vegas calculations - * again. - */ - if (tcp_is_vegas(tp) && tcp_packets_in_flight(tp) == 0) - tcp_vegas_enable(tp); + if (tcp_packets_in_flight(tp) == 0) + tcp_ca_event(tp, CA_EVENT_TX_START); th = (struct tcphdr *) skb_push(skb, tcp_header_size); skb->h.th = th; @@ -521,6 +515,7 @@ static int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len) * skbs, which it never sent before. --ANK */ TCP_SKB_CB(buff)->when = TCP_SKB_CB(skb)->when; + buff->stamp = skb->stamp; if (TCP_SKB_CB(skb)->sacked & TCPCB_LOST) { tp->lost_out -= tcp_skb_pcount(skb); @@ -1449,7 +1444,6 @@ static inline void tcp_connect_init(struct sock *sk) tp->window_clamp = dst_metric(dst, RTAX_WINDOW); tp->advmss = dst_metric(dst, RTAX_ADVMSS); tcp_initialize_rcv_mss(sk); - tcp_ca_init(tp); tcp_select_initial_window(tcp_full_space(sk), tp->advmss - (tp->rx_opt.ts_recent_stamp ? tp->tcp_header_len - sizeof(struct tcphdr) : 0), @@ -1503,7 +1497,6 @@ int tcp_connect(struct sock *sk) TCP_SKB_CB(buff)->end_seq = tp->write_seq; tp->snd_nxt = tp->write_seq; tp->pushed_seq = tp->write_seq; - tcp_ca_init(tp); /* Send it off. */ TCP_SKB_CB(buff)->when = tcp_time_stamp; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 2414937f2a83..fce56039b0e9 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2025,7 +2025,7 @@ static int tcp_v6_init_sock(struct sock *sk) sk->sk_state = TCP_CLOSE; tp->af_specific = &ipv6_specific; - + tp->ca_ops = &tcp_reno; sk->sk_write_space = sk_stream_write_space; sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); -- cgit v1.2.3 From 056ede6cface66b400cd3b8e60ed077cc5b85c18 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 12:21:28 -0700 Subject: [TCP]: Report congestion control algorithm in tcp_diag. Enhancement to the tcp_diag interface used by the iproute2 ss command to report the tcp congestion control being used by a socket. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/tcp_diag.h | 4 ++-- net/ipv4/tcp_diag.c | 5 +++++ 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tcp_diag.h b/include/linux/tcp_diag.h index ceee962e1d15..7a5996743946 100644 --- a/include/linux/tcp_diag.h +++ b/include/linux/tcp_diag.h @@ -99,9 +99,10 @@ enum TCPDIAG_MEMINFO, TCPDIAG_INFO, TCPDIAG_VEGASINFO, + TCPDIAG_CONG, }; -#define TCPDIAG_MAX TCPDIAG_VEGASINFO +#define TCPDIAG_MAX TCPDIAG_CONG /* TCPDIAG_MEM */ @@ -123,5 +124,4 @@ struct tcpvegas_info { __u32 tcpv_minrtt; }; - #endif /* _TCP_DIAG_H_ */ diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index a4e512036d88..f66945cb158f 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -65,6 +65,11 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk, if (ext & (1<<(TCPDIAG_INFO-1))) info = TCPDIAG_PUT(skb, TCPDIAG_INFO, sizeof(*info)); + if (ext & (1<<(TCPDIAG_CONG-1))) { + size_t len = strlen(tp->ca_ops->name); + strcpy(TCPDIAG_PUT(skb, TCPDIAG_CONG, len+1), + tp->ca_ops->name); + } } r->tcpdiag_family = sk->sk_family; r->tcpdiag_state = sk->sk_state; -- cgit v1.2.3 From c1ebcdb8c422cd73f54bcd2b9953e443a47667e5 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:08:59 -0700 Subject: [NET]: Remove obsolete fastroute stats. Remove last vestiages of fastroute code that is no longer used. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 ----- net/core/dev.c | 10 ++-------- net/core/sysctl_net_core.c | 1 - 3 files changed, 2 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d89816ad642f..c2e15e381a58 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -165,11 +165,6 @@ struct netif_rx_stats unsigned dropped; unsigned time_squeeze; unsigned throttled; - unsigned fastroute_hit; - unsigned fastroute_success; - unsigned fastroute_defer; - unsigned fastroute_deferred_out; - unsigned fastroute_latency_reduction; unsigned cpu_collision; }; diff --git a/net/core/dev.c b/net/core/dev.c index ab935778ce81..4f1ae2efe872 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2056,14 +2056,8 @@ static int softnet_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x\n", s->total, s->dropped, s->time_squeeze, s->throttled, - s->fastroute_hit, s->fastroute_success, s->fastroute_defer, - s->fastroute_deferred_out, -#if 0 - s->fastroute_latency_reduction -#else - s->cpu_collision -#endif - ); + 0, 0, 0, 0, /* was fastroute */ + s->cpu_collision ); return 0; } diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 880a88815211..76e9987474ca 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -18,7 +18,6 @@ extern int no_cong_thresh; extern int no_cong; extern int lo_cong; extern int mod_cong; -extern int netdev_fastroute; extern int net_msg_cost; extern int net_msg_burst; -- cgit v1.2.3 From 34008d8c631d067caffa136313260525f3ae48a2 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:10:00 -0700 Subject: [NET]: Remove obsolete netif_rx congestion sensing mechanism. Remove the congestion sensing mechanism from netif_rx, and always return either full or empty. Almost no driver checks the return value from netif_rx, and those that do only use it for debug messages. The original design of netif_rx was to do flow control based on the receive queue, but NAPI has supplanted this and no driver uses the feedback. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 -- net/core/dev.c | 88 +--------------------------------------------- net/core/sysctl_net_core.c | 36 ------------------- 3 files changed, 1 insertion(+), 125 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c2e15e381a58..718ad579c65c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -558,8 +558,6 @@ static inline int unregister_gifconf(unsigned int family) struct softnet_data { int throttle; - int cng_level; - int avg_blog; struct sk_buff_head input_pkt_queue; struct list_head poll_list; struct net_device *output_queue; diff --git a/net/core/dev.c b/net/core/dev.c index 4f1ae2efe872..3156df699f01 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -115,18 +115,6 @@ #endif /* CONFIG_NET_RADIO */ #include -/* This define, if set, will randomly drop a packet when congestion - * is more than moderate. It helps fairness in the multi-interface - * case when one of them is a hog, but it kills performance for the - * single interface case so it is off now by default. - */ -#undef RAND_LIE - -/* Setting this will sample the queue lengths and thus congestion - * via a timer instead of as each packet is received. - */ -#undef OFFLINE_SAMPLE - /* * The list of packet types we will receive (as opposed to discard) * and the routines to invoke. @@ -159,11 +147,6 @@ static DEFINE_SPINLOCK(ptype_lock); static struct list_head ptype_base[16]; /* 16 way hashed list */ static struct list_head ptype_all; /* Taps */ -#ifdef OFFLINE_SAMPLE -static void sample_queue(unsigned long dummy); -static struct timer_list samp_timer = TIMER_INITIALIZER(sample_queue, 0, 0); -#endif - /* * The @dev_base list is protected by @dev_base_lock and the rtln * semaphore. @@ -1365,69 +1348,10 @@ out: int netdev_max_backlog = 300; int weight_p = 64; /* old backlog weight */ -/* These numbers are selected based on intuition and some - * experimentatiom, if you have more scientific way of doing this - * please go ahead and fix things. - */ -int no_cong_thresh = 10; -int no_cong = 20; -int lo_cong = 100; -int mod_cong = 290; DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; -static void get_sample_stats(int cpu) -{ -#ifdef RAND_LIE - unsigned long rd; - int rq; -#endif - struct softnet_data *sd = &per_cpu(softnet_data, cpu); - int blog = sd->input_pkt_queue.qlen; - int avg_blog = sd->avg_blog; - - avg_blog = (avg_blog >> 1) + (blog >> 1); - - if (avg_blog > mod_cong) { - /* Above moderate congestion levels. */ - sd->cng_level = NET_RX_CN_HIGH; -#ifdef RAND_LIE - rd = net_random(); - rq = rd % netdev_max_backlog; - if (rq < avg_blog) /* unlucky bastard */ - sd->cng_level = NET_RX_DROP; -#endif - } else if (avg_blog > lo_cong) { - sd->cng_level = NET_RX_CN_MOD; -#ifdef RAND_LIE - rd = net_random(); - rq = rd % netdev_max_backlog; - if (rq < avg_blog) /* unlucky bastard */ - sd->cng_level = NET_RX_CN_HIGH; -#endif - } else if (avg_blog > no_cong) - sd->cng_level = NET_RX_CN_LOW; - else /* no congestion */ - sd->cng_level = NET_RX_SUCCESS; - - sd->avg_blog = avg_blog; -} - -#ifdef OFFLINE_SAMPLE -static void sample_queue(unsigned long dummy) -{ -/* 10 ms 0r 1ms -- i don't care -- JHS */ - int next_tick = 1; - int cpu = smp_processor_id(); - - get_sample_stats(cpu); - next_tick += jiffies; - mod_timer(&samp_timer, next_tick); -} -#endif - - /** * netif_rx - post buffer to the network code * @skb: buffer to post @@ -1476,11 +1400,8 @@ int netif_rx(struct sk_buff *skb) enqueue: dev_hold(skb->dev); __skb_queue_tail(&queue->input_pkt_queue, skb); -#ifndef OFFLINE_SAMPLE - get_sample_stats(this_cpu); -#endif local_irq_restore(flags); - return queue->cng_level; + return NET_RX_SUCCESS; } if (queue->throttle) @@ -3300,8 +3221,6 @@ static int __init net_dev_init(void) queue = &per_cpu(softnet_data, i); skb_queue_head_init(&queue->input_pkt_queue); queue->throttle = 0; - queue->cng_level = 0; - queue->avg_blog = 10; /* arbitrary non-zero */ queue->completion_queue = NULL; INIT_LIST_HEAD(&queue->poll_list); set_bit(__LINK_STATE_START, &queue->backlog_dev.state); @@ -3310,11 +3229,6 @@ static int __init net_dev_init(void) atomic_set(&queue->backlog_dev.refcnt, 1); } -#ifdef OFFLINE_SAMPLE - samp_timer.expires = jiffies + (10 * HZ); - add_timer(&samp_timer); -#endif - dev_boot_phase = 0; open_softirq(NET_TX_SOFTIRQ, net_tx_action, NULL); diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 76e9987474ca..fff63643a35c 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -14,10 +14,6 @@ extern int netdev_max_backlog; extern int weight_p; -extern int no_cong_thresh; -extern int no_cong; -extern int lo_cong; -extern int mod_cong; extern int net_msg_cost; extern int net_msg_burst; @@ -84,38 +80,6 @@ ctl_table core_table[] = { .mode = 0644, .proc_handler = &proc_dointvec }, - { - .ctl_name = NET_CORE_NO_CONG_THRESH, - .procname = "no_cong_thresh", - .data = &no_cong_thresh, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_CORE_NO_CONG, - .procname = "no_cong", - .data = &no_cong, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_CORE_LO_CONG, - .procname = "lo_cong", - .data = &lo_cong, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_CORE_MOD_CONG, - .procname = "mod_cong", - .data = &mod_cong, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, { .ctl_name = NET_CORE_MSG_COST, .procname = "message_cost", -- cgit v1.2.3 From 31aa02c53c84658f6694f319f09e232ede27be5a Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:12:48 -0700 Subject: [NET]: Eliminate netif_rx massive packet drops. Eliminate the throttling behaviour when the netif receive queue fills because it behaves badly when using high speed networks under load. The throttling cause multiple packet drops that cause TCP to go into slow start mode. The same effective patch has been part of BIC TCP and H-TCP as well as part of Web100. The existing code drops 100's of packets when the queue fills; this changes it to individual packet drop-tail. Signed-off-by: Stephen Hemmminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 +--- net/core/dev.c | 21 ++------------------- 2 files changed, 3 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 718ad579c65c..3a0ed7f9e801 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -164,7 +164,6 @@ struct netif_rx_stats unsigned total; unsigned dropped; unsigned time_squeeze; - unsigned throttled; unsigned cpu_collision; }; @@ -557,10 +556,9 @@ static inline int unregister_gifconf(unsigned int family) struct softnet_data { - int throttle; + struct net_device *output_queue; struct sk_buff_head input_pkt_queue; struct list_head poll_list; - struct net_device *output_queue; struct sk_buff *completion_queue; struct net_device backlog_dev; /* Sorry. 8) */ diff --git a/net/core/dev.c b/net/core/dev.c index 3156df699f01..1a64508e527f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -198,7 +198,7 @@ static struct notifier_block *netdev_chain; * Device drivers call our routines to queue packets here. We empty the * queue in the local softnet handler. */ -DEFINE_PER_CPU(struct softnet_data, softnet_data) = { 0, }; +DEFINE_PER_CPU(struct softnet_data, softnet_data) = { NULL }; #ifdef CONFIG_SYSFS extern int netdev_sysfs_init(void); @@ -1372,7 +1372,6 @@ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; int netif_rx(struct sk_buff *skb) { - int this_cpu; struct softnet_data *queue; unsigned long flags; @@ -1388,15 +1387,11 @@ int netif_rx(struct sk_buff *skb) * short when CPU is congested, but is still operating. */ local_irq_save(flags); - this_cpu = smp_processor_id(); queue = &__get_cpu_var(softnet_data); __get_cpu_var(netdev_rx_stat).total++; if (queue->input_pkt_queue.qlen <= netdev_max_backlog) { if (queue->input_pkt_queue.qlen) { - if (queue->throttle) - goto drop; - enqueue: dev_hold(skb->dev); __skb_queue_tail(&queue->input_pkt_queue, skb); @@ -1404,19 +1399,10 @@ enqueue: return NET_RX_SUCCESS; } - if (queue->throttle) - queue->throttle = 0; - netif_rx_schedule(&queue->backlog_dev); goto enqueue; } - if (!queue->throttle) { - queue->throttle = 1; - __get_cpu_var(netdev_rx_stat).throttled++; - } - -drop: __get_cpu_var(netdev_rx_stat).dropped++; local_irq_restore(flags); @@ -1701,8 +1687,6 @@ job_done: smp_mb__before_clear_bit(); netif_poll_enable(backlog_dev); - if (queue->throttle) - queue->throttle = 0; local_irq_enable(); return 0; } @@ -1976,7 +1960,7 @@ static int softnet_seq_show(struct seq_file *seq, void *v) struct netif_rx_stats *s = v; seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x\n", - s->total, s->dropped, s->time_squeeze, s->throttled, + s->total, s->dropped, s->time_squeeze, 0, 0, 0, 0, 0, /* was fastroute */ s->cpu_collision ); return 0; @@ -3220,7 +3204,6 @@ static int __init net_dev_init(void) queue = &per_cpu(softnet_data, i); skb_queue_head_init(&queue->input_pkt_queue); - queue->throttle = 0; queue->completion_queue = NULL; INIT_LIST_HEAD(&queue->poll_list); set_bit(__LINK_STATE_START, &queue->backlog_dev.state); -- cgit v1.2.3 From 51b0bdedb8e784d0d969a6b77151911130812400 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:14:40 -0700 Subject: [NET]: Separate two usages of netdev_max_backlog. Separate out the two uses of netdev_max_backlog. One controls the upper bound on packets processed per softirq, the new name for this is netdev_budget; the other controls the limit on packets queued via netif_rx. Increase the max_backlog default to account for faster processors. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/sysctl.h | 1 + net/core/dev.c | 6 +++--- net/core/sysctl_net_core.c | 9 +++++++++ 3 files changed, 13 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 72965bfe6cfb..ebfe1250f0a4 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -243,6 +243,7 @@ enum NET_CORE_MOD_CONG=16, NET_CORE_DEV_WEIGHT=17, NET_CORE_SOMAXCONN=18, + NET_CORE_BUDGET=19, }; /* /proc/sys/net/ethernet */ diff --git a/net/core/dev.c b/net/core/dev.c index 1a64508e527f..7016e0c36b3d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1346,7 +1346,8 @@ out: Receiver routines =======================================================================*/ -int netdev_max_backlog = 300; +int netdev_max_backlog = 1000; +int netdev_budget = 300; int weight_p = 64; /* old backlog weight */ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; @@ -1695,8 +1696,7 @@ static void net_rx_action(struct softirq_action *h) { struct softnet_data *queue = &__get_cpu_var(softnet_data); unsigned long start_time = jiffies; - int budget = netdev_max_backlog; - + int budget = netdev_budget; local_irq_disable(); diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index fff63643a35c..8f817ad9f546 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -13,6 +13,7 @@ #ifdef CONFIG_SYSCTL extern int netdev_max_backlog; +extern int netdev_budget; extern int weight_p; extern int net_msg_cost; extern int net_msg_burst; @@ -124,6 +125,14 @@ ctl_table core_table[] = { .mode = 0644, .proc_handler = &proc_dointvec }, + { + .ctl_name = NET_CORE_BUDGET, + .procname = "netdev_budget", + .data = &netdev_budget, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, { .ctl_name = 0 } }; -- cgit v1.2.3 From 5f8ef48d240963093451bcf83df89f1a1364f51d Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:37:36 -0700 Subject: [TCP]: Allow choosing TCP congestion control via sockopt. Allow using setsockopt to set TCP congestion control to use on a per socket basis. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/tcp.h | 1 + include/net/tcp.h | 3 ++- net/ipv4/tcp.c | 31 ++++++++++++++++++++++++++++++- net/ipv4/tcp_cong.c | 46 ++++++++++++++++++++++++++++++++++++++++++++-- net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 6 files changed, 79 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 3ea75dd6640a..dfd93d03f5d2 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -127,6 +127,7 @@ enum { #define TCP_WINDOW_CLAMP 10 /* Bound advertised window */ #define TCP_INFO 11 /* Information about this connection. */ #define TCP_QUICKACK 12 /* Block/reenable quick acks */ +#define TCP_CONGESTION 13 /* Congestion control algorithm */ #define TCPI_OPT_TIMESTAMPS 1 #define TCPI_OPT_SACK 2 diff --git a/include/net/tcp.h b/include/net/tcp.h index e427cf35915c..d04b21188ccb 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1162,8 +1162,9 @@ extern void tcp_init_congestion_control(struct tcp_sock *tp); extern void tcp_cleanup_congestion_control(struct tcp_sock *tp); extern int tcp_set_default_congestion_control(const char *name); extern void tcp_get_default_congestion_control(char *name); +extern int tcp_set_congestion_control(struct tcp_sock *tp, const char *name); -extern struct tcp_congestion_ops tcp_reno; +extern struct tcp_congestion_ops tcp_init_congestion_ops; extern u32 tcp_reno_ssthresh(struct tcp_sock *tp); extern void tcp_reno_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt, u32 in_flight, int flag); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index f3dbc8dc1263..882436da9a3a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1927,6 +1927,25 @@ int tcp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, return tp->af_specific->setsockopt(sk, level, optname, optval, optlen); + /* This is a string value all the others are int's */ + if (optname == TCP_CONGESTION) { + char name[TCP_CA_NAME_MAX]; + + if (optlen < 1) + return -EINVAL; + + val = strncpy_from_user(name, optval, + min(TCP_CA_NAME_MAX-1, optlen)); + if (val < 0) + return -EFAULT; + name[val] = 0; + + lock_sock(sk); + err = tcp_set_congestion_control(tp, name); + release_sock(sk); + return err; + } + if (optlen < sizeof(int)) return -EINVAL; @@ -2211,6 +2230,16 @@ int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, case TCP_QUICKACK: val = !tp->ack.pingpong; break; + + case TCP_CONGESTION: + if (get_user(len, optlen)) + return -EFAULT; + len = min_t(unsigned int, len, TCP_CA_NAME_MAX); + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, tp->ca_ops->name, len)) + return -EFAULT; + return 0; default: return -ENOPROTOOPT; }; @@ -2224,7 +2253,7 @@ int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, extern void __skb_cb_too_small_for_tcp(int, int); -extern void tcpdiag_init(void); +extern struct tcp_congestion_ops tcp_reno; static __initdata unsigned long thash_entries; static int __init set_thash_entries(char *str) diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 665394a63ae4..4970d10a7785 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -21,7 +21,7 @@ static struct tcp_congestion_ops *tcp_ca_find(const char *name) { struct tcp_congestion_ops *e; - list_for_each_entry(e, &tcp_cong_list, list) { + list_for_each_entry_rcu(e, &tcp_cong_list, list) { if (strcmp(e->name, name) == 0) return e; } @@ -77,6 +77,9 @@ void tcp_init_congestion_control(struct tcp_sock *tp) { struct tcp_congestion_ops *ca; + if (tp->ca_ops != &tcp_init_congestion_ops) + return; + rcu_read_lock(); list_for_each_entry_rcu(ca, &tcp_cong_list, list) { if (try_module_get(ca->owner)) { @@ -139,6 +142,34 @@ void tcp_get_default_congestion_control(char *name) rcu_read_unlock(); } +/* Change congestion control for socket */ +int tcp_set_congestion_control(struct tcp_sock *tp, const char *name) +{ + struct tcp_congestion_ops *ca; + int err = 0; + + rcu_read_lock(); + ca = tcp_ca_find(name); + if (ca == tp->ca_ops) + goto out; + + if (!ca) + err = -ENOENT; + + else if (!try_module_get(ca->owner)) + err = -EBUSY; + + else { + tcp_cleanup_congestion_control(tp); + tp->ca_ops = ca; + if (tp->ca_ops->init) + tp->ca_ops->init(tp); + } + out: + rcu_read_unlock(); + return err; +} + /* * TCP Reno congestion control * This is special case used for fallback as well. @@ -192,4 +223,15 @@ struct tcp_congestion_ops tcp_reno = { .min_cwnd = tcp_reno_min_cwnd, }; -EXPORT_SYMBOL_GPL(tcp_reno); +/* Initial congestion control used (until SYN) + * really reno under another name so we can tell difference + * during tcp_set_default_congestion_control + */ +struct tcp_congestion_ops tcp_init_congestion_ops = { + .name = "", + .owner = THIS_MODULE, + .ssthresh = tcp_reno_ssthresh, + .cong_avoid = tcp_reno_cong_avoid, + .min_cwnd = tcp_reno_min_cwnd, +}; +EXPORT_SYMBOL_GPL(tcp_init_congestion_ops); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 9122814c13ad..ebf112347a97 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2048,7 +2048,7 @@ static int tcp_v4_init_sock(struct sock *sk) tp->mss_cache_std = tp->mss_cache = 536; tp->reordering = sysctl_tcp_reordering; - tp->ca_ops = &tcp_reno; + tp->ca_ops = &tcp_init_congestion_ops; sk->sk_state = TCP_CLOSE; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index fce56039b0e9..9dac7fdf4726 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2025,7 +2025,7 @@ static int tcp_v6_init_sock(struct sock *sk) sk->sk_state = TCP_CLOSE; tp->af_specific = &ipv6_specific; - tp->ca_ops = &tcp_reno; + tp->ca_ops = &tcp_init_congestion_ops; sk->sk_write_space = sk_stream_write_space; sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); -- cgit v1.2.3 From 2de4ff7bd658c97fb357efa3095a509674dacb5a Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 20:49:30 -0700 Subject: [LIB]: Textsearch infrastructure. The textsearch infrastructure provides text searching facitilies for both linear and non-linear data. Individual search algorithms are implemented in modules and chosen by the user. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/textsearch.h | 180 +++++++++++++++++++++++++ lib/Kconfig | 8 +- lib/Makefile | 2 + lib/textsearch.c | 317 +++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 506 insertions(+), 1 deletion(-) create mode 100644 include/linux/textsearch.h create mode 100644 lib/textsearch.c (limited to 'include/linux') diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h new file mode 100644 index 000000000000..941f45ac117a --- /dev/null +++ b/include/linux/textsearch.h @@ -0,0 +1,180 @@ +#ifndef __LINUX_TEXTSEARCH_H +#define __LINUX_TEXTSEARCH_H + +#ifdef __KERNEL__ + +#include +#include +#include +#include +#include + +struct ts_config; + +/** + * TS_AUTOLOAD - Automatically load textsearch modules when needed + */ +#define TS_AUTOLOAD 1 + +/** + * struct ts_state - search state + * @offset: offset for next match + * @cb: control buffer, for persistant variables of get_next_block() + */ +struct ts_state +{ + unsigned int offset; + char cb[40]; +}; + +/** + * struct ts_ops - search module operations + * @name: name of search algorithm + * @init: initialization function to prepare a search + * @find: find the next occurrence of the pattern + * @destroy: destroy algorithm specific parts of a search configuration + * @get_pattern: return head of pattern + * @get_pattern_len: return length of pattern + * @owner: module reference to algorithm + */ +struct ts_ops +{ + const char *name; + struct ts_config * (*init)(const void *, unsigned int, int); + unsigned int (*find)(struct ts_config *, + struct ts_state *); + void (*destroy)(struct ts_config *); + void * (*get_pattern)(struct ts_config *); + unsigned int (*get_pattern_len)(struct ts_config *); + struct module *owner; + struct list_head list; +}; + +/** + * struct ts_config - search configuration + * @ops: operations of chosen algorithm + * @get_next_block: callback to fetch the next block to search in + * @finish: callback to finalize a search + */ +struct ts_config +{ + struct ts_ops *ops; + + /** + * get_next_block - fetch next block of data + * @consumed: number of bytes consumed by the caller + * @dst: destination buffer + * @conf: search configuration + * @state: search state + * + * Called repeatedly until 0 is returned. Must assign the + * head of the next block of data to &*dst and return the length + * of the block or 0 if at the end. consumed == 0 indicates + * a new search. May store/read persistant values in state->cb. + */ + unsigned int (*get_next_block)(unsigned int consumed, + const u8 **dst, + struct ts_config *conf, + struct ts_state *state); + + /** + * finish - finalize/clean a series of get_next_block() calls + * @conf: search configuration + * @state: search state + * + * Called after the last use of get_next_block(), may be used + * to cleanup any leftovers. + */ + void (*finish)(struct ts_config *conf, + struct ts_state *state); +}; + +/** + * textsearch_next - continue searching for a pattern + * @conf: search configuration + * @state: search state + * + * Continues a search looking for more occurrences of the pattern. + * textsearch_find() must be called to find the first occurrence + * in order to reset the state. + * + * Returns the position of the next occurrence of the pattern or + * UINT_MAX if not match was found. + */ +static inline unsigned int textsearch_next(struct ts_config *conf, + struct ts_state *state) +{ + unsigned int ret = conf->ops->find(conf, state); + + if (conf->finish) + conf->finish(conf, state); + + return ret; +} + +/** + * textsearch_find - start searching for a pattern + * @conf: search configuration + * @state: search state + * + * Returns the position of first occurrence of the pattern or + * UINT_MAX if no match was found. + */ +static inline unsigned int textsearch_find(struct ts_config *conf, + struct ts_state *state) +{ + state->offset = 0; + return textsearch_next(conf, state); +} + +/** + * textsearch_get_pattern - return head of the pattern + * @conf: search configuration + */ +static inline void *textsearch_get_pattern(struct ts_config *conf) +{ + return conf->ops->get_pattern(conf); +} + +/** + * textsearch_get_pattern_len - return length of the pattern + * @conf: search configuration + */ +static inline unsigned int textsearch_get_pattern_len(struct ts_config *conf) +{ + return conf->ops->get_pattern_len(conf); +} + +extern int textsearch_register(struct ts_ops *); +extern int textsearch_unregister(struct ts_ops *); +extern struct ts_config *textsearch_prepare(const char *, const void *, + unsigned int, int, int); +extern void textsearch_destroy(struct ts_config *conf); +extern unsigned int textsearch_find_continuous(struct ts_config *, + struct ts_state *, + const void *, unsigned int); + + +#define TS_PRIV_ALIGNTO 8 +#define TS_PRIV_ALIGN(len) (((len) + TS_PRIV_ALIGNTO-1) & ~(TS_PRIV_ALIGNTO-1)) + +static inline struct ts_config *alloc_ts_config(size_t payload, int gfp_mask) +{ + struct ts_config *conf; + + conf = kmalloc(TS_PRIV_ALIGN(sizeof(*conf)) + payload, gfp_mask); + if (conf == NULL) + return ERR_PTR(-ENOMEM); + + memset(conf, 0, TS_PRIV_ALIGN(sizeof(*conf)) + payload); + return conf; +} + +static inline void *ts_config_priv(struct ts_config *conf) +{ + return ((u8 *) conf + TS_PRIV_ALIGN(sizeof(struct ts_config))); +} + +#endif /* __KERNEL__ */ + +#endif diff --git a/lib/Kconfig b/lib/Kconfig index 2d4d4e3bc4aa..5bc2d523e6d1 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -63,5 +63,11 @@ config REED_SOLOMON_ENC16 config REED_SOLOMON_DEC16 boolean -endmenu +config TEXTSEARCH + boolean "Textsearch infrastructure" + default y + help + Say Y here if you want to provide a textsearch infrastructure + to other subsystems. +endmenu diff --git a/lib/Makefile b/lib/Makefile index dcb4231916e2..3e917436ad60 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -36,6 +36,8 @@ obj-$(CONFIG_ZLIB_INFLATE) += zlib_inflate/ obj-$(CONFIG_ZLIB_DEFLATE) += zlib_deflate/ obj-$(CONFIG_REED_SOLOMON) += reed_solomon/ +lib-$(CONFIG_TEXTSEARCH) += textsearch.o + hostprogs-y := gen_crc32table clean-files := crc32table.h diff --git a/lib/textsearch.c b/lib/textsearch.c new file mode 100644 index 000000000000..1e934c196f0f --- /dev/null +++ b/lib/textsearch.c @@ -0,0 +1,317 @@ +/* + * lib/textsearch.c Generic text search interface + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Authors: Thomas Graf + * Pablo Neira Ayuso + * + * ========================================================================== + * + * INTRODUCTION + * + * The textsearch infrastructure provides text searching facitilies for + * both linear and non-linear data. Individual search algorithms are + * implemented in modules and chosen by the user. + * + * ARCHITECTURE + * + * User + * +----------------+ + * | finish()|<--------------(6)-----------------+ + * |get_next_block()|<--------------(5)---------------+ | + * | | Algorithm | | + * | | +------------------------------+ + * | | | init() find() destroy() | + * | | +------------------------------+ + * | | Core API ^ ^ ^ + * | | +---------------+ (2) (4) (8) + * | (1)|----->| prepare() |---+ | | + * | (3)|----->| find()/next() |-----------+ | + * | (7)|----->| destroy() |----------------------+ + * +----------------+ +---------------+ + * + * (1) User configures a search by calling _prepare() specifying the + * search parameters such as the pattern and algorithm name. + * (2) Core requests the algorithm to allocate and initialize a search + * configuration according to the specified parameters. + * (3) User starts the search(es) by calling _find() or _next() to + * fetch subsequent occurrences. A state variable is provided + * to the algorihtm to store persistant variables. + * (4) Core eventually resets the search offset and forwards the find() + * request to the algorithm. + * (5) Algorithm calls get_next_block() provided by the user continously + * to fetch the data to be searched in block by block. + * (6) Algorithm invokes finish() after the last call to get_next_block + * to clean up any leftovers from get_next_block. (Optional) + * (7) User destroys the configuration by calling _destroy(). + * (8) Core notifies the algorithm to destroy algorithm specific + * allocations. (Optional) + * + * USAGE + * + * Before a search can be performed, a configuration must be created + * by calling textsearch_prepare() specyfing the searching algorithm and + * the pattern to look for. The returned configuration may then be used + * for an arbitary amount of times and even in parallel as long as a + * separate struct ts_state variable is provided to every instance. + * + * The actual search is performed by either calling textsearch_find_- + * continuous() for linear data or by providing an own get_next_block() + * implementation and calling textsearch_find(). Both functions return + * the position of the first occurrence of the patern or UINT_MAX if + * no match was found. Subsequent occurences can be found by calling + * textsearch_next() regardless of the linearity of the data. + * + * Once you're done using a configuration it must be given back via + * textsearch_destroy. + * + * EXAMPLE + * + * int pos; + * struct ts_config *conf; + * struct ts_state state; + * const char *pattern = "chicken"; + * const char *example = "We dance the funky chicken"; + * + * conf = textsearch_prepare("kmp", pattern, strlen(pattern), + * GFP_KERNEL, TS_AUTOLOAD); + * if (IS_ERR(conf)) { + * err = PTR_ERR(conf); + * goto errout; + * } + * + * pos = textsearch_find_continuous(conf, &state, example, strlen(example)); + * if (pos != UINT_MAX) + * panic("Oh my god, dancing chickens at %d\n", pos); + * + * textsearch_destroy(conf); + * + * ========================================================================== + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +static LIST_HEAD(ts_ops); +static DEFINE_SPINLOCK(ts_mod_lock); + +static inline struct ts_ops *lookup_ts_algo(const char *name) +{ + struct ts_ops *o; + + rcu_read_lock(); + list_for_each_entry_rcu(o, &ts_ops, list) { + if (!strcmp(name, o->name)) { + if (!try_module_get(o->owner)) + o = NULL; + rcu_read_unlock(); + return o; + } + } + rcu_read_unlock(); + + return NULL; +} + +/** + * textsearch_register - register a textsearch module + * @ops: operations lookup table + * + * This function must be called by textsearch modules to announce + * their presence. The specified &@ops must have %name set to a + * unique identifier and the callbacks find(), init(), get_pattern(), + * and get_pattern_len() must be implemented. + * + * Returns 0 or -EEXISTS if another module has already registered + * with same name. + */ +int textsearch_register(struct ts_ops *ops) +{ + int err = -EEXIST; + struct ts_ops *o; + + if (ops->name == NULL || ops->find == NULL || ops->init == NULL || + ops->get_pattern == NULL || ops->get_pattern_len == NULL) + return -EINVAL; + + spin_lock(&ts_mod_lock); + list_for_each_entry(o, &ts_ops, list) { + if (!strcmp(ops->name, o->name)) + goto errout; + } + + list_add_tail_rcu(&ops->list, &ts_ops); + err = 0; +errout: + spin_unlock(&ts_mod_lock); + return err; +} + +/** + * textsearch_unregister - unregister a textsearch module + * @ops: operations lookup table + * + * This function must be called by textsearch modules to announce + * their disappearance for examples when the module gets unloaded. + * The &ops parameter must be the same as the one during the + * registration. + * + * Returns 0 on success or -ENOENT if no matching textsearch + * registration was found. + */ +int textsearch_unregister(struct ts_ops *ops) +{ + int err = 0; + struct ts_ops *o; + + spin_lock(&ts_mod_lock); + list_for_each_entry(o, &ts_ops, list) { + if (o == ops) { + list_del_rcu(&o->list); + goto out; + } + } + + err = -ENOENT; +out: + spin_unlock(&ts_mod_lock); + return err; +} + +struct ts_linear_state +{ + unsigned int len; + const void *data; +}; + +static unsigned int get_linear_data(unsigned int consumed, const u8 **dst, + struct ts_config *conf, + struct ts_state *state) +{ + struct ts_linear_state *st = (struct ts_linear_state *) state->cb; + + if (likely(consumed < st->len)) { + *dst = st->data + consumed; + return st->len - consumed; + } + + return 0; +} + +/** + * textsearch_find_continuous - search a pattern in continuous/linear data + * @conf: search configuration + * @state: search state + * @data: data to search in + * @len: length of data + * + * A simplified version of textsearch_find() for continuous/linear data. + * Call textsearch_next() to retrieve subsequent matches. + * + * Returns the position of first occurrence of the pattern or + * UINT_MAX if no occurrence was found. + */ +unsigned int textsearch_find_continuous(struct ts_config *conf, + struct ts_state *state, + const void *data, unsigned int len) +{ + struct ts_linear_state *st = (struct ts_linear_state *) state->cb; + + conf->get_next_block = get_linear_data; + st->data = data; + st->len = len; + + return textsearch_find(conf, state); +} + +/** + * textsearch_prepare - Prepare a search + * @algo: name of search algorithm + * @pattern: pattern data + * @len: length of pattern + * @gfp_mask: allocation mask + * @flags: search flags + * + * Looks up the search algorithm module and creates a new textsearch + * configuration for the specified pattern. Upon completion all + * necessary refcnts are held and the configuration must be put back + * using textsearch_put() after usage. + * + * Note: The format of the pattern may not be compatible between + * the various search algorithms. + * + * Returns a new textsearch configuration according to the specified + * parameters or a ERR_PTR(). + */ +struct ts_config *textsearch_prepare(const char *algo, const void *pattern, + unsigned int len, int gfp_mask, int flags) +{ + int err = -ENOENT; + struct ts_config *conf; + struct ts_ops *ops; + + ops = lookup_ts_algo(algo); +#ifdef CONFIG_KMOD + /* + * Why not always autoload you may ask. Some users are + * in a situation where requesting a module may deadlock, + * especially when the module is located on a NFS mount. + */ + if (ops == NULL && flags & TS_AUTOLOAD) { + request_module("ts_%s", algo); + ops = lookup_ts_algo(algo); + } +#endif + + if (ops == NULL) + goto errout; + + conf = ops->init(pattern, len, gfp_mask); + if (IS_ERR(conf)) { + err = PTR_ERR(conf); + goto errout; + } + + conf->ops = ops; + return conf; + +errout: + if (ops) + module_put(ops->owner); + + return ERR_PTR(err); +} + +/** + * textsearch_destroy - destroy a search configuration + * @conf: search configuration + * + * Releases all references of the configuration and frees + * up the memory. + */ +void textsearch_destroy(struct ts_config *conf) +{ + if (conf->ops) { + if (conf->ops->destroy) + conf->ops->destroy(conf); + module_put(conf->ops->owner); + } + + kfree(conf); +} + +EXPORT_SYMBOL(textsearch_register); +EXPORT_SYMBOL(textsearch_unregister); +EXPORT_SYMBOL(textsearch_prepare); +EXPORT_SYMBOL(textsearch_find_continuous); +EXPORT_SYMBOL(textsearch_destroy); -- cgit v1.2.3 From 6408f79cce401e1bfecf923e7156f84f96e021e3 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 20:59:16 -0700 Subject: [LIB]: Naive finite state machine based textsearch A finite state machine consists of n states (struct ts_fsm_token) representing the pattern as a finite automation. The data is read sequentially on a octet basis. Every state token specifies the number of recurrences and the type of value accepted which can be either a specific character or ctype based set of characters. The available type of recurrences include 1, (0|1), [0 n], and [1 n]. The algorithm differs between strict/non-strict mode specyfing whether the pattern has to start at the first octect. Strict mode is enabled by default and can be disabled by inserting TS_FSM_HEAD_IGNORE as the first token in the chain. The runtime performance of the algorithm should be around O(n), however while in strict mode the average runtime can be better. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/textsearch_fsm.h | 48 ++++++ lib/Kconfig | 11 ++ lib/Makefile | 1 + lib/ts_fsm.c | 338 +++++++++++++++++++++++++++++++++++++++++ 4 files changed, 398 insertions(+) create mode 100644 include/linux/textsearch_fsm.h create mode 100644 lib/ts_fsm.c (limited to 'include/linux') diff --git a/include/linux/textsearch_fsm.h b/include/linux/textsearch_fsm.h new file mode 100644 index 000000000000..fdfa078c66e5 --- /dev/null +++ b/include/linux/textsearch_fsm.h @@ -0,0 +1,48 @@ +#ifndef __LINUX_TEXTSEARCH_FSM_H +#define __LINUX_TEXTSEARCH_FSM_H + +#include + +enum { + TS_FSM_SPECIFIC, /* specific character */ + TS_FSM_WILDCARD, /* any character */ + TS_FSM_DIGIT, /* isdigit() */ + TS_FSM_XDIGIT, /* isxdigit() */ + TS_FSM_PRINT, /* isprint() */ + TS_FSM_ALPHA, /* isalpha() */ + TS_FSM_ALNUM, /* isalnum() */ + TS_FSM_ASCII, /* isascii() */ + TS_FSM_CNTRL, /* iscntrl() */ + TS_FSM_GRAPH, /* isgraph() */ + TS_FSM_LOWER, /* islower() */ + TS_FSM_UPPER, /* isupper() */ + TS_FSM_PUNCT, /* ispunct() */ + TS_FSM_SPACE, /* isspace() */ + __TS_FSM_TYPE_MAX, +}; +#define TS_FSM_TYPE_MAX (__TS_FSM_TYPE_MAX - 1) + +enum { + TS_FSM_SINGLE, /* 1 occurrence */ + TS_FSM_PERHAPS, /* 1 or 0 occurrence */ + TS_FSM_ANY, /* 0..n occurrences */ + TS_FSM_MULTI, /* 1..n occurrences */ + TS_FSM_HEAD_IGNORE, /* 0..n ignored occurrences at head */ + __TS_FSM_RECUR_MAX, +}; +#define TS_FSM_RECUR_MAX (__TS_FSM_RECUR_MAX - 1) + +/** + * struct ts_fsm_token - state machine token (state) + * @type: type of token + * @recur: number of recurrences + * @value: character value for TS_FSM_SPECIFIC + */ +struct ts_fsm_token +{ + __u16 type; + __u8 recur; + __u8 value; +}; + +#endif diff --git a/lib/Kconfig b/lib/Kconfig index 16b8fa2175e4..455833a9e31a 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -80,4 +80,15 @@ config TEXTSEARCH_KMP To compile this code as a module, choose M here: the module will be called ts_kmp. +config TEXTSEARCH_FSM + depends on TEXTSEARCH + tristate "Finite state machine" + help + Say Y here if you want to be able to search text using a + naive finite state machine approach implementing a subset + of regular expressions. + + To compile this code as a module, choose M here: the + module will be called ts_fsm. + endmenu diff --git a/lib/Makefile b/lib/Makefile index 6cdb10f312df..7f6eda449102 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -38,6 +38,7 @@ obj-$(CONFIG_REED_SOLOMON) += reed_solomon/ lib-$(CONFIG_TEXTSEARCH) += textsearch.o obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o +obj-$(CONFIG_TEXTSEARCH_FSM) += ts_fsm.o hostprogs-y := gen_crc32table clean-files := crc32table.h diff --git a/lib/ts_fsm.c b/lib/ts_fsm.c new file mode 100644 index 000000000000..d27c0a072940 --- /dev/null +++ b/lib/ts_fsm.c @@ -0,0 +1,338 @@ +/* + * lib/ts_fsm.c A naive finite state machine text search approach + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Authors: Thomas Graf + * + * ========================================================================== + * + * A finite state machine consists of n states (struct ts_fsm_token) + * representing the pattern as a finite automation. The data is read + * sequentially on a octet basis. Every state token specifies the number + * of recurrences and the type of value accepted which can be either a + * specific character or ctype based set of characters. The available + * type of recurrences include 1, (0|1), [0 n], and [1 n]. + * + * The algorithm differs between strict/non-strict mode specyfing + * whether the pattern has to start at the first octect. Strict mode + * is enabled by default and can be disabled by inserting + * TS_FSM_HEAD_IGNORE as the first token in the chain. + * + * The runtime performance of the algorithm should be around O(n), + * however while in strict mode the average runtime can be better. + */ + +#include +#include +#include +#include +#include +#include +#include + +struct ts_fsm +{ + unsigned int ntokens; + struct ts_fsm_token tokens[0]; +}; + +/* other values derived from ctype.h */ +#define _A 0x100 /* ascii */ +#define _W 0x200 /* wildcard */ + +/* Map to _ctype flags and some magic numbers */ +static u16 token_map[TS_FSM_TYPE_MAX+1] = { + [TS_FSM_SPECIFIC] = 0, + [TS_FSM_WILDCARD] = _W, + [TS_FSM_CNTRL] = _C, + [TS_FSM_LOWER] = _L, + [TS_FSM_UPPER] = _U, + [TS_FSM_PUNCT] = _P, + [TS_FSM_SPACE] = _S, + [TS_FSM_DIGIT] = _D, + [TS_FSM_XDIGIT] = _D | _X, + [TS_FSM_ALPHA] = _U | _L, + [TS_FSM_ALNUM] = _U | _L | _D, + [TS_FSM_PRINT] = _P | _U | _L | _D | _SP, + [TS_FSM_GRAPH] = _P | _U | _L | _D, + [TS_FSM_ASCII] = _A, +}; + +static u16 token_lookup_tbl[256] = { +_W|_A|_C, _W|_A|_C, _W|_A|_C, _W|_A|_C, /* 0- 3 */ +_W|_A|_C, _W|_A|_C, _W|_A|_C, _W|_A|_C, /* 4- 7 */ +_W|_A|_C, _W|_A|_C|_S, _W|_A|_C|_S, _W|_A|_C|_S, /* 8- 11 */ +_W|_A|_C|_S, _W|_A|_C|_S, _W|_A|_C, _W|_A|_C, /* 12- 15 */ +_W|_A|_C, _W|_A|_C, _W|_A|_C, _W|_A|_C, /* 16- 19 */ +_W|_A|_C, _W|_A|_C, _W|_A|_C, _W|_A|_C, /* 20- 23 */ +_W|_A|_C, _W|_A|_C, _W|_A|_C, _W|_A|_C, /* 24- 27 */ +_W|_A|_C, _W|_A|_C, _W|_A|_C, _W|_A|_C, /* 28- 31 */ +_W|_A|_S|_SP, _W|_A|_P, _W|_A|_P, _W|_A|_P, /* 32- 35 */ +_W|_A|_P, _W|_A|_P, _W|_A|_P, _W|_A|_P, /* 36- 39 */ +_W|_A|_P, _W|_A|_P, _W|_A|_P, _W|_A|_P, /* 40- 43 */ +_W|_A|_P, _W|_A|_P, _W|_A|_P, _W|_A|_P, /* 44- 47 */ +_W|_A|_D, _W|_A|_D, _W|_A|_D, _W|_A|_D, /* 48- 51 */ +_W|_A|_D, _W|_A|_D, _W|_A|_D, _W|_A|_D, /* 52- 55 */ +_W|_A|_D, _W|_A|_D, _W|_A|_P, _W|_A|_P, /* 56- 59 */ +_W|_A|_P, _W|_A|_P, _W|_A|_P, _W|_A|_P, /* 60- 63 */ +_W|_A|_P, _W|_A|_U|_X, _W|_A|_U|_X, _W|_A|_U|_X, /* 64- 67 */ +_W|_A|_U|_X, _W|_A|_U|_X, _W|_A|_U|_X, _W|_A|_U, /* 68- 71 */ +_W|_A|_U, _W|_A|_U, _W|_A|_U, _W|_A|_U, /* 72- 75 */ +_W|_A|_U, _W|_A|_U, _W|_A|_U, _W|_A|_U, /* 76- 79 */ +_W|_A|_U, _W|_A|_U, _W|_A|_U, _W|_A|_U, /* 80- 83 */ +_W|_A|_U, _W|_A|_U, _W|_A|_U, _W|_A|_U, /* 84- 87 */ +_W|_A|_U, _W|_A|_U, _W|_A|_U, _W|_A|_P, /* 88- 91 */ +_W|_A|_P, _W|_A|_P, _W|_A|_P, _W|_A|_P, /* 92- 95 */ +_W|_A|_P, _W|_A|_L|_X, _W|_A|_L|_X, _W|_A|_L|_X, /* 96- 99 */ +_W|_A|_L|_X, _W|_A|_L|_X, _W|_A|_L|_X, _W|_A|_L, /* 100-103 */ +_W|_A|_L, _W|_A|_L, _W|_A|_L, _W|_A|_L, /* 104-107 */ +_W|_A|_L, _W|_A|_L, _W|_A|_L, _W|_A|_L, /* 108-111 */ +_W|_A|_L, _W|_A|_L, _W|_A|_L, _W|_A|_L, /* 112-115 */ +_W|_A|_L, _W|_A|_L, _W|_A|_L, _W|_A|_L, /* 116-119 */ +_W|_A|_L, _W|_A|_L, _W|_A|_L, _W|_A|_P, /* 120-123 */ +_W|_A|_P, _W|_A|_P, _W|_A|_P, _W|_A|_C, /* 124-127 */ +_W, _W, _W, _W, /* 128-131 */ +_W, _W, _W, _W, /* 132-135 */ +_W, _W, _W, _W, /* 136-139 */ +_W, _W, _W, _W, /* 140-143 */ +_W, _W, _W, _W, /* 144-147 */ +_W, _W, _W, _W, /* 148-151 */ +_W, _W, _W, _W, /* 152-155 */ +_W, _W, _W, _W, /* 156-159 */ +_W|_S|_SP, _W|_P, _W|_P, _W|_P, /* 160-163 */ +_W|_P, _W|_P, _W|_P, _W|_P, /* 164-167 */ +_W|_P, _W|_P, _W|_P, _W|_P, /* 168-171 */ +_W|_P, _W|_P, _W|_P, _W|_P, /* 172-175 */ +_W|_P, _W|_P, _W|_P, _W|_P, /* 176-179 */ +_W|_P, _W|_P, _W|_P, _W|_P, /* 180-183 */ +_W|_P, _W|_P, _W|_P, _W|_P, /* 184-187 */ +_W|_P, _W|_P, _W|_P, _W|_P, /* 188-191 */ +_W|_U, _W|_U, _W|_U, _W|_U, /* 192-195 */ +_W|_U, _W|_U, _W|_U, _W|_U, /* 196-199 */ +_W|_U, _W|_U, _W|_U, _W|_U, /* 200-203 */ +_W|_U, _W|_U, _W|_U, _W|_U, /* 204-207 */ +_W|_U, _W|_U, _W|_U, _W|_U, /* 208-211 */ +_W|_U, _W|_U, _W|_U, _W|_P, /* 212-215 */ +_W|_U, _W|_U, _W|_U, _W|_U, /* 216-219 */ +_W|_U, _W|_U, _W|_U, _W|_L, /* 220-223 */ +_W|_L, _W|_L, _W|_L, _W|_L, /* 224-227 */ +_W|_L, _W|_L, _W|_L, _W|_L, /* 228-231 */ +_W|_L, _W|_L, _W|_L, _W|_L, /* 232-235 */ +_W|_L, _W|_L, _W|_L, _W|_L, /* 236-239 */ +_W|_L, _W|_L, _W|_L, _W|_L, /* 240-243 */ +_W|_L, _W|_L, _W|_L, _W|_P, /* 244-247 */ +_W|_L, _W|_L, _W|_L, _W|_L, /* 248-251 */ +_W|_L, _W|_L, _W|_L, _W|_L}; /* 252-255 */ + +static inline int match_token(struct ts_fsm_token *t, u8 d) +{ + if (t->type) + return (token_lookup_tbl[d] & t->type) != 0; + else + return t->value == d; +} + +static unsigned int fsm_find(struct ts_config *conf, struct ts_state *state) +{ + struct ts_fsm *fsm = ts_config_priv(conf); + struct ts_fsm_token *cur = NULL, *next; + unsigned int match_start, block_idx = 0, tok_idx; + unsigned block_len = 0, strict, consumed = state->offset; + const u8 *data; + +#define GET_NEXT_BLOCK() \ +({ consumed += block_idx; \ + block_idx = 0; \ + block_len = conf->get_next_block(consumed, &data, conf, state); }) + +#define TOKEN_MISMATCH() \ + do { \ + if (strict) \ + goto no_match; \ + block_idx++; \ + goto startover; \ + } while(0) + +#define end_of_data() unlikely(block_idx >= block_len && !GET_NEXT_BLOCK()) + + if (end_of_data()) + goto no_match; + + strict = fsm->tokens[0].recur != TS_FSM_HEAD_IGNORE; + +startover: + match_start = consumed + block_idx; + + for (tok_idx = 0; tok_idx < fsm->ntokens; tok_idx++) { + cur = &fsm->tokens[tok_idx]; + + if (likely(tok_idx < (fsm->ntokens - 1))) + next = &fsm->tokens[tok_idx + 1]; + else + next = NULL; + + switch (cur->recur) { + case TS_FSM_SINGLE: + if (end_of_data()) + goto no_match; + + if (!match_token(cur, data[block_idx])) + TOKEN_MISMATCH(); + break; + + case TS_FSM_PERHAPS: + if (end_of_data() || + !match_token(cur, data[block_idx])) + continue; + break; + + case TS_FSM_MULTI: + if (end_of_data()) + goto no_match; + + if (!match_token(cur, data[block_idx])) + TOKEN_MISMATCH(); + + block_idx++; + /* fall through */ + + case TS_FSM_ANY: + if (next == NULL) + goto found_match; + + if (end_of_data()) + continue; + + while (!match_token(next, data[block_idx])) { + if (!match_token(cur, data[block_idx])) + TOKEN_MISMATCH(); + block_idx++; + if (end_of_data()) + goto no_match; + } + continue; + + /* + * Optimization: Prefer small local loop over jumping + * back and forth until garbage at head is munched. + */ + case TS_FSM_HEAD_IGNORE: + if (end_of_data()) + continue; + + while (!match_token(next, data[block_idx])) { + /* + * Special case, don't start over upon + * a mismatch, give the user the + * chance to specify the type of data + * allowed to be ignored. + */ + if (!match_token(cur, data[block_idx])) + goto no_match; + + block_idx++; + if (end_of_data()) + goto no_match; + } + + match_start = consumed + block_idx; + continue; + } + + block_idx++; + } + + if (end_of_data()) + goto found_match; + +no_match: + return UINT_MAX; + +found_match: + state->offset = consumed + block_idx; + return match_start; +} + +static struct ts_config *fsm_init(const void *pattern, unsigned int len, + int gfp_mask) +{ + int i, err = -EINVAL; + struct ts_config *conf; + struct ts_fsm *fsm; + struct ts_fsm_token *tokens = (struct ts_fsm_token *) pattern; + unsigned int ntokens = len / sizeof(*tokens); + size_t priv_size = sizeof(*fsm) + len; + + if (len % sizeof(struct ts_fsm_token) || ntokens < 1) + goto errout; + + for (i = 0; i < ntokens; i++) { + struct ts_fsm_token *t = &tokens[i]; + + if (t->type > TS_FSM_TYPE_MAX || t->recur > TS_FSM_RECUR_MAX) + goto errout; + + if (t->recur == TS_FSM_HEAD_IGNORE && + (i != 0 || i == (ntokens - 1))) + goto errout; + } + + conf = alloc_ts_config(priv_size, gfp_mask); + if (IS_ERR(conf)) + return conf; + + fsm = ts_config_priv(conf); + fsm->ntokens = ntokens; + memcpy(fsm->tokens, pattern, len); + + for (i = 0; i < fsm->ntokens; i++) { + struct ts_fsm_token *t = &fsm->tokens[i]; + t->type = token_map[t->type]; + } + + return conf; + +errout: + return ERR_PTR(err); +} + +static void *fsm_get_pattern(struct ts_config *conf) +{ + struct ts_fsm *fsm = ts_config_priv(conf); + return fsm->tokens; +} + +static unsigned int fsm_get_pattern_len(struct ts_config *conf) +{ + struct ts_fsm *fsm = ts_config_priv(conf); + return fsm->ntokens * sizeof(struct ts_fsm_token); +} + +static struct ts_ops fsm_ops = { + .name = "fsm", + .find = fsm_find, + .init = fsm_init, + .get_pattern = fsm_get_pattern, + .get_pattern_len = fsm_get_pattern_len, + .owner = THIS_MODULE, + .list = LIST_HEAD_INIT(fsm_ops.list) +}; + +static int __init init_fsm(void) +{ + return textsearch_register(&fsm_ops); +} + +static void __exit exit_fsm(void) +{ + textsearch_unregister(&fsm_ops); +} + +MODULE_LICENSE("GPL"); + +module_init(init_fsm); +module_exit(exit_fsm); -- cgit v1.2.3 From 677e90eda3bd8cfde0b748daaa46476162a03950 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 20:59:51 -0700 Subject: [NET]: Zerocopy sequential reading of skb data Implements sequential reading for both linear and non-linear skb data at zerocopy cost. The data is returned in chunks of arbitary length, therefore random access is not possible. Usage: from := 0 to := 128 state := undef data := undef len := undef consumed := 0 skb_prepare_seq_read(skb, from, to, &state) while (len = skb_seq_read(consumed, &data, &state)) != 0 do /* do something with 'data' of length 'len' */ if abort then /* abort read if we don't wait for * skb_seq_read() to return 0 */ skb_abort_seq_read(&state) return endif /* not necessary to consume all of 'len' */ consumed += len done Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/skbuff.h | 18 ++++++++ net/core/skbuff.c | 117 +++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 135 insertions(+) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index d7c839a21842..171a37dff83a 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -321,6 +321,24 @@ extern void skb_over_panic(struct sk_buff *skb, int len, extern void skb_under_panic(struct sk_buff *skb, int len, void *here); +struct skb_seq_state +{ + __u32 lower_offset; + __u32 upper_offset; + __u32 frag_idx; + __u32 stepped_offset; + struct sk_buff *root_skb; + struct sk_buff *cur_skb; + __u8 *frag_data; +}; + +extern void skb_prepare_seq_read(struct sk_buff *skb, + unsigned int from, unsigned int to, + struct skb_seq_state *st); +extern unsigned int skb_seq_read(unsigned int consumed, const u8 **data, + struct skb_seq_state *st); +extern void skb_abort_seq_read(struct skb_seq_state *st); + /* Internal */ #define skb_shinfo(SKB) ((struct skb_shared_info *)((SKB)->end)) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 6d68c03bc051..d285f2f7e812 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1500,6 +1500,120 @@ void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len) skb_split_no_header(skb, skb1, len, pos); } +/** + * skb_prepare_seq_read - Prepare a sequential read of skb data + * @skb: the buffer to read + * @from: lower offset of data to be read + * @to: upper offset of data to be read + * @st: state variable + * + * Initializes the specified state variable. Must be called before + * invoking skb_seq_read() for the first time. + */ +void skb_prepare_seq_read(struct sk_buff *skb, unsigned int from, + unsigned int to, struct skb_seq_state *st) +{ + st->lower_offset = from; + st->upper_offset = to; + st->root_skb = st->cur_skb = skb; + st->frag_idx = st->stepped_offset = 0; + st->frag_data = NULL; +} + +/** + * skb_seq_read - Sequentially read skb data + * @consumed: number of bytes consumed by the caller so far + * @data: destination pointer for data to be returned + * @st: state variable + * + * Reads a block of skb data at &consumed relative to the + * lower offset specified to skb_prepare_seq_read(). Assigns + * the head of the data block to &data and returns the length + * of the block or 0 if the end of the skb data or the upper + * offset has been reached. + * + * The caller is not required to consume all of the data + * returned, i.e. &consumed is typically set to the number + * of bytes already consumed and the next call to + * skb_seq_read() will return the remaining part of the block. + * + * Note: The size of each block of data returned can be arbitary, + * this limitation is the cost for zerocopy seqeuental + * reads of potentially non linear data. + * + * Note: Fragment lists within fragments are not implemented + * at the moment, state->root_skb could be replaced with + * a stack for this purpose. + */ +unsigned int skb_seq_read(unsigned int consumed, const u8 **data, + struct skb_seq_state *st) +{ + unsigned int block_limit, abs_offset = consumed + st->lower_offset; + skb_frag_t *frag; + + if (unlikely(abs_offset >= st->upper_offset)) + return 0; + +next_skb: + block_limit = skb_headlen(st->cur_skb); + + if (abs_offset < block_limit) { + *data = st->cur_skb->data + abs_offset; + return block_limit - abs_offset; + } + + if (st->frag_idx == 0 && !st->frag_data) + st->stepped_offset += skb_headlen(st->cur_skb); + + while (st->frag_idx < skb_shinfo(st->cur_skb)->nr_frags) { + frag = &skb_shinfo(st->cur_skb)->frags[st->frag_idx]; + block_limit = frag->size + st->stepped_offset; + + if (abs_offset < block_limit) { + if (!st->frag_data) + st->frag_data = kmap_skb_frag(frag); + + *data = (u8 *) st->frag_data + frag->page_offset + + (abs_offset - st->stepped_offset); + + return block_limit - abs_offset; + } + + if (st->frag_data) { + kunmap_skb_frag(st->frag_data); + st->frag_data = NULL; + } + + st->frag_idx++; + st->stepped_offset += frag->size; + } + + if (st->cur_skb->next) { + st->cur_skb = st->cur_skb->next; + st->frag_idx = 0; + goto next_skb; + } else if (st->root_skb == st->cur_skb && + skb_shinfo(st->root_skb)->frag_list) { + st->cur_skb = skb_shinfo(st->root_skb)->frag_list; + goto next_skb; + } + + return 0; +} + +/** + * skb_abort_seq_read - Abort a sequential read of skb data + * @st: state variable + * + * Must be called if skb_seq_read() was not called until it + * returned 0. + */ +void skb_abort_seq_read(struct skb_seq_state *st) +{ + if (st->frag_data) + kunmap_skb_frag(st->frag_data); +} + void __init skb_init(void) { skbuff_head_cache = kmem_cache_create("skbuff_head_cache", @@ -1538,3 +1652,6 @@ EXPORT_SYMBOL(skb_queue_tail); EXPORT_SYMBOL(skb_unlink); EXPORT_SYMBOL(skb_append); EXPORT_SYMBOL(skb_split); +EXPORT_SYMBOL(skb_prepare_seq_read); +EXPORT_SYMBOL(skb_seq_read); +EXPORT_SYMBOL(skb_abort_seq_read); -- cgit v1.2.3 From 3fc7e8a6d842f72d16d2623b1022814a635ab961 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 21:00:17 -0700 Subject: [NET]: skb_find_text() - Find a text pattern in skb data Finds a pattern in the skb data according to the specified textsearch configuration. Use textsearch_next() to retrieve subsequent occurrences of the pattern. Returns the offset to the first occurrence or UINT_MAX if no match was found. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/skbuff.h | 5 +++++ net/core/skbuff.c | 40 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 45 insertions(+) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 171a37dff83a..416a2e4024b2 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -27,6 +27,7 @@ #include #include #include +#include #include #define HAVE_ALLOC_SKB /* For the drivers to know */ @@ -339,6 +340,10 @@ extern unsigned int skb_seq_read(unsigned int consumed, const u8 **data, struct skb_seq_state *st); extern void skb_abort_seq_read(struct skb_seq_state *st); +extern unsigned int skb_find_text(struct sk_buff *skb, unsigned int from, + unsigned int to, struct ts_config *config, + struct ts_state *state); + /* Internal */ #define skb_shinfo(SKB) ((struct skb_shared_info *)((SKB)->end)) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index d285f2f7e812..bb73b2190ec7 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1614,6 +1614,45 @@ void skb_abort_seq_read(struct skb_seq_state *st) kunmap_skb_frag(st->frag_data); } +#define TS_SKB_CB(state) ((struct skb_seq_state *) &((state)->cb)) + +static unsigned int skb_ts_get_next_block(unsigned int offset, const u8 **text, + struct ts_config *conf, + struct ts_state *state) +{ + return skb_seq_read(offset, text, TS_SKB_CB(state)); +} + +static void skb_ts_finish(struct ts_config *conf, struct ts_state *state) +{ + skb_abort_seq_read(TS_SKB_CB(state)); +} + +/** + * skb_find_text - Find a text pattern in skb data + * @skb: the buffer to look in + * @from: search offset + * @to: search limit + * @config: textsearch configuration + * @state: uninitialized textsearch state variable + * + * Finds a pattern in the skb data according to the specified + * textsearch configuration. Use textsearch_next() to retrieve + * subsequent occurrences of the pattern. Returns the offset + * to the first occurrence or UINT_MAX if no match was found. + */ +unsigned int skb_find_text(struct sk_buff *skb, unsigned int from, + unsigned int to, struct ts_config *config, + struct ts_state *state) +{ + config->get_next_block = skb_ts_get_next_block; + config->finish = skb_ts_finish; + + skb_prepare_seq_read(skb, from, to, TS_SKB_CB(state)); + + return textsearch_find(config, state); +} + void __init skb_init(void) { skbuff_head_cache = kmem_cache_create("skbuff_head_cache", @@ -1655,3 +1694,4 @@ EXPORT_SYMBOL(skb_split); EXPORT_SYMBOL(skb_prepare_seq_read); EXPORT_SYMBOL(skb_seq_read); EXPORT_SYMBOL(skb_abort_seq_read); +EXPORT_SYMBOL(skb_find_text); -- cgit v1.2.3 From d675c989ed2d4ba23dff615330b04371aea83534 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 21:00:58 -0700 Subject: [PKT_SCHED]: Packet classification based on textsearch (ematch) Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/pkt_cls.h | 1 + include/linux/rtnetlink.h | 7 +- include/linux/tc_ematch/tc_em_text.h | 19 +++++ net/sched/Kconfig | 11 +++ net/sched/Makefile | 1 + net/sched/em_text.c | 157 +++++++++++++++++++++++++++++++++++ 6 files changed, 194 insertions(+), 2 deletions(-) create mode 100644 include/linux/tc_ematch/tc_em_text.h create mode 100644 net/sched/em_text.c (limited to 'include/linux') diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h index d2aa214d6803..25d2d67c1faf 100644 --- a/include/linux/pkt_cls.h +++ b/include/linux/pkt_cls.h @@ -408,6 +408,7 @@ enum TCF_EM_NBYTE, TCF_EM_U32, TCF_EM_META, + TCF_EM_TEXT, __TCF_EM_MAX }; diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index e68dbf0bf579..d021888b58f1 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -892,10 +892,13 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi goto rtattr_failure; \ __rta_fill(skb, attrtype, attrlen, data); }) -#define RTA_PUT_NOHDR(skb, attrlen, data) \ +#define RTA_APPEND(skb, attrlen, data) \ ({ if (unlikely(skb_tailroom(skb) < (int)(attrlen))) \ goto rtattr_failure; \ - memcpy(skb_put(skb, RTA_ALIGN(attrlen)), data, attrlen); }) + memcpy(skb_put(skb, attrlen), data, attrlen); }) + +#define RTA_PUT_NOHDR(skb, attrlen, data) \ + RTA_APPEND(skb, RTA_ALIGN(attrlen), data) #define RTA_PUT_U8(skb, attrtype, value) \ ({ u8 _tmp = (value); \ diff --git a/include/linux/tc_ematch/tc_em_text.h b/include/linux/tc_ematch/tc_em_text.h new file mode 100644 index 000000000000..7cd43e99c7f5 --- /dev/null +++ b/include/linux/tc_ematch/tc_em_text.h @@ -0,0 +1,19 @@ +#ifndef __LINUX_TC_EM_TEXT_H +#define __LINUX_TC_EM_TEXT_H + +#include + +#define TC_EM_TEXT_ALGOSIZ 16 + +struct tcf_em_text +{ + char algo[TC_EM_TEXT_ALGOSIZ]; + __u16 from_offset; + __u16 to_offset; + __u16 pattern_len; + __u8 from_layer:4; + __u8 to_layer:4; + __u8 pad; +}; + +#endif diff --git a/net/sched/Kconfig b/net/sched/Kconfig index b22c9beb604d..95d9bc5d8621 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -449,6 +449,17 @@ config NET_EMATCH_META To compile this code as a module, choose M here: the module will be called em_meta. +config NET_EMATCH_TEXT + tristate "Textsearch" + depends on NET_EMATCH + ---help--- + Say Y here if you want to be ablt to classify packets based on + textsearch comparisons. Please select the appropriate textsearch + algorithms in the Library section. + + To compile this code as a module, choose M here: the + module will be called em_text. + config NET_CLS_ACT bool "Packet ACTION" depends on EXPERIMENTAL && NET_CLS && NET_QOS diff --git a/net/sched/Makefile b/net/sched/Makefile index eb3fe583eba8..8f58cecd6266 100644 --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -40,3 +40,4 @@ obj-$(CONFIG_NET_EMATCH_CMP) += em_cmp.o obj-$(CONFIG_NET_EMATCH_NBYTE) += em_nbyte.o obj-$(CONFIG_NET_EMATCH_U32) += em_u32.o obj-$(CONFIG_NET_EMATCH_META) += em_meta.o +obj-$(CONFIG_NET_EMATCH_TEXT) += em_text.o diff --git a/net/sched/em_text.c b/net/sched/em_text.c new file mode 100644 index 000000000000..873840d8d072 --- /dev/null +++ b/net/sched/em_text.c @@ -0,0 +1,157 @@ +/* + * net/sched/em_text.c Textsearch ematch + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Authors: Thomas Graf + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +struct text_match +{ + u16 from_offset; + u16 to_offset; + u8 from_layer; + u8 to_layer; + struct ts_config *config; +}; + +#define EM_TEXT_PRIV(m) ((struct text_match *) (m)->data) + +static int em_text_match(struct sk_buff *skb, struct tcf_ematch *m, + struct tcf_pkt_info *info) +{ + struct text_match *tm = EM_TEXT_PRIV(m); + int from, to; + struct ts_state state; + + from = tcf_get_base_ptr(skb, tm->from_layer) - skb->data; + from += tm->from_offset; + + to = tcf_get_base_ptr(skb, tm->to_layer) - skb->data; + to += tm->to_offset; + + return skb_find_text(skb, from, to, tm->config, &state) != UINT_MAX; +} + +static int em_text_change(struct tcf_proto *tp, void *data, int len, + struct tcf_ematch *m) +{ + struct text_match *tm; + struct tcf_em_text *conf = data; + struct ts_config *ts_conf; + int flags = 0; + + printk("Configuring text: %s from %d:%d to %d:%d len %d\n", conf->algo, conf->from_offset, + conf->from_layer, conf->to_offset, conf->to_layer, conf->pattern_len); + + if (len < sizeof(*conf) || len < (sizeof(*conf) + conf->pattern_len)) + return -EINVAL; + + if (conf->from_layer > conf->to_layer) + return -EINVAL; + + if (conf->from_layer == conf->to_layer && + conf->from_offset > conf->to_offset) + return -EINVAL; + +retry: + ts_conf = textsearch_prepare(conf->algo, (u8 *) conf + sizeof(*conf), + conf->pattern_len, GFP_KERNEL, flags); + + if (flags & TS_AUTOLOAD) + rtnl_lock(); + + if (IS_ERR(ts_conf)) { + if (PTR_ERR(ts_conf) == -ENOENT && !(flags & TS_AUTOLOAD)) { + rtnl_unlock(); + flags |= TS_AUTOLOAD; + goto retry; + } else + return PTR_ERR(ts_conf); + } else if (flags & TS_AUTOLOAD) { + textsearch_destroy(ts_conf); + return -EAGAIN; + } + + tm = kmalloc(sizeof(*tm), GFP_KERNEL); + if (tm == NULL) { + textsearch_destroy(ts_conf); + return -ENOBUFS; + } + + tm->from_offset = conf->from_offset; + tm->to_offset = conf->to_offset; + tm->from_layer = conf->from_layer; + tm->to_layer = conf->to_layer; + tm->config = ts_conf; + + m->datalen = sizeof(*tm); + m->data = (unsigned long) tm; + + return 0; +} + +static void em_text_destroy(struct tcf_proto *tp, struct tcf_ematch *m) +{ + textsearch_destroy(EM_TEXT_PRIV(m)->config); +} + +static int em_text_dump(struct sk_buff *skb, struct tcf_ematch *m) +{ + struct text_match *tm = EM_TEXT_PRIV(m); + struct tcf_em_text conf; + + strncpy(conf.algo, tm->config->ops->name, sizeof(conf.algo) - 1); + conf.from_offset = tm->from_offset; + conf.to_offset = tm->to_offset; + conf.from_layer = tm->from_layer; + conf.to_layer = tm->to_layer; + conf.pattern_len = textsearch_get_pattern_len(tm->config); + conf.pad = 0; + + RTA_PUT_NOHDR(skb, sizeof(conf), &conf); + RTA_APPEND(skb, conf.pattern_len, textsearch_get_pattern(tm->config)); + return 0; + +rtattr_failure: + return -1; +} + +static struct tcf_ematch_ops em_text_ops = { + .kind = TCF_EM_TEXT, + .change = em_text_change, + .match = em_text_match, + .destroy = em_text_destroy, + .dump = em_text_dump, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_text_ops.link) +}; + +static int __init init_em_text(void) +{ + return tcf_em_register(&em_text_ops); +} + +static void __exit exit_em_text(void) +{ + tcf_em_unregister(&em_text_ops); +} + +MODULE_LICENSE("GPL"); + +module_init(init_em_text); +module_exit(exit_em_text); -- cgit v1.2.3 From 76d8aeabfeb1c42641a81c44280177b9a08670d8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 23 Jun 2005 22:00:49 -0700 Subject: [PATCH] keys: Discard key spinlock and use RCU for key payload The attached patch changes the key implementation in a number of ways: (1) It removes the spinlock from the key structure. (2) The key flags are now accessed using atomic bitops instead of write-locking the key spinlock and using C bitwise operators. The three instantiation flags are dealt with with the construction semaphore held during the request_key/instantiate/negate sequence, thus rendering the spinlock superfluous. The key flags are also now bit numbers not bit masks. (3) The key payload is now accessed using RCU. This permits the recursive keyring search algorithm to be simplified greatly since no locks need be taken other than the usual RCU preemption disablement. Searching now does not require any locks or semaphores to be held; merely that the starting keyring be pinned. (4) The keyring payload now includes an RCU head so that it can be disposed of by call_rcu(). This requires that the payload be copied on unlink to prevent introducing races in copy-down vs search-up. (5) The user key payload is now a structure with the data following it. It includes an RCU head like the keyring payload and for the same reason. It also contains a data length because the data length in the key may be changed on another CPU whilst an RCU protected read is in progress on the payload. This would then see the supposed RCU payload and the on-key data length getting out of sync. I'm tempted to drop the key's datalen entirely, except that it's used in conjunction with quota management and so is a little tricky to get rid of. (6) Update the keys documentation. Signed-Off-By: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/keys.txt | 285 ++++++++++++++++++++++++++----------------- include/linux/key-ui.h | 6 +- include/linux/key.h | 25 ++-- security/keys/key.c | 94 +++++++------- security/keys/keyctl.c | 23 ++-- security/keys/keyring.c | 245 ++++++++++++++++++++++--------------- security/keys/proc.c | 21 ++-- security/keys/process_keys.c | 12 +- security/keys/request_key.c | 32 +++-- security/keys/user_defined.c | 85 +++++++++---- 10 files changed, 480 insertions(+), 348 deletions(-) (limited to 'include/linux') diff --git a/Documentation/keys.txt b/Documentation/keys.txt index 36d80aeeaf28..3df40c1fe15a 100644 --- a/Documentation/keys.txt +++ b/Documentation/keys.txt @@ -22,6 +22,7 @@ This document has the following sections: - New procfs files - Userspace system call interface - Kernel services + - Notes on accessing payload contents - Defining a key type - Request-key callback service - Key access filesystem @@ -45,27 +46,26 @@ Each key has a number of attributes: - State. - (*) Each key is issued a serial number of type key_serial_t that is unique - for the lifetime of that key. All serial numbers are positive non-zero - 32-bit integers. + (*) Each key is issued a serial number of type key_serial_t that is unique for + the lifetime of that key. All serial numbers are positive non-zero 32-bit + integers. Userspace programs can use a key's serial numbers as a way to gain access to it, subject to permission checking. (*) Each key is of a defined "type". Types must be registered inside the - kernel by a kernel service (such as a filesystem) before keys of that - type can be added or used. Userspace programs cannot define new types - directly. + kernel by a kernel service (such as a filesystem) before keys of that type + can be added or used. Userspace programs cannot define new types directly. - Key types are represented in the kernel by struct key_type. This defines - a number of operations that can be performed on a key of that type. + Key types are represented in the kernel by struct key_type. This defines a + number of operations that can be performed on a key of that type. Should a type be removed from the system, all the keys of that type will be invalidated. (*) Each key has a description. This should be a printable string. The key - type provides an operation to perform a match between the description on - a key and a criterion string. + type provides an operation to perform a match between the description on a + key and a criterion string. (*) Each key has an owner user ID, a group ID and a permissions mask. These are used to control what a process may do to a key from userspace, and @@ -74,10 +74,10 @@ Each key has a number of attributes: (*) Each key can be set to expire at a specific time by the key type's instantiation function. Keys can also be immortal. - (*) Each key can have a payload. This is a quantity of data that represent - the actual "key". In the case of a keyring, this is a list of keys to - which the keyring links; in the case of a user-defined key, it's an - arbitrary blob of data. + (*) Each key can have a payload. This is a quantity of data that represent the + actual "key". In the case of a keyring, this is a list of keys to which + the keyring links; in the case of a user-defined key, it's an arbitrary + blob of data. Having a payload is not required; and the payload can, in fact, just be a value stored in the struct key itself. @@ -92,8 +92,8 @@ Each key has a number of attributes: (*) Each key can be in one of a number of basic states: - (*) Uninstantiated. The key exists, but does not have any data - attached. Keys being requested from userspace will be in this state. + (*) Uninstantiated. The key exists, but does not have any data attached. + Keys being requested from userspace will be in this state. (*) Instantiated. This is the normal state. The key is fully formed, and has data attached. @@ -140,10 +140,10 @@ The key service provides a number of features besides keys: clone, fork, vfork or execve occurs. A new keyring is created only when required. - The process-specific keyring is replaced with an empty one in the child - on clone, fork, vfork unless CLONE_THREAD is supplied, in which case it - is shared. execve also discards the process's process keyring and creates - a new one. + The process-specific keyring is replaced with an empty one in the child on + clone, fork, vfork unless CLONE_THREAD is supplied, in which case it is + shared. execve also discards the process's process keyring and creates a + new one. The session-specific keyring is persistent across clone, fork, vfork and execve, even when the latter executes a set-UID or set-GID binary. A @@ -177,11 +177,11 @@ The key service provides a number of features besides keys: If a system call that modifies a key or keyring in some way would put the user over quota, the operation is refused and error EDQUOT is returned. - (*) There's a system call interface by which userspace programs can create - and manipulate keys and keyrings. + (*) There's a system call interface by which userspace programs can create and + manipulate keys and keyrings. - (*) There's a kernel interface by which services can register types and - search for keys. + (*) There's a kernel interface by which services can register types and search + for keys. (*) There's a way for the a search done from the kernel to call back to userspace to request a key that can't be found in a process's keyrings. @@ -194,9 +194,9 @@ The key service provides a number of features besides keys: KEY ACCESS PERMISSIONS ====================== -Keys have an owner user ID, a group access ID, and a permissions mask. The -mask has up to eight bits each for user, group and other access. Only five of -each set of eight bits are defined. These permissions granted are: +Keys have an owner user ID, a group access ID, and a permissions mask. The mask +has up to eight bits each for user, group and other access. Only five of each +set of eight bits are defined. These permissions granted are: (*) View @@ -210,8 +210,8 @@ each set of eight bits are defined. These permissions granted are: (*) Write - This permits a key's payload to be instantiated or updated, or it allows - a link to be added to or removed from a keyring. + This permits a key's payload to be instantiated or updated, or it allows a + link to be added to or removed from a keyring. (*) Search @@ -238,8 +238,8 @@ about the status of the key service: (*) /proc/keys This lists all the keys on the system, giving information about their - type, description and permissions. The payload of the key is not - available this way: + type, description and permissions. The payload of the key is not available + this way: SERIAL FLAGS USAGE EXPY PERM UID GID TYPE DESCRIPTION: SUMMARY 00000001 I----- 39 perm 1f0000 0 0 keyring _uid_ses.0: 1/4 @@ -318,21 +318,21 @@ The main syscalls are: If a key of the same type and description as that proposed already exists in the keyring, this will try to update it with the given payload, or it will return error EEXIST if that function is not supported by the key - type. The process must also have permission to write to the key to be - able to update it. The new key will have all user permissions granted and - no group or third party permissions. + type. The process must also have permission to write to the key to be able + to update it. The new key will have all user permissions granted and no + group or third party permissions. - Otherwise, this will attempt to create a new key of the specified type - and description, and to instantiate it with the supplied payload and - attach it to the keyring. In this case, an error will be generated if the - process does not have permission to write to the keyring. + Otherwise, this will attempt to create a new key of the specified type and + description, and to instantiate it with the supplied payload and attach it + to the keyring. In this case, an error will be generated if the process + does not have permission to write to the keyring. The payload is optional, and the pointer can be NULL if not required by the type. The payload is plen in size, and plen can be zero for an empty payload. - A new keyring can be generated by setting type "keyring", the keyring - name as the description (or NULL) and setting the payload to NULL. + A new keyring can be generated by setting type "keyring", the keyring name + as the description (or NULL) and setting the payload to NULL. User defined keys can be created by specifying type "user". It is recommended that a user defined key's description by prefixed with a type @@ -369,9 +369,9 @@ The keyctl syscall functions are: key_serial_t keyctl(KEYCTL_GET_KEYRING_ID, key_serial_t id, int create); - The special key specified by "id" is looked up (with the key being - created if necessary) and the ID of the key or keyring thus found is - returned if it exists. + The special key specified by "id" is looked up (with the key being created + if necessary) and the ID of the key or keyring thus found is returned if + it exists. If the key does not yet exist, the key will be created if "create" is non-zero; and the error ENOKEY will be returned if "create" is zero. @@ -402,8 +402,8 @@ The keyctl syscall functions are: This will try to update the specified key with the given payload, or it will return error EOPNOTSUPP if that function is not supported by the key - type. The process must also have permission to write to the key to be - able to update it. + type. The process must also have permission to write to the key to be able + to update it. The payload is of length plen, and may be absent or empty as for add_key(). @@ -422,8 +422,8 @@ The keyctl syscall functions are: long keyctl(KEYCTL_CHOWN, key_serial_t key, uid_t uid, gid_t gid); - This function permits a key's owner and group ID to be changed. Either - one of uid or gid can be set to -1 to suppress that change. + This function permits a key's owner and group ID to be changed. Either one + of uid or gid can be set to -1 to suppress that change. Only the superuser can change a key's owner to something other than the key's current owner. Similarly, only the superuser can change a key's @@ -484,12 +484,12 @@ The keyctl syscall functions are: long keyctl(KEYCTL_LINK, key_serial_t keyring, key_serial_t key); - This function creates a link from the keyring to the key. The process - must have write permission on the keyring and must have link permission - on the key. + This function creates a link from the keyring to the key. The process must + have write permission on the keyring and must have link permission on the + key. - Should the keyring not be a keyring, error ENOTDIR will result; and if - the keyring is full, error ENFILE will result. + Should the keyring not be a keyring, error ENOTDIR will result; and if the + keyring is full, error ENFILE will result. The link procedure checks the nesting of the keyrings, returning ELOOP if it appears to deep or EDEADLK if the link would introduce a cycle. @@ -503,8 +503,8 @@ The keyctl syscall functions are: specified key, and removes it if found. Subsequent links to that key are ignored. The process must have write permission on the keyring. - If the keyring is not a keyring, error ENOTDIR will result; and if the - key is not present, error ENOENT will be the result. + If the keyring is not a keyring, error ENOTDIR will result; and if the key + is not present, error ENOENT will be the result. (*) Search a keyring tree for a key: @@ -513,9 +513,9 @@ The keyctl syscall functions are: const char *type, const char *description, key_serial_t dest_keyring); - This searches the keyring tree headed by the specified keyring until a - key is found that matches the type and description criteria. Each keyring - is checked for keys before recursion into its children occurs. + This searches the keyring tree headed by the specified keyring until a key + is found that matches the type and description criteria. Each keyring is + checked for keys before recursion into its children occurs. The process must have search permission on the top level keyring, or else error EACCES will result. Only keyrings that the process has search @@ -549,8 +549,8 @@ The keyctl syscall functions are: As much of the data as can be fitted into the buffer will be copied to userspace if the buffer pointer is not NULL. - On a successful return, the function will always return the amount of - data available rather than the amount copied. + On a successful return, the function will always return the amount of data + available rather than the amount copied. (*) Instantiate a partially constructed key. @@ -568,8 +568,8 @@ The keyctl syscall functions are: it, and the key must be uninstantiated. If a keyring is specified (non-zero), the key will also be linked into - that keyring, however all the constraints applying in KEYCTL_LINK apply - in this case too. + that keyring, however all the constraints applying in KEYCTL_LINK apply in + this case too. The payload and plen arguments describe the payload data as for add_key(). @@ -587,8 +587,8 @@ The keyctl syscall functions are: it, and the key must be uninstantiated. If a keyring is specified (non-zero), the key will also be linked into - that keyring, however all the constraints applying in KEYCTL_LINK apply - in this case too. + that keyring, however all the constraints applying in KEYCTL_LINK apply in + this case too. =============== @@ -601,17 +601,14 @@ be broken down into two areas: keys and key types. Dealing with keys is fairly straightforward. Firstly, the kernel service registers its type, then it searches for a key of that type. It should retain the key as long as it has need of it, and then it should release it. For a -filesystem or device file, a search would probably be performed during the -open call, and the key released upon close. How to deal with conflicting keys -due to two different users opening the same file is left to the filesystem -author to solve. - -When accessing a key's payload data, key->lock should be at least read locked, -or else the data may be changed by an update being performed from userspace -whilst the driver or filesystem is trying to access it. If no update method is -supplied, then the key's payload may be accessed without holding a lock as -there is no way to change it, provided it can be guaranteed that the key's -type definition won't go away. +filesystem or device file, a search would probably be performed during the open +call, and the key released upon close. How to deal with conflicting keys due to +two different users opening the same file is left to the filesystem author to +solve. + +When accessing a key's payload contents, certain precautions must be taken to +prevent access vs modification races. See the section "Notes on accessing +payload contents" for more information. (*) To search for a key, call: @@ -690,6 +687,54 @@ type definition won't go away. void unregister_key_type(struct key_type *type); +=================================== +NOTES ON ACCESSING PAYLOAD CONTENTS +=================================== + +The simplest payload is just a number in key->payload.value. In this case, +there's no need to indulge in RCU or locking when accessing the payload. + +More complex payload contents must be allocated and a pointer to them set in +key->payload.data. One of the following ways must be selected to access the +data: + + (1) Unmodifyable key type. + + If the key type does not have a modify method, then the key's payload can + be accessed without any form of locking, provided that it's known to be + instantiated (uninstantiated keys cannot be "found"). + + (2) The key's semaphore. + + The semaphore could be used to govern access to the payload and to control + the payload pointer. It must be write-locked for modifications and would + have to be read-locked for general access. The disadvantage of doing this + is that the accessor may be required to sleep. + + (3) RCU. + + RCU must be used when the semaphore isn't already held; if the semaphore + is held then the contents can't change under you unexpectedly as the + semaphore must still be used to serialise modifications to the key. The + key management code takes care of this for the key type. + + However, this means using: + + rcu_read_lock() ... rcu_dereference() ... rcu_read_unlock() + + to read the pointer, and: + + rcu_dereference() ... rcu_assign_pointer() ... call_rcu() + + to set the pointer and dispose of the old contents after a grace period. + Note that only the key type should ever modify a key's payload. + + Furthermore, an RCU controlled payload must hold a struct rcu_head for the + use of call_rcu() and, if the payload is of variable size, the length of + the payload. key->datalen cannot be relied upon to be consistent with the + payload just dereferenced if the key's semaphore is not held. + + =================== DEFINING A KEY TYPE =================== @@ -717,15 +762,15 @@ The structure has a number of fields, some of which are mandatory: int key_payload_reserve(struct key *key, size_t datalen); - With the revised data length. Error EDQUOT will be returned if this is - not viable. + With the revised data length. Error EDQUOT will be returned if this is not + viable. (*) int (*instantiate)(struct key *key, const void *data, size_t datalen); This method is called to attach a payload to a key during construction. - The payload attached need not bear any relation to the data passed to - this function. + The payload attached need not bear any relation to the data passed to this + function. If the amount of data attached to the key differs from the size in keytype->def_datalen, then key_payload_reserve() should be called. @@ -734,38 +779,47 @@ The structure has a number of fields, some of which are mandatory: The fact that KEY_FLAG_INSTANTIATED is not set in key->flags prevents anything else from gaining access to the key. - This method may sleep if it wishes. + It is safe to sleep in this method. (*) int (*duplicate)(struct key *key, const struct key *source); If this type of key can be duplicated, then this method should be - provided. It is called to copy the payload attached to the source into - the new key. The data length on the new key will have been updated and - the quota adjusted already. + provided. It is called to copy the payload attached to the source into the + new key. The data length on the new key will have been updated and the + quota adjusted already. This method will be called with the source key's semaphore read-locked to - prevent its payload from being changed. It is safe to sleep here. + prevent its payload from being changed, thus RCU constraints need not be + applied to the source key. + + This method does not have to lock the destination key in order to attach a + payload. The fact that KEY_FLAG_INSTANTIATED is not set in key->flags + prevents anything else from gaining access to the key. + + It is safe to sleep in this method. (*) int (*update)(struct key *key, const void *data, size_t datalen); - If this type of key can be updated, then this method should be - provided. It is called to update a key's payload from the blob of data - provided. + If this type of key can be updated, then this method should be provided. + It is called to update a key's payload from the blob of data provided. key_payload_reserve() should be called if the data length might change - before any changes are actually made. Note that if this succeeds, the - type is committed to changing the key because it's already been altered, - so all memory allocation must be done first. + before any changes are actually made. Note that if this succeeds, the type + is committed to changing the key because it's already been altered, so all + memory allocation must be done first. + + The key will have its semaphore write-locked before this method is called, + but this only deters other writers; any changes to the key's payload must + be made under RCU conditions, and call_rcu() must be used to dispose of + the old payload. - key_payload_reserve() should be called with the key->lock write locked, - and the changes to the key's attached payload should be made before the - key is locked. + key_payload_reserve() should be called before the changes are made, but + after all allocations and other potentially failing function calls are + made. - The key will have its semaphore write-locked before this method is - called. Any changes to the key should be made with the key's rwlock - write-locked also. It is safe to sleep here. + It is safe to sleep in this method. (*) int (*match)(const struct key *key, const void *desc); @@ -782,12 +836,12 @@ The structure has a number of fields, some of which are mandatory: (*) void (*destroy)(struct key *key); - This method is optional. It is called to discard the payload data on a - key when it is being destroyed. + This method is optional. It is called to discard the payload data on a key + when it is being destroyed. - This method does not need to lock the key; it can consider the key as - being inaccessible. Note that the key's type may have changed before this - function is called. + This method does not need to lock the key to access the payload; it can + consider the key as being inaccessible at this time. Note that the key's + type may have been changed before this function is called. It is not safe to sleep in this method; the caller may hold spinlocks. @@ -797,26 +851,31 @@ The structure has a number of fields, some of which are mandatory: This method is optional. It is called during /proc/keys reading to summarise a key's description and payload in text form. - This method will be called with the key's rwlock read-locked. This will - prevent the key's payload and state changing; also the description should - not change. This also means it is not safe to sleep in this method. + This method will be called with the RCU read lock held. rcu_dereference() + should be used to read the payload pointer if the payload is to be + accessed. key->datalen cannot be trusted to stay consistent with the + contents of the payload. + + The description will not change, though the key's state may. + + It is not safe to sleep in this method; the RCU read lock is held by the + caller. (*) long (*read)(const struct key *key, char __user *buffer, size_t buflen); This method is optional. It is called by KEYCTL_READ to translate the - key's payload into something a blob of data for userspace to deal - with. Ideally, the blob should be in the same format as that passed in to - the instantiate and update methods. + key's payload into something a blob of data for userspace to deal with. + Ideally, the blob should be in the same format as that passed in to the + instantiate and update methods. If successful, the blob size that could be produced should be returned rather than the size copied. - This method will be called with the key's semaphore read-locked. This - will prevent the key's payload changing. It is not necessary to also - read-lock key->lock when accessing the key's payload. It is safe to sleep - in this method, such as might happen when the userspace buffer is - accessed. + This method will be called with the key's semaphore read-locked. This will + prevent the key's payload changing. It is not necessary to use RCU locking + when accessing the key's payload. It is safe to sleep in this method, such + as might happen when the userspace buffer is accessed. ============================ @@ -853,8 +912,8 @@ If it returns with the key remaining in the unconstructed state, the key will be marked as being negative, it will be added to the session keyring, and an error will be returned to the key requestor. -Supplementary information may be provided from whoever or whatever invoked -this service. This will be passed as the parameter. If no such +Supplementary information may be provided from whoever or whatever invoked this +service. This will be passed as the parameter. If no such information was made available, then "-" will be passed as this parameter instead. diff --git a/include/linux/key-ui.h b/include/linux/key-ui.h index 60cc7b762e78..159ca8d54e9a 100644 --- a/include/linux/key-ui.h +++ b/include/linux/key-ui.h @@ -31,8 +31,10 @@ extern spinlock_t key_serial_lock; * subscribed */ struct keyring_list { - unsigned maxkeys; /* max keys this list can hold */ - unsigned nkeys; /* number of keys currently held */ + struct rcu_head rcu; /* RCU deletion hook */ + unsigned short maxkeys; /* max keys this list can hold */ + unsigned short nkeys; /* number of keys currently held */ + unsigned short delkey; /* key to be unlinked by RCU */ struct key *keys[0]; }; diff --git a/include/linux/key.h b/include/linux/key.h index 6aa46d0e812f..2c24ffaca86f 100644 --- a/include/linux/key.h +++ b/include/linux/key.h @@ -18,7 +18,7 @@ #include #include #include -#include +#include #include #ifdef __KERNEL__ @@ -78,7 +78,6 @@ struct key { key_serial_t serial; /* key serial number */ struct rb_node serial_node; struct key_type *type; /* type of key */ - rwlock_t lock; /* examination vs change lock */ struct rw_semaphore sem; /* change vs change sem */ struct key_user *user; /* owner of this key */ time_t expiry; /* time at which key expires (or 0) */ @@ -86,14 +85,10 @@ struct key { gid_t gid; key_perm_t perm; /* access permissions */ unsigned short quotalen; /* length added to quota */ - unsigned short datalen; /* payload data length */ - unsigned short flags; /* status flags (change with lock writelocked) */ -#define KEY_FLAG_INSTANTIATED 0x00000001 /* set if key has been instantiated */ -#define KEY_FLAG_DEAD 0x00000002 /* set if key type has been deleted */ -#define KEY_FLAG_REVOKED 0x00000004 /* set if key had been revoked */ -#define KEY_FLAG_IN_QUOTA 0x00000008 /* set if key consumes quota */ -#define KEY_FLAG_USER_CONSTRUCT 0x00000010 /* set if key is being constructed in userspace */ -#define KEY_FLAG_NEGATIVE 0x00000020 /* set if key is negative */ + unsigned short datalen; /* payload data length + * - may not match RCU dereferenced payload + * - payload should contain own length + */ #ifdef KEY_DEBUGGING unsigned magic; @@ -101,6 +96,14 @@ struct key { #define KEY_DEBUG_MAGIC_X 0xf8e9dacbu #endif + unsigned long flags; /* status flags (change with bitops) */ +#define KEY_FLAG_INSTANTIATED 0 /* set if key has been instantiated */ +#define KEY_FLAG_DEAD 1 /* set if key type has been deleted */ +#define KEY_FLAG_REVOKED 2 /* set if key had been revoked */ +#define KEY_FLAG_IN_QUOTA 3 /* set if key consumes quota */ +#define KEY_FLAG_USER_CONSTRUCT 4 /* set if key is being constructed in userspace */ +#define KEY_FLAG_NEGATIVE 5 /* set if key is negative */ + /* the description string * - this is used to match a key against search criteria * - this should be a printable string @@ -250,6 +253,8 @@ extern int keyring_add_key(struct key *keyring, extern struct key *key_lookup(key_serial_t id); +extern void keyring_replace_payload(struct key *key, void *replacement); + #define key_serial(key) ((key) ? (key)->serial : 0) /* diff --git a/security/keys/key.c b/security/keys/key.c index 59402c843203..1fdfccb3fe43 100644 --- a/security/keys/key.c +++ b/security/keys/key.c @@ -294,7 +294,6 @@ struct key *key_alloc(struct key_type *type, const char *desc, } atomic_set(&key->usage, 1); - rwlock_init(&key->lock); init_rwsem(&key->sem); key->type = type; key->user = user; @@ -308,7 +307,7 @@ struct key *key_alloc(struct key_type *type, const char *desc, key->payload.data = NULL; if (!not_in_quota) - key->flags |= KEY_FLAG_IN_QUOTA; + key->flags |= 1 << KEY_FLAG_IN_QUOTA; memset(&key->type_data, 0, sizeof(key->type_data)); @@ -359,7 +358,7 @@ int key_payload_reserve(struct key *key, size_t datalen) key_check(key); /* contemplate the quota adjustment */ - if (delta != 0 && key->flags & KEY_FLAG_IN_QUOTA) { + if (delta != 0 && test_bit(KEY_FLAG_IN_QUOTA, &key->flags)) { spin_lock(&key->user->lock); if (delta > 0 && @@ -405,23 +404,17 @@ static int __key_instantiate_and_link(struct key *key, down_write(&key_construction_sem); /* can't instantiate twice */ - if (!(key->flags & KEY_FLAG_INSTANTIATED)) { + if (!test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) { /* instantiate the key */ ret = key->type->instantiate(key, data, datalen); if (ret == 0) { /* mark the key as being instantiated */ - write_lock(&key->lock); - atomic_inc(&key->user->nikeys); - key->flags |= KEY_FLAG_INSTANTIATED; + set_bit(KEY_FLAG_INSTANTIATED, &key->flags); - if (key->flags & KEY_FLAG_USER_CONSTRUCT) { - key->flags &= ~KEY_FLAG_USER_CONSTRUCT; + if (test_and_clear_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags)) awaken = 1; - } - - write_unlock(&key->lock); /* and link it into the destination keyring */ if (keyring) @@ -486,21 +479,17 @@ int key_negate_and_link(struct key *key, down_write(&key_construction_sem); /* can't instantiate twice */ - if (!(key->flags & KEY_FLAG_INSTANTIATED)) { + if (!test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) { /* mark the key as being negatively instantiated */ - write_lock(&key->lock); - atomic_inc(&key->user->nikeys); - key->flags |= KEY_FLAG_INSTANTIATED | KEY_FLAG_NEGATIVE; + set_bit(KEY_FLAG_NEGATIVE, &key->flags); + set_bit(KEY_FLAG_INSTANTIATED, &key->flags); now = current_kernel_time(); key->expiry = now.tv_sec + timeout; - if (key->flags & KEY_FLAG_USER_CONSTRUCT) { - key->flags &= ~KEY_FLAG_USER_CONSTRUCT; + if (test_and_clear_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags)) awaken = 1; - } - write_unlock(&key->lock); ret = 0; /* and link it into the destination keyring */ @@ -553,8 +542,10 @@ static void key_cleanup(void *data) rb_erase(&key->serial_node, &key_serial_tree); spin_unlock(&key_serial_lock); + key_check(key); + /* deal with the user's key tracking and quota */ - if (key->flags & KEY_FLAG_IN_QUOTA) { + if (test_bit(KEY_FLAG_IN_QUOTA, &key->flags)) { spin_lock(&key->user->lock); key->user->qnkeys--; key->user->qnbytes -= key->quotalen; @@ -562,7 +553,7 @@ static void key_cleanup(void *data) } atomic_dec(&key->user->nkeys); - if (key->flags & KEY_FLAG_INSTANTIATED) + if (test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) atomic_dec(&key->user->nikeys); key_user_put(key->user); @@ -631,9 +622,9 @@ struct key *key_lookup(key_serial_t id) goto error; found: - /* pretent doesn't exist if it's dead */ + /* pretend it doesn't exist if it's dead */ if (atomic_read(&key->usage) == 0 || - (key->flags & KEY_FLAG_DEAD) || + test_bit(KEY_FLAG_DEAD, &key->flags) || key->type == &key_type_dead) goto not_found; @@ -708,12 +699,9 @@ static inline struct key *__key_update(struct key *key, const void *payload, ret = key->type->update(key, payload, plen); - if (ret == 0) { + if (ret == 0) /* updating a negative key instantiates it */ - write_lock(&key->lock); - key->flags &= ~KEY_FLAG_NEGATIVE; - write_unlock(&key->lock); - } + clear_bit(KEY_FLAG_NEGATIVE, &key->flags); up_write(&key->sem); @@ -841,12 +829,9 @@ int key_update(struct key *key, const void *payload, size_t plen) down_write(&key->sem); ret = key->type->update(key, payload, plen); - if (ret == 0) { + if (ret == 0) /* updating a negative key instantiates it */ - write_lock(&key->lock); - key->flags &= ~KEY_FLAG_NEGATIVE; - write_unlock(&key->lock); - } + clear_bit(KEY_FLAG_NEGATIVE, &key->flags); up_write(&key->sem); } @@ -892,10 +877,7 @@ struct key *key_duplicate(struct key *source, const char *desc) goto error2; atomic_inc(&key->user->nikeys); - - write_lock(&key->lock); - key->flags |= KEY_FLAG_INSTANTIATED; - write_unlock(&key->lock); + set_bit(KEY_FLAG_INSTANTIATED, &key->flags); error_k: up_read(&key_types_sem); @@ -922,9 +904,7 @@ void key_revoke(struct key *key) /* make sure no one's trying to change or use the key when we mark * it */ down_write(&key->sem); - write_lock(&key->lock); - key->flags |= KEY_FLAG_REVOKED; - write_unlock(&key->lock); + set_bit(KEY_FLAG_REVOKED, &key->flags); up_write(&key->sem); } /* end key_revoke() */ @@ -975,24 +955,33 @@ void unregister_key_type(struct key_type *ktype) /* withdraw the key type */ list_del_init(&ktype->link); - /* need to withdraw all keys of this type */ + /* mark all the keys of this type dead */ spin_lock(&key_serial_lock); for (_n = rb_first(&key_serial_tree); _n; _n = rb_next(_n)) { key = rb_entry(_n, struct key, serial_node); - if (key->type != ktype) - continue; + if (key->type == ktype) + key->type = &key_type_dead; + } + + spin_unlock(&key_serial_lock); + + /* make sure everyone revalidates their keys */ + synchronize_kernel(); + + /* we should now be able to destroy the payloads of all the keys of + * this type with impunity */ + spin_lock(&key_serial_lock); - write_lock(&key->lock); - key->type = &key_type_dead; - write_unlock(&key->lock); + for (_n = rb_first(&key_serial_tree); _n; _n = rb_next(_n)) { + key = rb_entry(_n, struct key, serial_node); - /* there shouldn't be anyone looking at the description or - * payload now */ - if (ktype->destroy) - ktype->destroy(key); - memset(&key->payload, 0xbd, sizeof(key->payload)); + if (key->type == ktype) { + if (ktype->destroy) + ktype->destroy(key); + memset(&key->payload, 0xbd, sizeof(key->payload)); + } } spin_unlock(&key_serial_lock); @@ -1037,4 +1026,5 @@ void __init key_init(void) /* link the two root keyrings together */ key_link(&root_session_keyring, &root_user_keyring); + } /* end key_init() */ diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c index dc0011b3fac9..cedb7326de29 100644 --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -728,7 +728,6 @@ long keyctl_chown_key(key_serial_t id, uid_t uid, gid_t gid) /* make the changes with the locks held to prevent chown/chown races */ ret = -EACCES; down_write(&key->sem); - write_lock(&key->lock); if (!capable(CAP_SYS_ADMIN)) { /* only the sysadmin can chown a key to some other UID */ @@ -755,7 +754,6 @@ long keyctl_chown_key(key_serial_t id, uid_t uid, gid_t gid) ret = 0; no_access: - write_unlock(&key->lock); up_write(&key->sem); key_put(key); error: @@ -784,26 +782,19 @@ long keyctl_setperm_key(key_serial_t id, key_perm_t perm) goto error; } - /* make the changes with the locks held to prevent chown/chmod - * races */ + /* make the changes with the locks held to prevent chown/chmod races */ ret = -EACCES; down_write(&key->sem); - write_lock(&key->lock); - /* if we're not the sysadmin, we can only chmod a key that we - * own */ - if (!capable(CAP_SYS_ADMIN) && key->uid != current->fsuid) - goto no_access; - - /* changing the permissions mask */ - key->perm = perm; - ret = 0; + /* if we're not the sysadmin, we can only change a key that we own */ + if (capable(CAP_SYS_ADMIN) || key->uid == current->fsuid) { + key->perm = perm; + ret = 0; + } - no_access: - write_unlock(&key->lock); up_write(&key->sem); key_put(key); - error: +error: return ret; } /* end keyctl_setperm_key() */ diff --git a/security/keys/keyring.c b/security/keys/keyring.c index e2ab4f8e7481..c9a5de197487 100644 --- a/security/keys/keyring.c +++ b/security/keys/keyring.c @@ -132,10 +132,17 @@ static int keyring_duplicate(struct key *keyring, const struct key *source) (PAGE_SIZE - sizeof(*klist)) / sizeof(struct key); ret = 0; - sklist = source->payload.subscriptions; - if (sklist && sklist->nkeys > 0) { + /* find out how many keys are currently linked */ + rcu_read_lock(); + sklist = rcu_dereference(source->payload.subscriptions); + max = 0; + if (sklist) max = sklist->nkeys; + rcu_read_unlock(); + + /* allocate a new payload and stuff load with key links */ + if (max > 0) { BUG_ON(max > limit); max = (max + 3) & ~3; @@ -148,6 +155,10 @@ static int keyring_duplicate(struct key *keyring, const struct key *source) if (!klist) goto error; + /* set links */ + rcu_read_lock(); + sklist = rcu_dereference(source->payload.subscriptions); + klist->maxkeys = max; klist->nkeys = sklist->nkeys; memcpy(klist->keys, @@ -157,7 +168,9 @@ static int keyring_duplicate(struct key *keyring, const struct key *source) for (loop = klist->nkeys - 1; loop >= 0; loop--) atomic_inc(&klist->keys[loop]->usage); - keyring->payload.subscriptions = klist; + rcu_read_unlock(); + + rcu_assign_pointer(keyring->payload.subscriptions, klist); ret = 0; } @@ -192,7 +205,7 @@ static void keyring_destroy(struct key *keyring) write_unlock(&keyring_name_lock); } - klist = keyring->payload.subscriptions; + klist = rcu_dereference(keyring->payload.subscriptions); if (klist) { for (loop = klist->nkeys - 1; loop >= 0; loop--) key_put(klist->keys[loop]); @@ -216,17 +229,20 @@ static void keyring_describe(const struct key *keyring, struct seq_file *m) seq_puts(m, "[anon]"); } - klist = keyring->payload.subscriptions; + rcu_read_lock(); + klist = rcu_dereference(keyring->payload.subscriptions); if (klist) seq_printf(m, ": %u/%u", klist->nkeys, klist->maxkeys); else seq_puts(m, ": empty"); + rcu_read_unlock(); } /* end keyring_describe() */ /*****************************************************************************/ /* * read a list of key IDs from the keyring's contents + * - the keyring's semaphore is read-locked */ static long keyring_read(const struct key *keyring, char __user *buffer, size_t buflen) @@ -237,7 +253,7 @@ static long keyring_read(const struct key *keyring, int loop, ret; ret = 0; - klist = keyring->payload.subscriptions; + klist = rcu_dereference(keyring->payload.subscriptions); if (klist) { /* calculate how much data we could return */ @@ -320,7 +336,7 @@ struct key *keyring_search_aux(struct key *keyring, key_match_func_t match) { struct { - struct key *keyring; + struct keyring_list *keylist; int kix; } stack[KEYRING_SEARCH_MAX_DEPTH]; @@ -328,10 +344,12 @@ struct key *keyring_search_aux(struct key *keyring, struct timespec now; struct key *key; long err; - int sp, psp, kix; + int sp, kix; key_check(keyring); + rcu_read_lock(); + /* top keyring must have search permission to begin the search */ key = ERR_PTR(-EACCES); if (!key_permission(keyring, KEY_SEARCH)) @@ -347,11 +365,10 @@ struct key *keyring_search_aux(struct key *keyring, /* start processing a new keyring */ descend: - read_lock(&keyring->lock); - if (keyring->flags & KEY_FLAG_REVOKED) + if (test_bit(KEY_FLAG_REVOKED, &keyring->flags)) goto not_this_keyring; - keylist = keyring->payload.subscriptions; + keylist = rcu_dereference(keyring->payload.subscriptions); if (!keylist) goto not_this_keyring; @@ -364,7 +381,7 @@ struct key *keyring_search_aux(struct key *keyring, continue; /* skip revoked keys and expired keys */ - if (key->flags & KEY_FLAG_REVOKED) + if (test_bit(KEY_FLAG_REVOKED, &key->flags)) continue; if (key->expiry && now.tv_sec >= key->expiry) @@ -379,7 +396,7 @@ struct key *keyring_search_aux(struct key *keyring, continue; /* we set a different error code if we find a negative key */ - if (key->flags & KEY_FLAG_NEGATIVE) { + if (test_bit(KEY_FLAG_NEGATIVE, &key->flags)) { err = -ENOKEY; continue; } @@ -390,48 +407,37 @@ struct key *keyring_search_aux(struct key *keyring, /* search through the keyrings nested in this one */ kix = 0; ascend: - while (kix < keylist->nkeys) { + for (; kix < keylist->nkeys; kix++) { key = keylist->keys[kix]; if (key->type != &key_type_keyring) - goto next; + continue; /* recursively search nested keyrings * - only search keyrings for which we have search permission */ if (sp >= KEYRING_SEARCH_MAX_DEPTH) - goto next; + continue; if (!key_permission(key, KEY_SEARCH)) - goto next; - - /* evade loops in the keyring tree */ - for (psp = 0; psp < sp; psp++) - if (stack[psp].keyring == keyring) - goto next; + continue; /* stack the current position */ - stack[sp].keyring = keyring; + stack[sp].keylist = keylist; stack[sp].kix = kix; sp++; /* begin again with the new keyring */ keyring = key; goto descend; - - next: - kix++; } /* the keyring we're looking at was disqualified or didn't contain a * matching key */ not_this_keyring: - read_unlock(&keyring->lock); - if (sp > 0) { /* resume the processing of a keyring higher up in the tree */ sp--; - keyring = stack[sp].keyring; - keylist = keyring->payload.subscriptions; + keylist = stack[sp].keylist; kix = stack[sp].kix + 1; goto ascend; } @@ -442,16 +448,9 @@ struct key *keyring_search_aux(struct key *keyring, /* we found a viable match */ found: atomic_inc(&key->usage); - read_unlock(&keyring->lock); - - /* unwind the keyring stack */ - while (sp > 0) { - sp--; - read_unlock(&stack[sp].keyring->lock); - } - key_check(key); error: + rcu_read_unlock(); return key; } /* end keyring_search_aux() */ @@ -489,7 +488,9 @@ struct key *__keyring_search_one(struct key *keyring, struct key *key; int loop; - klist = keyring->payload.subscriptions; + rcu_read_lock(); + + klist = rcu_dereference(keyring->payload.subscriptions); if (klist) { for (loop = 0; loop < klist->nkeys; loop++) { key = klist->keys[loop]; @@ -497,7 +498,7 @@ struct key *__keyring_search_one(struct key *keyring, if (key->type == ktype && key->type->match(key, description) && key_permission(key, perm) && - !(key->flags & KEY_FLAG_REVOKED) + !test_bit(KEY_FLAG_REVOKED, &key->flags) ) goto found; } @@ -509,6 +510,7 @@ struct key *__keyring_search_one(struct key *keyring, found: atomic_inc(&key->usage); error: + rcu_read_unlock(); return key; } /* end __keyring_search_one() */ @@ -540,7 +542,7 @@ struct key *find_keyring_by_name(const char *name, key_serial_t bound) &keyring_name_hash[bucket], type_data.link ) { - if (keyring->flags & KEY_FLAG_REVOKED) + if (test_bit(KEY_FLAG_REVOKED, &keyring->flags)) continue; if (strcmp(keyring->description, name) != 0) @@ -579,7 +581,7 @@ struct key *find_keyring_by_name(const char *name, key_serial_t bound) static int keyring_detect_cycle(struct key *A, struct key *B) { struct { - struct key *subtree; + struct keyring_list *keylist; int kix; } stack[KEYRING_SEARCH_MAX_DEPTH]; @@ -587,20 +589,21 @@ static int keyring_detect_cycle(struct key *A, struct key *B) struct key *subtree, *key; int sp, kix, ret; + rcu_read_lock(); + ret = -EDEADLK; if (A == B) - goto error; + goto cycle_detected; subtree = B; sp = 0; /* start processing a new keyring */ descend: - read_lock(&subtree->lock); - if (subtree->flags & KEY_FLAG_REVOKED) + if (test_bit(KEY_FLAG_REVOKED, &subtree->flags)) goto not_this_keyring; - keylist = subtree->payload.subscriptions; + keylist = rcu_dereference(subtree->payload.subscriptions); if (!keylist) goto not_this_keyring; kix = 0; @@ -619,7 +622,7 @@ static int keyring_detect_cycle(struct key *A, struct key *B) goto too_deep; /* stack the current position */ - stack[sp].subtree = subtree; + stack[sp].keylist = keylist; stack[sp].kix = kix; sp++; @@ -632,13 +635,10 @@ static int keyring_detect_cycle(struct key *A, struct key *B) /* the keyring we're looking at was disqualified or didn't contain a * matching key */ not_this_keyring: - read_unlock(&subtree->lock); - if (sp > 0) { /* resume the checking of a keyring higher up in the tree */ sp--; - subtree = stack[sp].subtree; - keylist = subtree->payload.subscriptions; + keylist = stack[sp].keylist; kix = stack[sp].kix + 1; goto ascend; } @@ -646,30 +646,36 @@ static int keyring_detect_cycle(struct key *A, struct key *B) ret = 0; /* no cycles detected */ error: + rcu_read_unlock(); return ret; too_deep: ret = -ELOOP; - goto error_unwind; + goto error; + cycle_detected: ret = -EDEADLK; - error_unwind: - read_unlock(&subtree->lock); - - /* unwind the keyring stack */ - while (sp > 0) { - sp--; - read_unlock(&stack[sp].subtree->lock); - } - goto error; } /* end keyring_detect_cycle() */ +/*****************************************************************************/ +/* + * dispose of a keyring list after the RCU grace period + */ +static void keyring_link_rcu_disposal(struct rcu_head *rcu) +{ + struct keyring_list *klist = + container_of(rcu, struct keyring_list, rcu); + + kfree(klist); + +} /* end keyring_link_rcu_disposal() */ + /*****************************************************************************/ /* * link a key into to a keyring - * - must be called with the keyring's semaphore held + * - must be called with the keyring's semaphore write-locked */ int __key_link(struct key *keyring, struct key *key) { @@ -679,7 +685,7 @@ int __key_link(struct key *keyring, struct key *key) int ret; ret = -EKEYREVOKED; - if (keyring->flags & KEY_FLAG_REVOKED) + if (test_bit(KEY_FLAG_REVOKED, &keyring->flags)) goto error; ret = -ENOTDIR; @@ -710,9 +716,10 @@ int __key_link(struct key *keyring, struct key *key) /* there's sufficient slack space to add directly */ atomic_inc(&key->usage); - write_lock(&keyring->lock); - klist->keys[klist->nkeys++] = key; - write_unlock(&keyring->lock); + klist->keys[klist->nkeys] = key; + smp_wmb(); + klist->nkeys++; + smp_wmb(); ret = 0; } @@ -723,6 +730,8 @@ int __key_link(struct key *keyring, struct key *key) max += klist->maxkeys; ret = -ENFILE; + if (max > 65535) + goto error3; size = sizeof(*klist) + sizeof(*key) * max; if (size > PAGE_SIZE) goto error3; @@ -743,14 +752,13 @@ int __key_link(struct key *keyring, struct key *key) /* add the key into the new space */ atomic_inc(&key->usage); - - write_lock(&keyring->lock); - keyring->payload.subscriptions = nklist; nklist->keys[nklist->nkeys++] = key; - write_unlock(&keyring->lock); + + rcu_assign_pointer(keyring->payload.subscriptions, nklist); /* dispose of the old keyring list */ - kfree(klist); + if (klist) + call_rcu(&klist->rcu, keyring_link_rcu_disposal); ret = 0; } @@ -789,13 +797,28 @@ int key_link(struct key *keyring, struct key *key) EXPORT_SYMBOL(key_link); +/*****************************************************************************/ +/* + * dispose of a keyring list after the RCU grace period, freeing the unlinked + * key + */ +static void keyring_unlink_rcu_disposal(struct rcu_head *rcu) +{ + struct keyring_list *klist = + container_of(rcu, struct keyring_list, rcu); + + key_put(klist->keys[klist->delkey]); + kfree(klist); + +} /* end keyring_unlink_rcu_disposal() */ + /*****************************************************************************/ /* * unlink the first link to a key from a keyring */ int key_unlink(struct key *keyring, struct key *key) { - struct keyring_list *klist; + struct keyring_list *klist, *nklist; int loop, ret; key_check(keyring); @@ -819,36 +842,69 @@ int key_unlink(struct key *keyring, struct key *key) ret = -ENOENT; goto error; - key_is_present: +key_is_present: + /* we need to copy the key list for RCU purposes */ + nklist = kmalloc(sizeof(*klist) + sizeof(*key) * klist->maxkeys, + GFP_KERNEL); + if (!nklist) + goto nomem; + nklist->maxkeys = klist->maxkeys; + nklist->nkeys = klist->nkeys - 1; + + if (loop > 0) + memcpy(&nklist->keys[0], + &klist->keys[0], + loop * sizeof(klist->keys[0])); + + if (loop < nklist->nkeys) + memcpy(&nklist->keys[loop], + &klist->keys[loop + 1], + (nklist->nkeys - loop) * sizeof(klist->keys[0])); + /* adjust the user's quota */ key_payload_reserve(keyring, keyring->datalen - KEYQUOTA_LINK_BYTES); - /* shuffle down the key pointers - * - it might be worth shrinking the allocated memory, but that runs - * the risk of ENOMEM as we would have to copy - */ - write_lock(&keyring->lock); + rcu_assign_pointer(keyring->payload.subscriptions, nklist); - klist->nkeys--; - if (loop < klist->nkeys) - memcpy(&klist->keys[loop], - &klist->keys[loop + 1], - (klist->nkeys - loop) * sizeof(struct key *)); + up_write(&keyring->sem); - write_unlock(&keyring->lock); + /* schedule for later cleanup */ + klist->delkey = loop; + call_rcu(&klist->rcu, keyring_unlink_rcu_disposal); - up_write(&keyring->sem); - key_put(key); ret = 0; - error: +error: return ret; +nomem: + ret = -ENOMEM; + up_write(&keyring->sem); + goto error; } /* end key_unlink() */ EXPORT_SYMBOL(key_unlink); +/*****************************************************************************/ +/* + * dispose of a keyring list after the RCU grace period, releasing the keys it + * links to + */ +static void keyring_clear_rcu_disposal(struct rcu_head *rcu) +{ + struct keyring_list *klist; + int loop; + + klist = container_of(rcu, struct keyring_list, rcu); + + for (loop = klist->nkeys - 1; loop >= 0; loop--) + key_put(klist->keys[loop]); + + kfree(klist); + +} /* end keyring_clear_rcu_disposal() */ + /*****************************************************************************/ /* * clear the specified process keyring @@ -857,7 +913,7 @@ EXPORT_SYMBOL(key_unlink); int keyring_clear(struct key *keyring) { struct keyring_list *klist; - int loop, ret; + int ret; ret = -ENOTDIR; if (keyring->type == &key_type_keyring) { @@ -870,20 +926,15 @@ int keyring_clear(struct key *keyring) key_payload_reserve(keyring, sizeof(struct keyring_list)); - write_lock(&keyring->lock); - keyring->payload.subscriptions = NULL; - write_unlock(&keyring->lock); + rcu_assign_pointer(keyring->payload.subscriptions, + NULL); } up_write(&keyring->sem); /* free the keys after the locks have been dropped */ - if (klist) { - for (loop = klist->nkeys - 1; loop >= 0; loop--) - key_put(klist->keys[loop]); - - kfree(klist); - } + if (klist) + call_rcu(&klist->rcu, keyring_clear_rcu_disposal); ret = 0; } diff --git a/security/keys/proc.c b/security/keys/proc.c index 91343b85c39c..c55cf1fd0826 100644 --- a/security/keys/proc.c +++ b/security/keys/proc.c @@ -140,7 +140,7 @@ static int proc_keys_show(struct seq_file *m, void *v) now = current_kernel_time(); - read_lock(&key->lock); + rcu_read_lock(); /* come up with a suitable timeout value */ if (key->expiry == 0) { @@ -164,14 +164,17 @@ static int proc_keys_show(struct seq_file *m, void *v) sprintf(xbuf, "%luw", timo / (60*60*24*7)); } +#define showflag(KEY, LETTER, FLAG) \ + (test_bit(FLAG, &(KEY)->flags) ? LETTER : '-') + seq_printf(m, "%08x %c%c%c%c%c%c %5d %4s %06x %5d %5d %-9.9s ", key->serial, - key->flags & KEY_FLAG_INSTANTIATED ? 'I' : '-', - key->flags & KEY_FLAG_REVOKED ? 'R' : '-', - key->flags & KEY_FLAG_DEAD ? 'D' : '-', - key->flags & KEY_FLAG_IN_QUOTA ? 'Q' : '-', - key->flags & KEY_FLAG_USER_CONSTRUCT ? 'U' : '-', - key->flags & KEY_FLAG_NEGATIVE ? 'N' : '-', + showflag(key, 'I', KEY_FLAG_INSTANTIATED), + showflag(key, 'R', KEY_FLAG_REVOKED), + showflag(key, 'D', KEY_FLAG_DEAD), + showflag(key, 'Q', KEY_FLAG_IN_QUOTA), + showflag(key, 'U', KEY_FLAG_USER_CONSTRUCT), + showflag(key, 'N', KEY_FLAG_NEGATIVE), atomic_read(&key->usage), xbuf, key->perm, @@ -179,11 +182,13 @@ static int proc_keys_show(struct seq_file *m, void *v) key->gid, key->type->name); +#undef showflag + if (key->type->describe) key->type->describe(key, m); seq_putc(m, '\n'); - read_unlock(&key->lock); + rcu_read_unlock(); return 0; diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c index 2eb0e471cd40..059c350cac46 100644 --- a/security/keys/process_keys.c +++ b/security/keys/process_keys.c @@ -38,10 +38,9 @@ struct key root_user_keyring = { .serial = 2, .type = &key_type_keyring, .user = &root_key_user, - .lock = RW_LOCK_UNLOCKED, .sem = __RWSEM_INITIALIZER(root_user_keyring.sem), .perm = KEY_USR_ALL, - .flags = KEY_FLAG_INSTANTIATED, + .flags = 1 << KEY_FLAG_INSTANTIATED, .description = "_uid.0", #ifdef KEY_DEBUGGING .magic = KEY_DEBUG_MAGIC, @@ -54,10 +53,9 @@ struct key root_session_keyring = { .serial = 1, .type = &key_type_keyring, .user = &root_key_user, - .lock = RW_LOCK_UNLOCKED, .sem = __RWSEM_INITIALIZER(root_session_keyring.sem), .perm = KEY_USR_ALL, - .flags = KEY_FLAG_INSTANTIATED, + .flags = 1 << KEY_FLAG_INSTANTIATED, .description = "_uid_ses.0", #ifdef KEY_DEBUGGING .magic = KEY_DEBUG_MAGIC, @@ -349,9 +347,7 @@ void key_fsuid_changed(struct task_struct *tsk) /* update the ownership of the thread keyring */ if (tsk->thread_keyring) { down_write(&tsk->thread_keyring->sem); - write_lock(&tsk->thread_keyring->lock); tsk->thread_keyring->uid = tsk->fsuid; - write_unlock(&tsk->thread_keyring->lock); up_write(&tsk->thread_keyring->sem); } @@ -366,9 +362,7 @@ void key_fsgid_changed(struct task_struct *tsk) /* update the ownership of the thread keyring */ if (tsk->thread_keyring) { down_write(&tsk->thread_keyring->sem); - write_lock(&tsk->thread_keyring->lock); tsk->thread_keyring->gid = tsk->fsgid; - write_unlock(&tsk->thread_keyring->lock); up_write(&tsk->thread_keyring->sem); } @@ -588,7 +582,7 @@ struct key *lookup_user_key(key_serial_t id, int create, int partial, } ret = -EIO; - if (!partial && !(key->flags & KEY_FLAG_INSTANTIATED)) + if (!partial && !test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) goto invalid_key; ret = -EACCES; diff --git a/security/keys/request_key.c b/security/keys/request_key.c index 9705b1aeba5d..1f6c0940297f 100644 --- a/security/keys/request_key.c +++ b/security/keys/request_key.c @@ -105,7 +105,7 @@ static struct key *__request_key_construction(struct key_type *type, struct key_construction cons; struct timespec now; struct key *key; - int ret, negative; + int ret, negated; /* create a key and add it to the queue */ key = key_alloc(type, description, @@ -113,9 +113,7 @@ static struct key *__request_key_construction(struct key_type *type, if (IS_ERR(key)) goto alloc_failed; - write_lock(&key->lock); - key->flags |= KEY_FLAG_USER_CONSTRUCT; - write_unlock(&key->lock); + set_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags); cons.key = key; list_add_tail(&cons.link, &key->user->consq); @@ -130,7 +128,7 @@ static struct key *__request_key_construction(struct key_type *type, /* if the key wasn't instantiated, then we want to give an error */ ret = -ENOKEY; - if (!(key->flags & KEY_FLAG_INSTANTIATED)) + if (!test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) goto request_failed; down_write(&key_construction_sem); @@ -139,7 +137,7 @@ static struct key *__request_key_construction(struct key_type *type, /* also give an error if the key was negatively instantiated */ check_not_negative: - if (key->flags & KEY_FLAG_NEGATIVE) { + if (test_bit(KEY_FLAG_NEGATIVE, &key->flags)) { key_put(key); key = ERR_PTR(-ENOKEY); } @@ -152,24 +150,23 @@ static struct key *__request_key_construction(struct key_type *type, * - remove from construction queue * - mark the key as dead */ - negative = 0; + negated = 0; down_write(&key_construction_sem); list_del(&cons.link); - write_lock(&key->lock); - key->flags &= ~KEY_FLAG_USER_CONSTRUCT; - /* check it didn't get instantiated between the check and the down */ - if (!(key->flags & KEY_FLAG_INSTANTIATED)) { - key->flags |= KEY_FLAG_INSTANTIATED | KEY_FLAG_NEGATIVE; - negative = 1; + if (!test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) { + set_bit(KEY_FLAG_NEGATIVE, &key->flags); + set_bit(KEY_FLAG_INSTANTIATED, &key->flags); + negated = 1; } - write_unlock(&key->lock); + clear_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags); + up_write(&key_construction_sem); - if (!negative) + if (!negated) goto check_not_negative; /* surprisingly, the key got * instantiated */ @@ -250,7 +247,7 @@ static struct key *request_key_construction(struct key_type *type, for (;;) { set_current_state(TASK_UNINTERRUPTIBLE); - if (!(ckey->flags & KEY_FLAG_USER_CONSTRUCT)) + if (!test_bit(KEY_FLAG_USER_CONSTRUCT, &ckey->flags)) break; schedule(); } @@ -339,7 +336,8 @@ int key_validate(struct key *key) if (key) { /* check it's still accessible */ ret = -EKEYREVOKED; - if (key->flags & (KEY_FLAG_REVOKED | KEY_FLAG_DEAD)) + if (test_bit(KEY_FLAG_REVOKED, &key->flags) || + test_bit(KEY_FLAG_DEAD, &key->flags)) goto error; /* check it hasn't expired */ diff --git a/security/keys/user_defined.c b/security/keys/user_defined.c index 8d65b3a28129..c33d3614a0db 100644 --- a/security/keys/user_defined.c +++ b/security/keys/user_defined.c @@ -42,12 +42,19 @@ struct key_type key_type_user = { .read = user_read, }; +struct user_key_payload { + struct rcu_head rcu; /* RCU destructor */ + unsigned short datalen; /* length of this data */ + char data[0]; /* actual data */ +}; + /*****************************************************************************/ /* * instantiate a user defined key */ static int user_instantiate(struct key *key, const void *data, size_t datalen) { + struct user_key_payload *upayload; int ret; ret = -EINVAL; @@ -58,13 +65,15 @@ static int user_instantiate(struct key *key, const void *data, size_t datalen) if (ret < 0) goto error; - /* attach the data */ ret = -ENOMEM; - key->payload.data = kmalloc(datalen, GFP_KERNEL); - if (!key->payload.data) + upayload = kmalloc(sizeof(*upayload) + datalen, GFP_KERNEL); + if (!upayload) goto error; - memcpy(key->payload.data, data, datalen); + /* attach the data */ + upayload->datalen = datalen; + memcpy(upayload->data, data, datalen); + rcu_assign_pointer(key->payload.data, upayload); ret = 0; error: @@ -75,18 +84,25 @@ static int user_instantiate(struct key *key, const void *data, size_t datalen) /*****************************************************************************/ /* * duplicate a user defined key + * - both keys' semaphores are locked against further modification + * - the new key cannot yet be accessed */ static int user_duplicate(struct key *key, const struct key *source) { + struct user_key_payload *upayload, *spayload; int ret; /* just copy the payload */ ret = -ENOMEM; - key->payload.data = kmalloc(source->datalen, GFP_KERNEL); + upayload = kmalloc(sizeof(*upayload) + source->datalen, GFP_KERNEL); + if (upayload) { + spayload = rcu_dereference(source->payload.data); + BUG_ON(source->datalen != spayload->datalen); - if (key->payload.data) { - key->datalen = source->datalen; - memcpy(key->payload.data, source->payload.data, source->datalen); + upayload->datalen = key->datalen = spayload->datalen; + memcpy(upayload->data, spayload->data, key->datalen); + + key->payload.data = upayload; ret = 0; } @@ -94,42 +110,56 @@ static int user_duplicate(struct key *key, const struct key *source) } /* end user_duplicate() */ +/*****************************************************************************/ +/* + * dispose of the old data from an updated user defined key + */ +static void user_update_rcu_disposal(struct rcu_head *rcu) +{ + struct user_key_payload *upayload; + + upayload = container_of(rcu, struct user_key_payload, rcu); + + kfree(upayload); + +} /* end user_update_rcu_disposal() */ + /*****************************************************************************/ /* * update a user defined key + * - the key's semaphore is write-locked */ static int user_update(struct key *key, const void *data, size_t datalen) { - void *new, *zap; + struct user_key_payload *upayload, *zap; int ret; ret = -EINVAL; if (datalen <= 0 || datalen > 32767 || !data) goto error; - /* copy the data */ + /* construct a replacement payload */ ret = -ENOMEM; - new = kmalloc(datalen, GFP_KERNEL); - if (!new) + upayload = kmalloc(sizeof(*upayload) + datalen, GFP_KERNEL); + if (!upayload) goto error; - memcpy(new, data, datalen); + upayload->datalen = datalen; + memcpy(upayload->data, data, datalen); /* check the quota and attach the new data */ - zap = new; - write_lock(&key->lock); + zap = upayload; ret = key_payload_reserve(key, datalen); if (ret == 0) { /* attach the new data, displacing the old */ zap = key->payload.data; - key->payload.data = new; + rcu_assign_pointer(key->payload.data, upayload); key->expiry = 0; } - write_unlock(&key->lock); - kfree(zap); + call_rcu(&zap->rcu, user_update_rcu_disposal); error: return ret; @@ -152,13 +182,15 @@ static int user_match(const struct key *key, const void *description) */ static void user_destroy(struct key *key) { - kfree(key->payload.data); + struct user_key_payload *upayload = key->payload.data; + + kfree(upayload); } /* end user_destroy() */ /*****************************************************************************/ /* - * describe the user + * describe the user key */ static void user_describe(const struct key *key, struct seq_file *m) { @@ -171,18 +203,23 @@ static void user_describe(const struct key *key, struct seq_file *m) /*****************************************************************************/ /* * read the key data + * - the key's semaphore is read-locked */ static long user_read(const struct key *key, char __user *buffer, size_t buflen) { - long ret = key->datalen; + struct user_key_payload *upayload; + long ret; + + upayload = rcu_dereference(key->payload.data); + ret = upayload->datalen; /* we can return the data as is */ if (buffer && buflen > 0) { - if (buflen > key->datalen) - buflen = key->datalen; + if (buflen > upayload->datalen) + buflen = upayload->datalen; - if (copy_to_user(buffer, key->payload.data, buflen) != 0) + if (copy_to_user(buffer, upayload->data, buflen) != 0) ret = -EFAULT; } -- cgit v1.2.3 From 7888e7ff4ee579442128d7d12a9c9dbf2cf7de6a Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 23 Jun 2005 22:00:51 -0700 Subject: [PATCH] Keys: Pass session keyring to call_usermodehelper() The attached patch makes it possible to pass a session keyring through to the process spawned by call_usermodehelper(). This allows patch 3/3 to pass an authorisation key through to /sbin/request-key, thus permitting better access controls when doing just-in-time key creation. Signed-Off-By: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/key.h | 10 +++++++++- include/linux/kmod.h | 13 ++++++++++++- kernel/kmod.c | 17 +++++++++++++---- security/keys/request_key.c | 2 +- 4 files changed, 35 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/key.h b/include/linux/key.h index 2c24ffaca86f..2bfbf88d2740 100644 --- a/include/linux/key.h +++ b/include/linux/key.h @@ -273,14 +273,22 @@ extern void key_fsuid_changed(struct task_struct *tsk); extern void key_fsgid_changed(struct task_struct *tsk); extern void key_init(void); +#define __install_session_keyring(tsk, keyring) \ +({ \ + struct key *old_session = tsk->signal->session_keyring; \ + tsk->signal->session_keyring = keyring; \ + old_session; \ +}) + #else /* CONFIG_KEYS */ #define key_validate(k) 0 #define key_serial(k) 0 -#define key_get(k) NULL +#define key_get(k) ({ NULL; }) #define key_put(k) do { } while(0) #define alloc_uid_keyring(u) 0 #define switch_uid_keyring(u) do { } while(0) +#define __install_session_keyring(t, k) ({ NULL; }) #define copy_keys(f,t) 0 #define copy_thread_group_keys(t) 0 #define exit_keys(t) do { } while(0) diff --git a/include/linux/kmod.h b/include/linux/kmod.h index 95d0e4b0814d..e4a231549407 100644 --- a/include/linux/kmod.h +++ b/include/linux/kmod.h @@ -19,6 +19,7 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ +#include #include #include #include @@ -34,7 +35,17 @@ static inline int request_module(const char * name, ...) { return -ENOSYS; } #endif #define try_then_request_module(x, mod...) ((x) ?: (request_module(mod), (x))) -extern int call_usermodehelper(char *path, char *argv[], char *envp[], int wait); + +struct key; +extern int call_usermodehelper_keys(char *path, char *argv[], char *envp[], + struct key *session_keyring, int wait); + +static inline int +call_usermodehelper(char *path, char **argv, char **envp, int wait) +{ + return call_usermodehelper_keys(path, argv, envp, NULL, wait); +} + extern void usermodehelper_init(void); #endif /* __LINUX_KMOD_H__ */ diff --git a/kernel/kmod.c b/kernel/kmod.c index eed53d4f5230..44166e3bb8af 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -120,6 +120,7 @@ struct subprocess_info { char *path; char **argv; char **envp; + struct key *ring; int wait; int retval; }; @@ -130,16 +131,21 @@ struct subprocess_info { static int ____call_usermodehelper(void *data) { struct subprocess_info *sub_info = data; + struct key *old_session; int retval; - /* Unblock all signals. */ + /* Unblock all signals and set the session keyring. */ + key_get(sub_info->ring); flush_signals(current); spin_lock_irq(¤t->sighand->siglock); + old_session = __install_session_keyring(current, sub_info->ring); flush_signal_handlers(current, 1); sigemptyset(¤t->blocked); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); + key_put(old_session); + /* We can run anywhere, unlike our parent keventd(). */ set_cpus_allowed(current, CPU_MASK_ALL); @@ -211,10 +217,11 @@ static void __call_usermodehelper(void *data) } /** - * call_usermodehelper - start a usermode application + * call_usermodehelper_keys - start a usermode application * @path: pathname for the application * @argv: null-terminated argument list * @envp: null-terminated environment list + * @session_keyring: session keyring for process (NULL for an empty keyring) * @wait: wait for the application to finish and return status. * * Runs a user-space application. The application is started @@ -224,7 +231,8 @@ static void __call_usermodehelper(void *data) * Must be called from process context. Returns a negative error code * if program was not execed successfully, or 0. */ -int call_usermodehelper(char *path, char **argv, char **envp, int wait) +int call_usermodehelper_keys(char *path, char **argv, char **envp, + struct key *session_keyring, int wait) { DECLARE_COMPLETION(done); struct subprocess_info sub_info = { @@ -232,6 +240,7 @@ int call_usermodehelper(char *path, char **argv, char **envp, int wait) .path = path, .argv = argv, .envp = envp, + .ring = session_keyring, .wait = wait, .retval = 0, }; @@ -247,7 +256,7 @@ int call_usermodehelper(char *path, char **argv, char **envp, int wait) wait_for_completion(&done); return sub_info.retval; } -EXPORT_SYMBOL(call_usermodehelper); +EXPORT_SYMBOL(call_usermodehelper_keys); void __init usermodehelper_init(void) { diff --git a/security/keys/request_key.c b/security/keys/request_key.c index 1f6c0940297f..1919540f047d 100644 --- a/security/keys/request_key.c +++ b/security/keys/request_key.c @@ -88,7 +88,7 @@ static int call_request_key(struct key *key, argv[i] = NULL; /* do it */ - return call_usermodehelper(argv[0], argv, envp, 1); + return call_usermodehelper_keys(argv[0], argv, envp, NULL, 1); } /* end call_request_key() */ -- cgit v1.2.3 From 3e30148c3d524a9c1c63ca28261bc24c457eb07a Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 23 Jun 2005 22:00:56 -0700 Subject: [PATCH] Keys: Make request-key create an authorisation key The attached patch makes the following changes: (1) There's a new special key type called ".request_key_auth". This is an authorisation key for when one process requests a key and another process is started to construct it. This type of key cannot be created by the user; nor can it be requested by kernel services. Authorisation keys hold two references: (a) Each refers to a key being constructed. When the key being constructed is instantiated the authorisation key is revoked, rendering it of no further use. (b) The "authorising process". This is either: (i) the process that called request_key(), or: (ii) if the process that called request_key() itself had an authorisation key in its session keyring, then the authorising process referred to by that authorisation key will also be referred to by the new authorisation key. This means that the process that initiated a chain of key requests will authorise the lot of them, and will, by default, wind up with the keys obtained from them in its keyrings. (2) request_key() creates an authorisation key which is then passed to /sbin/request-key in as part of a new session keyring. (3) When request_key() is searching for a key to hand back to the caller, if it comes across an authorisation key in the session keyring of the calling process, it will also search the keyrings of the process specified therein and it will use the specified process's credentials (fsuid, fsgid, groups) to do that rather than the calling process's credentials. This allows a process started by /sbin/request-key to find keys belonging to the authorising process. (4) A key can be read, even if the process executing KEYCTL_READ doesn't have direct read or search permission if that key is contained within the keyrings of a process specified by an authorisation key found within the calling process's session keyring, and is searchable using the credentials of the authorising process. This allows a process started by /sbin/request-key to read keys belonging to the authorising process. (5) The magic KEY_SPEC_*_KEYRING key IDs when passed to KEYCTL_INSTANTIATE or KEYCTL_NEGATE will specify a keyring of the authorising process, rather than the process doing the instantiation. (6) One of the process keyrings can be nominated as the default to which request_key() should attach new keys if not otherwise specified. This is done with KEYCTL_SET_REQKEY_KEYRING and one of the KEY_REQKEY_DEFL_* constants. The current setting can also be read using this call. (7) request_key() is partially interruptible. If it is waiting for another process to finish constructing a key, it can be interrupted. This permits a request-key cycle to be broken without recourse to rebooting. Signed-Off-By: David Howells Signed-Off-By: Benoit Boissinot Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/keys.txt | 34 ++++++++ include/linux/key-ui.h | 41 ++++++++- include/linux/key.h | 9 +- include/linux/keyctl.h | 11 +++ include/linux/sched.h | 8 +- kernel/sys.c | 2 +- security/keys/Makefile | 5 +- security/keys/compat.c | 7 +- security/keys/internal.h | 45 +++++++++- security/keys/key.c | 24 ++++-- security/keys/keyctl.c | 176 ++++++++++++++++++++++++------------- security/keys/keyring.c | 67 ++++++++++++-- security/keys/process_keys.c | 179 +++++++++++++++++++++++--------------- security/keys/request_key.c | 182 ++++++++++++++++++++++++++++++++------- security/keys/request_key_auth.c | 180 ++++++++++++++++++++++++++++++++++++++ 15 files changed, 779 insertions(+), 191 deletions(-) create mode 100644 security/keys/request_key_auth.c (limited to 'include/linux') diff --git a/Documentation/keys.txt b/Documentation/keys.txt index 3df40c1fe15a..0321ded4b9ae 100644 --- a/Documentation/keys.txt +++ b/Documentation/keys.txt @@ -591,6 +591,37 @@ The keyctl syscall functions are: this case too. + (*) Set the default request-key destination keyring. + + long keyctl(KEYCTL_SET_REQKEY_KEYRING, int reqkey_defl); + + This sets the default keyring to which implicitly requested keys will be + attached for this thread. reqkey_defl should be one of these constants: + + CONSTANT VALUE NEW DEFAULT KEYRING + ====================================== ====== ======================= + KEY_REQKEY_DEFL_NO_CHANGE -1 No change + KEY_REQKEY_DEFL_DEFAULT 0 Default[1] + KEY_REQKEY_DEFL_THREAD_KEYRING 1 Thread keyring + KEY_REQKEY_DEFL_PROCESS_KEYRING 2 Process keyring + KEY_REQKEY_DEFL_SESSION_KEYRING 3 Session keyring + KEY_REQKEY_DEFL_USER_KEYRING 4 User keyring + KEY_REQKEY_DEFL_USER_SESSION_KEYRING 5 User session keyring + KEY_REQKEY_DEFL_GROUP_KEYRING 6 Group keyring + + The old default will be returned if successful and error EINVAL will be + returned if reqkey_defl is not one of the above values. + + The default keyring can be overridden by the keyring indicated to the + request_key() system call. + + Note that this setting is inherited across fork/exec. + + [1] The default default is: the thread keyring if there is one, otherwise + the process keyring if there is one, otherwise the session keyring if + there is one, otherwise the user default session keyring. + + =============== KERNEL SERVICES =============== @@ -626,6 +657,9 @@ payload contents" for more information. Should the function fail error ENOKEY, EKEYEXPIRED or EKEYREVOKED will be returned. + If successful, the key will have been attached to the default keyring for + implicitly obtained request-key keys, as set by KEYCTL_SET_REQKEY_KEYRING. + (*) When it is no longer required, the key should be released using: diff --git a/include/linux/key-ui.h b/include/linux/key-ui.h index 159ca8d54e9a..cc326174a808 100644 --- a/include/linux/key-ui.h +++ b/include/linux/key-ui.h @@ -1,4 +1,4 @@ -/* key-ui.h: key userspace interface stuff for use by keyfs +/* key-ui.h: key userspace interface stuff * * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) @@ -84,8 +84,45 @@ static inline int key_any_permission(const struct key *key, key_perm_t perm) return kperm != 0; } +static inline int key_task_groups_search(struct task_struct *tsk, gid_t gid) +{ + int ret; + + task_lock(tsk); + ret = groups_search(tsk->group_info, gid); + task_unlock(tsk); + return ret; +} + +static inline int key_task_permission(const struct key *key, + struct task_struct *context, + key_perm_t perm) +{ + key_perm_t kperm; + + if (key->uid == context->fsuid) { + kperm = key->perm >> 16; + } + else if (key->gid != -1 && + key->perm & KEY_GRP_ALL && ( + key->gid == context->fsgid || + key_task_groups_search(context, key->gid) + ) + ) { + kperm = key->perm >> 8; + } + else { + kperm = key->perm; + } + + kperm = kperm & perm & KEY_ALL; + + return kperm == perm; + +} -extern struct key *lookup_user_key(key_serial_t id, int create, int part, +extern struct key *lookup_user_key(struct task_struct *context, + key_serial_t id, int create, int partial, key_perm_t perm); extern long join_session_keyring(const char *name); diff --git a/include/linux/key.h b/include/linux/key.h index 2bfbf88d2740..970bbd916cf4 100644 --- a/include/linux/key.h +++ b/include/linux/key.h @@ -199,10 +199,12 @@ extern int key_payload_reserve(struct key *key, size_t datalen); extern int key_instantiate_and_link(struct key *key, const void *data, size_t datalen, - struct key *keyring); + struct key *keyring, + struct key *instkey); extern int key_negate_and_link(struct key *key, unsigned timeout, - struct key *keyring); + struct key *keyring, + struct key *instkey); extern void key_revoke(struct key *key); extern void key_put(struct key *key); @@ -245,9 +247,6 @@ extern struct key *keyring_search(struct key *keyring, struct key_type *type, const char *description); -extern struct key *search_process_keyrings(struct key_type *type, - const char *description); - extern int keyring_add_key(struct key *keyring, struct key *key); diff --git a/include/linux/keyctl.h b/include/linux/keyctl.h index 381dedc370a3..8d7c59a29e09 100644 --- a/include/linux/keyctl.h +++ b/include/linux/keyctl.h @@ -20,6 +20,16 @@ #define KEY_SPEC_USER_SESSION_KEYRING -5 /* - key ID for UID-session keyring */ #define KEY_SPEC_GROUP_KEYRING -6 /* - key ID for GID-specific keyring */ +/* request-key default keyrings */ +#define KEY_REQKEY_DEFL_NO_CHANGE -1 +#define KEY_REQKEY_DEFL_DEFAULT 0 +#define KEY_REQKEY_DEFL_THREAD_KEYRING 1 +#define KEY_REQKEY_DEFL_PROCESS_KEYRING 2 +#define KEY_REQKEY_DEFL_SESSION_KEYRING 3 +#define KEY_REQKEY_DEFL_USER_KEYRING 4 +#define KEY_REQKEY_DEFL_USER_SESSION_KEYRING 5 +#define KEY_REQKEY_DEFL_GROUP_KEYRING 6 + /* keyctl commands */ #define KEYCTL_GET_KEYRING_ID 0 /* ask for a keyring's ID */ #define KEYCTL_JOIN_SESSION_KEYRING 1 /* join or start named session keyring */ @@ -35,5 +45,6 @@ #define KEYCTL_READ 11 /* read a key or keyring's contents */ #define KEYCTL_INSTANTIATE 12 /* instantiate a partially constructed key */ #define KEYCTL_NEGATE 13 /* negate a partially constructed key */ +#define KEYCTL_SET_REQKEY_KEYRING 14 /* set default request-key keyring */ #endif /* _LINUX_KEYCTL_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 901742f92389..2c69682b0444 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -561,9 +561,10 @@ struct group_info { groups_free(group_info); \ } while (0) -struct group_info *groups_alloc(int gidsetsize); -void groups_free(struct group_info *group_info); -int set_current_groups(struct group_info *group_info); +extern struct group_info *groups_alloc(int gidsetsize); +extern void groups_free(struct group_info *group_info); +extern int set_current_groups(struct group_info *group_info); +extern int groups_search(struct group_info *group_info, gid_t grp); /* access the groups "array" with this macro */ #define GROUP_AT(gi, i) \ ((gi)->blocks[(i)/NGROUPS_PER_BLOCK][(i)%NGROUPS_PER_BLOCK]) @@ -660,6 +661,7 @@ struct task_struct { struct user_struct *user; #ifdef CONFIG_KEYS struct key *thread_keyring; /* keyring private to this thread */ + unsigned char jit_keyring; /* default keyring to attach requested keys to */ #endif int oomkilladj; /* OOM kill score adjustment (bit shift). */ char comm[TASK_COMM_LEN]; /* executable name excluding path diff --git a/kernel/sys.c b/kernel/sys.c index 5a9d6b075016..da24bc1292db 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1259,7 +1259,7 @@ static void groups_sort(struct group_info *group_info) } /* a simple bsearch */ -static int groups_search(struct group_info *group_info, gid_t grp) +int groups_search(struct group_info *group_info, gid_t grp) { int left, right; diff --git a/security/keys/Makefile b/security/keys/Makefile index ddb495d65062..c392d750b208 100644 --- a/security/keys/Makefile +++ b/security/keys/Makefile @@ -7,8 +7,9 @@ obj-y := \ keyring.o \ keyctl.o \ process_keys.o \ - user_defined.o \ - request_key.o + request_key.o \ + request_key_auth.o \ + user_defined.o obj-$(CONFIG_KEYS_COMPAT) += compat.o obj-$(CONFIG_PROC_FS) += proc.o diff --git a/security/keys/compat.c b/security/keys/compat.c index aff8b22dcb5c..3303673c636e 100644 --- a/security/keys/compat.c +++ b/security/keys/compat.c @@ -1,6 +1,6 @@ /* compat.c: 32-bit compatibility syscall for 64-bit systems * - * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -24,7 +24,7 @@ * - if you can, you should call sys_keyctl directly */ asmlinkage long compat_sys_keyctl(u32 option, - u32 arg2, u32 arg3, u32 arg4, u32 arg5) + u32 arg2, u32 arg3, u32 arg4, u32 arg5) { switch (option) { case KEYCTL_GET_KEYRING_ID: @@ -71,6 +71,9 @@ asmlinkage long compat_sys_keyctl(u32 option, case KEYCTL_NEGATE: return keyctl_negate_key(arg2, arg3, arg4); + case KEYCTL_SET_REQKEY_KEYRING: + return keyctl_set_reqkey_keyring(arg2); + default: return -EOPNOTSUPP; } diff --git a/security/keys/internal.h b/security/keys/internal.h index 67b2b93a7489..46c8602661c9 100644 --- a/security/keys/internal.h +++ b/security/keys/internal.h @@ -1,6 +1,6 @@ /* internal.h: authentication token and access key management internal defs * - * Copyright (C) 2003 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2003-5 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -15,6 +15,16 @@ #include #include +#if 0 +#define kenter(FMT, a...) printk("==> %s("FMT")\n",__FUNCTION__ , ## a) +#define kleave(FMT, a...) printk("<== %s()"FMT"\n",__FUNCTION__ , ## a) +#define kdebug(FMT, a...) printk(FMT"\n" , ## a) +#else +#define kenter(FMT, a...) do {} while(0) +#define kleave(FMT, a...) do {} while(0) +#define kdebug(FMT, a...) do {} while(0) +#endif + extern struct key_type key_type_dead; extern struct key_type key_type_user; @@ -66,20 +76,46 @@ extern struct key *__keyring_search_one(struct key *keyring, const char *description, key_perm_t perm); +extern struct key *keyring_search_instkey(struct key *keyring, + key_serial_t target_id); + typedef int (*key_match_func_t)(const struct key *, const void *); extern struct key *keyring_search_aux(struct key *keyring, + struct task_struct *tsk, struct key_type *type, const void *description, key_match_func_t match); -extern struct key *search_process_keyrings_aux(struct key_type *type, - const void *description, - key_match_func_t match); +extern struct key *search_process_keyrings(struct key_type *type, + const void *description, + key_match_func_t match, + struct task_struct *tsk); extern struct key *find_keyring_by_name(const char *name, key_serial_t bound); extern int install_thread_keyring(struct task_struct *tsk); +extern int install_process_keyring(struct task_struct *tsk); + +extern struct key *request_key_and_link(struct key_type *type, + const char *description, + const char *callout_info, + struct key *dest_keyring); + +/* + * request_key authorisation + */ +struct request_key_auth { + struct key *target_key; + struct task_struct *context; + pid_t pid; +}; + +extern struct key_type key_type_request_key_auth; +extern struct key *request_key_auth_new(struct key *target, + struct key **_rkakey); + +extern struct key *key_get_instantiation_authkey(key_serial_t target_id); /* * keyctl functions @@ -100,6 +136,7 @@ extern long keyctl_setperm_key(key_serial_t, key_perm_t); extern long keyctl_instantiate_key(key_serial_t, const void __user *, size_t, key_serial_t); extern long keyctl_negate_key(key_serial_t, unsigned, key_serial_t); +extern long keyctl_set_reqkey_keyring(int); /* diff --git a/security/keys/key.c b/security/keys/key.c index 1fdfccb3fe43..3304d37bb379 100644 --- a/security/keys/key.c +++ b/security/keys/key.c @@ -1,6 +1,6 @@ /* key.c: basic authentication token and access key management * - * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -391,7 +391,8 @@ EXPORT_SYMBOL(key_payload_reserve); static int __key_instantiate_and_link(struct key *key, const void *data, size_t datalen, - struct key *keyring) + struct key *keyring, + struct key *instkey) { int ret, awaken; @@ -419,6 +420,10 @@ static int __key_instantiate_and_link(struct key *key, /* and link it into the destination keyring */ if (keyring) ret = __key_link(keyring, key); + + /* disable the authorisation key */ + if (instkey) + key_revoke(instkey); } } @@ -439,19 +444,21 @@ static int __key_instantiate_and_link(struct key *key, int key_instantiate_and_link(struct key *key, const void *data, size_t datalen, - struct key *keyring) + struct key *keyring, + struct key *instkey) { int ret; if (keyring) down_write(&keyring->sem); - ret = __key_instantiate_and_link(key, data, datalen, keyring); + ret = __key_instantiate_and_link(key, data, datalen, keyring, instkey); if (keyring) up_write(&keyring->sem); return ret; + } /* end key_instantiate_and_link() */ EXPORT_SYMBOL(key_instantiate_and_link); @@ -462,7 +469,8 @@ EXPORT_SYMBOL(key_instantiate_and_link); */ int key_negate_and_link(struct key *key, unsigned timeout, - struct key *keyring) + struct key *keyring, + struct key *instkey) { struct timespec now; int ret, awaken; @@ -495,6 +503,10 @@ int key_negate_and_link(struct key *key, /* and link it into the destination keyring */ if (keyring) ret = __key_link(keyring, key); + + /* disable the authorisation key */ + if (instkey) + key_revoke(instkey); } up_write(&key_construction_sem); @@ -781,7 +793,7 @@ struct key *key_create_or_update(struct key *keyring, } /* instantiate it and link it into the target keyring */ - ret = __key_instantiate_and_link(key, payload, plen, keyring); + ret = __key_instantiate_and_link(key, payload, plen, keyring, NULL); if (ret < 0) { key_put(key); key = ERR_PTR(ret); diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c index cedb7326de29..fea262860ea0 100644 --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -1,6 +1,6 @@ /* keyctl.c: userspace keyctl operations * - * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -49,6 +49,13 @@ asmlinkage long sys_add_key(const char __user *_type, goto error; type[31] = '\0'; + if (!type[0]) + goto error; + + ret = -EPERM; + if (type[0] == '.') + goto error; + ret = -EFAULT; dlen = strnlen_user(_description, PAGE_SIZE - 1); if (dlen <= 0) @@ -82,7 +89,7 @@ asmlinkage long sys_add_key(const char __user *_type, } /* find the target keyring (which must be writable) */ - keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE); + keyring = lookup_user_key(NULL, ringid, 1, 0, KEY_WRITE); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); goto error3; @@ -181,7 +188,7 @@ asmlinkage long sys_request_key(const char __user *_type, /* get the destination keyring if specified */ dest = NULL; if (destringid) { - dest = lookup_user_key(destringid, 1, 0, KEY_WRITE); + dest = lookup_user_key(NULL, destringid, 1, 0, KEY_WRITE); if (IS_ERR(dest)) { ret = PTR_ERR(dest); goto error3; @@ -196,23 +203,15 @@ asmlinkage long sys_request_key(const char __user *_type, } /* do the search */ - key = request_key(ktype, description, callout_info); + key = request_key_and_link(ktype, description, callout_info, dest); if (IS_ERR(key)) { ret = PTR_ERR(key); goto error5; } - /* link the resulting key to the destination keyring */ - if (dest) { - ret = key_link(dest, key); - if (ret < 0) - goto error6; - } - ret = key->serial; - error6: - key_put(key); + key_put(key); error5: key_type_put(ktype); error4: @@ -237,7 +236,7 @@ long keyctl_get_keyring_ID(key_serial_t id, int create) struct key *key; long ret; - key = lookup_user_key(id, create, 0, KEY_SEARCH); + key = lookup_user_key(NULL, id, create, 0, KEY_SEARCH); if (IS_ERR(key)) { ret = PTR_ERR(key); goto error; @@ -324,7 +323,7 @@ long keyctl_update_key(key_serial_t id, } /* find the target key (which must be writable) */ - key = lookup_user_key(id, 0, 0, KEY_WRITE); + key = lookup_user_key(NULL, id, 0, 0, KEY_WRITE); if (IS_ERR(key)) { ret = PTR_ERR(key); goto error2; @@ -352,7 +351,7 @@ long keyctl_revoke_key(key_serial_t id) struct key *key; long ret; - key = lookup_user_key(id, 0, 0, KEY_WRITE); + key = lookup_user_key(NULL, id, 0, 0, KEY_WRITE); if (IS_ERR(key)) { ret = PTR_ERR(key); goto error; @@ -378,7 +377,7 @@ long keyctl_keyring_clear(key_serial_t ringid) struct key *keyring; long ret; - keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE); + keyring = lookup_user_key(NULL, ringid, 1, 0, KEY_WRITE); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); goto error; @@ -404,13 +403,13 @@ long keyctl_keyring_link(key_serial_t id, key_serial_t ringid) struct key *keyring, *key; long ret; - keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE); + keyring = lookup_user_key(NULL, ringid, 1, 0, KEY_WRITE); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); goto error; } - key = lookup_user_key(id, 1, 0, KEY_LINK); + key = lookup_user_key(NULL, id, 1, 0, KEY_LINK); if (IS_ERR(key)) { ret = PTR_ERR(key); goto error2; @@ -438,13 +437,13 @@ long keyctl_keyring_unlink(key_serial_t id, key_serial_t ringid) struct key *keyring, *key; long ret; - keyring = lookup_user_key(ringid, 0, 0, KEY_WRITE); + keyring = lookup_user_key(NULL, ringid, 0, 0, KEY_WRITE); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); goto error; } - key = lookup_user_key(id, 0, 0, 0); + key = lookup_user_key(NULL, id, 0, 0, 0); if (IS_ERR(key)) { ret = PTR_ERR(key); goto error2; @@ -475,16 +474,29 @@ long keyctl_describe_key(key_serial_t keyid, char __user *buffer, size_t buflen) { - struct key *key; + struct key *key, *instkey; char *tmpbuf; long ret; - key = lookup_user_key(keyid, 0, 1, KEY_VIEW); + key = lookup_user_key(NULL, keyid, 0, 1, KEY_VIEW); if (IS_ERR(key)) { + /* viewing a key under construction is permitted if we have the + * authorisation token handy */ + if (PTR_ERR(key) == -EACCES) { + instkey = key_get_instantiation_authkey(keyid); + if (!IS_ERR(instkey)) { + key_put(instkey); + key = lookup_user_key(NULL, keyid, 0, 1, 0); + if (!IS_ERR(key)) + goto okay; + } + } + ret = PTR_ERR(key); goto error; } +okay: /* calculate how much description we're going to return */ ret = -ENOMEM; tmpbuf = kmalloc(PAGE_SIZE, GFP_KERNEL); @@ -568,7 +580,7 @@ long keyctl_keyring_search(key_serial_t ringid, goto error2; /* get the keyring at which to begin the search */ - keyring = lookup_user_key(ringid, 0, 0, KEY_SEARCH); + keyring = lookup_user_key(NULL, ringid, 0, 0, KEY_SEARCH); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); goto error2; @@ -577,7 +589,7 @@ long keyctl_keyring_search(key_serial_t ringid, /* get the destination keyring if specified */ dest = NULL; if (destringid) { - dest = lookup_user_key(destringid, 1, 0, KEY_WRITE); + dest = lookup_user_key(NULL, destringid, 1, 0, KEY_WRITE); if (IS_ERR(dest)) { ret = PTR_ERR(dest); goto error3; @@ -656,24 +668,23 @@ long keyctl_read_key(key_serial_t keyid, char __user *buffer, size_t buflen) long ret; /* find the key first */ - key = lookup_user_key(keyid, 0, 0, 0); + key = lookup_user_key(NULL, keyid, 0, 0, 0); if (!IS_ERR(key)) { /* see if we can read it directly */ if (key_permission(key, KEY_READ)) goto can_read_key; - /* can't; see if it's searchable from this process's - * keyrings */ - ret = -ENOKEY; - if (key_permission(key, KEY_SEARCH)) { - /* okay - we do have search permission on the key - * itself, but do we have the key? */ - skey = search_process_keyrings_aux(key->type, key, - keyctl_read_key_same); - if (!IS_ERR(skey)) - goto can_read_key2; - } - + /* we can't; see if it's searchable from this process's + * keyrings + * - we automatically take account of the fact that it may be + * dangling off an instantiation key + */ + skey = search_process_keyrings(key->type, key, + keyctl_read_key_same, current); + if (!IS_ERR(skey)) + goto can_read_key2; + + ret = PTR_ERR(skey); goto error2; } @@ -719,7 +730,7 @@ long keyctl_chown_key(key_serial_t id, uid_t uid, gid_t gid) if (uid == (uid_t) -1 && gid == (gid_t) -1) goto error; - key = lookup_user_key(id, 1, 1, 0); + key = lookup_user_key(NULL, id, 1, 1, 0); if (IS_ERR(key)) { ret = PTR_ERR(key); goto error; @@ -776,7 +787,7 @@ long keyctl_setperm_key(key_serial_t id, key_perm_t perm) if (perm & ~(KEY_USR_ALL | KEY_GRP_ALL | KEY_OTH_ALL)) goto error; - key = lookup_user_key(id, 1, 1, 0); + key = lookup_user_key(NULL, id, 1, 1, 0); if (IS_ERR(key)) { ret = PTR_ERR(key); goto error; @@ -809,7 +820,8 @@ long keyctl_instantiate_key(key_serial_t id, size_t plen, key_serial_t ringid) { - struct key *key, *keyring; + struct request_key_auth *rka; + struct key *instkey, *keyring; void *payload; long ret; @@ -831,18 +843,21 @@ long keyctl_instantiate_key(key_serial_t id, goto error2; } - /* find the target key (which must be writable) */ - key = lookup_user_key(id, 0, 1, KEY_WRITE); - if (IS_ERR(key)) { - ret = PTR_ERR(key); + /* find the instantiation authorisation key */ + instkey = key_get_instantiation_authkey(id); + if (IS_ERR(instkey)) { + ret = PTR_ERR(instkey); goto error2; } - /* find the destination keyring if present (which must also be - * writable) */ + rka = instkey->payload.data; + + /* find the destination keyring amongst those belonging to the + * requesting task */ keyring = NULL; if (ringid) { - keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE); + keyring = lookup_user_key(rka->context, ringid, 1, 0, + KEY_WRITE); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); goto error3; @@ -850,11 +865,12 @@ long keyctl_instantiate_key(key_serial_t id, } /* instantiate the key and link it into a keyring */ - ret = key_instantiate_and_link(key, payload, plen, keyring); + ret = key_instantiate_and_link(rka->target_key, payload, plen, + keyring, instkey); key_put(keyring); error3: - key_put(key); + key_put(instkey); error2: kfree(payload); error: @@ -869,21 +885,24 @@ long keyctl_instantiate_key(key_serial_t id, */ long keyctl_negate_key(key_serial_t id, unsigned timeout, key_serial_t ringid) { - struct key *key, *keyring; + struct request_key_auth *rka; + struct key *instkey, *keyring; long ret; - /* find the target key (which must be writable) */ - key = lookup_user_key(id, 0, 1, KEY_WRITE); - if (IS_ERR(key)) { - ret = PTR_ERR(key); + /* find the instantiation authorisation key */ + instkey = key_get_instantiation_authkey(id); + if (IS_ERR(instkey)) { + ret = PTR_ERR(instkey); goto error; } + rka = instkey->payload.data; + /* find the destination keyring if present (which must also be * writable) */ keyring = NULL; if (ringid) { - keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE); + keyring = lookup_user_key(NULL, ringid, 1, 0, KEY_WRITE); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); goto error2; @@ -891,16 +910,54 @@ long keyctl_negate_key(key_serial_t id, unsigned timeout, key_serial_t ringid) } /* instantiate the key and link it into a keyring */ - ret = key_negate_and_link(key, timeout, keyring); + ret = key_negate_and_link(rka->target_key, timeout, keyring, instkey); key_put(keyring); error2: - key_put(key); + key_put(instkey); error: return ret; } /* end keyctl_negate_key() */ +/*****************************************************************************/ +/* + * set the default keyring in which request_key() will cache keys + * - return the old setting + */ +long keyctl_set_reqkey_keyring(int reqkey_defl) +{ + int ret; + + switch (reqkey_defl) { + case KEY_REQKEY_DEFL_THREAD_KEYRING: + ret = install_thread_keyring(current); + if (ret < 0) + return ret; + goto set; + + case KEY_REQKEY_DEFL_PROCESS_KEYRING: + ret = install_process_keyring(current); + if (ret < 0) + return ret; + + case KEY_REQKEY_DEFL_DEFAULT: + case KEY_REQKEY_DEFL_SESSION_KEYRING: + case KEY_REQKEY_DEFL_USER_KEYRING: + case KEY_REQKEY_DEFL_USER_SESSION_KEYRING: + set: + current->jit_keyring = reqkey_defl; + + case KEY_REQKEY_DEFL_NO_CHANGE: + return current->jit_keyring; + + case KEY_REQKEY_DEFL_GROUP_KEYRING: + default: + return -EINVAL; + } + +} /* end keyctl_set_reqkey_keyring() */ + /*****************************************************************************/ /* * the key control system call @@ -971,6 +1028,9 @@ asmlinkage long sys_keyctl(int option, unsigned long arg2, unsigned long arg3, (unsigned) arg3, (key_serial_t) arg4); + case KEYCTL_SET_REQKEY_KEYRING: + return keyctl_set_reqkey_keyring(arg2); + default: return -EOPNOTSUPP; } diff --git a/security/keys/keyring.c b/security/keys/keyring.c index c9a5de197487..90a551e4da66 100644 --- a/security/keys/keyring.c +++ b/security/keys/keyring.c @@ -1,6 +1,6 @@ /* keyring.c: keyring handling * - * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -308,7 +308,7 @@ struct key *keyring_alloc(const char *description, uid_t uid, gid_t gid, uid, gid, KEY_USR_ALL, not_in_quota); if (!IS_ERR(keyring)) { - ret = key_instantiate_and_link(keyring, NULL, 0, dest); + ret = key_instantiate_and_link(keyring, NULL, 0, dest, NULL); if (ret < 0) { key_put(keyring); keyring = ERR_PTR(ret); @@ -326,11 +326,12 @@ struct key *keyring_alloc(const char *description, uid_t uid, gid_t gid, * - we only find keys on which we have search permission * - we use the supplied match function to see if the description (or other * feature of interest) matches - * - we readlock the keyrings as we search down the tree + * - we rely on RCU to prevent the keyring lists from disappearing on us * - we return -EAGAIN if we didn't find any matching key * - we return -ENOKEY if we only found negative matching keys */ struct key *keyring_search_aux(struct key *keyring, + struct task_struct *context, struct key_type *type, const void *description, key_match_func_t match) @@ -352,7 +353,7 @@ struct key *keyring_search_aux(struct key *keyring, /* top keyring must have search permission to begin the search */ key = ERR_PTR(-EACCES); - if (!key_permission(keyring, KEY_SEARCH)) + if (!key_task_permission(keyring, context, KEY_SEARCH)) goto error; key = ERR_PTR(-ENOTDIR); @@ -392,7 +393,7 @@ struct key *keyring_search_aux(struct key *keyring, continue; /* key must have search permissions */ - if (!key_permission(key, KEY_SEARCH)) + if (!key_task_permission(key, context, KEY_SEARCH)) continue; /* we set a different error code if we find a negative key */ @@ -418,7 +419,7 @@ struct key *keyring_search_aux(struct key *keyring, if (sp >= KEYRING_SEARCH_MAX_DEPTH) continue; - if (!key_permission(key, KEY_SEARCH)) + if (!key_task_permission(key, context, KEY_SEARCH)) continue; /* stack the current position */ @@ -468,7 +469,11 @@ struct key *keyring_search(struct key *keyring, struct key_type *type, const char *description) { - return keyring_search_aux(keyring, type, description, type->match); + if (!type->match) + return ERR_PTR(-ENOKEY); + + return keyring_search_aux(keyring, current, + type, description, type->match); } /* end keyring_search() */ @@ -496,7 +501,8 @@ struct key *__keyring_search_one(struct key *keyring, key = klist->keys[loop]; if (key->type == ktype && - key->type->match(key, description) && + (!key->type->match || + key->type->match(key, description)) && key_permission(key, perm) && !test_bit(KEY_FLAG_REVOKED, &key->flags) ) @@ -515,6 +521,51 @@ struct key *__keyring_search_one(struct key *keyring, } /* end __keyring_search_one() */ +/*****************************************************************************/ +/* + * search for an instantiation authorisation key matching a target key + * - the RCU read lock must be held by the caller + * - a target_id of zero specifies any valid token + */ +struct key *keyring_search_instkey(struct key *keyring, + key_serial_t target_id) +{ + struct request_key_auth *rka; + struct keyring_list *klist; + struct key *instkey; + int loop; + + klist = rcu_dereference(keyring->payload.subscriptions); + if (klist) { + for (loop = 0; loop < klist->nkeys; loop++) { + instkey = klist->keys[loop]; + + if (instkey->type != &key_type_request_key_auth) + continue; + + rka = instkey->payload.data; + if (target_id && rka->target_key->serial != target_id) + continue; + + /* the auth key is revoked during instantiation */ + if (!test_bit(KEY_FLAG_REVOKED, &instkey->flags)) + goto found; + + instkey = ERR_PTR(-EKEYREVOKED); + goto error; + } + } + + instkey = ERR_PTR(-EACCES); + goto error; + +found: + atomic_inc(&instkey->usage); +error: + return instkey; + +} /* end keyring_search_instkey() */ + /*****************************************************************************/ /* * find a keyring with the specified name diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c index 972e30172687..34db087bbcc7 100644 --- a/security/keys/process_keys.c +++ b/security/keys/process_keys.c @@ -165,7 +165,7 @@ int install_thread_keyring(struct task_struct *tsk) /* * make sure a process keyring is installed */ -static int install_process_keyring(struct task_struct *tsk) +int install_process_keyring(struct task_struct *tsk) { unsigned long flags; struct key *keyring; @@ -376,12 +376,13 @@ void key_fsgid_changed(struct task_struct *tsk) * - we return -EAGAIN if we didn't find any matching key * - we return -ENOKEY if we found only negative matching keys */ -struct key *search_process_keyrings_aux(struct key_type *type, - const void *description, - key_match_func_t match) +struct key *search_process_keyrings(struct key_type *type, + const void *description, + key_match_func_t match, + struct task_struct *context) { - struct task_struct *tsk = current; - struct key *key, *ret, *err; + struct request_key_auth *rka; + struct key *key, *ret, *err, *instkey; /* we want to return -EAGAIN or -ENOKEY if any of the keyrings were * searchable, but we failed to find a key or we found a negative key; @@ -395,9 +396,9 @@ struct key *search_process_keyrings_aux(struct key_type *type, err = ERR_PTR(-EAGAIN); /* search the thread keyring first */ - if (tsk->thread_keyring) { - key = keyring_search_aux(tsk->thread_keyring, type, - description, match); + if (context->thread_keyring) { + key = keyring_search_aux(context->thread_keyring, + context, type, description, match); if (!IS_ERR(key)) goto found; @@ -415,9 +416,9 @@ struct key *search_process_keyrings_aux(struct key_type *type, } /* search the process keyring second */ - if (tsk->signal->process_keyring) { - key = keyring_search_aux(tsk->signal->process_keyring, - type, description, match); + if (context->signal->process_keyring) { + key = keyring_search_aux(context->signal->process_keyring, + context, type, description, match); if (!IS_ERR(key)) goto found; @@ -434,53 +435,93 @@ struct key *search_process_keyrings_aux(struct key_type *type, } } - /* search the session keyring last */ - if (tsk->signal->session_keyring) { + /* search the session keyring */ + if (context->signal->session_keyring) { rcu_read_lock(); key = keyring_search_aux( - rcu_dereference(tsk->signal->session_keyring), - type, description, match); + rcu_dereference(context->signal->session_keyring), + context, type, description, match); rcu_read_unlock(); + + if (!IS_ERR(key)) + goto found; + + switch (PTR_ERR(key)) { + case -EAGAIN: /* no key */ + if (ret) + break; + case -ENOKEY: /* negative key */ + ret = key; + break; + default: + err = key; + break; + } + + /* if this process has a session keyring and that has an + * instantiation authorisation key in the bottom level, then we + * also search the keyrings of the process mentioned there */ + if (context != current) + goto no_key; + + rcu_read_lock(); + instkey = __keyring_search_one( + rcu_dereference(context->signal->session_keyring), + &key_type_request_key_auth, NULL, 0); + rcu_read_unlock(); + + if (IS_ERR(instkey)) + goto no_key; + + rka = instkey->payload.data; + + key = search_process_keyrings(type, description, match, + rka->context); + key_put(instkey); + + if (!IS_ERR(key)) + goto found; + + switch (PTR_ERR(key)) { + case -EAGAIN: /* no key */ + if (ret) + break; + case -ENOKEY: /* negative key */ + ret = key; + break; + default: + err = key; + break; + } } + /* or search the user-session keyring */ else { - key = keyring_search_aux(tsk->user->session_keyring, - type, description, match); - } - - if (!IS_ERR(key)) - goto found; + key = keyring_search_aux(context->user->session_keyring, + context, type, description, match); + if (!IS_ERR(key)) + goto found; - switch (PTR_ERR(key)) { - case -EAGAIN: /* no key */ - if (ret) + switch (PTR_ERR(key)) { + case -EAGAIN: /* no key */ + if (ret) + break; + case -ENOKEY: /* negative key */ + ret = key; break; - case -ENOKEY: /* negative key */ - ret = key; - break; - default: - err = key; - break; + default: + err = key; + break; + } } + +no_key: /* no key - decide on the error we're going to go for */ key = ret ? ret : err; - found: +found: return key; -} /* end search_process_keyrings_aux() */ - -/*****************************************************************************/ -/* - * search the process keyrings for the first matching key - * - we return -EAGAIN if we didn't find any matching key - * - we return -ENOKEY if we found only negative matching keys - */ -struct key *search_process_keyrings(struct key_type *type, - const char *description) -{ - return search_process_keyrings_aux(type, description, type->match); - } /* end search_process_keyrings() */ /*****************************************************************************/ @@ -489,72 +530,73 @@ struct key *search_process_keyrings(struct key_type *type, * - don't create special keyrings unless so requested * - partially constructed keys aren't found unless requested */ -struct key *lookup_user_key(key_serial_t id, int create, int partial, - key_perm_t perm) +struct key *lookup_user_key(struct task_struct *context, key_serial_t id, + int create, int partial, key_perm_t perm) { - struct task_struct *tsk = current; - unsigned long flags; struct key *key; int ret; + if (!context) + context = current; + key = ERR_PTR(-ENOKEY); switch (id) { case KEY_SPEC_THREAD_KEYRING: - if (!tsk->thread_keyring) { + if (!context->thread_keyring) { if (!create) goto error; - ret = install_thread_keyring(tsk); + ret = install_thread_keyring(context); if (ret < 0) { key = ERR_PTR(ret); goto error; } } - key = tsk->thread_keyring; + key = context->thread_keyring; atomic_inc(&key->usage); break; case KEY_SPEC_PROCESS_KEYRING: - if (!tsk->signal->process_keyring) { + if (!context->signal->process_keyring) { if (!create) goto error; - ret = install_process_keyring(tsk); + ret = install_process_keyring(context); if (ret < 0) { key = ERR_PTR(ret); goto error; } } - key = tsk->signal->process_keyring; + key = context->signal->process_keyring; atomic_inc(&key->usage); break; case KEY_SPEC_SESSION_KEYRING: - if (!tsk->signal->session_keyring) { + if (!context->signal->session_keyring) { /* always install a session keyring upon access if one * doesn't exist yet */ ret = install_session_keyring( - tsk, tsk->user->session_keyring); + context, context->user->session_keyring); if (ret < 0) goto error; } - spin_lock_irqsave(&tsk->sighand->siglock, flags); - key = tsk->signal->session_keyring; + rcu_read_lock(); + key = rcu_dereference(context->signal->session_keyring); atomic_inc(&key->usage); - spin_unlock_irqrestore(&tsk->sighand->siglock, flags); + rcu_read_unlock(); break; case KEY_SPEC_USER_KEYRING: - key = tsk->user->uid_keyring; + key = context->user->uid_keyring; atomic_inc(&key->usage); break; case KEY_SPEC_USER_SESSION_KEYRING: - key = tsk->user->session_keyring; + key = context->user->session_keyring; atomic_inc(&key->usage); break; @@ -574,7 +616,7 @@ struct key *lookup_user_key(key_serial_t id, int create, int partial, break; } - /* check the status and permissions */ + /* check the status */ if (perm) { ret = key_validate(key); if (ret < 0) @@ -585,8 +627,10 @@ struct key *lookup_user_key(key_serial_t id, int create, int partial, if (!partial && !test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) goto invalid_key; + /* check the permissions */ ret = -EACCES; - if (!key_permission(key, perm)) + + if (!key_task_permission(key, context, perm)) goto invalid_key; error: @@ -609,7 +653,6 @@ struct key *lookup_user_key(key_serial_t id, int create, int partial, long join_session_keyring(const char *name) { struct task_struct *tsk = current; - unsigned long flags; struct key *keyring; long ret; @@ -619,9 +662,9 @@ long join_session_keyring(const char *name) if (ret < 0) goto error; - spin_lock_irqsave(&tsk->sighand->siglock, flags); - ret = tsk->signal->session_keyring->serial; - spin_unlock_irqrestore(&tsk->sighand->siglock, flags); + rcu_read_lock(); + ret = rcu_dereference(tsk->signal->session_keyring)->serial; + rcu_read_unlock(); goto error; } diff --git a/security/keys/request_key.c b/security/keys/request_key.c index 54aa7b70e63b..dfcd983af1fd 100644 --- a/security/keys/request_key.c +++ b/security/keys/request_key.c @@ -1,6 +1,6 @@ /* request_key.c: request a key from userspace * - * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -13,6 +13,7 @@ #include #include #include +#include #include "internal.h" struct key_construction { @@ -27,18 +28,26 @@ DECLARE_WAIT_QUEUE_HEAD(request_key_conswq); /* * request userspace finish the construction of a key * - execute "/sbin/request-key " - * - if callout_info is an empty string, it'll be rendered as a "-" instead */ static int call_request_key(struct key *key, const char *op, const char *callout_info) { struct task_struct *tsk = current; - unsigned long flags; key_serial_t prkey, sskey; + struct key *session_keyring, *rkakey; char *argv[10], *envp[3], uid_str[12], gid_str[12]; char key_str[12], keyring_str[3][12]; - int i; + int ret, i; + + kenter("{%d},%s,%s", key->serial, op, callout_info); + + /* generate a new session keyring with an auth key in it */ + session_keyring = request_key_auth_new(key, &rkakey); + if (IS_ERR(session_keyring)) { + ret = PTR_ERR(session_keyring); + goto error; + } /* record the UID and GID */ sprintf(uid_str, "%d", current->fsuid); @@ -55,17 +64,17 @@ static int call_request_key(struct key *key, if (tsk->signal->process_keyring) prkey = tsk->signal->process_keyring->serial; - sskey = 0; - spin_lock_irqsave(&tsk->sighand->siglock, flags); - if (tsk->signal->session_keyring) - sskey = tsk->signal->session_keyring->serial; - spin_unlock_irqrestore(&tsk->sighand->siglock, flags); - + sprintf(keyring_str[1], "%d", prkey); - if (!sskey) + if (tsk->signal->session_keyring) { + rcu_read_lock(); + sskey = rcu_dereference(tsk->signal->session_keyring)->serial; + rcu_read_unlock(); + } + else { sskey = tsk->user->session_keyring->serial; + } - sprintf(keyring_str[1], "%d", prkey); sprintf(keyring_str[2], "%d", sskey); /* set up a minimal environment */ @@ -84,11 +93,20 @@ static int call_request_key(struct key *key, argv[i++] = keyring_str[0]; argv[i++] = keyring_str[1]; argv[i++] = keyring_str[2]; - argv[i++] = callout_info[0] ? (char *) callout_info : "-"; + argv[i++] = (char *) callout_info; argv[i] = NULL; /* do it */ - return call_usermodehelper_keys(argv[0], argv, envp, NULL, 1); + ret = call_usermodehelper_keys(argv[0], argv, envp, session_keyring, 1); + + /* dispose of the special keys */ + key_revoke(rkakey); + key_put(rkakey); + key_put(session_keyring); + + error: + kleave(" = %d", ret); + return ret; } /* end call_request_key() */ @@ -107,6 +125,8 @@ static struct key *__request_key_construction(struct key_type *type, struct key *key; int ret, negated; + kenter("%s,%s,%s", type->name, description, callout_info); + /* create a key and add it to the queue */ key = key_alloc(type, description, current->fsuid, current->fsgid, KEY_USR_ALL, 0); @@ -143,6 +163,7 @@ static struct key *__request_key_construction(struct key_type *type, } out: + kleave(" = %p", key); return key; request_failed: @@ -216,6 +237,9 @@ static struct key *request_key_construction(struct key_type *type, DECLARE_WAITQUEUE(myself, current); + kenter("%s,%s,{%d},%s", + type->name, description, user->uid, callout_info); + /* see if there's such a key under construction already */ down_write(&key_construction_sem); @@ -232,6 +256,7 @@ static struct key *request_key_construction(struct key_type *type, /* see about getting userspace to construct the key */ key = __request_key_construction(type, description, callout_info); error: + kleave(" = %p", key); return key; /* someone else has the same key under construction @@ -245,9 +270,11 @@ static struct key *request_key_construction(struct key_type *type, add_wait_queue(&request_key_conswq, &myself); for (;;) { - set_current_state(TASK_UNINTERRUPTIBLE); + set_current_state(TASK_INTERRUPTIBLE); if (!test_bit(KEY_FLAG_USER_CONSTRUCT, &ckey->flags)) break; + if (signal_pending(current)) + break; schedule(); } @@ -265,23 +292,85 @@ static struct key *request_key_construction(struct key_type *type, } /* end request_key_construction() */ +/*****************************************************************************/ +/* + * link a freshly minted key to an appropriate destination keyring + */ +static void request_key_link(struct key *key, struct key *dest_keyring) +{ + struct task_struct *tsk = current; + struct key *drop = NULL; + + kenter("{%d},%p", key->serial, dest_keyring); + + /* find the appropriate keyring */ + if (!dest_keyring) { + switch (tsk->jit_keyring) { + case KEY_REQKEY_DEFL_DEFAULT: + case KEY_REQKEY_DEFL_THREAD_KEYRING: + dest_keyring = tsk->thread_keyring; + if (dest_keyring) + break; + + case KEY_REQKEY_DEFL_PROCESS_KEYRING: + dest_keyring = tsk->signal->process_keyring; + if (dest_keyring) + break; + + case KEY_REQKEY_DEFL_SESSION_KEYRING: + rcu_read_lock(); + dest_keyring = key_get( + rcu_dereference(tsk->signal->session_keyring)); + rcu_read_unlock(); + drop = dest_keyring; + + if (dest_keyring) + break; + + case KEY_REQKEY_DEFL_USER_SESSION_KEYRING: + dest_keyring = current->user->session_keyring; + break; + + case KEY_REQKEY_DEFL_USER_KEYRING: + dest_keyring = current->user->uid_keyring; + break; + + case KEY_REQKEY_DEFL_GROUP_KEYRING: + default: + BUG(); + } + } + + /* and attach the key to it */ + key_link(dest_keyring, key); + + key_put(drop); + + kleave(""); + +} /* end request_key_link() */ + /*****************************************************************************/ /* * request a key * - search the process's keyrings * - check the list of keys being created or updated - * - call out to userspace for a key if requested (supplementary info can be - * passed) + * - call out to userspace for a key if supplementary info was provided + * - cache the key in an appropriate keyring */ -struct key *request_key(struct key_type *type, - const char *description, - const char *callout_info) +struct key *request_key_and_link(struct key_type *type, + const char *description, + const char *callout_info, + struct key *dest_keyring) { struct key_user *user; struct key *key; + kenter("%s,%s,%s,%p", + type->name, description, callout_info, dest_keyring); + /* search all the process keyrings for a key */ - key = search_process_keyrings_aux(type, description, type->match); + key = search_process_keyrings(type, description, type->match, current); if (PTR_ERR(key) == -EAGAIN) { /* the search failed, but the keyrings were searchable, so we @@ -292,12 +381,13 @@ struct key *request_key(struct key_type *type, /* - get hold of the user's construction queue */ user = key_user_lookup(current->fsuid); - if (!user) { - key = ERR_PTR(-ENOMEM); - goto error; - } + if (!user) + goto nomem; + + do { + if (signal_pending(current)) + goto interrupted; - for (;;) { /* ask userspace (returns NULL if it waited on a key * being constructed) */ key = request_key_construction(type, description, @@ -307,18 +397,46 @@ struct key *request_key(struct key_type *type, /* someone else made the key we want, so we need to * search again as it might now be available to us */ - key = search_process_keyrings_aux(type, description, - type->match); - if (PTR_ERR(key) != -EAGAIN) - break; - } + key = search_process_keyrings(type, description, + type->match, current); + + } while (PTR_ERR(key) == -EAGAIN); key_user_put(user); + + /* link the new key into the appropriate keyring */ + if (!PTR_ERR(key)) + request_key_link(key, dest_keyring); } - error: +error: + kleave(" = %p", key); return key; +nomem: + key = ERR_PTR(-ENOMEM); + goto error; + +interrupted: + key_user_put(user); + key = ERR_PTR(-EINTR); + goto error; + +} /* end request_key_and_link() */ + +/*****************************************************************************/ +/* + * request a key + * - search the process's keyrings + * - check the list of keys being created or updated + * - call out to userspace for a key if supplementary info was provided + */ +struct key *request_key(struct key_type *type, + const char *description, + const char *callout_info) +{ + return request_key_and_link(type, description, callout_info, NULL); + } /* end request_key() */ EXPORT_SYMBOL(request_key); diff --git a/security/keys/request_key_auth.c b/security/keys/request_key_auth.c new file mode 100644 index 000000000000..f22264632229 --- /dev/null +++ b/security/keys/request_key_auth.c @@ -0,0 +1,180 @@ +/* request_key_auth.c: request key authorisation controlling key def + * + * Copyright (C) 2005 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include +#include "internal.h" + +static int request_key_auth_instantiate(struct key *, const void *, size_t); +static void request_key_auth_describe(const struct key *, struct seq_file *); +static void request_key_auth_destroy(struct key *); + +/* + * the request-key authorisation key type definition + */ +struct key_type key_type_request_key_auth = { + .name = ".request_key_auth", + .def_datalen = sizeof(struct request_key_auth), + .instantiate = request_key_auth_instantiate, + .describe = request_key_auth_describe, + .destroy = request_key_auth_destroy, +}; + +/*****************************************************************************/ +/* + * instantiate a request-key authorisation record + */ +static int request_key_auth_instantiate(struct key *key, + const void *data, + size_t datalen) +{ + struct request_key_auth *rka, *irka; + struct key *instkey; + int ret; + + ret = -ENOMEM; + rka = kmalloc(sizeof(*rka), GFP_KERNEL); + if (rka) { + /* see if the calling process is already servicing the key + * request of another process */ + instkey = key_get_instantiation_authkey(0); + if (!IS_ERR(instkey)) { + /* it is - use that instantiation context here too */ + irka = instkey->payload.data; + rka->context = irka->context; + rka->pid = irka->pid; + key_put(instkey); + } + else { + /* it isn't - use this process as the context */ + rka->context = current; + rka->pid = current->pid; + } + + rka->target_key = key_get((struct key *) data); + key->payload.data = rka; + ret = 0; + } + + return ret; + +} /* end request_key_auth_instantiate() */ + +/*****************************************************************************/ +/* + * + */ +static void request_key_auth_describe(const struct key *key, + struct seq_file *m) +{ + struct request_key_auth *rka = key->payload.data; + + seq_puts(m, "key:"); + seq_puts(m, key->description); + seq_printf(m, " pid:%d", rka->pid); + +} /* end request_key_auth_describe() */ + +/*****************************************************************************/ +/* + * destroy an instantiation authorisation token key + */ +static void request_key_auth_destroy(struct key *key) +{ + struct request_key_auth *rka = key->payload.data; + + kenter("{%d}", key->serial); + + key_put(rka->target_key); + +} /* end request_key_auth_destroy() */ + +/*****************************************************************************/ +/* + * create a session keyring to be for the invokation of /sbin/request-key and + * stick an authorisation token in it + */ +struct key *request_key_auth_new(struct key *target, struct key **_rkakey) +{ + struct key *keyring, *rkakey = NULL; + char desc[20]; + int ret; + + kenter("%d,", target->serial); + + /* allocate a new session keyring */ + sprintf(desc, "_req.%u", target->serial); + + keyring = keyring_alloc(desc, current->fsuid, current->fsgid, 1, NULL); + if (IS_ERR(keyring)) { + kleave("= %ld", PTR_ERR(keyring)); + return keyring; + } + + /* allocate the auth key */ + sprintf(desc, "%x", target->serial); + + rkakey = key_alloc(&key_type_request_key_auth, desc, + current->fsuid, current->fsgid, + KEY_USR_VIEW, 1); + if (IS_ERR(rkakey)) { + key_put(keyring); + kleave("= %ld", PTR_ERR(rkakey)); + return rkakey; + } + + /* construct and attach to the keyring */ + ret = key_instantiate_and_link(rkakey, target, 0, keyring, NULL); + if (ret < 0) { + key_revoke(rkakey); + key_put(rkakey); + key_put(keyring); + kleave("= %d", ret); + return ERR_PTR(ret); + } + + *_rkakey = rkakey; + kleave(" = {%d} ({%d})", keyring->serial, rkakey->serial); + return keyring; + +} /* end request_key_auth_new() */ + +/*****************************************************************************/ +/* + * get the authorisation key for instantiation of a specific key if attached to + * the current process's keyrings + * - this key is inserted into a keyring and that is set as /sbin/request-key's + * session keyring + * - a target_id of zero specifies any valid token + */ +struct key *key_get_instantiation_authkey(key_serial_t target_id) +{ + struct task_struct *tsk = current; + struct key *instkey; + + /* we must have our own personal session keyring */ + if (!tsk->signal->session_keyring) + return ERR_PTR(-EACCES); + + /* and it must contain a suitable request authorisation key + * - lock RCU against session keyring changing + */ + rcu_read_lock(); + + instkey = keyring_search_instkey( + rcu_dereference(tsk->signal->session_keyring), target_id); + + rcu_read_unlock(); + return instkey; + +} /* end key_get_instantiation_authkey() */ -- cgit v1.2.3 From 92198f7eaa5df3479341dd8fa20c2c81aa3b1e25 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Jun 2005 22:00:59 -0700 Subject: [PATCH] pass iocb to dio_iodone_t XFS will have to look at iocb->private to fix aio+dio. No other filesystem is using the blockdev_direct_IO* end_io callback. Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/direct-io.c | 2 +- fs/xfs/linux-2.6/xfs_aops.c | 3 ++- include/linux/fs.h | 4 ++-- 3 files changed, 5 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/fs/direct-io.c b/fs/direct-io.c index 1d55e7e67342..0d06097bc995 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -215,7 +215,7 @@ static struct page *dio_get_page(struct dio *dio) static void dio_complete(struct dio *dio, loff_t offset, ssize_t bytes) { if (dio->end_io && dio->result) - dio->end_io(dio->inode, offset, bytes, dio->map_bh.b_private); + dio->end_io(dio->iocb, offset, bytes, dio->map_bh.b_private); if (dio->lock_type == DIO_LOCKING) up_read(&dio->inode->i_alloc_sem); } diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 93ce257cd149..a3a4b5aaf5d9 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -149,11 +149,12 @@ linvfs_unwritten_convert( */ STATIC void linvfs_unwritten_convert_direct( - struct inode *inode, + struct kiocb *iocb, loff_t offset, ssize_t size, void *private) { + struct inode *inode = iocb->ki_filp->f_dentry->d_inode; ASSERT(!private || inode == (struct inode *)private); /* private indicates an unwritten extent lay beneath this IO */ diff --git a/include/linux/fs.h b/include/linux/fs.h index 517bf4966bf5..83857d8070d3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -220,6 +220,7 @@ extern int dir_notify_enable; struct iovec; struct nameidata; +struct kiocb; struct pipe_inode_info; struct poll_table_struct; struct kstatfs; @@ -240,7 +241,7 @@ typedef int (get_block_t)(struct inode *inode, sector_t iblock, typedef int (get_blocks_t)(struct inode *inode, sector_t iblock, unsigned long max_blocks, struct buffer_head *bh_result, int create); -typedef void (dio_iodone_t)(struct inode *inode, loff_t offset, +typedef void (dio_iodone_t)(struct kiocb *iocb, loff_t offset, ssize_t bytes, void *private); /* @@ -302,7 +303,6 @@ struct iattr { struct page; struct address_space; struct writeback_control; -struct kiocb; struct address_space_operations { int (*writepage)(struct page *page, struct writeback_control *wbc); -- cgit v1.2.3 From 4e5117ba0af4582b6ec9164874f719d7f3f1eb2b Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 23 Jun 2005 22:01:03 -0700 Subject: [PATCH] quota: improve credits estimates Improve estimates on the number of needed credits for quota transaction. Now we distinguish blocks that might need to be allocated and blocks that only need to be rewritten. Also we distinguish deleting of a quota structure and creating of a new one. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/dqblk_v1.h | 6 ++++++ include/linux/dqblk_v2.h | 6 ++++++ include/linux/quota.h | 7 +++++-- 3 files changed, 17 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dqblk_v1.h b/include/linux/dqblk_v1.h index 42fbf4797156..57f1250d5a52 100644 --- a/include/linux/dqblk_v1.h +++ b/include/linux/dqblk_v1.h @@ -11,6 +11,12 @@ /* Root squash turned on */ #define V1_DQF_RSQUASH 1 +/* Numbers of blocks needed for updates */ +#define V1_INIT_ALLOC 1 +#define V1_INIT_REWRITE 1 +#define V1_DEL_ALLOC 0 +#define V1_DEL_REWRITE 2 + /* Special information about quotafile */ struct v1_mem_dqinfo { }; diff --git a/include/linux/dqblk_v2.h b/include/linux/dqblk_v2.h index 4a6c5f6867bb..4f853322cb7f 100644 --- a/include/linux/dqblk_v2.h +++ b/include/linux/dqblk_v2.h @@ -10,6 +10,12 @@ /* id numbers of quota format */ #define QFMT_VFS_V0 2 +/* Numbers of blocks needed for updates */ +#define V2_INIT_ALLOC 4 +#define V2_INIT_REWRITE 2 +#define V2_DEL_ALLOC 0 +#define V2_DEL_REWRITE 6 + /* Inmemory copy of version specific information */ struct v2_mem_dqinfo { unsigned int dqi_blocks; diff --git a/include/linux/quota.h b/include/linux/quota.h index ac5b90f4f256..700ead45084f 100644 --- a/include/linux/quota.h +++ b/include/linux/quota.h @@ -138,8 +138,11 @@ struct if_dqinfo { #include /* Maximal numbers of writes for quota operation (insert/delete/update) - * (over all formats) - info block, 4 pointer blocks, data block */ -#define DQUOT_MAX_WRITES 6 + * (over VFS all formats) */ +#define DQUOT_INIT_ALLOC max(V1_INIT_ALLOC, V2_INIT_ALLOC) +#define DQUOT_INIT_REWRITE max(V1_INIT_REWRITE, V2_INIT_REWRITE) +#define DQUOT_DEL_ALLOC max(V1_DEL_ALLOC, V2_DEL_ALLOC) +#define DQUOT_DEL_REWRITE max(V1_DEL_REWRITE, V2_DEL_REWRITE) /* * Data for one user/group kept in memory -- cgit v1.2.3 From 1f54587bea84a35125c95e19b98c2f464c50871b Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 23 Jun 2005 22:01:04 -0700 Subject: [PATCH] quota: ext3: Improve quota credit estimates Use improved credits estimates for quota operations. Also reserve a space for a quota operation in a transaction only if filesystem was mounted with some quota options. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/acl.c | 5 +++-- fs/ext3/inode.c | 7 ++++--- fs/ext3/namei.c | 25 +++++++++++++------------ fs/ext3/super.c | 37 +++++++++++++++++++++++++++---------- fs/ext3/xattr.c | 2 +- include/linux/ext3_fs.h | 1 + include/linux/ext3_jbd.h | 19 +++++++++++-------- 7 files changed, 60 insertions(+), 36 deletions(-) (limited to 'include/linux') diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c index 133f5aa581bb..3ac38266fc9e 100644 --- a/fs/ext3/acl.c +++ b/fs/ext3/acl.c @@ -393,7 +393,8 @@ ext3_acl_chmod(struct inode *inode) int retries = 0; retry: - handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS); + handle = ext3_journal_start(inode, + EXT3_DATA_TRANS_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) { error = PTR_ERR(handle); ext3_std_error(inode->i_sb, error); @@ -503,7 +504,7 @@ ext3_xattr_set_acl(struct inode *inode, int type, const void *value, acl = NULL; retry: - handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS); + handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); error = ext3_set_acl(handle, inode, type, acl); diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 0d5fa73b18dc..0b2db4f618cb 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -128,7 +128,7 @@ static unsigned long blocks_for_truncate(struct inode *inode) if (needed > EXT3_MAX_TRANS_DATA) needed = EXT3_MAX_TRANS_DATA; - return EXT3_DATA_TRANS_BLOCKS + needed; + return EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + needed; } /* @@ -2763,7 +2763,8 @@ int ext3_setattr(struct dentry *dentry, struct iattr *attr) /* (user+group)*(old+new) structure, inode write (sb, * inode block, ? - but truncate inode update has it) */ - handle = ext3_journal_start(inode, 4*EXT3_QUOTA_INIT_BLOCKS+3); + handle = ext3_journal_start(inode, 2*(EXT3_QUOTA_INIT_BLOCKS(inode->i_sb)+ + EXT3_QUOTA_DEL_BLOCKS(inode->i_sb))+3); if (IS_ERR(handle)) { error = PTR_ERR(handle); goto err_out; @@ -2861,7 +2862,7 @@ static int ext3_writepage_trans_blocks(struct inode *inode) #ifdef CONFIG_QUOTA /* We know that structure was already allocated during DQUOT_INIT so * we will be updating only the data blocks + inodes */ - ret += 2*EXT3_QUOTA_TRANS_BLOCKS; + ret += 2*EXT3_QUOTA_TRANS_BLOCKS(inode->i_sb); #endif return ret; diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 60e44e6dd7a6..50378d8ff84b 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1645,9 +1645,9 @@ static int ext3_create (struct inode * dir, struct dentry * dentry, int mode, int err, retries = 0; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS + + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + - 2*EXT3_QUOTA_INIT_BLOCKS); + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -1679,9 +1679,9 @@ static int ext3_mknod (struct inode * dir, struct dentry *dentry, return -EINVAL; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS + + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + - 2*EXT3_QUOTA_INIT_BLOCKS); + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -1715,9 +1715,9 @@ static int ext3_mkdir(struct inode * dir, struct dentry * dentry, int mode) return -EMLINK; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS + + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + - 2*EXT3_QUOTA_INIT_BLOCKS); + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2006,7 +2006,7 @@ static int ext3_rmdir (struct inode * dir, struct dentry *dentry) /* Initialize quotas before so that eventual writes go in * separate transaction */ DQUOT_INIT(dentry->d_inode); - handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS); + handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2065,7 +2065,7 @@ static int ext3_unlink(struct inode * dir, struct dentry *dentry) /* Initialize quotas before so that eventual writes go * in separate transaction */ DQUOT_INIT(dentry->d_inode); - handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS); + handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2120,9 +2120,9 @@ static int ext3_symlink (struct inode * dir, return -ENAMETOOLONG; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS + + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 + - 2*EXT3_QUOTA_INIT_BLOCKS); + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2174,7 +2174,7 @@ static int ext3_link (struct dentry * old_dentry, return -EMLINK; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS + + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2216,7 +2216,8 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, * in separate transaction */ if (new_dentry->d_inode) DQUOT_INIT(new_dentry->d_inode); - handle = ext3_journal_start(old_dir, 2 * EXT3_DATA_TRANS_BLOCKS + + handle = ext3_journal_start(old_dir, 2 * + EXT3_DATA_TRANS_BLOCKS(old_dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2); if (IS_ERR(handle)) return PTR_ERR(handle); diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 9630fbfdc24a..b4b3e8a39131 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -589,7 +589,7 @@ enum { Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, - Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, + Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize, }; @@ -634,10 +634,10 @@ static match_table_t tokens = { {Opt_grpjquota, "grpjquota=%s"}, {Opt_jqfmt_vfsold, "jqfmt=vfsold"}, {Opt_jqfmt_vfsv0, "jqfmt=vfsv0"}, - {Opt_ignore, "grpquota"}, - {Opt_ignore, "noquota"}, - {Opt_ignore, "quota"}, - {Opt_ignore, "usrquota"}, + {Opt_quota, "grpquota"}, + {Opt_noquota, "noquota"}, + {Opt_quota, "quota"}, + {Opt_quota, "usrquota"}, {Opt_barrier, "barrier=%u"}, {Opt_err, NULL}, {Opt_resize, "resize"}, @@ -876,6 +876,7 @@ set_qf_name: sbi->s_qf_names[qtype] = NULL; return 0; } + set_opt(sbi->s_mount_opt, QUOTA); break; case Opt_offusrjquota: qtype = USRQUOTA; @@ -898,6 +899,17 @@ clear_qf_name: case Opt_jqfmt_vfsv0: sbi->s_jquota_fmt = QFMT_VFS_V0; break; + case Opt_quota: + set_opt(sbi->s_mount_opt, QUOTA); + break; + case Opt_noquota: + if (sb_any_quota_enabled(sb)) { + printk(KERN_ERR "EXT3-fs: Cannot change quota " + "options when quota turned on.\n"); + return 0; + } + clear_opt(sbi->s_mount_opt, QUOTA); + break; #else case Opt_usrjquota: case Opt_grpjquota: @@ -909,6 +921,9 @@ clear_qf_name: "EXT3-fs: journalled quota options not " "supported.\n"); break; + case Opt_quota: + case Opt_noquota: + break; #endif case Opt_abort: set_opt(sbi->s_mount_opt, ABORT); @@ -2238,7 +2253,7 @@ static int ext3_dquot_initialize(struct inode *inode, int type) int ret, err; /* We may create quota structure so we need to reserve enough blocks */ - handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS); + handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_initialize(inode, type); @@ -2254,7 +2269,7 @@ static int ext3_dquot_drop(struct inode *inode) int ret, err; /* We may delete quota structure so we need to reserve enough blocks */ - handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS); + handle = ext3_journal_start(inode, 2*EXT3_QUOTA_DEL_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_drop(inode); @@ -2272,7 +2287,7 @@ static int ext3_write_dquot(struct dquot *dquot) inode = dquot_to_inode(dquot); handle = ext3_journal_start(inode, - EXT3_QUOTA_TRANS_BLOCKS); + EXT3_QUOTA_TRANS_BLOCKS(dquot->dq_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_commit(dquot); @@ -2288,7 +2303,7 @@ static int ext3_acquire_dquot(struct dquot *dquot) handle_t *handle; handle = ext3_journal_start(dquot_to_inode(dquot), - EXT3_QUOTA_INIT_BLOCKS); + EXT3_QUOTA_INIT_BLOCKS(dquot->dq_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_acquire(dquot); @@ -2304,7 +2319,7 @@ static int ext3_release_dquot(struct dquot *dquot) handle_t *handle; handle = ext3_journal_start(dquot_to_inode(dquot), - EXT3_QUOTA_INIT_BLOCKS); + EXT3_QUOTA_DEL_BLOCKS(dquot->dq_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_release(dquot); @@ -2361,6 +2376,8 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, int err; struct nameidata nd; + if (!test_opt(sb, QUOTA)) + return -EINVAL; /* Not journalling quota? */ if (!EXT3_SB(sb)->s_qf_names[USRQUOTA] && !EXT3_SB(sb)->s_qf_names[GRPQUOTA]) diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c index 4cbc6d0212d3..3f9dfa643b19 100644 --- a/fs/ext3/xattr.c +++ b/fs/ext3/xattr.c @@ -1044,7 +1044,7 @@ ext3_xattr_set(struct inode *inode, int name_index, const char *name, int error, retries = 0; retry: - handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS); + handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) { error = PTR_ERR(handle); } else { diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h index 74ad31781e3e..4b6e1ab216a5 100644 --- a/include/linux/ext3_fs.h +++ b/include/linux/ext3_fs.h @@ -358,6 +358,7 @@ struct ext3_inode { #define EXT3_MOUNT_RESERVATION 0x10000 /* Preallocation */ #define EXT3_MOUNT_BARRIER 0x20000 /* Use block barriers */ #define EXT3_MOUNT_NOBH 0x40000 /* No bufferheads */ +#define EXT3_MOUNT_QUOTA 0x80000 /* Some quota option set */ /* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H diff --git a/include/linux/ext3_jbd.h b/include/linux/ext3_jbd.h index e8292af9033b..c8307c02dd07 100644 --- a/include/linux/ext3_jbd.h +++ b/include/linux/ext3_jbd.h @@ -42,15 +42,15 @@ * superblock only gets updated once, of course, so don't bother * counting that again for the quota updates. */ -#define EXT3_DATA_TRANS_BLOCKS (EXT3_SINGLEDATA_TRANS_BLOCKS + \ +#define EXT3_DATA_TRANS_BLOCKS(sb) (EXT3_SINGLEDATA_TRANS_BLOCKS + \ EXT3_XATTR_TRANS_BLOCKS - 2 + \ - 2*EXT3_QUOTA_TRANS_BLOCKS) + 2*EXT3_QUOTA_TRANS_BLOCKS(sb)) /* Delete operations potentially hit one directory's namespace plus an * entire inode, plus arbitrary amounts of bitmap/indirection data. Be * generous. We can grow the delete transaction later if necessary. */ -#define EXT3_DELETE_TRANS_BLOCKS (2 * EXT3_DATA_TRANS_BLOCKS + 64) +#define EXT3_DELETE_TRANS_BLOCKS(sb) (2 * EXT3_DATA_TRANS_BLOCKS(sb) + 64) /* Define an arbitrary limit for the amount of data we will anticipate * writing to any given transaction. For unbounded transactions such as @@ -74,14 +74,17 @@ #ifdef CONFIG_QUOTA /* Amount of blocks needed for quota update - we know that the structure was * allocated so we need to update only inode+data */ -#define EXT3_QUOTA_TRANS_BLOCKS 2 +#define EXT3_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) /* Amount of blocks needed for quota insert/delete - we do some block writes * but inode, sb and group updates are done only once */ -#define EXT3_QUOTA_INIT_BLOCKS (DQUOT_MAX_WRITES*\ - (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3) +#define EXT3_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ + (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_INIT_REWRITE) : 0) +#define EXT3_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ + (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_DEL_REWRITE) : 0) #else -#define EXT3_QUOTA_TRANS_BLOCKS 0 -#define EXT3_QUOTA_INIT_BLOCKS 0 +#define EXT3_QUOTA_TRANS_BLOCKS(sb) 0 +#define EXT3_QUOTA_INIT_BLOCKS(sb) 0 +#define EXT3_QUOTA_DEL_BLOCKS(sb) 0 #endif int -- cgit v1.2.3 From 556a2a45bce1740f035befaa7201e4ad836c7257 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 23 Jun 2005 22:01:06 -0700 Subject: [PATCH] quota: reiserfs: improve quota credit estimates Use improved credits estimates for quota operations. Also reserve space for a quota operation in a transaction only if filesystem was mounted with some quota option. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/file.c | 4 ++-- fs/reiserfs/inode.c | 11 ++++++----- fs/reiserfs/namei.c | 25 ++++++++++++++----------- fs/reiserfs/super.c | 35 +++++++++++++++++++++++------------ include/linux/reiserfs_fs.h | 15 +++++++++++---- include/linux/reiserfs_fs_sb.h | 2 ++ 6 files changed, 58 insertions(+), 34 deletions(-) (limited to 'include/linux') diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 2230afff1870..12e91209544e 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -201,7 +201,7 @@ static int reiserfs_allocate_blocks_for_region( /* If we came here, it means we absolutely need to open a transaction, since we need to allocate some blocks */ reiserfs_write_lock(inode->i_sb); // Journaling stuff and we need that. - res = journal_begin(th, inode->i_sb, JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS); // Wish I know if this number enough + res = journal_begin(th, inode->i_sb, JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS(inode->i_sb)); // Wish I know if this number enough if (res) goto error_exit; reiserfs_update_inode_transaction(inode) ; @@ -576,7 +576,7 @@ error_exit: int err; // update any changes we made to blk count reiserfs_update_sd(th, inode); - err = journal_end(th, inode->i_sb, JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS); + err = journal_end(th, inode->i_sb, JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS(inode->i_sb)); if (err) res = err; } diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 073425e6e0a9..0d5817f81972 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -28,7 +28,7 @@ static int reiserfs_prepare_write(struct file *f, struct page *page, void reiserfs_delete_inode (struct inode * inode) { /* We need blocks for transaction + (user+group) quota update (possibly delete) */ - int jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 * REISERFS_QUOTA_INIT_BLOCKS; + int jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 * REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb); struct reiserfs_transaction_handle th ; reiserfs_write_lock(inode->i_sb); @@ -591,7 +591,7 @@ int reiserfs_get_block (struct inode * inode, sector_t block, XXX in practically impossible worst case direct2indirect() can incur (much) more than 3 balancings. quota update for user, group */ - int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS; + int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS(inode->i_sb); int version; int dangle = 1; loff_t new_offset = (((loff_t)block) << inode->i_sb->s_blocksize_bits) + 1 ; @@ -2796,14 +2796,15 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr) { if (!error) { struct reiserfs_transaction_handle th; + int jbegin_count = 2*(REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb)+REISERFS_QUOTA_DEL_BLOCKS(inode->i_sb))+2; /* (user+group)*(old+new) structure - we count quota info and , inode write (sb, inode) */ - error = journal_begin(&th, inode->i_sb, 4*REISERFS_QUOTA_INIT_BLOCKS+2); + error = journal_begin(&th, inode->i_sb, jbegin_count); if (error) goto out; error = DQUOT_TRANSFER(inode, attr) ? -EDQUOT : 0; if (error) { - journal_end(&th, inode->i_sb, 4*REISERFS_QUOTA_INIT_BLOCKS+2); + journal_end(&th, inode->i_sb, jbegin_count); goto out; } /* Update corresponding info in inode so that everything is in @@ -2813,7 +2814,7 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr) { if (attr->ia_valid & ATTR_GID) inode->i_gid = attr->ia_gid; mark_inode_dirty(inode); - error = journal_end(&th, inode->i_sb, 4*REISERFS_QUOTA_INIT_BLOCKS+2); + error = journal_end(&th, inode->i_sb, jbegin_count); } } if (!error) diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index 7d4dc5f5aa8b..4a333255f27a 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -586,7 +586,7 @@ static int reiserfs_create (struct inode * dir, struct dentry *dentry, int mode, int retval; struct inode * inode; /* We need blocks for transaction + (user+group)*(quotas for new inode + update of quota for directory owner) */ - int jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS); + int jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 * (REISERFS_QUOTA_INIT_BLOCKS(dir->i_sb)+REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb)); struct reiserfs_transaction_handle th ; int locked; @@ -653,7 +653,7 @@ static int reiserfs_mknod (struct inode * dir, struct dentry *dentry, int mode, struct inode * inode; struct reiserfs_transaction_handle th ; /* We need blocks for transaction + (user+group)*(quotas for new inode + update of quota for directory owner) */ - int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS); + int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS(dir->i_sb)+REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb)); int locked; if (!new_valid_dev(rdev)) @@ -727,7 +727,7 @@ static int reiserfs_mkdir (struct inode * dir, struct dentry *dentry, int mode) struct inode * inode; struct reiserfs_transaction_handle th ; /* We need blocks for transaction + (user+group)*(quotas for new inode + update of quota for directory owner) */ - int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS); + int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS(dir->i_sb)+REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb)); int locked; #ifdef DISPLACE_NEW_PACKING_LOCALITIES @@ -829,8 +829,10 @@ static int reiserfs_rmdir (struct inode * dir, struct dentry *dentry) /* we will be doing 2 balancings and update 2 stat data, we change quotas - * of the owner of the directory and of the owner of the parent directory */ - jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS); + * of the owner of the directory and of the owner of the parent directory. + * The quota structure is possibly deleted only on last iput => outside + * of this transaction */ + jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 + 4 * REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb); reiserfs_write_lock(dir->i_sb); retval = journal_begin(&th, dir->i_sb, jbegin_count) ; @@ -913,9 +915,10 @@ static int reiserfs_unlink (struct inode * dir, struct dentry *dentry) inode = dentry->d_inode; /* in this transaction we can be doing at max two balancings and update - two stat datas, we change quotas of the owner of the directory and of - the owner of the parent directory */ - jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS); + * two stat datas, we change quotas of the owner of the directory and of + * the owner of the parent directory. The quota structure is possibly + * deleted only on iput => outside of this transaction */ + jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 + 4 * REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb); reiserfs_write_lock(dir->i_sb); retval = journal_begin(&th, dir->i_sb, jbegin_count) ; @@ -1000,7 +1003,7 @@ static int reiserfs_symlink (struct inode * parent_dir, struct reiserfs_transaction_handle th ; int mode = S_IFLNK | S_IRWXUGO; /* We need blocks for transaction + (user+group)*(quotas for new inode + update of quota for directory owner) */ - int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS); + int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS(parent_dir->i_sb)+REISERFS_QUOTA_TRANS_BLOCKS(parent_dir->i_sb)); if (!(inode = new_inode(parent_dir->i_sb))) { return -ENOMEM ; @@ -1076,7 +1079,7 @@ static int reiserfs_link (struct dentry * old_dentry, struct inode * dir, struct struct inode *inode = old_dentry->d_inode; struct reiserfs_transaction_handle th ; /* We need blocks for transaction + update of quotas for the owners of the directory */ - int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * REISERFS_QUOTA_TRANS_BLOCKS; + int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb); reiserfs_write_lock(dir->i_sb); if (inode->i_nlink >= REISERFS_LINK_MAX) { @@ -1196,7 +1199,7 @@ static int reiserfs_rename (struct inode * old_dir, struct dentry *old_dentry, pointed initially and (5) maybe block containing ".." of renamed directory quota updates: two parent directories */ - jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 5 + 4 * REISERFS_QUOTA_TRANS_BLOCKS; + jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 5 + 4 * REISERFS_QUOTA_TRANS_BLOCKS(old_dir->i_sb); old_inode = old_dentry->d_inode; new_dentry_inode = new_dentry->d_inode; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 031577fb41a1..660aefca1fd2 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -866,8 +866,9 @@ static int reiserfs_parse_options (struct super_block * s, char * options, /* st {"jdev", .arg_required = 'j', .values = NULL}, {"nolargeio", .arg_required = 'w', .values = NULL}, {"commit", .arg_required = 'c', .values = NULL}, - {"usrquota",}, - {"grpquota",}, + {"usrquota", .setmask = 1<s_qf_names[qtype], arg); + *mount_options |= 1<s_qf_names[qtype]) { @@ -995,7 +997,13 @@ static int reiserfs_parse_options (struct super_block * s, char * options, /* st reiserfs_warning(s, "reiserfs_parse_options: journalled quota format not specified."); return 0; } + /* This checking is not precise wrt the quota type but for our purposes it is sufficient */ + if (!(*mount_options & (1<i_sb); - ret = journal_begin(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS); + ret = journal_begin(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb)); if (ret) goto out; ret = dquot_initialize(inode, type); - err = journal_end(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS); + err = journal_end(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb)); if (!ret && err) ret = err; out: @@ -1864,11 +1873,11 @@ static int reiserfs_dquot_drop(struct inode *inode) /* We may delete quota structure so we need to reserve enough blocks */ reiserfs_write_lock(inode->i_sb); - ret = journal_begin(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS); + ret = journal_begin(&th, inode->i_sb, 2*REISERFS_QUOTA_DEL_BLOCKS(inode->i_sb)); if (ret) goto out; ret = dquot_drop(inode); - err = journal_end(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS); + err = journal_end(&th, inode->i_sb, 2*REISERFS_QUOTA_DEL_BLOCKS(inode->i_sb)); if (!ret && err) ret = err; out: @@ -1882,11 +1891,11 @@ static int reiserfs_write_dquot(struct dquot *dquot) int ret, err; reiserfs_write_lock(dquot->dq_sb); - ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_TRANS_BLOCKS); + ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_TRANS_BLOCKS(dquot->dq_sb)); if (ret) goto out; ret = dquot_commit(dquot); - err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_TRANS_BLOCKS); + err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_TRANS_BLOCKS(dquot->dq_sb)); if (!ret && err) ret = err; out: @@ -1900,11 +1909,11 @@ static int reiserfs_acquire_dquot(struct dquot *dquot) int ret, err; reiserfs_write_lock(dquot->dq_sb); - ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS); + ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS(dquot->dq_sb)); if (ret) goto out; ret = dquot_acquire(dquot); - err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS); + err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS(dquot->dq_sb)); if (!ret && err) ret = err; out: @@ -1918,11 +1927,11 @@ static int reiserfs_release_dquot(struct dquot *dquot) int ret, err; reiserfs_write_lock(dquot->dq_sb); - ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS); + ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_DEL_BLOCKS(dquot->dq_sb)); if (ret) goto out; ret = dquot_release(dquot); - err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS); + err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_DEL_BLOCKS(dquot->dq_sb)); if (!ret && err) ret = err; out: @@ -1978,6 +1987,8 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, ch int err; struct nameidata nd; + if (!(REISERFS_SB(sb)->s_mount_opt & (1<s_mount_opt & (1<s_mount_opt & (1<s_mount_opt & (1< Date: Thu, 23 Jun 2005 22:01:42 -0700 Subject: [PATCH] ipmi: add power cycle capability This patch to adds "power cycle" functionality to the IPMI power off module ipmi_poweroff. It also contains changes to support procfs control of the feature. The power cycle action is considered an optional chassis control in the IPMI specification. However, it is definitely useful when the hardware supports it. A power cycle is usually required in order to reset a firmware in a bad state. This action is critical to allow remote management of servers. The implementation adds power cycle as optional to the ipmi_poweroff module. It can be modified dynamically through the proc entry mentioned above. During a power down and enabled, the power cycle command is sent to the BMC firmware. If it fails either due to non-support or some error, it will retry to send the command as power off. Signed-off-by: Christopher A. Poblete Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/IPMI.txt | 20 +++++++ drivers/char/ipmi/ipmi_msghandler.c | 29 +++++++++- drivers/char/ipmi/ipmi_poweroff.c | 112 +++++++++++++++++++++++++++++++++--- include/linux/ipmi.h | 5 ++ 4 files changed, 155 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/Documentation/IPMI.txt b/Documentation/IPMI.txt index 2f99fe6299ab..84d3d4d10c17 100644 --- a/Documentation/IPMI.txt +++ b/Documentation/IPMI.txt @@ -594,3 +594,23 @@ an event generator, the event receiver from the local management controller will be queried and the events sent to the SEL on that device. Otherwise, the events go nowhere since there is nowhere to send them. + + +Poweroff +-------- + +If the poweroff capability is selected, the IPMI driver will install +a shutdown function into the standard poweroff function pointer. This +is in the ipmi_poweroff module. When the system requests a powerdown, +it will send the proper IPMI commands to do this. This is supported on +several platforms. + +There is a module parameter named "poweroff_control" that may either be zero +(do a power down) or 2 (do a power cycle, power the system off, then power +it on in a few seconds). Setting ipmi_poweroff.poweroff_control=x will do +the same thing on the kernel command line. The parameter is also available +via the proc filesystem in /proc/ipmi/poweroff_control. Note that if the +system does not support power cycling, it will always to the power off. + +Note that if you have ACPI enabled, the system will prefer using ACPI to +power off. diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index ed75e96d0035..1813d0d198f1 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -54,7 +54,9 @@ static int ipmi_init_msghandler(void); static int initialized = 0; -static struct proc_dir_entry *proc_ipmi_root = NULL; +#ifdef CONFIG_PROC_FS +struct proc_dir_entry *proc_ipmi_root = NULL; +#endif /* CONFIG_PROC_FS */ #define MAX_EVENTS_IN_QUEUE 25 @@ -124,11 +126,13 @@ struct ipmi_channel unsigned char protocol; }; +#ifdef CONFIG_PROC_FS struct ipmi_proc_entry { char *name; struct ipmi_proc_entry *next; }; +#endif #define IPMI_IPMB_NUM_SEQ 64 #define IPMI_MAX_CHANNELS 8 @@ -156,10 +160,13 @@ struct ipmi_smi struct ipmi_smi_handlers *handlers; void *send_info; +#ifdef CONFIG_PROC_FS /* A list of proc entries for this interface. This does not need a lock, only one thread creates it and only one thread destroys it. */ + spinlock_t proc_entry_lock; struct ipmi_proc_entry *proc_entries; +#endif /* A table of sequence numbers for this interface. We use the sequence numbers for IPMB messages that go out of the @@ -1470,8 +1477,9 @@ int ipmi_smi_add_proc_entry(ipmi_smi_t smi, char *name, read_proc_t *read_proc, write_proc_t *write_proc, void *data, struct module *owner) { - struct proc_dir_entry *file; int rv = 0; +#ifdef CONFIG_PROC_FS + struct proc_dir_entry *file; struct ipmi_proc_entry *entry; /* Create a list element. */ @@ -1497,10 +1505,13 @@ int ipmi_smi_add_proc_entry(ipmi_smi_t smi, char *name, file->write_proc = write_proc; file->owner = owner; + spin_lock(&smi->proc_entry_lock); /* Stick it on the list. */ entry->next = smi->proc_entries; smi->proc_entries = entry; + spin_unlock(&smi->proc_entry_lock); } +#endif /* CONFIG_PROC_FS */ return rv; } @@ -1509,6 +1520,7 @@ static int add_proc_entries(ipmi_smi_t smi, int num) { int rv = 0; +#ifdef CONFIG_PROC_FS sprintf(smi->proc_dir_name, "%d", num); smi->proc_dir = proc_mkdir(smi->proc_dir_name, proc_ipmi_root); if (!smi->proc_dir) @@ -1531,14 +1543,17 @@ static int add_proc_entries(ipmi_smi_t smi, int num) rv = ipmi_smi_add_proc_entry(smi, "version", version_file_read_proc, NULL, smi, THIS_MODULE); +#endif /* CONFIG_PROC_FS */ return rv; } static void remove_proc_entries(ipmi_smi_t smi) { +#ifdef CONFIG_PROC_FS struct ipmi_proc_entry *entry; + spin_lock(&smi->proc_entry_lock); while (smi->proc_entries) { entry = smi->proc_entries; smi->proc_entries = entry->next; @@ -1547,7 +1562,9 @@ static void remove_proc_entries(ipmi_smi_t smi) kfree(entry->name); kfree(entry); } + spin_unlock(&smi->proc_entry_lock); remove_proc_entry(smi->proc_dir_name, proc_ipmi_root); +#endif /* CONFIG_PROC_FS */ } static int @@ -1694,6 +1711,9 @@ int ipmi_register_smi(struct ipmi_smi_handlers *handlers, new_intf->seq_table[j].seqid = 0; } new_intf->curr_seq = 0; +#ifdef CONFIG_PROC_FS + spin_lock_init(&(new_intf->proc_entry_lock)); +#endif spin_lock_init(&(new_intf->waiting_msgs_lock)); INIT_LIST_HEAD(&(new_intf->waiting_msgs)); spin_lock_init(&(new_intf->events_lock)); @@ -3085,6 +3105,7 @@ static int ipmi_init_msghandler(void) ipmi_interfaces[i] = NULL; } +#ifdef CONFIG_PROC_FS proc_ipmi_root = proc_mkdir("ipmi", NULL); if (!proc_ipmi_root) { printk(KERN_ERR PFX "Unable to create IPMI proc dir"); @@ -3092,6 +3113,7 @@ static int ipmi_init_msghandler(void) } proc_ipmi_root->owner = THIS_MODULE; +#endif /* CONFIG_PROC_FS */ init_timer(&ipmi_timer); ipmi_timer.data = 0; @@ -3129,7 +3151,9 @@ static __exit void cleanup_ipmi(void) atomic_inc(&stop_operation); del_timer_sync(&ipmi_timer); +#ifdef CONFIG_PROC_FS remove_proc_entry(proc_ipmi_root->name, &proc_root); +#endif /* CONFIG_PROC_FS */ initialized = 0; @@ -3170,4 +3194,5 @@ EXPORT_SYMBOL(ipmi_get_my_address); EXPORT_SYMBOL(ipmi_set_my_LUN); EXPORT_SYMBOL(ipmi_get_my_LUN); EXPORT_SYMBOL(ipmi_smi_add_proc_entry); +EXPORT_SYMBOL(proc_ipmi_root); EXPORT_SYMBOL(ipmi_user_set_run_to_completion); diff --git a/drivers/char/ipmi/ipmi_poweroff.c b/drivers/char/ipmi/ipmi_poweroff.c index cb5cdc6f14bf..61329b55c4a9 100644 --- a/drivers/char/ipmi/ipmi_poweroff.c +++ b/drivers/char/ipmi/ipmi_poweroff.c @@ -34,6 +34,8 @@ #include #include #include +#include +#include #include #include #include @@ -44,6 +46,18 @@ /* Where to we insert our poweroff function? */ extern void (*pm_power_off)(void); +/* Definitions for controlling power off (if the system supports it). It + * conveniently matches the IPMI chassis control values. */ +#define IPMI_CHASSIS_POWER_DOWN 0 /* power down, the default. */ +#define IPMI_CHASSIS_POWER_CYCLE 0x02 /* power cycle */ + +/* the IPMI data command */ +static int poweroff_control = IPMI_CHASSIS_POWER_DOWN; + +/* parameter definition to allow user to flag power cycle */ +module_param(poweroff_control, int, IPMI_CHASSIS_POWER_DOWN); +MODULE_PARM_DESC(poweroff_control, " Set to 2 to enable power cycle instead of power down. Power cycle is contingent on hardware support, otherwise it defaults back to power down."); + /* Stuff from the get device id command. */ static unsigned int mfg_id; static unsigned int prod_id; @@ -349,26 +363,38 @@ static void ipmi_poweroff_chassis (ipmi_user_t user) smi_addr.channel = IPMI_BMC_CHANNEL; smi_addr.lun = 0; - printk(KERN_INFO PFX "Powering down via IPMI chassis control command\n"); + powercyclefailed: + printk(KERN_INFO PFX "Powering %s via IPMI chassis control command\n", + ((poweroff_control != IPMI_CHASSIS_POWER_CYCLE) ? "down" : "cycle")); /* * Power down */ send_msg.netfn = IPMI_NETFN_CHASSIS_REQUEST; send_msg.cmd = IPMI_CHASSIS_CONTROL_CMD; - data[0] = 0; /* Power down */ + data[0] = poweroff_control; send_msg.data = data; send_msg.data_len = sizeof(data); rv = ipmi_request_in_rc_mode(user, (struct ipmi_addr *) &smi_addr, &send_msg); if (rv) { - printk(KERN_ERR PFX "Unable to send chassis powerdown message," - " IPMI error 0x%x\n", rv); - goto out; + switch (poweroff_control) { + case IPMI_CHASSIS_POWER_CYCLE: + /* power cycle failed, default to power down */ + printk(KERN_ERR PFX "Unable to send chassis power " \ + "cycle message, IPMI error 0x%x\n", rv); + poweroff_control = IPMI_CHASSIS_POWER_DOWN; + goto powercyclefailed; + + case IPMI_CHASSIS_POWER_DOWN: + default: + printk(KERN_ERR PFX "Unable to send chassis power " \ + "down message, IPMI error 0x%x\n", rv); + break; + } } - out: return; } @@ -430,7 +456,8 @@ static void ipmi_po_new_smi(int if_num) if (ready) return; - rv = ipmi_create_user(if_num, &ipmi_poweroff_handler, NULL, &ipmi_user); + rv = ipmi_create_user(if_num, &ipmi_poweroff_handler, NULL, + &ipmi_user); if (rv) { printk(KERN_ERR PFX "could not create IPMI user, error %d\n", rv); @@ -509,21 +536,84 @@ static struct ipmi_smi_watcher smi_watcher = }; +#ifdef CONFIG_PROC_FS +/* displays properties to proc */ +static int proc_read_chassctrl(char *page, char **start, off_t off, int count, + int *eof, void *data) +{ + return sprintf(page, "%d\t[ 0=powerdown 2=powercycle ]\n", + poweroff_control); +} + +/* process property writes from proc */ +static int proc_write_chassctrl(struct file *file, const char *buffer, + unsigned long count, void *data) +{ + int rv = count; + unsigned int newval = 0; + + sscanf(buffer, "%d", &newval); + switch (newval) { + case IPMI_CHASSIS_POWER_CYCLE: + printk(KERN_INFO PFX "power cycle is now enabled\n"); + poweroff_control = newval; + break; + + case IPMI_CHASSIS_POWER_DOWN: + poweroff_control = IPMI_CHASSIS_POWER_DOWN; + break; + + default: + rv = -EINVAL; + break; + } + + return rv; +} +#endif /* CONFIG_PROC_FS */ + /* * Startup and shutdown functions. */ static int ipmi_poweroff_init (void) { - int rv; + int rv; + struct proc_dir_entry *file; printk ("Copyright (C) 2004 MontaVista Software -" " IPMI Powerdown via sys_reboot version " IPMI_POWEROFF_VERSION ".\n"); + switch (poweroff_control) { + case IPMI_CHASSIS_POWER_CYCLE: + printk(KERN_INFO PFX "Power cycle is enabled.\n"); + break; + + case IPMI_CHASSIS_POWER_DOWN: + default: + poweroff_control = IPMI_CHASSIS_POWER_DOWN; + break; + } + rv = ipmi_smi_watcher_register(&smi_watcher); - if (rv) + if (rv) { printk(KERN_ERR PFX "Unable to register SMI watcher: %d\n", rv); + goto out_err; + } + +#ifdef CONFIG_PROC_FS + file = create_proc_entry("poweroff_control", 0, proc_ipmi_root); + if (!file) { + printk(KERN_ERR PFX "Unable to create proc power control\n"); + } else { + file->nlink = 1; + file->read_proc = proc_read_chassctrl; + file->write_proc = proc_write_chassctrl; + file->owner = THIS_MODULE; + } +#endif + out_err: return rv; } @@ -532,6 +622,10 @@ static __exit void ipmi_poweroff_cleanup(void) { int rv; +#ifdef CONFIG_PROC_FS + remove_proc_entry("poweroff_control", proc_ipmi_root); +#endif + ipmi_smi_watcher_unregister(&smi_watcher); if (ready) { diff --git a/include/linux/ipmi.h b/include/linux/ipmi.h index 2ec265e1045f..596ca6130159 100644 --- a/include/linux/ipmi.h +++ b/include/linux/ipmi.h @@ -209,6 +209,11 @@ struct kernel_ipmi_msg #include #include +#ifdef CONFIG_PROC_FS +#include +extern struct proc_dir_entry *proc_ipmi_root; +#endif /* CONFIG_PROC_FS */ + /* Opaque type for a IPMI message user. One of these is needed to send and receive messages. */ typedef struct ipmi_user *ipmi_user_t; -- cgit v1.2.3 From a6df7da8f7ee99e6fd1995fad852bacb978a6447 Mon Sep 17 00:00:00 2001 From: Kylene Hall Date: Thu, 23 Jun 2005 22:02:04 -0700 Subject: [PATCH] tpm: TPMs on additional LPC bus Add support for TPMs on additional LPC buses. Signed-off-by: Kylene Hall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/tpm/tpm_atmel.c | 1 + include/linux/pci_ids.h | 1 + 2 files changed, 2 insertions(+) (limited to 'include/linux') diff --git a/drivers/char/tpm/tpm_atmel.c b/drivers/char/tpm/tpm_atmel.c index 68974577a6a6..13248400b9c3 100644 --- a/drivers/char/tpm/tpm_atmel.c +++ b/drivers/char/tpm/tpm_atmel.c @@ -205,6 +205,7 @@ static struct pci_device_id tpm_pci_tbl[] __devinitdata = { {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801DB_12)}, {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801EB_0)}, {PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8111_LPC)}, + {PCI_DEVICE(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_CSB6LPC)}, {0,} }; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 63e89e47b8e9..bf608808a60c 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1568,6 +1568,7 @@ #define PCI_DEVICE_ID_SERVERWORKS_OSB4USB 0x0220 #define PCI_DEVICE_ID_SERVERWORKS_CSB5USB PCI_DEVICE_ID_SERVERWORKS_OSB4USB #define PCI_DEVICE_ID_SERVERWORKS_CSB6USB 0x0221 +#define PCI_DEVICE_ID_SERVERWORKS_CSB6LPC 0x0227 #define PCI_DEVICE_ID_SERVERWORKS_GCLE 0x0225 #define PCI_DEVICE_ID_SERVERWORKS_GCLE2 0x0227 #define PCI_DEVICE_ID_SERVERWORKS_CSB5ISA 0x0230 -- cgit v1.2.3 From 61fbfa8129c1771061a0e9f47747854293081c5b Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:11 -0700 Subject: [PATCH] I2O: bugfixes and compability enhancements Changes: - Fixed sysfs bug where user and parent links where added to the I2O device itself - Fixed bug when calculating TID for the event handler and cleaned up the workflow of i2o_driver_dispatch() - Fixed oops when no I2O device could be found for an event delivered to Exec-OSM - Fixed initialization of spinlock in Exec-OSM - Fixed memory leak in i2o_cfg_passthru() and i2o_cfg_passthru() - Removed MTRR support - Added PCI ID of Promise SX6000 with firmware >= 1.20.x.x - Turn of caching for ioremapped memory of in_queue - Added initialization sequence for Promise controllers - Moved definition of u8 / u16 / u32 for raidutils before first use Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/i2o/device.c | 10 +++-- drivers/message/i2o/driver.c | 89 +++++++++++++++++++------------------- drivers/message/i2o/exec-osm.c | 9 ++-- drivers/message/i2o/i2o_config.c | 48 +++++++++++++-------- drivers/message/i2o/i2o_scsi.c | 3 +- drivers/message/i2o/pci.c | 93 ++++++++++++++-------------------------- include/linux/i2o-dev.h | 16 +++---- include/linux/i2o.h | 5 --- 8 files changed, 124 insertions(+), 149 deletions(-) (limited to 'include/linux') diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c index eb907e87bc7b..280627ae6cf7 100644 --- a/drivers/message/i2o/device.c +++ b/drivers/message/i2o/device.c @@ -401,25 +401,27 @@ static int i2o_device_class_add(struct class_device *cd) /* create user entries for this device */ tmp = i2o_iop_find_device(i2o_dev->iop, i2o_dev->lct_data.user_tid); - if (tmp) + if (tmp && (tmp != i2o_dev)) sysfs_create_link(&i2o_dev->device.kobj, &tmp->device.kobj, "user"); /* create user entries refering to this device */ list_for_each_entry(tmp, &c->devices, list) - if (tmp->lct_data.user_tid == i2o_dev->lct_data.tid) + if ((tmp->lct_data.user_tid == i2o_dev->lct_data.tid) + && (tmp != i2o_dev)) sysfs_create_link(&tmp->device.kobj, &i2o_dev->device.kobj, "user"); /* create parent entries for this device */ tmp = i2o_iop_find_device(i2o_dev->iop, i2o_dev->lct_data.parent_tid); - if (tmp) + if (tmp && (tmp != i2o_dev)) sysfs_create_link(&i2o_dev->device.kobj, &tmp->device.kobj, "parent"); /* create parent entries refering to this device */ list_for_each_entry(tmp, &c->devices, list) - if (tmp->lct_data.parent_tid == i2o_dev->lct_data.tid) + if ((tmp->lct_data.parent_tid == i2o_dev->lct_data.tid) + && (tmp != i2o_dev)) sysfs_create_link(&tmp->device.kobj, &i2o_dev->device.kobj, "parent"); diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c index 91f4edbb2a27..c71e68f70e7d 100644 --- a/drivers/message/i2o/driver.c +++ b/drivers/message/i2o/driver.c @@ -18,6 +18,8 @@ #include #include +#define OSM_NAME "core" + /* max_drivers - Maximum I2O drivers (OSMs) which could be registered */ unsigned int i2o_max_drivers = I2O_MAX_DRIVERS; module_param_named(max_drivers, i2o_max_drivers, uint, 0); @@ -182,62 +184,59 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m, struct i2o_driver *drv; u32 context = readl(&msg->u.s.icntxt); - if (likely(context < i2o_max_drivers)) { - spin_lock(&i2o_drivers_lock); - drv = i2o_drivers[context]; - spin_unlock(&i2o_drivers_lock); - - if (unlikely(!drv)) { - printk(KERN_WARNING "%s: Spurious reply to unknown " - "driver %d\n", c->name, context); - return -EIO; - } + if (unlikely(context >= i2o_max_drivers)) { + printk(KERN_WARNING "%s: Spurious reply to unknown driver " + "%d\n", c->name, readl(&msg->u.s.icntxt)); + return -EIO; + } - if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_UTIL_EVT_REGISTER) { - struct i2o_device *dev, *tmp; - struct i2o_event *evt; - u16 size; - u16 tid; + spin_lock(&i2o_drivers_lock); + drv = i2o_drivers[context]; + spin_unlock(&i2o_drivers_lock); - tid = readl(&msg->u.head[1]) & 0x1fff; + if (unlikely(!drv)) { + osm_warn("Spurious reply to unknown driver %d\n", context); + return -EIO; + } - pr_debug("%s: event received from device %d\n", c->name, - tid); + if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_UTIL_EVT_REGISTER) { + struct i2o_device *dev, *tmp; + struct i2o_event *evt; + u16 size; + u16 tid = readl(&msg->u.head[1]) & 0xfff; - /* cut of header from message size (in 32-bit words) */ - size = (readl(&msg->u.head[0]) >> 16) - 5; + osm_debug("event received from device %d\n", tid); - evt = kmalloc(size * 4 + sizeof(*evt), GFP_ATOMIC); - if (!evt) - return -ENOMEM; - memset(evt, 0, size * 4 + sizeof(*evt)); + /* cut of header from message size (in 32-bit words) */ + size = (readl(&msg->u.head[0]) >> 16) - 5; - evt->size = size; - memcpy_fromio(&evt->tcntxt, &msg->u.s.tcntxt, - (size + 2) * 4); + evt = kmalloc(size * 4 + sizeof(*evt), GFP_ATOMIC | __GFP_ZERO); + if (!evt) + return -ENOMEM; - list_for_each_entry_safe(dev, tmp, &c->devices, list) - if (dev->lct_data.tid == tid) { - evt->i2o_dev = dev; - break; - } + evt->size = size; + evt->tcntxt = readl(&msg->u.s.tcntxt); + evt->event_indicator = readl(&msg->body[0]); + memcpy_fromio(&evt->tcntxt, &msg->u.s.tcntxt, size * 4); - INIT_WORK(&evt->work, (void (*)(void *))drv->event, - evt); - queue_work(drv->event_queue, &evt->work); - return 1; + list_for_each_entry_safe(dev, tmp, &c->devices, list) + if (dev->lct_data.tid == tid) { + evt->i2o_dev = dev; + break; } - if (likely(drv->reply)) - return drv->reply(c, m, msg); - else - pr_debug("%s: Reply to driver %s, but no reply function" - " defined!\n", c->name, drv->name); + INIT_WORK(&evt->work, (void (*)(void *))drv->event, evt); + queue_work(drv->event_queue, &evt->work); + return 1; + } + + if (unlikely(!drv->reply)) { + pr_debug("%s: Reply to driver %s, but no reply function" + " defined!\n", c->name, drv->name); return -EIO; - } else - printk(KERN_WARNING "%s: Spurious reply to unknown driver " - "%d\n", c->name, readl(&msg->u.s.icntxt)); - return -EIO; + } + + return drv->reply(c, m, msg); } /** diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index 79c1cbfb8f44..1e28e886f1ca 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -204,12 +204,10 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, struct i2o_message __iomem *msg) { struct i2o_exec_wait *wait, *tmp; - static spinlock_t lock; + static spinlock_t lock = SPIN_LOCK_UNLOCKED; int rc = 1; u32 context; - spin_lock_init(&lock); - context = readl(&msg->u.s.tcntxt); /* @@ -381,8 +379,9 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m, */ static void i2o_exec_event(struct i2o_event *evt) { - osm_info("Event received from device: %d\n", - evt->i2o_dev->lct_data.tid); + if(likely(evt->i2o_dev)) + osm_info("Event received from device: %d\n", + evt->i2o_dev->lct_data.tid); kfree(evt); }; diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c index 1fb5cdf67f8f..46d373287a30 100644 --- a/drivers/message/i2o/i2o_config.c +++ b/drivers/message/i2o/i2o_config.c @@ -555,6 +555,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar u32 sg_offset = 0; u32 sg_count = 0; u32 i = 0; + u32 sg_index = 0; i2o_status_block *sb; struct i2o_message *msg; u32 m; @@ -634,8 +635,8 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar if (sg_count > SG_TABLESIZE) { printk(KERN_DEBUG "%s:IOCTL SG List too large (%u)\n", c->name, sg_count); - kfree(reply); - return -EINVAL; + rcode = -EINVAL; + goto cleanup; } for (i = 0; i < sg_count; i++) { @@ -651,7 +652,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar goto cleanup; } sg_size = sg[i].flag_count & 0xffffff; - p = &(sg_list[i]); + p = &(sg_list[sg_index++]); /* Allocate memory for the transfer */ if (i2o_dma_alloc (&c->pdev->dev, p, sg_size, @@ -660,7 +661,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar "%s: Could not allocate SG buffer - size = %d buffer number %d of %d\n", c->name, sg_size, i, sg_count); rcode = -ENOMEM; - goto cleanup; + goto sg_list_cleanup; } /* Copy in the user's SG buffer if necessary */ if (sg[i]. @@ -673,7 +674,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar "%s: Could not copy SG buf %d FROM user\n", c->name, i); rcode = -EFAULT; - goto cleanup; + goto sg_list_cleanup; } } //TODO 64bit fix @@ -683,10 +684,10 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar rcode = i2o_msg_post_wait(c, m, 60); if (rcode) - goto cleanup; + goto sg_list_cleanup; if (sg_offset) { - u32 msg[128]; + u32 msg[MSG_FRAME_SIZE]; /* Copy back the Scatter Gather buffers back to user space */ u32 j; // TODO 64bit fix @@ -698,14 +699,14 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar // get user msg size in u32s if (get_user(size, &user_msg[0])) { rcode = -EFAULT; - goto cleanup; + goto sg_list_cleanup; } size = size >> 16; size *= 4; /* Copy in the user's I2O command */ if (copy_from_user(msg, user_msg, size)) { rcode = -EFAULT; - goto cleanup; + goto sg_list_cleanup; } sg_count = (size - sg_offset * 4) / sizeof(struct sg_simple_element); @@ -727,7 +728,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar c->name, sg_list[j].virt, sg[j].addr_bus); rcode = -EFAULT; - goto cleanup; + goto sg_list_cleanup; } } } @@ -741,6 +742,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar "%s: Could not copy message context FROM user\n", c->name); rcode = -EFAULT; + goto sg_list_cleanup; } if (copy_to_user(user_reply, reply, reply_size)) { printk(KERN_WARNING @@ -749,6 +751,10 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar } } + sg_list_cleanup: + for (i = 0; i < sg_index; i++) + i2o_dma_free(&c->pdev->dev, &sg_list[i]); + cleanup: kfree(reply); return rcode; @@ -862,8 +868,8 @@ static int i2o_cfg_passthru(unsigned long arg) if (sg_count > SG_TABLESIZE) { printk(KERN_DEBUG "%s:IOCTL SG List too large (%u)\n", c->name, sg_count); - kfree(reply); - return -EINVAL; + rcode = -EINVAL; + goto cleanup; } for (i = 0; i < sg_count; i++) { @@ -875,7 +881,7 @@ static int i2o_cfg_passthru(unsigned long arg) "%s:Bad SG element %d - not simple (%x)\n", c->name, i, sg[i].flag_count); rcode = -EINVAL; - goto cleanup; + goto sg_list_cleanup; } sg_size = sg[i].flag_count & 0xffffff; /* Allocate memory for the transfer */ @@ -885,7 +891,7 @@ static int i2o_cfg_passthru(unsigned long arg) "%s: Could not allocate SG buffer - size = %d buffer number %d of %d\n", c->name, sg_size, i, sg_count); rcode = -ENOMEM; - goto cleanup; + goto sg_list_cleanup; } sg_list[sg_index++] = p; // sglist indexed with input frame, not our internal frame. /* Copy in the user's SG buffer if necessary */ @@ -899,7 +905,7 @@ static int i2o_cfg_passthru(unsigned long arg) "%s: Could not copy SG buf %d FROM user\n", c->name, i); rcode = -EFAULT; - goto cleanup; + goto sg_list_cleanup; } } //TODO 64bit fix @@ -909,7 +915,7 @@ static int i2o_cfg_passthru(unsigned long arg) rcode = i2o_msg_post_wait(c, m, 60); if (rcode) - goto cleanup; + goto sg_list_cleanup; if (sg_offset) { u32 msg[128]; @@ -924,14 +930,14 @@ static int i2o_cfg_passthru(unsigned long arg) // get user msg size in u32s if (get_user(size, &user_msg[0])) { rcode = -EFAULT; - goto cleanup; + goto sg_list_cleanup; } size = size >> 16; size *= 4; /* Copy in the user's I2O command */ if (copy_from_user(msg, user_msg, size)) { rcode = -EFAULT; - goto cleanup; + goto sg_list_cleanup; } sg_count = (size - sg_offset * 4) / sizeof(struct sg_simple_element); @@ -953,7 +959,7 @@ static int i2o_cfg_passthru(unsigned long arg) c->name, sg_list[j], sg[j].addr_bus); rcode = -EFAULT; - goto cleanup; + goto sg_list_cleanup; } } } @@ -975,6 +981,10 @@ static int i2o_cfg_passthru(unsigned long arg) } } + sg_list_cleanup: + for (i = 0; i < sg_index; i++) + kfree(sg_list[i]); + cleanup: kfree(reply); return rcode; diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c index 43f5875e0be5..af40f1c1ec77 100644 --- a/drivers/message/i2o/i2o_scsi.c +++ b/drivers/message/i2o/i2o_scsi.c @@ -103,7 +103,8 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c) list_for_each_entry(i2o_dev, &c->devices, list) if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER_PORT) { - if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) || (type == 1)) /* SCSI bus */ + if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) + && (type == 0x01)) /* SCSI bus */ max_channel++; } diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c index e772752f056d..579a8b7a2120 100644 --- a/drivers/message/i2o/pci.c +++ b/drivers/message/i2o/pci.c @@ -31,10 +31,6 @@ #include #include -#ifdef CONFIG_MTRR -#include -#endif // CONFIG_MTRR - /* Module internal functions from other sources */ extern struct i2o_controller *i2o_iop_alloc(void); extern void i2o_iop_free(struct i2o_controller *); @@ -49,6 +45,8 @@ extern int i2o_driver_dispatch(struct i2o_controller *, u32, static struct pci_device_id __devinitdata i2o_pci_ids[] = { {PCI_DEVICE_CLASS(PCI_CLASS_INTELLIGENT_I2O << 8, 0xffff00)}, {PCI_DEVICE(PCI_VENDOR_ID_DPT, 0xa511)}, + {.vendor = PCI_VENDOR_ID_INTEL,.device = 0x1962, + .subvendor = PCI_VENDOR_ID_PROMISE,.subdevice = PCI_ANY_ID}, {0} }; @@ -97,13 +95,6 @@ static void i2o_pci_free(struct i2o_controller *c) i2o_dma_free(dev, &c->hrt); i2o_dma_free(dev, &c->status); -#ifdef CONFIG_MTRR - if (c->mtrr_reg0 >= 0) - mtrr_del(c->mtrr_reg0, 0, 0); - if (c->mtrr_reg1 >= 0) - mtrr_del(c->mtrr_reg1, 0, 0); -#endif - if (c->raptor && c->in_queue.virt) iounmap(c->in_queue.virt); @@ -178,14 +169,15 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c) c->name, (unsigned long)c->base.phys, (unsigned long)c->base.len); - c->base.virt = ioremap(c->base.phys, c->base.len); + c->base.virt = ioremap_nocache(c->base.phys, c->base.len); if (!c->base.virt) { printk(KERN_ERR "%s: Unable to map controller.\n", c->name); return -ENOMEM; } if (c->raptor) { - c->in_queue.virt = ioremap(c->in_queue.phys, c->in_queue.len); + c->in_queue.virt = + ioremap_nocache(c->in_queue.phys, c->in_queue.len); if (!c->in_queue.virt) { printk(KERN_ERR "%s: Unable to map controller.\n", c->name); @@ -199,40 +191,6 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c) c->post_port = c->base.virt + 0x40; c->reply_port = c->base.virt + 0x44; -#ifdef CONFIG_MTRR - /* Enable Write Combining MTRR for IOP's memory region */ - c->mtrr_reg0 = mtrr_add(c->in_queue.phys, c->in_queue.len, - MTRR_TYPE_WRCOMB, 1); - c->mtrr_reg1 = -1; - - if (c->mtrr_reg0 < 0) - printk(KERN_WARNING "%s: could not enable write combining " - "MTRR\n", c->name); - else - printk(KERN_INFO "%s: using write combining MTRR\n", c->name); - - /* - * If it is an INTEL i960 I/O processor then set the first 64K to - * Uncacheable since the region contains the messaging unit which - * shouldn't be cached. - */ - if ((pdev->vendor == PCI_VENDOR_ID_INTEL || - pdev->vendor == PCI_VENDOR_ID_DPT) && !c->raptor) { - printk(KERN_INFO "%s: MTRR workaround for Intel i960 processor" - "\n", c->name); - c->mtrr_reg1 = mtrr_add(c->base.phys, 0x10000, - MTRR_TYPE_UNCACHABLE, 1); - - if (c->mtrr_reg1 < 0) { - printk(KERN_WARNING "%s: Error in setting " - "MTRR_TYPE_UNCACHABLE\n", c->name); - mtrr_del(c->mtrr_reg0, c->in_queue.phys, - c->in_queue.len); - c->mtrr_reg0 = -1; - } - } -#endif - if (i2o_dma_alloc(dev, &c->status, 8, GFP_KERNEL)) { i2o_pci_free(c); return -ENOMEM; @@ -385,28 +343,25 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, { struct i2o_controller *c; int rc; + struct pci_dev *i960 = NULL; printk(KERN_INFO "i2o: Checking for PCI I2O controllers...\n"); if ((pdev->class & 0xff) > 1) { - printk(KERN_WARNING "i2o: I2O controller found but does not " - "support I2O 1.5 (skipping).\n"); + printk(KERN_WARNING "i2o: %s does not support I2O 1.5 " + "(skipping).\n", pci_name(pdev)); return -ENODEV; } if ((rc = pci_enable_device(pdev))) { - printk(KERN_WARNING "i2o: I2O controller found but could not be" - " enabled.\n"); + printk(KERN_WARNING "i2o: couldn't enable device %s\n", + pci_name(pdev)); return rc; } - printk(KERN_INFO "i2o: I2O controller found on bus %d at %d.\n", - pdev->bus->number, pdev->devfn); - if (pci_set_dma_mask(pdev, DMA_32BIT_MASK)) { - printk(KERN_WARNING "i2o: I2O controller on bus %d at %d: No " - "suitable DMA available!\n", pdev->bus->number, - pdev->devfn); + printk(KERN_WARNING "i2o: no suitable DMA found for %s\n", + pci_name(pdev)); rc = -ENODEV; goto disable; } @@ -415,11 +370,13 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, c = i2o_iop_alloc(); if (IS_ERR(c)) { - printk(KERN_ERR "i2o: memory for I2O controller could not be " - "allocated\n"); + printk(KERN_ERR "i2o: couldn't allocate memory for %s\n", + pci_name(pdev)); rc = PTR_ERR(c); goto disable; - } + } else + printk(KERN_INFO "%s: controller found (%s)\n", c->name, + pci_name(pdev)); c->pdev = pdev; c->device = pdev->dev; @@ -432,9 +389,18 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, } if (pdev->subsystem_vendor == PCI_VENDOR_ID_PROMISE) { + /* + * Expose the ship behind i960 for initialization, or it will + * failed + */ + i960 = + pci_find_slot(c->pdev->bus->number, + PCI_DEVFN(PCI_SLOT(c->pdev->devfn), 0)); + + if (i960) + pci_write_config_word(i960, 0x42, 0); + c->promise = 1; - printk(KERN_INFO "%s: Promise workarounds activated.\n", - c->name); } /* Cards that go bananas if you quiesce them before you reset them. */ @@ -459,6 +425,9 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, if ((rc = i2o_iop_add(c))) goto uninstall; + if (i960) + pci_write_config_word(i960, 0x42, 0x03ff); + return 0; uninstall: diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h index ef7f644dd873..3414325bdcfd 100644 --- a/include/linux/i2o-dev.h +++ b/include/linux/i2o-dev.h @@ -24,6 +24,14 @@ #define MAX_I2O_CONTROLLERS 32 //#include +#ifndef __KERNEL__ + +typedef unsigned char u8; +typedef unsigned short u16; +typedef unsigned int u32; + +#endif /* __KERNEL__ */ + /* * I2O Control IOCTLs and structures @@ -126,14 +134,6 @@ struct i2o_evt_get { #define I2O_BUS_CARDBUS 7 #define I2O_BUS_UNKNOWN 0x80 -#ifndef __KERNEL__ - -typedef unsigned char u8; -typedef unsigned short u16; -typedef unsigned int u32; - -#endif /* __KERNEL__ */ - typedef struct _i2o_pci_bus { u8 PciFunctionNumber; u8 PciDeviceNumber; diff --git a/include/linux/i2o.h b/include/linux/i2o.h index ea9a3ad4b67f..40e45a83d3fb 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -152,11 +152,6 @@ struct i2o_controller { unsigned int raptor:1; /* split bar */ unsigned int promise:1; /* Promise controller */ -#ifdef CONFIG_MTRR - int mtrr_reg0; - int mtrr_reg1; -#endif - struct list_head devices; /* list of I2O devices */ struct notifier_block *event_notifer; /* Events */ -- cgit v1.2.3 From f88e119c4b824a5017456fa094950d0f4092d96c Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:14 -0700 Subject: [PATCH] I2O: first code cleanup of spare warnings and unused functions Changes: - Removed unnecessary checking of NULL before calling kfree() - Make some functions static - Changed pr_debug() into osm_debug() - Use i2o_msg_in_to_virt() for getting a pointer to the message frame - Cleaned up some comments - Changed some le32_to_cpu() into readl() where necessary - Make error messages of OSM's look the same - Cleaned up error handling in i2o_block_end_request() - Removed unused error handling of failed messages in Block-OSM, which are not allowed by the I2O spec - Corrected the blocksize detection in i2o_block - Added hrt and lct sysfs-attribute to controller - Call done() function in SCSI-OSM after freeing DMA buffers - Removed unneeded variable for message size calculation in i2o_scsi_queuecommand() - Make some changes to remove sparse warnings - Reordered some functions - Cleaned up controller initialization - Replaced some magic numbers by defines - Removed unnecessary dma_sync_single_for_cpu() call on coherent DMA - Removed some unused fields in i2o_controller and removed some unused functions Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/i2o/device.c | 9 +- drivers/message/i2o/driver.c | 46 +++++---- drivers/message/i2o/exec-osm.c | 47 +++++---- drivers/message/i2o/i2o_block.c | 211 +++++++++++++-------------------------- drivers/message/i2o/i2o_block.h | 2 +- drivers/message/i2o/i2o_config.c | 118 +++++++++++++++++++++- drivers/message/i2o/i2o_scsi.c | 31 +++--- drivers/message/i2o/iop.c | 87 ++++++++++------ drivers/message/i2o/pci.c | 67 ++++++------- include/linux/i2o.h | 74 +++----------- 10 files changed, 356 insertions(+), 336 deletions(-) (limited to 'include/linux') diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c index 280627ae6cf7..f1b7eb63d54b 100644 --- a/drivers/message/i2o/device.c +++ b/drivers/message/i2o/device.c @@ -282,8 +282,7 @@ int i2o_device_parse_lct(struct i2o_controller *c) down(&c->lct_lock); - if (c->lct) - kfree(c->lct); + kfree(c->lct); lct = c->dlct.virt; @@ -447,8 +446,8 @@ static struct class_interface i2o_device_class_interface = { * ResultCount, ErrorInfoSize, BlockStatus and BlockSize. */ -int i2o_parm_issue(struct i2o_device *i2o_dev, int cmd, void *oplist, - int oplen, void *reslist, int reslen) +static int i2o_parm_issue(struct i2o_device *i2o_dev, int cmd, void *oplist, + int oplen, void *reslist, int reslen) { struct i2o_message __iomem *msg; u32 m; @@ -540,7 +539,7 @@ int i2o_parm_field_get(struct i2o_device *i2o_dev, int group, int field, opblk[4] = -1; size = i2o_parm_issue(i2o_dev, I2O_CMD_UTIL_PARAMS_GET, opblk, - sizeof(opblk), resblk, sizeof(resblk)); + sizeof(opblk), resblk, buflen + 8); memcpy(buf, resblk + 8, buflen); /* cut off header */ diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c index c71e68f70e7d..bebdd509b5d8 100644 --- a/drivers/message/i2o/driver.c +++ b/drivers/message/i2o/driver.c @@ -18,7 +18,7 @@ #include #include -#define OSM_NAME "core" +#define OSM_NAME "i2o" /* max_drivers - Maximum I2O drivers (OSMs) which could be registered */ unsigned int i2o_max_drivers = I2O_MAX_DRIVERS; @@ -78,17 +78,16 @@ int i2o_driver_register(struct i2o_driver *drv) int rc = 0; unsigned long flags; - pr_debug("i2o: Register driver %s\n", drv->name); + osm_debug("Register driver %s\n", drv->name); if (drv->event) { drv->event_queue = create_workqueue(drv->name); if (!drv->event_queue) { - printk(KERN_ERR "i2o: Could not initialize event queue " - "for driver %s\n", drv->name); + osm_err("Could not initialize event queue for driver " + "%s\n", drv->name); return -EFAULT; } - pr_debug("i2o: Event queue initialized for driver %s\n", - drv->name); + osm_debug("Event queue initialized for driver %s\n", drv->name); } else drv->event_queue = NULL; @@ -99,8 +98,8 @@ int i2o_driver_register(struct i2o_driver *drv) for (i = 0; i2o_drivers[i]; i++) if (i >= i2o_max_drivers) { - printk(KERN_ERR "i2o: too many drivers registered, " - "increase max_drivers\n"); + osm_err("too many drivers registered, increase " + "max_drivers\n"); spin_unlock_irqrestore(&i2o_drivers_lock, flags); return -EFAULT; } @@ -110,8 +109,7 @@ int i2o_driver_register(struct i2o_driver *drv) spin_unlock_irqrestore(&i2o_drivers_lock, flags); - pr_debug("i2o: driver %s gets context id %d\n", drv->name, - drv->context); + osm_debug("driver %s gets context id %d\n", drv->name, drv->context); list_for_each_entry(c, &i2o_controllers, list) { struct i2o_device *i2o_dev; @@ -141,7 +139,7 @@ void i2o_driver_unregister(struct i2o_driver *drv) struct i2o_controller *c; unsigned long flags; - pr_debug("i2o: unregister driver %s\n", drv->name); + osm_debug("unregister driver %s\n", drv->name); driver_unregister(&drv->driver); @@ -161,7 +159,7 @@ void i2o_driver_unregister(struct i2o_driver *drv) if (drv->event_queue) { destroy_workqueue(drv->event_queue); drv->event_queue = NULL; - pr_debug("i2o: event queue removed for %s\n", drv->name); + osm_debug("event queue removed for %s\n", drv->name); } }; @@ -178,15 +176,15 @@ void i2o_driver_unregister(struct i2o_driver *drv) * on success and if the message should be flushed afterwords. Returns * negative error code on failure (the message will be flushed too). */ -int i2o_driver_dispatch(struct i2o_controller *c, u32 m, - struct i2o_message __iomem *msg) +int i2o_driver_dispatch(struct i2o_controller *c, u32 m) { struct i2o_driver *drv; + struct i2o_message __iomem *msg = i2o_msg_out_to_virt(c, m); u32 context = readl(&msg->u.s.icntxt); if (unlikely(context >= i2o_max_drivers)) { - printk(KERN_WARNING "%s: Spurious reply to unknown driver " - "%d\n", c->name, readl(&msg->u.s.icntxt)); + osm_warn("%s: Spurious reply to unknown driver %d\n", c->name, + context); return -EIO; } @@ -195,7 +193,8 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m, spin_unlock(&i2o_drivers_lock); if (unlikely(!drv)) { - osm_warn("Spurious reply to unknown driver %d\n", context); + osm_warn("%s: Spurious reply to unknown driver %d\n", c->name, + context); return -EIO; } @@ -207,6 +206,9 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m, osm_debug("event received from device %d\n", tid); + if (!drv->event) + return -EIO; + /* cut of header from message size (in 32-bit words) */ size = (readl(&msg->u.head[0]) >> 16) - 5; @@ -231,8 +233,8 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m, } if (unlikely(!drv->reply)) { - pr_debug("%s: Reply to driver %s, but no reply function" - " defined!\n", c->name, drv->name); + osm_debug("%s: Reply to driver %s, but no reply function" + " defined!\n", c->name, drv->name); return -EIO; } @@ -333,11 +335,11 @@ int __init i2o_driver_init(void) if ((i2o_max_drivers < 2) || (i2o_max_drivers > 64) || ((i2o_max_drivers ^ (i2o_max_drivers - 1)) != (2 * i2o_max_drivers - 1))) { - printk(KERN_WARNING "i2o: max_drivers set to %d, but must be " - ">=2 and <= 64 and a power of 2\n", i2o_max_drivers); + osm_warn("max_drivers set to %d, but must be >=2 and <= 64 and " + "a power of 2\n", i2o_max_drivers); i2o_max_drivers = I2O_MAX_DRIVERS; } - printk(KERN_INFO "i2o: max drivers = %d\n", i2o_max_drivers); + osm_info("max drivers = %d\n", i2o_max_drivers); i2o_drivers = kmalloc(i2o_max_drivers * sizeof(*i2o_drivers), GFP_KERNEL); diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index 1e28e886f1ca..5581344fbba6 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -108,7 +108,8 @@ static void i2o_exec_wait_free(struct i2o_exec_wait *wait) * buffer must not be freed. Instead the event completion will free them * for you. In all other cases the buffer are your problem. * - * Returns 0 on success or negative error code on failure. + * Returns 0 on success, negative error code on timeout or positive error + * code from reply. */ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long timeout, struct i2o_dma *dma) @@ -116,7 +117,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long DECLARE_WAIT_QUEUE_HEAD(wq); struct i2o_exec_wait *wait; static u32 tcntxt = 0x80000000; - struct i2o_message __iomem *msg = c->in_queue.virt + m; + struct i2o_message __iomem *msg = i2o_msg_in_to_virt(c, m); int rc = 0; wait = i2o_exec_wait_alloc(); @@ -161,8 +162,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long barrier(); if (wait->complete) { - if (readl(&wait->msg->body[0]) >> 24) - rc = readl(&wait->msg->body[0]) & 0xff; + rc = readl(&wait->msg->body[0]) >> 24; i2o_flush_reply(c, wait->m); i2o_exec_wait_free(wait); } else { @@ -187,6 +187,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long * @c: I2O controller which answers * @m: message id * @msg: pointer to the I2O reply message + * @context: transaction context of request * * This function is called in interrupt context only. If the reply reached * before the timeout, the i2o_exec_wait struct is filled with the message @@ -201,14 +202,12 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long * message must also be given back to the controller. */ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, - struct i2o_message __iomem *msg) + struct i2o_message __iomem *msg, + u32 context) { struct i2o_exec_wait *wait, *tmp; static spinlock_t lock = SPIN_LOCK_UNLOCKED; int rc = 1; - u32 context; - - context = readl(&msg->u.s.tcntxt); /* * We need to search through the i2o_exec_wait_list to see if the given @@ -251,7 +250,7 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, spin_unlock(&lock); - pr_debug("%s: Bogus reply in POST WAIT (tr-context: %08x)!\n", c->name, + osm_warn("%s: Bogus reply in POST WAIT (tr-context: %08x)!\n", c->name, context); return -1; @@ -321,29 +320,35 @@ static void i2o_exec_lct_modified(struct i2o_controller *c) * code on failure and if the reply should be flushed. */ static int i2o_exec_reply(struct i2o_controller *c, u32 m, - struct i2o_message *msg) + struct i2o_message __iomem *msg) { - if (le32_to_cpu(msg->u.head[0]) & MSG_FAIL) { // Fail bit is set - struct i2o_message __iomem *pmsg; /* preserved message */ + u32 context; + + if (readl(&msg->u.head[0]) & MSG_FAIL) { + /* + * If Fail bit is set we must take the transaction context of + * the preserved message to find the right request again. + */ + struct i2o_message __iomem *pmsg; u32 pm; - pm = le32_to_cpu(msg->body[3]); + pm = readl(&msg->body[3]); pmsg = i2o_msg_in_to_virt(c, pm); i2o_report_status(KERN_INFO, "i2o_core", msg); - /* Release the preserved msg by resubmitting it as a NOP */ - i2o_msg_nop(c, pm); + context = readl(&pmsg->u.s.tcntxt); - /* If reply to i2o_post_wait failed, return causes a timeout */ - return -1; - } + /* Release the preserved msg */ + i2o_msg_nop(c, pm); + } else + context = readl(&msg->u.s.tcntxt); - if (le32_to_cpu(msg->u.s.tcntxt) & 0x80000000) - return i2o_msg_post_wait_complete(c, m, msg); + if (context & 0x80000000) + return i2o_msg_post_wait_complete(c, m, msg, context); - if ((le32_to_cpu(msg->u.head[1]) >> 24) == I2O_CMD_LCT_NOTIFY) { + if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_LCT_NOTIFY) { struct work_struct *work; pr_debug("%s: LCT notify received\n", c->name); diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c index 4830b7759061..e69421e36ac5 100644 --- a/drivers/message/i2o/i2o_block.c +++ b/drivers/message/i2o/i2o_block.c @@ -104,7 +104,8 @@ static int i2o_block_remove(struct device *dev) struct i2o_device *i2o_dev = to_i2o_device(dev); struct i2o_block_device *i2o_blk_dev = dev_get_drvdata(dev); - osm_info("Device removed %s\n", i2o_blk_dev->gd->disk_name); + osm_info("device removed (TID: %03x): %s\n", i2o_dev->lct_data.tid, + i2o_blk_dev->gd->disk_name); i2o_event_register(i2o_dev, &i2o_block_driver, 0, 0); @@ -400,71 +401,62 @@ static void i2o_block_delayed_request_fn(void *delayed_request) }; /** - * i2o_block_reply - Block OSM reply handler. - * @c: I2O controller from which the message arrives - * @m: message id of reply - * qmsg: the actuall I2O message reply + * i2o_block_end_request - Post-processing of completed commands + * @req: request which should be completed + * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error + * @nr_bytes: number of bytes to complete * - * This function gets all the message replies. + * Mark the request as complete. The lock must not be held when entering. * */ -static int i2o_block_reply(struct i2o_controller *c, u32 m, - struct i2o_message *msg) +static void i2o_block_end_request(struct request *req, int uptodate, + int nr_bytes) { - struct i2o_block_request *ireq; - struct request *req; - struct i2o_block_device *dev; - struct request_queue *q; - u8 st; + struct i2o_block_request *ireq = req->special; + struct i2o_block_device *dev = ireq->i2o_blk_dev; + request_queue_t *q = dev->gd->queue; unsigned long flags; - /* FAILed message */ - if (unlikely(le32_to_cpu(msg->u.head[0]) & (1 << 13))) { - struct i2o_message *pmsg; - u32 pm; - - /* - * FAILed message from controller - * We increment the error count and abort it - * - * In theory this will never happen. The I2O block class - * specification states that block devices never return - * FAILs but instead use the REQ status field...but - * better be on the safe side since no one really follows - * the spec to the book :) - */ - pm = le32_to_cpu(msg->body[3]); - pmsg = i2o_msg_in_to_virt(c, pm); + if (end_that_request_chunk(req, uptodate, nr_bytes)) { + int leftover = (req->hard_nr_sectors << 9); - req = i2o_cntxt_list_get(c, le32_to_cpu(pmsg->u.s.tcntxt)); - if (unlikely(!req)) { - osm_err("NULL reply received!\n"); - return -1; - } + if (blk_pc_request(req)) + leftover = req->data_len; - ireq = req->special; - dev = ireq->i2o_blk_dev; - q = dev->gd->queue; + if (end_io_error(uptodate)) + end_that_request_chunk(req, 0, leftover); + } - req->errors++; + add_disk_randomness(req->rq_disk); - spin_lock_irqsave(q->queue_lock, flags); + spin_lock_irqsave(q->queue_lock, flags); - while (end_that_request_chunk(req, !req->errors, - le32_to_cpu(pmsg->body[1]))) ; - end_that_request_last(req); + end_that_request_last(req); + dev->open_queue_depth--; + list_del(&ireq->queue); - dev->open_queue_depth--; - list_del(&ireq->queue); - blk_start_queue(q); + blk_start_queue(q); - spin_unlock_irqrestore(q->queue_lock, flags); + spin_unlock_irqrestore(q->queue_lock, flags); - /* Now flush the message by making it a NOP */ - i2o_msg_nop(c, pm); + i2o_block_sglist_free(ireq); + i2o_block_request_free(ireq); +}; - return -1; - } +/** + * i2o_block_reply - Block OSM reply handler. + * @c: I2O controller from which the message arrives + * @m: message id of reply + * qmsg: the actuall I2O message reply + * + * This function gets all the message replies. + * + */ +static int i2o_block_reply(struct i2o_controller *c, u32 m, + struct i2o_message *msg) +{ + struct request *req; + int uptodate = 1; req = i2o_cntxt_list_get(c, le32_to_cpu(msg->u.s.tcntxt)); if (unlikely(!req)) { @@ -472,61 +464,13 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m, return -1; } - ireq = req->special; - dev = ireq->i2o_blk_dev; - q = dev->gd->queue; - - if (unlikely(!dev->i2o_dev)) { - /* - * This is HACK, but Intel Integrated RAID allows user - * to delete a volume that is claimed, locked, and in use - * by the OS. We have to check for a reply from a - * non-existent device and flag it as an error or the system - * goes kaput... - */ - req->errors++; - osm_warn("Data transfer to deleted device!\n"); - spin_lock_irqsave(q->queue_lock, flags); - while (end_that_request_chunk - (req, !req->errors, le32_to_cpu(msg->body[1]))) ; - end_that_request_last(req); - - dev->open_queue_depth--; - list_del(&ireq->queue); - blk_start_queue(q); - - spin_unlock_irqrestore(q->queue_lock, flags); - return -1; - } - /* * Lets see what is cooking. We stuffed the * request in the context. */ - st = le32_to_cpu(msg->body[0]) >> 24; - - if (st != 0) { - int err; - char *bsa_errors[] = { - "Success", - "Media Error", - "Failure communicating to device", - "Device Failure", - "Device is not ready", - "Media not present", - "Media is locked by another user", - "Media has failed", - "Failure communicating to device", - "Device bus failure", - "Device is locked by another user", - "Device is write protected", - "Device has reset", - "Volume has changed, waiting for acknowledgement" - }; - - err = le32_to_cpu(msg->body[0]) & 0xffff; - + if ((le32_to_cpu(msg->body[0]) >> 24) != 0) { + u32 status = le32_to_cpu(msg->body[0]); /* * Device not ready means two things. One is that the * the thing went offline (but not a removal media) @@ -539,40 +483,23 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m, * Don't stick a supertrak100 into cache aggressive modes */ - osm_err("block-osm: /dev/%s error: %s", dev->gd->disk_name, - bsa_errors[le32_to_cpu(msg->body[0]) & 0xffff]); - if (le32_to_cpu(msg->body[0]) & 0x00ff0000) - printk(KERN_ERR " - DDM attempted %d retries", - (le32_to_cpu(msg->body[0]) >> 16) & 0x00ff); - printk(KERN_ERR ".\n"); - req->errors++; - } else - req->errors = 0; - - if (!end_that_request_chunk - (req, !req->errors, le32_to_cpu(msg->body[1]))) { - add_disk_randomness(req->rq_disk); - spin_lock_irqsave(q->queue_lock, flags); + osm_err("%03x error status: %02x, detailed status: %04x\n", + (le32_to_cpu(msg->u.head[1]) >> 12 & 0xfff), + status >> 24, status & 0xffff); - end_that_request_last(req); + req->errors++; - dev->open_queue_depth--; - list_del(&ireq->queue); - blk_start_queue(q); + uptodate = 0; + } - spin_unlock_irqrestore(q->queue_lock, flags); - - i2o_block_sglist_free(ireq); - i2o_block_request_free(ireq); - } else - osm_err("still remaining chunks\n"); + i2o_block_end_request(req, uptodate, le32_to_cpu(msg->body[1])); return 1; }; static void i2o_block_event(struct i2o_event *evt) { - osm_info("block-osm: event received\n"); + osm_info("event received\n"); kfree(evt); }; @@ -875,9 +802,7 @@ static int i2o_block_transfer(struct request *req) sg++; } - writel(I2O_MESSAGE_SIZE - (((unsigned long)mptr - - (unsigned long)&msg->u.head[0]) >> 2) | SGL_OFFSET_8, + writel(I2O_MESSAGE_SIZE(mptr - &msg->u.head[0]) | SGL_OFFSET_8, &msg->u.head[0]); list_add_tail(&ireq->queue, &dev->open_queue); @@ -1048,7 +973,6 @@ static int i2o_block_probe(struct device *dev) int rc; u64 size; u32 blocksize; - u16 power; u32 flags, status; int segments; @@ -1058,8 +982,6 @@ static int i2o_block_probe(struct device *dev) return -ENODEV; } - osm_info("New device detected (TID: %03x)\n", i2o_dev->lct_data.tid); - if (i2o_device_claim(i2o_dev)) { osm_warn("Unable to claim device. Installation aborted\n"); rc = -EFAULT; @@ -1111,15 +1033,21 @@ static int i2o_block_probe(struct device *dev) * Ask for the current media data. If that isn't supported * then we ask for the device capacity data */ - if (i2o_parm_field_get(i2o_dev, 0x0004, 1, &blocksize, 4) != 0 - || i2o_parm_field_get(i2o_dev, 0x0004, 0, &size, 8) != 0) { - i2o_parm_field_get(i2o_dev, 0x0000, 3, &blocksize, 4); - i2o_parm_field_get(i2o_dev, 0x0000, 4, &size, 8); - } - osm_debug("blocksize = %d\n", blocksize); + if (!i2o_parm_field_get(i2o_dev, 0x0004, 0, &size, 8)) + if (!i2o_parm_field_get(i2o_dev, 0x0000, 4, &size, 8)) { + osm_warn("could not get size of %s\n", gd->disk_name); + size = 0; + } - if (i2o_parm_field_get(i2o_dev, 0x0000, 2, &power, 2)) - power = 0; + if (!i2o_parm_field_get(i2o_dev, 0x0004, 1, &blocksize, 4)) + if (!i2o_parm_field_get(i2o_dev, 0x0000, 3, &blocksize, 4)) { + osm_warn("unable to get blocksize of %s\n", + gd->disk_name); + blocksize = 0; + } + + if (!i2o_parm_field_get(i2o_dev, 0x0000, 2, &i2o_blk_dev->power, 2)) + i2o_blk_dev->power = 0; i2o_parm_field_get(i2o_dev, 0x0000, 5, &flags, 4); i2o_parm_field_get(i2o_dev, 0x0000, 6, &status, 4); @@ -1131,6 +1059,9 @@ static int i2o_block_probe(struct device *dev) unit++; + osm_info("device added (TID: %03x): %s\n", i2o_dev->lct_data.tid, + i2o_blk_dev->gd->disk_name); + return 0; claim_release: diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h index ddd9a15679c0..712111ffa638 100644 --- a/drivers/message/i2o/i2o_block.h +++ b/drivers/message/i2o/i2o_block.h @@ -74,7 +74,7 @@ struct i2o_block_device { int rcache; /* read cache flags */ int wcache; /* write cache flags */ int flags; - int power; /* power state */ + u16 power; /* power state */ int media_change_flag; /* media changed flag */ }; diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c index 46d373287a30..383e89a5c9f0 100644 --- a/drivers/message/i2o/i2o_config.c +++ b/drivers/message/i2o/i2o_config.c @@ -80,13 +80,123 @@ struct i2o_cfg_info { static struct i2o_cfg_info *open_files = NULL; static ulong i2o_cfg_info_id = 0; -/* - * Each of these describes an i2o message handler. They are - * multiplexed by the i2o_core code +/** + * i2o_config_read_hrt - Returns the HRT of the controller + * @kob: kernel object handle + * @buf: buffer into which the HRT should be copied + * @off: file offset + * @count: number of bytes to read + * + * Put @count bytes starting at @off into @buf from the HRT of the I2O + * controller corresponding to @kobj. + * + * Returns number of bytes copied into buffer. + */ +static ssize_t i2o_config_read_hrt(struct kobject *kobj, char *buf, + loff_t offset, size_t count) +{ + struct i2o_controller *c = to_i2o_controller(container_of(kobj, + struct device, + kobj)); + i2o_hrt *hrt = c->hrt.virt; + + u32 size = (hrt->num_entries * hrt->entry_len + 2) * 4; + + if(offset > size) + return 0; + + if(offset + count > size) + count = size - offset; + + memcpy(buf, (u8 *) hrt + offset, count); + + return count; +}; + +/** + * i2o_config_read_lct - Returns the LCT of the controller + * @kob: kernel object handle + * @buf: buffer into which the LCT should be copied + * @off: file offset + * @count: number of bytes to read + * + * Put @count bytes starting at @off into @buf from the LCT of the I2O + * controller corresponding to @kobj. + * + * Returns number of bytes copied into buffer. + */ +static ssize_t i2o_config_read_lct(struct kobject *kobj, char *buf, + loff_t offset, size_t count) +{ + struct i2o_controller *c = to_i2o_controller(container_of(kobj, + struct device, + kobj)); + u32 size = c->lct->table_size * 4; + + if(offset > size) + return 0; + + if(offset + count > size) + count = size - offset; + + memcpy(buf, (u8 *) c->lct + offset, count); + + return count; +}; + +/* attribute for HRT in sysfs */ +static struct bin_attribute i2o_config_hrt_attr = { + .attr = { + .name = "hrt", + .mode = S_IRUGO, + .owner = THIS_MODULE + }, + .size = 0, + .read = i2o_config_read_hrt +}; + +/* attribute for LCT in sysfs */ +static struct bin_attribute i2o_config_lct_attr = { + .attr = { + .name = "lct", + .mode = S_IRUGO, + .owner = THIS_MODULE + }, + .size = 0, + .read = i2o_config_read_lct +}; + +/** + * i2o_config_notify_controller_add - Notify of added controller + * @c: the controller which was added + * + * If a I2O controller is added, we catch the notification to add sysfs + * entries. + */ +static void i2o_config_notify_controller_add(struct i2o_controller *c) +{ + sysfs_create_bin_file(&(c->device.kobj), &i2o_config_hrt_attr); + sysfs_create_bin_file(&(c->device.kobj), &i2o_config_lct_attr); +}; + +/** + * i2o_config_notify_controller_remove - Notify of removed controller + * @c: the controller which was removed + * + * If a I2O controller is removed, we catch the notification to remove the + * sysfs entries. */ +static void i2o_config_notify_controller_remove(struct i2o_controller *c) +{ + sysfs_remove_bin_file(&c->device.kobj, &i2o_config_lct_attr); + sysfs_remove_bin_file(&c->device.kobj, &i2o_config_hrt_attr); +}; +/* Config OSM driver struct */ static struct i2o_driver i2o_config_driver = { - .name = OSM_NAME + .name = OSM_NAME, + .notify_controller_add = i2o_config_notify_controller_add, + .notify_controller_remove = i2o_config_notify_controller_remove }; static int i2o_cfg_getiops(unsigned long arg) diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c index af40f1c1ec77..812c29ec86d3 100644 --- a/drivers/message/i2o/i2o_scsi.c +++ b/drivers/message/i2o/i2o_scsi.c @@ -40,6 +40,7 @@ * Fix the resource management problems. */ +#define DEBUG 1 #include #include #include @@ -179,6 +180,8 @@ static int i2o_scsi_remove(struct device *dev) struct i2o_scsi_host *i2o_shost; struct scsi_device *scsi_dev; + osm_info("device removed (TID: %03x)\n", i2o_dev->lct_data.tid); + i2o_shost = i2o_scsi_get_host(c); shost_for_each_device(scsi_dev, i2o_shost->scsi_host) @@ -262,8 +265,8 @@ static int i2o_scsi_probe(struct device *dev) return -EFAULT; } - osm_debug("added new SCSI device %03x (cannel: %d, id: %d, lun: %d)\n", - i2o_dev->lct_data.tid, channel, id, (unsigned int)lun); + osm_info("device added (TID: %03x) channel: %d, id: %d, lun: %d\n", + i2o_dev->lct_data.tid, channel, id, (unsigned int)lun); return 0; }; @@ -439,8 +442,6 @@ static int i2o_scsi_reply(struct i2o_controller *c, u32 m, cmd->result = DID_OK << 16 | ds; - cmd->scsi_done(cmd); - dev = &c->pdev->dev; if (cmd->use_sg) dma_unmap_sg(dev, (struct scatterlist *)cmd->buffer, @@ -449,6 +450,8 @@ static int i2o_scsi_reply(struct i2o_controller *c, u32 m, dma_unmap_single(dev, (dma_addr_t) ((long)cmd->SCp.ptr), cmd->request_bufflen, cmd->sc_data_direction); + cmd->scsi_done(cmd); + return 1; }; @@ -502,7 +505,7 @@ static void i2o_scsi_notify_controller_remove(struct i2o_controller *c) scsi_remove_host(i2o_shost->scsi_host); scsi_host_put(i2o_shost->scsi_host); - pr_info("I2O SCSI host removed\n"); + osm_debug("I2O SCSI host removed\n"); }; /* SCSI OSM driver struct */ @@ -545,7 +548,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, u32 scsi_flags, sg_flags; u32 __iomem *mptr; u32 __iomem *lenptr; - u32 len, reqlen; + u32 len; int i; /* @@ -580,12 +583,12 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, if (m == I2O_QUEUE_EMPTY) return SCSI_MLQUEUE_HOST_BUSY; + mptr = &msg->body[0]; + /* * Put together a scsi execscb message */ - len = SCpnt->request_bufflen; - switch (SCpnt->sc_data_direction) { case PCI_DMA_NONE: scsi_flags = 0x00000000; // DATA NO XFER @@ -637,17 +640,13 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, */ /* Direction, disconnect ok, tag, CDBLen */ - writel(scsi_flags | 0x20200000 | SCpnt->cmd_len, &msg->body[0]); - - mptr = &msg->body[1]; + writel(scsi_flags | 0x20200000 | SCpnt->cmd_len, mptr ++); /* Write SCSI command into the message - always 16 byte block */ memcpy_toio(mptr, SCpnt->cmnd, 16); mptr += 4; lenptr = mptr++; /* Remember me - fill in when we know */ - reqlen = 12; // SINGLE SGE - /* Now fill in the SGList and command */ if (SCpnt->use_sg) { struct scatterlist *sg; @@ -671,7 +670,6 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, sg++; } - reqlen = mptr - &msg->u.head[0]; writel(len, lenptr); } else { len = SCpnt->request_bufflen; @@ -691,12 +689,11 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, sg_flags |= 0xC0000000; writel(sg_flags | SCpnt->request_bufflen, mptr++); writel(dma_addr, mptr++); - } else - reqlen = 9; + } } /* Stick the headers on */ - writel(reqlen << 16 | SGL_OFFSET_10, &msg->u.head[0]); + writel((mptr - &msg->u.head[0]) << 16 | SGL_OFFSET_10, &msg->u.head[0]); /* Queue the message */ i2o_msg_post(c, m); diff --git a/drivers/message/i2o/iop.c b/drivers/message/i2o/iop.c index 50c8cedf7a2d..62b0d8bed186 100644 --- a/drivers/message/i2o/iop.c +++ b/drivers/message/i2o/iop.c @@ -68,7 +68,7 @@ extern void i2o_device_exit(void); */ void i2o_msg_nop(struct i2o_controller *c, u32 m) { - struct i2o_message __iomem *msg = c->in_queue.virt + m; + struct i2o_message __iomem *msg = i2o_msg_in_to_virt(c, m); writel(THREE_WORD_MSG_SIZE | SGL_OFFSET_0, &msg->u.head[0]); writel(I2O_CMD_UTIL_NOP << 24 | HOST_TID << 12 | ADAPTER_TID, @@ -452,8 +452,6 @@ static int i2o_iop_clear(struct i2o_controller *c) /* Enable all IOPs */ i2o_iop_enable_all(); - i2o_status_get(c); - return rc; } @@ -591,12 +589,11 @@ static int i2o_iop_init_outbound_queue(struct i2o_controller *c) if (m == I2O_QUEUE_EMPTY) return -ETIMEDOUT; - writel(EIGHT_WORD_MSG_SIZE | TRL_OFFSET_6, &msg->u.head[0]); + writel(EIGHT_WORD_MSG_SIZE | SGL_OFFSET_6, &msg->u.head[0]); writel(I2O_CMD_OUTBOUND_INIT << 24 | HOST_TID << 12 | ADAPTER_TID, &msg->u.head[1]); writel(i2o_exec_driver.context, &msg->u.s.icntxt); - writel(0x0106, &msg->u.s.tcntxt); /* FIXME: why 0x0106, maybe in - Spec? */ + writel(0x00000000, &msg->u.s.tcntxt); writel(PAGE_SIZE, &msg->body[0]); writel(MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]); /* Outbound msg frame size in words and Initcode */ @@ -891,8 +888,12 @@ void i2o_iop_remove(struct i2o_controller *c) list_for_each_entry_safe(dev, tmp, &c->devices, list) i2o_device_remove(dev); + device_del(&c->device); + /* Ask the IOP to switch to RESET state */ i2o_iop_reset(c); + + put_device(&c->device); } /** @@ -971,8 +972,10 @@ static int i2o_systab_build(void) systab->iops[count].frame_size = sb->inbound_frame_size; systab->iops[count].last_changed = change_ind; systab->iops[count].iop_capabilities = sb->iop_capabilities; - systab->iops[count].inbound_low = i2o_ptr_low(c->post_port); - systab->iops[count].inbound_high = i2o_ptr_high(c->post_port); + systab->iops[count].inbound_low = + i2o_dma_low(c->base.phys + I2O_IN_PORT); + systab->iops[count].inbound_high = + i2o_dma_high(c->base.phys + I2O_IN_PORT); count++; } @@ -1109,6 +1112,30 @@ static int i2o_hrt_get(struct i2o_controller *c) return -EBUSY; } +/** + * i2o_iop_free - Free the i2o_controller struct + * @c: I2O controller to free + */ +void i2o_iop_free(struct i2o_controller *c) +{ + kfree(c); +}; + + +/** + * i2o_iop_release - release the memory for a I2O controller + * @dev: I2O controller which should be released + * + * Release the allocated memory. This function is called if refcount of + * device reaches 0 automatically. + */ +static void i2o_iop_release(struct device *dev) +{ + struct i2o_controller *c = to_i2o_controller(dev); + + i2o_iop_free(c); +}; + /** * i2o_iop_alloc - Allocate and initialize a i2o_controller struct * @@ -1137,6 +1164,10 @@ struct i2o_controller *i2o_iop_alloc(void) c->unit = unit++; sprintf(c->name, "iop%d", c->unit); + device_initialize(&c->device); + c->device.release = &i2o_iop_release; + snprintf(c->device.bus_id, BUS_ID_SIZE, "iop%d", c->unit); + #if BITS_PER_LONG == 64 spin_lock_init(&c->context_list_lock); atomic_set(&c->context_list_counter, 0); @@ -1146,15 +1177,6 @@ struct i2o_controller *i2o_iop_alloc(void) return c; }; -/** - * i2o_iop_free - Free the i2o_controller struct - * @c: I2O controller to free - */ -void i2o_iop_free(struct i2o_controller *c) -{ - kfree(c); -}; - /** * i2o_iop_add - Initialize the I2O controller and add him to the I2O core * @c: controller @@ -1168,6 +1190,11 @@ int i2o_iop_add(struct i2o_controller *c) { int rc; + if((rc = device_add(&c->device))) { + printk(KERN_ERR "%s: could not register controller\n", c->name); + goto iop_reset; + } + printk(KERN_INFO "%s: Activating I2O controller...\n", c->name); printk(KERN_INFO "%s: This may take a few minutes if there are many " "devices\n", c->name); @@ -1175,30 +1202,23 @@ int i2o_iop_add(struct i2o_controller *c) if ((rc = i2o_iop_activate(c))) { printk(KERN_ERR "%s: could not activate controller\n", c->name); - i2o_iop_reset(c); - return rc; + goto iop_reset; } pr_debug("%s: building sys table...\n", c->name); - if ((rc = i2o_systab_build())) { - i2o_iop_reset(c); - return rc; - } + if ((rc = i2o_systab_build())) + goto iop_reset; pr_debug("%s: online controller...\n", c->name); - if ((rc = i2o_iop_online(c))) { - i2o_iop_reset(c); - return rc; - } + if ((rc = i2o_iop_online(c))) + goto iop_reset; pr_debug("%s: getting LCT...\n", c->name); - if ((rc = i2o_exec_lct_get(c))) { - i2o_iop_reset(c); - return rc; - } + if ((rc = i2o_exec_lct_get(c))) + goto iop_reset; list_add(&c->list, &i2o_controllers); @@ -1207,6 +1227,11 @@ int i2o_iop_add(struct i2o_controller *c) printk(KERN_INFO "%s: Controller added\n", c->name); return 0; + +iop_reset: + i2o_iop_reset(c); + + return rc; }; /** diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c index 579a8b7a2120..f33fd81f77a4 100644 --- a/drivers/message/i2o/pci.c +++ b/drivers/message/i2o/pci.c @@ -38,8 +38,7 @@ extern void i2o_iop_free(struct i2o_controller *); extern int i2o_iop_add(struct i2o_controller *); extern void i2o_iop_remove(struct i2o_controller *); -extern int i2o_driver_dispatch(struct i2o_controller *, u32, - struct i2o_message *); +extern int i2o_driver_dispatch(struct i2o_controller *, u32); /* PCI device id table for all I2O controllers */ static struct pci_device_id __devinitdata i2o_pci_ids[] = { @@ -89,8 +88,7 @@ static void i2o_pci_free(struct i2o_controller *c) i2o_dma_free(dev, &c->out_queue); i2o_dma_free(dev, &c->status_block); - if (c->lct) - kfree(c->lct); + kfree(c->lct); i2o_dma_free(dev, &c->dlct); i2o_dma_free(dev, &c->hrt); i2o_dma_free(dev, &c->status); @@ -187,9 +185,9 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c) } else c->in_queue = c->base; - c->irq_mask = c->base.virt + 0x34; - c->post_port = c->base.virt + 0x40; - c->reply_port = c->base.virt + 0x44; + c->irq_mask = c->base.virt + I2O_IRQ_MASK; + c->in_port = c->base.virt + I2O_IN_PORT; + c->out_port = c->base.virt + I2O_OUT_PORT; if (i2o_dma_alloc(dev, &c->status, 8, GFP_KERNEL)) { i2o_pci_free(c); @@ -235,49 +233,34 @@ static irqreturn_t i2o_pci_interrupt(int irq, void *dev_id, struct pt_regs *r) { struct i2o_controller *c = dev_id; struct device *dev = &c->pdev->dev; - struct i2o_message *m; - u32 mv; + u32 mv = readl(c->out_port); /* * Old 960 steppings had a bug in the I2O unit that caused * the queue to appear empty when it wasn't. */ - mv = I2O_REPLY_READ32(c); if (mv == I2O_QUEUE_EMPTY) { - mv = I2O_REPLY_READ32(c); - if (unlikely(mv == I2O_QUEUE_EMPTY)) { + mv = readl(c->out_port); + if (unlikely(mv == I2O_QUEUE_EMPTY)) return IRQ_NONE; - } else + else pr_debug("%s: 960 bug detected\n", c->name); } while (mv != I2O_QUEUE_EMPTY) { - /* - * Map the message from the page frame map to kernel virtual. - * Because bus_to_virt is deprecated, we have calculate the - * location by ourself! - */ - m = i2o_msg_out_to_virt(c, mv); - - /* - * Ensure this message is seen coherently but cachably by - * the processor - */ - dma_sync_single_for_cpu(dev, mv, MSG_FRAME_SIZE * 4, - PCI_DMA_FROMDEVICE); - /* dispatch it */ - if (i2o_driver_dispatch(c, mv, m)) + if (i2o_driver_dispatch(c, mv)) /* flush it if result != 0 */ i2o_flush_reply(c, mv); /* * That 960 bug again... */ - mv = I2O_REPLY_READ32(c); + mv = readl(c->out_port); if (mv == I2O_QUEUE_EMPTY) - mv = I2O_REPLY_READ32(c); + mv = readl(c->out_port); } + return IRQ_HANDLED; } @@ -294,7 +277,9 @@ static int i2o_pci_irq_enable(struct i2o_controller *c) struct pci_dev *pdev = c->pdev; int rc; - I2O_IRQ_WRITE32(c, 0xffffffff); + wmb(); + writel(0xffffffff, c->irq_mask); + wmb(); if (pdev->irq) { rc = request_irq(pdev->irq, i2o_pci_interrupt, SA_SHIRQ, @@ -306,7 +291,8 @@ static int i2o_pci_irq_enable(struct i2o_controller *c) } } - I2O_IRQ_WRITE32(c, 0x00000000); + writel(0x00000000, c->irq_mask); + wmb(); printk(KERN_INFO "%s: Installed at IRQ %d\n", c->name, pdev->irq); @@ -321,7 +307,9 @@ static int i2o_pci_irq_enable(struct i2o_controller *c) */ static void i2o_pci_irq_disable(struct i2o_controller *c) { - I2O_IRQ_WRITE32(c, 0xffffffff); + wmb(); + writel(0xffffffff, c->irq_mask); + wmb(); if (c->pdev->irq > 0) free_irq(c->pdev->irq, c); @@ -379,7 +367,7 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, pci_name(pdev)); c->pdev = pdev; - c->device = pdev->dev; + c->device.parent = get_device(&pdev->dev); /* Cards that fall apart if you hit them with large I/O loads... */ if (pdev->vendor == PCI_VENDOR_ID_NCR && pdev->device == 0x0630) { @@ -428,6 +416,8 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, if (i960) pci_write_config_word(i960, 0x42, 0x03ff); + get_device(&c->device); + return 0; uninstall: @@ -438,6 +428,7 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, free_controller: i2o_iop_free(c); + put_device(c->device.parent); disable: pci_disable_device(pdev); @@ -461,15 +452,17 @@ static void __devexit i2o_pci_remove(struct pci_dev *pdev) i2o_pci_irq_disable(c); i2o_pci_free(c); + pci_disable_device(pdev); + printk(KERN_INFO "%s: Controller removed.\n", c->name); - i2o_iop_free(c); - pci_disable_device(pdev); + put_device(c->device.parent); + put_device(&c->device); }; /* PCI driver for I2O controller */ static struct pci_driver i2o_pci_driver = { - .name = "I2O controller", + .name = "PCI_I2O", .id_table = i2o_pci_ids, .probe = i2o_pci_probe, .remove = __devexit_p(i2o_pci_remove), diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 40e45a83d3fb..e8cd11290010 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -153,12 +153,10 @@ struct i2o_controller { unsigned int promise:1; /* Promise controller */ struct list_head devices; /* list of I2O devices */ - - struct notifier_block *event_notifer; /* Events */ - atomic_t users; struct list_head list; /* Controller list */ - void __iomem *post_port; /* Inbout port address */ - void __iomem *reply_port; /* Outbound port address */ + + void __iomem *in_port; /* Inbout port address */ + void __iomem *out_port; /* Outbound port address */ void __iomem *irq_mask; /* Interrupt register address */ /* Dynamic LCT related data */ @@ -182,9 +180,6 @@ struct i2o_controller { struct resource io_resource; /* I/O resource allocated to the IOP */ struct resource mem_resource; /* Mem resource allocated to the IOP */ - struct proc_dir_entry *proc_entry; /* /proc dir */ - - struct list_head bus_list; /* list of busses on IOP */ struct device device; struct i2o_device *exec; /* Executive */ #if BITS_PER_LONG == 64 @@ -380,49 +375,10 @@ extern int i2o_device_claim_release(struct i2o_device *); /* Exec OSM functions */ extern int i2o_exec_lct_get(struct i2o_controller *); -/* device to i2o_device and driver to i2o_driver convertion functions */ +/* device / driver conversion functions */ #define to_i2o_driver(drv) container_of(drv,struct i2o_driver, driver) #define to_i2o_device(dev) container_of(dev, struct i2o_device, device) - -/* - * Messenger inlines - */ -static inline u32 I2O_POST_READ32(struct i2o_controller *c) -{ - rmb(); - return readl(c->post_port); -}; - -static inline void I2O_POST_WRITE32(struct i2o_controller *c, u32 val) -{ - wmb(); - writel(val, c->post_port); -}; - -static inline u32 I2O_REPLY_READ32(struct i2o_controller *c) -{ - rmb(); - return readl(c->reply_port); -}; - -static inline void I2O_REPLY_WRITE32(struct i2o_controller *c, u32 val) -{ - wmb(); - writel(val, c->reply_port); -}; - -static inline u32 I2O_IRQ_READ32(struct i2o_controller *c) -{ - rmb(); - return readl(c->irq_mask); -}; - -static inline void I2O_IRQ_WRITE32(struct i2o_controller *c, u32 val) -{ - wmb(); - writel(val, c->irq_mask); - wmb(); -}; +#define to_i2o_controller(dev) container_of(dev, struct i2o_controller, device) /** * i2o_msg_get - obtain an I2O message from the IOP @@ -440,10 +396,12 @@ static inline void I2O_IRQ_WRITE32(struct i2o_controller *c, u32 val) static inline u32 i2o_msg_get(struct i2o_controller *c, struct i2o_message __iomem **msg) { - u32 m; + u32 m = readl(c->in_port); - if ((m = I2O_POST_READ32(c)) != I2O_QUEUE_EMPTY) + if (m != I2O_QUEUE_EMPTY) { *msg = c->in_queue.virt + m; + rmb(); + } return m; }; @@ -457,7 +415,8 @@ static inline u32 i2o_msg_get(struct i2o_controller *c, */ static inline void i2o_msg_post(struct i2o_controller *c, u32 m) { - I2O_POST_WRITE32(c, m); + wmb(); + writel(m, c->in_port); }; /** @@ -486,12 +445,10 @@ static inline int i2o_msg_post_wait(struct i2o_controller *c, u32 m, * The I2O controller must be informed that the reply message is not needed * anymore. If you forget to flush the reply, the message frame can't be * used by the controller anymore and is therefore lost. - * - * FIXME: is there a timeout after which the controller reuse the message? */ static inline void i2o_flush_reply(struct i2o_controller *c, u32 m) { - I2O_REPLY_WRITE32(c, m); + writel(m, c->out_port); }; /** @@ -505,8 +462,9 @@ static inline void i2o_flush_reply(struct i2o_controller *c, u32 m) * work for sender side messages as they are ioremap objects * provided by the I2O controller. */ -static inline struct i2o_message *i2o_msg_out_to_virt(struct i2o_controller *c, - u32 m) +static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct + i2o_controller *c, + u32 m) { BUG_ON(m < c->out_queue.phys || m >= c->out_queue.phys + c->out_queue.len); @@ -917,7 +875,7 @@ extern void i2o_debug_state(struct i2o_controller *c); #define I2OVER15 0x0001 #define I2OVER20 0x0002 -/* Default is 1.5, FIXME: Need support for both 1.5 and 2.0 */ +/* Default is 1.5 */ #define I2OVERSION I2OVER15 #define SGL_OFFSET_0 I2OVERSION -- cgit v1.2.3 From f10378fff658f61307496e0ae00095041725cf07 Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:16 -0700 Subject: [PATCH] I2O: new sysfs attributes and Adaptec specific block device access and 64-bit DMA support Changes: - Added Bus-OSM which could be used by user space programs to reset a channel on the controller - Make ioctl's in Config-OSM obsolete in prefer for sysfs attributes and move those to its own file - Added sysfs attribute for firmware read and write access for I2O controllers - Added special handling of firmware read and write access for Adaptec controllers - Added vendor id and product id as sysfs-attribute to Executive classes - Added automatic notification of LCT change handling to Exec-OSM - Added flushing function to Block-OSM for later barrier implementation - Use PRIVATE messages for Block access on Adaptec controllers, which are faster then BLOCK class access - Cleaned up support for Promise controller - New messages are now detected using the IRQ status register as suggested by the I2O spec - Added i2o_dma_high() and i2o_dma_low() functions - Added facility for SG tablesize calculation when using 32-bit and 64-bit DMA addresses - Added i2o_dma_map_single() and i2o_dma_map_sg() which could build the SG list for 32-bit as well as 64-bit DMA addresses Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/i2o/Kconfig | 18 ++ drivers/message/i2o/Makefile | 3 + drivers/message/i2o/bus-osm.c | 164 +++++++++++ drivers/message/i2o/config-osm.c | 579 +++++++++++++++++++++++++++++++++++++++ drivers/message/i2o/driver.c | 12 +- drivers/message/i2o/exec-osm.c | 74 ++++- drivers/message/i2o/i2o_block.c | 277 ++++++++++++------- drivers/message/i2o/i2o_block.h | 4 +- drivers/message/i2o/i2o_config.c | 156 +---------- drivers/message/i2o/i2o_proc.c | 4 +- drivers/message/i2o/i2o_scsi.c | 30 +- drivers/message/i2o/iop.c | 263 ++++++++---------- drivers/message/i2o/pci.c | 67 ++--- include/linux/i2o-dev.h | 6 +- include/linux/i2o.h | 321 ++++++++++++++++++---- 15 files changed, 1446 insertions(+), 532 deletions(-) create mode 100644 drivers/message/i2o/bus-osm.c create mode 100644 drivers/message/i2o/config-osm.c (limited to 'include/linux') diff --git a/drivers/message/i2o/Kconfig b/drivers/message/i2o/Kconfig index 8d132b0d6b12..ce278e060aca 100644 --- a/drivers/message/i2o/Kconfig +++ b/drivers/message/i2o/Kconfig @@ -35,6 +35,24 @@ config I2O_CONFIG To compile this support as a module, choose M here: the module will be called i2o_config. +config I2O_CONFIG_OLD_IOCTL + bool "Enable ioctls (OBSOLETE)" + depends on I2O_CONFIG + default y + ---help--- + Enables old ioctls. + +config I2O_BUS + tristate "I2O Bus Adapter OSM" + depends on I2O + ---help--- + Include support for the I2O Bus Adapter OSM. The Bus Adapter OSM + provides access to the busses on the I2O controller. The main purpose + is to rescan the bus to find new devices. + + To compile this support as a module, choose M here: the + module will be called i2o_bus. + config I2O_BLOCK tristate "I2O Block OSM" depends on I2O diff --git a/drivers/message/i2o/Makefile b/drivers/message/i2o/Makefile index aabc6cdc3fce..2c2e39aa1efa 100644 --- a/drivers/message/i2o/Makefile +++ b/drivers/message/i2o/Makefile @@ -6,8 +6,11 @@ # i2o_core-y += iop.o driver.o device.o debug.o pci.o exec-osm.o +i2o_bus-y += bus-osm.o +i2o_config-y += config-osm.o obj-$(CONFIG_I2O) += i2o_core.o obj-$(CONFIG_I2O_CONFIG)+= i2o_config.o +obj-$(CONFIG_I2O_BUS) += i2o_bus.o obj-$(CONFIG_I2O_BLOCK) += i2o_block.o obj-$(CONFIG_I2O_SCSI) += i2o_scsi.o obj-$(CONFIG_I2O_PROC) += i2o_proc.o diff --git a/drivers/message/i2o/bus-osm.c b/drivers/message/i2o/bus-osm.c new file mode 100644 index 000000000000..d43c35894ae9 --- /dev/null +++ b/drivers/message/i2o/bus-osm.c @@ -0,0 +1,164 @@ +/* + * Bus Adapter OSM + * + * Copyright (C) 2005 Markus Lidel + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * Fixes/additions: + * Markus Lidel + * initial version. + */ + +#include +#include + +#define OSM_NAME "bus-osm" +#define OSM_VERSION "$Rev$" +#define OSM_DESCRIPTION "I2O Bus Adapter OSM" + +static struct i2o_driver i2o_bus_driver; + +/* Bus OSM class handling definition */ +static struct i2o_class_id i2o_bus_class_id[] = { + {I2O_CLASS_BUS_ADAPTER}, + {I2O_CLASS_END} +}; + +/** + * i2o_bus_scan - Scan the bus for new devices + * @dev: I2O device of the bus, which should be scanned + * + * Scans the bus dev for new / removed devices. After the scan a new LCT + * will be fetched automatically. + * + * Returns 0 on success or negative error code on failure. + */ +static int i2o_bus_scan(struct i2o_device *dev) +{ + struct i2o_message __iomem *msg; + u32 m; + + m = i2o_msg_get_wait(dev->iop, &msg, I2O_TIMEOUT_MESSAGE_GET); + if (m == I2O_QUEUE_EMPTY) + return -ETIMEDOUT; + + writel(FIVE_WORD_MSG_SIZE | SGL_OFFSET_0, &msg->u.head[0]); + writel(I2O_CMD_BUS_SCAN << 24 | HOST_TID << 12 | dev->lct_data.tid, + &msg->u.head[1]); + + return i2o_msg_post_wait(dev->iop, m, 60); +}; + +/** + * i2o_bus_store_scan - Scan the I2O Bus Adapter + * @d: device which should be scanned + * + * Returns count. + */ +static ssize_t i2o_bus_store_scan(struct device *d, const char *buf, + size_t count) +{ + struct i2o_device *i2o_dev = to_i2o_device(d); + int rc; + + if ((rc = i2o_bus_scan(i2o_dev))) + osm_warn("bus scan failed %d\n", rc); + + return count; +} + +/* Bus Adapter OSM device attributes */ +static DEVICE_ATTR(scan, S_IWUSR, NULL, i2o_bus_store_scan); + +/** + * i2o_bus_probe - verify if dev is a I2O Bus Adapter device and install it + * @dev: device to verify if it is a I2O Bus Adapter device + * + * Because we want all Bus Adapters always return 0. + * + * Returns 0. + */ +static int i2o_bus_probe(struct device *dev) +{ + struct i2o_device *i2o_dev = to_i2o_device(get_device(dev)); + + device_create_file(dev, &dev_attr_scan); + + osm_info("device added (TID: %03x)\n", i2o_dev->lct_data.tid); + + return 0; +}; + +/** + * i2o_bus_remove - remove the I2O Bus Adapter device from the system again + * @dev: I2O Bus Adapter device which should be removed + * + * Always returns 0. + */ +static int i2o_bus_remove(struct device *dev) +{ + struct i2o_device *i2o_dev = to_i2o_device(dev); + + device_remove_file(dev, &dev_attr_scan); + + put_device(dev); + + osm_info("device removed (TID: %03x)\n", i2o_dev->lct_data.tid); + + return 0; +}; + +/* Bus Adapter OSM driver struct */ +static struct i2o_driver i2o_bus_driver = { + .name = OSM_NAME, + .classes = i2o_bus_class_id, + .driver = { + .probe = i2o_bus_probe, + .remove = i2o_bus_remove, + }, +}; + +/** + * i2o_bus_init - Bus Adapter OSM initialization function + * + * Only register the Bus Adapter OSM in the I2O core. + * + * Returns 0 on success or negative error code on failure. + */ +static int __init i2o_bus_init(void) +{ + int rc; + + printk(KERN_INFO OSM_DESCRIPTION " v" OSM_VERSION "\n"); + + /* Register Bus Adapter OSM into I2O core */ + rc = i2o_driver_register(&i2o_bus_driver); + if (rc) { + osm_err("Could not register Bus Adapter OSM\n"); + return rc; + } + + return 0; +}; + +/** + * i2o_bus_exit - Bus Adapter OSM exit function + * + * Unregisters Bus Adapter OSM from I2O core. + */ +static void __exit i2o_bus_exit(void) +{ + i2o_driver_unregister(&i2o_bus_driver); +}; + +MODULE_AUTHOR("Markus Lidel "); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION(OSM_DESCRIPTION); +MODULE_VERSION(OSM_VERSION); + +module_init(i2o_bus_init); +module_exit(i2o_bus_exit); diff --git a/drivers/message/i2o/config-osm.c b/drivers/message/i2o/config-osm.c new file mode 100644 index 000000000000..d0267609a949 --- /dev/null +++ b/drivers/message/i2o/config-osm.c @@ -0,0 +1,579 @@ +/* + * Configuration OSM + * + * Copyright (C) 2005 Markus Lidel + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * Fixes/additions: + * Markus Lidel + * initial version. + */ + +#include +#include +#include + +#include + +#define OSM_NAME "config-osm" +#define OSM_VERSION "1.248" +#define OSM_DESCRIPTION "I2O Configuration OSM" + +/* access mode user rw */ +#define S_IWRSR (S_IRUSR | S_IWUSR) + +static struct i2o_driver i2o_config_driver; + +/* Special file operations for sysfs */ +struct fops_attribute { + struct bin_attribute bin; + struct file_operations fops; +}; + +/** + * sysfs_read_dummy + */ +static ssize_t sysfs_read_dummy(struct kobject *kobj, char *buf, loff_t offset, + size_t count) +{ + return 0; +}; + +/** + * sysfs_write_dummy + */ +static ssize_t sysfs_write_dummy(struct kobject *kobj, char *buf, loff_t offset, + size_t count) +{ + return 0; +}; + +/** + * sysfs_create_fops_file - Creates attribute with special file operations + * @kobj: kobject which should contains the attribute + * @attr: attributes which should be used to create file + * + * First creates attribute @attr in kobject @kobj. If it is the first time + * this function is called, merge old fops from sysfs with new one and + * write it back. Afterwords the new fops will be set for the created + * attribute. + * + * Returns 0 on success or negative error code on failure. + */ +static int sysfs_create_fops_file(struct kobject *kobj, + struct fops_attribute *attr) +{ + struct file_operations tmp, *fops; + struct dentry *d; + struct qstr qstr; + int rc; + + fops = &attr->fops; + + if (fops->read) + attr->bin.read = sysfs_read_dummy; + + if (fops->write) + attr->bin.write = sysfs_write_dummy; + + if ((rc = sysfs_create_bin_file(kobj, &attr->bin))) + return rc; + + qstr.name = attr->bin.attr.name; + qstr.len = strlen(qstr.name); + qstr.hash = full_name_hash(qstr.name, qstr.len); + + if ((d = lookup_hash(&qstr, kobj->dentry))) { + if (!fops->owner) { + memcpy(&tmp, d->d_inode->i_fop, sizeof(tmp)); + if (fops->read) + tmp.read = fops->read; + if (fops->write) + tmp.write = fops->write; + memcpy(fops, &tmp, sizeof(tmp)); + } + + d->d_inode->i_fop = fops; + } else + sysfs_remove_bin_file(kobj, &attr->bin); + + return -ENOENT; +}; + +/** + * sysfs_remove_fops_file - Remove attribute with special file operations + * @kobj: kobject which contains the attribute + * @attr: attributes which are used to create file + * + * Only wrapper arround sysfs_remove_bin_file() + * + * Returns 0 on success or negative error code on failure. + */ +static inline int sysfs_remove_fops_file(struct kobject *kobj, + struct fops_attribute *attr) +{ + return sysfs_remove_bin_file(kobj, &attr->bin); +}; + +/** + * i2o_config_read_hrt - Returns the HRT of the controller + * @kob: kernel object handle + * @buf: buffer into which the HRT should be copied + * @off: file offset + * @count: number of bytes to read + * + * Put @count bytes starting at @off into @buf from the HRT of the I2O + * controller corresponding to @kobj. + * + * Returns number of bytes copied into buffer. + */ +static ssize_t i2o_config_read_hrt(struct kobject *kobj, char *buf, + loff_t offset, size_t count) +{ + struct i2o_controller *c = kobj_to_i2o_device(kobj)->iop; + i2o_hrt *hrt = c->hrt.virt; + + u32 size = (hrt->num_entries * hrt->entry_len + 2) * 4; + + if (offset > size) + return 0; + + if (offset + count > size) + count = size - offset; + + memcpy(buf, (u8 *) hrt + offset, count); + + return count; +}; + +/** + * i2o_config_read_lct - Returns the LCT of the controller + * @kob: kernel object handle + * @buf: buffer into which the LCT should be copied + * @off: file offset + * @count: number of bytes to read + * + * Put @count bytes starting at @off into @buf from the LCT of the I2O + * controller corresponding to @kobj. + * + * Returns number of bytes copied into buffer. + */ +static ssize_t i2o_config_read_lct(struct kobject *kobj, char *buf, + loff_t offset, size_t count) +{ + struct i2o_controller *c = kobj_to_i2o_device(kobj)->iop; + u32 size = c->lct->table_size * 4; + + if (offset > size) + return 0; + + if (offset + count > size) + count = size - offset; + + memcpy(buf, (u8 *) c->lct + offset, count); + + return count; +}; + +#define I2O_CONFIG_SW_ATTR(_name,_mode,_type,_swid) \ +static ssize_t i2o_config_##_name##_read(struct file *file, char __user *buf, size_t count, loff_t * offset) { \ + return i2o_config_sw_read(file, buf, count, offset, _type, _swid); \ +};\ +\ +static ssize_t i2o_config_##_name##_write(struct file *file, const char __user *buf, size_t count, loff_t * offset) { \ + return i2o_config_sw_write(file, buf, count, offset, _type, _swid); \ +}; \ +\ +static struct fops_attribute i2o_config_attr_##_name = { \ + .bin = { .attr = { .name = __stringify(_name), .mode = _mode, \ + .owner = THIS_MODULE }, \ + .size = 0, }, \ + .fops = { .write = i2o_config_##_name##_write, \ + .read = i2o_config_##_name##_read} \ +}; + +#ifdef CONFIG_I2O_EXT_ADAPTEC + +/** + * i2o_config_dpt_reagion - Converts type and id to flash region + * @swtype: type of software module reading + * @swid: id of software which should be read + * + * Converts type and id from I2O spec to the matching region for DPT / + * Adaptec controllers. + * + * Returns region which match type and id or -1 on error. + */ +static u32 i2o_config_dpt_region(u8 swtype, u8 swid) +{ + switch (swtype) { + case I2O_SOFTWARE_MODULE_IRTOS: + /* + * content: operation firmware + * region size: + * 0xbc000 for 2554, 3754, 2564, 3757 + * 0x170000 for 2865 + * 0x17c000 for 3966 + */ + if (!swid) + return 0; + + break; + + case I2O_SOFTWARE_MODULE_IOP_PRIVATE: + /* + * content: BIOS and SMOR + * BIOS size: first 0x8000 bytes + * region size: + * 0x40000 for 2554, 3754, 2564, 3757 + * 0x80000 for 2865, 3966 + */ + if (!swid) + return 1; + + break; + + case I2O_SOFTWARE_MODULE_IOP_CONFIG: + switch (swid) { + case 0: + /* + * content: NVRAM defaults + * region size: 0x2000 bytes + */ + return 2; + case 1: + /* + * content: serial number + * region size: 0x2000 bytes + */ + return 3; + } + break; + } + + return -1; +}; + +#endif + +/** + * i2o_config_sw_read - Read a software module from controller + * @file: file pointer + * @buf: buffer into which the data should be copied + * @count: number of bytes to read + * @off: file offset + * @swtype: type of software module reading + * @swid: id of software which should be read + * + * Transfers @count bytes at offset @offset from IOP into buffer using + * type @swtype and id @swid as described in I2O spec. + * + * Returns number of bytes copied into buffer or error code on failure. + */ +static ssize_t i2o_config_sw_read(struct file *file, char __user * buf, + size_t count, loff_t * offset, u8 swtype, + u32 swid) +{ + struct sysfs_dirent *sd = file->f_dentry->d_parent->d_fsdata; + struct kobject *kobj = sd->s_element; + struct i2o_controller *c = kobj_to_i2o_device(kobj)->iop; + u32 m, function = I2O_CMD_SW_UPLOAD; + struct i2o_dma buffer; + struct i2o_message __iomem *msg; + u32 __iomem *mptr; + int rc, status; + + m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET); + if (m == I2O_QUEUE_EMPTY) + return -EBUSY; + + mptr = &msg->body[3]; + + if ((rc = i2o_dma_alloc(&c->pdev->dev, &buffer, count, GFP_KERNEL))) { + i2o_msg_nop(c, m); + return rc; + } +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (c->adaptec) { + mptr = &msg->body[4]; + function = I2O_CMD_PRIVATE; + + writel(TEN_WORD_MSG_SIZE | SGL_OFFSET_8, &msg->u.head[0]); + + writel(I2O_VENDOR_DPT << 16 | I2O_DPT_FLASH_READ, + &msg->body[0]); + writel(i2o_config_dpt_region(swtype, swid), &msg->body[1]); + writel(*offset, &msg->body[2]); + writel(count, &msg->body[3]); + } else +#endif + writel(NINE_WORD_MSG_SIZE | SGL_OFFSET_7, &msg->u.head[0]); + + writel(0xD0000000 | count, mptr++); + writel(buffer.phys, mptr); + + writel(function << 24 | HOST_TID << 12 | ADAPTER_TID, &msg->u.head[1]); + writel(i2o_config_driver.context, &msg->u.head[2]); + writel(0, &msg->u.head[3]); + +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (!c->adaptec) +#endif + { + writel((u32) swtype << 16 | (u32) 1 << 8, &msg->body[0]); + writel(0, &msg->body[1]); + writel(swid, &msg->body[2]); + } + + status = i2o_msg_post_wait_mem(c, m, 60, &buffer); + + if (status == I2O_POST_WAIT_OK) { + if (!(rc = copy_to_user(buf, buffer.virt, count))) { + rc = count; + *offset += count; + } + } else + rc = -EIO; + + if (status != -ETIMEDOUT) + i2o_dma_free(&c->pdev->dev, &buffer); + + return rc; +}; + +/** + * i2o_config_sw_write - Write a software module to controller + * @file: file pointer + * @buf: buffer into which the data should be copied + * @count: number of bytes to read + * @off: file offset + * @swtype: type of software module writing + * @swid: id of software which should be written + * + * Transfers @count bytes at offset @offset from buffer to IOP using + * type @swtype and id @swid as described in I2O spec. + * + * Returns number of bytes copied from buffer or error code on failure. + */ +static ssize_t i2o_config_sw_write(struct file *file, const char __user * buf, + size_t count, loff_t * offset, u8 swtype, + u32 swid) +{ + struct sysfs_dirent *sd = file->f_dentry->d_parent->d_fsdata; + struct kobject *kobj = sd->s_element; + struct i2o_controller *c = kobj_to_i2o_device(kobj)->iop; + u32 m, function = I2O_CMD_SW_DOWNLOAD; + struct i2o_dma buffer; + struct i2o_message __iomem *msg; + u32 __iomem *mptr; + int rc, status; + + m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET); + if (m == I2O_QUEUE_EMPTY) + return -EBUSY; + + mptr = &msg->body[3]; + + if ((rc = i2o_dma_alloc(&c->pdev->dev, &buffer, count, GFP_KERNEL))) + goto nop_msg; + + if ((rc = copy_from_user(buffer.virt, buf, count))) + goto free_buffer; + +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (c->adaptec) { + mptr = &msg->body[4]; + function = I2O_CMD_PRIVATE; + + writel(TEN_WORD_MSG_SIZE | SGL_OFFSET_8, &msg->u.head[0]); + + writel(I2O_VENDOR_DPT << 16 | I2O_DPT_FLASH_WRITE, + &msg->body[0]); + writel(i2o_config_dpt_region(swtype, swid), &msg->body[1]); + writel(*offset, &msg->body[2]); + writel(count, &msg->body[3]); + } else +#endif + writel(NINE_WORD_MSG_SIZE | SGL_OFFSET_7, &msg->u.head[0]); + + writel(0xD4000000 | count, mptr++); + writel(buffer.phys, mptr); + + writel(function << 24 | HOST_TID << 12 | ADAPTER_TID, &msg->u.head[1]); + writel(i2o_config_driver.context, &msg->u.head[2]); + writel(0, &msg->u.head[3]); + +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (!c->adaptec) +#endif + { + writel((u32) swtype << 16 | (u32) 1 << 8, &msg->body[0]); + writel(0, &msg->body[1]); + writel(swid, &msg->body[2]); + } + + status = i2o_msg_post_wait_mem(c, m, 60, &buffer); + + if (status != -ETIMEDOUT) + i2o_dma_free(&c->pdev->dev, &buffer); + + if (status != I2O_POST_WAIT_OK) + return -EIO; + + *offset += count; + + return count; + + free_buffer: + i2o_dma_free(&c->pdev->dev, &buffer); + + nop_msg: + i2o_msg_nop(c, m); + + return rc; +}; + +/* attribute for HRT in sysfs */ +static struct bin_attribute i2o_config_hrt_attr = { + .attr = { + .name = "hrt", + .mode = S_IRUGO, + .owner = THIS_MODULE}, + .size = 0, + .read = i2o_config_read_hrt +}; + +/* attribute for LCT in sysfs */ +static struct bin_attribute i2o_config_lct_attr = { + .attr = { + .name = "lct", + .mode = S_IRUGO, + .owner = THIS_MODULE}, + .size = 0, + .read = i2o_config_read_lct +}; + +/* IRTOS firmware access */ +I2O_CONFIG_SW_ATTR(irtos, S_IWRSR, I2O_SOFTWARE_MODULE_IRTOS, 0); + +#ifdef CONFIG_I2O_EXT_ADAPTEC + +/* + * attribute for BIOS / SMOR, nvram and serial number access on DPT / Adaptec + * controllers + */ +I2O_CONFIG_SW_ATTR(bios, S_IWRSR, I2O_SOFTWARE_MODULE_IOP_PRIVATE, 0); +I2O_CONFIG_SW_ATTR(nvram, S_IWRSR, I2O_SOFTWARE_MODULE_IOP_CONFIG, 0); +I2O_CONFIG_SW_ATTR(serial, S_IWRSR, I2O_SOFTWARE_MODULE_IOP_CONFIG, 1); + +#endif + +/** + * i2o_config_notify_controller_add - Notify of added controller + * @c: the controller which was added + * + * If a I2O controller is added, we catch the notification to add sysfs + * entries. + */ +static void i2o_config_notify_controller_add(struct i2o_controller *c) +{ + struct kobject *kobj = &c->exec->device.kobj; + + sysfs_create_bin_file(kobj, &i2o_config_hrt_attr); + sysfs_create_bin_file(kobj, &i2o_config_lct_attr); + + sysfs_create_fops_file(kobj, &i2o_config_attr_irtos); +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (c->adaptec) { + sysfs_create_fops_file(kobj, &i2o_config_attr_bios); + sysfs_create_fops_file(kobj, &i2o_config_attr_nvram); + sysfs_create_fops_file(kobj, &i2o_config_attr_serial); + } +#endif +}; + +/** + * i2o_config_notify_controller_remove - Notify of removed controller + * @c: the controller which was removed + * + * If a I2O controller is removed, we catch the notification to remove the + * sysfs entries. + */ +static void i2o_config_notify_controller_remove(struct i2o_controller *c) +{ + struct kobject *kobj = &c->exec->device.kobj; + +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (c->adaptec) { + sysfs_remove_fops_file(kobj, &i2o_config_attr_serial); + sysfs_remove_fops_file(kobj, &i2o_config_attr_nvram); + sysfs_remove_fops_file(kobj, &i2o_config_attr_bios); + } +#endif + sysfs_remove_fops_file(kobj, &i2o_config_attr_irtos); + + sysfs_remove_bin_file(kobj, &i2o_config_lct_attr); + sysfs_remove_bin_file(kobj, &i2o_config_hrt_attr); +}; + +/* Config OSM driver struct */ +static struct i2o_driver i2o_config_driver = { + .name = OSM_NAME, + .notify_controller_add = i2o_config_notify_controller_add, + .notify_controller_remove = i2o_config_notify_controller_remove +}; + +#ifdef CONFIG_I2O_CONFIG_OLD_IOCTL +#include "i2o_config.c" +#endif + +/** + * i2o_config_init - Configuration OSM initialization function + * + * Registers Configuration OSM in the I2O core and if old ioctl's are + * compiled in initialize them. + * + * Returns 0 on success or negative error code on failure. + */ +static int __init i2o_config_init(void) +{ + printk(KERN_INFO OSM_DESCRIPTION " v" OSM_VERSION "\n"); + + if (i2o_driver_register(&i2o_config_driver)) { + osm_err("handler register failed.\n"); + return -EBUSY; + } +#ifdef CONFIG_I2O_CONFIG_OLD_IOCTL + if (i2o_config_old_init()) + i2o_driver_unregister(&i2o_config_driver); +#endif + + return 0; +} + +/** + * i2o_config_exit - Configuration OSM exit function + * + * If old ioctl's are compiled in exit remove them and unregisters + * Configuration OSM from I2O core. + */ +static void i2o_config_exit(void) +{ +#ifdef CONFIG_I2O_CONFIG_OLD_IOCTL + i2o_config_old_exit(); +#endif + + i2o_driver_unregister(&i2o_config_driver); +} + +MODULE_AUTHOR("Markus Lidel "); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION(OSM_DESCRIPTION); +MODULE_VERSION(OSM_VERSION); + +module_init(i2o_config_init); +module_exit(i2o_config_exit); diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c index bebdd509b5d8..393be8e2914c 100644 --- a/drivers/message/i2o/driver.c +++ b/drivers/message/i2o/driver.c @@ -180,7 +180,13 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m) { struct i2o_driver *drv; struct i2o_message __iomem *msg = i2o_msg_out_to_virt(c, m); - u32 context = readl(&msg->u.s.icntxt); + u32 context; + unsigned long flags; + + if(unlikely(!msg)) + return -EIO; + + context = readl(&msg->u.s.icntxt); if (unlikely(context >= i2o_max_drivers)) { osm_warn("%s: Spurious reply to unknown driver %d\n", c->name, @@ -188,9 +194,9 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m) return -EIO; } - spin_lock(&i2o_drivers_lock); + spin_lock_irqsave(&i2o_drivers_lock, flags); drv = i2o_drivers[context]; - spin_unlock(&i2o_drivers_lock); + spin_unlock_irqrestore(&i2o_drivers_lock, flags); if (unlikely(!drv)) { osm_warn("%s: Spurious reply to unknown driver %d\n", c->name, diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index 5581344fbba6..0160221c802a 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -206,6 +206,7 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, u32 context) { struct i2o_exec_wait *wait, *tmp; + unsigned long flags; static spinlock_t lock = SPIN_LOCK_UNLOCKED; int rc = 1; @@ -216,11 +217,13 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, * already expired. Not much we can do about that except log it for * debug purposes, increase timeout, and recompile. */ - spin_lock(&lock); + spin_lock_irqsave(&lock, flags); list_for_each_entry_safe(wait, tmp, &i2o_exec_wait_list, list) { if (wait->tcntxt == context) { list_del(&wait->list); + spin_unlock_irqrestore(&lock, flags); + wait->m = m; wait->msg = msg; wait->complete = 1; @@ -242,13 +245,11 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, rc = -1; } - spin_unlock(&lock); - return rc; } } - spin_unlock(&lock); + spin_unlock_irqrestore(&lock, flags); osm_warn("%s: Bogus reply in POST WAIT (tr-context: %08x)!\n", c->name, context); @@ -256,6 +257,50 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, return -1; }; +/** + * i2o_exec_show_vendor_id - Displays Vendor ID of controller + * @d: device of which the Vendor ID should be displayed + * @buf: buffer into which the Vendor ID should be printed + * + * Returns number of bytes printed into buffer. + */ +static ssize_t i2o_exec_show_vendor_id(struct device *d, char *buf) +{ + struct i2o_device *dev = to_i2o_device(d); + u16 id; + + if (i2o_parm_field_get(dev, 0x0000, 0, &id, 2)) { + sprintf(buf, "0x%04x", id); + return strlen(buf) + 1; + } + + return 0; +}; + +/** + * i2o_exec_show_product_id - Displays Product ID of controller + * @d: device of which the Product ID should be displayed + * @buf: buffer into which the Product ID should be printed + * + * Returns number of bytes printed into buffer. + */ +static ssize_t i2o_exec_show_product_id(struct device *d, char *buf) +{ + struct i2o_device *dev = to_i2o_device(d); + u16 id; + + if (i2o_parm_field_get(dev, 0x0000, 1, &id, 2)) { + sprintf(buf, "0x%04x", id); + return strlen(buf) + 1; + } + + return 0; +}; + +/* Exec-OSM device attributes */ +static DEVICE_ATTR(vendor_id, S_IRUGO, i2o_exec_show_vendor_id, NULL); +static DEVICE_ATTR(product_id, S_IRUGO, i2o_exec_show_product_id, NULL); + /** * i2o_exec_probe - Called if a new I2O device (executive class) appears * @dev: I2O device which should be probed @@ -268,10 +313,16 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, static int i2o_exec_probe(struct device *dev) { struct i2o_device *i2o_dev = to_i2o_device(dev); + struct i2o_controller *c = i2o_dev->iop; i2o_event_register(i2o_dev, &i2o_exec_driver, 0, 0xffffffff); - i2o_dev->iop->exec = i2o_dev; + c->exec = i2o_dev; + + i2o_exec_lct_notify(c, c->lct->change_ind + 1); + + device_create_file(dev, &dev_attr_vendor_id); + device_create_file(dev, &dev_attr_product_id); return 0; }; @@ -286,6 +337,9 @@ static int i2o_exec_probe(struct device *dev) */ static int i2o_exec_remove(struct device *dev) { + device_remove_file(dev, &dev_attr_product_id); + device_remove_file(dev, &dev_attr_vendor_id); + i2o_event_register(to_i2o_device(dev), &i2o_exec_driver, 0, 0); return 0; @@ -297,12 +351,16 @@ static int i2o_exec_remove(struct device *dev) * * This function handles asynchronus LCT NOTIFY replies. It parses the * new LCT and if the buffer for the LCT was to small sends a LCT NOTIFY - * again. + * again, otherwise send LCT NOTIFY to get informed on next LCT change. */ static void i2o_exec_lct_modified(struct i2o_controller *c) { - if (i2o_device_parse_lct(c) == -EAGAIN) - i2o_exec_lct_notify(c, 0); + u32 change_ind = 0; + + if (i2o_device_parse_lct(c) != -EAGAIN) + change_ind = c->lct->change_ind + 1; + + i2o_exec_lct_notify(c, change_ind); }; /** diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c index e69421e36ac5..1dd2b9dad50e 100644 --- a/drivers/message/i2o/i2o_block.c +++ b/drivers/message/i2o/i2o_block.c @@ -146,6 +146,29 @@ static int i2o_block_device_flush(struct i2o_device *dev) return i2o_msg_post_wait(dev->iop, m, 60); }; +/** + * i2o_block_issue_flush - device-flush interface for block-layer + * @queue: the request queue of the device which should be flushed + * @disk: gendisk + * @error_sector: error offset + * + * Helper function to provide flush functionality to block-layer. + * + * Returns 0 on success or negative error code on failure. + */ + +static int i2o_block_issue_flush(request_queue_t * queue, struct gendisk *disk, + sector_t * error_sector) +{ + struct i2o_block_device *i2o_blk_dev = queue->queuedata; + int rc = -ENODEV; + + if (likely(i2o_blk_dev)) + rc = i2o_block_device_flush(i2o_blk_dev->i2o_dev); + + return rc; +} + /** * i2o_block_device_mount - Mount (load) the media of device dev * @dev: I2O device which should receive the mount request @@ -299,28 +322,31 @@ static inline void i2o_block_request_free(struct i2o_block_request *ireq) /** * i2o_block_sglist_alloc - Allocate the SG list and map it + * @c: I2O controller to which the request belongs * @ireq: I2O block request * - * Builds the SG list and map it into to be accessable by the controller. + * Builds the SG list and map it to be accessable by the controller. * - * Returns the number of elements in the SG list or 0 on failure. + * Returns 0 on failure or 1 on success. */ -static inline int i2o_block_sglist_alloc(struct i2o_block_request *ireq) +static inline int i2o_block_sglist_alloc(struct i2o_controller *c, + struct i2o_block_request *ireq, + u32 __iomem ** mptr) { - struct device *dev = &ireq->i2o_blk_dev->i2o_dev->iop->pdev->dev; int nents; + enum dma_data_direction direction; + ireq->dev = &c->pdev->dev; nents = blk_rq_map_sg(ireq->req->q, ireq->req, ireq->sg_table); if (rq_data_dir(ireq->req) == READ) - ireq->sg_dma_direction = PCI_DMA_FROMDEVICE; + direction = PCI_DMA_FROMDEVICE; else - ireq->sg_dma_direction = PCI_DMA_TODEVICE; + direction = PCI_DMA_TODEVICE; - ireq->sg_nents = dma_map_sg(dev, ireq->sg_table, nents, - ireq->sg_dma_direction); + ireq->sg_nents = nents; - return ireq->sg_nents; + return i2o_dma_map_sg(c, ireq->sg_table, nents, direction, mptr); }; /** @@ -331,10 +357,14 @@ static inline int i2o_block_sglist_alloc(struct i2o_block_request *ireq) */ static inline void i2o_block_sglist_free(struct i2o_block_request *ireq) { - struct device *dev = &ireq->i2o_blk_dev->i2o_dev->iop->pdev->dev; + enum dma_data_direction direction; - dma_unmap_sg(dev, ireq->sg_table, ireq->sg_nents, - ireq->sg_dma_direction); + if (rq_data_dir(ireq->req) == READ) + direction = PCI_DMA_FROMDEVICE; + else + direction = PCI_DMA_TODEVICE; + + dma_unmap_sg(ireq->dev, ireq->sg_table, ireq->sg_nents, direction); }; /** @@ -352,6 +382,11 @@ static int i2o_block_prep_req_fn(struct request_queue *q, struct request *req) struct i2o_block_device *i2o_blk_dev = q->queuedata; struct i2o_block_request *ireq; + if (unlikely(!i2o_blk_dev)) { + osm_err("block device already removed\n"); + return BLKPREP_KILL; + } + /* request is already processed by us, so return */ if (req->flags & REQ_SPECIAL) { osm_debug("REQ_SPECIAL already set!\n"); @@ -414,11 +449,11 @@ static void i2o_block_end_request(struct request *req, int uptodate, { struct i2o_block_request *ireq = req->special; struct i2o_block_device *dev = ireq->i2o_blk_dev; - request_queue_t *q = dev->gd->queue; + request_queue_t *q = req->q; unsigned long flags; if (end_that_request_chunk(req, uptodate, nr_bytes)) { - int leftover = (req->hard_nr_sectors << 9); + int leftover = (req->hard_nr_sectors << KERNEL_SECTOR_SHIFT); if (blk_pc_request(req)) leftover = req->data_len; @@ -432,8 +467,11 @@ static void i2o_block_end_request(struct request *req, int uptodate, spin_lock_irqsave(q->queue_lock, flags); end_that_request_last(req); - dev->open_queue_depth--; - list_del(&ireq->queue); + + if (likely(dev)) { + dev->open_queue_depth--; + list_del(&ireq->queue); + } blk_start_queue(q); @@ -483,8 +521,8 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m, * Don't stick a supertrak100 into cache aggressive modes */ - osm_err("%03x error status: %02x, detailed status: %04x\n", - (le32_to_cpu(msg->u.head[1]) >> 12 & 0xfff), + osm_err("TID %03x error status: 0x%02x, detailed status: " + "0x%04x\n", (le32_to_cpu(msg->u.head[1]) >> 12 & 0xfff), status >> 24, status & 0xffff); req->errors++; @@ -705,18 +743,25 @@ static int i2o_block_media_changed(struct gendisk *disk) static int i2o_block_transfer(struct request *req) { struct i2o_block_device *dev = req->rq_disk->private_data; - struct i2o_controller *c = dev->i2o_dev->iop; + struct i2o_controller *c; int tid = dev->i2o_dev->lct_data.tid; struct i2o_message __iomem *msg; - void __iomem *mptr; + u32 __iomem *mptr; struct i2o_block_request *ireq = req->special; - struct scatterlist *sg; - int sgnum; - int i; u32 m; u32 tcntxt; - u32 sg_flags; + u32 sgl_offset = SGL_OFFSET_8; + u32 ctl_flags = 0x00000000; int rc; + u32 cmd; + + if (unlikely(!dev->i2o_dev)) { + osm_err("transfer to removed drive\n"); + rc = -ENODEV; + goto exit; + } + + c = dev->i2o_dev->iop; m = i2o_msg_get(c, &msg); if (m == I2O_QUEUE_EMPTY) { @@ -730,80 +775,109 @@ static int i2o_block_transfer(struct request *req) goto nop_msg; } - if ((sgnum = i2o_block_sglist_alloc(ireq)) <= 0) { - rc = -ENOMEM; - goto context_remove; - } - - /* Build the message based on the request. */ writel(i2o_block_driver.context, &msg->u.s.icntxt); writel(tcntxt, &msg->u.s.tcntxt); - writel(req->nr_sectors << 9, &msg->body[1]); - writel((((u64) req->sector) << 9) & 0xffffffff, &msg->body[2]); - writel(req->sector >> 23, &msg->body[3]); - - mptr = &msg->body[4]; - - sg = ireq->sg_table; + mptr = &msg->body[0]; if (rq_data_dir(req) == READ) { - writel(I2O_CMD_BLOCK_READ << 24 | HOST_TID << 12 | tid, - &msg->u.head[1]); - sg_flags = 0x10000000; + cmd = I2O_CMD_BLOCK_READ << 24; + switch (dev->rcache) { - case CACHE_NULL: - writel(0, &msg->body[0]); - break; case CACHE_PREFETCH: - writel(0x201F0008, &msg->body[0]); + ctl_flags = 0x201F0008; break; + case CACHE_SMARTFETCH: if (req->nr_sectors > 16) - writel(0x201F0008, &msg->body[0]); + ctl_flags = 0x201F0008; else - writel(0x001F0000, &msg->body[0]); + ctl_flags = 0x001F0000; + break; + + default: break; } } else { - writel(I2O_CMD_BLOCK_WRITE << 24 | HOST_TID << 12 | tid, - &msg->u.head[1]); - sg_flags = 0x14000000; + cmd = I2O_CMD_BLOCK_WRITE << 24; + switch (dev->wcache) { - case CACHE_NULL: - writel(0, &msg->body[0]); - break; case CACHE_WRITETHROUGH: - writel(0x001F0008, &msg->body[0]); + ctl_flags = 0x001F0008; break; case CACHE_WRITEBACK: - writel(0x001F0010, &msg->body[0]); + ctl_flags = 0x001F0010; break; case CACHE_SMARTBACK: if (req->nr_sectors > 16) - writel(0x001F0004, &msg->body[0]); + ctl_flags = 0x001F0004; else - writel(0x001F0010, &msg->body[0]); + ctl_flags = 0x001F0010; break; case CACHE_SMARTTHROUGH: if (req->nr_sectors > 16) - writel(0x001F0004, &msg->body[0]); + ctl_flags = 0x001F0004; else - writel(0x001F0010, &msg->body[0]); + ctl_flags = 0x001F0010; + default: + break; + } + } + +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (c->adaptec) { + u8 cmd[10]; + u32 scsi_flags; + u16 hwsec = queue_hardsect_size(req->q) >> KERNEL_SECTOR_SHIFT; + + memset(cmd, 0, 10); + + sgl_offset = SGL_OFFSET_12; + + writel(I2O_CMD_PRIVATE << 24 | HOST_TID << 12 | tid, + &msg->u.head[1]); + + writel(I2O_VENDOR_DPT << 16 | I2O_CMD_SCSI_EXEC, mptr++); + writel(tid, mptr++); + + /* + * ENABLE_DISCONNECT + * SIMPLE_TAG + * RETURN_SENSE_DATA_IN_REPLY_MESSAGE_FRAME + */ + if (rq_data_dir(req) == READ) { + cmd[0] = 0x28; + scsi_flags = 0x60a0000a; + } else { + cmd[0] = 0x2A; + scsi_flags = 0xa0a0000a; } + + writel(scsi_flags, mptr++); + + *((u32 *) & cmd[2]) = cpu_to_be32(req->sector * hwsec); + *((u16 *) & cmd[7]) = cpu_to_be16(req->nr_sectors * hwsec); + + memcpy_toio(mptr, cmd, 10); + mptr += 4; + writel(req->nr_sectors << KERNEL_SECTOR_SHIFT, mptr++); + } else +#endif + { + writel(cmd | HOST_TID << 12 | tid, &msg->u.head[1]); + writel(ctl_flags, mptr++); + writel(req->nr_sectors << KERNEL_SECTOR_SHIFT, mptr++); + writel((u32) (req->sector << KERNEL_SECTOR_SHIFT), mptr++); + writel(req->sector >> (32 - KERNEL_SECTOR_SHIFT), mptr++); } - for (i = sgnum; i > 0; i--) { - if (i == 1) - sg_flags |= 0x80000000; - writel(sg_flags | sg_dma_len(sg), mptr); - writel(sg_dma_address(sg), mptr + 4); - mptr += 8; - sg++; + if (!i2o_block_sglist_alloc(c, ireq, &mptr)) { + rc = -ENOMEM; + goto context_remove; } - writel(I2O_MESSAGE_SIZE(mptr - &msg->u.head[0]) | SGL_OFFSET_8, - &msg->u.head[0]); + writel(I2O_MESSAGE_SIZE(mptr - &msg->u.head[0]) | + sgl_offset, &msg->u.head[0]); list_add_tail(&ireq->queue, &dev->open_queue); dev->open_queue_depth++; @@ -846,11 +920,13 @@ static void i2o_block_request_fn(struct request_queue *q) queue_depth = ireq->i2o_blk_dev->open_queue_depth; - if (queue_depth < I2O_BLOCK_MAX_OPEN_REQUESTS) + if (queue_depth < I2O_BLOCK_MAX_OPEN_REQUESTS) { if (!i2o_block_transfer(req)) { blkdev_dequeue_request(req); continue; - } + } else + osm_info("transfer error\n"); + } if (queue_depth) break; @@ -933,6 +1009,7 @@ static struct i2o_block_device *i2o_block_device_alloc(void) } blk_queue_prep_rq(queue, i2o_block_prep_req_fn); + blk_queue_issue_flush_fn(queue, i2o_block_issue_flush); gd->major = I2O_MAJOR; gd->queue = queue; @@ -974,7 +1051,18 @@ static int i2o_block_probe(struct device *dev) u64 size; u32 blocksize; u32 flags, status; - int segments; + u16 body_size = 4; + unsigned short max_sectors; + +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (c->adaptec) + body_size = 8; +#endif + + if (c->limit_sectors) + max_sectors = I2O_MAX_SECTORS_LIMITED; + else + max_sectors = I2O_MAX_SECTORS; /* skip devices which are used by IOP */ if (i2o_dev->lct_data.user_tid != 0xfff) { @@ -1009,50 +1097,35 @@ static int i2o_block_probe(struct device *dev) queue = gd->queue; queue->queuedata = i2o_blk_dev; - blk_queue_max_phys_segments(queue, I2O_MAX_SEGMENTS); - blk_queue_max_sectors(queue, I2O_MAX_SECTORS); - - if (c->short_req) - segments = 8; - else { - i2o_status_block *sb; + blk_queue_max_phys_segments(queue, I2O_MAX_PHYS_SEGMENTS); + blk_queue_max_sectors(queue, max_sectors); + blk_queue_max_hw_segments(queue, i2o_sg_tablesize(c, body_size)); - sb = c->status_block.virt; - - segments = (sb->inbound_frame_size - - sizeof(struct i2o_message) / 4 - 4) / 2; - } - - blk_queue_max_hw_segments(queue, segments); - - osm_debug("max sectors = %d\n", I2O_MAX_SECTORS); - osm_debug("phys segments = %d\n", I2O_MAX_SEGMENTS); - osm_debug("hw segments = %d\n", segments); + osm_debug("max sectors = %d\n", queue->max_phys_segments); + osm_debug("phys segments = %d\n", queue->max_sectors); + osm_debug("max hw segments = %d\n", queue->max_hw_segments); /* * Ask for the current media data. If that isn't supported * then we ask for the device capacity data */ - if (!i2o_parm_field_get(i2o_dev, 0x0004, 0, &size, 8)) - if (!i2o_parm_field_get(i2o_dev, 0x0000, 4, &size, 8)) { - osm_warn("could not get size of %s\n", gd->disk_name); - size = 0; - } + if (i2o_parm_field_get(i2o_dev, 0x0004, 1, &blocksize, 4) || + i2o_parm_field_get(i2o_dev, 0x0000, 3, &blocksize, 4)) { + blk_queue_hardsect_size(queue, blocksize); + } else + osm_warn("unable to get blocksize of %s\n", gd->disk_name); - if (!i2o_parm_field_get(i2o_dev, 0x0004, 1, &blocksize, 4)) - if (!i2o_parm_field_get(i2o_dev, 0x0000, 3, &blocksize, 4)) { - osm_warn("unable to get blocksize of %s\n", - gd->disk_name); - blocksize = 0; - } + if (i2o_parm_field_get(i2o_dev, 0x0004, 0, &size, 8) || + i2o_parm_field_get(i2o_dev, 0x0000, 4, &size, 8)) { + set_capacity(gd, size >> KERNEL_SECTOR_SHIFT); + } else + osm_warn("could not get size of %s\n", gd->disk_name); if (!i2o_parm_field_get(i2o_dev, 0x0000, 2, &i2o_blk_dev->power, 2)) i2o_blk_dev->power = 0; i2o_parm_field_get(i2o_dev, 0x0000, 5, &flags, 4); i2o_parm_field_get(i2o_dev, 0x0000, 6, &status, 4); - set_capacity(gd, size >> 9); - i2o_event_register(i2o_dev, &i2o_block_driver, 0, 0xffffffff); add_disk(gd); @@ -1109,7 +1182,7 @@ static int __init i2o_block_init(void) goto exit; } - i2o_blk_req_pool.pool = mempool_create(I2O_REQ_MEMPOOL_SIZE, + i2o_blk_req_pool.pool = mempool_create(I2O_BLOCK_REQ_MEMPOOL_SIZE, mempool_alloc_slab, mempool_free_slab, i2o_blk_req_pool.slab); diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h index 712111ffa638..9e1a95fb0833 100644 --- a/drivers/message/i2o/i2o_block.h +++ b/drivers/message/i2o/i2o_block.h @@ -84,9 +84,9 @@ struct i2o_block_request struct list_head queue; struct request *req; /* corresponding request */ struct i2o_block_device *i2o_blk_dev; /* I2O block device */ - int sg_dma_direction; /* direction of DMA buffer read/write */ + struct device *dev; /* device used for DMA */ int sg_nents; /* number of SG elements */ - struct scatterlist sg_table[I2O_MAX_SEGMENTS]; /* SG table */ + struct scatterlist sg_table[I2O_MAX_PHYS_SEGMENTS]; /* SG table */ }; /* I2O Block device delayed request */ diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c index 383e89a5c9f0..849d90aad779 100644 --- a/drivers/message/i2o/i2o_config.c +++ b/drivers/message/i2o/i2o_config.c @@ -30,27 +30,11 @@ * 2 of the License, or (at your option) any later version. */ -#include -#include -#include -#include -#include -#include -#include #include -#include -#include #include -#include #include -#include #include -#include - -#define OSM_NAME "config-osm" -#define OSM_VERSION "$Rev$" -#define OSM_DESCRIPTION "I2O Configuration OSM" extern int i2o_parm_issue(struct i2o_device *, int, void *, int, void *, int); @@ -80,125 +64,6 @@ struct i2o_cfg_info { static struct i2o_cfg_info *open_files = NULL; static ulong i2o_cfg_info_id = 0; -/** - * i2o_config_read_hrt - Returns the HRT of the controller - * @kob: kernel object handle - * @buf: buffer into which the HRT should be copied - * @off: file offset - * @count: number of bytes to read - * - * Put @count bytes starting at @off into @buf from the HRT of the I2O - * controller corresponding to @kobj. - * - * Returns number of bytes copied into buffer. - */ -static ssize_t i2o_config_read_hrt(struct kobject *kobj, char *buf, - loff_t offset, size_t count) -{ - struct i2o_controller *c = to_i2o_controller(container_of(kobj, - struct device, - kobj)); - i2o_hrt *hrt = c->hrt.virt; - - u32 size = (hrt->num_entries * hrt->entry_len + 2) * 4; - - if(offset > size) - return 0; - - if(offset + count > size) - count = size - offset; - - memcpy(buf, (u8 *) hrt + offset, count); - - return count; -}; - -/** - * i2o_config_read_lct - Returns the LCT of the controller - * @kob: kernel object handle - * @buf: buffer into which the LCT should be copied - * @off: file offset - * @count: number of bytes to read - * - * Put @count bytes starting at @off into @buf from the LCT of the I2O - * controller corresponding to @kobj. - * - * Returns number of bytes copied into buffer. - */ -static ssize_t i2o_config_read_lct(struct kobject *kobj, char *buf, - loff_t offset, size_t count) -{ - struct i2o_controller *c = to_i2o_controller(container_of(kobj, - struct device, - kobj)); - u32 size = c->lct->table_size * 4; - - if(offset > size) - return 0; - - if(offset + count > size) - count = size - offset; - - memcpy(buf, (u8 *) c->lct + offset, count); - - return count; -}; - -/* attribute for HRT in sysfs */ -static struct bin_attribute i2o_config_hrt_attr = { - .attr = { - .name = "hrt", - .mode = S_IRUGO, - .owner = THIS_MODULE - }, - .size = 0, - .read = i2o_config_read_hrt -}; - -/* attribute for LCT in sysfs */ -static struct bin_attribute i2o_config_lct_attr = { - .attr = { - .name = "lct", - .mode = S_IRUGO, - .owner = THIS_MODULE - }, - .size = 0, - .read = i2o_config_read_lct -}; - -/** - * i2o_config_notify_controller_add - Notify of added controller - * @c: the controller which was added - * - * If a I2O controller is added, we catch the notification to add sysfs - * entries. - */ -static void i2o_config_notify_controller_add(struct i2o_controller *c) -{ - sysfs_create_bin_file(&(c->device.kobj), &i2o_config_hrt_attr); - sysfs_create_bin_file(&(c->device.kobj), &i2o_config_lct_attr); -}; - -/** - * i2o_config_notify_controller_remove - Notify of removed controller - * @c: the controller which was removed - * - * If a I2O controller is removed, we catch the notification to remove the - * sysfs entries. - */ -static void i2o_config_notify_controller_remove(struct i2o_controller *c) -{ - sysfs_remove_bin_file(&c->device.kobj, &i2o_config_lct_attr); - sysfs_remove_bin_file(&c->device.kobj, &i2o_config_hrt_attr); -}; - -/* Config OSM driver struct */ -static struct i2o_driver i2o_config_driver = { - .name = OSM_NAME, - .notify_controller_add = i2o_config_notify_controller_add, - .notify_controller_remove = i2o_config_notify_controller_remove -}; - static int i2o_cfg_getiops(unsigned long arg) { struct i2o_controller *c; @@ -1257,37 +1122,20 @@ static struct miscdevice i2o_miscdev = { &config_fops }; -static int __init i2o_config_init(void) +static int __init i2o_config_old_init(void) { - printk(KERN_INFO OSM_DESCRIPTION " v" OSM_VERSION "\n"); - spin_lock_init(&i2o_config_lock); if (misc_register(&i2o_miscdev) < 0) { osm_err("can't register device.\n"); return -EBUSY; } - /* - * Install our handler - */ - if (i2o_driver_register(&i2o_config_driver)) { - osm_err("handler register failed.\n"); - misc_deregister(&i2o_miscdev); - return -EBUSY; - } return 0; } -static void i2o_config_exit(void) +static void i2o_config_old_exit(void) { misc_deregister(&i2o_miscdev); - i2o_driver_unregister(&i2o_config_driver); } MODULE_AUTHOR("Red Hat Software"); -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION(OSM_DESCRIPTION); -MODULE_VERSION(OSM_VERSION); - -module_init(i2o_config_init); -module_exit(i2o_config_exit); diff --git a/drivers/message/i2o/i2o_proc.c b/drivers/message/i2o/i2o_proc.c index b176d0eeff7f..e5b74452c495 100644 --- a/drivers/message/i2o/i2o_proc.c +++ b/drivers/message/i2o/i2o_proc.c @@ -228,7 +228,7 @@ static const char *i2o_get_class_name(int class) case I2O_CLASS_FLOPPY_DEVICE: idx = 12; break; - case I2O_CLASS_BUS_ADAPTER_PORT: + case I2O_CLASS_BUS_ADAPTER: idx = 13; break; case I2O_CLASS_PEER_TRANSPORT_AGENT: @@ -490,7 +490,7 @@ static int i2o_seq_show_lct(struct seq_file *seq, void *v) seq_printf(seq, ", Unknown Device Type"); break; - case I2O_CLASS_BUS_ADAPTER_PORT: + case I2O_CLASS_BUS_ADAPTER: if (lct->lct_entry[i].sub_class < BUS_TABLE_SIZE) seq_printf(seq, ", %s", bus_ports[lct->lct_entry[i]. diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c index 812c29ec86d3..c3b0c29ac02d 100644 --- a/drivers/message/i2o/i2o_scsi.c +++ b/drivers/message/i2o/i2o_scsi.c @@ -103,7 +103,7 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c) i2o_status_block *sb; list_for_each_entry(i2o_dev, &c->devices, list) - if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER_PORT) { + if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER) { if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) && (type == 0x01)) /* SCSI bus */ max_channel++; @@ -139,7 +139,7 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c) i = 0; list_for_each_entry(i2o_dev, &c->devices, list) - if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER_PORT) { + if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER) { if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) || (type == 1)) /* only SCSI bus */ i2o_shost->channel[i++] = i2o_dev; @@ -186,6 +186,7 @@ static int i2o_scsi_remove(struct device *dev) shost_for_each_device(scsi_dev, i2o_shost->scsi_host) if (scsi_dev->hostdata == i2o_dev) { + sysfs_remove_link(&i2o_dev->device.kobj, "scsi"); scsi_remove_device(scsi_dev); scsi_device_put(scsi_dev); break; @@ -259,12 +260,14 @@ static int i2o_scsi_probe(struct device *dev) scsi_dev = __scsi_add_device(i2o_shost->scsi_host, channel, id, lun, i2o_dev); - if (!scsi_dev) { + if (IS_ERR(scsi_dev)) { osm_warn("can not add SCSI device %03x\n", i2o_dev->lct_data.tid); - return -EFAULT; + return PTR_ERR(scsi_dev); } + sysfs_create_link(&i2o_dev->device.kobj, &scsi_dev->sdev_gendev.kobj, "scsi"); + osm_info("device added (TID: %03x) channel: %d, id: %d, lun: %d\n", i2o_dev->lct_data.tid, channel, id, (unsigned int)lun); @@ -545,7 +548,13 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, int tid; struct i2o_message __iomem *msg; u32 m; - u32 scsi_flags, sg_flags; + /* + * ENABLE_DISCONNECT + * SIMPLE_TAG + * RETURN_SENSE_DATA_IN_REPLY_MESSAGE_FRAME + */ + u32 scsi_flags = 0x20a00000; + u32 sg_flags; u32 __iomem *mptr; u32 __iomem *lenptr; u32 len; @@ -591,17 +600,19 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, switch (SCpnt->sc_data_direction) { case PCI_DMA_NONE: - scsi_flags = 0x00000000; // DATA NO XFER + /* DATA NO XFER */ sg_flags = 0x00000000; break; case PCI_DMA_TODEVICE: - scsi_flags = 0x80000000; // DATA OUT (iop-->dev) + /* DATA OUT (iop-->dev) */ + scsi_flags |= 0x80000000; sg_flags = 0x14000000; break; case PCI_DMA_FROMDEVICE: - scsi_flags = 0x40000000; // DATA IN (iop<--dev) + /* DATA IN (iop<--dev) */ + scsi_flags |= 0x40000000; sg_flags = 0x10000000; break; @@ -639,8 +650,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, } */ - /* Direction, disconnect ok, tag, CDBLen */ - writel(scsi_flags | 0x20200000 | SCpnt->cmd_len, mptr ++); + writel(scsi_flags | SCpnt->cmd_len, mptr++); /* Write SCSI command into the message - always 16 byte block */ memcpy_toio(mptr, SCpnt->cmnd, 16); diff --git a/drivers/message/i2o/iop.c b/drivers/message/i2o/iop.c index 62b0d8bed186..40312053b38d 100644 --- a/drivers/message/i2o/iop.c +++ b/drivers/message/i2o/iop.c @@ -455,6 +455,70 @@ static int i2o_iop_clear(struct i2o_controller *c) return rc; } +/** + * i2o_iop_init_outbound_queue - setup the outbound message queue + * @c: I2O controller + * + * Clear and (re)initialize IOP's outbound queue and post the message + * frames to the IOP. + * + * Returns 0 on success or a negative errno code on failure. + */ +static int i2o_iop_init_outbound_queue(struct i2o_controller *c) +{ + u8 *status = c->status.virt; + u32 m; + struct i2o_message __iomem *msg; + ulong timeout; + int i; + + osm_debug("%s: Initializing Outbound Queue...\n", c->name); + + memset(status, 0, 4); + + m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET); + if (m == I2O_QUEUE_EMPTY) + return -ETIMEDOUT; + + writel(EIGHT_WORD_MSG_SIZE | TRL_OFFSET_6, &msg->u.head[0]); + writel(I2O_CMD_OUTBOUND_INIT << 24 | HOST_TID << 12 | ADAPTER_TID, + &msg->u.head[1]); + writel(i2o_exec_driver.context, &msg->u.s.icntxt); + writel(0x0106, &msg->u.s.tcntxt); /* FIXME: why 0x0106, maybe in + Spec? */ + writel(PAGE_SIZE, &msg->body[0]); + /* Outbound msg frame size in words and Initcode */ + writel(MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]); + writel(0xd0000004, &msg->body[2]); + writel(i2o_dma_low(c->status.phys), &msg->body[3]); + writel(i2o_dma_high(c->status.phys), &msg->body[4]); + + i2o_msg_post(c, m); + + timeout = jiffies + I2O_TIMEOUT_INIT_OUTBOUND_QUEUE * HZ; + while (*status <= I2O_CMD_IN_PROGRESS) { + if (time_after(jiffies, timeout)) { + osm_warn("%s: Timeout Initializing\n", c->name); + return -ETIMEDOUT; + } + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(1); + + rmb(); + } + + m = c->out_queue.phys; + + /* Post frames */ + for (i = 0; i < NMBR_MSG_FRAMES; i++) { + i2o_flush_reply(c, m); + udelay(1); /* Promise */ + m += MSG_FRAME_SIZE * 4; + } + + return 0; +} + /** * i2o_iop_reset - reset an I2O controller * @c: controller to reset @@ -491,25 +555,16 @@ static int i2o_iop_reset(struct i2o_controller *c) writel(0, &msg->u.s.tcntxt); //FIXME: use reasonable transaction context writel(0, &msg->body[0]); writel(0, &msg->body[1]); - writel(i2o_ptr_low((void *)c->status.phys), &msg->body[2]); - writel(i2o_ptr_high((void *)c->status.phys), &msg->body[3]); + writel(i2o_dma_low(c->status.phys), &msg->body[2]); + writel(i2o_dma_high(c->status.phys), &msg->body[3]); i2o_msg_post(c, m); /* Wait for a reply */ timeout = jiffies + I2O_TIMEOUT_RESET * HZ; while (!*status) { - if (time_after(jiffies, timeout)) { - printk(KERN_ERR "%s: IOP reset timeout.\n", c->name); - rc = -ETIMEDOUT; - goto exit; - } - - /* Promise bug */ - if (status[1] || status[4]) { - *status = 0; + if (time_after(jiffies, timeout)) break; - } set_current_state(TASK_UNINTERRUPTIBLE); schedule_timeout(1); @@ -517,14 +572,20 @@ static int i2o_iop_reset(struct i2o_controller *c) rmb(); } - if (*status == I2O_CMD_IN_PROGRESS) { + switch (*status) { + case I2O_CMD_REJECTED: + osm_warn("%s: IOP reset rejected\n", c->name); + rc = -EPERM; + break; + + case I2O_CMD_IN_PROGRESS: /* * Once the reset is sent, the IOP goes into the INIT state - * which is indeterminate. We need to wait until the IOP - * has rebooted before we can let the system talk to - * it. We read the inbound Free_List until a message is - * available. If we can't read one in the given ammount of - * time, we assume the IOP could not reboot properly. + * which is indeterminate. We need to wait until the IOP has + * rebooted before we can let the system talk to it. We read + * the inbound Free_List until a message is available. If we + * can't read one in the given ammount of time, we assume the + * IOP could not reboot properly. */ pr_debug("%s: Reset in progress, waiting for reboot...\n", c->name); @@ -543,19 +604,26 @@ static int i2o_iop_reset(struct i2o_controller *c) m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_RESET); } i2o_msg_nop(c, m); - } - /* from here all quiesce commands are safe */ - c->no_quiesce = 0; + /* from here all quiesce commands are safe */ + c->no_quiesce = 0; - /* If IopReset was rejected or didn't perform reset, try IopClear */ - i2o_status_get(c); - if (*status == I2O_CMD_REJECTED || sb->iop_state != ADAPTER_STATE_RESET) { - printk(KERN_WARNING "%s: Reset rejected, trying to clear\n", - c->name); - i2o_iop_clear(c); - } else - pr_debug("%s: Reset completed.\n", c->name); + /* verify if controller is in state RESET */ + i2o_status_get(c); + + if (!c->promise && (sb->iop_state != ADAPTER_STATE_RESET)) + osm_warn("%s: reset completed, but adapter not in RESET" + " state.\n", c->name); + else + osm_debug("%s: reset completed.\n", c->name); + + break; + + default: + osm_err("%s: IOP reset timeout.\n", c->name); + rc = -ETIMEDOUT; + break; + } exit: /* Enable all IOPs */ @@ -564,87 +632,6 @@ static int i2o_iop_reset(struct i2o_controller *c) return rc; }; -/** - * i2o_iop_init_outbound_queue - setup the outbound message queue - * @c: I2O controller - * - * Clear and (re)initialize IOP's outbound queue and post the message - * frames to the IOP. - * - * Returns 0 on success or a negative errno code on failure. - */ -static int i2o_iop_init_outbound_queue(struct i2o_controller *c) -{ - u8 *status = c->status.virt; - u32 m; - struct i2o_message __iomem *msg; - ulong timeout; - int i; - - pr_debug("%s: Initializing Outbound Queue...\n", c->name); - - memset(status, 0, 4); - - m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET); - if (m == I2O_QUEUE_EMPTY) - return -ETIMEDOUT; - - writel(EIGHT_WORD_MSG_SIZE | SGL_OFFSET_6, &msg->u.head[0]); - writel(I2O_CMD_OUTBOUND_INIT << 24 | HOST_TID << 12 | ADAPTER_TID, - &msg->u.head[1]); - writel(i2o_exec_driver.context, &msg->u.s.icntxt); - writel(0x00000000, &msg->u.s.tcntxt); - writel(PAGE_SIZE, &msg->body[0]); - writel(MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]); /* Outbound msg frame - size in words and Initcode */ - writel(0xd0000004, &msg->body[2]); - writel(i2o_ptr_low((void *)c->status.phys), &msg->body[3]); - writel(i2o_ptr_high((void *)c->status.phys), &msg->body[4]); - - i2o_msg_post(c, m); - - timeout = jiffies + I2O_TIMEOUT_INIT_OUTBOUND_QUEUE * HZ; - while (*status <= I2O_CMD_IN_PROGRESS) { - if (time_after(jiffies, timeout)) { - printk(KERN_WARNING "%s: Timeout Initializing\n", - c->name); - return -ETIMEDOUT; - } - set_current_state(TASK_UNINTERRUPTIBLE); - schedule_timeout(1); - - rmb(); - } - - m = c->out_queue.phys; - - /* Post frames */ - for (i = 0; i < NMBR_MSG_FRAMES; i++) { - i2o_flush_reply(c, m); - udelay(1); /* Promise */ - m += MSG_FRAME_SIZE * 4; - } - - return 0; -} - -/** - * i2o_iop_send_nop - send a core NOP message - * @c: controller - * - * Send a no-operation message with a reply set to cause no - * action either. Needed for bringing up promise controllers. - */ -static int i2o_iop_send_nop(struct i2o_controller *c) -{ - struct i2o_message __iomem *msg; - u32 m = i2o_msg_get_wait(c, &msg, HZ); - if (m == I2O_QUEUE_EMPTY) - return -ETIMEDOUT; - i2o_msg_nop(c, m); - return 0; -} - /** * i2o_iop_activate - Bring controller up to HOLD * @c: controller @@ -656,26 +643,9 @@ static int i2o_iop_send_nop(struct i2o_controller *c) */ static int i2o_iop_activate(struct i2o_controller *c) { - struct pci_dev *i960 = NULL; i2o_status_block *sb = c->status_block.virt; int rc; - - if (c->promise) { - /* Beat up the hardware first of all */ - i960 = - pci_find_slot(c->pdev->bus->number, - PCI_DEVFN(PCI_SLOT(c->pdev->devfn), 0)); - if (i960) - pci_write_config_word(i960, 0x42, 0); - - /* Follow this sequence precisely or the controller - ceases to perform useful functions until reboot */ - if ((rc = i2o_iop_send_nop(c))) - return rc; - - if ((rc = i2o_iop_reset(c))) - return rc; - } + int state; /* In INIT state, Wait Inbound Q to initialize (in i2o_status_get) */ /* In READY state, Get status */ @@ -684,7 +654,8 @@ static int i2o_iop_activate(struct i2o_controller *c) if (rc) { printk(KERN_INFO "%s: Unable to obtain status, " "attempting a reset.\n", c->name); - if (i2o_iop_reset(c)) + rc = i2o_iop_reset(c); + if (rc) return rc; } @@ -697,37 +668,37 @@ static int i2o_iop_activate(struct i2o_controller *c) switch (sb->iop_state) { case ADAPTER_STATE_FAULTED: printk(KERN_CRIT "%s: hardware fault\n", c->name); - return -ENODEV; + return -EFAULT; case ADAPTER_STATE_READY: case ADAPTER_STATE_OPERATIONAL: case ADAPTER_STATE_HOLD: case ADAPTER_STATE_FAILED: pr_debug("%s: already running, trying to reset...\n", c->name); - if (i2o_iop_reset(c)) - return -ENODEV; + rc = i2o_iop_reset(c); + if (rc) + return rc; } + /* preserve state */ + state = sb->iop_state; + rc = i2o_iop_init_outbound_queue(c); if (rc) return rc; - if (c->promise) { - if ((rc = i2o_iop_send_nop(c))) - return rc; + /* if adapter was not in RESET state clear now */ + if (state != ADAPTER_STATE_RESET) + i2o_iop_clear(c); - if ((rc = i2o_status_get(c))) - return rc; + i2o_status_get(c); - if (i960) - pci_write_config_word(i960, 0x42, 0x3FF); + if (sb->iop_state != ADAPTER_STATE_HOLD) { + osm_err("%s: failed to bring IOP into HOLD state\n", c->name); + return -EIO; } - /* In HOLD state */ - - rc = i2o_hrt_get(c); - - return rc; + return i2o_hrt_get(c); }; /** @@ -1030,8 +1001,8 @@ int i2o_status_get(struct i2o_controller *c) writel(0, &msg->u.s.tcntxt); // FIXME: use resonable transaction context writel(0, &msg->body[0]); writel(0, &msg->body[1]); - writel(i2o_ptr_low((void *)c->status_block.phys), &msg->body[2]); - writel(i2o_ptr_high((void *)c->status_block.phys), &msg->body[3]); + writel(i2o_dma_low(c->status_block.phys), &msg->body[2]); + writel(i2o_dma_high(c->status_block.phys), &msg->body[3]); writel(sizeof(i2o_status_block), &msg->body[4]); /* always 88 bytes */ i2o_msg_post(c, m); diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c index f33fd81f77a4..a499af096a68 100644 --- a/drivers/message/i2o/pci.c +++ b/drivers/message/i2o/pci.c @@ -49,30 +49,6 @@ static struct pci_device_id __devinitdata i2o_pci_ids[] = { {0} }; -/** - * i2o_dma_realloc - Realloc DMA memory - * @dev: struct device pointer to the PCI device of the I2O controller - * @addr: pointer to a i2o_dma struct DMA buffer - * @len: new length of memory - * @gfp_mask: GFP mask - * - * If there was something allocated in the addr, free it first. If len > 0 - * than try to allocate it and write the addresses back to the addr - * structure. If len == 0 set the virtual address to NULL. - * - * Returns the 0 on success or negative error code on failure. - */ -int i2o_dma_realloc(struct device *dev, struct i2o_dma *addr, size_t len, - unsigned int gfp_mask) -{ - i2o_dma_free(dev, addr); - - if (len) - return i2o_dma_alloc(dev, addr, len, gfp_mask); - - return 0; -}; - /** * i2o_pci_free - Frees the DMA memory for the I2O controller * @c: I2O controller to free @@ -185,6 +161,7 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c) } else c->in_queue = c->base; + c->irq_status = c->base.virt + I2O_IRQ_STATUS; c->irq_mask = c->base.virt + I2O_IRQ_MASK; c->in_port = c->base.virt + I2O_IN_PORT; c->out_port = c->base.virt + I2O_OUT_PORT; @@ -232,36 +209,30 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c) static irqreturn_t i2o_pci_interrupt(int irq, void *dev_id, struct pt_regs *r) { struct i2o_controller *c = dev_id; - struct device *dev = &c->pdev->dev; - u32 mv = readl(c->out_port); - - /* - * Old 960 steppings had a bug in the I2O unit that caused - * the queue to appear empty when it wasn't. - */ - if (mv == I2O_QUEUE_EMPTY) { - mv = readl(c->out_port); - if (unlikely(mv == I2O_QUEUE_EMPTY)) - return IRQ_NONE; - else - pr_debug("%s: 960 bug detected\n", c->name); - } + u32 m; + irqreturn_t rc = IRQ_NONE; + + while (readl(c->irq_status) & I2O_IRQ_OUTBOUND_POST) { + m = readl(c->out_port); + if (m == I2O_QUEUE_EMPTY) { + /* + * Old 960 steppings had a bug in the I2O unit that + * caused the queue to appear empty when it wasn't. + */ + m = readl(c->out_port); + if (unlikely(m == I2O_QUEUE_EMPTY)) + break; + } - while (mv != I2O_QUEUE_EMPTY) { /* dispatch it */ - if (i2o_driver_dispatch(c, mv)) + if (i2o_driver_dispatch(c, m)) /* flush it if result != 0 */ - i2o_flush_reply(c, mv); + i2o_flush_reply(c, m); - /* - * That 960 bug again... - */ - mv = readl(c->out_port); - if (mv == I2O_QUEUE_EMPTY) - mv = readl(c->out_port); + rc = IRQ_HANDLED; } - return IRQ_HANDLED; + return rc; } /** diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h index 3414325bdcfd..90c984ecd521 100644 --- a/include/linux/i2o-dev.h +++ b/include/linux/i2o-dev.h @@ -32,6 +32,10 @@ typedef unsigned int u32; #endif /* __KERNEL__ */ +/* + * Vendors + */ +#define I2O_VENDOR_DPT 0x001b /* * I2O Control IOCTLs and structures @@ -333,7 +337,7 @@ typedef struct _i2o_status_block { #define I2O_CLASS_ATE_PERIPHERAL 0x061 #define I2O_CLASS_FLOPPY_CONTROLLER 0x070 #define I2O_CLASS_FLOPPY_DEVICE 0x071 -#define I2O_CLASS_BUS_ADAPTER_PORT 0x080 +#define I2O_CLASS_BUS_ADAPTER 0x080 #define I2O_CLASS_PEER_TRANSPORT_AGENT 0x090 #define I2O_CLASS_PEER_TRANSPORT 0x091 #define I2O_CLASS_END 0xfff diff --git a/include/linux/i2o.h b/include/linux/i2o.h index e8cd11290010..497ea574f96b 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -157,7 +157,8 @@ struct i2o_controller { void __iomem *in_port; /* Inbout port address */ void __iomem *out_port; /* Outbound port address */ - void __iomem *irq_mask; /* Interrupt register address */ + void __iomem *irq_status; /* Interrupt status register address */ + void __iomem *irq_mask; /* Interrupt mask register address */ /* Dynamic LCT related data */ @@ -242,15 +243,6 @@ extern int i2o_msg_post_wait_mem(struct i2o_controller *, u32, unsigned long, extern void i2o_msg_nop(struct i2o_controller *, u32); static inline void i2o_flush_reply(struct i2o_controller *, u32); -/* DMA handling functions */ -static inline int i2o_dma_alloc(struct device *, struct i2o_dma *, size_t, - unsigned int); -static inline void i2o_dma_free(struct device *, struct i2o_dma *); -int i2o_dma_realloc(struct device *, struct i2o_dma *, size_t, unsigned int); - -static inline int i2o_dma_map(struct device *, struct i2o_dma *); -static inline void i2o_dma_unmap(struct device *, struct i2o_dma *); - /* IOP functions */ extern int i2o_status_get(struct i2o_controller *); @@ -275,6 +267,16 @@ static inline u32 i2o_ptr_high(void *ptr) { return (u32) ((u64) ptr >> 32); }; + +static inline u32 i2o_dma_low(dma_addr_t dma_addr) +{ + return (u32) (u64) dma_addr; +}; + +static inline u32 i2o_dma_high(dma_addr_t dma_addr) +{ + return (u32) ((u64) dma_addr >> 32); +}; #else static inline u32 i2o_cntxt_list_add(struct i2o_controller *c, void *ptr) { @@ -305,8 +307,246 @@ static inline u32 i2o_ptr_high(void *ptr) { return 0; }; + +static inline u32 i2o_dma_low(dma_addr_t dma_addr) +{ + return (u32) dma_addr; +}; + +static inline u32 i2o_dma_high(dma_addr_t dma_addr) +{ + return 0; +}; +#endif + +/** + * i2o_sg_tablesize - Calculate the maximum number of elements in a SGL + * @c: I2O controller for which the calculation should be done + * @body_size: maximum body size used for message in 32-bit words. + * + * Return the maximum number of SG elements in a SG list. + */ +static inline u16 i2o_sg_tablesize(struct i2o_controller *c, u16 body_size) +{ + i2o_status_block *sb = c->status_block.virt; + u16 sg_count = + (sb->inbound_frame_size - sizeof(struct i2o_message) / 4) - + body_size; + + if (c->pae_support) { + /* + * for 64-bit a SG attribute element must be added and each + * SG element needs 12 bytes instead of 8. + */ + sg_count -= 2; + sg_count /= 3; + } else + sg_count /= 2; + + if (c->short_req && (sg_count > 8)) + sg_count = 8; + + return sg_count; +}; + +/** + * i2o_dma_map_single - Map pointer to controller and fill in I2O message. + * @c: I2O controller + * @ptr: pointer to the data which should be mapped + * @size: size of data in bytes + * @direction: DMA_TO_DEVICE / DMA_FROM_DEVICE + * @sg_ptr: pointer to the SG list inside the I2O message + * + * This function does all necessary DMA handling and also writes the I2O + * SGL elements into the I2O message. For details on DMA handling see also + * dma_map_single(). The pointer sg_ptr will only be set to the end of the + * SG list if the allocation was successful. + * + * Returns DMA address which must be checked for failures using + * dma_mapping_error(). + */ +static inline dma_addr_t i2o_dma_map_single(struct i2o_controller *c, void *ptr, + size_t size, + enum dma_data_direction direction, + u32 __iomem ** sg_ptr) +{ + u32 sg_flags; + u32 __iomem *mptr = *sg_ptr; + dma_addr_t dma_addr; + + switch (direction) { + case DMA_TO_DEVICE: + sg_flags = 0xd4000000; + break; + case DMA_FROM_DEVICE: + sg_flags = 0xd0000000; + break; + default: + return 0; + } + + dma_addr = dma_map_single(&c->pdev->dev, ptr, size, direction); + if (!dma_mapping_error(dma_addr)) { +#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64 + if ((sizeof(dma_addr_t) > 4) && c->pae_support) { + writel(0x7C020002, mptr++); + writel(PAGE_SIZE, mptr++); + } +#endif + + writel(sg_flags | size, mptr++); + writel(i2o_dma_low(dma_addr), mptr++); +#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64 + if ((sizeof(dma_addr_t) > 4) && c->pae_support) + writel(i2o_dma_high(dma_addr), mptr++); +#endif + *sg_ptr = mptr; + } + return dma_addr; +}; + +/** + * i2o_dma_map_sg - Map a SG List to controller and fill in I2O message. + * @c: I2O controller + * @sg: SG list to be mapped + * @sg_count: number of elements in the SG list + * @direction: DMA_TO_DEVICE / DMA_FROM_DEVICE + * @sg_ptr: pointer to the SG list inside the I2O message + * + * This function does all necessary DMA handling and also writes the I2O + * SGL elements into the I2O message. For details on DMA handling see also + * dma_map_sg(). The pointer sg_ptr will only be set to the end of the SG + * list if the allocation was successful. + * + * Returns 0 on failure or 1 on success. + */ +static inline int i2o_dma_map_sg(struct i2o_controller *c, + struct scatterlist *sg, int sg_count, + enum dma_data_direction direction, + u32 __iomem ** sg_ptr) +{ + u32 sg_flags; + u32 __iomem *mptr = *sg_ptr; + + switch (direction) { + case DMA_TO_DEVICE: + sg_flags = 0x14000000; + break; + case DMA_FROM_DEVICE: + sg_flags = 0x10000000; + break; + default: + return 0; + } + + sg_count = dma_map_sg(&c->pdev->dev, sg, sg_count, direction); + if (!sg_count) + return 0; + +#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64 + if ((sizeof(dma_addr_t) > 4) && c->pae_support) { + writel(0x7C020002, mptr++); + writel(PAGE_SIZE, mptr++); + } #endif + while (sg_count-- > 0) { + if (!sg_count) + sg_flags |= 0xC0000000; + writel(sg_flags | sg_dma_len(sg), mptr++); + writel(i2o_dma_low(sg_dma_address(sg)), mptr++); +#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64 + if ((sizeof(dma_addr_t) > 4) && c->pae_support) + writel(i2o_dma_high(sg_dma_address(sg)), mptr++); +#endif + sg++; + } + *sg_ptr = mptr; + + return 1; +}; + +/** + * i2o_dma_alloc - Allocate DMA memory + * @dev: struct device pointer to the PCI device of the I2O controller + * @addr: i2o_dma struct which should get the DMA buffer + * @len: length of the new DMA memory + * @gfp_mask: GFP mask + * + * Allocate a coherent DMA memory and write the pointers into addr. + * + * Returns 0 on success or -ENOMEM on failure. + */ +static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr, + size_t len, unsigned int gfp_mask) +{ + struct pci_dev *pdev = to_pci_dev(dev); + int dma_64 = 0; + + if ((sizeof(dma_addr_t) > 4) && (pdev->dma_mask == DMA_64BIT_MASK)) { + dma_64 = 1; + if (pci_set_dma_mask(pdev, DMA_32BIT_MASK)) + return -ENOMEM; + } + + addr->virt = dma_alloc_coherent(dev, len, &addr->phys, gfp_mask); + + if ((sizeof(dma_addr_t) > 4) && dma_64) + if (pci_set_dma_mask(pdev, DMA_64BIT_MASK)) + printk(KERN_WARNING "i2o: unable to set 64-bit DMA"); + + if (!addr->virt) + return -ENOMEM; + + memset(addr->virt, 0, len); + addr->len = len; + + return 0; +}; + +/** + * i2o_dma_free - Free DMA memory + * @dev: struct device pointer to the PCI device of the I2O controller + * @addr: i2o_dma struct which contains the DMA buffer + * + * Free a coherent DMA memory and set virtual address of addr to NULL. + */ +static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr) +{ + if (addr->virt) { + if (addr->phys) + dma_free_coherent(dev, addr->len, addr->virt, + addr->phys); + else + kfree(addr->virt); + addr->virt = NULL; + } +}; + +/** + * i2o_dma_realloc - Realloc DMA memory + * @dev: struct device pointer to the PCI device of the I2O controller + * @addr: pointer to a i2o_dma struct DMA buffer + * @len: new length of memory + * @gfp_mask: GFP mask + * + * If there was something allocated in the addr, free it first. If len > 0 + * than try to allocate it and write the addresses back to the addr + * structure. If len == 0 set the virtual address to NULL. + * + * Returns the 0 on success or negative error code on failure. + */ +static inline int i2o_dma_realloc(struct device *dev, struct i2o_dma *addr, + size_t len, unsigned int gfp_mask) +{ + i2o_dma_free(dev, addr); + + if (len) + return i2o_dma_alloc(dev, addr, len, gfp_mask); + + return 0; +}; + /* I2O driver (OSM) functions */ extern int i2o_driver_register(struct i2o_driver *); extern void i2o_driver_unregister(struct i2o_driver *); @@ -375,10 +615,11 @@ extern int i2o_device_claim_release(struct i2o_device *); /* Exec OSM functions */ extern int i2o_exec_lct_get(struct i2o_controller *); -/* device / driver conversion functions */ +/* device / driver / kobject conversion functions */ #define to_i2o_driver(drv) container_of(drv,struct i2o_driver, driver) #define to_i2o_device(dev) container_of(dev, struct i2o_device, device) #define to_i2o_controller(dev) container_of(dev, struct i2o_controller, device) +#define kobj_to_i2o_device(kobj) to_i2o_device(container_of(kobj, struct device, kobj)) /** * i2o_msg_get - obtain an I2O message from the IOP @@ -466,8 +707,10 @@ static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct i2o_controller *c, u32 m) { - BUG_ON(m < c->out_queue.phys - || m >= c->out_queue.phys + c->out_queue.len); + if (unlikely + (m < c->out_queue.phys + || m >= c->out_queue.phys + c->out_queue.len)) + return NULL; return c->out_queue.virt + (m - c->out_queue.phys); }; @@ -532,48 +775,6 @@ static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr) } }; -/** - * i2o_dma_map - Map the memory to DMA - * @dev: struct device pointer to the PCI device of the I2O controller - * @addr: i2o_dma struct which should be mapped - * - * Map the memory in addr->virt to coherent DMA memory and write the - * physical address into addr->phys. - * - * Returns 0 on success or -ENOMEM on failure. - */ -static inline int i2o_dma_map(struct device *dev, struct i2o_dma *addr) -{ - if (!addr->virt) - return -EFAULT; - - if (!addr->phys) - addr->phys = dma_map_single(dev, addr->virt, addr->len, - DMA_BIDIRECTIONAL); - if (!addr->phys) - return -ENOMEM; - - return 0; -}; - -/** - * i2o_dma_unmap - Unmap the DMA memory - * @dev: struct device pointer to the PCI device of the I2O controller - * @addr: i2o_dma struct which should be unmapped - * - * Unmap the memory in addr->virt from DMA memory. - */ -static inline void i2o_dma_unmap(struct device *dev, struct i2o_dma *addr) -{ - if (!addr->virt) - return; - - if (addr->phys) { - dma_unmap_single(dev, addr->phys, addr->len, DMA_BIDIRECTIONAL); - addr->phys = 0; - } -}; - /* * Endian handling wrapped into the macro - keeps the core code * cleaner. @@ -725,6 +926,14 @@ extern void i2o_debug_state(struct i2o_controller *c); #define I2O_CMD_SCSI_ABORT 0x83 #define I2O_CMD_SCSI_BUSRESET 0x27 +/* + * Bus Adapter Class + */ +#define I2O_CMD_BUS_ADAPTER_RESET 0x85 +#define I2O_CMD_BUS_RESET 0x87 +#define I2O_CMD_BUS_SCAN 0x89 +#define I2O_CMD_BUS_QUIESCE 0x8b + /* * Random Block Storage Class */ @@ -948,7 +1157,7 @@ extern void i2o_debug_state(struct i2o_controller *c); /* request queue sizes */ #define I2O_MAX_SECTORS 1024 -#define I2O_MAX_SEGMENTS 128 +#define I2O_MAX_PHYS_SEGMENTS MAX_PHYS_SEGMENTS #define I2O_REQ_MEMPOOL_SIZE 32 -- cgit v1.2.3 From b2aaee33fbb354a2f08121aa1c1be55841102761 Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:19 -0700 Subject: [PATCH] I2O: Adaptec specific SG_IO access, firmware access through sysfs and 2400A workaround Changes: - Provide SG_IO access to BLOCK and EXECUTIVE class on Adaptec controllers - Use PRIVATE messages in SCSI-OSM because on some controllers normal SCSI class commands like READ or READ CAPACITY cause errors - Use new DMA and SG list creation function - Added workaround to limit sectors per request for Adaptec 2400A controllers Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/i2o/Kconfig | 18 +++ drivers/message/i2o/i2o_block.h | 6 + drivers/message/i2o/i2o_config.c | 4 + drivers/message/i2o/i2o_scsi.c | 263 +++++++++++++++++++++++---------------- drivers/message/i2o/pci.c | 22 ++++ include/linux/i2o-dev.h | 22 ++++ include/linux/i2o.h | 37 ++++-- include/scsi/sg_request.h | 26 ++++ 8 files changed, 282 insertions(+), 116 deletions(-) create mode 100644 include/scsi/sg_request.h (limited to 'include/linux') diff --git a/drivers/message/i2o/Kconfig b/drivers/message/i2o/Kconfig index ce278e060aca..94b6d676c5cb 100644 --- a/drivers/message/i2o/Kconfig +++ b/drivers/message/i2o/Kconfig @@ -24,6 +24,24 @@ config I2O If unsure, say N. +config I2O_EXT_ADAPTEC + bool "Enable Adaptec extensions" + depends on I2O + default y + ---help--- + Say Y for support of raidutils for Adaptec I2O controllers. You also + have to say Y to "I2O Configuration support", "I2O SCSI OSM" below + and to "SCSI generic support" under "SCSI device configuration". + +config I2O_EXT_ADAPTEC_DMA64 + bool "Enable 64-bit DMA" + depends on I2O_EXT_ADAPTEC && ( 64BIT || HIGHMEM64G ) + default y + ---help--- + Say Y for support of 64-bit DMA transfer mode on Adaptec I2O + controllers. + Note: You need at least firmware version 3709. + config I2O_CONFIG tristate "I2O Configuration support" depends on PCI && I2O diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h index 9e1a95fb0833..e45cc40ce384 100644 --- a/drivers/message/i2o/i2o_block.h +++ b/drivers/message/i2o/i2o_block.h @@ -56,6 +56,12 @@ #define I2O_BLOCK_RETRY_TIME HZ/4 #define I2O_BLOCK_MAX_OPEN_REQUESTS 50 +/* request queue sizes */ +#define I2O_BLOCK_REQ_MEMPOOL_SIZE 32 + +#define KERNEL_SECTOR_SHIFT 9 +#define KERNEL_SECTOR_SIZE (1 << KERNEL_SECTOR_SHIFT) + /* I2O Block OSM mempool struct */ struct i2o_block_mempool { kmem_cache_t *slab; diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c index 849d90aad779..7636833b4623 100644 --- a/drivers/message/i2o/i2o_config.c +++ b/drivers/message/i2o/i2o_config.c @@ -515,6 +515,7 @@ static int i2o_cfg_evt_get(unsigned long arg, struct file *fp) return 0; } +#ifdef CONFIG_I2O_EXT_ADAPTEC #ifdef CONFIG_COMPAT static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long arg) { @@ -964,6 +965,7 @@ static int i2o_cfg_passthru(unsigned long arg) kfree(reply); return rcode; } +#endif /* * IOCTL Handler @@ -1018,9 +1020,11 @@ static int i2o_cfg_ioctl(struct inode *inode, struct file *fp, unsigned int cmd, ret = i2o_cfg_evt_get(arg, fp); break; +#ifdef CONFIG_I2O_EXT_ADAPTEC case I2OPASSTHRU: ret = i2o_cfg_passthru(arg); break; +#endif default: osm_debug("unknown ioctl called!\n"); diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c index c3b0c29ac02d..fef53b509a61 100644 --- a/drivers/message/i2o/i2o_scsi.c +++ b/drivers/message/i2o/i2o_scsi.c @@ -55,6 +55,7 @@ #include #include #include +#include #include #include @@ -65,19 +66,23 @@ #include #include #include +#include +#include +#include #define OSM_NAME "scsi-osm" -#define OSM_VERSION "$Rev$" +#define OSM_VERSION "1.282" #define OSM_DESCRIPTION "I2O SCSI Peripheral OSM" static struct i2o_driver i2o_scsi_driver; -static int i2o_scsi_max_id = 16; -static int i2o_scsi_max_lun = 8; +static unsigned int i2o_scsi_max_id = 16; +static unsigned int i2o_scsi_max_lun = 255; struct i2o_scsi_host { struct Scsi_Host *scsi_host; /* pointer to the SCSI host */ struct i2o_controller *iop; /* pointer to the I2O controller */ + unsigned int lun; /* lun's used for block devices */ struct i2o_device *channel[0]; /* channel->i2o_dev mapping table */ }; @@ -100,12 +105,17 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c) u8 type; int i; size_t size; - i2o_status_block *sb; + u16 body_size = 6; + +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (c->adaptec) + body_size = 8; +#endif list_for_each_entry(i2o_dev, &c->devices, list) if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER) { if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) - && (type == 0x01)) /* SCSI bus */ + && (type == 0x01)) /* SCSI bus */ max_channel++; } @@ -127,20 +137,18 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c) scsi_host->max_id = i2o_scsi_max_id; scsi_host->max_lun = i2o_scsi_max_lun; scsi_host->this_id = c->unit; - - sb = c->status_block.virt; - - scsi_host->sg_tablesize = (sb->inbound_frame_size - - sizeof(struct i2o_message) / 4 - 6) / 2; + scsi_host->sg_tablesize = i2o_sg_tablesize(c, body_size); i2o_shost = (struct i2o_scsi_host *)scsi_host->hostdata; i2o_shost->scsi_host = scsi_host; i2o_shost->iop = c; + i2o_shost->lun = 1; i = 0; list_for_each_entry(i2o_dev, &c->devices, list) if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER) { - if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) || (type == 1)) /* only SCSI bus */ + if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) + && (type == 0x01)) /* only SCSI bus */ i2o_shost->channel[i++] = i2o_dev; if (i >= max_channel) @@ -212,8 +220,8 @@ static int i2o_scsi_probe(struct device *dev) struct Scsi_Host *scsi_host; struct i2o_device *parent; struct scsi_device *scsi_dev; - u32 id; - u64 lun; + u32 id = -1; + u64 lun = -1; int channel = -1; int i; @@ -223,8 +231,56 @@ static int i2o_scsi_probe(struct device *dev) scsi_host = i2o_shost->scsi_host; - if (i2o_parm_field_get(i2o_dev, 0, 3, &id, 4) < 0) + switch (i2o_dev->lct_data.class_id) { + case I2O_CLASS_RANDOM_BLOCK_STORAGE: + case I2O_CLASS_EXECUTIVE: +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (c->adaptec) { + u8 type; + struct i2o_device *d = i2o_shost->channel[0]; + + if (i2o_parm_field_get(d, 0x0000, 0, &type, 1) + && (type == 0x01)) /* SCSI bus */ + if (i2o_parm_field_get(d, 0x0200, 4, &id, 4)) { + channel = 0; + if (i2o_dev->lct_data.class_id == + I2O_CLASS_RANDOM_BLOCK_STORAGE) + lun = i2o_shost->lun++; + else + lun = 0; + } + } +#endif + break; + + case I2O_CLASS_SCSI_PERIPHERAL: + if (i2o_parm_field_get(i2o_dev, 0x0000, 3, &id, 4) < 0) + return -EFAULT; + + if (i2o_parm_field_get(i2o_dev, 0x0000, 4, &lun, 8) < 0) + return -EFAULT; + + parent = i2o_iop_find_device(c, i2o_dev->lct_data.parent_tid); + if (!parent) { + osm_warn("can not find parent of device %03x\n", + i2o_dev->lct_data.tid); + return -EFAULT; + } + + for (i = 0; i <= i2o_shost->scsi_host->max_channel; i++) + if (i2o_shost->channel[i] == parent) + channel = i; + break; + + default: + return -EFAULT; + } + + if (channel == -1) { + osm_warn("can not find channel of device %03x\n", + i2o_dev->lct_data.tid); return -EFAULT; + } if (id >= scsi_host->max_id) { osm_warn("SCSI device id (%d) >= max_id of I2O host (%d)", id, @@ -232,31 +288,12 @@ static int i2o_scsi_probe(struct device *dev) return -EFAULT; } - if (i2o_parm_field_get(i2o_dev, 0, 4, &lun, 8) < 0) - return -EFAULT; if (lun >= scsi_host->max_lun) { osm_warn("SCSI device id (%d) >= max_lun of I2O host (%d)", (unsigned int)lun, scsi_host->max_lun); return -EFAULT; } - parent = i2o_iop_find_device(c, i2o_dev->lct_data.parent_tid); - if (!parent) { - osm_warn("can not find parent of device %03x\n", - i2o_dev->lct_data.tid); - return -EFAULT; - } - - for (i = 0; i <= i2o_shost->scsi_host->max_channel; i++) - if (i2o_shost->channel[i] == parent) - channel = i; - - if (channel == -1) { - osm_warn("can not find channel of device %03x\n", - i2o_dev->lct_data.tid); - return -EFAULT; - } - scsi_dev = __scsi_add_device(i2o_shost->scsi_host, channel, id, lun, i2o_dev); @@ -266,7 +303,8 @@ static int i2o_scsi_probe(struct device *dev) return PTR_ERR(scsi_dev); } - sysfs_create_link(&i2o_dev->device.kobj, &scsi_dev->sdev_gendev.kobj, "scsi"); + sysfs_create_link(&i2o_dev->device.kobj, &scsi_dev->sdev_gendev.kobj, + "scsi"); osm_info("device added (TID: %03x) channel: %d, id: %d, lun: %d\n", i2o_dev->lct_data.tid, channel, id, (unsigned int)lun); @@ -542,9 +580,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, void (*done) (struct scsi_cmnd *)) { struct i2o_controller *c; - struct Scsi_Host *host; struct i2o_device *i2o_dev; - struct device *dev; int tid; struct i2o_message __iomem *msg; u32 m; @@ -554,20 +590,16 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, * RETURN_SENSE_DATA_IN_REPLY_MESSAGE_FRAME */ u32 scsi_flags = 0x20a00000; - u32 sg_flags; + u32 sgl_offset; u32 __iomem *mptr; - u32 __iomem *lenptr; - u32 len; - int i; + u32 cmd = I2O_CMD_SCSI_EXEC << 24; + int rc = 0; /* * Do the incoming paperwork */ - i2o_dev = SCpnt->device->hostdata; - host = SCpnt->device->host; c = i2o_dev->iop; - dev = &c->pdev->dev; SCpnt->scsi_done = done; @@ -575,7 +607,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, osm_warn("no I2O device in request\n"); SCpnt->result = DID_NO_CONNECT << 16; done(SCpnt); - return 0; + goto exit; } tid = i2o_dev->lct_data.tid; @@ -583,47 +615,86 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, osm_debug("qcmd: Tid = %03x\n", tid); osm_debug("Real scsi messages.\n"); - /* - * Obtain an I2O message. If there are none free then - * throw it back to the scsi layer - */ - - m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET); - if (m == I2O_QUEUE_EMPTY) - return SCSI_MLQUEUE_HOST_BUSY; - - mptr = &msg->body[0]; - /* * Put together a scsi execscb message */ - switch (SCpnt->sc_data_direction) { case PCI_DMA_NONE: /* DATA NO XFER */ - sg_flags = 0x00000000; + sgl_offset = SGL_OFFSET_0; break; case PCI_DMA_TODEVICE: /* DATA OUT (iop-->dev) */ scsi_flags |= 0x80000000; - sg_flags = 0x14000000; + sgl_offset = SGL_OFFSET_10; break; case PCI_DMA_FROMDEVICE: /* DATA IN (iop<--dev) */ scsi_flags |= 0x40000000; - sg_flags = 0x10000000; + sgl_offset = SGL_OFFSET_10; break; default: /* Unknown - kill the command */ SCpnt->result = DID_NO_CONNECT << 16; done(SCpnt); - return 0; + goto exit; } - writel(I2O_CMD_SCSI_EXEC << 24 | HOST_TID << 12 | tid, &msg->u.head[1]); + /* + * Obtain an I2O message. If there are none free then + * throw it back to the scsi layer + */ + + m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET); + if (m == I2O_QUEUE_EMPTY) { + rc = SCSI_MLQUEUE_HOST_BUSY; + goto exit; + } + + mptr = &msg->body[0]; + +#ifdef CONFIG_I2O_EXT_ADAPTEC + if (c->adaptec) { + u32 adpt_flags = 0; + + if (SCpnt->sc_request && SCpnt->sc_request->upper_private_data) { + i2o_sg_io_hdr_t __user *usr_ptr = + ((Sg_request *) (SCpnt->sc_request-> + upper_private_data))->header. + usr_ptr; + + if (usr_ptr) + get_user(adpt_flags, &usr_ptr->flags); + } + + switch (i2o_dev->lct_data.class_id) { + case I2O_CLASS_EXECUTIVE: + case I2O_CLASS_RANDOM_BLOCK_STORAGE: + /* interpret flag has to be set for executive */ + adpt_flags ^= I2O_DPT_SG_FLAG_INTERPRET; + break; + + default: + break; + } + + /* + * for Adaptec controllers we use the PRIVATE command, because + * the normal SCSI EXEC doesn't support all SCSI commands on + * all controllers (for example READ CAPACITY). + */ + if (sgl_offset == SGL_OFFSET_10) + sgl_offset = SGL_OFFSET_12; + cmd = I2O_CMD_PRIVATE << 24; + writel(I2O_VENDOR_DPT << 16 | I2O_CMD_SCSI_EXEC, mptr++); + writel(adpt_flags | tid, mptr++); + } +#endif + + writel(cmd | HOST_TID << 12 | tid, &msg->u.head[1]); writel(i2o_scsi_driver.context, &msg->u.s.icntxt); /* We want the SCSI control block back */ @@ -655,55 +726,30 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, /* Write SCSI command into the message - always 16 byte block */ memcpy_toio(mptr, SCpnt->cmnd, 16); mptr += 4; - lenptr = mptr++; /* Remember me - fill in when we know */ - - /* Now fill in the SGList and command */ - if (SCpnt->use_sg) { - struct scatterlist *sg; - int sg_count; - - sg = SCpnt->request_buffer; - len = 0; - sg_count = dma_map_sg(dev, sg, SCpnt->use_sg, - SCpnt->sc_data_direction); - - if (unlikely(sg_count <= 0)) - return -ENOMEM; - - for (i = SCpnt->use_sg; i > 0; i--) { - if (i == 1) - sg_flags |= 0xC0000000; - writel(sg_flags | sg_dma_len(sg), mptr++); - writel(sg_dma_address(sg), mptr++); - len += sg_dma_len(sg); - sg++; - } - - writel(len, lenptr); - } else { - len = SCpnt->request_bufflen; - - writel(len, lenptr); - - if (len > 0) { - dma_addr_t dma_addr; - - dma_addr = dma_map_single(dev, SCpnt->request_buffer, - SCpnt->request_bufflen, - SCpnt->sc_data_direction); - if (!dma_addr) - return -ENOMEM; - - SCpnt->SCp.ptr = (void *)(unsigned long)dma_addr; - sg_flags |= 0xC0000000; - writel(sg_flags | SCpnt->request_bufflen, mptr++); - writel(dma_addr, mptr++); + if (sgl_offset != SGL_OFFSET_0) { + /* write size of data addressed by SGL */ + writel(SCpnt->request_bufflen, mptr++); + + /* Now fill in the SGList and command */ + if (SCpnt->use_sg) { + if (!i2o_dma_map_sg(c, SCpnt->request_buffer, + SCpnt->use_sg, + SCpnt->sc_data_direction, &mptr)) + goto nomem; + } else { + SCpnt->SCp.dma_handle = + i2o_dma_map_single(c, SCpnt->request_buffer, + SCpnt->request_bufflen, + SCpnt->sc_data_direction, &mptr); + if (dma_mapping_error(SCpnt->SCp.dma_handle)) + goto nomem; } } /* Stick the headers on */ - writel((mptr - &msg->u.head[0]) << 16 | SGL_OFFSET_10, &msg->u.head[0]); + writel(I2O_MESSAGE_SIZE(mptr - &msg->u.head[0]) | sgl_offset, + &msg->u.head[0]); /* Queue the message */ i2o_msg_post(c, m); @@ -711,6 +757,13 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, osm_debug("Issued %ld\n", SCpnt->serial_number); return 0; + + nomem: + rc = -ENOMEM; + i2o_msg_nop(c, m); + + exit: + return rc; }; /** diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c index a499af096a68..964fe481849e 100644 --- a/drivers/message/i2o/pci.c +++ b/drivers/message/i2o/pci.c @@ -362,11 +362,33 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, c->promise = 1; } + if (pdev->subsystem_vendor == PCI_VENDOR_ID_DPT) + c->adaptec = 1; + /* Cards that go bananas if you quiesce them before you reset them. */ if (pdev->vendor == PCI_VENDOR_ID_DPT) { c->no_quiesce = 1; if (pdev->device == 0xa511) c->raptor = 1; + + if (pdev->subsystem_device == 0xc05a) { + c->limit_sectors = 1; + printk(KERN_INFO + "%s: limit sectors per request to %d\n", c->name, + I2O_MAX_SECTORS_LIMITED); + } +#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64 + if (sizeof(dma_addr_t) > 4) { + if (pci_set_dma_mask(pdev, DMA_64BIT_MASK)) + printk(KERN_INFO "%s: 64-bit DMA unavailable\n", + c->name); + else { + c->pae_support = 1; + printk(KERN_INFO "%s: using 64-bit DMA\n", + c->name); + } + } +#endif } if ((rc = i2o_pci_alloc(c))) { diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h index 90c984ecd521..d4a08d29e36d 100644 --- a/include/linux/i2o-dev.h +++ b/include/linux/i2o-dev.h @@ -32,6 +32,13 @@ typedef unsigned int u32; #endif /* __KERNEL__ */ +/* + * Software module types + */ +#define I2O_SOFTWARE_MODULE_IRTOS 0x11 +#define I2O_SOFTWARE_MODULE_IOP_PRIVATE 0x22 +#define I2O_SOFTWARE_MODULE_IOP_CONFIG 0x23 + /* * Vendors */ @@ -125,6 +132,10 @@ struct i2o_evt_get { int lost; }; +typedef struct i2o_sg_io_hdr { + unsigned int flags; /* see I2O_DPT_SG_IO_FLAGS */ +} i2o_sg_io_hdr_t; + /************************************************************************** * HRT related constants and structures **************************************************************************/ @@ -403,4 +414,15 @@ typedef struct _i2o_status_block { #define ADAPTER_STATE_FAILED 0x10 #define ADAPTER_STATE_FAULTED 0x11 + +/* + * DPT / Adaptec specific values for i2o_sg_io_hdr flags. + */ +#define I2O_DPT_SG_FLAG_INTERPRET 0x00010000 +#define I2O_DPT_SG_FLAG_PHYSICAL 0x00020000 + +#define I2O_DPT_FLASH_FRAG_SIZE 0x10000 +#define I2O_DPT_FLASH_READ 0x0101 +#define I2O_DPT_FLASH_WRITE 0x0102 + #endif /* _I2O_DEV_H */ diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 497ea574f96b..2039a87c2b91 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -147,10 +147,13 @@ struct i2o_controller { struct pci_dev *pdev; /* PCI device */ - unsigned int short_req:1; /* use small block sizes */ - unsigned int no_quiesce:1; /* dont quiesce before reset */ - unsigned int raptor:1; /* split bar */ unsigned int promise:1; /* Promise controller */ + unsigned int adaptec:1; /* DPT / Adaptec controller */ + unsigned int raptor:1; /* split bar */ + unsigned int no_quiesce:1; /* dont quiesce before reset */ + unsigned int short_req:1; /* use small block sizes */ + unsigned int limit_sectors:1; /* limit number of sectors / request */ + unsigned int pae_support:1; /* controller has 64-bit SGL support */ struct list_head devices; /* list of I2O devices */ struct list_head list; /* Controller list */ @@ -746,7 +749,21 @@ static inline struct i2o_message __iomem *i2o_msg_in_to_virt(struct i2o_controll static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr, size_t len, unsigned int gfp_mask) { + struct pci_dev *pdev = to_pci_dev(dev); + int dma_64 = 0; + + if ((sizeof(dma_addr_t) > 4) && (pdev->dma_mask == DMA_64BIT_MASK)) { + dma_64 = 1; + if(pci_set_dma_mask(pdev, DMA_32BIT_MASK)) + return -ENOMEM; + } + addr->virt = dma_alloc_coherent(dev, len, &addr->phys, gfp_mask); + + if ((sizeof(dma_addr_t) > 4) && dma_64) + if(pci_set_dma_mask(pdev, DMA_64BIT_MASK)) + printk(KERN_WARNING "i2o: unable to set 64-bit DMA"); + if (!addr->virt) return -ENOMEM; @@ -946,7 +963,7 @@ extern void i2o_debug_state(struct i2o_controller *c); #define I2O_CMD_BLOCK_MEJECT 0x43 #define I2O_CMD_BLOCK_POWER 0x70 -#define I2O_PRIVATE_MSG 0xFF +#define I2O_CMD_PRIVATE 0xFF /* Command status values */ @@ -1095,9 +1112,9 @@ extern void i2o_debug_state(struct i2o_controller *c); #define SGL_OFFSET_8 (0x0080 | I2OVERSION) #define SGL_OFFSET_9 (0x0090 | I2OVERSION) #define SGL_OFFSET_10 (0x00A0 | I2OVERSION) - -#define TRL_OFFSET_5 (0x0050 | I2OVERSION) -#define TRL_OFFSET_6 (0x0060 | I2OVERSION) +#define SGL_OFFSET_11 (0x00B0 | I2OVERSION) +#define SGL_OFFSET_12 (0x00C0 | I2OVERSION) +#define SGL_OFFSET(x) (((x)<<4) | I2OVERSION) /* Transaction Reply Lists (TRL) Control Word structure */ #define TRL_SINGLE_FIXED_LENGTH 0x00 @@ -1130,7 +1147,6 @@ extern void i2o_debug_state(struct i2o_controller *c); #define HOST_TID 1 #define MSG_FRAME_SIZE 128 /* i2o_scsi assumes >= 32 */ -#define REPLY_FRAME_SIZE 17 #define SG_TABLESIZE 30 #define NMBR_MSG_FRAMES 128 @@ -1155,11 +1171,10 @@ extern void i2o_debug_state(struct i2o_controller *c); #define I2O_HRT_GET_TRIES 3 #define I2O_LCT_GET_TRIES 3 -/* request queue sizes */ +/* defines for max_sectors and max_phys_segments */ #define I2O_MAX_SECTORS 1024 +#define I2O_MAX_SECTORS_LIMITED 256 #define I2O_MAX_PHYS_SEGMENTS MAX_PHYS_SEGMENTS -#define I2O_REQ_MEMPOOL_SIZE 32 - #endif /* __KERNEL__ */ #endif /* _I2O_H */ diff --git a/include/scsi/sg_request.h b/include/scsi/sg_request.h new file mode 100644 index 000000000000..57ff525bdd3b --- /dev/null +++ b/include/scsi/sg_request.h @@ -0,0 +1,26 @@ +typedef struct scsi_request Scsi_Request; + +static Scsi_Request *dummy_cmdp; /* only used for sizeof */ + +typedef struct sg_scatter_hold { /* holding area for scsi scatter gather info */ + unsigned short k_use_sg; /* Count of kernel scatter-gather pieces */ + unsigned short sglist_len; /* size of malloc'd scatter-gather list ++ */ + unsigned bufflen; /* Size of (aggregate) data buffer */ + unsigned b_malloc_len; /* actual len malloc'ed in buffer */ + void *buffer; /* Data buffer or scatter list (k_use_sg>0) */ + char dio_in_use; /* 0->indirect IO (or mmap), 1->dio */ + unsigned char cmd_opcode; /* first byte of command */ +} Sg_scatter_hold; + +typedef struct sg_request { /* SG_MAX_QUEUE requests outstanding per file */ + Scsi_Request *my_cmdp; /* != 0 when request with lower levels */ + struct sg_request *nextrp; /* NULL -> tail request (slist) */ + struct sg_fd *parentfp; /* NULL -> not in use */ + Sg_scatter_hold data; /* hold buffer, perhaps scatter list */ + sg_io_hdr_t header; /* scsi command+info, see */ + unsigned char sense_b[sizeof (dummy_cmdp->sr_sense_buffer)]; + char res_used; /* 1 -> using reserve buffer, 0 -> not ... */ + char orphan; /* 1 -> drop on sight, 0 -> normal */ + char sg_io_owned; /* 1 -> packet belongs to SG_IO */ + volatile char done; /* 0->before bh, 1->before read, 2->read */ +} Sg_request; -- cgit v1.2.3 From 9e87545f06930c1d294423a8091d1077e7444a47 Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:21 -0700 Subject: [PATCH] I2O: second code cleanup of sparse warnings and unneeded syncronization Changes: - Added header "core.h" for i2o_core.ko internal definitions - More sparse fixes - Changed display of TID's in sysfs attributes from XXX to 0xXXX - Use the right functions for accessing I/O and normal memory - Removed error handling of SCSI device errors and let the SCSI layer take care of it - Added new device / removed device handling to SCSI-OSM - Make status access volatile - Cleaned up activation of I2O controller - Removed unnecessary wmb() and rmb() calls - Use own struct i2o_io for I/O memory instead of struct i2o_dma Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/i2o/core.h | 55 +++++++++++ drivers/message/i2o/debug.c | 3 - drivers/message/i2o/device.c | 22 +++-- drivers/message/i2o/driver.c | 24 ++--- drivers/message/i2o/exec-osm.c | 27 +++--- drivers/message/i2o/i2o_block.c | 4 +- drivers/message/i2o/i2o_config.c | 8 +- drivers/message/i2o/i2o_scsi.c | 202 +++++++++++++-------------------------- drivers/message/i2o/iop.c | 128 +++++++++++++------------ drivers/message/i2o/pci.c | 21 +--- include/linux/i2o-dev.h | 23 +++-- include/linux/i2o.h | 116 ++++++---------------- 12 files changed, 275 insertions(+), 358 deletions(-) create mode 100644 drivers/message/i2o/core.h (limited to 'include/linux') diff --git a/drivers/message/i2o/core.h b/drivers/message/i2o/core.h new file mode 100644 index 000000000000..49851cccc48d --- /dev/null +++ b/drivers/message/i2o/core.h @@ -0,0 +1,55 @@ +/* + * I2O core internal declarations + * + * Copyright (C) 2005 Markus Lidel + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * Fixes/additions: + * Markus Lidel + * initial version. + */ + +/* Exec-OSM */ +extern struct bus_type i2o_bus_type; + +extern struct i2o_driver i2o_exec_driver; +extern int i2o_exec_lct_get(struct i2o_controller *); + +extern int __init i2o_exec_init(void); +extern void __exit i2o_exec_exit(void); + +/* driver */ +extern int i2o_driver_dispatch(struct i2o_controller *, u32); + +extern int __init i2o_driver_init(void); +extern void __exit i2o_driver_exit(void); + +/* PCI */ +extern int __init i2o_pci_init(void); +extern void __exit i2o_pci_exit(void); + +/* device */ +extern void i2o_device_remove(struct i2o_device *); +extern int i2o_device_parse_lct(struct i2o_controller *); + +extern int i2o_device_init(void); +extern void i2o_device_exit(void); + +/* IOP */ +extern struct i2o_controller *i2o_iop_alloc(void); +extern void i2o_iop_free(struct i2o_controller *); + +extern int i2o_iop_add(struct i2o_controller *); +extern void i2o_iop_remove(struct i2o_controller *); + +/* control registers relative to c->base */ +#define I2O_IRQ_STATUS 0x30 +#define I2O_IRQ_MASK 0x34 +#define I2O_IN_PORT 0x40 +#define I2O_OUT_PORT 0x44 + +#define I2O_IRQ_OUTBOUND_POST 0x00000008 diff --git a/drivers/message/i2o/debug.c b/drivers/message/i2o/debug.c index 2a5d478fc60e..018ca887ca85 100644 --- a/drivers/message/i2o/debug.c +++ b/drivers/message/i2o/debug.c @@ -4,8 +4,6 @@ #include #include -extern struct i2o_driver **i2o_drivers; -extern unsigned int i2o_max_drivers; static void i2o_report_util_cmd(u8 cmd); static void i2o_report_exec_cmd(u8 cmd); static void i2o_report_fail_status(u8 req_status, u32 * msg); @@ -23,7 +21,6 @@ void i2o_report_status(const char *severity, const char *str, u8 cmd = (msg[1] >> 24) & 0xFF; u8 req_status = (msg[4] >> 24) & 0xFF; u16 detailed_status = msg[4] & 0xFFFF; - //struct i2o_driver *h = i2o_drivers[msg[2] & (i2o_max_drivers-1)]; if (cmd == I2O_CMD_UTIL_EVT_REGISTER) return; // No status in this reply diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c index f1b7eb63d54b..0ee342ea29bc 100644 --- a/drivers/message/i2o/device.c +++ b/drivers/message/i2o/device.c @@ -16,9 +16,7 @@ #include #include #include - -/* Exec OSM functions */ -extern struct bus_type i2o_bus_type; +#include "core.h" /** * i2o_device_issue_claim - claim or release a device @@ -293,12 +291,12 @@ int i2o_device_parse_lct(struct i2o_controller *c) } if (lct->table_size * 4 > c->dlct.len) { - memcpy_fromio(c->lct, c->dlct.virt, c->dlct.len); + memcpy(c->lct, c->dlct.virt, c->dlct.len); up(&c->lct_lock); return -EAGAIN; } - memcpy_fromio(c->lct, c->dlct.virt, lct->table_size * 4); + memcpy(c->lct, c->dlct.virt, lct->table_size * 4); lct = c->lct; @@ -353,7 +351,7 @@ static ssize_t i2o_device_class_show_class_id(struct class_device *cd, { struct i2o_device *dev = to_i2o_device(cd->dev); - sprintf(buf, "%03x\n", dev->lct_data.class_id); + sprintf(buf, "0x%03x\n", dev->lct_data.class_id); return strlen(buf) + 1; }; @@ -368,7 +366,7 @@ static ssize_t i2o_device_class_show_tid(struct class_device *cd, char *buf) { struct i2o_device *dev = to_i2o_device(cd->dev); - sprintf(buf, "%03x\n", dev->lct_data.tid); + sprintf(buf, "0x%03x\n", dev->lct_data.tid); return strlen(buf) + 1; }; @@ -490,7 +488,7 @@ static int i2o_parm_issue(struct i2o_device *i2o_dev, int cmd, void *oplist, if (rc == -ETIMEDOUT) return rc; - memcpy_fromio(reslist, res.virt, res.len); + memcpy(reslist, res.virt, res.len); i2o_dma_free(dev, &res); /* Query failed */ @@ -532,17 +530,23 @@ int i2o_parm_field_get(struct i2o_device *i2o_dev, int group, int field, void *buf, int buflen) { u16 opblk[] = { 1, 0, I2O_PARAMS_FIELD_GET, group, 1, field }; - u8 resblk[8 + buflen]; /* 8 bytes for header */ + u8 *resblk; /* 8 bytes for header */ int size; if (field == -1) /* whole group */ opblk[4] = -1; + resblk = kmalloc(buflen + 8, GFP_KERNEL | GFP_ATOMIC); + if (!resblk) + return -ENOMEM; + size = i2o_parm_issue(i2o_dev, I2O_CMD_UTIL_PARAMS_GET, opblk, sizeof(opblk), resblk, buflen + 8); memcpy(buf, resblk + 8, buflen); /* cut off header */ + kfree(resblk); + if (size > buflen) return buflen; diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c index 393be8e2914c..c32f9dbc5744 100644 --- a/drivers/message/i2o/driver.c +++ b/drivers/message/i2o/driver.c @@ -17,11 +17,12 @@ #include #include #include +#include "core.h" #define OSM_NAME "i2o" /* max_drivers - Maximum I2O drivers (OSMs) which could be registered */ -unsigned int i2o_max_drivers = I2O_MAX_DRIVERS; +static unsigned int i2o_max_drivers = I2O_MAX_DRIVERS; module_param_named(max_drivers, i2o_max_drivers, uint, 0); MODULE_PARM_DESC(max_drivers, "maximum number of OSM's to support"); @@ -179,15 +180,10 @@ void i2o_driver_unregister(struct i2o_driver *drv) int i2o_driver_dispatch(struct i2o_controller *c, u32 m) { struct i2o_driver *drv; - struct i2o_message __iomem *msg = i2o_msg_out_to_virt(c, m); - u32 context; + struct i2o_message *msg = i2o_msg_out_to_virt(c, m); + u32 context = le32_to_cpu(msg->u.s.icntxt); unsigned long flags; - if(unlikely(!msg)) - return -EIO; - - context = readl(&msg->u.s.icntxt); - if (unlikely(context >= i2o_max_drivers)) { osm_warn("%s: Spurious reply to unknown driver %d\n", c->name, context); @@ -204,11 +200,11 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m) return -EIO; } - if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_UTIL_EVT_REGISTER) { + if ((le32_to_cpu(msg->u.head[1]) >> 24) == I2O_CMD_UTIL_EVT_REGISTER) { struct i2o_device *dev, *tmp; struct i2o_event *evt; u16 size; - u16 tid = readl(&msg->u.head[1]) & 0xfff; + u16 tid = le32_to_cpu(msg->u.head[1]) & 0xfff; osm_debug("event received from device %d\n", tid); @@ -216,16 +212,16 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m) return -EIO; /* cut of header from message size (in 32-bit words) */ - size = (readl(&msg->u.head[0]) >> 16) - 5; + size = (le32_to_cpu(msg->u.head[0]) >> 16) - 5; evt = kmalloc(size * 4 + sizeof(*evt), GFP_ATOMIC | __GFP_ZERO); if (!evt) return -ENOMEM; evt->size = size; - evt->tcntxt = readl(&msg->u.s.tcntxt); - evt->event_indicator = readl(&msg->body[0]); - memcpy_fromio(&evt->tcntxt, &msg->u.s.tcntxt, size * 4); + evt->tcntxt = le32_to_cpu(msg->u.s.tcntxt); + evt->event_indicator = le32_to_cpu(msg->body[0]); + memcpy(&evt->tcntxt, &msg->u.s.tcntxt, size * 4); list_for_each_entry_safe(dev, tmp, &c->devices, list) if (dev->lct_data.tid == tid) { diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index 0160221c802a..ffe0cecfa060 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -30,6 +30,7 @@ #include #include #include +#include "core.h" #define OSM_NAME "exec-osm" @@ -37,9 +38,6 @@ struct i2o_driver i2o_exec_driver; static int i2o_exec_lct_notify(struct i2o_controller *c, u32 change_ind); -/* Module internal functions from other sources */ -extern int i2o_device_parse_lct(struct i2o_controller *); - /* global wait list for POST WAIT */ static LIST_HEAD(i2o_exec_wait_list); @@ -50,7 +48,7 @@ struct i2o_exec_wait { u32 tcntxt; /* transaction context from reply */ int complete; /* 1 if reply received otherwise 0 */ u32 m; /* message id */ - struct i2o_message __iomem *msg; /* pointer to the reply message */ + struct i2o_message *msg; /* pointer to the reply message */ struct list_head list; /* node in global wait list */ }; @@ -162,7 +160,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long barrier(); if (wait->complete) { - rc = readl(&wait->msg->body[0]) >> 24; + rc = le32_to_cpu(wait->msg->body[0]) >> 24; i2o_flush_reply(c, wait->m); i2o_exec_wait_free(wait); } else { @@ -202,8 +200,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long * message must also be given back to the controller. */ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, - struct i2o_message __iomem *msg, - u32 context) + struct i2o_message *msg, u32 context) { struct i2o_exec_wait *wait, *tmp; unsigned long flags; @@ -378,11 +375,11 @@ static void i2o_exec_lct_modified(struct i2o_controller *c) * code on failure and if the reply should be flushed. */ static int i2o_exec_reply(struct i2o_controller *c, u32 m, - struct i2o_message __iomem *msg) + struct i2o_message *msg) { u32 context; - if (readl(&msg->u.head[0]) & MSG_FAIL) { + if (le32_to_cpu(msg->u.head[0]) & MSG_FAIL) { /* * If Fail bit is set we must take the transaction context of * the preserved message to find the right request again. @@ -390,7 +387,7 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m, struct i2o_message __iomem *pmsg; u32 pm; - pm = readl(&msg->body[3]); + pm = le32_to_cpu(msg->body[3]); pmsg = i2o_msg_in_to_virt(c, pm); @@ -401,12 +398,12 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m, /* Release the preserved msg */ i2o_msg_nop(c, pm); } else - context = readl(&msg->u.s.tcntxt); + context = le32_to_cpu(msg->u.s.tcntxt); if (context & 0x80000000) return i2o_msg_post_wait_complete(c, m, msg, context); - if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_LCT_NOTIFY) { + if ((le32_to_cpu(msg->u.head[1]) >> 24) == I2O_CMD_LCT_NOTIFY) { struct work_struct *work; pr_debug("%s: LCT notify received\n", c->name); @@ -442,9 +439,9 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m, */ static void i2o_exec_event(struct i2o_event *evt) { - if(likely(evt->i2o_dev)) - osm_info("Event received from device: %d\n", - evt->i2o_dev->lct_data.tid); + if (likely(evt->i2o_dev)) + osm_debug("Event received from device: %d\n", + evt->i2o_dev->lct_data.tid); kfree(evt); }; diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c index 1dd2b9dad50e..28b3918dbc16 100644 --- a/drivers/message/i2o/i2o_block.c +++ b/drivers/message/i2o/i2o_block.c @@ -62,7 +62,7 @@ #include "i2o_block.h" #define OSM_NAME "block-osm" -#define OSM_VERSION "$Rev$" +#define OSM_VERSION "1.287" #define OSM_DESCRIPTION "I2O Block Device OSM" static struct i2o_driver i2o_block_driver; @@ -537,7 +537,7 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m, static void i2o_block_event(struct i2o_event *evt) { - osm_info("event received\n"); + osm_debug("event received\n"); kfree(evt); }; diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c index 7636833b4623..8160a1f6c73a 100644 --- a/drivers/message/i2o/i2o_config.c +++ b/drivers/message/i2o/i2o_config.c @@ -36,6 +36,8 @@ #include +#define SG_TABLESIZE 30 + extern int i2o_parm_issue(struct i2o_device *, int, void *, int, void *, int); static int i2o_cfg_ioctl(struct inode *inode, struct file *fp, unsigned int cmd, @@ -663,7 +665,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar goto sg_list_cleanup; if (sg_offset) { - u32 msg[MSG_FRAME_SIZE]; + u32 msg[I2O_OUTBOUND_MSG_FRAME_SIZE]; /* Copy back the Scatter Gather buffers back to user space */ u32 j; // TODO 64bit fix @@ -671,7 +673,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar int sg_size; // re-acquire the original message to handle correctly the sg copy operation - memset(&msg, 0, MSG_FRAME_SIZE * 4); + memset(&msg, 0, I2O_OUTBOUND_MSG_FRAME_SIZE * 4); // get user msg size in u32s if (get_user(size, &user_msg[0])) { rcode = -EFAULT; @@ -902,7 +904,7 @@ static int i2o_cfg_passthru(unsigned long arg) int sg_size; // re-acquire the original message to handle correctly the sg copy operation - memset(&msg, 0, MSG_FRAME_SIZE * 4); + memset(&msg, 0, I2O_OUTBOUND_MSG_FRAME_SIZE * 4); // get user msg size in u32s if (get_user(size, &user_msg[0])) { rcode = -EFAULT; diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c index fef53b509a61..9f1744c3933b 100644 --- a/drivers/message/i2o/i2o_scsi.c +++ b/drivers/message/i2o/i2o_scsi.c @@ -40,7 +40,6 @@ * Fix the resource management problems. */ -#define DEBUG 1 #include #include #include @@ -338,162 +337,89 @@ static int i2o_scsi_reply(struct i2o_controller *c, u32 m, struct i2o_message *msg) { struct scsi_cmnd *cmd; + u32 error; struct device *dev; - u8 as, ds, st; cmd = i2o_cntxt_list_get(c, le32_to_cpu(msg->u.s.tcntxt)); - - if (msg->u.head[0] & (1 << 13)) { - struct i2o_message __iomem *pmsg; /* preserved message */ - u32 pm; - int err = DID_ERROR; - - pm = le32_to_cpu(msg->body[3]); - - pmsg = i2o_msg_in_to_virt(c, pm); - - osm_err("IOP fail.\n"); - osm_err("From %d To %d Cmd %d.\n", - (msg->u.head[1] >> 12) & 0xFFF, - msg->u.head[1] & 0xFFF, msg->u.head[1] >> 24); - osm_err("Failure Code %d.\n", msg->body[0] >> 24); - if (msg->body[0] & (1 << 16)) - osm_err("Format error.\n"); - if (msg->body[0] & (1 << 17)) - osm_err("Path error.\n"); - if (msg->body[0] & (1 << 18)) - osm_err("Path State.\n"); - if (msg->body[0] & (1 << 18)) - { - osm_err("Congestion.\n"); - err = DID_BUS_BUSY; - } - - osm_debug("Failing message is %p.\n", pmsg); - - cmd = i2o_cntxt_list_get(c, readl(&pmsg->u.s.tcntxt)); - if (!cmd) - return 1; - - cmd->result = err << 16; - cmd->scsi_done(cmd); - - /* Now flush the message by making it a NOP */ - i2o_msg_nop(c, pm); - - return 1; + if (unlikely(!cmd)) { + osm_err("NULL reply received!\n"); + return -1; } /* * Low byte is device status, next is adapter status, * (then one byte reserved), then request status. */ - ds = (u8) le32_to_cpu(msg->body[0]); - as = (u8) (le32_to_cpu(msg->body[0]) >> 8); - st = (u8) (le32_to_cpu(msg->body[0]) >> 24); + error = le32_to_cpu(msg->body[0]); + osm_debug("Completed %ld\n", cmd->serial_number); + + cmd->result = error & 0xff; /* - * Is this a control request coming back - eg an abort ? + * if DeviceStatus is not SCSI_SUCCESS copy over the sense data and let + * the SCSI layer handle the error */ + if (cmd->result) + memcpy(cmd->sense_buffer, &msg->body[3], + min(sizeof(cmd->sense_buffer), (size_t) 40)); - if (!cmd) { - if (st) - osm_warn("SCSI abort: %08X", le32_to_cpu(msg->body[0])); - osm_info("SCSI abort completed.\n"); - return -EFAULT; - } + /* only output error code if AdapterStatus is not HBA_SUCCESS */ + if ((error >> 8) & 0xff) + osm_err("SCSI error %08x\n", error); - osm_debug("Completed %ld\n", cmd->serial_number); + dev = &c->pdev->dev; + if (cmd->use_sg) + dma_unmap_sg(dev, cmd->request_buffer, cmd->use_sg, + cmd->sc_data_direction); + else if (cmd->SCp.dma_handle) + dma_unmap_single(dev, cmd->SCp.dma_handle, cmd->request_bufflen, + cmd->sc_data_direction); - if (st) { - u32 count, error; - /* An error has occurred */ - - switch (st) { - case 0x06: - count = le32_to_cpu(msg->body[1]); - if (count < cmd->underflow) { - int i; - - osm_err("SCSI underflow 0x%08X 0x%08X\n", count, - cmd->underflow); - osm_debug("Cmd: "); - for (i = 0; i < 15; i++) - pr_debug("%02X ", cmd->cmnd[i]); - pr_debug(".\n"); - cmd->result = (DID_ERROR << 16); - } - break; + cmd->scsi_done(cmd); - default: - error = le32_to_cpu(msg->body[0]); - - osm_err("SCSI error %08x\n", error); - - if ((error & 0xff) == 0x02 /*CHECK_CONDITION */ ) { - int i; - u32 len = sizeof(cmd->sense_buffer); - len = (len > 40) ? 40 : len; - // Copy over the sense data - memcpy(cmd->sense_buffer, (void *)&msg->body[3], - len); - for (i = 0; i <= len; i++) - osm_info("%02x\n", - cmd->sense_buffer[i]); - if (cmd->sense_buffer[0] == 0x70 - && cmd->sense_buffer[2] == DATA_PROTECT) { - /* This is to handle an array failed */ - cmd->result = (DID_TIME_OUT << 16); - printk(KERN_WARNING "%s: SCSI Data " - "Protect-Device (%d,%d,%d) " - "hba_status=0x%x, dev_status=" - "0x%x, cmd=0x%x\n", c->name, - (u32) cmd->device->channel, - (u32) cmd->device->id, - (u32) cmd->device->lun, - (error >> 8) & 0xff, - error & 0xff, cmd->cmnd[0]); - } else - cmd->result = (DID_ERROR << 16); - - break; - } - - switch (as) { - case 0x0E: - /* SCSI Reset */ - cmd->result = DID_RESET << 16; - break; - - case 0x0F: - cmd->result = DID_PARITY << 16; - break; - - default: - cmd->result = DID_ERROR << 16; - break; - } + return 1; +}; - break; - } +/** + * i2o_scsi_notify_device_add - Retrieve notifications of added devices + * @i2o_dev: the I2O device which was added + * + * If a I2O device is added we catch the notification, because I2O classes + * other then SCSI peripheral will not be received through + * i2o_scsi_probe(). + */ +static void i2o_scsi_notify_device_add(struct i2o_device *i2o_dev) +{ + switch (i2o_dev->lct_data.class_id) { + case I2O_CLASS_EXECUTIVE: + case I2O_CLASS_RANDOM_BLOCK_STORAGE: + i2o_scsi_probe(&i2o_dev->device); + break; - cmd->scsi_done(cmd); - return 1; + default: + break; } +}; - cmd->result = DID_OK << 16 | ds; - - dev = &c->pdev->dev; - if (cmd->use_sg) - dma_unmap_sg(dev, (struct scatterlist *)cmd->buffer, - cmd->use_sg, cmd->sc_data_direction); - else if (cmd->request_bufflen) - dma_unmap_single(dev, (dma_addr_t) ((long)cmd->SCp.ptr), - cmd->request_bufflen, cmd->sc_data_direction); - - cmd->scsi_done(cmd); +/** + * i2o_scsi_notify_device_remove - Retrieve notifications of removed + * devices + * @i2o_dev: the I2O device which was removed + * + * If a I2O device is removed, we catch the notification to remove the + * corresponding SCSI device. + */ +static void i2o_scsi_notify_device_remove(struct i2o_device *i2o_dev) +{ + switch (i2o_dev->lct_data.class_id) { + case I2O_CLASS_EXECUTIVE: + case I2O_CLASS_RANDOM_BLOCK_STORAGE: + i2o_scsi_remove(&i2o_dev->device); + break; - return 1; + default: + break; + } }; /** @@ -554,6 +480,8 @@ static struct i2o_driver i2o_scsi_driver = { .name = OSM_NAME, .reply = i2o_scsi_reply, .classes = i2o_scsi_class_id, + .notify_device_add = i2o_scsi_notify_device_add, + .notify_device_remove = i2o_scsi_notify_device_remove, .notify_controller_add = i2o_scsi_notify_controller_add, .notify_controller_remove = i2o_scsi_notify_controller_remove, .driver = { @@ -712,7 +640,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt, */ /* Attach tags to the devices */ - /* + /* FIXME: implement if(SCpnt->device->tagged_supported) { if(SCpnt->tag == HEAD_OF_QUEUE_TAG) scsi_flags |= 0x01000000; diff --git a/drivers/message/i2o/iop.c b/drivers/message/i2o/iop.c index 40312053b38d..c32022bc2a21 100644 --- a/drivers/message/i2o/iop.c +++ b/drivers/message/i2o/iop.c @@ -28,8 +28,10 @@ #include #include #include +#include "core.h" -#define OSM_VERSION "$Rev$" +#define OSM_NAME "i2o" +#define OSM_VERSION "1.288" #define OSM_DESCRIPTION "I2O subsystem" /* global I2O controller list */ @@ -43,20 +45,6 @@ static struct i2o_dma i2o_systab; static int i2o_hrt_get(struct i2o_controller *c); -/* Module internal functions from other sources */ -extern struct i2o_driver i2o_exec_driver; -extern int i2o_exec_lct_get(struct i2o_controller *); -extern void i2o_device_remove(struct i2o_device *); - -extern int __init i2o_driver_init(void); -extern void __exit i2o_driver_exit(void); -extern int __init i2o_exec_init(void); -extern void __exit i2o_exec_exit(void); -extern int __init i2o_pci_init(void); -extern void __exit i2o_pci_exit(void); -extern int i2o_device_init(void); -extern void i2o_device_exit(void); - /** * i2o_msg_nop - Returns a message which is not used * @c: I2O controller from which the message was created @@ -92,16 +80,16 @@ void i2o_msg_nop(struct i2o_controller *c, u32 m) * address from the read port (see the i2o spec). If no message is * available returns I2O_QUEUE_EMPTY and msg is leaved untouched. */ -u32 i2o_msg_get_wait(struct i2o_controller *c, struct i2o_message __iomem **msg, - int wait) +u32 i2o_msg_get_wait(struct i2o_controller *c, + struct i2o_message __iomem ** msg, int wait) { unsigned long timeout = jiffies + wait * HZ; u32 m; while ((m = i2o_msg_get(c, msg)) == I2O_QUEUE_EMPTY) { if (time_after(jiffies, timeout)) { - pr_debug("%s: Timeout waiting for message frame.\n", - c->name); + osm_debug("%s: Timeout waiting for message frame.\n", + c->name); return I2O_QUEUE_EMPTY; } set_current_state(TASK_UNINTERRUPTIBLE); @@ -466,7 +454,7 @@ static int i2o_iop_clear(struct i2o_controller *c) */ static int i2o_iop_init_outbound_queue(struct i2o_controller *c) { - u8 *status = c->status.virt; + volatile u8 *status = c->status.virt; u32 m; struct i2o_message __iomem *msg; ulong timeout; @@ -474,21 +462,20 @@ static int i2o_iop_init_outbound_queue(struct i2o_controller *c) osm_debug("%s: Initializing Outbound Queue...\n", c->name); - memset(status, 0, 4); + memset(c->status.virt, 0, 4); m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET); if (m == I2O_QUEUE_EMPTY) return -ETIMEDOUT; - writel(EIGHT_WORD_MSG_SIZE | TRL_OFFSET_6, &msg->u.head[0]); + writel(EIGHT_WORD_MSG_SIZE | SGL_OFFSET_6, &msg->u.head[0]); writel(I2O_CMD_OUTBOUND_INIT << 24 | HOST_TID << 12 | ADAPTER_TID, &msg->u.head[1]); writel(i2o_exec_driver.context, &msg->u.s.icntxt); - writel(0x0106, &msg->u.s.tcntxt); /* FIXME: why 0x0106, maybe in - Spec? */ + writel(0x00000000, &msg->u.s.tcntxt); writel(PAGE_SIZE, &msg->body[0]); /* Outbound msg frame size in words and Initcode */ - writel(MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]); + writel(I2O_OUTBOUND_MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]); writel(0xd0000004, &msg->body[2]); writel(i2o_dma_low(c->status.phys), &msg->body[3]); writel(i2o_dma_high(c->status.phys), &msg->body[4]); @@ -503,17 +490,15 @@ static int i2o_iop_init_outbound_queue(struct i2o_controller *c) } set_current_state(TASK_UNINTERRUPTIBLE); schedule_timeout(1); - - rmb(); } m = c->out_queue.phys; /* Post frames */ - for (i = 0; i < NMBR_MSG_FRAMES; i++) { + for (i = 0; i < I2O_MAX_OUTBOUND_MSG_FRAMES; i++) { i2o_flush_reply(c, m); udelay(1); /* Promise */ - m += MSG_FRAME_SIZE * 4; + m += I2O_OUTBOUND_MSG_FRAME_SIZE * sizeof(u32); } return 0; @@ -530,20 +515,20 @@ static int i2o_iop_init_outbound_queue(struct i2o_controller *c) */ static int i2o_iop_reset(struct i2o_controller *c) { - u8 *status = c->status.virt; + volatile u8 *status = c->status.virt; struct i2o_message __iomem *msg; u32 m; unsigned long timeout; i2o_status_block *sb = c->status_block.virt; int rc = 0; - pr_debug("%s: Resetting controller\n", c->name); + osm_debug("%s: Resetting controller\n", c->name); m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET); if (m == I2O_QUEUE_EMPTY) return -ETIMEDOUT; - memset(status, 0, 8); + memset(c->status_block.virt, 0, 8); /* Quiesce all IOPs first */ i2o_iop_quiesce_all(); @@ -568,8 +553,6 @@ static int i2o_iop_reset(struct i2o_controller *c) set_current_state(TASK_UNINTERRUPTIBLE); schedule_timeout(1); - - rmb(); } switch (*status) { @@ -984,11 +967,11 @@ int i2o_status_get(struct i2o_controller *c) { struct i2o_message __iomem *msg; u32 m; - u8 *status_block; + volatile u8 *status_block; unsigned long timeout; status_block = (u8 *) c->status_block.virt; - memset(status_block, 0, sizeof(i2o_status_block)); + memset(c->status_block.virt, 0, sizeof(i2o_status_block)); m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET); if (m == I2O_QUEUE_EMPTY) @@ -1017,8 +1000,6 @@ int i2o_status_get(struct i2o_controller *c) set_current_state(TASK_UNINTERRUPTIBLE); schedule_timeout(1); - - rmb(); } #ifdef DEBUG @@ -1107,6 +1088,11 @@ static void i2o_iop_release(struct device *dev) i2o_iop_free(c); }; +/* I2O controller class */ +static struct class i2o_controller_class = { + .name = "i2o_controller", +}; + /** * i2o_iop_alloc - Allocate and initialize a i2o_controller struct * @@ -1136,8 +1122,14 @@ struct i2o_controller *i2o_iop_alloc(void) sprintf(c->name, "iop%d", c->unit); device_initialize(&c->device); + class_device_initialize(&c->classdev); + c->device.release = &i2o_iop_release; + c->classdev.class = &i2o_controller_class; + c->classdev.dev = &c->device; + snprintf(c->device.bus_id, BUS_ID_SIZE, "iop%d", c->unit); + snprintf(c->classdev.class_id, BUS_ID_SIZE, "iop%d", c->unit); #if BITS_PER_LONG == 64 spin_lock_init(&c->context_list_lock); @@ -1161,45 +1153,55 @@ int i2o_iop_add(struct i2o_controller *c) { int rc; - if((rc = device_add(&c->device))) { - printk(KERN_ERR "%s: could not register controller\n", c->name); + if ((rc = device_add(&c->device))) { + osm_err("%s: could not add controller\n", c->name); goto iop_reset; } - printk(KERN_INFO "%s: Activating I2O controller...\n", c->name); - printk(KERN_INFO "%s: This may take a few minutes if there are many " - "devices\n", c->name); + if ((rc = class_device_add(&c->classdev))) { + osm_err("%s: could not add controller class\n", c->name); + goto device_del; + } + + osm_info("%s: Activating I2O controller...\n", c->name); + osm_info("%s: This may take a few minutes if there are many devices\n", + c->name); if ((rc = i2o_iop_activate(c))) { - printk(KERN_ERR "%s: could not activate controller\n", - c->name); - goto iop_reset; + osm_err("%s: could not activate controller\n", c->name); + goto class_del; } - pr_debug("%s: building sys table...\n", c->name); + osm_debug("%s: building sys table...\n", c->name); if ((rc = i2o_systab_build())) - goto iop_reset; + goto class_del; - pr_debug("%s: online controller...\n", c->name); + osm_debug("%s: online controller...\n", c->name); if ((rc = i2o_iop_online(c))) - goto iop_reset; + goto class_del; - pr_debug("%s: getting LCT...\n", c->name); + osm_debug("%s: getting LCT...\n", c->name); if ((rc = i2o_exec_lct_get(c))) - goto iop_reset; + goto class_del; list_add(&c->list, &i2o_controllers); i2o_driver_notify_controller_add_all(c); - printk(KERN_INFO "%s: Controller added\n", c->name); + osm_info("%s: Controller added\n", c->name); return 0; -iop_reset: + class_del: + class_device_del(&c->classdev); + + device_del: + device_del(&c->device); + + iop_reset: i2o_iop_reset(c); return rc; @@ -1260,16 +1262,18 @@ static int __init i2o_iop_init(void) if (rc) goto exit; - rc = i2o_driver_init(); - if (rc) + if ((rc = class_register(&i2o_controller_class))) { + osm_err("can't register class i2o_controller\n"); goto device_exit; + } - rc = i2o_exec_init(); - if (rc) + if ((rc = i2o_driver_init())) + goto class_exit; + + if ((rc = i2o_exec_init())) goto driver_exit; - rc = i2o_pci_init(); - if (rc < 0) + if ((rc = i2o_pci_init())) goto exec_exit; return 0; @@ -1280,6 +1284,9 @@ static int __init i2o_iop_init(void) driver_exit: i2o_driver_exit(); + class_exit: + class_unregister(&i2o_controller_class); + device_exit: i2o_device_exit(); @@ -1297,6 +1304,7 @@ static void __exit i2o_iop_exit(void) i2o_pci_exit(); i2o_exec_exit(); i2o_driver_exit(); + class_unregister(&i2o_controller_class); i2o_device_exit(); }; diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c index 964fe481849e..442e34506b90 100644 --- a/drivers/message/i2o/pci.c +++ b/drivers/message/i2o/pci.c @@ -30,15 +30,7 @@ #include #include #include - -/* Module internal functions from other sources */ -extern struct i2o_controller *i2o_iop_alloc(void); -extern void i2o_iop_free(struct i2o_controller *); - -extern int i2o_iop_add(struct i2o_controller *); -extern void i2o_iop_remove(struct i2o_controller *); - -extern int i2o_driver_dispatch(struct i2o_controller *, u32); +#include "core.h" /* PCI device id table for all I2O controllers */ static struct pci_device_id __devinitdata i2o_pci_ids[] = { @@ -248,9 +240,7 @@ static int i2o_pci_irq_enable(struct i2o_controller *c) struct pci_dev *pdev = c->pdev; int rc; - wmb(); writel(0xffffffff, c->irq_mask); - wmb(); if (pdev->irq) { rc = request_irq(pdev->irq, i2o_pci_interrupt, SA_SHIRQ, @@ -263,7 +253,6 @@ static int i2o_pci_irq_enable(struct i2o_controller *c) } writel(0x00000000, c->irq_mask); - wmb(); printk(KERN_INFO "%s: Installed at IRQ %d\n", c->name, pdev->irq); @@ -278,9 +267,7 @@ static int i2o_pci_irq_enable(struct i2o_controller *c) */ static void i2o_pci_irq_disable(struct i2o_controller *c) { - wmb(); writel(0xffffffff, c->irq_mask); - wmb(); if (c->pdev->irq > 0) free_irq(c->pdev->irq, c); @@ -406,11 +393,11 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, if ((rc = i2o_iop_add(c))) goto uninstall; + get_device(&c->device); + if (i960) pci_write_config_word(i960, 0x42, 0x03ff); - get_device(&c->device); - return 0; uninstall: @@ -478,6 +465,4 @@ void __exit i2o_pci_exit(void) { pci_unregister_driver(&i2o_pci_driver); }; - -EXPORT_SYMBOL(i2o_dma_realloc); MODULE_DEVICE_TABLE(pci, i2o_pci_ids); diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h index d4a08d29e36d..36fd18cdad28 100644 --- a/include/linux/i2o-dev.h +++ b/include/linux/i2o-dev.h @@ -32,18 +32,6 @@ typedef unsigned int u32; #endif /* __KERNEL__ */ -/* - * Software module types - */ -#define I2O_SOFTWARE_MODULE_IRTOS 0x11 -#define I2O_SOFTWARE_MODULE_IOP_PRIVATE 0x22 -#define I2O_SOFTWARE_MODULE_IOP_CONFIG 0x23 - -/* - * Vendors - */ -#define I2O_VENDOR_DPT 0x001b - /* * I2O Control IOCTLs and structures */ @@ -414,6 +402,17 @@ typedef struct _i2o_status_block { #define ADAPTER_STATE_FAILED 0x10 #define ADAPTER_STATE_FAULTED 0x11 +/* + * Software module types + */ +#define I2O_SOFTWARE_MODULE_IRTOS 0x11 +#define I2O_SOFTWARE_MODULE_IOP_PRIVATE 0x22 +#define I2O_SOFTWARE_MODULE_IOP_CONFIG 0x23 + +/* + * Vendors + */ +#define I2O_VENDOR_DPT 0x001b /* * DPT / Adaptec specific values for i2o_sg_io_hdr flags. diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 2039a87c2b91..be937d0372a7 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -119,12 +119,21 @@ struct i2o_driver { }; /* - * Contains all information which are necessary for DMA operations + * Contains DMA mapped address information */ struct i2o_dma { void *virt; dma_addr_t phys; - u32 len; + size_t len; +}; + +/* + * Contains IO mapped address information + */ +struct i2o_io { + void __iomem *virt; + unsigned long phys; + unsigned long len; }; /* @@ -173,8 +182,8 @@ struct i2o_controller { struct semaphore lct_lock; /* Lock for LCT updates */ struct i2o_dma status_block; /* IOP status block */ - struct i2o_dma base; /* controller messaging unit */ - struct i2o_dma in_queue; /* inbound message queue Host->IOP */ + struct i2o_io base; /* controller messaging unit */ + struct i2o_io in_queue; /* inbound message queue Host->IOP */ struct i2o_dma out_queue; /* outbound message queue IOP->Host */ unsigned int battery:1; /* Has a battery backup */ @@ -185,6 +194,7 @@ struct i2o_controller { struct resource mem_resource; /* Mem resource allocated to the IOP */ struct device device; + struct class_device classdev; /* I2O controller class */ struct i2o_device *exec; /* Executive */ #if BITS_PER_LONG == 64 spinlock_t context_list_lock; /* lock for context_list */ @@ -235,9 +245,10 @@ struct i2o_sys_tbl { extern struct list_head i2o_controllers; /* Message functions */ -static inline u32 i2o_msg_get(struct i2o_controller *, struct i2o_message __iomem **); -extern u32 i2o_msg_get_wait(struct i2o_controller *, struct i2o_message __iomem **, - int); +static inline u32 i2o_msg_get(struct i2o_controller *, + struct i2o_message __iomem **); +extern u32 i2o_msg_get_wait(struct i2o_controller *, + struct i2o_message __iomem **, int); static inline void i2o_msg_post(struct i2o_controller *, u32); static inline int i2o_msg_post_wait(struct i2o_controller *, u32, unsigned long); @@ -638,14 +649,12 @@ extern int i2o_exec_lct_get(struct i2o_controller *); * available returns I2O_QUEUE_EMPTY and msg is leaved untouched. */ static inline u32 i2o_msg_get(struct i2o_controller *c, - struct i2o_message __iomem **msg) + struct i2o_message __iomem ** msg) { u32 m = readl(c->in_port); - if (m != I2O_QUEUE_EMPTY) { + if (m != I2O_QUEUE_EMPTY) *msg = c->in_queue.virt + m; - rmb(); - } return m; }; @@ -659,7 +668,6 @@ static inline u32 i2o_msg_get(struct i2o_controller *c, */ static inline void i2o_msg_post(struct i2o_controller *c, u32 m) { - wmb(); writel(m, c->in_port); }; @@ -706,14 +714,11 @@ static inline void i2o_flush_reply(struct i2o_controller *c, u32 m) * work for sender side messages as they are ioremap objects * provided by the I2O controller. */ -static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct - i2o_controller *c, - u32 m) +static inline struct i2o_message *i2o_msg_out_to_virt(struct i2o_controller *c, + u32 m) { - if (unlikely - (m < c->out_queue.phys - || m >= c->out_queue.phys + c->out_queue.len)) - return NULL; + BUG_ON(m < c->out_queue.phys + || m >= c->out_queue.phys + c->out_queue.len); return c->out_queue.virt + (m - c->out_queue.phys); }; @@ -729,69 +734,13 @@ static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct * work for receive side messages as they are kmalloc objects * in a different pool. */ -static inline struct i2o_message __iomem *i2o_msg_in_to_virt(struct i2o_controller *c, - u32 m) +static inline struct i2o_message __iomem *i2o_msg_in_to_virt(struct + i2o_controller *c, + u32 m) { return c->in_queue.virt + m; }; -/** - * i2o_dma_alloc - Allocate DMA memory - * @dev: struct device pointer to the PCI device of the I2O controller - * @addr: i2o_dma struct which should get the DMA buffer - * @len: length of the new DMA memory - * @gfp_mask: GFP mask - * - * Allocate a coherent DMA memory and write the pointers into addr. - * - * Returns 0 on success or -ENOMEM on failure. - */ -static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr, - size_t len, unsigned int gfp_mask) -{ - struct pci_dev *pdev = to_pci_dev(dev); - int dma_64 = 0; - - if ((sizeof(dma_addr_t) > 4) && (pdev->dma_mask == DMA_64BIT_MASK)) { - dma_64 = 1; - if(pci_set_dma_mask(pdev, DMA_32BIT_MASK)) - return -ENOMEM; - } - - addr->virt = dma_alloc_coherent(dev, len, &addr->phys, gfp_mask); - - if ((sizeof(dma_addr_t) > 4) && dma_64) - if(pci_set_dma_mask(pdev, DMA_64BIT_MASK)) - printk(KERN_WARNING "i2o: unable to set 64-bit DMA"); - - if (!addr->virt) - return -ENOMEM; - - memset(addr->virt, 0, len); - addr->len = len; - - return 0; -}; - -/** - * i2o_dma_free - Free DMA memory - * @dev: struct device pointer to the PCI device of the I2O controller - * @addr: i2o_dma struct which contains the DMA buffer - * - * Free a coherent DMA memory and set virtual address of addr to NULL. - */ -static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr) -{ - if (addr->virt) { - if (addr->phys) - dma_free_coherent(dev, addr->len, addr->virt, - addr->phys); - else - kfree(addr->virt); - addr->virt = NULL; - } -}; - /* * Endian handling wrapped into the macro - keeps the core code * cleaner. @@ -1141,16 +1090,13 @@ extern void i2o_debug_state(struct i2o_controller *c); #define ELEVEN_WORD_MSG_SIZE 0x000B0000 #define I2O_MESSAGE_SIZE(x) ((x)<<16) -/* Special TID Assignments */ - +/* special TID assignments */ #define ADAPTER_TID 0 #define HOST_TID 1 -#define MSG_FRAME_SIZE 128 /* i2o_scsi assumes >= 32 */ -#define SG_TABLESIZE 30 -#define NMBR_MSG_FRAMES 128 - -#define MSG_POOL_SIZE (MSG_FRAME_SIZE*NMBR_MSG_FRAMES*sizeof(u32)) +/* outbound queue defines */ +#define I2O_MAX_OUTBOUND_MSG_FRAMES 128 +#define I2O_OUTBOUND_MSG_FRAME_SIZE 128 /* in 32-bit words */ #define I2O_POST_WAIT_OK 0 #define I2O_POST_WAIT_TIMEOUT -ETIMEDOUT -- cgit v1.2.3 From f33213ecf49c98da4e85121b592c3bea8057c2e6 Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:23 -0700 Subject: [PATCH] I2O: Lindent run and replacement of printk through osm printing functions Lindent run and replaced printk() through the corresponding osm_*() function Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/i2o/Kconfig | 10 +-- drivers/message/i2o/device.c | 1 - drivers/message/i2o/driver.c | 3 +- drivers/message/i2o/exec-osm.c | 2 +- drivers/message/i2o/i2o_block.c | 3 +- drivers/message/i2o/i2o_block.h | 28 ++++----- drivers/message/i2o/i2o_config.c | 20 +++--- drivers/message/i2o/i2o_proc.c | 2 +- drivers/message/i2o/iop.c | 128 ++++++++++++++++++--------------------- drivers/message/i2o/pci.c | 5 +- include/linux/i2o.h | 8 +-- 11 files changed, 100 insertions(+), 110 deletions(-) (limited to 'include/linux') diff --git a/drivers/message/i2o/Kconfig b/drivers/message/i2o/Kconfig index 94b6d676c5cb..06e8eb19a05c 100644 --- a/drivers/message/i2o/Kconfig +++ b/drivers/message/i2o/Kconfig @@ -44,8 +44,8 @@ config I2O_EXT_ADAPTEC_DMA64 config I2O_CONFIG tristate "I2O Configuration support" - depends on PCI && I2O - help + depends on I2O + ---help--- Say Y for support of the configuration interface for the I2O adapters. If you have a RAID controller from Adaptec and you want to use the raidutils to manage your RAID array, you have to say Y here. @@ -74,7 +74,7 @@ config I2O_BUS config I2O_BLOCK tristate "I2O Block OSM" depends on I2O - help + ---help--- Include support for the I2O Block OSM. The Block OSM presents disk and other structured block devices to the operating system. If you are using an RAID controller, you could access the array only by @@ -87,7 +87,7 @@ config I2O_BLOCK config I2O_SCSI tristate "I2O SCSI OSM" depends on I2O && SCSI - help + ---help--- Allows direct SCSI access to SCSI devices on a SCSI or FibreChannel I2O controller. You can use both the SCSI and Block OSM together if you wish. To access a RAID array, you must use the Block OSM driver. @@ -99,7 +99,7 @@ config I2O_SCSI config I2O_PROC tristate "I2O /proc support" depends on I2O - help + ---help--- If you say Y here and to "/proc file system support", you will be able to read I2O related information from the virtual directory /proc/i2o. diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c index 0ee342ea29bc..d8d6e89a91cc 100644 --- a/drivers/message/i2o/device.c +++ b/drivers/message/i2o/device.c @@ -443,7 +443,6 @@ static struct class_interface i2o_device_class_interface = { * Note that the minimum sized reslist is 8 bytes and contains * ResultCount, ErrorInfoSize, BlockStatus and BlockSize. */ - static int i2o_parm_issue(struct i2o_device *i2o_dev, int cmd, void *oplist, int oplen, void *reslist, int reslen) { diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c index c32f9dbc5744..739bfdef0c6d 100644 --- a/drivers/message/i2o/driver.c +++ b/drivers/message/i2o/driver.c @@ -117,10 +117,9 @@ int i2o_driver_register(struct i2o_driver *drv) i2o_driver_notify_controller_add(drv, c); list_for_each_entry(i2o_dev, &c->devices, list) - i2o_driver_notify_device_add(drv, i2o_dev); + i2o_driver_notify_device_add(drv, i2o_dev); } - rc = driver_register(&drv->driver); if (rc) destroy_workqueue(drv->event_queue); diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index ffe0cecfa060..1b7389876e70 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -152,7 +152,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long list_add(&wait->list, &i2o_exec_wait_list); wait_event_interruptible_timeout(wq, wait->complete, - timeout * HZ); + timeout * HZ); wait->wq = NULL; } diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c index 28b3918dbc16..f283b5bafdd3 100644 --- a/drivers/message/i2o/i2o_block.c +++ b/drivers/message/i2o/i2o_block.c @@ -940,7 +940,6 @@ static void i2o_block_request_fn(struct request_queue *q) INIT_WORK(&dreq->work, i2o_block_delayed_request_fn, dreq); - osm_info("transfer error\n"); if (!queue_delayed_work(i2o_block_driver.event_queue, &dreq->work, I2O_BLOCK_RETRY_TIME)) @@ -1042,8 +1041,8 @@ static struct i2o_block_device *i2o_block_device_alloc(void) static int i2o_block_probe(struct device *dev) { struct i2o_device *i2o_dev = to_i2o_device(dev); - struct i2o_block_device *i2o_blk_dev; struct i2o_controller *c = i2o_dev->iop; + struct i2o_block_device *i2o_blk_dev; struct gendisk *gd; struct request_queue *queue; static int unit = 0; diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h index e45cc40ce384..4fdaa5bda412 100644 --- a/drivers/message/i2o/i2o_block.h +++ b/drivers/message/i2o/i2o_block.h @@ -64,40 +64,38 @@ /* I2O Block OSM mempool struct */ struct i2o_block_mempool { - kmem_cache_t *slab; - mempool_t *pool; + kmem_cache_t *slab; + mempool_t *pool; }; /* I2O Block device descriptor */ struct i2o_block_device { struct i2o_device *i2o_dev; /* pointer to I2O device */ struct gendisk *gd; - spinlock_t lock; /* queue lock */ + spinlock_t lock; /* queue lock */ struct list_head open_queue; /* list of transfered, but unfinished requests */ unsigned int open_queue_depth; /* number of requests in the queue */ - int rcache; /* read cache flags */ - int wcache; /* write cache flags */ + int rcache; /* read cache flags */ + int wcache; /* write cache flags */ int flags; - u16 power; /* power state */ - int media_change_flag; /* media changed flag */ + u16 power; /* power state */ + int media_change_flag; /* media changed flag */ }; /* I2O Block device request */ -struct i2o_block_request -{ +struct i2o_block_request { struct list_head queue; - struct request *req; /* corresponding request */ + struct request *req; /* corresponding request */ struct i2o_block_device *i2o_blk_dev; /* I2O block device */ - struct device *dev; /* device used for DMA */ - int sg_nents; /* number of SG elements */ - struct scatterlist sg_table[I2O_MAX_PHYS_SEGMENTS]; /* SG table */ + struct device *dev; /* device used for DMA */ + int sg_nents; /* number of SG elements */ + struct scatterlist sg_table[I2O_MAX_PHYS_SEGMENTS]; /* SG table */ }; /* I2O Block device delayed request */ -struct i2o_block_delayed_request -{ +struct i2o_block_delayed_request { struct work_struct work; struct request_queue *queue; }; diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c index 8160a1f6c73a..8ebc86ff1002 100644 --- a/drivers/message/i2o/i2o_config.c +++ b/drivers/message/i2o/i2o_config.c @@ -368,9 +368,9 @@ static int i2o_cfg_swul(unsigned long arg) i2o_dma_free(&c->pdev->dev, &buffer); -return_ret: + return_ret: return ret; -return_fault: + return_fault: ret = -EFAULT; goto return_ret; }; @@ -519,7 +519,8 @@ static int i2o_cfg_evt_get(unsigned long arg, struct file *fp) #ifdef CONFIG_I2O_EXT_ADAPTEC #ifdef CONFIG_COMPAT -static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long arg) +static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, + unsigned long arg) { struct i2o_cmd_passthru32 __user *cmd; struct i2o_controller *c; @@ -646,8 +647,9 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar flag_count & 0x04000000 /*I2O_SGL_FLAGS_DIR */ ) { // TODO 64bit fix if (copy_from_user - (p->virt, (void __user *)(unsigned long)sg[i].addr_bus, - sg_size)) { + (p->virt, + (void __user *)(unsigned long)sg[i]. + addr_bus, sg_size)) { printk(KERN_DEBUG "%s: Could not copy SG buf %d FROM user\n", c->name, i); @@ -738,11 +740,12 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar return rcode; } -static long i2o_cfg_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg) +static long i2o_cfg_compat_ioctl(struct file *file, unsigned cmd, + unsigned long arg) { int ret; - lock_kernel(); - switch (cmd) { + lock_kernel(); + switch (cmd) { case I2OGETIOPS: ret = i2o_cfg_ioctl(NULL, file, cmd, arg); break; @@ -1136,6 +1139,7 @@ static int __init i2o_config_old_init(void) osm_err("can't register device.\n"); return -EBUSY; } + return 0; } diff --git a/drivers/message/i2o/i2o_proc.c b/drivers/message/i2o/i2o_proc.c index e5b74452c495..d559a1758363 100644 --- a/drivers/message/i2o/i2o_proc.c +++ b/drivers/message/i2o/i2o_proc.c @@ -28,7 +28,7 @@ */ #define OSM_NAME "proc-osm" -#define OSM_VERSION "$Rev$" +#define OSM_VERSION "1.145" #define OSM_DESCRIPTION "I2O ProcFS OSM" #define I2O_MAX_MODULES 4 diff --git a/drivers/message/i2o/iop.c b/drivers/message/i2o/iop.c index c32022bc2a21..42f8b810d6e5 100644 --- a/drivers/message/i2o/iop.c +++ b/drivers/message/i2o/iop.c @@ -117,13 +117,13 @@ u32 i2o_cntxt_list_add(struct i2o_controller * c, void *ptr) unsigned long flags; if (!ptr) - printk(KERN_ERR "%s: couldn't add NULL pointer to context list!" - "\n", c->name); + osm_err("%s: couldn't add NULL pointer to context list!\n", + c->name); entry = kmalloc(sizeof(*entry), GFP_ATOMIC); if (!entry) { - printk(KERN_ERR "%s: Could not allocate memory for context " - "list element\n", c->name); + osm_err("%s: Could not allocate memory for context list element" + "\n", c->name); return 0; } @@ -142,7 +142,7 @@ u32 i2o_cntxt_list_add(struct i2o_controller * c, void *ptr) spin_unlock_irqrestore(&c->context_list_lock, flags); - pr_debug("%s: Add context to list %p -> %d\n", c->name, ptr, context); + osm_debug("%s: Add context to list %p -> %d\n", c->name, ptr, context); return entry->context; }; @@ -174,11 +174,11 @@ u32 i2o_cntxt_list_remove(struct i2o_controller * c, void *ptr) spin_unlock_irqrestore(&c->context_list_lock, flags); if (!context) - printk(KERN_WARNING "%s: Could not remove nonexistent ptr " - "%p\n", c->name, ptr); + osm_warn("%s: Could not remove nonexistent ptr %p\n", c->name, + ptr); - pr_debug("%s: remove ptr from context list %d -> %p\n", c->name, - context, ptr); + osm_debug("%s: remove ptr from context list %d -> %p\n", c->name, + context, ptr); return context; }; @@ -208,11 +208,10 @@ void *i2o_cntxt_list_get(struct i2o_controller *c, u32 context) spin_unlock_irqrestore(&c->context_list_lock, flags); if (!ptr) - printk(KERN_WARNING "%s: context id %d not found\n", c->name, - context); + osm_warn("%s: context id %d not found\n", c->name, context); - pr_debug("%s: get ptr from context list %d -> %p\n", c->name, context, - ptr); + osm_debug("%s: get ptr from context list %d -> %p\n", c->name, context, + ptr); return ptr; }; @@ -240,11 +239,11 @@ u32 i2o_cntxt_list_get_ptr(struct i2o_controller * c, void *ptr) spin_unlock_irqrestore(&c->context_list_lock, flags); if (!context) - printk(KERN_WARNING "%s: Could not find nonexistent ptr " - "%p\n", c->name, ptr); + osm_warn("%s: Could not find nonexistent ptr %p\n", c->name, + ptr); - pr_debug("%s: get context id from context list %p -> %d\n", c->name, - ptr, context); + osm_debug("%s: get context id from context list %p -> %d\n", c->name, + ptr, context); return context; }; @@ -324,10 +323,9 @@ static int i2o_iop_quiesce(struct i2o_controller *c) /* Long timeout needed for quiesce if lots of devices */ if ((rc = i2o_msg_post_wait(c, m, 240))) - printk(KERN_INFO "%s: Unable to quiesce (status=%#x).\n", - c->name, -rc); + osm_info("%s: Unable to quiesce (status=%#x).\n", c->name, -rc); else - pr_debug("%s: Quiesced.\n", c->name); + osm_debug("%s: Quiesced.\n", c->name); i2o_status_get(c); // Entered READY state @@ -365,10 +363,9 @@ static int i2o_iop_enable(struct i2o_controller *c) /* How long of a timeout do we need? */ if ((rc = i2o_msg_post_wait(c, m, 240))) - printk(KERN_ERR "%s: Could not enable (status=%#x).\n", - c->name, -rc); + osm_err("%s: Could not enable (status=%#x).\n", c->name, -rc); else - pr_debug("%s: Enabled.\n", c->name); + osm_debug("%s: Enabled.\n", c->name); i2o_status_get(c); // entered OPERATIONAL state @@ -432,10 +429,9 @@ static int i2o_iop_clear(struct i2o_controller *c) &msg->u.head[1]); if ((rc = i2o_msg_post_wait(c, m, 30))) - printk(KERN_INFO "%s: Unable to clear (status=%#x).\n", - c->name, -rc); + osm_info("%s: Unable to clear (status=%#x).\n", c->name, -rc); else - pr_debug("%s: Cleared.\n", c->name); + osm_debug("%s: Cleared.\n", c->name); /* Enable all IOPs */ i2o_iop_enable_all(); @@ -570,14 +566,13 @@ static int i2o_iop_reset(struct i2o_controller *c) * can't read one in the given ammount of time, we assume the * IOP could not reboot properly. */ - pr_debug("%s: Reset in progress, waiting for reboot...\n", - c->name); + osm_debug("%s: Reset in progress, waiting for reboot...\n", + c->name); m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_RESET); while (m == I2O_QUEUE_EMPTY) { if (time_after(jiffies, timeout)) { - printk(KERN_ERR "%s: IOP reset timeout.\n", - c->name); + osm_err("%s: IOP reset timeout.\n", c->name); rc = -ETIMEDOUT; goto exit; } @@ -635,29 +630,29 @@ static int i2o_iop_activate(struct i2o_controller *c) rc = i2o_status_get(c); if (rc) { - printk(KERN_INFO "%s: Unable to obtain status, " - "attempting a reset.\n", c->name); + osm_info("%s: Unable to obtain status, attempting a reset.\n", + c->name); rc = i2o_iop_reset(c); if (rc) return rc; } if (sb->i2o_version > I2OVER15) { - printk(KERN_ERR "%s: Not running version 1.5 of the I2O " - "Specification.\n", c->name); + osm_err("%s: Not running version 1.5 of the I2O Specification." + "\n", c->name); return -ENODEV; } switch (sb->iop_state) { case ADAPTER_STATE_FAULTED: - printk(KERN_CRIT "%s: hardware fault\n", c->name); + osm_err("%s: hardware fault\n", c->name); return -EFAULT; case ADAPTER_STATE_READY: case ADAPTER_STATE_OPERATIONAL: case ADAPTER_STATE_HOLD: case ADAPTER_STATE_FAILED: - pr_debug("%s: already running, trying to reset...\n", c->name); + osm_debug("%s: already running, trying to reset...\n", c->name); rc = i2o_iop_reset(c); if (rc) return rc; @@ -707,20 +702,18 @@ static int i2o_iop_systab_set(struct i2o_controller *c) res->flags = IORESOURCE_MEM; res->start = 0; res->end = 0; - printk(KERN_INFO "%s: requires private memory resources.\n", - c->name); + osm_info("%s: requires private memory resources.\n", c->name); root = pci_find_parent_resource(c->pdev, res); if (root == NULL) - printk(KERN_WARNING "%s: Can't find parent resource!\n", - c->name); + osm_warn("%s: Can't find parent resource!\n", c->name); if (root && allocate_resource(root, res, sb->desired_mem_size, sb->desired_mem_size, sb->desired_mem_size, 1 << 20, /* Unspecified, so use 1Mb and play safe */ NULL, NULL) >= 0) { c->mem_alloc = 1; sb->current_mem_size = 1 + res->end - res->start; sb->current_mem_base = res->start; - printk(KERN_INFO "%s: allocated %ld bytes of PCI memory" - " at 0x%08lX.\n", c->name, - 1 + res->end - res->start, res->start); + osm_info("%s: allocated %ld bytes of PCI memory at " + "0x%08lX.\n", c->name, + 1 + res->end - res->start, res->start); } } @@ -730,20 +723,18 @@ static int i2o_iop_systab_set(struct i2o_controller *c) res->flags = IORESOURCE_IO; res->start = 0; res->end = 0; - printk(KERN_INFO "%s: requires private memory resources.\n", - c->name); + osm_info("%s: requires private memory resources.\n", c->name); root = pci_find_parent_resource(c->pdev, res); if (root == NULL) - printk(KERN_WARNING "%s: Can't find parent resource!\n", - c->name); + osm_warn("%s: Can't find parent resource!\n", c->name); if (root && allocate_resource(root, res, sb->desired_io_size, sb->desired_io_size, sb->desired_io_size, 1 << 20, /* Unspecified, so use 1Mb and play safe */ NULL, NULL) >= 0) { c->io_alloc = 1; sb->current_io_size = 1 + res->end - res->start; sb->current_mem_base = res->start; - printk(KERN_INFO "%s: allocated %ld bytes of PCI I/O at" - " 0x%08lX.\n", c->name, - 1 + res->end - res->start, res->start); + osm_info("%s: allocated %ld bytes of PCI I/O at 0x%08lX" + ".\n", c->name, 1 + res->end - res->start, + res->start); } } @@ -787,10 +778,10 @@ static int i2o_iop_systab_set(struct i2o_controller *c) PCI_DMA_TODEVICE); if (rc < 0) - printk(KERN_ERR "%s: Unable to set SysTab (status=%#x).\n", - c->name, -rc); + osm_err("%s: Unable to set SysTab (status=%#x).\n", c->name, + -rc); else - pr_debug("%s: SysTab set.\n", c->name); + osm_debug("%s: SysTab set.\n", c->name); i2o_status_get(c); // Entered READY state @@ -814,7 +805,7 @@ static int i2o_iop_online(struct i2o_controller *c) return rc; /* In READY state */ - pr_debug("%s: Attempting to enable...\n", c->name); + osm_debug("%s: Attempting to enable...\n", c->name); rc = i2o_iop_enable(c); if (rc) return rc; @@ -833,7 +824,7 @@ void i2o_iop_remove(struct i2o_controller *c) { struct i2o_device *dev, *tmp; - pr_debug("%s: deleting controller\n", c->name); + osm_debug("%s: deleting controller\n", c->name); i2o_driver_notify_controller_remove_all(c); @@ -882,8 +873,7 @@ static int i2o_systab_build(void) systab = i2o_systab.virt = kmalloc(i2o_systab.len, GFP_KERNEL); if (!systab) { - printk(KERN_ERR "i2o: unable to allocate memory for System " - "Table\n"); + osm_err("unable to allocate memory for System Table\n"); return -ENOMEM; } memset(systab, 0, i2o_systab.len); @@ -895,8 +885,8 @@ static int i2o_systab_build(void) i2o_status_block *sb; if (count >= num_controllers) { - printk(KERN_ERR "i2o: controller added while building " - "system table\n"); + osm_err("controller added while building system table" + "\n"); break; } @@ -910,9 +900,8 @@ static int i2o_systab_build(void) * it is techninically not part of the I2O subsystem... */ if (unlikely(i2o_status_get(c))) { - printk(KERN_ERR "%s: Deleting b/c could not get status" - " while attempting to build system table\n", - c->name); + osm_err("%s: Deleting b/c could not get status while " + "attempting to build system table\n", c->name); i2o_iop_remove(c); continue; // try the next one } @@ -994,7 +983,7 @@ int i2o_status_get(struct i2o_controller *c) timeout = jiffies + I2O_TIMEOUT_STATUS_GET * HZ; while (status_block[87] != 0xFF) { if (time_after(jiffies, timeout)) { - printk(KERN_ERR "%s: Get status timeout.\n", c->name); + osm_err("%s: Get status timeout.\n", c->name); return -ETIMEDOUT; } @@ -1043,8 +1032,8 @@ static int i2o_hrt_get(struct i2o_controller *c) rc = i2o_msg_post_wait_mem(c, m, 20, &c->hrt); if (rc < 0) { - printk(KERN_ERR "%s: Unable to get HRT (status=%#x)\n", - c->name, -rc); + osm_err("%s: Unable to get HRT (status=%#x)\n", c->name, + -rc); return rc; } @@ -1058,8 +1047,8 @@ static int i2o_hrt_get(struct i2o_controller *c) return i2o_parse_hrt(c); } - printk(KERN_ERR "%s: Unable to get HRT after %d tries, giving up\n", - c->name, I2O_HRT_GET_TRIES); + osm_err("%s: Unable to get HRT after %d tries, giving up\n", c->name, + I2O_HRT_GET_TRIES); return -EBUSY; } @@ -1073,7 +1062,6 @@ void i2o_iop_free(struct i2o_controller *c) kfree(c); }; - /** * i2o_iop_release - release the memory for a I2O controller * @dev: I2O controller which should be released @@ -1109,8 +1097,8 @@ struct i2o_controller *i2o_iop_alloc(void) c = kmalloc(sizeof(*c), GFP_KERNEL); if (!c) { - printk(KERN_ERR "i2o: Insufficient memory to allocate a I2O " - "controller.\n"); + osm_err("i2o: Insufficient memory to allocate a I2O controller." + "\n"); return ERR_PTR(-ENOMEM); } memset(c, 0, sizeof(*c)); diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c index 442e34506b90..9971430e5184 100644 --- a/drivers/message/i2o/pci.c +++ b/drivers/message/i2o/pci.c @@ -179,7 +179,10 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c) return -ENOMEM; } - if (i2o_dma_alloc(dev, &c->out_queue, MSG_POOL_SIZE, GFP_KERNEL)) { + if (i2o_dma_alloc + (dev, &c->out_queue, + I2O_MAX_OUTBOUND_MSG_FRAMES * I2O_OUTBOUND_MSG_FRAME_SIZE * + sizeof(u32), GFP_KERNEL)) { i2o_pci_free(c); return -ENOMEM; } diff --git a/include/linux/i2o.h b/include/linux/i2o.h index be937d0372a7..bdc286ec947c 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -156,8 +156,8 @@ struct i2o_controller { struct pci_dev *pdev; /* PCI device */ - unsigned int promise:1; /* Promise controller */ - unsigned int adaptec:1; /* DPT / Adaptec controller */ + unsigned int promise:1; /* Promise controller */ + unsigned int adaptec:1; /* DPT / Adaptec controller */ unsigned int raptor:1; /* split bar */ unsigned int no_quiesce:1; /* dont quiesce before reset */ unsigned int short_req:1; /* use small block sizes */ @@ -174,7 +174,7 @@ struct i2o_controller { /* Dynamic LCT related data */ - struct i2o_dma status; /* status of IOP */ + struct i2o_dma status; /* IOP status block */ struct i2o_dma hrt; /* HW Resource Table */ i2o_lct *lct; /* Logical Config Table */ @@ -186,7 +186,7 @@ struct i2o_controller { struct i2o_io in_queue; /* inbound message queue Host->IOP */ struct i2o_dma out_queue; /* outbound message queue IOP->Host */ - unsigned int battery:1; /* Has a battery backup */ + unsigned int battery:1; /* Has a battery backup */ unsigned int io_alloc:1; /* An I/O resource was allocated */ unsigned int mem_alloc:1; /* A memory resource was allocated */ -- cgit v1.2.3 From 496400014f22c4dbdbc1e89249a2feba46939708 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:02:58 -0700 Subject: [PATCH] nfsd4: fix fh_expire_type We're returning NFS4_FH_NOEXPIRE_WITH_OPEN | NFS4_FH_VOL_RENAME for the fh_expire_type attribute. This is incorrect: 1. The spec actually only allows NOEXPIRE_WITH_OPEN when VOLATILE_ANY is also set. 2. Filehandles for open files can expire, if the file is removed and there is a reboot. 3. Filehandles are only volatile on rename in the nosubtree check case. Unfortunately, there's no way to indicate that we only expire on remove. So our only choice is FH4_VOLATILE_ANY. Although it's redundant, we also set FH4_VOL_RENAME in the subtree check case, since subtreecheck does actually cause problems in practice and it seems possibly useful to give clients some way to distinguish that case. Fix a mispelled #define while we're at it. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4xdr.c | 5 ++++- include/linux/nfs4.h | 2 +- 2 files changed, 5 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 36a058a112d5..0ae1467c3bc3 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1366,7 +1366,10 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, if (bmval0 & FATTR4_WORD0_FH_EXPIRE_TYPE) { if ((buflen -= 4) < 0) goto out_resource; - WRITE32( NFS4_FH_NOEXPIRE_WITH_OPEN | NFS4_FH_VOL_RENAME ); + if (exp->ex_flags & NFSEXP_NOSUBTREECHECK) + WRITE32(NFS4_FH_VOLATILE_ANY); + else + WRITE32(NFS4_FH_VOLATILE_ANY|NFS4_FH_VOL_RENAME); } if (bmval0 & FATTR4_WORD0_CHANGE) { /* diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index 5bb5b2fd7ba2..0c1c306cdaec 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -28,7 +28,7 @@ #define NFS4_ACCESS_DELETE 0x0010 #define NFS4_ACCESS_EXECUTE 0x0020 -#define NFS4_FH_PERISTENT 0x0000 +#define NFS4_FH_PERSISTENT 0x0000 #define NFS4_FH_NOEXPIRE_WITH_OPEN 0x0001 #define NFS4_FH_VOLATILE_ANY 0x0002 #define NFS4_FH_VOL_MIGRATION 0x0004 -- cgit v1.2.3 From 8beefa249371f55432394ac96864c83b0b309c28 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:08 -0700 Subject: [PATCH] nfsd4: rename nfs4_file fields Trivial renaming patch: I can never remember, while looking at various lists relating the nfsd4 state structures, which are the "heads" and which are items on other lists, or which structures are actually on the various lists. The following convention helps me: given structures foo and bar, with foo containing the head of a list of bars, use "bars" for the name of the head of the list contained in the struct foo, and use "per_foo" for the entries in the struct bars. Go ahead and do this for struct nfs4_file. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 22 +++++++++++----------- include/linux/nfsd/state.h | 4 ++-- 2 files changed, 13 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index f03a4180fa11..a84a80e8c0cf 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -153,7 +153,7 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f current_fh->fh_handle.fh_size); dp->dl_time = 0; atomic_set(&dp->dl_count, 1); - list_add(&dp->dl_del_perfile, &fp->fi_del_perfile); + list_add(&dp->dl_del_perfile, &fp->fi_delegations); list_add(&dp->dl_del_perclnt, &clp->cl_del_perclnt); return dp; } @@ -954,8 +954,8 @@ alloc_init_file(struct inode *ino) fp = kmem_cache_alloc(file_slab, GFP_KERNEL); if (fp) { INIT_LIST_HEAD(&fp->fi_hash); - INIT_LIST_HEAD(&fp->fi_perfile); - INIT_LIST_HEAD(&fp->fi_del_perfile); + INIT_LIST_HEAD(&fp->fi_stateids); + INIT_LIST_HEAD(&fp->fi_delegations); list_add(&fp->fi_hash, &file_hashtbl[hashval]); fp->fi_inode = igrab(ino); fp->fi_id = current_fileid++; @@ -974,7 +974,7 @@ release_all_files(void) while (!list_empty(&file_hashtbl[i])) { fp = list_entry(file_hashtbl[i].next, struct nfs4_file, fi_hash); /* this should never be more than once... */ - if (!list_empty(&fp->fi_perfile) || !list_empty(&fp->fi_del_perfile)) { + if (!list_empty(&fp->fi_stateids) || !list_empty(&fp->fi_delegations)) { printk("ERROR: release_all_files: file %p is open, creating dangling state !!!\n",fp); } release_file(fp); @@ -1139,7 +1139,7 @@ init_stateid(struct nfs4_stateid *stp, struct nfs4_file *fp, struct nfsd4_open * INIT_LIST_HEAD(&stp->st_perfile); list_add(&stp->st_hash, &stateid_hashtbl[hashval]); list_add(&stp->st_perfilestate, &sop->so_perfilestate); - list_add(&stp->st_perfile, &fp->fi_perfile); + list_add(&stp->st_perfile, &fp->fi_stateids); stp->st_stateowner = sop; stp->st_file = fp; stp->st_stateid.si_boot = boot_time; @@ -1204,7 +1204,7 @@ release_state_owner(struct nfs4_stateid *stp, int flag) if (sop->so_confirmed && list_empty(&sop->so_perfilestate)) move_to_close_lru(sop); /* unused nfs4_file's are releseed. XXX slab cache? */ - if (list_empty(&fp->fi_perfile) && list_empty(&fp->fi_del_perfile)) { + if (list_empty(&fp->fi_stateids) && list_empty(&fp->fi_delegations)) { release_file(fp); } } @@ -1294,7 +1294,7 @@ nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type) fp = find_file(ino); if (fp) { /* Search for conflicting share reservations */ - list_for_each_entry(stp, &fp->fi_perfile, st_perfile) { + list_for_each_entry(stp, &fp->fi_stateids, st_perfile) { if (test_bit(deny_type, &stp->st_deny_bmap) || test_bit(NFS4_SHARE_DENY_BOTH, &stp->st_deny_bmap)) return nfserr_share_denied; @@ -1545,7 +1545,7 @@ find_delegation_file(struct nfs4_file *fp, stateid_t *stid) { struct nfs4_delegation *dp; - list_for_each_entry(dp, &fp->fi_del_perfile, dl_del_perfile) { + list_for_each_entry(dp, &fp->fi_delegations, dl_del_perfile) { if (dp->dl_stateid.si_stateownerid == stid->si_stateownerid) return dp; } @@ -1583,7 +1583,7 @@ nfs4_check_open(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_state int status = nfserr_share_denied; struct nfs4_stateowner *sop = open->op_stateowner; - list_for_each_entry(local, &fp->fi_perfile, st_perfile) { + list_for_each_entry(local, &fp->fi_stateids, st_perfile) { /* ignore lock owners */ if (local->st_stateowner->so_is_open_owner == 0) continue; @@ -1830,7 +1830,7 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf stp->st_stateid.si_fileid, stp->st_stateid.si_generation); out: /* take the opportunity to clean up unused state */ - if (fp && list_empty(&fp->fi_perfile) && list_empty(&fp->fi_del_perfile)) + if (fp && list_empty(&fp->fi_stateids) && list_empty(&fp->fi_delegations)) release_file(fp); /* CLAIM_PREVIOUS has different error returns */ @@ -2633,7 +2633,7 @@ alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struc INIT_LIST_HEAD(&stp->st_perfilestate); INIT_LIST_HEAD(&stp->st_perlockowner); /* not used */ list_add(&stp->st_hash, &lockstateid_hashtbl[hashval]); - list_add(&stp->st_perfile, &fp->fi_perfile); + list_add(&stp->st_perfile, &fp->fi_stateids); list_add(&stp->st_perfilestate, &sop->so_perfilestate); stp->st_stateowner = sop; stp->st_file = fp; diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index b6b2fe1e7c63..2c3b42674a4c 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -218,8 +218,8 @@ struct nfs4_stateowner { */ struct nfs4_file { struct list_head fi_hash; /* hash by "struct inode *" */ - struct list_head fi_perfile; /* list: nfs4_stateid */ - struct list_head fi_del_perfile; /* list: nfs4_delegation */ + struct list_head fi_stateids; + struct list_head fi_delegations; struct inode *fi_inode; u32 fi_id; /* used with stateowner->so_id * for stateid_hashtbl hash */ -- cgit v1.2.3 From 13cd21845d6a9729ca95e36ae6e8c669623fbfd4 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:10 -0700 Subject: [PATCH] nfsd4: reference count struct nfs4_file Add a struct kref to each nfs4_file and take a reference to it from each stateid and delegation that refers to it. The atomicity guarantees are overkill given that all this stuff is done under the single nfsd4 state lock, but a) we'd like finer-grained locking some day, and b) this simplifies the cleanup of the structures a bit, something that has previously been a bit complicated and bug-prone. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 100 +++++++++++++++++++++++---------------------- include/linux/nfsd/state.h | 1 + 2 files changed, 52 insertions(+), 49 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index a84a80e8c0cf..6ba428afa433 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -111,7 +111,6 @@ opaque_hashval(const void *ptr, int nbytes) /* forward declarations */ static void release_stateowner(struct nfs4_stateowner *sop); static void release_stateid(struct nfs4_stateid *stp, int flags); -static void release_file(struct nfs4_file *fp); /* * Delegation state @@ -121,6 +120,27 @@ static void release_file(struct nfs4_file *fp); spinlock_t recall_lock; static struct list_head del_recall_lru; +static void +free_nfs4_file(struct kref *kref) +{ + struct nfs4_file *fp = container_of(kref, struct nfs4_file, fi_ref); + list_del(&fp->fi_hash); + iput(fp->fi_inode); + kmem_cache_free(file_slab, fp); +} + +static inline void +put_nfs4_file(struct nfs4_file *fi) +{ + kref_put(&fi->fi_ref, free_nfs4_file); +} + +static inline void +get_nfs4_file(struct nfs4_file *fi) +{ + kref_get(&fi->fi_ref); +} + static struct nfs4_delegation * alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_fh *current_fh, u32 type) { @@ -136,6 +156,7 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f INIT_LIST_HEAD(&dp->dl_del_perclnt); INIT_LIST_HEAD(&dp->dl_recall_lru); dp->dl_client = clp; + get_nfs4_file(fp); dp->dl_file = fp; dp->dl_flock = NULL; get_file(stp->st_vfs_file); @@ -163,6 +184,7 @@ nfs4_put_delegation(struct nfs4_delegation *dp) { if (atomic_dec_and_test(&dp->dl_count)) { dprintk("NFSD: freeing dp %p\n",dp); + put_nfs4_file(dp->dl_file); kmem_cache_free(deleg_slab, dp); } } @@ -953,6 +975,7 @@ alloc_init_file(struct inode *ino) fp = kmem_cache_alloc(file_slab, GFP_KERNEL); if (fp) { + kref_init(&fp->fi_ref); INIT_LIST_HEAD(&fp->fi_hash); INIT_LIST_HEAD(&fp->fi_stateids); INIT_LIST_HEAD(&fp->fi_delegations); @@ -964,24 +987,6 @@ alloc_init_file(struct inode *ino) return NULL; } -static void -release_all_files(void) -{ - int i; - struct nfs4_file *fp; - - for (i=0;ifi_stateids) || !list_empty(&fp->fi_delegations)) { - printk("ERROR: release_all_files: file %p is open, creating dangling state !!!\n",fp); - } - release_file(fp); - } - } -} - static void nfsd4_free_slab(kmem_cache_t **slab) { @@ -1141,6 +1146,7 @@ init_stateid(struct nfs4_stateid *stp, struct nfs4_file *fp, struct nfsd4_open * list_add(&stp->st_perfilestate, &sop->so_perfilestate); list_add(&stp->st_perfile, &fp->fi_stateids); stp->st_stateowner = sop; + get_nfs4_file(fp); stp->st_file = fp; stp->st_stateid.si_boot = boot_time; stp->st_stateid.si_stateownerid = sop->so_id; @@ -1166,18 +1172,11 @@ release_stateid(struct nfs4_stateid *stp, int flags) nfsd_close(filp); } else if (flags & LOCK_STATE) locks_remove_posix(filp, (fl_owner_t) stp->st_stateowner); + put_nfs4_file(stp->st_file); kmem_cache_free(stateid_slab, stp); stp = NULL; } -static void -release_file(struct nfs4_file *fp) -{ - list_del(&fp->fi_hash); - iput(fp->fi_inode); - kmem_cache_free(file_slab, fp); -} - void move_to_close_lru(struct nfs4_stateowner *sop) { @@ -1192,7 +1191,6 @@ void release_state_owner(struct nfs4_stateid *stp, int flag) { struct nfs4_stateowner *sop = stp->st_stateowner; - struct nfs4_file *fp = stp->st_file; dprintk("NFSD: release_state_owner\n"); release_stateid(stp, flag); @@ -1203,10 +1201,6 @@ release_state_owner(struct nfs4_stateid *stp, int flag) */ if (sop->so_confirmed && list_empty(&sop->so_perfilestate)) move_to_close_lru(sop); - /* unused nfs4_file's are releseed. XXX slab cache? */ - if (list_empty(&fp->fi_stateids) && list_empty(&fp->fi_delegations)) { - release_file(fp); - } } static int @@ -1236,8 +1230,10 @@ find_file(struct inode *ino) struct nfs4_file *fp; list_for_each_entry(fp, &file_hashtbl[hashval], fi_hash) { - if (fp->fi_inode == ino) + if (fp->fi_inode == ino) { + get_nfs4_file(fp); return fp; + } } return NULL; } @@ -1288,19 +1284,24 @@ nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type) struct inode *ino = current_fh->fh_dentry->d_inode; struct nfs4_file *fp; struct nfs4_stateid *stp; + int ret; dprintk("NFSD: nfs4_share_conflict\n"); fp = find_file(ino); - if (fp) { + if (!fp) + return nfs_ok; + ret = nfserr_share_denied; /* Search for conflicting share reservations */ - list_for_each_entry(stp, &fp->fi_stateids, st_perfile) { - if (test_bit(deny_type, &stp->st_deny_bmap) || - test_bit(NFS4_SHARE_DENY_BOTH, &stp->st_deny_bmap)) - return nfserr_share_denied; - } + list_for_each_entry(stp, &fp->fi_stateids, st_perfile) { + if (test_bit(deny_type, &stp->st_deny_bmap) || + test_bit(NFS4_SHARE_DENY_BOTH, &stp->st_deny_bmap)) + goto out; } - return nfs_ok; + ret = nfs_ok; +out: + put_nfs4_file(fp); + return ret; } static inline void @@ -1829,10 +1830,8 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf stp->st_stateid.si_boot, stp->st_stateid.si_stateownerid, stp->st_stateid.si_fileid, stp->st_stateid.si_generation); out: - /* take the opportunity to clean up unused state */ - if (fp && list_empty(&fp->fi_stateids) && list_empty(&fp->fi_delegations)) - release_file(fp); - + if (fp) + put_nfs4_file(fp); /* CLAIM_PREVIOUS has different error returns */ nfs4_set_claim_prev(open, &status); /* @@ -2480,16 +2479,19 @@ find_stateid(stateid_t *stid, int flags) static struct nfs4_delegation * find_delegation_stateid(struct inode *ino, stateid_t *stid) { - struct nfs4_file *fp = NULL; + struct nfs4_file *fp; + struct nfs4_delegation *dl; dprintk("NFSD:find_delegation_stateid stateid=(%08x/%08x/%08x/%08x)\n", stid->si_boot, stid->si_stateownerid, stid->si_fileid, stid->si_generation); fp = find_file(ino); - if (fp) - return find_delegation_file(fp, stid); - return NULL; + if (!fp) + return NULL; + dl = find_delegation_file(fp, stid); + put_nfs4_file(fp); + return dl; } /* @@ -2636,6 +2638,7 @@ alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struc list_add(&stp->st_perfile, &fp->fi_stateids); list_add(&stp->st_perfilestate, &sop->so_perfilestate); stp->st_stateowner = sop; + get_nfs4_file(fp); stp->st_file = fp; stp->st_stateid.si_boot = boot_time; stp->st_stateid.si_stateownerid = sop->so_id; @@ -3287,7 +3290,6 @@ __nfs4_state_shutdown(void) unhash_delegation(dp); } - release_all_files(); cancel_delayed_work(&laundromat_work); flush_scheduled_work(); nfs4_init = 0; diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 2c3b42674a4c..296e6429fc3b 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -217,6 +217,7 @@ struct nfs4_stateowner { * share_acces, share_deny on the file. */ struct nfs4_file { + struct kref fi_ref; struct list_head fi_hash; /* hash by "struct inode *" */ struct list_head fi_stateids; struct list_head fi_delegations; -- cgit v1.2.3 From 7b190fecfa33d72bcf74c9473134c2ad14ae9545 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:23 -0700 Subject: [PATCH] knfsd: nfsd4: delegation recovery Allow recovery of delegations after reboot. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 36 ++++++++++++++++++++++++++++-------- fs/nfsd/nfs4xdr.c | 2 +- include/linux/nfsd/xdr4.h | 1 + 3 files changed, 30 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 16c9a43218c3..0f6119714c8c 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1709,14 +1709,30 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta int status, flag = 0; flag = NFS4_OPEN_DELEGATE_NONE; - if (open->op_claim_type != NFS4_OPEN_CLAIM_NULL - || !atomic_read(&cb->cb_set) || !sop->so_confirmed) - goto out; - - if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) - flag = NFS4_OPEN_DELEGATE_WRITE; - else - flag = NFS4_OPEN_DELEGATE_READ; + open->op_recall = 0; + switch (open->op_claim_type) { + case NFS4_OPEN_CLAIM_PREVIOUS: + if (!atomic_read(&cb->cb_set)) + open->op_recall = 1; + flag = open->op_delegate_type; + if (flag == NFS4_OPEN_DELEGATE_NONE) + goto out; + break; + case NFS4_OPEN_CLAIM_NULL: + /* Let's not give out any delegations till everyone's + * had the chance to reclaim theirs.... */ + if (nfs4_in_grace()) + goto out; + if (!atomic_read(&cb->cb_set) || !sop->so_confirmed) + goto out; + if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) + flag = NFS4_OPEN_DELEGATE_WRITE; + else + flag = NFS4_OPEN_DELEGATE_READ; + break; + default: + goto out; + } dp = alloc_init_deleg(sop->so_client, stp, fh, flag); if (dp == NULL) { @@ -1750,6 +1766,10 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta dp->dl_stateid.si_fileid, dp->dl_stateid.si_generation); out: + if (open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS + && flag == NFS4_OPEN_DELEGATE_NONE + && open->op_delegate_type != NFS4_OPEN_DELEGATE_NONE) + printk("NFSD: WARNING: refusing delegation reclaim\n"); open->op_delegate_type = flag; } diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 0ae1467c3bc3..cfe978a72cea 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1972,7 +1972,7 @@ nfsd4_encode_open(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open case NFS4_OPEN_DELEGATE_READ: RESERVE_SPACE(20 + sizeof(stateid_t)); WRITEMEM(&open->op_delegate_stateid, sizeof(stateid_t)); - WRITE32(0); + WRITE32(open->op_recall); /* * TODO: ACE's in delegations diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h index a1f5ad0be1bf..4d24d65c0e88 100644 --- a/include/linux/nfsd/xdr4.h +++ b/include/linux/nfsd/xdr4.h @@ -210,6 +210,7 @@ struct nfsd4_open { u32 op_share_access; /* request */ u32 op_share_deny; /* request */ stateid_t op_stateid; /* response */ + u32 op_recall; /* recall */ struct nfsd4_change_info op_cinfo; /* response */ u32 op_rflags; /* response */ int op_truncate; /* used during processing */ -- cgit v1.2.3 From 76a3550ec50ed86885a10a767ebaebb7c9104721 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:26 -0700 Subject: [PATCH] knfsd: nfsd4: rename nfs4_state_init Somewhat gratuitous rename to simplify following patch. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 6 +++--- fs/nfsd/nfssvc.c | 2 +- include/linux/nfsd/nfsd.h | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 0f6119714c8c..e00b3472851c 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -3185,7 +3185,7 @@ nfs4_check_open_reclaim(clientid_t *clid) */ static void -__nfs4_state_init(void) +__nfs4_state_start(void) { int i; time_t grace_time; @@ -3235,7 +3235,7 @@ __nfs4_state_init(void) } int -nfs4_state_init(void) +nfs4_state_start(void) { int status; @@ -3244,7 +3244,7 @@ nfs4_state_init(void) status = nfsd4_init_slabs(); if (status) return status; - __nfs4_state_init(); + __nfs4_state_start(); nfs4_init = 1; return 0; } diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 904df604e86b..07b9a065e9da 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -95,7 +95,7 @@ nfsd_svc(unsigned short port, int nrservs) error = nfsd_racache_init(2*nrservs); if (error<0) goto out; - error = nfs4_state_init(); + error = nfs4_state_start(); if (error<0) goto out; if (!nfsd_serv) { diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 4bf931d5ff56..3855fdc5af77 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -145,12 +145,12 @@ int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *); * NFSv4 State */ #ifdef CONFIG_NFSD_V4 -int nfs4_state_init(void); +int nfs4_state_start(void); void nfs4_state_shutdown(void); time_t nfs4_lease_time(void); void nfs4_reset_lease(time_t leasetime); #else -static inline int nfs4_state_init(void){return 0;} +static inline int nfs4_state_start(void){return 0;} static inline void nfs4_state_shutdown(void){} static inline time_t nfs4_lease_time(void){return 0;} static inline void nfs4_reset_lease(time_t leasetime){} -- cgit v1.2.3 From ac4d8ff2a57179de3ef7834c6ab3fac430b0a05d Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:30 -0700 Subject: [PATCH] knfsd: nfsd4: clean up state initialization Separate out stuff that needs initialization on startup from stuff that only needs initialization on module init from static data. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 35 +++++++++++++++++------------------ fs/nfsd/nfsctl.c | 1 + include/linux/nfsd/nfsd.h | 2 ++ 3 files changed, 20 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index e00b3472851c..1f68ce36e724 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -117,7 +117,7 @@ static void release_stateid(struct nfs4_stateid *stp, int flags); */ /* recall_lock protects the del_recall_lru */ -spinlock_t recall_lock; +spinlock_t recall_lock = SPIN_LOCK_UNLOCKED; static struct list_head del_recall_lru; static void @@ -3179,23 +3179,13 @@ nfs4_check_open_reclaim(clientid_t *clid) return nfs4_find_reclaim_client(clid) ? nfs_ok : nfserr_reclaim_bad; } +/* initialization to perform at module load time: */ -/* - * Start and stop routines - */ - -static void -__nfs4_state_start(void) +void +nfs4_state_init(void) { int i; - time_t grace_time; - if (!nfs4_reclaim_init) { - for (i = 0; i < CLIENT_HASH_SIZE; i++) - INIT_LIST_HEAD(&reclaim_str_hashtbl[i]); - reclaim_str_hashtbl_size = 0; - nfs4_reclaim_init = 1; - } for (i = 0; i < CLIENT_HASH_SIZE; i++) { INIT_LIST_HEAD(&conf_id_hashtbl[i]); INIT_LIST_HEAD(&conf_str_hashtbl[i]); @@ -3217,19 +3207,28 @@ __nfs4_state_start(void) INIT_LIST_HEAD(&lock_ownerid_hashtbl[i]); INIT_LIST_HEAD(&lock_ownerstr_hashtbl[i]); } - memset(&zerostateid, 0, sizeof(stateid_t)); memset(&onestateid, ~0, sizeof(stateid_t)); - INIT_LIST_HEAD(&close_lru); INIT_LIST_HEAD(&client_lru); INIT_LIST_HEAD(&del_recall_lru); - spin_lock_init(&recall_lock); + for (i = 0; i < CLIENT_HASH_SIZE; i++) + INIT_LIST_HEAD(&reclaim_str_hashtbl[i]); + reclaim_str_hashtbl_size = 0; + nfs4_reclaim_init = 1; +} + +/* initialization to perform when the nfsd service is started: */ + +static void +__nfs4_state_start(void) +{ + time_t grace_time; + boot_time = get_seconds(); grace_time = max(user_lease_time, lease_time); lease_time = user_lease_time; printk("NFSD: starting %ld-second grace period\n", grace_time); grace_end = boot_time + grace_time; - INIT_WORK(&laundromat_work,laundromat_main, NULL); laundry_wq = create_singlethread_workqueue("nfsd4"); queue_delayed_work(laundry_wq, &laundromat_work, NFSD_LEASE_TIME*HZ); } diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 161afdcb8f7d..3d56531a7a03 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -397,6 +397,7 @@ static int __init init_nfsd(void) nfsd_cache_init(); /* RPC reply cache */ nfsd_export_init(); /* Exports table */ nfsd_lockd_init(); /* lockd->nfsd callbacks */ + nfs4_state_init(); /* NFSv4 locking state */ #ifdef CONFIG_NFSD_V4 nfsd_idmap_init(); /* Name to ID mapping */ #endif /* CONFIG_NFSD_V4 */ diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 3855fdc5af77..21c6e9d86e4f 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -145,11 +145,13 @@ int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *); * NFSv4 State */ #ifdef CONFIG_NFSD_V4 +void nfs4_state_init(void); int nfs4_state_start(void); void nfs4_state_shutdown(void); time_t nfs4_lease_time(void); void nfs4_reset_lease(time_t leasetime); #else +static inline void nfs4_state_init(void){}; static inline int nfs4_state_start(void){return 0;} static inline void nfs4_state_shutdown(void){} static inline time_t nfs4_lease_time(void){return 0;} -- cgit v1.2.3 From bd0b1e954e3ba3e5d2cab941458cf98206471bd2 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:35 -0700 Subject: [PATCH] knfsd: nfsd4: idmap initialization Adopt standard kernel style by defining a no-op function instead of putting ifdef's in the code where the function is called. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfsctl.c | 4 ---- include/linux/nfsd_idmap.h | 5 +++++ 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 3d56531a7a03..3da43a3ed32c 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -398,9 +398,7 @@ static int __init init_nfsd(void) nfsd_export_init(); /* Exports table */ nfsd_lockd_init(); /* lockd->nfsd callbacks */ nfs4_state_init(); /* NFSv4 locking state */ -#ifdef CONFIG_NFSD_V4 nfsd_idmap_init(); /* Name to ID mapping */ -#endif /* CONFIG_NFSD_V4 */ if (proc_mkdir("fs/nfs", NULL)) { struct proc_dir_entry *entry; entry = create_proc_entry("fs/nfs/exports", 0, NULL); @@ -427,9 +425,7 @@ static void __exit exit_nfsd(void) remove_proc_entry("fs/nfs", NULL); nfsd_stat_shutdown(); nfsd_lockd_shutdown(); -#ifdef CONFIG_NFSD_V4 nfsd_idmap_shutdown(); -#endif /* CONFIG_NFSD_V4 */ unregister_filesystem(&nfsd_fs_type); } diff --git a/include/linux/nfsd_idmap.h b/include/linux/nfsd_idmap.h index 9bb7f30e923b..e82746fcad14 100644 --- a/include/linux/nfsd_idmap.h +++ b/include/linux/nfsd_idmap.h @@ -43,8 +43,13 @@ /* XXX from linux/nfs_idmap.h */ #define IDMAP_NAMESZ 128 +#ifdef CONFIG_NFSD_V4 void nfsd_idmap_init(void); void nfsd_idmap_shutdown(void); +#else +static inline void nfsd_idmap_init(void) {}; +static inline void nfsd_idmap_shutdown(void) {}; +#endif int nfsd_map_name_to_uid(struct svc_rqst *, const char *, size_t, __u32 *); int nfsd_map_name_to_gid(struct svc_rqst *, const char *, size_t, __u32 *); -- cgit v1.2.3 From a55370a3c0106106a975c5a09cee800611d0cf50 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:52 -0700 Subject: [PATCH] knfsd: nfsd4: reboot hash For the purposes of reboot recovery we keep a directory with subdirectories each having a name that is the ascii hex representation of the md5 sum of a client identifier for an active client. This adds the code to calculate that name. We also use it for the purposes of comparing clients, so if someone ever manages to find two client names that are md5 collisions, then we'll return clid_inuse to the second. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/Kconfig | 2 + fs/nfsd/Makefile | 2 +- fs/nfsd/nfs4recover.c | 97 ++++++++++++++++++++++++++++++++++++++++++++++ fs/nfsd/nfs4state.c | 80 ++++++++++++++++++-------------------- include/linux/nfsd/state.h | 6 ++- 5 files changed, 143 insertions(+), 44 deletions(-) create mode 100644 fs/nfsd/nfs4recover.c (limited to 'include/linux') diff --git a/fs/Kconfig b/fs/Kconfig index a7c0cc3203cb..5c704d05627a 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1413,6 +1413,8 @@ config NFSD_V4 bool "Provide NFSv4 server support (EXPERIMENTAL)" depends on NFSD_V3 && EXPERIMENTAL select NFSD_TCP + select CRYPTO_MD5 + select CRYPTO help If you would like to include the NFSv4 server as well as the NFSv2 and NFSv3 servers, say Y here. This feature is experimental, and diff --git a/fs/nfsd/Makefile b/fs/nfsd/Makefile index 9f043f44c92f..ce341dc76d5e 100644 --- a/fs/nfsd/Makefile +++ b/fs/nfsd/Makefile @@ -10,5 +10,5 @@ nfsd-$(CONFIG_NFSD_V2_ACL) += nfs2acl.o nfsd-$(CONFIG_NFSD_V3) += nfs3proc.o nfs3xdr.o nfsd-$(CONFIG_NFSD_V3_ACL) += nfs3acl.o nfsd-$(CONFIG_NFSD_V4) += nfs4proc.o nfs4xdr.o nfs4state.o nfs4idmap.o \ - nfs4acl.o nfs4callback.o + nfs4acl.o nfs4callback.o nfs4recover.o nfsd-objs := $(nfsd-y) diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c new file mode 100644 index 000000000000..841a305d7948 --- /dev/null +++ b/fs/nfsd/nfs4recover.c @@ -0,0 +1,97 @@ +/* +* linux/fs/nfsd/nfs4recover.c +* +* Copyright (c) 2004 The Regents of the University of Michigan. +* All rights reserved. +* +* Andy Adamson +* +* Redistribution and use in source and binary forms, with or without +* modification, are permitted provided that the following conditions +* are met: +* +* 1. Redistributions of source code must retain the above copyright +* notice, this list of conditions and the following disclaimer. +* 2. Redistributions in binary form must reproduce the above copyright +* notice, this list of conditions and the following disclaimer in the +* documentation and/or other materials provided with the distribution. +* 3. Neither the name of the University nor the names of its +* contributors may be used to endorse or promote products derived +* from this software without specific prior written permission. +* +* THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED +* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF +* MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +* DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE +* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR +* BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF +* LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +* NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +* SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +* +*/ + + +#include +#include +#include +#include +#include +#include +#include +#include + + +#define NFSDDBG_FACILITY NFSDDBG_PROC + +static void +md5_to_hex(char *out, char *md5) +{ + int i; + + for (i=0; i<16; i++) { + unsigned char c = md5[i]; + + *out++ = '0' + ((c&0xf0)>>4) + (c>=0xa0)*('a'-'9'-1); + *out++ = '0' + (c&0x0f) + ((c&0x0f)>=0x0a)*('a'-'9'-1); + } + *out = '\0'; +} + +int +nfs4_make_rec_clidname(char *dname, struct xdr_netobj *clname) +{ + struct xdr_netobj cksum; + struct crypto_tfm *tfm; + struct scatterlist sg[1]; + int status = nfserr_resource; + + dprintk("NFSD: nfs4_make_rec_clidname for %.*s\n", + clname->len, clname->data); + tfm = crypto_alloc_tfm("md5", 0); + if (tfm == NULL) + goto out; + cksum.len = crypto_tfm_alg_digestsize(tfm); + cksum.data = kmalloc(cksum.len, GFP_KERNEL); + if (cksum.data == NULL) + goto out; + crypto_digest_init(tfm); + + sg[0].page = virt_to_page(clname->data); + sg[0].offset = offset_in_page(clname->data); + sg[0].length = clname->len; + + crypto_digest_update(tfm, sg, 1); + crypto_digest_final(tfm, cksum.data); + + md5_to_hex(dname, cksum.data); + + kfree(cksum.data); + status = nfs_ok; +out: + if (tfm) + crypto_free_tfm(tfm); + return status; +} diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 2a59d176e69a..0be0b37c84e9 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -231,8 +231,8 @@ unhash_delegation(struct nfs4_delegation *dp) #define clientid_hashval(id) \ ((id) & CLIENT_HASH_MASK) -#define clientstr_hashval(name, namelen) \ - (opaque_hashval((name), (namelen)) & CLIENT_HASH_MASK) +#define clientstr_hashval(name) \ + (opaque_hashval((name), 8) & CLIENT_HASH_MASK) /* * reclaim_str_hashtbl[] holds known client info from previous reset/reboot * used in reboot/reset lease grace period processing @@ -366,11 +366,12 @@ expire_client(struct nfs4_client *clp) } static struct nfs4_client * -create_client(struct xdr_netobj name) { +create_client(struct xdr_netobj name, char *recdir) { struct nfs4_client *clp; if (!(clp = alloc_client(name))) goto out; + memcpy(clp->cl_recdir, recdir, HEXDIR_LEN); atomic_set(&clp->cl_count, 1); atomic_set(&clp->cl_callback.cb_set, 0); clp->cl_callback.cb_parsed = 0; @@ -403,11 +404,9 @@ copy_cred(struct svc_cred *target, struct svc_cred *source) { get_group_info(target->cr_group_info); } -static int -cmp_name(struct xdr_netobj *n1, struct xdr_netobj *n2) { - if (!n1 || !n2) - return 0; - return((n1->len == n2->len) && !memcmp(n1->data, n2->data, n2->len)); +static inline int +same_name(const char *n1, const char *n2) { + return 0 == memcmp(n1, n2, HEXDIR_LEN); } static int @@ -479,8 +478,7 @@ move_to_confirmed(struct nfs4_client *clp) list_del_init(&clp->cl_strhash); list_del_init(&clp->cl_idhash); list_add(&clp->cl_idhash, &conf_id_hashtbl[idhashval]); - strhashval = clientstr_hashval(clp->cl_name.data, - clp->cl_name.len); + strhashval = clientstr_hashval(clp->cl_recdir); list_add(&clp->cl_strhash, &conf_str_hashtbl[strhashval]); renew_client(clp); } @@ -651,22 +649,27 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) unsigned int strhashval; struct nfs4_client * conf, * unconf, * new, * clp; int status; + char dname[HEXDIR_LEN]; status = nfserr_inval; if (!check_name(clname)) goto out; + status = nfs4_make_rec_clidname(dname, &clname); + if (status) + goto out; + /* * XXX The Duplicate Request Cache (DRC) has been checked (??) * We get here on a DRC miss. */ - strhashval = clientstr_hashval(clname.data, clname.len); + strhashval = clientstr_hashval(dname); conf = NULL; nfs4_lock_state(); list_for_each_entry(clp, &conf_str_hashtbl[strhashval], cl_strhash) { - if (!cmp_name(&clp->cl_name, &clname)) + if (!same_name(clp->cl_recdir, dname)) continue; /* * CASE 0: @@ -686,7 +689,7 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) } unconf = NULL; list_for_each_entry(clp, &unconf_str_hashtbl[strhashval], cl_strhash) { - if (!cmp_name(&clp->cl_name, &clname)) + if (!same_name(clp->cl_recdir, dname)) continue; /* cl_name match from a previous SETCLIENTID operation */ unconf = clp; @@ -700,7 +703,8 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) */ if (unconf) expire_client(unconf); - if (!(new = create_client(clname))) + new = create_client(clname, dname); + if (new == NULL) goto out; copy_verf(new, &clverifier); new->cl_addr = ip_addr; @@ -728,7 +732,8 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) cmp_clid(&unconf->cl_clientid, &conf->cl_clientid)) { expire_client(unconf); } - if (!(new = create_client(clname))) + new = create_client(clname, dname); + if (new == NULL) goto out; copy_verf(new,&conf->cl_verifier); new->cl_addr = ip_addr; @@ -746,7 +751,8 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) * using input clverifier, clname, and callback info * and generate a new cl_clientid and cl_confirm. */ - if (!(new = create_client(clname))) + new = create_client(clname, dname); + if (new == NULL) goto out; copy_verf(new,&clverifier); new->cl_addr = ip_addr; @@ -772,7 +778,8 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) * new cl_verifier and a new cl_confirm */ expire_client(unconf); - if (!(new = create_client(clname))) + new = create_client(clname, dname); + if (new == NULL) goto out; copy_verf(new,&clverifier); new->cl_addr = ip_addr; @@ -856,7 +863,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confi if ((conf && unconf) && (cmp_verf(&unconf->cl_confirm, &confirm)) && (cmp_verf(&conf->cl_verifier, &unconf->cl_verifier)) && - (cmp_name(&conf->cl_name,&unconf->cl_name)) && + (same_name(conf->cl_recdir,unconf->cl_recdir)) && (!cmp_verf(&conf->cl_confirm, &unconf->cl_confirm))) { if (!cmp_creds(&conf->cl_cred, &unconf->cl_cred)) status = nfserr_clid_inuse; @@ -876,7 +883,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confi if ((conf && !unconf) || ((conf && unconf) && (!cmp_verf(&conf->cl_verifier, &unconf->cl_verifier) || - !cmp_name(&conf->cl_name, &unconf->cl_name)))) { + !same_name(conf->cl_recdir, unconf->cl_recdir)))) { if (!cmp_creds(&conf->cl_cred,&rqstp->rq_cred)) { status = nfserr_clid_inuse; } else { @@ -3074,39 +3081,28 @@ out: } static inline struct nfs4_client_reclaim * -alloc_reclaim(int namelen) +alloc_reclaim(void) { - struct nfs4_client_reclaim *crp = NULL; - - crp = kmalloc(sizeof(struct nfs4_client_reclaim), GFP_KERNEL); - if (!crp) - return NULL; - crp->cr_name.data = kmalloc(namelen, GFP_KERNEL); - if (!crp->cr_name.data) { - kfree(crp); - return NULL; - } - return crp; + return kmalloc(sizeof(struct nfs4_client_reclaim), GFP_KERNEL); } /* * failure => all reset bets are off, nfserr_no_grace... */ static int -nfs4_client_to_reclaim(char *name, int namlen) +nfs4_client_to_reclaim(char *name) { unsigned int strhashval; struct nfs4_client_reclaim *crp = NULL; - dprintk("NFSD nfs4_client_to_reclaim NAME: %.*s\n", namlen, name); - crp = alloc_reclaim(namlen); + dprintk("NFSD nfs4_client_to_reclaim NAME: %.*s\n", HEXDIR_LEN, name); + crp = alloc_reclaim(); if (!crp) return 0; - strhashval = clientstr_hashval(name, namlen); + strhashval = clientstr_hashval(name); INIT_LIST_HEAD(&crp->cr_strhash); list_add(&crp->cr_strhash, &reclaim_str_hashtbl[strhashval]); - memcpy(crp->cr_name.data, name, namlen); - crp->cr_name.len = namlen; + memcpy(crp->cr_recdir, name, HEXDIR_LEN); reclaim_str_hashtbl_size++; return 1; } @@ -3122,7 +3118,6 @@ nfs4_release_reclaim(void) crp = list_entry(reclaim_str_hashtbl[i].next, struct nfs4_client_reclaim, cr_strhash); list_del(&crp->cr_strhash); - kfree(crp->cr_name.data); kfree(crp); reclaim_str_hashtbl_size--; } @@ -3145,13 +3140,14 @@ nfs4_find_reclaim_client(clientid_t *clid) if (clp == NULL) return NULL; - dprintk("NFSD: nfs4_find_reclaim_client for %.*s\n", - clp->cl_name.len, clp->cl_name.data); + dprintk("NFSD: nfs4_find_reclaim_client for %.*s with recdir %s\n", + clp->cl_name.len, clp->cl_name.data, + clp->cl_recdir); /* find clp->cl_name in reclaim_str_hashtbl */ - strhashval = clientstr_hashval(clp->cl_name.data, clp->cl_name.len); + strhashval = clientstr_hashval(clp->cl_recdir); list_for_each_entry(crp, &reclaim_str_hashtbl[strhashval], cr_strhash) { - if (cmp_name(&crp->cr_name, &clp->cl_name)) { + if (same_name(crp->cr_recdir, clp->cl_recdir)) { return crp; } } diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 296e6429fc3b..fdaa84addadb 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -109,6 +109,8 @@ struct nfs4_callback { struct rpc_clnt * cb_client; }; +#define HEXDIR_LEN 33 /* hex version of 16 byte md5 of cl_name plus '\0' */ + /* * struct nfs4_client - one per client. Clientids live here. * o Each nfs4_client is hashed by clientid. @@ -126,6 +128,7 @@ struct nfs4_client { struct list_head cl_del_perclnt; /* list: delegations */ struct list_head cl_lru; /* tail queue */ struct xdr_netobj cl_name; /* id generated by client */ + char cl_recdir[HEXDIR_LEN]; /* recovery dir */ nfs4_verifier cl_verifier; /* generated by client */ time_t cl_time; /* time of last lease renewal */ u32 cl_addr; /* client ipaddress */ @@ -143,7 +146,7 @@ struct nfs4_client { */ struct nfs4_client_reclaim { struct list_head cr_strhash; /* hash by cr_name */ - struct xdr_netobj cr_name; /* id generated by client */ + char cr_recdir[HEXDIR_LEN]; /* recover dir */ }; static inline void @@ -283,6 +286,7 @@ extern void nfs4_free_stateowner(struct kref *kref); extern void nfsd4_probe_callback(struct nfs4_client *clp); extern void nfsd4_cb_recall(struct nfs4_delegation *dp); extern void nfs4_put_delegation(struct nfs4_delegation *dp); +extern int nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname); static inline void nfs4_put_stateowner(struct nfs4_stateowner *so) -- cgit v1.2.3 From fd39ca9a808c6026989bc2188868a0574eb37108 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:03 -0700 Subject: [PATCH] knfsd: nfsd4: make needlessly global code static This patch contains the following possible cleanups: - make needlessly global code static Signed-off-by: Adrian Bunk Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4acl.c | 4 ++-- fs/nfsd/nfs4callback.c | 7 +++--- fs/nfsd/nfs4idmap.c | 12 +++++----- fs/nfsd/nfs4state.c | 57 ++++++++++++++++++++++++---------------------- fs/nfsd/nfs4xdr.c | 4 ++-- include/linux/nfsd/state.h | 8 ------- 6 files changed, 43 insertions(+), 49 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c index 11ebf6c4aa54..4a2105552ac4 100644 --- a/fs/nfsd/nfs4acl.c +++ b/fs/nfsd/nfs4acl.c @@ -125,7 +125,7 @@ static short ace2type(struct nfs4_ace *); static int _posix_to_nfsv4_one(struct posix_acl *, struct nfs4_acl *, unsigned int); static struct posix_acl *_nfsv4_to_posix_one(struct nfs4_acl *, unsigned int); int nfs4_acl_add_ace(struct nfs4_acl *, u32, u32, u32, int, uid_t); -int nfs4_acl_split(struct nfs4_acl *, struct nfs4_acl *); +static int nfs4_acl_split(struct nfs4_acl *, struct nfs4_acl *); struct nfs4_acl * nfs4_acl_posix_to_nfsv4(struct posix_acl *pacl, struct posix_acl *dpacl, @@ -775,7 +775,7 @@ out_err: return pacl; } -int +static int nfs4_acl_split(struct nfs4_acl *acl, struct nfs4_acl *dacl) { struct list_head *h, *n; diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 38c3e1c47d83..68bb245491f6 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -54,7 +54,6 @@ /* declarations */ static void nfs4_cb_null(struct rpc_task *task); -extern spinlock_t recall_lock; /* Index of predefined Linux callback client operations */ @@ -329,12 +328,12 @@ out: .p_bufsiz = MAX(NFS4_##argtype##_sz,NFS4_##restype##_sz) << 2, \ } -struct rpc_procinfo nfs4_cb_procedures[] = { +static struct rpc_procinfo nfs4_cb_procedures[] = { PROC(CB_NULL, NULL, enc_cb_null, dec_cb_null), PROC(CB_RECALL, COMPOUND, enc_cb_recall, dec_cb_recall), }; -struct rpc_version nfs_cb_version4 = { +static struct rpc_version nfs_cb_version4 = { .number = 1, .nrprocs = sizeof(nfs4_cb_procedures)/sizeof(nfs4_cb_procedures[0]), .procs = nfs4_cb_procedures @@ -348,7 +347,7 @@ static struct rpc_version * nfs_cb_version[] = { /* * Use the SETCLIENTID credential */ -struct rpc_cred * +static struct rpc_cred * nfsd4_lookupcred(struct nfs4_client *clp, int taskflags) { struct auth_cred acred; diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c index 4ba540841cf6..5605a26efc57 100644 --- a/fs/nfsd/nfs4idmap.c +++ b/fs/nfsd/nfs4idmap.c @@ -104,7 +104,7 @@ ent_update(struct ent *new, struct ent *itm) ent_init(new, itm); } -void +static void ent_put(struct cache_head *ch, struct cache_detail *cd) { if (cache_put(ch, cd)) { @@ -186,7 +186,7 @@ warn_no_idmapd(struct cache_detail *detail) static int idtoname_parse(struct cache_detail *, char *, int); static struct ent *idtoname_lookup(struct ent *, int); -struct cache_detail idtoname_cache = { +static struct cache_detail idtoname_cache = { .hash_size = ENT_HASHMAX, .hash_table = idtoname_table, .name = "nfs4.idtoname", @@ -277,7 +277,7 @@ nametoid_hash(struct ent *ent) return hash_str(ent->name, ENT_HASHBITS); } -void +static void nametoid_request(struct cache_detail *cd, struct cache_head *ch, char **bpp, int *blen) { @@ -317,9 +317,9 @@ nametoid_show(struct seq_file *m, struct cache_detail *cd, struct cache_head *h) } static struct ent *nametoid_lookup(struct ent *, int); -int nametoid_parse(struct cache_detail *, char *, int); +static int nametoid_parse(struct cache_detail *, char *, int); -struct cache_detail nametoid_cache = { +static struct cache_detail nametoid_cache = { .hash_size = ENT_HASHMAX, .hash_table = nametoid_table, .name = "nfs4.nametoid", @@ -330,7 +330,7 @@ struct cache_detail nametoid_cache = { .warn_no_listener = warn_no_idmapd, }; -int +static int nametoid_parse(struct cache_detail *cd, char *buf, int buflen) { struct ent ent, *res; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 1b2f67f5eef6..8a5f777b1e96 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -54,18 +54,21 @@ /* Globals */ static time_t lease_time = 90; /* default lease time */ static time_t user_lease_time = 90; -time_t boot_time; +static time_t boot_time; static int in_grace = 1; static u32 current_clientid = 1; static u32 current_ownerid = 1; static u32 current_fileid = 1; static u32 current_delegid = 1; static u32 nfs4_init; -stateid_t zerostateid; /* bits all 0 */ -stateid_t onestateid; /* bits all 1 */ +static stateid_t zerostateid; /* bits all 0 */ +static stateid_t onestateid; /* bits all 1 */ + +#define ZERO_STATEID(stateid) (!memcmp((stateid), &zerostateid, sizeof(stateid_t))) +#define ONE_STATEID(stateid) (!memcmp((stateid), &onestateid, sizeof(stateid_t))) /* forward declarations */ -struct nfs4_stateid * find_stateid(stateid_t *stid, int flags); +static struct nfs4_stateid * find_stateid(stateid_t *stid, int flags); static struct nfs4_delegation * find_delegation_stateid(struct inode *ino, stateid_t *stid); static void release_stateid_lockowners(struct nfs4_stateid *open_stp); @@ -77,10 +80,10 @@ static void release_stateid_lockowners(struct nfs4_stateid *open_stp); */ static DECLARE_MUTEX(client_sema); -kmem_cache_t *stateowner_slab = NULL; -kmem_cache_t *file_slab = NULL; -kmem_cache_t *stateid_slab = NULL; -kmem_cache_t *deleg_slab = NULL; +static kmem_cache_t *stateowner_slab = NULL; +static kmem_cache_t *file_slab = NULL; +static kmem_cache_t *stateid_slab = NULL; +static kmem_cache_t *deleg_slab = NULL; void nfs4_lock_state(void) @@ -116,7 +119,7 @@ static void release_stateid(struct nfs4_stateid *stp, int flags); */ /* recall_lock protects the del_recall_lru */ -spinlock_t recall_lock = SPIN_LOCK_UNLOCKED; +static spinlock_t recall_lock = SPIN_LOCK_UNLOCKED; static struct list_head del_recall_lru; static void @@ -456,7 +459,7 @@ check_name(struct xdr_netobj name) { return 1; } -void +static void add_to_unconfirmed(struct nfs4_client *clp, unsigned int strhashval) { unsigned int idhashval; @@ -468,7 +471,7 @@ add_to_unconfirmed(struct nfs4_client *clp, unsigned int strhashval) clp->cl_time = get_seconds(); } -void +static void move_to_confirmed(struct nfs4_client *clp) { unsigned int idhashval = clientid_hashval(clp->cl_clientid.cl_id); @@ -567,7 +570,7 @@ parse_octet(unsigned int *lenp, char **addrp) } /* parse and set the setclientid ipv4 callback address */ -int +static int parse_ipv4(unsigned int addr_len, char *addr_val, unsigned int *cbaddrp, unsigned short *cbportp) { int temp = 0; @@ -603,7 +606,7 @@ parse_ipv4(unsigned int addr_len, char *addr_val, unsigned int *cbaddrp, unsigne return 1; } -void +static void gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se) { struct nfs4_callback *cb = &clp->cl_callback; @@ -1186,7 +1189,7 @@ release_stateid(struct nfs4_stateid *stp, int flags) stp = NULL; } -void +static void move_to_close_lru(struct nfs4_stateowner *sop) { dprintk("NFSD: move_to_close_lru nfs4_stateowner %p\n", sop); @@ -1196,7 +1199,7 @@ move_to_close_lru(struct nfs4_stateowner *sop) sop->so_time = get_seconds(); } -void +static void release_state_owner(struct nfs4_stateid *stp, int flag) { struct nfs4_stateowner *sop = stp->st_stateowner; @@ -1250,7 +1253,7 @@ find_file(struct inode *ino) #define TEST_ACCESS(x) ((x > 0 || x < 4)?1:0) #define TEST_DENY(x) ((x >= 0 || x < 5)?1:0) -void +static void set_access(unsigned int *access, unsigned long bmap) { int i; @@ -1261,7 +1264,7 @@ set_access(unsigned int *access, unsigned long bmap) { } } -void +static void set_deny(unsigned int *deny, unsigned long bmap) { int i; @@ -1287,7 +1290,7 @@ test_share(struct nfs4_stateid *stp, struct nfsd4_open *open) { * Called to check deny when READ with all zero stateid or * WRITE with all zero or all one stateid */ -int +static int nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type) { struct inode *ino = current_fh->fh_dentry->d_inode; @@ -1442,7 +1445,7 @@ int nfsd_change_deleg_cb(struct file_lock **onlist, int arg) return -EAGAIN; } -struct lock_manager_operations nfsd_lease_mng_ops = { +static struct lock_manager_operations nfsd_lease_mng_ops = { .fl_break = nfsd_break_deleg_cb, .fl_release_private = nfsd_release_deleg_cb, .fl_copy_lock = nfsd_copy_lock_deleg_cb, @@ -1915,7 +1918,7 @@ end_grace(void) in_grace = 0; } -time_t +static time_t nfs4_laundromat(void) { struct nfs4_client *clp; @@ -1996,7 +1999,7 @@ laundromat_main(void *not_used) /* search ownerid_hashtbl[] and close_lru for stateid owner * (stateid->si_stateownerid) */ -struct nfs4_stateowner * +static struct nfs4_stateowner * find_openstateowner_id(u32 st_id, int flags) { struct nfs4_stateowner *local = NULL; @@ -2170,7 +2173,7 @@ out: /* * Checks for sequence id mutating operations. */ -int +static int nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *stateid, int flags, struct nfs4_stateowner **sopp, struct nfs4_stateid **stpp, clientid_t *lockclid) { int status; @@ -2486,7 +2489,7 @@ static struct list_head lock_ownerid_hashtbl[LOCK_HASH_SIZE]; static struct list_head lock_ownerstr_hashtbl[LOCK_HASH_SIZE]; static struct list_head lockstateid_hashtbl[STATEID_HASH_SIZE]; -struct nfs4_stateid * +static struct nfs4_stateid * find_stateid(stateid_t *stid, int flags) { struct nfs4_stateid *local = NULL; @@ -2550,7 +2553,7 @@ nfs4_transform_lock_offset(struct file_lock *lock) lock->fl_end = OFFSET_MAX; } -int +static int nfs4_verify_lock_stateowner(struct nfs4_stateowner *sop, unsigned int hashval) { struct nfs4_stateowner *local = NULL; @@ -2660,7 +2663,7 @@ alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp, str return sop; } -struct nfs4_stateid * +static struct nfs4_stateid * alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struct nfs4_stateid *open_stp) { struct nfs4_stateid *stp; @@ -2691,7 +2694,7 @@ out: return stp; } -int +static int check_lock_length(u64 offset, u64 length) { return ((length == 0) || ((length != ~(u64)0) && @@ -3149,7 +3152,7 @@ nfs4_release_reclaim(void) /* * called from OPEN, CLAIM_PREVIOUS with a new clientid. */ -struct nfs4_client_reclaim * +static struct nfs4_client_reclaim * nfs4_find_reclaim_client(clientid_t *clid) { unsigned int strhashval; diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index cfe978a72cea..91fb171d2ace 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -136,7 +136,7 @@ xdr_error: \ } \ } while (0) -u32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) +static u32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) { /* We want more bytes than seem to be available. * Maybe we need a new page, maybe we have just run out @@ -190,7 +190,7 @@ defer_free(struct nfsd4_compoundargs *argp, return 0; } -char *savemem(struct nfsd4_compoundargs *argp, u32 *p, int nbytes) +static char *savemem(struct nfsd4_compoundargs *argp, u32 *p, int nbytes) { void *new = NULL; if (p == argp->tmp) { diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index fdaa84addadb..0e18ae22127d 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -61,11 +61,6 @@ typedef struct { #define si_stateownerid si_opaque.so_stateownerid #define si_fileid si_opaque.so_fileid -extern stateid_t zerostateid; -extern stateid_t onestateid; - -#define ZERO_STATEID(stateid) (!memcmp((stateid), &zerostateid, sizeof(stateid_t))) -#define ONE_STATEID(stateid) (!memcmp((stateid), &onestateid, sizeof(stateid_t))) struct nfs4_cb_recall { u32 cbr_ident; @@ -271,12 +266,9 @@ struct nfs4_stateid { ((err) != nfserr_stale_stateid) && \ ((err) != nfserr_bad_stateid)) -extern time_t nfs4_laundromat(void); extern int nfsd4_renew(clientid_t *clid); extern int nfs4_preprocess_stateid_op(struct svc_fh *current_fh, stateid_t *stateid, int flags, struct file **filp); -extern int nfs4_share_conflict(struct svc_fh *current_fh, - unsigned int deny_type); extern void nfs4_lock_state(void); extern void nfs4_unlock_state(void); extern int nfs4_in_grace(void); -- cgit v1.2.3 From ea1da636e956ad1591a74904f23d98bbc26a644b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:17 -0700 Subject: [PATCH] knfsd: nfsd4: rename state list fields Trivial renaming patch: I can never remember, while looking at various lists relating the nfsd4 state structures, which are the "heads" and which are items on other lists, or which structures are actually on the various lists. The following convention helps me: given structures foo and bar, with foo containing the head of a list of bars, use "bars" for the name of the head of the list contained in the struct foo, and use "per_foo" for the entries in the struct bars. Already done for struct nfs4_file; go ahead and do it for the other nfsd4 state structures. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 78 +++++++++++++++++++++++----------------------- include/linux/nfsd/state.h | 18 +++++------ 2 files changed, 48 insertions(+), 48 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 884115198116..22e76e3f06a5 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -154,8 +154,8 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f dp = kmem_cache_alloc(deleg_slab, GFP_KERNEL); if (dp == NULL) return dp; - INIT_LIST_HEAD(&dp->dl_del_perfile); - INIT_LIST_HEAD(&dp->dl_del_perclnt); + INIT_LIST_HEAD(&dp->dl_perfile); + INIT_LIST_HEAD(&dp->dl_perclnt); INIT_LIST_HEAD(&dp->dl_recall_lru); dp->dl_client = clp; get_nfs4_file(fp); @@ -176,8 +176,8 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f current_fh->fh_handle.fh_size); dp->dl_time = 0; atomic_set(&dp->dl_count, 1); - list_add(&dp->dl_del_perfile, &fp->fi_delegations); - list_add(&dp->dl_del_perclnt, &clp->cl_del_perclnt); + list_add(&dp->dl_perfile, &fp->fi_delegations); + list_add(&dp->dl_perclnt, &clp->cl_delegations); return dp; } @@ -214,8 +214,8 @@ nfs4_close_delegation(struct nfs4_delegation *dp) static void unhash_delegation(struct nfs4_delegation *dp) { - list_del_init(&dp->dl_del_perfile); - list_del_init(&dp->dl_del_perclnt); + list_del_init(&dp->dl_perfile); + list_del_init(&dp->dl_perclnt); spin_lock(&recall_lock); list_del_init(&dp->dl_recall_lru); spin_unlock(&recall_lock); @@ -345,11 +345,11 @@ expire_client(struct nfs4_client *clp) INIT_LIST_HEAD(&reaplist); spin_lock(&recall_lock); - while (!list_empty(&clp->cl_del_perclnt)) { - dp = list_entry(clp->cl_del_perclnt.next, struct nfs4_delegation, dl_del_perclnt); + while (!list_empty(&clp->cl_delegations)) { + dp = list_entry(clp->cl_delegations.next, struct nfs4_delegation, dl_perclnt); dprintk("NFSD: expire client. dp %p, fp %p\n", dp, dp->dl_flock); - list_del_init(&dp->dl_del_perclnt); + list_del_init(&dp->dl_perclnt); list_move(&dp->dl_recall_lru, &reaplist); } spin_unlock(&recall_lock); @@ -361,8 +361,8 @@ expire_client(struct nfs4_client *clp) list_del(&clp->cl_idhash); list_del(&clp->cl_strhash); list_del(&clp->cl_lru); - while (!list_empty(&clp->cl_perclient)) { - sop = list_entry(clp->cl_perclient.next, struct nfs4_stateowner, so_perclient); + while (!list_empty(&clp->cl_openowners)) { + sop = list_entry(clp->cl_openowners.next, struct nfs4_stateowner, so_perclient); release_stateowner(sop); } put_nfs4_client(clp); @@ -380,8 +380,8 @@ create_client(struct xdr_netobj name, char *recdir) { clp->cl_callback.cb_parsed = 0; INIT_LIST_HEAD(&clp->cl_idhash); INIT_LIST_HEAD(&clp->cl_strhash); - INIT_LIST_HEAD(&clp->cl_perclient); - INIT_LIST_HEAD(&clp->cl_del_perclnt); + INIT_LIST_HEAD(&clp->cl_openowners); + INIT_LIST_HEAD(&clp->cl_delegations); INIT_LIST_HEAD(&clp->cl_lru); out: return clp; @@ -1074,13 +1074,13 @@ alloc_init_open_stateowner(unsigned int strhashval, struct nfs4_client *clp, str INIT_LIST_HEAD(&sop->so_idhash); INIT_LIST_HEAD(&sop->so_strhash); INIT_LIST_HEAD(&sop->so_perclient); - INIT_LIST_HEAD(&sop->so_perfilestate); - INIT_LIST_HEAD(&sop->so_perlockowner); /* not used */ + INIT_LIST_HEAD(&sop->so_stateids); + INIT_LIST_HEAD(&sop->so_perstateid); /* not used */ INIT_LIST_HEAD(&sop->so_close_lru); sop->so_time = 0; list_add(&sop->so_idhash, &ownerid_hashtbl[idhashval]); list_add(&sop->so_strhash, &ownerstr_hashtbl[strhashval]); - list_add(&sop->so_perclient, &clp->cl_perclient); + list_add(&sop->so_perclient, &clp->cl_openowners); sop->so_is_open_owner = 1; sop->so_id = current_ownerid++; sop->so_client = clp; @@ -1098,10 +1098,10 @@ release_stateid_lockowners(struct nfs4_stateid *open_stp) { struct nfs4_stateowner *lock_sop; - while (!list_empty(&open_stp->st_perlockowner)) { - lock_sop = list_entry(open_stp->st_perlockowner.next, - struct nfs4_stateowner, so_perlockowner); - /* list_del(&open_stp->st_perlockowner); */ + while (!list_empty(&open_stp->st_lockowners)) { + lock_sop = list_entry(open_stp->st_lockowners.next, + struct nfs4_stateowner, so_perstateid); + /* list_del(&open_stp->st_lockowners); */ BUG_ON(lock_sop->so_is_open_owner); release_stateowner(lock_sop); } @@ -1116,10 +1116,10 @@ unhash_stateowner(struct nfs4_stateowner *sop) list_del(&sop->so_strhash); if (sop->so_is_open_owner) list_del(&sop->so_perclient); - list_del(&sop->so_perlockowner); - while (!list_empty(&sop->so_perfilestate)) { - stp = list_entry(sop->so_perfilestate.next, - struct nfs4_stateid, st_perfilestate); + list_del(&sop->so_perstateid); + while (!list_empty(&sop->so_stateids)) { + stp = list_entry(sop->so_stateids.next, + struct nfs4_stateid, st_perstateowner); if (sop->so_is_open_owner) release_stateid(stp, OPEN_STATE); else @@ -1141,11 +1141,11 @@ init_stateid(struct nfs4_stateid *stp, struct nfs4_file *fp, struct nfsd4_open * unsigned int hashval = stateid_hashval(sop->so_id, fp->fi_id); INIT_LIST_HEAD(&stp->st_hash); - INIT_LIST_HEAD(&stp->st_perfilestate); - INIT_LIST_HEAD(&stp->st_perlockowner); + INIT_LIST_HEAD(&stp->st_perstateowner); + INIT_LIST_HEAD(&stp->st_lockowners); INIT_LIST_HEAD(&stp->st_perfile); list_add(&stp->st_hash, &stateid_hashtbl[hashval]); - list_add(&stp->st_perfilestate, &sop->so_perfilestate); + list_add(&stp->st_perstateowner, &sop->so_stateids); list_add(&stp->st_perfile, &fp->fi_stateids); stp->st_stateowner = sop; get_nfs4_file(fp); @@ -1167,7 +1167,7 @@ release_stateid(struct nfs4_stateid *stp, int flags) list_del(&stp->st_hash); list_del(&stp->st_perfile); - list_del(&stp->st_perfilestate); + list_del(&stp->st_perstateowner); if (flags & OPEN_STATE) { release_stateid_lockowners(stp); stp->st_vfs_file = NULL; @@ -1201,7 +1201,7 @@ release_state_owner(struct nfs4_stateid *stp, int flag) * released by the laundromat service after the lease period * to enable us to handle CLOSE replay */ - if (sop->so_confirmed && list_empty(&sop->so_perfilestate)) + if (sop->so_confirmed && list_empty(&sop->so_stateids)) move_to_close_lru(sop); } @@ -1548,7 +1548,7 @@ find_delegation_file(struct nfs4_file *fp, stateid_t *stid) { struct nfs4_delegation *dp; - list_for_each_entry(dp, &fp->fi_delegations, dl_del_perfile) { + list_for_each_entry(dp, &fp->fi_delegations, dl_perfile) { if (dp->dl_stateid.si_stateownerid == stid->si_stateownerid) return dp; } @@ -1892,7 +1892,7 @@ nfsd4_renew(clientid_t *clid) } renew_client(clp); status = nfserr_cb_path_down; - if (!list_empty(&clp->cl_del_perclnt) + if (!list_empty(&clp->cl_delegations) && !atomic_read(&clp->cl_callback.cb_set)) goto out; status = nfs_ok; @@ -2634,13 +2634,13 @@ alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp, str INIT_LIST_HEAD(&sop->so_idhash); INIT_LIST_HEAD(&sop->so_strhash); INIT_LIST_HEAD(&sop->so_perclient); - INIT_LIST_HEAD(&sop->so_perfilestate); - INIT_LIST_HEAD(&sop->so_perlockowner); + INIT_LIST_HEAD(&sop->so_stateids); + INIT_LIST_HEAD(&sop->so_perstateid); INIT_LIST_HEAD(&sop->so_close_lru); /* not used */ sop->so_time = 0; list_add(&sop->so_idhash, &lock_ownerid_hashtbl[idhashval]); list_add(&sop->so_strhash, &lock_ownerstr_hashtbl[strhashval]); - list_add(&sop->so_perlockowner, &open_stp->st_perlockowner); + list_add(&sop->so_perstateid, &open_stp->st_lockowners); sop->so_is_open_owner = 0; sop->so_id = current_ownerid++; sop->so_client = clp; @@ -2664,11 +2664,11 @@ alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struc goto out; INIT_LIST_HEAD(&stp->st_hash); INIT_LIST_HEAD(&stp->st_perfile); - INIT_LIST_HEAD(&stp->st_perfilestate); - INIT_LIST_HEAD(&stp->st_perlockowner); /* not used */ + INIT_LIST_HEAD(&stp->st_perstateowner); + INIT_LIST_HEAD(&stp->st_lockowners); /* not used */ list_add(&stp->st_hash, &lockstateid_hashtbl[hashval]); list_add(&stp->st_perfile, &fp->fi_stateids); - list_add(&stp->st_perfilestate, &sop->so_perfilestate); + list_add(&stp->st_perstateowner, &sop->so_stateids); stp->st_stateowner = sop; get_nfs4_file(fp); stp->st_file = fp; @@ -3081,8 +3081,8 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner * /* check for any locks held by any stateid * associated with the (lock) stateowner */ status = nfserr_locks_held; - list_for_each_entry(stp, &local->so_perfilestate, - st_perfilestate) { + list_for_each_entry(stp, &local->so_stateids, + st_perstateowner) { if (check_for_locks(stp->st_vfs_file, local)) goto out; } diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 0e18ae22127d..f4f27b76ee64 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -72,8 +72,8 @@ struct nfs4_cb_recall { }; struct nfs4_delegation { - struct list_head dl_del_perfile; /* nfs4_file->fi_del_perfile */ - struct list_head dl_del_perclnt; /* nfs4_client->cl_del_perclnt*/ + struct list_head dl_perfile; + struct list_head dl_perclnt; struct list_head dl_recall_lru; /* delegation recalled */ atomic_t dl_count; /* ref count */ struct nfs4_client *dl_client; @@ -119,8 +119,8 @@ struct nfs4_callback { struct nfs4_client { struct list_head cl_idhash; /* hash by cl_clientid.id */ struct list_head cl_strhash; /* hash by cl_name */ - struct list_head cl_perclient; /* list: stateowners */ - struct list_head cl_del_perclnt; /* list: delegations */ + struct list_head cl_openowners; + struct list_head cl_delegations; struct list_head cl_lru; /* tail queue */ struct xdr_netobj cl_name; /* id generated by client */ char cl_recdir[HEXDIR_LEN]; /* recovery dir */ @@ -195,9 +195,9 @@ struct nfs4_stateowner { struct kref so_ref; struct list_head so_idhash; /* hash by so_id */ struct list_head so_strhash; /* hash by op_name */ - struct list_head so_perclient; /* nfs4_client->cl_perclient */ - struct list_head so_perfilestate; /* list: nfs4_stateid */ - struct list_head so_perlockowner; /* nfs4_stateid->st_perlockowner */ + struct list_head so_perclient; + struct list_head so_stateids; + struct list_head so_perstateid; /* for lockowners only */ struct list_head so_close_lru; /* tail queue */ time_t so_time; /* time of placement on so_close_lru */ int so_is_open_owner; /* 1=openowner,0=lockowner */ @@ -240,8 +240,8 @@ struct nfs4_file { struct nfs4_stateid { struct list_head st_hash; struct list_head st_perfile; - struct list_head st_perfilestate; - struct list_head st_perlockowner; + struct list_head st_perstateowner; + struct list_head st_lockowners; struct nfs4_stateowner * st_stateowner; struct nfs4_file * st_file; stateid_t st_stateid; -- cgit v1.2.3 From cb36d6345752fa24827044c68e15f6708a40d9f6 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:23 -0700 Subject: [PATCH] knfsd: nfsd4: remove cb_parsed The cb_parsed field is only used by probe_callback, to determine whether the callback information has been filled in by setclientid. But there is no way that probe_callback() can be called without that having already happened, so that check is superfluous, as is cb_parsed. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4callback.c | 4 +--- fs/nfsd/nfs4state.c | 5 +---- include/linux/nfsd/state.h | 1 - 3 files changed, 2 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 68bb245491f6..583c0710e45e 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -386,9 +386,7 @@ nfsd4_probe_callback(struct nfs4_client *clp) char hostname[32]; int status; - dprintk("NFSD: probe_callback. cb_parsed %d cb_set %d\n", - cb->cb_parsed, atomic_read(&cb->cb_set)); - if (!cb->cb_parsed || atomic_read(&cb->cb_set)) + if (atomic_read(&cb->cb_set)) return; /* Initialize address */ diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 26d00465c28a..0b47a97e953d 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -377,7 +377,6 @@ create_client(struct xdr_netobj name, char *recdir) { memcpy(clp->cl_recdir, recdir, HEXDIR_LEN); atomic_set(&clp->cl_count, 1); atomic_set(&clp->cl_callback.cb_set, 0); - clp->cl_callback.cb_parsed = 0; INIT_LIST_HEAD(&clp->cl_idhash); INIT_LIST_HEAD(&clp->cl_strhash); INIT_LIST_HEAD(&clp->cl_openowners); @@ -620,14 +619,12 @@ gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se) goto out_err; cb->cb_prog = se->se_callback_prog; cb->cb_ident = se->se_callback_ident; - cb->cb_parsed = 1; return; out_err: printk(KERN_INFO "NFSD: this client (clientid %08x/%08x) " "will not receive delegations\n", clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id); - cb->cb_parsed = 0; return; } @@ -872,7 +869,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confi else { /* XXX: We just turn off callbacks until we can handle * change request correctly. */ - conf->cl_callback.cb_parsed = 0; + atomic_set(&conf->cl_callback.cb_set, 0); gen_confirm(conf); expire_client(unconf); status = nfs_ok; diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index f4f27b76ee64..83d29ec03a58 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -92,7 +92,6 @@ struct nfs4_delegation { /* client delegation callback info */ struct nfs4_callback { /* SETCLIENTID info */ - u32 cb_parsed; /* addr parsed */ u32 cb_addr; unsigned short cb_port; u32 cb_prog; -- cgit v1.2.3 From 190e4fbf96037e5e526ba3210f2bcc2a3b6fe964 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:25 -0700 Subject: [PATCH] knfsd: nfsd4: initialize recovery directory NFSv4 clients are required to know what state they have on the server so that they can reclaim it on server reboot. However, it is possible for pathalogical combinations of server reboots and network partitions to leave a client in a state where it cannot know whether it has lost its state on the server. For this reason, rfc3530 requires that we store some information about clients to stable storage. So we maintain a directory /var/lib/nfs/v4recovery with a subdirectory for each client with active state. We leave open the possibility of including files underneath each such subdirectory with information about the client, but for now the subdirectories are empty. We create a client subdirectory whenever a client makes its first non-reclaim open_confirm. We remove a client subdirectory whenever either a) its lease expires, or b) the grace period ends without it reclaiming anything. When handling reclaims, we allow the reclaim if and only if the client doing the reclaim has a subdirectory. This patch adds just the code to scan the recovery directory on nfsd startup. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4recover.c | 166 +++++++++++++++++++++++++++++++++++++++++++++ fs/nfsd/nfs4state.c | 18 ++++- include/linux/nfsd/state.h | 4 ++ 3 files changed, 186 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 841a305d7948..2dc9851a1d37 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -39,6 +39,9 @@ #include #include #include +#include +#include +#include #include #include #include @@ -46,6 +49,27 @@ #define NFSDDBG_FACILITY NFSDDBG_PROC +/* Globals */ +char recovery_dirname[PATH_MAX] = "/var/lib/nfs/v4recovery"; +static struct nameidata rec_dir; +static int rec_dir_init = 0; + +static void +nfs4_save_user(uid_t *saveuid, gid_t *savegid) +{ + *saveuid = current->fsuid; + *savegid = current->fsgid; + current->fsuid = 0; + current->fsgid = 0; +} + +static void +nfs4_reset_user(uid_t saveuid, gid_t savegid) +{ + current->fsuid = saveuid; + current->fsgid = savegid; +} + static void md5_to_hex(char *out, char *md5) { @@ -95,3 +119,145 @@ out: crypto_free_tfm(tfm); return status; } + +typedef int (recdir_func)(struct dentry *, struct dentry *); + +struct dentry_list { + struct dentry *dentry; + struct list_head list; +}; + +struct dentry_list_arg { + struct list_head dentries; + struct dentry *parent; +}; + +static int +nfsd4_build_dentrylist(void *arg, const char *name, int namlen, + loff_t offset, ino_t ino, unsigned int d_type) +{ + struct dentry_list_arg *dla = arg; + struct list_head *dentries = &dla->dentries; + struct dentry *parent = dla->parent; + struct dentry *dentry; + struct dentry_list *child; + + if (name && isdotent(name, namlen)) + return nfs_ok; + dentry = lookup_one_len(name, parent, namlen); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + child = kmalloc(sizeof(*child), GFP_KERNEL); + if (child == NULL) + return -ENOMEM; + child->dentry = dentry; + list_add(&child->list, dentries); + return 0; +} + +static int +nfsd4_list_rec_dir(struct dentry *dir, recdir_func *f) +{ + struct file *filp; + struct dentry_list_arg dla = { + .parent = dir, + }; + struct list_head *dentries = &dla.dentries; + struct dentry_list *child; + uid_t uid; + gid_t gid; + int status; + + if (!rec_dir_init) + return 0; + + nfs4_save_user(&uid, &gid); + + filp = dentry_open(dget(dir), mntget(rec_dir.mnt), + O_RDWR); + status = PTR_ERR(filp); + if (IS_ERR(filp)) + goto out; + INIT_LIST_HEAD(dentries); + status = vfs_readdir(filp, nfsd4_build_dentrylist, &dla); + fput(filp); + while (!list_empty(dentries)) { + child = list_entry(dentries->next, struct dentry_list, list); + status = f(dir, child->dentry); + if (status) + goto out; + list_del(&child->list); + dput(child->dentry); + kfree(child); + } +out: + while (!list_empty(dentries)) { + child = list_entry(dentries->next, struct dentry_list, list); + list_del(&child->list); + dput(child->dentry); + kfree(child); + } + nfs4_reset_user(uid, gid); + return status; +} + +static int +load_recdir(struct dentry *parent, struct dentry *child) +{ + if (child->d_name.len != HEXDIR_LEN - 1) { + printk("nfsd4: illegal name %s in recovery directory\n", + child->d_name.name); + /* Keep trying; maybe the others are OK: */ + return nfs_ok; + } + nfs4_client_to_reclaim(child->d_name.name); + return nfs_ok; +} + +int +nfsd4_recdir_load(void) { + int status; + + status = nfsd4_list_rec_dir(rec_dir.dentry, load_recdir); + if (status) + printk("nfsd4: failed loading clients from recovery" + " directory %s\n", rec_dir.dentry->d_name.name); + return status; +} + +/* + * Hold reference to the recovery directory. + */ + +void +nfsd4_init_recdir(char *rec_dirname) +{ + uid_t uid = 0; + gid_t gid = 0; + int status; + + printk("NFSD: Using %s as the NFSv4 state recovery directory\n", + rec_dirname); + + BUG_ON(rec_dir_init); + + nfs4_save_user(&uid, &gid); + + status = path_lookup(rec_dirname, LOOKUP_FOLLOW, &rec_dir); + if (status == -ENOENT) + printk("NFSD: recovery directory %s doesn't exist\n", + rec_dirname); + + if (!status) + rec_dir_init = 1; + nfs4_reset_user(uid, gid); +} + +void +nfsd4_shutdown_recdir(void) +{ + if (!rec_dir_init) + return; + rec_dir_init = 0; + path_release(&rec_dir); +} diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 0b47a97e953d..6b9d23c39afe 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -71,6 +71,7 @@ static stateid_t onestateid; /* bits all 1 */ static struct nfs4_stateid * find_stateid(stateid_t *stid, int flags); static struct nfs4_delegation * find_delegation_stateid(struct inode *ino, stateid_t *stid); static void release_stateid_lockowners(struct nfs4_stateid *open_stp); +extern char recovery_dirname[]; /* Locking: * @@ -3091,8 +3092,8 @@ alloc_reclaim(void) /* * failure => all reset bets are off, nfserr_no_grace... */ -static int -nfs4_client_to_reclaim(char *name) +int +nfs4_client_to_reclaim(const char *name) { unsigned int strhashval; struct nfs4_client_reclaim *crp = NULL; @@ -3202,6 +3203,17 @@ nfs4_state_init(void) reclaim_str_hashtbl_size = 0; } +static void +nfsd4_load_reboot_recovery_data(void) +{ + int status; + + nfsd4_init_recdir(recovery_dirname); + status = nfsd4_recdir_load(); + if (status) + printk("NFSD: Failure reading reboot recovery data\n"); +} + /* initialization to perform when the nfsd service is started: */ static void @@ -3228,6 +3240,7 @@ nfs4_state_start(void) status = nfsd4_init_slabs(); if (status) return status; + nfsd4_load_reboot_recovery_data(); __nfs4_state_start(); nfs4_init = 1; return 0; @@ -3286,6 +3299,7 @@ __nfs4_state_shutdown(void) cancel_delayed_work(&laundromat_work); flush_workqueue(laundry_wq); destroy_workqueue(laundry_wq); + nfsd4_shutdown_recdir(); nfs4_init = 0; } diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 83d29ec03a58..19481ab122df 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -278,6 +278,10 @@ extern void nfsd4_probe_callback(struct nfs4_client *clp); extern void nfsd4_cb_recall(struct nfs4_delegation *dp); extern void nfs4_put_delegation(struct nfs4_delegation *dp); extern int nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname); +extern void nfsd4_init_recdir(char *recdir_name); +extern int nfsd4_recdir_load(void); +extern void nfsd4_shutdown_recdir(void); +extern int nfs4_client_to_reclaim(const char *name); static inline void nfs4_put_stateowner(struct nfs4_stateowner *so) -- cgit v1.2.3 From c7b9a45927e74c81d6562153f7fde9d32da00159 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:30 -0700 Subject: [PATCH] knfsd: nfsd4: reboot recovery This patch adds the code to create and remove client subdirectories from the recovery directory, as described in the previous patch comment. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4recover.c | 169 +++++++++++++++++++++++++++++++++++++++++++++ fs/nfsd/nfs4state.c | 16 +++++ include/linux/nfsd/state.h | 5 ++ 3 files changed, 190 insertions(+) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 2dc9851a1d37..2805c5245eac 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -120,6 +120,70 @@ out: return status; } +static int +nfsd4_rec_fsync(struct dentry *dentry) +{ + struct file *filp; + int status = nfs_ok; + + dprintk("NFSD: nfs4_fsync_rec_dir\n"); + filp = dentry_open(dget(dentry), mntget(rec_dir.mnt), O_RDWR); + if (IS_ERR(filp)) { + status = PTR_ERR(filp); + goto out; + } + if (filp->f_op && filp->f_op->fsync) + status = filp->f_op->fsync(filp, filp->f_dentry, 0); + fput(filp); +out: + if (status) + printk("nfsd4: unable to sync recovery directory\n"); + return status; +} + +int +nfsd4_create_clid_dir(struct nfs4_client *clp) +{ + char *dname = clp->cl_recdir; + struct dentry *dentry; + uid_t uid; + gid_t gid; + int status; + + dprintk("NFSD: nfsd4_create_clid_dir for \"%s\"\n", dname); + + if (!rec_dir_init || clp->cl_firststate) + return 0; + + nfs4_save_user(&uid, &gid); + + /* lock the parent */ + down(&rec_dir.dentry->d_inode->i_sem); + + dentry = lookup_one_len(dname, rec_dir.dentry, HEXDIR_LEN-1); + if (IS_ERR(dentry)) { + status = PTR_ERR(dentry); + goto out_unlock; + } + status = -EEXIST; + if (dentry->d_inode) { + dprintk("NFSD: nfsd4_create_clid_dir: DIRECTORY EXISTS\n"); + goto out_put; + } + status = vfs_mkdir(rec_dir.dentry->d_inode, dentry, S_IRWXU); +out_put: + dput(dentry); +out_unlock: + up(&rec_dir.dentry->d_inode->i_sem); + if (status == 0) { + clp->cl_firststate = 1; + status = nfsd4_rec_fsync(rec_dir.dentry); + } + nfs4_reset_user(uid, gid); + dprintk("NFSD: nfsd4_create_clid_dir returns %d\n", status); + return status; +} + typedef int (recdir_func)(struct dentry *, struct dentry *); struct dentry_list { @@ -201,6 +265,111 @@ out: return status; } +static int +nfsd4_remove_clid_file(struct dentry *dir, struct dentry *dentry) +{ + int status; + + if (!S_ISREG(dir->d_inode->i_mode)) { + printk("nfsd4: non-file found in client recovery directory\n"); + return -EINVAL; + } + down(&dir->d_inode->i_sem); + status = vfs_unlink(dir->d_inode, dentry); + up(&dir->d_inode->i_sem); + return status; +} + +static int +nfsd4_clear_clid_dir(struct dentry *dir, struct dentry *dentry) +{ + int status; + + /* For now this directory should already be empty, but we empty it of + * any regular files anyway, just in case the directory was created by + * a kernel from the future.... */ + nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file); + down(&dir->d_inode->i_sem); + status = vfs_rmdir(dir->d_inode, dentry); + up(&dir->d_inode->i_sem); + return status; +} + +static int +nfsd4_unlink_clid_dir(char *name, int namlen) +{ + struct dentry *dentry; + int status; + + dprintk("NFSD: nfsd4_unlink_clid_dir. name %.*s\n", namlen, name); + + dentry = lookup_one_len(name, rec_dir.dentry, namlen); + if (IS_ERR(dentry)) { + status = PTR_ERR(dentry); + return status; + } + status = -ENOENT; + if (!dentry->d_inode) + goto out; + + status = nfsd4_clear_clid_dir(rec_dir.dentry, dentry); +out: + dput(dentry); + return status; +} + +void +nfsd4_remove_clid_dir(struct nfs4_client *clp) +{ + uid_t uid; + gid_t gid; + int status; + + if (!rec_dir_init || !clp->cl_firststate) + return; + + nfs4_save_user(&uid, &gid); + status = nfsd4_unlink_clid_dir(clp->cl_recdir, HEXDIR_LEN-1); + nfs4_reset_user(uid, gid); + if (status == 0) + status = nfsd4_rec_fsync(rec_dir.dentry); + if (status) + printk("NFSD: Failed to remove expired client state directory" + " %.*s\n", HEXDIR_LEN, clp->cl_recdir); + return; +} + +static int +purge_old(struct dentry *parent, struct dentry *child) +{ + int status; + + if (nfs4_has_reclaimed_state(child->d_name.name)) + return nfs_ok; + + status = nfsd4_clear_clid_dir(parent, child); + if (status) + printk("failed to remove client recovery directory %s\n", + child->d_name.name); + /* Keep trying, success or failure: */ + return nfs_ok; +} + +void +nfsd4_recdir_purge_old(void) { + int status; + + if (!rec_dir_init) + return; + status = nfsd4_list_rec_dir(rec_dir.dentry, purge_old); + if (status == 0) + status = nfsd4_rec_fsync(rec_dir.dentry); + if (status) + printk("nfsd4: failed to purge old clients from recovery" + " directory %s\n", rec_dir.dentry->d_name.name); + return; +} + static int load_recdir(struct dentry *parent, struct dentry *child) { diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 6b9d23c39afe..6cca358cd650 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -905,6 +905,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confi conf = find_confirmed_client_by_str(unconf->cl_recdir, hash); if (conf) { + nfsd4_remove_clid_dir(conf); expire_client(conf); } move_to_confirmed(unconf); @@ -1691,6 +1692,7 @@ nfs4_set_claim_prev(struct nfsd4_open *open, int *status) *status = nfserr_reclaim_bad; else { open->op_stateowner->so_confirmed = 1; + open->op_stateowner->so_client->cl_firststate = 1; open->op_stateowner->so_seqid--; } } @@ -1903,6 +1905,7 @@ static void end_grace(void) { dprintk("NFSD: end of grace period\n"); + nfsd4_recdir_purge_old(); in_grace = 0; } @@ -1932,6 +1935,7 @@ nfs4_laundromat(void) } dprintk("NFSD: purging unused client (clientid %08x)\n", clp->cl_clientid.cl_id); + nfsd4_remove_clid_dir(clp); expire_client(clp); } INIT_LIST_HEAD(&reaplist); @@ -2320,6 +2324,8 @@ nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs stp->st_stateid.si_stateownerid, stp->st_stateid.si_fileid, stp->st_stateid.si_generation); + + nfsd4_create_clid_dir(sop->so_client); out: if (oc->oc_stateowner) nfs4_get_stateowner(oc->oc_stateowner); @@ -3089,6 +3095,16 @@ alloc_reclaim(void) return kmalloc(sizeof(struct nfs4_client_reclaim), GFP_KERNEL); } +int +nfs4_has_reclaimed_state(const char *name) +{ + unsigned int strhashval = clientstr_hashval(name); + struct nfs4_client *clp; + + clp = find_confirmed_client_by_str(name, strhashval); + return clp ? 1 : 0; +} + /* * failure => all reset bets are off, nfserr_no_grace... */ diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 19481ab122df..a84a3fa99be1 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -131,6 +131,7 @@ struct nfs4_client { nfs4_verifier cl_confirm; /* generated by server */ struct nfs4_callback cl_callback; /* callback info */ atomic_t cl_count; /* ref count */ + u32 cl_firststate; /* recovery dir creation */ }; /* struct nfs4_client_reset @@ -282,6 +283,10 @@ extern void nfsd4_init_recdir(char *recdir_name); extern int nfsd4_recdir_load(void); extern void nfsd4_shutdown_recdir(void); extern int nfs4_client_to_reclaim(const char *name); +extern int nfs4_has_reclaimed_state(const char *name); +extern void nfsd4_recdir_purge_old(void); +extern int nfsd4_create_clid_dir(struct nfs4_client *clp); +extern void nfsd4_remove_clid_dir(struct nfs4_client *clp); static inline void nfs4_put_stateowner(struct nfs4_stateowner *so) -- cgit v1.2.3 From 0964a3d3f1aa96468091924f6b0c391a46dc6d0b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:32 -0700 Subject: [PATCH] knfsd: nfsd4 reboot dirname fix Set the recovery directory via /proc/fs/nfsd/nfs4recoverydir. It may be changed any time, but is used only on startup. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4recover.c | 1 - fs/nfsd/nfs4state.c | 37 +++++++++++++++++++++++++++++++++++-- fs/nfsd/nfsctl.c | 23 +++++++++++++++++++++++ include/linux/nfsd/nfsd.h | 2 ++ 4 files changed, 60 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 2805c5245eac..095f1740f3ae 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -50,7 +50,6 @@ #define NFSDDBG_FACILITY NFSDDBG_PROC /* Globals */ -char recovery_dirname[PATH_MAX] = "/var/lib/nfs/v4recovery"; static struct nameidata rec_dir; static int rec_dir_init = 0; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 6cca358cd650..89e36526d7f2 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -48,6 +48,7 @@ #include #include #include +#include #define NFSDDBG_FACILITY NFSDDBG_PROC @@ -71,7 +72,8 @@ static stateid_t onestateid; /* bits all 1 */ static struct nfs4_stateid * find_stateid(stateid_t *stid, int flags); static struct nfs4_delegation * find_delegation_stateid(struct inode *ino, stateid_t *stid); static void release_stateid_lockowners(struct nfs4_stateid *open_stp); -extern char recovery_dirname[]; +static char user_recovery_dirname[PATH_MAX] = "/var/lib/nfs/v4recovery"; +static void nfs4_set_recdir(char *recdir); /* Locking: * @@ -3224,8 +3226,10 @@ nfsd4_load_reboot_recovery_data(void) { int status; - nfsd4_init_recdir(recovery_dirname); + nfs4_lock_state(); + nfsd4_init_recdir(user_recovery_dirname); status = nfsd4_recdir_load(); + nfs4_unlock_state(); if (status) printk("NFSD: Failure reading reboot recovery data\n"); } @@ -3329,6 +3333,35 @@ nfs4_state_shutdown(void) nfs4_unlock_state(); } +static void +nfs4_set_recdir(char *recdir) +{ + nfs4_lock_state(); + strcpy(user_recovery_dirname, recdir); + nfs4_unlock_state(); +} + +/* + * Change the NFSv4 recovery directory to recdir. + */ +int +nfs4_reset_recoverydir(char *recdir) +{ + int status; + struct nameidata nd; + + status = path_lookup(recdir, LOOKUP_FOLLOW, &nd); + if (status) + return status; + status = -ENOTDIR; + if (S_ISDIR(nd.dentry->d_inode->i_mode)) { + nfs4_set_recdir(recdir); + status = 0; + } + path_release(&nd); + return status; +} + /* * Called when leasetime is changed. * diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 3da43a3ed32c..841c562991e8 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -51,6 +51,7 @@ enum { NFSD_Fh, NFSD_Threads, NFSD_Leasetime, + NFSD_RecoveryDir, }; /* @@ -66,6 +67,7 @@ static ssize_t write_getfs(struct file *file, char *buf, size_t size); static ssize_t write_filehandle(struct file *file, char *buf, size_t size); static ssize_t write_threads(struct file *file, char *buf, size_t size); static ssize_t write_leasetime(struct file *file, char *buf, size_t size); +static ssize_t write_recoverydir(struct file *file, char *buf, size_t size); static ssize_t (*write_op[])(struct file *, char *, size_t) = { [NFSD_Svc] = write_svc, @@ -78,6 +80,7 @@ static ssize_t (*write_op[])(struct file *, char *, size_t) = { [NFSD_Fh] = write_filehandle, [NFSD_Threads] = write_threads, [NFSD_Leasetime] = write_leasetime, + [NFSD_RecoveryDir] = write_recoverydir, }; static ssize_t nfsctl_transaction_write(struct file *file, const char __user *buf, size_t size, loff_t *pos) @@ -349,6 +352,25 @@ static ssize_t write_leasetime(struct file *file, char *buf, size_t size) return strlen(buf); } +static ssize_t write_recoverydir(struct file *file, char *buf, size_t size) +{ + char *mesg = buf; + char *recdir; + int len, status; + + if (size > PATH_MAX || buf[size-1] != '\n') + return -EINVAL; + buf[size-1] = 0; + + recdir = mesg; + len = qword_get(&mesg, recdir, size); + if (len <= 0) + return -EINVAL; + + status = nfs4_reset_recoverydir(recdir); + return strlen(buf); +} + /*----------------------------------------------------------------------------*/ /* * populating the filesystem. @@ -369,6 +391,7 @@ static int nfsd_fill_super(struct super_block * sb, void * data, int silent) [NFSD_Threads] = {"threads", &transaction_ops, S_IWUSR|S_IRUSR}, #ifdef CONFIG_NFSD_V4 [NFSD_Leasetime] = {"nfsv4leasetime", &transaction_ops, S_IWUSR|S_IRUSR}, + [NFSD_RecoveryDir] = {"nfsv4recoverydir", &transaction_ops, S_IWUSR|S_IRUSR}, #endif /* last one */ {""} }; diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 21c6e9d86e4f..5791dfd30dd0 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -150,12 +150,14 @@ int nfs4_state_start(void); void nfs4_state_shutdown(void); time_t nfs4_lease_time(void); void nfs4_reset_lease(time_t leasetime); +int nfs4_reset_recoverydir(char *recdir); #else static inline void nfs4_state_init(void){}; static inline int nfs4_state_start(void){return 0;} static inline void nfs4_state_shutdown(void){} static inline time_t nfs4_lease_time(void){return 0;} static inline void nfs4_reset_lease(time_t leasetime){} +static inline int nfs4_reset_recoverydir(char *recdir) {return 0;} #endif /* -- cgit v1.2.3 From c988d2b2845495373f666a381d354a7f80981d62 Mon Sep 17 00:00:00 2001 From: Matt Domsch Date: Thu, 23 Jun 2005 22:05:15 -0700 Subject: [PATCH] modules: add version and srcversion to sysfs This patch adds version and srcversion files to /sys/module/${modulename} containing the version and srcversion fields of the module's modinfo section (if present). /sys/module/e1000 |-- srcversion `-- version This patch differs slightly from the version posted in January, as it now uses the new kstrdup() call in -mm. Why put this in sysfs? a) Tools like DKMS, which deal with changing out individual kernel modules without replacing the whole kernel, can behave smarter if they can tell the version of a given module. The autoinstaller feature, for example, which determines if your system has a "good" version of a driver (i.e. if the one provided by DKMS has a newer verson than that provided by the kernel package installed), and to automatically compile and install a newer version if DKMS has it but your kernel doesn't yet have that version. b) Because sysadmins manually, or with tools like DKMS, can switch out modules on the file system, you can't count on 'modinfo foo.ko', which looks at /lib/modules/${kernelver}/... actually matching what is loaded into the kernel already. Hence asking sysfs for this. c) as the unbind-driver-from-device work takes shape, it will be possible to rebind a driver that's built-in (no .ko to modinfo for the version) to a newly loaded module. sysfs will have the currently-built-in version info, for comparison. d) tech support scripts can then easily grab the version info for what's running presently - a question I get often. There has been renewed interest in this patch on linux-scsi by driver authors. As the idea originated from GregKH, I leave his Signed-off-by: intact, though the implementation is nearly completely new. Compiled and run on x86 and x86_64. From: Matthew Dobson build fix From: Thierry Vignaud build fix From: Matthew Dobson warning fix Signed-off-by: Greg Kroah-Hartman Signed-off-by: Matt Domsch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/module.h | 5 +++ kernel/module.c | 95 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 100 insertions(+) (limited to 'include/linux') diff --git a/include/linux/module.h b/include/linux/module.h index 0e432a0f4aee..f05372b7fe77 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -51,6 +51,9 @@ struct module_attribute { ssize_t (*show)(struct module_attribute *, struct module *, char *); ssize_t (*store)(struct module_attribute *, struct module *, const char *, size_t count); + void (*setup)(struct module *, const char *); + int (*test)(struct module *); + void (*free)(struct module *); }; struct module_kobject @@ -239,6 +242,8 @@ struct module /* Sysfs stuff. */ struct module_kobject mkobj; struct module_param_attrs *param_attrs; + const char *version; + const char *srcversion; /* Exported symbols */ const struct kernel_symbol *syms; diff --git a/kernel/module.c b/kernel/module.c index a566745dde62..0494c89a0d26 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include #include @@ -370,6 +371,43 @@ static inline void percpu_modcopy(void *pcpudst, const void *src, #endif /* CONFIG_SMP */ #ifdef CONFIG_MODULE_UNLOAD +#define MODINFO_ATTR(field) \ +static void setup_modinfo_##field(struct module *mod, const char *s) \ +{ \ + mod->field = kstrdup(s, GFP_KERNEL); \ +} \ +static ssize_t show_modinfo_##field(struct module_attribute *mattr, \ + struct module *mod, char *buffer) \ +{ \ + return sprintf(buffer, "%s\n", mod->field); \ +} \ +static int modinfo_##field##_exists(struct module *mod) \ +{ \ + return mod->field != NULL; \ +} \ +static void free_modinfo_##field(struct module *mod) \ +{ \ + kfree(mod->field); \ + mod->field = NULL; \ +} \ +static struct module_attribute modinfo_##field = { \ + .attr = { .name = __stringify(field), .mode = 0444, \ + .owner = THIS_MODULE }, \ + .show = show_modinfo_##field, \ + .setup = setup_modinfo_##field, \ + .test = modinfo_##field##_exists, \ + .free = free_modinfo_##field, \ +}; + +MODINFO_ATTR(version); +MODINFO_ATTR(srcversion); + +static struct module_attribute *modinfo_attrs[] = { + &modinfo_version, + &modinfo_srcversion, + NULL, +}; + /* Init the unload section of the module. */ static void module_unload_init(struct module *mod) { @@ -1031,6 +1069,32 @@ static void module_remove_refcnt_attr(struct module *mod) } #endif +#ifdef CONFIG_MODULE_UNLOAD +static int module_add_modinfo_attrs(struct module *mod) +{ + struct module_attribute *attr; + int error = 0; + int i; + + for (i = 0; (attr = modinfo_attrs[i]) && !error; i++) { + if (!attr->test || + (attr->test && attr->test(mod))) + error = sysfs_create_file(&mod->mkobj.kobj,&attr->attr); + } + return error; +} + +static void module_remove_modinfo_attrs(struct module *mod) +{ + struct module_attribute *attr; + int i; + + for (i = 0; (attr = modinfo_attrs[i]); i++) { + sysfs_remove_file(&mod->mkobj.kobj,&attr->attr); + attr->free(mod); + } +} +#endif static int mod_sysfs_setup(struct module *mod, struct kernel_param *kparam, @@ -1056,6 +1120,12 @@ static int mod_sysfs_setup(struct module *mod, if (err) goto out_unreg; +#ifdef CONFIG_MODULE_UNLOAD + err = module_add_modinfo_attrs(mod); + if (err) + goto out_unreg; +#endif + return 0; out_unreg: @@ -1066,6 +1136,9 @@ out: static void mod_kobject_remove(struct module *mod) { +#ifdef CONFIG_MODULE_UNLOAD + module_remove_modinfo_attrs(mod); +#endif module_remove_refcnt_attr(mod); module_param_sysfs_remove(mod); @@ -1311,6 +1384,23 @@ static char *get_modinfo(Elf_Shdr *sechdrs, return NULL; } +#ifdef CONFIG_MODULE_UNLOAD +static void setup_modinfo(struct module *mod, Elf_Shdr *sechdrs, + unsigned int infoindex) +{ + struct module_attribute *attr; + int i; + + for (i = 0; (attr = modinfo_attrs[i]); i++) { + if (attr->setup) + attr->setup(mod, + get_modinfo(sechdrs, + infoindex, + attr->attr.name)); + } +} +#endif + #ifdef CONFIG_KALLSYMS int is_exported(const char *name, const struct module *mod) { @@ -1615,6 +1705,11 @@ static struct module *load_module(void __user *umod, /* Set up license info based on the info section */ set_license(mod, get_modinfo(sechdrs, infoindex, "license")); +#ifdef CONFIG_MODULE_UNLOAD + /* Set up MODINFO_ATTR fields */ + setup_modinfo(mod, sechdrs, infoindex); +#endif + /* Fix up syms, so that st_value is a pointer to location. */ err = simplify_symbols(sechdrs, symindex, strtab, versindex, pcpuindex, mod); -- cgit v1.2.3 From 420edbcc09008342c7b2665453f6b370739aadb0 Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Thu, 23 Jun 2005 22:05:23 -0700 Subject: [PATCH] xip: bdev: execute in place This is the block device related part. The block device operation direct_access now has a struct block_device as first parameter. Signed-off-by: Carsten Otte Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/s390/block/dcssblk.c | 44 ++++++++++++++++++++++++++++++++++++++++---- include/linux/fs.h | 1 + 2 files changed, 41 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c index 16ab8d363ac6..6bc27d52326f 100644 --- a/drivers/s390/block/dcssblk.c +++ b/drivers/s390/block/dcssblk.c @@ -35,14 +35,17 @@ static int dcssblk_open(struct inode *inode, struct file *filp); static int dcssblk_release(struct inode *inode, struct file *filp); static int dcssblk_make_request(struct request_queue *q, struct bio *bio); +static int dcssblk_direct_access(struct block_device *bdev, sector_t secnum, + unsigned long *data); static char dcssblk_segments[DCSSBLK_PARM_LEN] = "\0"; static int dcssblk_major; static struct block_device_operations dcssblk_devops = { - .owner = THIS_MODULE, - .open = dcssblk_open, - .release = dcssblk_release, + .owner = THIS_MODULE, + .open = dcssblk_open, + .release = dcssblk_release, + .direct_access = dcssblk_direct_access, }; static ssize_t dcssblk_add_store(struct device * dev, struct device_attribute *attr, const char * buf, @@ -641,6 +644,20 @@ dcssblk_make_request(request_queue_t *q, struct bio *bio) /* Request beyond end of DCSS segment. */ goto fail; } + /* verify data transfer direction */ + if (dev_info->is_shared) { + switch (dev_info->segment_type) { + case SEG_TYPE_SR: + case SEG_TYPE_ER: + case SEG_TYPE_SC: + /* cannot write to these segments */ + if (bio_data_dir(bio) == WRITE) { + PRINT_WARN("rejecting write to ro segment %s\n", dev_info->dev.bus_id); + goto fail; + } + } + } + index = (bio->bi_sector >> 3); bio_for_each_segment(bvec, bio, i) { page_addr = (unsigned long) @@ -661,7 +678,26 @@ dcssblk_make_request(request_queue_t *q, struct bio *bio) bio_endio(bio, bytes_done, 0); return 0; fail: - bio_io_error(bio, bytes_done); + bio_io_error(bio, bio->bi_size); + return 0; +} + +static int +dcssblk_direct_access (struct block_device *bdev, sector_t secnum, + unsigned long *data) +{ + struct dcssblk_dev_info *dev_info; + unsigned long pgoff; + + dev_info = bdev->bd_disk->private_data; + if (!dev_info) + return -ENODEV; + if (secnum % (PAGE_SIZE/512)) + return -EINVAL; + pgoff = secnum / (PAGE_SIZE / 512); + if ((pgoff+1)*PAGE_SIZE-1 > dev_info->end - dev_info->start) + return -ERANGE; + *data = (unsigned long) (dev_info->start+pgoff*PAGE_SIZE); return 0; } diff --git a/include/linux/fs.h b/include/linux/fs.h index 83857d8070d3..929bf8d20c87 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -885,6 +885,7 @@ struct block_device_operations { int (*ioctl) (struct inode *, struct file *, unsigned, unsigned long); long (*unlocked_ioctl) (struct file *, unsigned, unsigned long); long (*compat_ioctl) (struct file *, unsigned, unsigned long); + int (*direct_access) (struct block_device *, sector_t, unsigned long *); int (*media_changed) (struct gendisk *); int (*revalidate_disk) (struct gendisk *); struct module *owner; -- cgit v1.2.3 From ceffc078528befc008c6f2c2c4decda79eabd534 Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Thu, 23 Jun 2005 22:05:25 -0700 Subject: [PATCH] xip: fs/mm: execute in place - generic_file* file operations do no longer have a xip/non-xip split - filemap_xip.c implements a new set of fops that require get_xip_page aop to work proper. all new fops are exported GPL-only (don't like to see whatever code use those except GPL modules) - __xip_unmap now uses page_check_address, which is no longer static in rmap.c, and defined in linux/rmap.h - mm/filemap.h is now much more clean, plainly having just Linus' inline funcs moved here from filemap.c - fix includes in filemap_xip to make it build cleanly on i386 Signed-off-by: Carsten Otte Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/open.c | 4 +- include/linux/fs.h | 18 ++ include/linux/rmap.h | 6 + mm/Makefile | 1 + mm/filemap.c | 74 +------ mm/filemap.h | 94 +++++++++ mm/filemap_xip.c | 581 +++++++++++++++++++++++++++++++++++++++++++++++++++ mm/rmap.c | 4 +- 8 files changed, 707 insertions(+), 75 deletions(-) create mode 100644 mm/filemap.h create mode 100644 mm/filemap_xip.c (limited to 'include/linux') diff --git a/fs/open.c b/fs/open.c index 8ec63f735918..3f4a4286fdc4 100644 --- a/fs/open.c +++ b/fs/open.c @@ -808,7 +808,9 @@ struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags) /* NB: we're sure to have correct a_ops only after f_op->open */ if (f->f_flags & O_DIRECT) { - if (!f->f_mapping->a_ops || !f->f_mapping->a_ops->direct_IO) { + if (!f->f_mapping->a_ops || + ((!f->f_mapping->a_ops->direct_IO) && + (!f->f_mapping->a_ops->get_xip_page))) { fput(f); f = ERR_PTR(-EINVAL); } diff --git a/include/linux/fs.h b/include/linux/fs.h index 929bf8d20c87..79c0fafc0211 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -330,6 +330,8 @@ struct address_space_operations { int (*releasepage) (struct page *, int); ssize_t (*direct_IO)(int, struct kiocb *, const struct iovec *iov, loff_t offset, unsigned long nr_segs); + struct page* (*get_xip_page)(struct address_space *, sector_t, + int); }; struct backing_dev_info; @@ -1497,6 +1499,22 @@ extern loff_t remote_llseek(struct file *file, loff_t offset, int origin); extern int generic_file_open(struct inode * inode, struct file * filp); extern int nonseekable_open(struct inode * inode, struct file * filp); +#ifdef CONFIG_FS_XIP +extern ssize_t xip_file_aio_read(struct kiocb *iocb, char __user *buf, + size_t count, loff_t pos); +extern ssize_t xip_file_readv(struct file *filp, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos); +extern ssize_t xip_file_sendfile(struct file *in_file, loff_t *ppos, + size_t count, read_actor_t actor, + void *target); +extern int xip_file_mmap(struct file * file, struct vm_area_struct * vma); +extern ssize_t xip_file_aio_write(struct kiocb *iocb, const char __user *buf, + size_t count, loff_t pos); +extern ssize_t xip_file_writev(struct file *file, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos); +extern int xip_truncate_page(struct address_space *mapping, loff_t from); +#endif + static inline void do_generic_file_read(struct file * filp, loff_t *ppos, read_descriptor_t * desc, read_actor_t actor) diff --git a/include/linux/rmap.h b/include/linux/rmap.h index 11b484e37ac9..e80fb7ee6efd 100644 --- a/include/linux/rmap.h +++ b/include/linux/rmap.h @@ -92,6 +92,12 @@ static inline void page_dup_rmap(struct page *page) int page_referenced(struct page *, int is_locked, int ignore_token); int try_to_unmap(struct page *); +/* + * Called from mm/filemap_xip.c to unmap empty zero page + */ +pte_t *page_check_address(struct page *, struct mm_struct *, unsigned long); + + /* * Used by swapoff to help locate where page is expected in vma. */ diff --git a/mm/Makefile b/mm/Makefile index 8f70ffd763c8..4cd69e3ce421 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -19,3 +19,4 @@ obj-$(CONFIG_SPARSEMEM) += sparse.o obj-$(CONFIG_SHMEM) += shmem.o obj-$(CONFIG_TINY_SHMEM) += tiny-shmem.o +obj-$(CONFIG_FS_XIP) += filemap_xip.o diff --git a/mm/filemap.c b/mm/filemap.c index a3598b542a31..7332194d7afd 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -28,6 +28,7 @@ #include #include #include +#include "filemap.h" /* * FIXME: remove all knowledge of the buffer layer from the core VM */ @@ -1714,32 +1715,7 @@ int remove_suid(struct dentry *dentry) } EXPORT_SYMBOL(remove_suid); -/* - * Copy as much as we can into the page and return the number of bytes which - * were sucessfully copied. If a fault is encountered then clear the page - * out to (offset+bytes) and return the number of bytes which were copied. - */ -static inline size_t -filemap_copy_from_user(struct page *page, unsigned long offset, - const char __user *buf, unsigned bytes) -{ - char *kaddr; - int left; - - kaddr = kmap_atomic(page, KM_USER0); - left = __copy_from_user_inatomic(kaddr + offset, buf, bytes); - kunmap_atomic(kaddr, KM_USER0); - - if (left != 0) { - /* Do it the slow way */ - kaddr = kmap(page); - left = __copy_from_user(kaddr + offset, buf, bytes); - kunmap(page); - } - return bytes - left; -} - -static size_t +size_t __filemap_copy_from_user_iovec(char *vaddr, const struct iovec *iov, size_t base, size_t bytes) { @@ -1766,52 +1742,6 @@ __filemap_copy_from_user_iovec(char *vaddr, return copied - left; } -/* - * This has the same sideeffects and return value as filemap_copy_from_user(). - * The difference is that on a fault we need to memset the remainder of the - * page (out to offset+bytes), to emulate filemap_copy_from_user()'s - * single-segment behaviour. - */ -static inline size_t -filemap_copy_from_user_iovec(struct page *page, unsigned long offset, - const struct iovec *iov, size_t base, size_t bytes) -{ - char *kaddr; - size_t copied; - - kaddr = kmap_atomic(page, KM_USER0); - copied = __filemap_copy_from_user_iovec(kaddr + offset, iov, - base, bytes); - kunmap_atomic(kaddr, KM_USER0); - if (copied != bytes) { - kaddr = kmap(page); - copied = __filemap_copy_from_user_iovec(kaddr + offset, iov, - base, bytes); - kunmap(page); - } - return copied; -} - -static inline void -filemap_set_next_iovec(const struct iovec **iovp, size_t *basep, size_t bytes) -{ - const struct iovec *iov = *iovp; - size_t base = *basep; - - while (bytes) { - int copy = min(bytes, iov->iov_len - base); - - bytes -= copy; - base += copy; - if (iov->iov_len == base) { - iov++; - base = 0; - } - } - *iovp = iov; - *basep = base; -} - /* * Performs necessary checks before doing a write * diff --git a/mm/filemap.h b/mm/filemap.h new file mode 100644 index 000000000000..c2d0546a57eb --- /dev/null +++ b/mm/filemap.h @@ -0,0 +1,94 @@ +/* + * linux/mm/filemap.h + * + * Copyright (C) 1994-1999 Linus Torvalds + */ + +#ifndef __FILEMAP_H +#define __FILEMAP_H + +#include +#include +#include +#include +#include +#include +#include + +extern size_t +__filemap_copy_from_user_iovec(char *vaddr, + const struct iovec *iov, + size_t base, + size_t bytes); + +/* + * Copy as much as we can into the page and return the number of bytes which + * were sucessfully copied. If a fault is encountered then clear the page + * out to (offset+bytes) and return the number of bytes which were copied. + */ +static inline size_t +filemap_copy_from_user(struct page *page, unsigned long offset, + const char __user *buf, unsigned bytes) +{ + char *kaddr; + int left; + + kaddr = kmap_atomic(page, KM_USER0); + left = __copy_from_user_inatomic(kaddr + offset, buf, bytes); + kunmap_atomic(kaddr, KM_USER0); + + if (left != 0) { + /* Do it the slow way */ + kaddr = kmap(page); + left = __copy_from_user(kaddr + offset, buf, bytes); + kunmap(page); + } + return bytes - left; +} + +/* + * This has the same sideeffects and return value as filemap_copy_from_user(). + * The difference is that on a fault we need to memset the remainder of the + * page (out to offset+bytes), to emulate filemap_copy_from_user()'s + * single-segment behaviour. + */ +static inline size_t +filemap_copy_from_user_iovec(struct page *page, unsigned long offset, + const struct iovec *iov, size_t base, size_t bytes) +{ + char *kaddr; + size_t copied; + + kaddr = kmap_atomic(page, KM_USER0); + copied = __filemap_copy_from_user_iovec(kaddr + offset, iov, + base, bytes); + kunmap_atomic(kaddr, KM_USER0); + if (copied != bytes) { + kaddr = kmap(page); + copied = __filemap_copy_from_user_iovec(kaddr + offset, iov, + base, bytes); + kunmap(page); + } + return copied; +} + +static inline void +filemap_set_next_iovec(const struct iovec **iovp, size_t *basep, size_t bytes) +{ + const struct iovec *iov = *iovp; + size_t base = *basep; + + while (bytes) { + int copy = min(bytes, iov->iov_len - base); + + bytes -= copy; + base += copy; + if (iov->iov_len == base) { + iov++; + base = 0; + } + } + *iovp = iov; + *basep = base; +} +#endif diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c new file mode 100644 index 000000000000..7d63acd48817 --- /dev/null +++ b/mm/filemap_xip.c @@ -0,0 +1,581 @@ +/* + * linux/mm/filemap_xip.c + * + * Copyright (C) 2005 IBM Corporation + * Author: Carsten Otte + * + * derived from linux/mm/filemap.c - Copyright (C) Linus Torvalds + * + */ + +#include +#include +#include +#include +#include +#include +#include "filemap.h" + +/* + * This is a file read routine for execute in place files, and uses + * the mapping->a_ops->get_xip_page() function for the actual low-level + * stuff. + * + * Note the struct file* is not used at all. It may be NULL. + */ +static void +do_xip_mapping_read(struct address_space *mapping, + struct file_ra_state *_ra, + struct file *filp, + loff_t *ppos, + read_descriptor_t *desc, + read_actor_t actor) +{ + struct inode *inode = mapping->host; + unsigned long index, end_index, offset; + loff_t isize; + + BUG_ON(!mapping->a_ops->get_xip_page); + + index = *ppos >> PAGE_CACHE_SHIFT; + offset = *ppos & ~PAGE_CACHE_MASK; + + isize = i_size_read(inode); + if (!isize) + goto out; + + end_index = (isize - 1) >> PAGE_CACHE_SHIFT; + for (;;) { + struct page *page; + unsigned long nr, ret; + + /* nr is the maximum number of bytes to copy from this page */ + nr = PAGE_CACHE_SIZE; + if (index >= end_index) { + if (index > end_index) + goto out; + nr = ((isize - 1) & ~PAGE_CACHE_MASK) + 1; + if (nr <= offset) { + goto out; + } + } + nr = nr - offset; + + page = mapping->a_ops->get_xip_page(mapping, + index*(PAGE_SIZE/512), 0); + if (!page) + goto no_xip_page; + if (unlikely(IS_ERR(page))) { + if (PTR_ERR(page) == -ENODATA) { + /* sparse */ + page = virt_to_page(empty_zero_page); + } else { + desc->error = PTR_ERR(page); + goto out; + } + } else + BUG_ON(!PageUptodate(page)); + + /* If users can be writing to this page using arbitrary + * virtual addresses, take care about potential aliasing + * before reading the page on the kernel side. + */ + if (mapping_writably_mapped(mapping)) + flush_dcache_page(page); + + /* + * Ok, we have the page, and it's up-to-date, so + * now we can copy it to user space... + * + * The actor routine returns how many bytes were actually used.. + * NOTE! This may not be the same as how much of a user buffer + * we filled up (we may be padding etc), so we can only update + * "pos" here (the actor routine has to update the user buffer + * pointers and the remaining count). + */ + ret = actor(desc, page, offset, nr); + offset += ret; + index += offset >> PAGE_CACHE_SHIFT; + offset &= ~PAGE_CACHE_MASK; + + if (ret == nr && desc->count) + continue; + goto out; + +no_xip_page: + /* Did not get the page. Report it */ + desc->error = -EIO; + goto out; + } + +out: + *ppos = ((loff_t) index << PAGE_CACHE_SHIFT) + offset; + if (filp) + file_accessed(filp); +} + +/* + * This is the "read()" routine for all filesystems + * that uses the get_xip_page address space operation. + */ +static ssize_t +__xip_file_aio_read(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos) +{ + struct file *filp = iocb->ki_filp; + ssize_t retval; + unsigned long seg; + size_t count; + + count = 0; + for (seg = 0; seg < nr_segs; seg++) { + const struct iovec *iv = &iov[seg]; + + /* + * If any segment has a negative length, or the cumulative + * length ever wraps negative then return -EINVAL. + */ + count += iv->iov_len; + if (unlikely((ssize_t)(count|iv->iov_len) < 0)) + return -EINVAL; + if (access_ok(VERIFY_WRITE, iv->iov_base, iv->iov_len)) + continue; + if (seg == 0) + return -EFAULT; + nr_segs = seg; + count -= iv->iov_len; /* This segment is no good */ + break; + } + + retval = 0; + if (count) { + for (seg = 0; seg < nr_segs; seg++) { + read_descriptor_t desc; + + desc.written = 0; + desc.arg.buf = iov[seg].iov_base; + desc.count = iov[seg].iov_len; + if (desc.count == 0) + continue; + desc.error = 0; + do_xip_mapping_read(filp->f_mapping, &filp->f_ra, filp, + ppos, &desc, file_read_actor); + retval += desc.written; + if (!retval) { + retval = desc.error; + break; + } + } + } + return retval; +} + +ssize_t +xip_file_aio_read(struct kiocb *iocb, char __user *buf, size_t count, + loff_t pos) +{ + struct iovec local_iov = { .iov_base = buf, .iov_len = count }; + + BUG_ON(iocb->ki_pos != pos); + return __xip_file_aio_read(iocb, &local_iov, 1, &iocb->ki_pos); +} +EXPORT_SYMBOL_GPL(xip_file_aio_read); + +ssize_t +xip_file_readv(struct file *filp, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos) +{ + struct kiocb kiocb; + + init_sync_kiocb(&kiocb, filp); + return __xip_file_aio_read(&kiocb, iov, nr_segs, ppos); +} +EXPORT_SYMBOL_GPL(xip_file_readv); + +ssize_t +xip_file_sendfile(struct file *in_file, loff_t *ppos, + size_t count, read_actor_t actor, void *target) +{ + read_descriptor_t desc; + + if (!count) + return 0; + + desc.written = 0; + desc.count = count; + desc.arg.data = target; + desc.error = 0; + + do_xip_mapping_read(in_file->f_mapping, &in_file->f_ra, in_file, + ppos, &desc, actor); + if (desc.written) + return desc.written; + return desc.error; +} +EXPORT_SYMBOL_GPL(xip_file_sendfile); + +/* + * __xip_unmap is invoked from xip_unmap and + * xip_write + * + * This function walks all vmas of the address_space and unmaps the + * empty_zero_page when found at pgoff. Should it go in rmap.c? + */ +static void +__xip_unmap (struct address_space * mapping, + unsigned long pgoff) +{ + struct vm_area_struct *vma; + struct mm_struct *mm; + struct prio_tree_iter iter; + unsigned long address; + pte_t *pte; + pte_t pteval; + + spin_lock(&mapping->i_mmap_lock); + vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff) { + mm = vma->vm_mm; + address = vma->vm_start + + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT); + BUG_ON(address < vma->vm_start || address >= vma->vm_end); + /* + * We need the page_table_lock to protect us from page faults, + * munmap, fork, etc... + */ + pte = page_check_address(virt_to_page(empty_zero_page), mm, + address); + if (!IS_ERR(pte)) { + /* Nuke the page table entry. */ + flush_cache_page(vma, address, pte_pfn(pte)); + pteval = ptep_clear_flush(vma, address, pte); + BUG_ON(pte_dirty(pteval)); + pte_unmap(pte); + spin_unlock(&mm->page_table_lock); + } + } + spin_unlock(&mapping->i_mmap_lock); +} + +/* + * xip_nopage() is invoked via the vma operations vector for a + * mapped memory region to read in file data during a page fault. + * + * This function is derived from filemap_nopage, but used for execute in place + */ +static struct page * +xip_file_nopage(struct vm_area_struct * area, + unsigned long address, + int *type) +{ + struct file *file = area->vm_file; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + struct page *page; + unsigned long size, pgoff, endoff; + + pgoff = ((address - area->vm_start) >> PAGE_CACHE_SHIFT) + + area->vm_pgoff; + endoff = ((area->vm_end - area->vm_start) >> PAGE_CACHE_SHIFT) + + area->vm_pgoff; + + size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + if (pgoff >= size) { + return NULL; + } + + page = mapping->a_ops->get_xip_page(mapping, pgoff*(PAGE_SIZE/512), 0); + if (!IS_ERR(page)) { + BUG_ON(!PageUptodate(page)); + return page; + } + if (PTR_ERR(page) != -ENODATA) + return NULL; + + /* sparse block */ + if ((area->vm_flags & (VM_WRITE | VM_MAYWRITE)) && + (area->vm_flags & (VM_SHARED| VM_MAYSHARE)) && + (!(mapping->host->i_sb->s_flags & MS_RDONLY))) { + /* maybe shared writable, allocate new block */ + page = mapping->a_ops->get_xip_page (mapping, + pgoff*(PAGE_SIZE/512), 1); + if (IS_ERR(page)) + return NULL; + BUG_ON(!PageUptodate(page)); + /* unmap page at pgoff from all other vmas */ + __xip_unmap(mapping, pgoff); + } else { + /* not shared and writable, use empty_zero_page */ + page = virt_to_page(empty_zero_page); + } + + return page; +} + +static struct vm_operations_struct xip_file_vm_ops = { + .nopage = xip_file_nopage, +}; + +int xip_file_mmap(struct file * file, struct vm_area_struct * vma) +{ + BUG_ON(!file->f_mapping->a_ops->get_xip_page); + + file_accessed(file); + vma->vm_ops = &xip_file_vm_ops; + return 0; +} +EXPORT_SYMBOL_GPL(xip_file_mmap); + +static ssize_t +do_xip_file_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos, loff_t *ppos, + size_t count) +{ + struct file *file = iocb->ki_filp; + struct address_space * mapping = file->f_mapping; + struct address_space_operations *a_ops = mapping->a_ops; + struct inode *inode = mapping->host; + long status = 0; + struct page *page; + size_t bytes; + const struct iovec *cur_iov = iov; /* current iovec */ + size_t iov_base = 0; /* offset in the current iovec */ + char __user *buf; + ssize_t written = 0; + + BUG_ON(!mapping->a_ops->get_xip_page); + + buf = iov->iov_base; + do { + unsigned long index; + unsigned long offset; + size_t copied; + + offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */ + index = pos >> PAGE_CACHE_SHIFT; + bytes = PAGE_CACHE_SIZE - offset; + if (bytes > count) + bytes = count; + + /* + * Bring in the user page that we will copy from _first_. + * Otherwise there's a nasty deadlock on copying from the + * same page as we're writing to, without it being marked + * up-to-date. + */ + fault_in_pages_readable(buf, bytes); + + page = a_ops->get_xip_page(mapping, + index*(PAGE_SIZE/512), 0); + if (IS_ERR(page) && (PTR_ERR(page) == -ENODATA)) { + /* we allocate a new page unmap it */ + page = a_ops->get_xip_page(mapping, + index*(PAGE_SIZE/512), 1); + if (!IS_ERR(page)) + /* unmap page at pgoff from all other vmas */ + __xip_unmap(mapping, index); + + } + + if (IS_ERR(page)) { + status = PTR_ERR(page); + break; + } + + BUG_ON(!PageUptodate(page)); + + if (likely(nr_segs == 1)) + copied = filemap_copy_from_user(page, offset, + buf, bytes); + else + copied = filemap_copy_from_user_iovec(page, offset, + cur_iov, iov_base, bytes); + flush_dcache_page(page); + if (likely(copied > 0)) { + status = copied; + + if (status >= 0) { + written += status; + count -= status; + pos += status; + buf += status; + if (unlikely(nr_segs > 1)) + filemap_set_next_iovec(&cur_iov, + &iov_base, status); + } + } + if (unlikely(copied != bytes)) + if (status >= 0) + status = -EFAULT; + if (status < 0) + break; + } while (count); + *ppos = pos; + /* + * No need to use i_size_read() here, the i_size + * cannot change under us because we hold i_sem. + */ + if (pos > inode->i_size) { + i_size_write(inode, pos); + mark_inode_dirty(inode); + } + + return written ? written : status; +} + +static ssize_t +xip_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos) +{ + struct file *file = iocb->ki_filp; + struct address_space * mapping = file->f_mapping; + size_t ocount; /* original count */ + size_t count; /* after file limit checks */ + struct inode *inode = mapping->host; + unsigned long seg; + loff_t pos; + ssize_t written; + ssize_t err; + + ocount = 0; + for (seg = 0; seg < nr_segs; seg++) { + const struct iovec *iv = &iov[seg]; + + /* + * If any segment has a negative length, or the cumulative + * length ever wraps negative then return -EINVAL. + */ + ocount += iv->iov_len; + if (unlikely((ssize_t)(ocount|iv->iov_len) < 0)) + return -EINVAL; + if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len)) + continue; + if (seg == 0) + return -EFAULT; + nr_segs = seg; + ocount -= iv->iov_len; /* This segment is no good */ + break; + } + + count = ocount; + pos = *ppos; + + vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE); + + written = 0; + + err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); + if (err) + goto out; + + if (count == 0) + goto out; + + err = remove_suid(file->f_dentry); + if (err) + goto out; + + inode_update_time(inode, 1); + + /* use execute in place to copy directly to disk */ + written = do_xip_file_write (iocb, iov, + nr_segs, pos, ppos, count); + out: + return written ? written : err; +} + +static ssize_t +__xip_file_write_nolock(struct file *file, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos) +{ + struct kiocb kiocb; + + init_sync_kiocb(&kiocb, file); + return xip_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos); +} + +ssize_t +xip_file_aio_write(struct kiocb *iocb, const char __user *buf, + size_t count, loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + ssize_t ret; + struct iovec local_iov = { .iov_base = (void __user *)buf, + .iov_len = count }; + + BUG_ON(iocb->ki_pos != pos); + + down(&inode->i_sem); + ret = xip_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos); + up(&inode->i_sem); + return ret; +} +EXPORT_SYMBOL_GPL(xip_file_aio_write); + +ssize_t xip_file_writev(struct file *file, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos) +{ + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + ssize_t ret; + + down(&inode->i_sem); + ret = __xip_file_write_nolock(file, iov, nr_segs, ppos); + up(&inode->i_sem); + return ret; +} +EXPORT_SYMBOL_GPL(xip_file_writev); + +/* + * truncate a page used for execute in place + * functionality is analog to block_truncate_page but does use get_xip_page + * to get the page instead of page cache + */ +int +xip_truncate_page(struct address_space *mapping, loff_t from) +{ + pgoff_t index = from >> PAGE_CACHE_SHIFT; + unsigned offset = from & (PAGE_CACHE_SIZE-1); + unsigned blocksize; + unsigned length; + struct page *page; + void *kaddr; + int err; + + BUG_ON(!mapping->a_ops->get_xip_page); + + blocksize = 1 << mapping->host->i_blkbits; + length = offset & (blocksize - 1); + + /* Block boundary? Nothing to do */ + if (!length) + return 0; + + length = blocksize - length; + + page = mapping->a_ops->get_xip_page(mapping, + index*(PAGE_SIZE/512), 0); + err = -ENOMEM; + if (!page) + goto out; + if (unlikely(IS_ERR(page))) { + if (PTR_ERR(page) == -ENODATA) { + /* Hole? No need to truncate */ + return 0; + } else { + err = PTR_ERR(page); + goto out; + } + } else + BUG_ON(!PageUptodate(page)); + kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + offset, 0, length); + kunmap_atomic(kaddr, KM_USER0); + + flush_dcache_page(page); + err = 0; +out: + return err; +} +EXPORT_SYMBOL_GPL(xip_truncate_page); diff --git a/mm/rmap.c b/mm/rmap.c index 89770bd25f31..08ac5c7fa91f 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -247,8 +247,8 @@ unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma) * * On success returns with mapped pte and locked mm->page_table_lock. */ -static pte_t *page_check_address(struct page *page, struct mm_struct *mm, - unsigned long address) +pte_t *page_check_address(struct page *page, struct mm_struct *mm, + unsigned long address) { pgd_t *pgd; pud_t *pud; -- cgit v1.2.3 From 6d79125bba55ee82701f1c7d4ebbc1aa20ecbe4e Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Thu, 23 Jun 2005 22:05:26 -0700 Subject: [PATCH] xip: ext2: execute in place These are the ext2 related parts. Ext2 now uses the xip_* file operations along with the get_xip_page aop when mounted with -o xip. Signed-off-by: Carsten Otte Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/Kconfig | 17 +++++++++++ fs/ext2/Makefile | 1 + fs/ext2/ext2.h | 2 ++ fs/ext2/file.c | 18 +++++++++++ fs/ext2/inode.c | 31 ++++++++++++++++--- fs/ext2/namei.c | 12 ++++++-- fs/ext2/super.c | 27 ++++++++++++++++- fs/ext2/xip.c | 80 +++++++++++++++++++++++++++++++++++++++++++++++++ fs/ext2/xip.h | 25 ++++++++++++++++ include/linux/ext2_fs.h | 25 ++++++++-------- include/linux/fs.h | 5 ++++ 11 files changed, 223 insertions(+), 20 deletions(-) create mode 100644 fs/ext2/xip.c create mode 100644 fs/ext2/xip.h (limited to 'include/linux') diff --git a/fs/Kconfig b/fs/Kconfig index 5c704d05627a..8157f2e2d515 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -50,6 +50,23 @@ config EXT2_FS_SECURITY If you are not using a security module that requires using extended attributes for file security labels, say N. +config EXT2_FS_XIP + bool "Ext2 execute in place support" + depends on EXT2_FS + help + Execute in place can be used on memory-backed block devices. If you + enable this option, you can select to mount block devices which are + capable of this feature without using the page cache. + + If you do not use a block device that is capable of using this, + or if unsure, say N. + +config FS_XIP +# execute in place + bool + depends on EXT2_FS_XIP + default y + config EXT3_FS tristate "Ext3 journalling file system support" help diff --git a/fs/ext2/Makefile b/fs/ext2/Makefile index ee240a14e70f..c5d02da73bc3 100644 --- a/fs/ext2/Makefile +++ b/fs/ext2/Makefile @@ -10,3 +10,4 @@ ext2-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ ext2-$(CONFIG_EXT2_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o ext2-$(CONFIG_EXT2_FS_POSIX_ACL) += acl.o ext2-$(CONFIG_EXT2_FS_SECURITY) += xattr_security.o +ext2-$(CONFIG_EXT2_FS_XIP) += xip.o diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index 8f0fd726c3f1..eed521d22cf0 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -147,9 +147,11 @@ extern struct file_operations ext2_dir_operations; /* file.c */ extern struct inode_operations ext2_file_inode_operations; extern struct file_operations ext2_file_operations; +extern struct file_operations ext2_xip_file_operations; /* inode.c */ extern struct address_space_operations ext2_aops; +extern struct address_space_operations ext2_aops_xip; extern struct address_space_operations ext2_nobh_aops; /* namei.c */ diff --git a/fs/ext2/file.c b/fs/ext2/file.c index f5e86141ec54..2b3d572365af 100644 --- a/fs/ext2/file.c +++ b/fs/ext2/file.c @@ -55,6 +55,24 @@ struct file_operations ext2_file_operations = { .sendfile = generic_file_sendfile, }; +#ifdef CONFIG_EXT2_FS_XIP +struct file_operations ext2_xip_file_operations = { + .llseek = generic_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = xip_file_aio_read, + .aio_write = xip_file_aio_write, + .ioctl = ext2_ioctl, + .mmap = xip_file_mmap, + .open = generic_file_open, + .release = ext2_release_file, + .fsync = ext2_sync_file, + .readv = xip_file_readv, + .writev = xip_file_writev, + .sendfile = xip_file_sendfile, +}; +#endif + struct inode_operations ext2_file_inode_operations = { .truncate = ext2_truncate, #ifdef CONFIG_EXT2_FS_XATTR diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index a50d9db4b6e4..53dceb0c6593 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -33,6 +33,7 @@ #include #include "ext2.h" #include "acl.h" +#include "xip.h" MODULE_AUTHOR("Remy Card and others"); MODULE_DESCRIPTION("Second Extended Filesystem"); @@ -594,6 +595,16 @@ out: if (err) goto cleanup; + if (ext2_use_xip(inode->i_sb)) { + /* + * we need to clear the block + */ + err = ext2_clear_xip_target (inode, + le32_to_cpu(chain[depth-1].key)); + if (err) + goto cleanup; + } + if (ext2_splice_branch(inode, iblock, chain, partial, left) < 0) goto changed; @@ -691,6 +702,11 @@ struct address_space_operations ext2_aops = { .writepages = ext2_writepages, }; +struct address_space_operations ext2_aops_xip = { + .bmap = ext2_bmap, + .get_xip_page = ext2_get_xip_page, +}; + struct address_space_operations ext2_nobh_aops = { .readpage = ext2_readpage, .readpages = ext2_readpages, @@ -910,7 +926,9 @@ void ext2_truncate (struct inode * inode) iblock = (inode->i_size + blocksize-1) >> EXT2_BLOCK_SIZE_BITS(inode->i_sb); - if (test_opt(inode->i_sb, NOBH)) + if (mapping_is_xip(inode->i_mapping)) + xip_truncate_page(inode->i_mapping, inode->i_size); + else if (test_opt(inode->i_sb, NOBH)) nobh_truncate_page(inode->i_mapping, inode->i_size); else block_truncate_page(inode->i_mapping, @@ -1110,11 +1128,16 @@ void ext2_read_inode (struct inode * inode) if (S_ISREG(inode->i_mode)) { inode->i_op = &ext2_file_inode_operations; - inode->i_fop = &ext2_file_operations; - if (test_opt(inode->i_sb, NOBH)) + if (ext2_use_xip(inode->i_sb)) { + inode->i_mapping->a_ops = &ext2_aops_xip; + inode->i_fop = &ext2_xip_file_operations; + } else if (test_opt(inode->i_sb, NOBH)) { inode->i_mapping->a_ops = &ext2_nobh_aops; - else + inode->i_fop = &ext2_file_operations; + } else { inode->i_mapping->a_ops = &ext2_aops; + inode->i_fop = &ext2_file_operations; + } } else if (S_ISDIR(inode->i_mode)) { inode->i_op = &ext2_dir_inode_operations; inode->i_fop = &ext2_dir_operations; diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c index 3176b3d3ffa8..c5513953c825 100644 --- a/fs/ext2/namei.c +++ b/fs/ext2/namei.c @@ -34,6 +34,7 @@ #include "ext2.h" #include "xattr.h" #include "acl.h" +#include "xip.h" /* * Couple of helper functions - make the code slightly cleaner. @@ -127,11 +128,16 @@ static int ext2_create (struct inode * dir, struct dentry * dentry, int mode, st int err = PTR_ERR(inode); if (!IS_ERR(inode)) { inode->i_op = &ext2_file_inode_operations; - inode->i_fop = &ext2_file_operations; - if (test_opt(inode->i_sb, NOBH)) + if (ext2_use_xip(inode->i_sb)) { + inode->i_mapping->a_ops = &ext2_aops_xip; + inode->i_fop = &ext2_xip_file_operations; + } else if (test_opt(inode->i_sb, NOBH)) { inode->i_mapping->a_ops = &ext2_nobh_aops; - else + inode->i_fop = &ext2_file_operations; + } else { inode->i_mapping->a_ops = &ext2_aops; + inode->i_fop = &ext2_file_operations; + } mark_inode_dirty(inode); err = ext2_add_nondir(dentry, inode); } diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 661c3d98d946..876e391f2871 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -31,6 +31,7 @@ #include "ext2.h" #include "xattr.h" #include "acl.h" +#include "xip.h" static void ext2_sync_super(struct super_block *sb, struct ext2_super_block *es); @@ -257,7 +258,7 @@ enum { Opt_bsd_df, Opt_minix_df, Opt_grpid, Opt_nogrpid, Opt_resgid, Opt_resuid, Opt_sb, Opt_err_cont, Opt_err_panic, Opt_err_ro, Opt_nouid32, Opt_check, Opt_nocheck, Opt_debug, Opt_oldalloc, Opt_orlov, Opt_nobh, - Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, + Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, Opt_xip, Opt_ignore, Opt_err, }; @@ -286,6 +287,7 @@ static match_table_t tokens = { {Opt_nouser_xattr, "nouser_xattr"}, {Opt_acl, "acl"}, {Opt_noacl, "noacl"}, + {Opt_xip, "xip"}, {Opt_ignore, "grpquota"}, {Opt_ignore, "noquota"}, {Opt_ignore, "quota"}, @@ -397,6 +399,13 @@ static int parse_options (char * options, printk("EXT2 (no)acl options not supported\n"); break; #endif + case Opt_xip: +#ifdef CONFIG_EXT2_FS_XIP + set_opt (sbi->s_mount_opt, XIP); +#else + printk("EXT2 xip option not supported\n"); +#endif + break; case Opt_ignore: break; default: @@ -640,6 +649,9 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) ((EXT2_SB(sb)->s_mount_opt & EXT2_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); + ext2_xip_verify_sb(sb); /* see if bdev supports xip, unset + EXT2_MOUNT_XIP if not */ + if (le32_to_cpu(es->s_rev_level) == EXT2_GOOD_OLD_REV && (EXT2_HAS_COMPAT_FEATURE(sb, ~0U) || EXT2_HAS_RO_COMPAT_FEATURE(sb, ~0U) || @@ -668,6 +680,13 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) blocksize = BLOCK_SIZE << le32_to_cpu(sbi->s_es->s_log_block_size); + if ((ext2_use_xip(sb)) && ((blocksize != PAGE_SIZE) || + (sb->s_blocksize != blocksize))) { + if (!silent) + printk("XIP: Unsupported blocksize\n"); + goto failed_mount; + } + /* If the blocksize doesn't match, re-read the thing.. */ if (sb->s_blocksize != blocksize) { brelse(bh); @@ -916,6 +935,7 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data) { struct ext2_sb_info * sbi = EXT2_SB(sb); struct ext2_super_block * es; + unsigned long old_mount_opt = sbi->s_mount_opt; /* * Allow the "check" option to be passed as a remount option. @@ -927,6 +947,11 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data) ((sbi->s_mount_opt & EXT2_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); es = sbi->s_es; + if (((sbi->s_mount_opt & EXT2_MOUNT_XIP) != + (old_mount_opt & EXT2_MOUNT_XIP)) && + invalidate_inodes(sb)) + ext2_warning(sb, __FUNCTION__, "busy inodes while remounting "\ + "xip remain in cache (no functional problem)"); if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY)) return 0; if (*flags & MS_RDONLY) { diff --git a/fs/ext2/xip.c b/fs/ext2/xip.c new file mode 100644 index 000000000000..d44431d1a338 --- /dev/null +++ b/fs/ext2/xip.c @@ -0,0 +1,80 @@ +/* + * linux/fs/ext2/xip.c + * + * Copyright (C) 2005 IBM Corporation + * Author: Carsten Otte (cotte@de.ibm.com) + */ + +#include +#include +#include +#include +#include +#include +#include "ext2.h" +#include "xip.h" + +static inline int +__inode_direct_access(struct inode *inode, sector_t sector, unsigned long *data) { + BUG_ON(!inode->i_sb->s_bdev->bd_disk->fops->direct_access); + return inode->i_sb->s_bdev->bd_disk->fops + ->direct_access(inode->i_sb->s_bdev,sector,data); +} + +int +ext2_clear_xip_target(struct inode *inode, int block) { + sector_t sector = block*(PAGE_SIZE/512); + unsigned long data; + int rc; + + rc = __inode_direct_access(inode, sector, &data); + if (rc) + return rc; + clear_page((void*)data); + return 0; +} + +void ext2_xip_verify_sb(struct super_block *sb) +{ + struct ext2_sb_info *sbi = EXT2_SB(sb); + + if ((sbi->s_mount_opt & EXT2_MOUNT_XIP)) { + if ((sb->s_bdev == NULL) || + sb->s_bdev->bd_disk == NULL || + sb->s_bdev->bd_disk->fops == NULL || + sb->s_bdev->bd_disk->fops->direct_access == NULL) { + sbi->s_mount_opt &= (~EXT2_MOUNT_XIP); + ext2_warning(sb, __FUNCTION__, + "ignoring xip option - not supported by bdev"); + } + } +} + +struct page* +ext2_get_xip_page(struct address_space *mapping, sector_t blockno, + int create) +{ + int rc; + unsigned long data; + struct buffer_head tmp; + + tmp.b_state = 0; + tmp.b_blocknr = 0; + rc = ext2_get_block(mapping->host, blockno/(PAGE_SIZE/512) , &tmp, + create); + if (rc) + return ERR_PTR(rc); + if (tmp.b_blocknr == 0) { + /* SPARSE block */ + BUG_ON(create); + return ERR_PTR(-ENODATA); + } + + rc = __inode_direct_access + (mapping->host,tmp.b_blocknr*(PAGE_SIZE/512) ,&data); + if (rc) + return ERR_PTR(rc); + + SetPageUptodate(virt_to_page(data)); + return virt_to_page(data); +} diff --git a/fs/ext2/xip.h b/fs/ext2/xip.h new file mode 100644 index 000000000000..aa85331d6c56 --- /dev/null +++ b/fs/ext2/xip.h @@ -0,0 +1,25 @@ +/* + * linux/fs/ext2/xip.h + * + * Copyright (C) 2005 IBM Corporation + * Author: Carsten Otte (cotte@de.ibm.com) + */ + +#ifdef CONFIG_EXT2_FS_XIP +extern void ext2_xip_verify_sb (struct super_block *); +extern int ext2_clear_xip_target (struct inode *, int); + +static inline int ext2_use_xip (struct super_block *sb) +{ + struct ext2_sb_info *sbi = EXT2_SB(sb); + return (sbi->s_mount_opt & EXT2_MOUNT_XIP); +} +struct page* ext2_get_xip_page (struct address_space *, sector_t, int); +#define mapping_is_xip(map) unlikely(map->a_ops->get_xip_page) +#else +#define mapping_is_xip(map) 0 +#define ext2_xip_verify_sb(sb) do { } while (0) +#define ext2_use_xip(sb) 0 +#define ext2_clear_xip_target(inode, chain) 0 +#define ext2_get_xip_page NULL +#endif diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h index fab43527e597..a657130ba03a 100644 --- a/include/linux/ext2_fs.h +++ b/include/linux/ext2_fs.h @@ -300,18 +300,19 @@ struct ext2_inode { /* * Mount flags */ -#define EXT2_MOUNT_CHECK 0x0001 /* Do mount-time checks */ -#define EXT2_MOUNT_OLDALLOC 0x0002 /* Don't use the new Orlov allocator */ -#define EXT2_MOUNT_GRPID 0x0004 /* Create files with directory's group */ -#define EXT2_MOUNT_DEBUG 0x0008 /* Some debugging messages */ -#define EXT2_MOUNT_ERRORS_CONT 0x0010 /* Continue on errors */ -#define EXT2_MOUNT_ERRORS_RO 0x0020 /* Remount fs ro on errors */ -#define EXT2_MOUNT_ERRORS_PANIC 0x0040 /* Panic on errors */ -#define EXT2_MOUNT_MINIX_DF 0x0080 /* Mimics the Minix statfs */ -#define EXT2_MOUNT_NOBH 0x0100 /* No buffer_heads */ -#define EXT2_MOUNT_NO_UID32 0x0200 /* Disable 32-bit UIDs */ -#define EXT2_MOUNT_XATTR_USER 0x4000 /* Extended user attributes */ -#define EXT2_MOUNT_POSIX_ACL 0x8000 /* POSIX Access Control Lists */ +#define EXT2_MOUNT_CHECK 0x000001 /* Do mount-time checks */ +#define EXT2_MOUNT_OLDALLOC 0x000002 /* Don't use the new Orlov allocator */ +#define EXT2_MOUNT_GRPID 0x000004 /* Create files with directory's group */ +#define EXT2_MOUNT_DEBUG 0x000008 /* Some debugging messages */ +#define EXT2_MOUNT_ERRORS_CONT 0x000010 /* Continue on errors */ +#define EXT2_MOUNT_ERRORS_RO 0x000020 /* Remount fs ro on errors */ +#define EXT2_MOUNT_ERRORS_PANIC 0x000040 /* Panic on errors */ +#define EXT2_MOUNT_MINIX_DF 0x000080 /* Mimics the Minix statfs */ +#define EXT2_MOUNT_NOBH 0x000100 /* No buffer_heads */ +#define EXT2_MOUNT_NO_UID32 0x000200 /* Disable 32-bit UIDs */ +#define EXT2_MOUNT_XATTR_USER 0x004000 /* Extended user attributes */ +#define EXT2_MOUNT_POSIX_ACL 0x008000 /* POSIX Access Control Lists */ +#define EXT2_MOUNT_XIP 0x010000 /* Execute in place */ #define clear_opt(o, opt) o &= ~EXT2_MOUNT_##opt #define set_opt(o, opt) o |= EXT2_MOUNT_##opt diff --git a/include/linux/fs.h b/include/linux/fs.h index 79c0fafc0211..7e0501895f35 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1513,6 +1513,11 @@ extern ssize_t xip_file_aio_write(struct kiocb *iocb, const char __user *buf, extern ssize_t xip_file_writev(struct file *file, const struct iovec *iov, unsigned long nr_segs, loff_t *ppos); extern int xip_truncate_page(struct address_space *mapping, loff_t from); +#else +static inline int xip_truncate_page(struct address_space *mapping, loff_t from) +{ + return 0; +} #endif static inline void do_generic_file_read(struct file * filp, loff_t *ppos, -- cgit v1.2.3 From eb6fe0c388e43b02e261f0fdee60e42f6298d7f7 Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Thu, 23 Jun 2005 22:05:28 -0700 Subject: [PATCH] xip: reduce code duplication This patch reworks filemap_xip.c with the goal to reduce code duplication from mm/filemap.c. It applies agains 2.6.12-rc6-mm1. Instead of implementing the aio functions, this one implements the synchronous read/write functions only. For readv and writev, the generic fallback is used. For aio, we rely on the application doing the fallback. Since our "synchronous" function does memcpy immediately anyway, there is no performance difference between using the fallbacks or implementing each operation. Signed-off-by: Carsten Otte Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/file.c | 8 +- include/linux/fs.h | 12 +-- mm/filemap.h | 2 +- mm/filemap_xip.c | 246 ++++++++++++----------------------------------------- 4 files changed, 63 insertions(+), 205 deletions(-) (limited to 'include/linux') diff --git a/fs/ext2/file.c b/fs/ext2/file.c index 2b3d572365af..a484412fc782 100644 --- a/fs/ext2/file.c +++ b/fs/ext2/file.c @@ -58,17 +58,13 @@ struct file_operations ext2_file_operations = { #ifdef CONFIG_EXT2_FS_XIP struct file_operations ext2_xip_file_operations = { .llseek = generic_file_llseek, - .read = do_sync_read, - .write = do_sync_write, - .aio_read = xip_file_aio_read, - .aio_write = xip_file_aio_write, + .read = xip_file_read, + .write = xip_file_write, .ioctl = ext2_ioctl, .mmap = xip_file_mmap, .open = generic_file_open, .release = ext2_release_file, .fsync = ext2_sync_file, - .readv = xip_file_readv, - .writev = xip_file_writev, .sendfile = xip_file_sendfile, }; #endif diff --git a/include/linux/fs.h b/include/linux/fs.h index 7e0501895f35..3ae8e37bdfc8 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1500,18 +1500,14 @@ extern int generic_file_open(struct inode * inode, struct file * filp); extern int nonseekable_open(struct inode * inode, struct file * filp); #ifdef CONFIG_FS_XIP -extern ssize_t xip_file_aio_read(struct kiocb *iocb, char __user *buf, - size_t count, loff_t pos); -extern ssize_t xip_file_readv(struct file *filp, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos); +extern ssize_t xip_file_read(struct file *filp, char __user *buf, size_t len, + loff_t *ppos); extern ssize_t xip_file_sendfile(struct file *in_file, loff_t *ppos, size_t count, read_actor_t actor, void *target); extern int xip_file_mmap(struct file * file, struct vm_area_struct * vma); -extern ssize_t xip_file_aio_write(struct kiocb *iocb, const char __user *buf, - size_t count, loff_t pos); -extern ssize_t xip_file_writev(struct file *file, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos); +extern ssize_t xip_file_write(struct file *filp, const char __user *buf, + size_t len, loff_t *ppos); extern int xip_truncate_page(struct address_space *mapping, loff_t from); #else static inline int xip_truncate_page(struct address_space *mapping, loff_t from) diff --git a/mm/filemap.h b/mm/filemap.h index c2d0546a57eb..13793ba0ce17 100644 --- a/mm/filemap.h +++ b/mm/filemap.h @@ -15,7 +15,7 @@ #include #include -extern size_t +size_t __filemap_copy_from_user_iovec(char *vaddr, const struct iovec *iov, size_t base, diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c index 7d63acd48817..3b6e384b98a6 100644 --- a/mm/filemap_xip.c +++ b/mm/filemap_xip.c @@ -114,83 +114,28 @@ out: file_accessed(filp); } -/* - * This is the "read()" routine for all filesystems - * that uses the get_xip_page address space operation. - */ -static ssize_t -__xip_file_aio_read(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos) -{ - struct file *filp = iocb->ki_filp; - ssize_t retval; - unsigned long seg; - size_t count; - - count = 0; - for (seg = 0; seg < nr_segs; seg++) { - const struct iovec *iv = &iov[seg]; - - /* - * If any segment has a negative length, or the cumulative - * length ever wraps negative then return -EINVAL. - */ - count += iv->iov_len; - if (unlikely((ssize_t)(count|iv->iov_len) < 0)) - return -EINVAL; - if (access_ok(VERIFY_WRITE, iv->iov_base, iv->iov_len)) - continue; - if (seg == 0) - return -EFAULT; - nr_segs = seg; - count -= iv->iov_len; /* This segment is no good */ - break; - } - - retval = 0; - if (count) { - for (seg = 0; seg < nr_segs; seg++) { - read_descriptor_t desc; - - desc.written = 0; - desc.arg.buf = iov[seg].iov_base; - desc.count = iov[seg].iov_len; - if (desc.count == 0) - continue; - desc.error = 0; - do_xip_mapping_read(filp->f_mapping, &filp->f_ra, filp, - ppos, &desc, file_read_actor); - retval += desc.written; - if (!retval) { - retval = desc.error; - break; - } - } - } - return retval; -} - ssize_t -xip_file_aio_read(struct kiocb *iocb, char __user *buf, size_t count, - loff_t pos) +xip_file_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos) { - struct iovec local_iov = { .iov_base = buf, .iov_len = count }; + read_descriptor_t desc; - BUG_ON(iocb->ki_pos != pos); - return __xip_file_aio_read(iocb, &local_iov, 1, &iocb->ki_pos); -} -EXPORT_SYMBOL_GPL(xip_file_aio_read); + if (!access_ok(VERIFY_WRITE, buf, len)) + return -EFAULT; -ssize_t -xip_file_readv(struct file *filp, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos) -{ - struct kiocb kiocb; + desc.written = 0; + desc.arg.buf = buf; + desc.count = len; + desc.error = 0; - init_sync_kiocb(&kiocb, filp); - return __xip_file_aio_read(&kiocb, iov, nr_segs, ppos); + do_xip_mapping_read(filp->f_mapping, &filp->f_ra, filp, + ppos, &desc, file_read_actor); + + if (desc.written) + return desc.written; + else + return desc.error; } -EXPORT_SYMBOL_GPL(xip_file_readv); +EXPORT_SYMBOL_GPL(xip_file_read); ssize_t xip_file_sendfile(struct file *in_file, loff_t *ppos, @@ -326,25 +271,19 @@ int xip_file_mmap(struct file * file, struct vm_area_struct * vma) EXPORT_SYMBOL_GPL(xip_file_mmap); static ssize_t -do_xip_file_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos, loff_t *ppos, - size_t count) +__xip_file_write(struct file *filp, const char __user *buf, + size_t count, loff_t pos, loff_t *ppos) { - struct file *file = iocb->ki_filp; - struct address_space * mapping = file->f_mapping; + struct address_space * mapping = filp->f_mapping; struct address_space_operations *a_ops = mapping->a_ops; struct inode *inode = mapping->host; long status = 0; struct page *page; size_t bytes; - const struct iovec *cur_iov = iov; /* current iovec */ - size_t iov_base = 0; /* offset in the current iovec */ - char __user *buf; ssize_t written = 0; BUG_ON(!mapping->a_ops->get_xip_page); - buf = iov->iov_base; do { unsigned long index; unsigned long offset; @@ -365,15 +304,14 @@ do_xip_file_write(struct kiocb *iocb, const struct iovec *iov, fault_in_pages_readable(buf, bytes); page = a_ops->get_xip_page(mapping, - index*(PAGE_SIZE/512), 0); + index*(PAGE_SIZE/512), 0); if (IS_ERR(page) && (PTR_ERR(page) == -ENODATA)) { /* we allocate a new page unmap it */ page = a_ops->get_xip_page(mapping, - index*(PAGE_SIZE/512), 1); + index*(PAGE_SIZE/512), 1); if (!IS_ERR(page)) - /* unmap page at pgoff from all other vmas */ - __xip_unmap(mapping, index); - + /* unmap page at pgoff from all other vmas */ + __xip_unmap(mapping, index); } if (IS_ERR(page)) { @@ -383,12 +321,7 @@ do_xip_file_write(struct kiocb *iocb, const struct iovec *iov, BUG_ON(!PageUptodate(page)); - if (likely(nr_segs == 1)) - copied = filemap_copy_from_user(page, offset, - buf, bytes); - else - copied = filemap_copy_from_user_iovec(page, offset, - cur_iov, iov_base, bytes); + copied = filemap_copy_from_user(page, offset, buf, bytes); flush_dcache_page(page); if (likely(copied > 0)) { status = copied; @@ -398,9 +331,6 @@ do_xip_file_write(struct kiocb *iocb, const struct iovec *iov, count -= status; pos += status; buf += status; - if (unlikely(nr_segs > 1)) - filemap_set_next_iovec(&cur_iov, - &iov_base, status); } } if (unlikely(copied != bytes)) @@ -422,110 +352,52 @@ do_xip_file_write(struct kiocb *iocb, const struct iovec *iov, return written ? written : status; } -static ssize_t -xip_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos) +ssize_t +xip_file_write(struct file *filp, const char __user *buf, size_t len, + loff_t *ppos) { - struct file *file = iocb->ki_filp; - struct address_space * mapping = file->f_mapping; - size_t ocount; /* original count */ - size_t count; /* after file limit checks */ - struct inode *inode = mapping->host; - unsigned long seg; - loff_t pos; - ssize_t written; - ssize_t err; + struct address_space *mapping = filp->f_mapping; + struct inode *inode = mapping->host; + size_t count; + loff_t pos; + ssize_t ret; - ocount = 0; - for (seg = 0; seg < nr_segs; seg++) { - const struct iovec *iv = &iov[seg]; + down(&inode->i_sem); - /* - * If any segment has a negative length, or the cumulative - * length ever wraps negative then return -EINVAL. - */ - ocount += iv->iov_len; - if (unlikely((ssize_t)(ocount|iv->iov_len) < 0)) - return -EINVAL; - if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len)) - continue; - if (seg == 0) - return -EFAULT; - nr_segs = seg; - ocount -= iv->iov_len; /* This segment is no good */ - break; + if (!access_ok(VERIFY_READ, buf, len)) { + ret=-EFAULT; + goto out_up; } - count = ocount; pos = *ppos; + count = len; vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE); - written = 0; - - err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); - if (err) - goto out; + /* We can write back this queue in page reclaim */ + current->backing_dev_info = mapping->backing_dev_info; + ret = generic_write_checks(filp, &pos, &count, S_ISBLK(inode->i_mode)); + if (ret) + goto out_backing; if (count == 0) - goto out; + goto out_backing; - err = remove_suid(file->f_dentry); - if (err) - goto out; + ret = remove_suid(filp->f_dentry); + if (ret) + goto out_backing; inode_update_time(inode, 1); - /* use execute in place to copy directly to disk */ - written = do_xip_file_write (iocb, iov, - nr_segs, pos, ppos, count); - out: - return written ? written : err; -} - -static ssize_t -__xip_file_write_nolock(struct file *file, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos) -{ - struct kiocb kiocb; - - init_sync_kiocb(&kiocb, file); - return xip_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos); -} - -ssize_t -xip_file_aio_write(struct kiocb *iocb, const char __user *buf, - size_t count, loff_t pos) -{ - struct file *file = iocb->ki_filp; - struct address_space *mapping = file->f_mapping; - struct inode *inode = mapping->host; - ssize_t ret; - struct iovec local_iov = { .iov_base = (void __user *)buf, - .iov_len = count }; + ret = __xip_file_write (filp, buf, count, pos, ppos); - BUG_ON(iocb->ki_pos != pos); - - down(&inode->i_sem); - ret = xip_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos); + out_backing: + current->backing_dev_info = NULL; + out_up: up(&inode->i_sem); return ret; } -EXPORT_SYMBOL_GPL(xip_file_aio_write); - -ssize_t xip_file_writev(struct file *file, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos) -{ - struct address_space *mapping = file->f_mapping; - struct inode *inode = mapping->host; - ssize_t ret; - - down(&inode->i_sem); - ret = __xip_file_write_nolock(file, iov, nr_segs, ppos); - up(&inode->i_sem); - return ret; -} -EXPORT_SYMBOL_GPL(xip_file_writev); +EXPORT_SYMBOL_GPL(xip_file_write); /* * truncate a page used for execute in place @@ -541,7 +413,6 @@ xip_truncate_page(struct address_space *mapping, loff_t from) unsigned length; struct page *page; void *kaddr; - int err; BUG_ON(!mapping->a_ops->get_xip_page); @@ -556,17 +427,14 @@ xip_truncate_page(struct address_space *mapping, loff_t from) page = mapping->a_ops->get_xip_page(mapping, index*(PAGE_SIZE/512), 0); - err = -ENOMEM; if (!page) - goto out; + return -ENOMEM; if (unlikely(IS_ERR(page))) { - if (PTR_ERR(page) == -ENODATA) { + if (PTR_ERR(page) == -ENODATA) /* Hole? No need to truncate */ return 0; - } else { - err = PTR_ERR(page); - goto out; - } + else + return PTR_ERR(page); } else BUG_ON(!PageUptodate(page)); kaddr = kmap_atomic(page, KM_USER0); @@ -574,8 +442,6 @@ xip_truncate_page(struct address_space *mapping, loff_t from) kunmap_atomic(kaddr, KM_USER0); flush_dcache_page(page); - err = 0; -out: - return err; + return 0; } EXPORT_SYMBOL_GPL(xip_truncate_page); -- cgit v1.2.3 From 52c1da39534fb382c061de58b65f678ad74b59f5 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 23 Jun 2005 22:05:33 -0700 Subject: [PATCH] make various thing static Another rollup of patches which give various symbols static scope Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/media/common/saa7146_fops.c | 2 +- drivers/media/video/tvaudio.c | 22 +++++++++++----------- drivers/scsi/hosts.c | 2 +- drivers/scsi/scsi.c | 6 ++++-- drivers/scsi/scsi_debug.c | 2 +- drivers/scsi/scsi_lib.c | 2 +- drivers/scsi/scsi_priv.h | 4 ---- drivers/scsi/scsi_sysfs.c | 4 ++-- fs/namespace.c | 2 +- fs/reiserfs/stree.c | 2 +- include/linux/irq.h | 1 - include/linux/namespace.h | 1 - include/net/sctp/sm.h | 6 ------ kernel/irq/spurious.c | 2 +- kernel/module.c | 2 +- kernel/power/swsusp.c | 2 +- net/sctp/sm_statefuns.c | 16 ++++++++++++++-- 17 files changed, 40 insertions(+), 38 deletions(-) (limited to 'include/linux') diff --git a/drivers/media/common/saa7146_fops.c b/drivers/media/common/saa7146_fops.c index cb826c9adfe7..c04fd11526e0 100644 --- a/drivers/media/common/saa7146_fops.c +++ b/drivers/media/common/saa7146_fops.c @@ -403,7 +403,7 @@ static struct file_operations video_fops = .llseek = no_llseek, }; -void vv_callback(struct saa7146_dev *dev, unsigned long status) +static void vv_callback(struct saa7146_dev *dev, unsigned long status) { u32 isr = status; diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c index 5430b25b910d..9a493bea76d8 100644 --- a/drivers/media/video/tvaudio.c +++ b/drivers/media/video/tvaudio.c @@ -1236,17 +1236,17 @@ static int ta8874z_checkit(struct CHIPSTATE *chip) /* audio chip descriptions - struct CHIPDESC */ /* insmod options to enable/disable individual audio chips */ -int tda8425 = 1; -int tda9840 = 1; -int tda9850 = 1; -int tda9855 = 1; -int tda9873 = 1; -int tda9874a = 1; -int tea6300 = 0; // address clash with msp34xx -int tea6320 = 0; // address clash with msp34xx -int tea6420 = 1; -int pic16c54 = 1; -int ta8874z = 0; // address clash with tda9840 +static int tda8425 = 1; +static int tda9840 = 1; +static int tda9850 = 1; +static int tda9855 = 1; +static int tda9873 = 1; +static int tda9874a = 1; +static int tea6300 = 0; // address clash with msp34xx +static int tea6320 = 0; // address clash with msp34xx +static int tea6420 = 1; +static int pic16c54 = 1; +static int ta8874z = 0; // address clash with tda9840 module_param(tda8425, int, 0444); module_param(tda9840, int, 0444); diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index ba347576d99b..d7a38b6713f9 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -56,7 +56,7 @@ static struct class shost_class = { * @shost: pointer to struct Scsi_Host * recovery: recovery requested to run. **/ -void scsi_host_cancel(struct Scsi_Host *shost, int recovery) +static void scsi_host_cancel(struct Scsi_Host *shost, int recovery) { struct scsi_device *sdev; diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index 5578ae9a9e45..1cb5f7d4f278 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -68,6 +68,8 @@ #include "scsi_priv.h" #include "scsi_logging.h" +static void scsi_done(struct scsi_cmnd *cmd); +static int scsi_retry_command(struct scsi_cmnd *cmd); /* * Definitions and constants. @@ -741,7 +743,7 @@ static DEFINE_PER_CPU(struct list_head, scsi_done_q); * * This function is interrupt context safe. */ -void scsi_done(struct scsi_cmnd *cmd) +static void scsi_done(struct scsi_cmnd *cmd) { /* * We don't have to worry about this one timing out any more. @@ -836,7 +838,7 @@ static void scsi_softirq(struct softirq_action *h) * level drivers should not become re-entrant as a result of * this. */ -int scsi_retry_command(struct scsi_cmnd *cmd) +static int scsi_retry_command(struct scsi_cmnd *cmd) { /* * Restore the SCSI command state. diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c index e0208886b45e..322b5a41a36f 100644 --- a/drivers/scsi/scsi_debug.c +++ b/drivers/scsi/scsi_debug.c @@ -1783,7 +1783,7 @@ static void __exit scsi_debug_exit(void) device_initcall(scsi_debug_init); module_exit(scsi_debug_exit); -void pseudo_0_release(struct device * dev) +static void pseudo_0_release(struct device * dev) { if (SCSI_DEBUG_OPT_NOISE & scsi_debug_opts) printk(KERN_INFO "scsi_debug: pseudo_0_release() called\n"); diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 9f996499fa9d..621dee8b8cb2 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -44,7 +44,7 @@ struct scsi_host_sg_pool { #endif #define SP(x) { x, "sgpool-" #x } -struct scsi_host_sg_pool scsi_sg_pools[] = { +static struct scsi_host_sg_pool scsi_sg_pools[] = { SP(8), SP(16), SP(32), diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h index c01580df4476..96d4f745975c 100644 --- a/drivers/scsi/scsi_priv.h +++ b/drivers/scsi/scsi_priv.h @@ -61,8 +61,6 @@ extern void scsi_exit_hosts(void); extern int scsi_dispatch_cmd(struct scsi_cmnd *cmd); extern int scsi_setup_command_freelist(struct Scsi_Host *shost); extern void scsi_destroy_command_freelist(struct Scsi_Host *shost); -extern void scsi_done(struct scsi_cmnd *cmd); -extern int scsi_retry_command(struct scsi_cmnd *cmd); extern int scsi_insert_special_req(struct scsi_request *sreq, int); extern void scsi_init_cmd_from_req(struct scsi_cmnd *cmd, struct scsi_request *sreq); @@ -136,7 +134,6 @@ extern void scsi_exit_sysctl(void); #endif /* CONFIG_SYSCTL */ /* scsi_sysfs.c */ -extern void scsi_device_dev_release(struct device *); extern int scsi_sysfs_add_sdev(struct scsi_device *); extern int scsi_sysfs_add_host(struct Scsi_Host *); extern int scsi_sysfs_register(void); @@ -145,7 +142,6 @@ extern void scsi_sysfs_device_initialize(struct scsi_device *); extern int scsi_sysfs_target_initialize(struct scsi_device *); extern struct scsi_transport_template blank_transport_template; -extern struct class sdev_class; extern struct bus_type scsi_bus_type; /* diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c index 93b41100a6d8..beed7fbe1cbe 100644 --- a/drivers/scsi/scsi_sysfs.c +++ b/drivers/scsi/scsi_sysfs.c @@ -150,7 +150,7 @@ static void scsi_device_cls_release(struct class_device *class_dev) put_device(&sdev->sdev_gendev); } -void scsi_device_dev_release(struct device *dev) +static void scsi_device_dev_release(struct device *dev) { struct scsi_device *sdev; struct device *parent; @@ -185,7 +185,7 @@ void scsi_device_dev_release(struct device *dev) put_device(parent); } -struct class sdev_class = { +static struct class sdev_class = { .name = "scsi_device", .release = scsi_device_cls_release, }; diff --git a/fs/namespace.c b/fs/namespace.c index 3b93e5d750eb..208c079e9fdb 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -337,7 +337,7 @@ int may_umount(struct vfsmount *mnt) EXPORT_SYMBOL(may_umount); -void umount_tree(struct vfsmount *mnt) +static void umount_tree(struct vfsmount *mnt) { struct vfsmount *p; LIST_HEAD(kill); diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c index c47f8fd31a2d..63158491e152 100644 --- a/fs/reiserfs/stree.c +++ b/fs/reiserfs/stree.c @@ -223,7 +223,7 @@ extern struct tree_balance * cur_tb; const struct reiserfs_key MIN_KEY = {0, 0, {{0, 0},}}; /* Maximal possible key. It is never in the tree. */ -const struct reiserfs_key MAX_KEY = { +static const struct reiserfs_key MAX_KEY = { __constant_cpu_to_le32(0xffffffff), __constant_cpu_to_le32(0xffffffff), {{__constant_cpu_to_le32(0xffffffff), diff --git a/include/linux/irq.h b/include/linux/irq.h index 7fc1022be9ee..12277799c007 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -88,7 +88,6 @@ extern fastcall int handle_IRQ_event(unsigned int irq, struct pt_regs *regs, struct irqaction *action); extern fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs); extern void note_interrupt(unsigned int irq, irq_desc_t *desc, int action_ret); -extern void report_bad_irq(unsigned int irq, irq_desc_t *desc, int action_ret); extern int can_request_irq(unsigned int irq, unsigned long irqflags); extern void init_irq_proc(void); diff --git a/include/linux/namespace.h b/include/linux/namespace.h index 9eca1558d72f..697991b69f9b 100644 --- a/include/linux/namespace.h +++ b/include/linux/namespace.h @@ -12,7 +12,6 @@ struct namespace { struct rw_semaphore sem; }; -extern void umount_tree(struct vfsmount *); extern int copy_namespace(int, struct task_struct *); extern void __put_namespace(struct namespace *namespace); diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index a53e08a45e32..88d9fe5975d5 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -131,7 +131,6 @@ sctp_state_fn_t sctp_sf_do_ecne; sctp_state_fn_t sctp_sf_ootb; sctp_state_fn_t sctp_sf_pdiscard; sctp_state_fn_t sctp_sf_violation; -sctp_state_fn_t sctp_sf_violation_chunklen; sctp_state_fn_t sctp_sf_discard_chunk; sctp_state_fn_t sctp_sf_do_5_2_1_siminit; sctp_state_fn_t sctp_sf_do_5_2_2_dupinit; @@ -259,11 +258,6 @@ struct sctp_chunk *sctp_make_fwdtsn(const struct sctp_association *asoc, void sctp_chunk_assign_tsn(struct sctp_chunk *); void sctp_chunk_assign_ssn(struct sctp_chunk *); -sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, - __u16 error, - const struct sctp_association *asoc, - struct sctp_transport *transport); - /* Prototypes for statetable processing. */ int sctp_do_sm(sctp_event_t event_type, sctp_subtype_t subtype, diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index f6297c306905..ba039e827d58 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -45,7 +45,7 @@ __report_bad_irq(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret) } } -void report_bad_irq(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret) +static void report_bad_irq(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret) { static int count = 100; diff --git a/kernel/module.c b/kernel/module.c index 0494c89a0d26..068e271ab3a5 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -730,7 +730,7 @@ static int obsparm_copy_string(const char *val, struct kernel_param *kp) return 0; } -int set_obsolete(const char *val, struct kernel_param *kp) +static int set_obsolete(const char *val, struct kernel_param *kp) { unsigned int min, max; unsigned int size, maxsize; diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index 90b3b68dee3f..53f9f8720ee4 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -81,7 +81,7 @@ static int nr_copy_pages_check; extern char resume_file[]; /* Local variables that should not be affected by save */ -unsigned int nr_copy_pages __nosavedata = 0; +static unsigned int nr_copy_pages __nosavedata = 0; /* Suspend pagedir is allocated before final copy, therefore it must be freed after resume diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 058189684c7c..86073df418f5 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -92,6 +92,17 @@ static sctp_disposition_t sctp_sf_shut_8_4_5(const struct sctp_endpoint *ep, sctp_cmd_seq_t *commands); static struct sctp_sackhdr *sctp_sm_pull_sack(struct sctp_chunk *chunk); +static sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, + __u16 error, + const struct sctp_association *asoc, + struct sctp_transport *transport); + +static sctp_disposition_t sctp_sf_violation_chunklen( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands); /* Small helper function that checks if the chunk length * is of the appropriate length. The 'required_length' argument @@ -2328,7 +2339,7 @@ sctp_disposition_t sctp_sf_cookie_echoed_abort(const struct sctp_endpoint *ep, * * This is common code called by several sctp_sf_*_abort() functions above. */ -sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, +static sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error, const struct sctp_association *asoc, struct sctp_transport *transport) @@ -3687,7 +3698,8 @@ sctp_disposition_t sctp_sf_violation(const struct sctp_endpoint *ep, * * Generate an ABORT chunk and terminate the association. */ -sctp_disposition_t sctp_sf_violation_chunklen(const struct sctp_endpoint *ep, +static sctp_disposition_t sctp_sf_violation_chunklen( + const struct sctp_endpoint *ep, const struct sctp_association *asoc, const sctp_subtype_t type, void *arg, -- cgit v1.2.3 From 75043cb5b386e5a01fd03b88f647dd992de02f97 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 24 Jun 2005 20:52:52 +0000 Subject: [PATCH] fs/qnx4/*: fix sparse warnings This patch fixes sparse warnings in the qnx4fs (and might even make qnx4fs work on big-endian boxes) Signed-off-by: Alexey Dobriyan Signed-off-by: Domen Puncer Signed-off-by: Anders Larsen Signed-off-by: Linus Torvalds --- fs/qnx4/dir.c | 2 +- fs/qnx4/inode.c | 4 ++-- include/linux/qnx4_fs.h | 18 +++++++++--------- include/linux/qnxtypes.h | 16 ++++++++-------- 4 files changed, 20 insertions(+), 20 deletions(-) (limited to 'include/linux') diff --git a/fs/qnx4/dir.c b/fs/qnx4/dir.c index cd66147cca04..7a8f5595c26f 100644 --- a/fs/qnx4/dir.c +++ b/fs/qnx4/dir.c @@ -61,7 +61,7 @@ static int qnx4_readdir(struct file *filp, void *dirent, filldir_t filldir) ino = blknum * QNX4_INODES_PER_BLOCK + ix - 1; else { le = (struct qnx4_link_info*)de; - ino = ( le->dl_inode_blk - 1 ) * + ino = ( le32_to_cpu(le->dl_inode_blk) - 1 ) * QNX4_INODES_PER_BLOCK + le->dl_inode_ndx; } diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index aa92d6b76a9a..b79162a35478 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -236,7 +236,7 @@ unsigned long qnx4_block_map( struct inode *inode, long iblock ) struct buffer_head *bh = NULL; struct qnx4_xblk *xblk = NULL; struct qnx4_inode_entry *qnx4_inode = qnx4_raw_inode(inode); - qnx4_nxtnt_t nxtnt = le16_to_cpu(qnx4_inode->di_num_xtnts); + u16 nxtnt = le16_to_cpu(qnx4_inode->di_num_xtnts); if ( iblock < le32_to_cpu(qnx4_inode->di_first_xtnt.xtnt_size) ) { // iblock is in the first extent. This is easy. @@ -372,7 +372,7 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent) printk("qnx4: unable to read the superblock\n"); goto outnobh; } - if ( le32_to_cpu( *(__u32*)bh->b_data ) != QNX4_SUPER_MAGIC ) { + if ( le32_to_cpup((__le32*) bh->b_data) != QNX4_SUPER_MAGIC ) { if (!silent) printk("qnx4: wrong fsid in superblock.\n"); goto out; diff --git a/include/linux/qnx4_fs.h b/include/linux/qnx4_fs.h index 22ba580b0ae8..fc610bb0f733 100644 --- a/include/linux/qnx4_fs.h +++ b/include/linux/qnx4_fs.h @@ -46,11 +46,11 @@ struct qnx4_inode_entry { char di_fname[QNX4_SHORT_NAME_MAX]; qnx4_off_t di_size; qnx4_xtnt_t di_first_xtnt; - __u32 di_xblk; - __s32 di_ftime; - __s32 di_mtime; - __s32 di_atime; - __s32 di_ctime; + __le32 di_xblk; + __le32 di_ftime; + __le32 di_mtime; + __le32 di_atime; + __le32 di_ctime; qnx4_nxtnt_t di_num_xtnts; qnx4_mode_t di_mode; qnx4_muid_t di_uid; @@ -63,18 +63,18 @@ struct qnx4_inode_entry { struct qnx4_link_info { char dl_fname[QNX4_NAME_MAX]; - __u32 dl_inode_blk; + __le32 dl_inode_blk; __u8 dl_inode_ndx; __u8 dl_spare[10]; __u8 dl_status; }; struct qnx4_xblk { - __u32 xblk_next_xblk; - __u32 xblk_prev_xblk; + __le32 xblk_next_xblk; + __le32 xblk_prev_xblk; __u8 xblk_num_xtnts; __u8 xblk_spare[3]; - __s32 xblk_num_blocks; + __le32 xblk_num_blocks; qnx4_xtnt_t xblk_xtnts[QNX4_MAX_XTNTS_PER_XBLK]; char xblk_signature[8]; qnx4_xtnt_t xblk_first_xtnt; diff --git a/include/linux/qnxtypes.h b/include/linux/qnxtypes.h index fb518e318c7c..a3eb1137857b 100644 --- a/include/linux/qnxtypes.h +++ b/include/linux/qnxtypes.h @@ -12,18 +12,18 @@ #ifndef _QNX4TYPES_H #define _QNX4TYPES_H -typedef __u16 qnx4_nxtnt_t; +typedef __le16 qnx4_nxtnt_t; typedef __u8 qnx4_ftype_t; typedef struct { - __u32 xtnt_blk; - __u32 xtnt_size; + __le32 xtnt_blk; + __le32 xtnt_size; } qnx4_xtnt_t; -typedef __u16 qnx4_mode_t; -typedef __u16 qnx4_muid_t; -typedef __u16 qnx4_mgid_t; -typedef __u32 qnx4_off_t; -typedef __u16 qnx4_nlink_t; +typedef __le16 qnx4_mode_t; +typedef __le16 qnx4_muid_t; +typedef __le16 qnx4_mgid_t; +typedef __le32 qnx4_off_t; +typedef __le16 qnx4_nlink_t; #endif -- cgit v1.2.3 From e70c9d5e61c6cb2272c866fc1303e62975006752 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Sat, 25 Jun 2005 14:54:25 -0700 Subject: [PATCH] I8K: use standard DMI interface I8K: Change to use stock dmi infrastructure instead of homegrown parsing code. The driver now requires box's DMI data to match list of supported models so driver can be safely compiled-in by default without fear of it poking into random SMM BIOS code. DMI checks can be ignored with i8k.ignore_dmi option. Signed-off-by: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 3 + arch/i386/kernel/dmi_scan.c | 6 +- drivers/char/i8k.c | 302 +++++++----------------------------- include/linux/dmi.h | 1 + 4 files changed, 60 insertions(+), 252 deletions(-) (limited to 'include/linux') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 4924d387a657..86db43fd6b0f 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -548,6 +548,9 @@ running once the system is up. i810= [HW,DRM] + i8k.ignore_dmi [HW] Continue probing hardware even if DMI data + indicates that the driver is running on unsupported + hardware. i8k.force [HW] Activate i8k driver even if SMM BIOS signature does not match list of supported models. i8k.power_status diff --git a/arch/i386/kernel/dmi_scan.c b/arch/i386/kernel/dmi_scan.c index 6ed7e28f306c..3facd20212bb 100644 --- a/arch/i386/kernel/dmi_scan.c +++ b/arch/i386/kernel/dmi_scan.c @@ -414,6 +414,7 @@ static void __init dmi_decode(struct dmi_header *dm) dmi_save_ident(dm, DMI_PRODUCT_VERSION, 6); dmi_printk(("Serial Number: %s\n", dmi_string(dm, data[7]))); + dmi_save_ident(dm, DMI_PRODUCT_SERIAL, 7); break; case 2: dmi_printk(("Board Vendor: %s\n", @@ -470,7 +471,6 @@ fail: d++; return count; } - EXPORT_SYMBOL(dmi_check_system); /** @@ -480,8 +480,8 @@ EXPORT_SYMBOL(dmi_check_system); * Returns one DMI data value, can be used to perform * complex DMI data checks. */ -char * dmi_get_system_info(int field) +char *dmi_get_system_info(int field) { return dmi_ident[field]; } - +EXPORT_SYMBOL(dmi_get_system_info); diff --git a/drivers/char/i8k.c b/drivers/char/i8k.c index bf5e43beca62..81d2f675fb77 100644 --- a/drivers/char/i8k.c +++ b/drivers/char/i8k.c @@ -20,7 +20,7 @@ #include #include #include -#include +#include #include #include @@ -52,18 +52,7 @@ #define I8K_TEMPERATURE_BUG 1 -#define DELL_SIGNATURE "Dell Computer" - -static char *supported_models[] = { - "Inspiron", - "Latitude", - NULL -}; - -static char system_vendor[48] = "?"; -static char product_name[48] = "?"; -static char bios_version[4] = "?"; -static char serial_number[16] = "?"; +static char bios_version[4]; MODULE_AUTHOR("Massimo Dal Zotto (dz@debian.org)"); MODULE_DESCRIPTION("Driver for accessing SMM BIOS on Dell laptops"); @@ -73,6 +62,10 @@ static int force; module_param(force, bool, 0); MODULE_PARM_DESC(force, "Force loading without checking for supported models"); +static int ignore_dmi; +module_param(ignore_dmi, bool, 0); +MODULE_PARM_DESC(ignore_dmi, "Continue probing hardware even if DMI data does not match"); + static int restricted; module_param(restricted, bool, 0); MODULE_PARM_DESC(restricted, "Allow fan control if SYS_ADMIN capability set"); @@ -99,11 +92,10 @@ typedef struct { unsigned int edi __attribute__ ((packed)); } SMMRegisters; -typedef struct { - u8 type; - u8 length; - u16 handle; -} DMIHeader; +static inline char *i8k_get_dmi_data(int field) +{ + return dmi_get_system_info(field) ? : "N/A"; +} /* * Call the System Management Mode BIOS. Code provided by Jonathan Buzzard. @@ -162,15 +154,6 @@ static int i8k_get_bios_version(void) return regs.eax; } -/* - * Read the machine id. - */ -static int i8k_get_serial_number(unsigned char *buff) -{ - strlcpy(buff, serial_number, sizeof(serial_number)); - return 0; -} - /* * Read the Fn key status. */ @@ -328,7 +311,7 @@ static int i8k_get_dell_signature(void) static int i8k_ioctl(struct inode *ip, struct file *fp, unsigned int cmd, unsigned long arg) { - int val; + int val = 0; int speed; unsigned char buff[16]; int __user *argp = (int __user *)arg; @@ -343,7 +326,7 @@ static int i8k_ioctl(struct inode *ip, struct file *fp, unsigned int cmd, case I8K_MACHINE_ID: memset(buff, 0, 16); - val = i8k_get_serial_number(buff); + strlcpy(buff, i8k_get_dmi_data(DMI_PRODUCT_SERIAL), sizeof(buff)); break; case I8K_FN_STATUS: @@ -451,10 +434,10 @@ static int i8k_get_info(char *buffer, char **start, off_t fpos, int length) n = sprintf(buffer, "%s %s %s %d %d %d %d %d %d %d\n", I8K_PROC_FMT, bios_version, - serial_number, + dmi_get_system_info(DMI_PRODUCT_SERIAL) ? : "N/A", cpu_temp, - left_fan, - right_fan, left_speed, right_speed, ac_power, fn_key); + left_fan, right_fan, left_speed, right_speed, + ac_power, fn_key); return n; } @@ -486,201 +469,23 @@ static ssize_t i8k_read(struct file *f, char __user * buffer, size_t len, return len; } -static char *__init string_trim(char *s, int size) -{ - int len; - char *p; - - if ((len = strlen(s)) > size) { - len = size; - } - - for (p = s + len - 1; len && (*p == ' '); len--, p--) { - *p = '\0'; - } - - return s; -} - -/* DMI code, stolen from arch/i386/kernel/dmi_scan.c */ - -/* - * |<-- dmi->length -->| - * | | - * |dmi header s=N | string1,\0, ..., stringN,\0, ..., \0 - * | | - * +-----------------------+ - */ -static char *__init dmi_string(DMIHeader * dmi, u8 s) -{ - u8 *p; - - if (!s) { - return ""; - } - s--; - - p = (u8 *) dmi + dmi->length; - while (s > 0) { - p += strlen(p); - p++; - s--; - } - - return p; -} - -static void __init dmi_decode(DMIHeader * dmi) -{ - u8 *data = (u8 *) dmi; - char *p; - -#ifdef I8K_DEBUG - int i; - printk("%08x ", (int)data); - for (i = 0; i < data[1] && i < 64; i++) { - printk("%02x ", data[i]); - } - printk("\n"); -#endif - - switch (dmi->type) { - case 0: /* BIOS Information */ - p = dmi_string(dmi, data[5]); - if (*p) { - strlcpy(bios_version, p, sizeof(bios_version)); - string_trim(bios_version, sizeof(bios_version)); - } - break; - case 1: /* System Information */ - p = dmi_string(dmi, data[4]); - if (*p) { - strlcpy(system_vendor, p, sizeof(system_vendor)); - string_trim(system_vendor, sizeof(system_vendor)); - } - p = dmi_string(dmi, data[5]); - if (*p) { - strlcpy(product_name, p, sizeof(product_name)); - string_trim(product_name, sizeof(product_name)); - } - p = dmi_string(dmi, data[7]); - if (*p) { - strlcpy(serial_number, p, sizeof(serial_number)); - string_trim(serial_number, sizeof(serial_number)); - } - break; - } -} - -static int __init dmi_table(u32 base, int len, int num, - void (*fn) (DMIHeader *)) -{ - u8 *buf; - u8 *data; - DMIHeader *dmi; - int i = 1; - - buf = ioremap(base, len); - if (buf == NULL) { - return -1; - } - data = buf; - - /* - * Stop when we see al the items the table claimed to have - * or we run off the end of the table (also happens) - */ - while ((i < num) && ((data - buf) < len)) { - dmi = (DMIHeader *) data; - /* - * Avoid misparsing crud if the length of the last - * record is crap - */ - if ((data - buf + dmi->length) >= len) { - break; - } - fn(dmi); - data += dmi->length; - /* - * Don't go off the end of the data if there is - * stuff looking like string fill past the end - */ - while (((data - buf) < len) && (*data || data[1])) { - data++; - } - data += 2; - i++; - } - iounmap(buf); - - return 0; -} - -static int __init dmi_iterate(void (*decode) (DMIHeader *)) -{ - unsigned char buf[20]; - void __iomem *p = ioremap(0xe0000, 0x20000), *q; - - if (!p) - return -1; - - for (q = p; q < p + 0x20000; q += 16) { - memcpy_fromio(buf, q, 20); - if (memcmp(buf, "_DMI_", 5) == 0) { - u16 num = buf[13] << 8 | buf[12]; - u16 len = buf[7] << 8 | buf[6]; - u32 base = buf[11] << 24 | buf[10] << 16 | buf[9] << 8 | buf[8]; -#ifdef I8K_DEBUG - printk(KERN_INFO "DMI %d.%d present.\n", - buf[14] >> 4, buf[14] & 0x0F); - printk(KERN_INFO "%d structures occupying %d bytes.\n", - buf[13] << 8 | buf[12], buf[7] << 8 | buf[6]); - printk(KERN_INFO "DMI table at 0x%08X.\n", - buf[11] << 24 | buf[10] << 16 | buf[9] << 8 | - buf[8]); -#endif - if (dmi_table(base, len, num, decode) == 0) { - iounmap(p); - return 0; - } - } - } - iounmap(p); - return -1; -} - -/* end of DMI code */ - -/* - * Get DMI information. - */ -static int __init i8k_dmi_probe(void) -{ - char **p; - - if (dmi_iterate(dmi_decode) != 0) { - printk(KERN_INFO "i8k: unable to get DMI information\n"); - return -ENODEV; - } - - if (strncmp(system_vendor, DELL_SIGNATURE, strlen(DELL_SIGNATURE)) != 0) { - printk(KERN_INFO "i8k: not running on a Dell system\n"); - return -ENODEV; - } - - for (p = supported_models;; p++) { - if (!*p) { - printk(KERN_INFO "i8k: unsupported model: %s\n", - product_name); - return -ENODEV; - } - if (strncmp(product_name, *p, strlen(*p)) == 0) { - break; - } - } - - return 0; -} +static struct dmi_system_id __initdata i8k_dmi_table[] = { + { + .ident = "Dell Inspiron", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer"), + DMI_MATCH(DMI_PRODUCT_NAME, "Inspiron"), + }, + }, + { + .ident = "Dell Latitude", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer"), + DMI_MATCH(DMI_PRODUCT_NAME, "Latitude"), + }, + }, + { } +}; /* * Probe for the presence of a supported laptop. @@ -689,23 +494,30 @@ static int __init i8k_probe(void) { char buff[4]; int version; - int smm_found = 0; /* * Get DMI information */ - if (i8k_dmi_probe() != 0) { + if (!dmi_check_system(i8k_dmi_table)) { + if (!ignore_dmi && !force) + return -ENODEV; + + printk(KERN_INFO "i8k: not running on a supported Dell system.\n"); printk(KERN_INFO "i8k: vendor=%s, model=%s, version=%s\n", - system_vendor, product_name, bios_version); + i8k_get_dmi_data(DMI_SYS_VENDOR), + i8k_get_dmi_data(DMI_PRODUCT_NAME), + i8k_get_dmi_data(DMI_BIOS_VERSION)); } + strlcpy(bios_version, i8k_get_dmi_data(DMI_BIOS_VERSION), sizeof(bios_version)); + /* * Get SMM Dell signature */ if (i8k_get_dell_signature() != 0) { - printk(KERN_INFO "i8k: unable to get SMM Dell signature\n"); - } else { - smm_found = 1; + printk(KERN_ERR "i8k: unable to get SMM Dell signature\n"); + if (!force) + return -ENODEV; } /* @@ -713,9 +525,8 @@ static int __init i8k_probe(void) */ version = i8k_get_bios_version(); if (version <= 0) { - printk(KERN_INFO "i8k: unable to get SMM BIOS version\n"); + printk(KERN_WARNING "i8k: unable to get SMM BIOS version\n"); } else { - smm_found = 1; buff[0] = (version >> 16) & 0xff; buff[1] = (version >> 8) & 0xff; buff[2] = (version) & 0xff; @@ -723,21 +534,15 @@ static int __init i8k_probe(void) /* * If DMI BIOS version is unknown use SMM BIOS version. */ - if (bios_version[0] == '?') { - strcpy(bios_version, buff); - } + if (!dmi_get_system_info(DMI_BIOS_VERSION)) + strlcpy(bios_version, buff, sizeof(bios_version)); + /* * Check if the two versions match. */ - if (strncmp(buff, bios_version, sizeof(bios_version)) != 0) { - printk(KERN_INFO - "i8k: BIOS version mismatch: %s != %s\n", buff, - bios_version); - } - } - - if (!smm_found && !force) { - return -ENODEV; + if (strncmp(buff, bios_version, sizeof(bios_version)) != 0) + printk(KERN_WARNING "i8k: BIOS version mismatch: %s != %s\n", + buff, bios_version); } return 0; @@ -751,9 +556,8 @@ int __init i8k_init(void) struct proc_dir_entry *proc_i8k; /* Are we running on an supported laptop? */ - if (i8k_probe() != 0) { + if (i8k_probe()) return -ENODEV; - } /* Register the proc entry */ proc_i8k = create_proc_info_entry("i8k", 0, NULL, i8k_get_info); diff --git a/include/linux/dmi.h b/include/linux/dmi.h index d2bcf556088b..5e93e6dce9a4 100644 --- a/include/linux/dmi.h +++ b/include/linux/dmi.h @@ -9,6 +9,7 @@ enum dmi_field { DMI_SYS_VENDOR, DMI_PRODUCT_NAME, DMI_PRODUCT_VERSION, + DMI_PRODUCT_SERIAL, DMI_BOARD_VENDOR, DMI_BOARD_NAME, DMI_BOARD_VERSION, -- cgit v1.2.3 From 52a119feaad92d44a0e97d01b22afbcbaf3fc079 Mon Sep 17 00:00:00 2001 From: Ashok Raj Date: Sat, 25 Jun 2005 14:54:57 -0700 Subject: [PATCH] make smp_prepare_cpu to a weak function I really wish smp_prepare_cpu() would disappear eventually. In the interim this is ideally a weak function, so we dont end up changing several places to define this dummy in headers. Today since the dummy declaration is done only in drivers/base/cpu.c but the function is called in kernel/power/smp.c i get undefined reference in my cpu hotplug code for x86_64 under development. Signed-off-by: Ashok Raj Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/cpu.c | 9 +++++---- include/asm-i386/smp.h | 3 --- include/linux/cpu.h | 1 + 3 files changed, 6 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index bdd7e9f55c81..0bf2dc11cdb8 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -16,9 +16,10 @@ struct sysdev_class cpu_sysdev_class = { EXPORT_SYMBOL(cpu_sysdev_class); #ifdef CONFIG_HOTPLUG_CPU -#ifndef __HAVE_ARCH_SMP_PREPARE_CPU -#define smp_prepare_cpu(cpu) (0) -#endif +int __attribute__((weak)) smp_prepare_cpu (int cpu) +{ + return 0; +} static ssize_t show_online(struct sys_device *dev, char *buf) { @@ -41,7 +42,7 @@ static ssize_t store_online(struct sys_device *dev, const char *buf, break; case '1': ret = smp_prepare_cpu(cpu->sysdev.id); - if (ret == 0) + if (!ret) ret = cpu_up(cpu->sysdev.id); break; default: diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h index c9996eda5408..edad9b4712fa 100644 --- a/include/asm-i386/smp.h +++ b/include/asm-i386/smp.h @@ -51,9 +51,6 @@ extern u8 x86_cpu_to_apicid[]; #ifdef CONFIG_HOTPLUG_CPU extern void cpu_exit_clear(void); extern void cpu_uninit(void); - -#define __HAVE_ARCH_SMP_PREPARE_CPU -extern int smp_prepare_cpu(int cpu); #endif /* diff --git a/include/linux/cpu.h b/include/linux/cpu.h index fe0298e5dae1..e8904c0da686 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -69,6 +69,7 @@ extern struct semaphore cpucontrol; register_cpu_notifier(&fn##_nb); \ } int cpu_down(unsigned int cpu); +extern int __attribute__((weak)) smp_prepare_cpu(int cpu); #define cpu_is_offline(cpu) unlikely(!cpu_online(cpu)) #else #define lock_cpu_hotplug() do { } while (0) -- cgit v1.2.3 From e6982c671c560da4a0bc5c908cbcbec12bd5991d Mon Sep 17 00:00:00 2001 From: Ashok Raj Date: Sat, 25 Jun 2005 14:54:58 -0700 Subject: [PATCH] x86_64: Change init sections for CPU hotplug support This patch adds __cpuinit and __cpuinitdata sections that need to exist past boot to support cpu hotplug. Caveat: This is done *only* for EM64T CPU Hotplug support, on request from Andi Kleen. Much of the generic hotplug code in kernel, and none of the other archs that support CPU hotplug today, i386, ia64, ppc64, s390 and parisc dont mark sections with __cpuinit, but only mark them as __devinit, and __devinitdata. If someone is motivated to change generic code, we need to make sure all existing hotplug code does not break, on other arch's that dont use __cpuinit, and __cpudevinit. Signed-off-by: Ashok Raj Acked-by: Andi Kleen Acked-by: Zwane Mwaikambo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/apic.c | 8 ++++---- arch/x86_64/kernel/i387.c | 2 +- arch/x86_64/kernel/mce.c | 8 ++++---- arch/x86_64/kernel/mce_intel.c | 4 ++-- arch/x86_64/kernel/nmi.c | 4 ++-- arch/x86_64/kernel/process.c | 2 +- arch/x86_64/kernel/setup.c | 18 +++++++++--------- arch/x86_64/kernel/setup64.c | 6 +++--- arch/x86_64/kernel/smpboot.c | 15 +++++---------- arch/x86_64/mm/numa.c | 2 +- include/linux/init.h | 12 ++++++++++++ 11 files changed, 44 insertions(+), 37 deletions(-) (limited to 'include/linux') diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c index f8e6cc4fecd4..2a6c893ccf60 100644 --- a/arch/x86_64/kernel/apic.c +++ b/arch/x86_64/kernel/apic.c @@ -285,7 +285,7 @@ void __init init_bsp_APIC(void) apic_write_around(APIC_LVT1, value); } -void __init setup_local_APIC (void) +void __cpuinit setup_local_APIC (void) { unsigned int value, ver, maxlvt; @@ -534,7 +534,7 @@ static struct sys_device device_lapic = { .cls = &lapic_sysclass, }; -static void __init apic_pm_activate(void) +static void __cpuinit apic_pm_activate(void) { apic_pm_state.active = 1; } @@ -774,14 +774,14 @@ void __init setup_boot_APIC_clock (void) local_irq_enable(); } -void __init setup_secondary_APIC_clock(void) +void __cpuinit setup_secondary_APIC_clock(void) { local_irq_disable(); /* FIXME: Do we need this? --RR */ setup_APIC_timer(calibration_result); local_irq_enable(); } -void __init disable_APIC_timer(void) +void __cpuinit disable_APIC_timer(void) { if (using_apic_timer) { unsigned long v; diff --git a/arch/x86_64/kernel/i387.c b/arch/x86_64/kernel/i387.c index ba139cac57ce..d9b22b633e39 100644 --- a/arch/x86_64/kernel/i387.c +++ b/arch/x86_64/kernel/i387.c @@ -42,7 +42,7 @@ void mxcsr_feature_mask_init(void) * Called at bootup to set up the initial FPU state that is later cloned * into all processes. */ -void __init fpu_init(void) +void __cpuinit fpu_init(void) { unsigned long oldcr0 = read_cr0(); extern void __bad_fxsave_alignment(void); diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index 3a89d735a4f6..7ab15c8ab95f 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -327,7 +327,7 @@ static void mce_init(void *dummy) } /* Add per CPU specific workarounds here */ -static void __init mce_cpu_quirks(struct cpuinfo_x86 *c) +static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c) { /* This should be disabled by the BIOS, but isn't always */ if (c->x86_vendor == X86_VENDOR_AMD && c->x86 == 15) { @@ -337,7 +337,7 @@ static void __init mce_cpu_quirks(struct cpuinfo_x86 *c) } } -static void __init mce_cpu_features(struct cpuinfo_x86 *c) +static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c) { switch (c->x86_vendor) { case X86_VENDOR_INTEL: @@ -352,7 +352,7 @@ static void __init mce_cpu_features(struct cpuinfo_x86 *c) * Called for each booted CPU to set up machine checks. * Must be called with preempt off. */ -void __init mcheck_init(struct cpuinfo_x86 *c) +void __cpuinit mcheck_init(struct cpuinfo_x86 *c) { static cpumask_t mce_cpus __initdata = CPU_MASK_NONE; @@ -542,7 +542,7 @@ ACCESSOR(bank4ctl,bank[4],mce_restart()) ACCESSOR(tolerant,tolerant,) ACCESSOR(check_interval,check_interval,mce_restart()) -static __init int mce_init_device(void) +static __cpuinit int mce_init_device(void) { int err; if (!mce_available(&boot_cpu_data)) diff --git a/arch/x86_64/kernel/mce_intel.c b/arch/x86_64/kernel/mce_intel.c index 4db9a640069f..0be0a7959814 100644 --- a/arch/x86_64/kernel/mce_intel.c +++ b/arch/x86_64/kernel/mce_intel.c @@ -42,7 +42,7 @@ done: irq_exit(); } -static void __init intel_init_thermal(struct cpuinfo_x86 *c) +static void __cpuinit intel_init_thermal(struct cpuinfo_x86 *c) { u32 l, h; int tm2 = 0; @@ -93,7 +93,7 @@ static void __init intel_init_thermal(struct cpuinfo_x86 *c) return; } -void __init mce_intel_feature_init(struct cpuinfo_x86 *c) +void __cpuinit mce_intel_feature_init(struct cpuinfo_x86 *c) { intel_init_thermal(c); } diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index 31c0f2e6ac91..4e44d6e6b7e5 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -98,7 +98,7 @@ static unsigned int nmi_p4_cccr_val; (P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT| \ P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE) -static __init inline int nmi_known_cpu(void) +static __cpuinit inline int nmi_known_cpu(void) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: @@ -110,7 +110,7 @@ static __init inline int nmi_known_cpu(void) } /* Run after command line and cpu_init init, but before all other checks */ -void __init nmi_watchdog_default(void) +void __cpuinit nmi_watchdog_default(void) { if (nmi_watchdog != NMI_DEFAULT) return; diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index e59d1f9d6163..15fa35a81a72 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -204,7 +204,7 @@ static void mwait_idle(void) } } -void __init select_idle_routine(const struct cpuinfo_x86 *c) +void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) { static int printed; if (cpu_has(c, X86_FEATURE_MWAIT)) { diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index 000015dd5a8b..ca9a915a6d4a 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -676,7 +676,7 @@ void __init setup_arch(char **cmdline_p) #endif } -static int __init get_model_name(struct cpuinfo_x86 *c) +static int __cpuinit get_model_name(struct cpuinfo_x86 *c) { unsigned int *v; @@ -692,7 +692,7 @@ static int __init get_model_name(struct cpuinfo_x86 *c) } -static void __init display_cacheinfo(struct cpuinfo_x86 *c) +static void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) { unsigned int n, dummy, eax, ebx, ecx, edx; @@ -803,7 +803,7 @@ static int __init init_amd(struct cpuinfo_x86 *c) return r; } -static void __init detect_ht(struct cpuinfo_x86 *c) +static void __cpuinit detect_ht(struct cpuinfo_x86 *c) { #ifdef CONFIG_SMP u32 eax, ebx, ecx, edx; @@ -864,7 +864,7 @@ static void __init detect_ht(struct cpuinfo_x86 *c) /* * find out the number of processor cores on the die */ -static int __init intel_num_cpu_cores(struct cpuinfo_x86 *c) +static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c) { unsigned int eax; @@ -882,7 +882,7 @@ static int __init intel_num_cpu_cores(struct cpuinfo_x86 *c) return 1; } -static void __init init_intel(struct cpuinfo_x86 *c) +static void __cpuinit init_intel(struct cpuinfo_x86 *c) { /* Cache sizes */ unsigned n; @@ -902,7 +902,7 @@ static void __init init_intel(struct cpuinfo_x86 *c) c->x86_num_cores = intel_num_cpu_cores(c); } -void __init get_cpu_vendor(struct cpuinfo_x86 *c) +void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) { char *v = c->x86_vendor_id; @@ -923,7 +923,7 @@ struct cpu_model_info { /* Do some early cpuid on the boot CPU to get some parameter that are needed before check_bugs. Everything advanced is in identify_cpu below. */ -void __init early_identify_cpu(struct cpuinfo_x86 *c) +void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) { u32 tfms; @@ -977,7 +977,7 @@ void __init early_identify_cpu(struct cpuinfo_x86 *c) /* * This does the hard work of actually picking apart the CPU stuff... */ -void __init identify_cpu(struct cpuinfo_x86 *c) +void __cpuinit identify_cpu(struct cpuinfo_x86 *c) { int i; u32 xlvl; @@ -1054,7 +1054,7 @@ void __init identify_cpu(struct cpuinfo_x86 *c) } -void __init print_cpu_info(struct cpuinfo_x86 *c) +void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) { if (c->x86_model_id[0]) printk("%s", c->x86_model_id); diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c index 678b7ac33b8b..f2b9c6bc999a 100644 --- a/arch/x86_64/kernel/setup64.c +++ b/arch/x86_64/kernel/setup64.c @@ -29,7 +29,7 @@ char x86_boot_params[BOOT_PARAM_SIZE] __initdata = {0,}; -cpumask_t cpu_initialized __initdata = CPU_MASK_NONE; +cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE; struct x8664_pda cpu_pda[NR_CPUS] __cacheline_aligned; @@ -171,7 +171,7 @@ void syscall_init(void) wrmsrl(MSR_SYSCALL_MASK, EF_TF|EF_DF|EF_IE|0x3000); } -void __init check_efer(void) +void __cpuinit check_efer(void) { unsigned long efer; @@ -188,7 +188,7 @@ void __init check_efer(void) * 'CPU state barrier', nothing should get across. * A lot of state is already set up in PDA init. */ -void __init cpu_init (void) +void __cpuinit cpu_init (void) { #ifdef CONFIG_SMP int cpu = stack_smp_processor_id(); diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index f1ec0f345941..bc98a6722cba 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -58,11 +58,6 @@ #include #include -/* Change for real CPU hotplug. Note other files need to be fixed - first too. */ -#define __cpuinit __init -#define __cpuinitdata __initdata - /* Number of siblings per CPU package */ int smp_num_siblings = 1; /* Package ID of each logical CPU */ @@ -823,7 +818,7 @@ static __cpuinit void smp_cleanup_boot(void) * * RED-PEN audit/test this more. I bet there is more state messed up here. */ -static __cpuinit void disable_smp(void) +static __init void disable_smp(void) { cpu_present_map = cpumask_of_cpu(0); cpu_possible_map = cpumask_of_cpu(0); @@ -838,7 +833,7 @@ static __cpuinit void disable_smp(void) /* * Handle user cpus=... parameter. */ -static __cpuinit void enforce_max_cpus(unsigned max_cpus) +static __init void enforce_max_cpus(unsigned max_cpus) { int i, k; k = 0; @@ -855,7 +850,7 @@ static __cpuinit void enforce_max_cpus(unsigned max_cpus) /* * Various sanity checks. */ -static int __cpuinit smp_sanity_check(unsigned max_cpus) +static int __init smp_sanity_check(unsigned max_cpus) { if (!physid_isset(hard_smp_processor_id(), phys_cpu_present_map)) { printk("weird, boot CPU (#%d) not listed by the BIOS.\n", @@ -913,7 +908,7 @@ static int __cpuinit smp_sanity_check(unsigned max_cpus) * Prepare for SMP bootup. The MP table or ACPI has been read * earlier. Just do some sanity checking here and enable APIC mode. */ -void __cpuinit smp_prepare_cpus(unsigned int max_cpus) +void __init smp_prepare_cpus(unsigned int max_cpus) { int i; @@ -1019,7 +1014,7 @@ int __cpuinit __cpu_up(unsigned int cpu) /* * Finish the SMP boot. */ -void __cpuinit smp_cpus_done(unsigned int max_cpus) +void __init smp_cpus_done(unsigned int max_cpus) { zap_low_mappings(); smp_cleanup_boot(); diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c index 84cde796ecb1..ac61c186eb02 100644 --- a/arch/x86_64/mm/numa.c +++ b/arch/x86_64/mm/numa.c @@ -251,7 +251,7 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn) setup_node_bootmem(0, start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT); } -__init void numa_add_cpu(int cpu) +__cpuinit void numa_add_cpu(int cpu) { /* BP is initialized elsewhere */ if (cpu) diff --git a/include/linux/init.h b/include/linux/init.h index 05c83e0521ca..59008c3826cf 100644 --- a/include/linux/init.h +++ b/include/linux/init.h @@ -229,6 +229,18 @@ void __init parse_early_param(void); #define __devexitdata __exitdata #endif +#ifdef CONFIG_HOTPLUG_CPU +#define __cpuinit +#define __cpuinitdata +#define __cpuexit +#define __cpuexitdata +#else +#define __cpuinit __init +#define __cpuinitdata __initdata +#define __cpuexit __exit +#define __cpuexitdata __exitdata +#endif + /* Functions marked as __devexit may be discarded at kernel link time, depending on config options. Newer versions of binutils detect references from retained sections to discarded sections and flag an error. Pointers to -- cgit v1.2.3 From 5a72e04df5470df0ec646029d31e5528167ab1a7 Mon Sep 17 00:00:00 2001 From: Li Shaohua Date: Sat, 25 Jun 2005 14:55:06 -0700 Subject: [PATCH] suspend/resume SMP support Using CPU hotplug to support suspend/resume SMP. Both S3 and S4 use disable/enable_nonboot_cpus API. The S4 part is based on Pavel's original S4 SMP patch. Signed-off-by: Li Shaohua Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/cpu/mcheck/k7.c | 2 +- arch/i386/kernel/cpu/mcheck/mce.c | 2 +- arch/i386/kernel/cpu/mcheck/p4.c | 4 +- arch/i386/kernel/cpu/mcheck/p6.c | 2 +- arch/i386/kernel/cpu/mcheck/winchip.c | 2 +- drivers/acpi/Kconfig | 2 +- include/linux/suspend.h | 2 +- kernel/power/Kconfig | 6 ++- kernel/power/Makefile | 6 +-- kernel/power/disk.c | 35 +++++++------- kernel/power/main.c | 16 ++++--- kernel/power/smp.c | 89 +++++++++++++---------------------- kernel/power/swsusp.c | 2 + 13 files changed, 80 insertions(+), 90 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/cpu/mcheck/k7.c b/arch/i386/kernel/cpu/mcheck/k7.c index 8df52e86c4d2..c4abe7657397 100644 --- a/arch/i386/kernel/cpu/mcheck/k7.c +++ b/arch/i386/kernel/cpu/mcheck/k7.c @@ -69,7 +69,7 @@ static fastcall void k7_machine_check(struct pt_regs * regs, long error_code) /* AMD K7 machine check is Intel like */ -void __init amd_mcheck_init(struct cpuinfo_x86 *c) +void __devinit amd_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; int i; diff --git a/arch/i386/kernel/cpu/mcheck/mce.c b/arch/i386/kernel/cpu/mcheck/mce.c index 7218a7341fbc..2cf25d2ba0f1 100644 --- a/arch/i386/kernel/cpu/mcheck/mce.c +++ b/arch/i386/kernel/cpu/mcheck/mce.c @@ -16,7 +16,7 @@ #include "mce.h" -int mce_disabled __initdata = 0; +int mce_disabled __devinitdata = 0; int nr_mce_banks; EXPORT_SYMBOL_GPL(nr_mce_banks); /* non-fatal.o */ diff --git a/arch/i386/kernel/cpu/mcheck/p4.c b/arch/i386/kernel/cpu/mcheck/p4.c index 8b16ceb929b4..0abccb6fdf9e 100644 --- a/arch/i386/kernel/cpu/mcheck/p4.c +++ b/arch/i386/kernel/cpu/mcheck/p4.c @@ -78,7 +78,7 @@ fastcall void smp_thermal_interrupt(struct pt_regs *regs) } /* P4/Xeon Thermal regulation detect and init */ -static void __init intel_init_thermal(struct cpuinfo_x86 *c) +static void __devinit intel_init_thermal(struct cpuinfo_x86 *c) { u32 l, h; unsigned int cpu = smp_processor_id(); @@ -232,7 +232,7 @@ static fastcall void intel_machine_check(struct pt_regs * regs, long error_code) } -void __init intel_p4_mcheck_init(struct cpuinfo_x86 *c) +void __devinit intel_p4_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; int i; diff --git a/arch/i386/kernel/cpu/mcheck/p6.c b/arch/i386/kernel/cpu/mcheck/p6.c index 46640f8c2494..f01b73f947e1 100644 --- a/arch/i386/kernel/cpu/mcheck/p6.c +++ b/arch/i386/kernel/cpu/mcheck/p6.c @@ -80,7 +80,7 @@ static fastcall void intel_machine_check(struct pt_regs * regs, long error_code) } /* Set up machine check reporting for processors with Intel style MCE */ -void __init intel_p6_mcheck_init(struct cpuinfo_x86 *c) +void __devinit intel_p6_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; int i; diff --git a/arch/i386/kernel/cpu/mcheck/winchip.c b/arch/i386/kernel/cpu/mcheck/winchip.c index 753fa7acb984..7bae68fa168f 100644 --- a/arch/i386/kernel/cpu/mcheck/winchip.c +++ b/arch/i386/kernel/cpu/mcheck/winchip.c @@ -23,7 +23,7 @@ static fastcall void winchip_machine_check(struct pt_regs * regs, long error_cod } /* Set up machine check reporting on the Winchip C6 series */ -void __init winchip_mcheck_init(struct cpuinfo_x86 *c) +void __devinit winchip_mcheck_init(struct cpuinfo_x86 *c) { u32 lo, hi; machine_check_vector = winchip_machine_check; diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig index 670fdb5142d1..86c52520ed34 100644 --- a/drivers/acpi/Kconfig +++ b/drivers/acpi/Kconfig @@ -55,7 +55,7 @@ if ACPI_INTERPRETER config ACPI_SLEEP bool "Sleep States (EXPERIMENTAL)" - depends on X86 + depends on X86 && (!SMP || SUSPEND_SMP) depends on EXPERIMENTAL && PM default y ---help--- diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 2bf0d5fabcdb..f2e96fdfaae0 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -58,7 +58,7 @@ static inline int software_suspend(void) } #endif -#ifdef CONFIG_SMP +#ifdef CONFIG_SUSPEND_SMP extern void disable_nonboot_cpus(void); extern void enable_nonboot_cpus(void); #else diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 696387ffe49c..fdb377636505 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -28,7 +28,7 @@ config PM_DEBUG config SOFTWARE_SUSPEND bool "Software Suspend (EXPERIMENTAL)" - depends on EXPERIMENTAL && PM && SWAP + depends on EXPERIMENTAL && PM && SWAP && (SUSPEND_SMP || !SMP) ---help--- Enable the possibility of suspending the machine. It doesn't need APM. @@ -72,3 +72,7 @@ config PM_STD_PARTITION suspended image to. It will simply pick the first available swap device. +config SUSPEND_SMP + bool + depends on HOTPLUG_CPU && X86 && PM + default y diff --git a/kernel/power/Makefile b/kernel/power/Makefile index fbdc634135a7..2f438d0eaa13 100644 --- a/kernel/power/Makefile +++ b/kernel/power/Makefile @@ -3,9 +3,9 @@ ifeq ($(CONFIG_PM_DEBUG),y) EXTRA_CFLAGS += -DDEBUG endif -swsusp-smp-$(CONFIG_SMP) += smp.o - obj-y := main.o process.o console.o pm.o -obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o $(swsusp-smp-y) disk.o +obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o disk.o + +obj-$(CONFIG_SUSPEND_SMP) += smp.o obj-$(CONFIG_MAGIC_SYSRQ) += poweroff.o diff --git a/kernel/power/disk.c b/kernel/power/disk.c index 02b6764034dc..fb8de63c2919 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -117,8 +117,8 @@ static void finish(void) { device_resume(); platform_finish(); - enable_nonboot_cpus(); thaw_processes(); + enable_nonboot_cpus(); pm_restore_console(); } @@ -131,28 +131,35 @@ static int prepare_processes(void) sys_sync(); + disable_nonboot_cpus(); + if (freeze_processes()) { error = -EBUSY; - return error; + goto thaw; } if (pm_disk_mode == PM_DISK_PLATFORM) { if (pm_ops && pm_ops->prepare) { if ((error = pm_ops->prepare(PM_SUSPEND_DISK))) - return error; + goto thaw; } } /* Free memory before shutting down devices. */ free_some_memory(); - return 0; +thaw: + thaw_processes(); + enable_nonboot_cpus(); + pm_restore_console(); + return error; } static void unprepare_processes(void) { - enable_nonboot_cpus(); + platform_finish(); thaw_processes(); + enable_nonboot_cpus(); pm_restore_console(); } @@ -160,15 +167,9 @@ static int prepare_devices(void) { int error; - disable_nonboot_cpus(); - if ((error = device_suspend(PMSG_FREEZE))) { + if ((error = device_suspend(PMSG_FREEZE))) printk("Some devices failed to suspend\n"); - platform_finish(); - enable_nonboot_cpus(); - return error; - } - - return 0; + return error; } /** @@ -185,9 +186,9 @@ int pm_suspend_disk(void) int error; error = prepare_processes(); - if (!error) { - error = prepare_devices(); - } + if (error) + return error; + error = prepare_devices(); if (error) { unprepare_processes(); @@ -250,7 +251,7 @@ static int software_resume(void) if ((error = prepare_processes())) { swsusp_close(); - goto Cleanup; + goto Done; } pr_debug("PM: Reading swsusp image.\n"); diff --git a/kernel/power/main.c b/kernel/power/main.c index 4cdebc972ff2..c94cb9e95090 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -55,6 +55,13 @@ static int suspend_prepare(suspend_state_t state) pm_prepare_console(); + disable_nonboot_cpus(); + + if (num_online_cpus() != 1) { + error = -EPERM; + goto Enable_cpu; + } + if (freeze_processes()) { error = -EAGAIN; goto Thaw; @@ -75,6 +82,8 @@ static int suspend_prepare(suspend_state_t state) pm_ops->finish(state); Thaw: thaw_processes(); + Enable_cpu: + enable_nonboot_cpus(); pm_restore_console(); return error; } @@ -113,6 +122,7 @@ static void suspend_finish(suspend_state_t state) if (pm_ops && pm_ops->finish) pm_ops->finish(state); thaw_processes(); + enable_nonboot_cpus(); pm_restore_console(); } @@ -150,12 +160,6 @@ static int enter_state(suspend_state_t state) goto Unlock; } - /* Suspend is hard to get right on SMP. */ - if (num_online_cpus() != 1) { - error = -EPERM; - goto Unlock; - } - pr_debug("PM: Preparing system for %s sleep\n", pm_states[state]); if ((error = suspend_prepare(state))) goto Unlock; diff --git a/kernel/power/smp.c b/kernel/power/smp.c index 457c2302ed42..bbe23079c62c 100644 --- a/kernel/power/smp.c +++ b/kernel/power/smp.c @@ -13,73 +13,52 @@ #include #include #include +#include #include #include -static atomic_t cpu_counter, freeze; - - -static void smp_pause(void * data) -{ - struct saved_context ctxt; - __save_processor_state(&ctxt); - printk("Sleeping in:\n"); - dump_stack(); - atomic_inc(&cpu_counter); - while (atomic_read(&freeze)) { - /* FIXME: restore takes place at random piece inside this. - This should probably be written in assembly, and - preserve general-purpose registers, too - - What about stack? We may need to move to new stack here. - - This should better be ran with interrupts disabled. - */ - cpu_relax(); - barrier(); - } - atomic_dec(&cpu_counter); - __restore_processor_state(&ctxt); -} - -static cpumask_t oldmask; +/* This is protected by pm_sem semaphore */ +static cpumask_t frozen_cpus; void disable_nonboot_cpus(void) { - oldmask = current->cpus_allowed; - set_cpus_allowed(current, cpumask_of_cpu(0)); - printk("Freezing CPUs (at %d)", raw_smp_processor_id()); - current->state = TASK_INTERRUPTIBLE; - schedule_timeout(HZ); - printk("..."); - BUG_ON(raw_smp_processor_id() != 0); - - /* FIXME: for this to work, all the CPUs must be running - * "idle" thread (or we deadlock). Is that guaranteed? */ + int cpu, error; - atomic_set(&cpu_counter, 0); - atomic_set(&freeze, 1); - smp_call_function(smp_pause, NULL, 0, 0); - while (atomic_read(&cpu_counter) < (num_online_cpus() - 1)) { - cpu_relax(); - barrier(); + error = 0; + cpus_clear(frozen_cpus); + printk("Freezing cpus ...\n"); + for_each_online_cpu(cpu) { + if (cpu == 0) + continue; + error = cpu_down(cpu); + if (!error) { + cpu_set(cpu, frozen_cpus); + printk("CPU%d is down\n", cpu); + continue; + } + printk("Error taking cpu %d down: %d\n", cpu, error); } - printk("ok\n"); + BUG_ON(smp_processor_id() != 0); + if (error) + panic("cpus not sleeping"); } void enable_nonboot_cpus(void) { - printk("Restarting CPUs"); - atomic_set(&freeze, 0); - while (atomic_read(&cpu_counter)) { - cpu_relax(); - barrier(); - } - printk("..."); - set_cpus_allowed(current, oldmask); - schedule(); - printk("ok\n"); + int cpu, error; + printk("Thawing cpus ...\n"); + for_each_cpu_mask(cpu, frozen_cpus) { + error = smp_prepare_cpu(cpu); + if (!error) + error = cpu_up(cpu); + if (!error) { + printk("CPU%d is up\n", cpu); + continue; + } + printk("Error taking cpu %d up: %d\n", cpu, error); + panic("Not enough cpus"); + } + cpus_clear(frozen_cpus); } - diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index 53f9f8720ee4..339b5c3735bd 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -1193,8 +1193,10 @@ static const char * sanity_check(void) return "version"; if (strcmp(swsusp_info.uts.machine,system_utsname.machine)) return "machine"; +#if 0 if(swsusp_info.cpus != num_online_cpus()) return "number of cpus"; +#endif return NULL; } -- cgit v1.2.3 From 620b03276488c3cf103caf1e326bd21f00d3df84 Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Sat, 25 Jun 2005 14:55:11 -0700 Subject: [PATCH] properly stop devices before poweroff Without this patch, Linux provokes emergency disk shutdowns and similar nastiness. It was in SuSE kernels for some time, IIRC. Signed-off-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pm.h | 33 +++++++++++++++++++++------------ kernel/sys.c | 3 +++ 2 files changed, 24 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pm.h b/include/linux/pm.h index ed2b76e75199..14479325e3f3 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -103,7 +103,8 @@ extern int pm_active; /* * Register a device with power management */ -struct pm_dev __deprecated *pm_register(pm_dev_t type, unsigned long id, pm_callback callback); +struct pm_dev __deprecated * +pm_register(pm_dev_t type, unsigned long id, pm_callback callback); /* * Unregister a device with power management @@ -190,17 +191,18 @@ typedef u32 __bitwise pm_message_t; /* * There are 4 important states driver can be in: * ON -- driver is working - * FREEZE -- stop operations and apply whatever policy is applicable to a suspended driver - * of that class, freeze queues for block like IDE does, drop packets for - * ethernet, etc... stop DMA engine too etc... so a consistent image can be - * saved; but do not power any hardware down. - * SUSPEND - like FREEZE, but hardware is doing as much powersaving as possible. Roughly - * pci D3. + * FREEZE -- stop operations and apply whatever policy is applicable to a + * suspended driver of that class, freeze queues for block like IDE + * does, drop packets for ethernet, etc... stop DMA engine too etc... + * so a consistent image can be saved; but do not power any hardware + * down. + * SUSPEND - like FREEZE, but hardware is doing as much powersaving as + * possible. Roughly pci D3. * - * Unfortunately, current drivers only recognize numeric values 0 (ON) and 3 (SUSPEND). - * We'll need to fix the drivers. So yes, putting 3 to all diferent defines is intentional, - * and will go away as soon as drivers are fixed. Also note that typedef is neccessary, - * we'll probably want to switch to + * Unfortunately, current drivers only recognize numeric values 0 (ON) and 3 + * (SUSPEND). We'll need to fix the drivers. So yes, putting 3 to all different + * defines is intentional, and will go away as soon as drivers are fixed. Also + * note that typedef is neccessary, we'll probably want to switch to * typedef struct pm_message_t { int event; int flags; } pm_message_t * or something similar soon. */ @@ -222,11 +224,18 @@ struct dev_pm_info { extern void device_pm_set_parent(struct device * dev, struct device * parent); -extern int device_suspend(pm_message_t state); extern int device_power_down(pm_message_t state); extern void device_power_up(void); extern void device_resume(void); +#ifdef CONFIG_PM +extern int device_suspend(pm_message_t state); +#else +static inline int device_suspend(pm_message_t state) +{ + return 0; +} +#endif #endif /* __KERNEL__ */ diff --git a/kernel/sys.c b/kernel/sys.c index da24bc1292db..dac10161ca23 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -405,6 +405,7 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user case LINUX_REBOOT_CMD_HALT: notifier_call_chain(&reboot_notifier_list, SYS_HALT, NULL); system_state = SYSTEM_HALT; + device_suspend(PMSG_SUSPEND); device_shutdown(); printk(KERN_EMERG "System halted.\n"); machine_halt(); @@ -415,6 +416,7 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user case LINUX_REBOOT_CMD_POWER_OFF: notifier_call_chain(&reboot_notifier_list, SYS_POWER_OFF, NULL); system_state = SYSTEM_POWER_OFF; + device_suspend(PMSG_SUSPEND); device_shutdown(); printk(KERN_EMERG "Power down.\n"); machine_power_off(); @@ -431,6 +433,7 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user notifier_call_chain(&reboot_notifier_list, SYS_RESTART, buffer); system_state = SYSTEM_RESTART; + device_suspend(PMSG_FREEZE); device_shutdown(); printk(KERN_EMERG "Restarting system with command '%s'.\n", buffer); machine_restart(buffer); -- cgit v1.2.3 From b2b18660066997420b716c1881a6be8b82700d97 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 25 Jun 2005 14:55:38 -0700 Subject: [PATCH] RCU: clean up a few remaining synchronize_kernel() calls 2.6.12-rc6-mm1 has a few remaining synchronize_kernel()s, some (but not all) in comments. This patch changes these synchronize_kernel() calls (and comments) to synchronize_rcu() or synchronize_sched() as follows: - arch/x86_64/kernel/mce.c mce_read(): change to synchronize_sched() to handle races with machine-check exceptions (synchronize_rcu() would not cut it given RCU implementations intended for hardcore realtime use. - drivers/input/serio/i8042.c i8042_stop(): change to synchronize_sched() to handle races with i8042_interrupt() interrupt handler. Again, synchronize_rcu() would not cut it given RCU implementations intended for hardcore realtime use. - include/*/kdebug.h comments: change to synchronize_sched() to handle races with NMIs. As before, synchronize_rcu() would not cut it... - include/linux/list.h comment: change to synchronize_rcu(), since this comment is for list_del_rcu(). - security/keys/key.c unregister_key_type(): change to synchronize_rcu(), since this is interacting with RCU read side. - security/keys/process_keys.c install_session_keyring(): change to synchronize_rcu(), since this is interacting with RCU read side. Signed-off-by: "Paul E. McKenney" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/mce.c | 2 +- drivers/input/serio/i8042.c | 2 +- include/asm-i386/kdebug.h | 2 +- include/asm-ppc64/kdebug.h | 2 +- include/asm-sparc64/kdebug.h | 2 +- include/asm-x86_64/kdebug.h | 2 +- include/linux/list.h | 2 +- security/keys/key.c | 2 +- security/keys/process_keys.c | 2 +- 9 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index 7ab15c8ab95f..21e70625a495 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -411,7 +411,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, loff memset(mcelog.entry, 0, next * sizeof(struct mce)); mcelog.next = 0; - synchronize_kernel(); + synchronize_sched(); /* Collect entries that were still getting written before the synchronize. */ diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c index 5900de3c3f4f..a9bf549c8dc5 100644 --- a/drivers/input/serio/i8042.c +++ b/drivers/input/serio/i8042.c @@ -396,7 +396,7 @@ static void i8042_stop(struct serio *serio) struct i8042_port *port = serio->port_data; port->exists = 0; - synchronize_kernel(); + synchronize_sched(); port->serio = NULL; } diff --git a/include/asm-i386/kdebug.h b/include/asm-i386/kdebug.h index de6498b0d493..b3f8d5f59d5d 100644 --- a/include/asm-i386/kdebug.h +++ b/include/asm-i386/kdebug.h @@ -18,7 +18,7 @@ struct die_args { }; /* Note - you should never unregister because that can race with NMIs. - If you really want to do it first unregister - then synchronize_kernel - then free. + If you really want to do it first unregister - then synchronize_sched - then free. */ int register_die_notifier(struct notifier_block *nb); extern struct notifier_block *i386die_chain; diff --git a/include/asm-ppc64/kdebug.h b/include/asm-ppc64/kdebug.h index 488634258a72..d383d161cf8d 100644 --- a/include/asm-ppc64/kdebug.h +++ b/include/asm-ppc64/kdebug.h @@ -17,7 +17,7 @@ struct die_args { /* Note - you should never unregister because that can race with NMIs. - If you really want to do it first unregister - then synchronize_kernel - + If you really want to do it first unregister - then synchronize_sched - then free. */ int register_die_notifier(struct notifier_block *nb); diff --git a/include/asm-sparc64/kdebug.h b/include/asm-sparc64/kdebug.h index f70d3dad01f9..6321f5a0198d 100644 --- a/include/asm-sparc64/kdebug.h +++ b/include/asm-sparc64/kdebug.h @@ -16,7 +16,7 @@ struct die_args { }; /* Note - you should never unregister because that can race with NMIs. - * If you really want to do it first unregister - then synchronize_kernel + * If you really want to do it first unregister - then synchronize_sched * - then free. */ int register_die_notifier(struct notifier_block *nb); diff --git a/include/asm-x86_64/kdebug.h b/include/asm-x86_64/kdebug.h index 6277f75cbb4b..b90341994d80 100644 --- a/include/asm-x86_64/kdebug.h +++ b/include/asm-x86_64/kdebug.h @@ -14,7 +14,7 @@ struct die_args { }; /* Note - you should never unregister because that can race with NMIs. - If you really want to do it first unregister - then synchronize_kernel - then free. + If you really want to do it first unregister - then synchronize_sched - then free. */ int register_die_notifier(struct notifier_block *nb); extern struct notifier_block *die_chain; diff --git a/include/linux/list.h b/include/linux/list.h index 399b51d17218..aab2db21b013 100644 --- a/include/linux/list.h +++ b/include/linux/list.h @@ -185,7 +185,7 @@ static inline void list_del(struct list_head *entry) * list_for_each_entry_rcu(). * * Note that the caller is not permitted to immediately free - * the newly deleted entry. Instead, either synchronize_kernel() + * the newly deleted entry. Instead, either synchronize_rcu() * or call_rcu() must be used to defer freeing until an RCU * grace period has elapsed. */ diff --git a/security/keys/key.c b/security/keys/key.c index 3304d37bb379..fb89f9844465 100644 --- a/security/keys/key.c +++ b/security/keys/key.c @@ -980,7 +980,7 @@ void unregister_key_type(struct key_type *ktype) spin_unlock(&key_serial_lock); /* make sure everyone revalidates their keys */ - synchronize_kernel(); + synchronize_rcu(); /* we should now be able to destroy the payloads of all the keys of * this type with impunity */ diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c index 34db087bbcc7..9b0369c5a223 100644 --- a/security/keys/process_keys.c +++ b/security/keys/process_keys.c @@ -234,7 +234,7 @@ static int install_session_keyring(struct task_struct *tsk, ret = 0; /* we're using RCU on the pointer */ - synchronize_kernel(); + synchronize_rcu(); key_put(old); error: return ret; -- cgit v1.2.3 From 7897986bad8f6cd50d6149345aca7f6480f49464 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:13 -0700 Subject: [PATCH] sched: balance timers Do CPU load averaging over a number of different intervals. Allow each interval to be chosen by sending a parameter to source_load and target_load. 0 is instantaneous, idx > 0 returns a decaying average with the most recent sample weighted at 2^(idx-1). To a maximum of 3 (could be easily increased). So generally a higher number will result in more conservative balancing. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-i386/topology.h | 4 ++ include/asm-x86_64/topology.h | 6 +- include/linux/sched.h | 4 ++ include/linux/topology.h | 8 +++ kernel/sched.c | 138 ++++++++++++++++++++++-------------------- 5 files changed, 95 insertions(+), 65 deletions(-) (limited to 'include/linux') diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h index 6d0f67507b21..0055fbfeec7b 100644 --- a/include/asm-i386/topology.h +++ b/include/asm-i386/topology.h @@ -74,6 +74,10 @@ static inline int node_to_first_cpu(int node) .imbalance_pct = 125, \ .cache_hot_time = (10*1000000), \ .cache_nice_tries = 1, \ + .busy_idx = 3, \ + .idle_idx = 1, \ + .newidle_idx = 2, \ + .wake_idx = 1, \ .per_cpu_gain = 100, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_EXEC \ diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h index 8f77e9f6bc23..fe8d80a15751 100644 --- a/include/asm-x86_64/topology.h +++ b/include/asm-x86_64/topology.h @@ -39,7 +39,11 @@ extern int __node_distance(int, int); .busy_factor = 32, \ .imbalance_pct = 125, \ .cache_hot_time = (10*1000000), \ - .cache_nice_tries = 1, \ + .cache_nice_tries = 2, \ + .busy_idx = 3, \ + .idle_idx = 2, \ + .newidle_idx = 1, \ + .wake_idx = 1, \ .per_cpu_gain = 100, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ diff --git a/include/linux/sched.h b/include/linux/sched.h index 2c69682b0444..664981ac1fb6 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -488,6 +488,10 @@ struct sched_domain { unsigned long long cache_hot_time; /* Task considered cache hot (ns) */ unsigned int cache_nice_tries; /* Leave cache hot tasks for # tries */ unsigned int per_cpu_gain; /* CPU % gained by adding domain cpus */ + unsigned int busy_idx; + unsigned int idle_idx; + unsigned int newidle_idx; + unsigned int wake_idx; int flags; /* See SD_* */ /* Runtime fields. */ diff --git a/include/linux/topology.h b/include/linux/topology.h index d70e8972c67f..ae9c2216dfa6 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -89,6 +89,10 @@ .cache_hot_time = 0, \ .cache_nice_tries = 0, \ .per_cpu_gain = 25, \ + .busy_idx = 0, \ + .idle_idx = 0, \ + .newidle_idx = 0, \ + .wake_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ @@ -115,6 +119,10 @@ .cache_hot_time = (5*1000000/2), \ .cache_nice_tries = 1, \ .per_cpu_gain = 100, \ + .busy_idx = 2, \ + .idle_idx = 0, \ + .newidle_idx = 1, \ + .wake_idx = 1, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ diff --git a/kernel/sched.c b/kernel/sched.c index f665de34ed82..b597b07e7911 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -206,7 +206,7 @@ struct runqueue { */ unsigned long nr_running; #ifdef CONFIG_SMP - unsigned long cpu_load; + unsigned long cpu_load[3]; #endif unsigned long long nr_switches; @@ -886,23 +886,27 @@ void kick_process(task_t *p) * We want to under-estimate the load of migration sources, to * balance conservatively. */ -static inline unsigned long source_load(int cpu) +static inline unsigned long source_load(int cpu, int type) { runqueue_t *rq = cpu_rq(cpu); unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE; + if (type == 0) + return load_now; - return min(rq->cpu_load, load_now); + return min(rq->cpu_load[type-1], load_now); } /* * Return a high guess at the load of a migration-target cpu */ -static inline unsigned long target_load(int cpu) +static inline unsigned long target_load(int cpu, int type) { runqueue_t *rq = cpu_rq(cpu); unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE; + if (type == 0) + return load_now; - return max(rq->cpu_load, load_now); + return max(rq->cpu_load[type-1], load_now); } #endif @@ -967,7 +971,7 @@ static int try_to_wake_up(task_t * p, unsigned int state, int sync) runqueue_t *rq; #ifdef CONFIG_SMP unsigned long load, this_load; - struct sched_domain *sd; + struct sched_domain *sd, *this_sd = NULL; int new_cpu; #endif @@ -986,72 +990,64 @@ static int try_to_wake_up(task_t * p, unsigned int state, int sync) if (unlikely(task_running(rq, p))) goto out_activate; -#ifdef CONFIG_SCHEDSTATS + new_cpu = cpu; + schedstat_inc(rq, ttwu_cnt); if (cpu == this_cpu) { schedstat_inc(rq, ttwu_local); - } else { - for_each_domain(this_cpu, sd) { - if (cpu_isset(cpu, sd->span)) { - schedstat_inc(sd, ttwu_wake_remote); - break; - } + goto out_set_cpu; + } + + for_each_domain(this_cpu, sd) { + if (cpu_isset(cpu, sd->span)) { + schedstat_inc(sd, ttwu_wake_remote); + this_sd = sd; + break; } } -#endif - new_cpu = cpu; - if (cpu == this_cpu || unlikely(!cpu_isset(this_cpu, p->cpus_allowed))) + if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed))) goto out_set_cpu; - load = source_load(cpu); - this_load = target_load(this_cpu); - /* - * If sync wakeup then subtract the (maximum possible) effect of - * the currently running task from the load of the current CPU: + * Check for affine wakeup and passive balancing possibilities. */ - if (sync) - this_load -= SCHED_LOAD_SCALE; - - /* Don't pull the task off an idle CPU to a busy one */ - if (load < SCHED_LOAD_SCALE/2 && this_load > SCHED_LOAD_SCALE/2) - goto out_set_cpu; + if (this_sd) { + int idx = this_sd->wake_idx; + unsigned int imbalance; - new_cpu = this_cpu; /* Wake to this CPU if we can */ + load = source_load(cpu, idx); + this_load = target_load(this_cpu, idx); - /* - * Scan domains for affine wakeup and passive balancing - * possibilities. - */ - for_each_domain(this_cpu, sd) { - unsigned int imbalance; /* - * Start passive balancing when half the imbalance_pct - * limit is reached. + * If sync wakeup then subtract the (maximum possible) effect of + * the currently running task from the load of the current CPU: */ - imbalance = sd->imbalance_pct + (sd->imbalance_pct - 100) / 2; + if (sync) + this_load -= SCHED_LOAD_SCALE; + + /* Don't pull the task off an idle CPU to a busy one */ + if (load < SCHED_LOAD_SCALE/2 && this_load > SCHED_LOAD_SCALE/2) + goto out_set_cpu; - if ((sd->flags & SD_WAKE_AFFINE) && - !task_hot(p, rq->timestamp_last_tick, sd)) { + new_cpu = this_cpu; /* Wake to this CPU if we can */ + + if ((this_sd->flags & SD_WAKE_AFFINE) && + !task_hot(p, rq->timestamp_last_tick, this_sd)) { /* * This domain has SD_WAKE_AFFINE and p is cache cold * in this domain. */ - if (cpu_isset(cpu, sd->span)) { - schedstat_inc(sd, ttwu_move_affine); - goto out_set_cpu; - } - } else if ((sd->flags & SD_WAKE_BALANCE) && + schedstat_inc(this_sd, ttwu_move_affine); + goto out_set_cpu; + } else if ((this_sd->flags & SD_WAKE_BALANCE) && imbalance*this_load <= 100*load) { /* * This domain has SD_WAKE_BALANCE and there is * an imbalance. */ - if (cpu_isset(cpu, sd->span)) { - schedstat_inc(sd, ttwu_move_balance); - goto out_set_cpu; - } + schedstat_inc(this_sd, ttwu_move_balance); + goto out_set_cpu; } } @@ -1509,7 +1505,7 @@ static int find_idlest_cpu(struct task_struct *p, int this_cpu, cpus_and(mask, sd->span, p->cpus_allowed); for_each_cpu_mask(i, mask) { - load = target_load(i); + load = target_load(i, sd->wake_idx); if (load < min_load) { min_cpu = i; @@ -1522,7 +1518,7 @@ static int find_idlest_cpu(struct task_struct *p, int this_cpu, } /* add +1 to account for the new task */ - this_load = source_load(this_cpu) + SCHED_LOAD_SCALE; + this_load = source_load(this_cpu, sd->wake_idx) + SCHED_LOAD_SCALE; /* * Would with the addition of the new task to the @@ -1767,8 +1763,15 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, { struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups; unsigned long max_load, avg_load, total_load, this_load, total_pwr; + int load_idx; max_load = this_load = total_load = total_pwr = 0; + if (idle == NOT_IDLE) + load_idx = sd->busy_idx; + else if (idle == NEWLY_IDLE) + load_idx = sd->newidle_idx; + else + load_idx = sd->idle_idx; do { unsigned long load; @@ -1783,9 +1786,9 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, for_each_cpu_mask(i, group->cpumask) { /* Bias balancing toward cpus of our domain */ if (local_group) - load = target_load(i); + load = target_load(i, load_idx); else - load = source_load(i); + load = source_load(i, load_idx); avg_load += load; } @@ -1895,7 +1898,7 @@ static runqueue_t *find_busiest_queue(struct sched_group *group) int i; for_each_cpu_mask(i, group->cpumask) { - load = source_load(i); + load = source_load(i, 0); if (load > max_load) { max_load = load; @@ -2150,18 +2153,23 @@ static void rebalance_tick(int this_cpu, runqueue_t *this_rq, unsigned long old_load, this_load; unsigned long j = jiffies + CPU_OFFSET(this_cpu); struct sched_domain *sd; + int i; - /* Update our load */ - old_load = this_rq->cpu_load; this_load = this_rq->nr_running * SCHED_LOAD_SCALE; - /* - * Round up the averaging division if load is increasing. This - * prevents us from getting stuck on 9 if the load is 10, for - * example. - */ - if (this_load > old_load) - old_load++; - this_rq->cpu_load = (old_load + this_load) / 2; + /* Update our load */ + for (i = 0; i < 3; i++) { + unsigned long new_load = this_load; + int scale = 1 << i; + old_load = this_rq->cpu_load[i]; + /* + * Round up the averaging division if load is increasing. This + * prevents us from getting stuck on 9 if the load is 10, for + * example. + */ + if (new_load > old_load) + new_load += scale-1; + this_rq->cpu_load[i] = (old_load*(scale-1) + new_load) / scale; + } for_each_domain(this_cpu, sd) { unsigned long interval; @@ -4921,13 +4929,15 @@ void __init sched_init(void) rq = cpu_rq(i); spin_lock_init(&rq->lock); + rq->nr_running = 0; rq->active = rq->arrays; rq->expired = rq->arrays + 1; rq->best_expired_prio = MAX_PRIO; #ifdef CONFIG_SMP rq->sd = &sched_domain_dummy; - rq->cpu_load = 0; + for (j = 1; j < 3; j++) + rq->cpu_load[j] = 0; rq->active_balance = 0; rq->push_cpu = 0; rq->migration_thread = NULL; -- cgit v1.2.3 From cafb20c1f9976a70d633bb1e1c8c24eab00e4e80 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:17 -0700 Subject: [PATCH] sched: no aggressive idle balancing Remove the very aggressive idle stuff that has recently gone into 2.6 - it is going against the direction we are trying to go. Hopefully we can regain performance through other methods. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-i386/topology.h | 1 - include/asm-x86_64/topology.h | 1 - include/linux/topology.h | 1 - kernel/sched.c | 21 ++------------------- 4 files changed, 2 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h index 0055fbfeec7b..5eb6f61dcefc 100644 --- a/include/asm-i386/topology.h +++ b/include/asm-i386/topology.h @@ -82,7 +82,6 @@ static inline int node_to_first_cpu(int node) .flags = SD_LOAD_BALANCE \ | SD_BALANCE_EXEC \ | SD_BALANCE_NEWIDLE \ - | SD_WAKE_IDLE \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 1, \ diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h index fe8d80a15751..9cb7459ce722 100644 --- a/include/asm-x86_64/topology.h +++ b/include/asm-x86_64/topology.h @@ -48,7 +48,6 @@ extern int __node_distance(int, int); .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ - | SD_WAKE_IDLE \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 1, \ diff --git a/include/linux/topology.h b/include/linux/topology.h index ae9c2216dfa6..b23ec64df7f1 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -127,7 +127,6 @@ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ | SD_WAKE_AFFINE \ - | SD_WAKE_IDLE \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 1, \ diff --git a/kernel/sched.c b/kernel/sched.c index 5ae3568eed0b..396724a2519f 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -414,22 +414,6 @@ static inline runqueue_t *this_rq_lock(void) return rq; } -#ifdef CONFIG_SCHED_SMT -static int cpu_and_siblings_are_idle(int cpu) -{ - int sib; - for_each_cpu_mask(sib, cpu_sibling_map[cpu]) { - if (idle_cpu(sib)) - continue; - return 0; - } - - return 1; -} -#else -#define cpu_and_siblings_are_idle(A) idle_cpu(A) -#endif - #ifdef CONFIG_SCHEDSTATS /* * Called when a process is dequeued from the active array and given @@ -1652,12 +1636,11 @@ int can_migrate_task(task_t *p, runqueue_t *rq, int this_cpu, /* * Aggressive migration if: - * 1) the [whole] cpu is idle, or + * 1) task is cache cold, or * 2) too many balance attempts have failed. */ - if (cpu_and_siblings_are_idle(this_cpu) || \ - sd->nr_balance_failed > sd->cache_nice_tries) + if (sd->nr_balance_failed > sd->cache_nice_tries) return 1; if (task_hot(p, rq->timestamp_last_tick, sd)) -- cgit v1.2.3 From 147cbb4bbe991452698f0772d8292f22825710ba Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:19 -0700 Subject: [PATCH] sched: balance on fork Reimplement the balance on exec balancing to be sched-domains aware. Use this to also do balance on fork balancing. Make x86_64 do balance on fork over the NUMA domain. The problem that the non sched domains aware blancing became apparent on dual core, multi socket opterons. What we want is for the new tasks to be sent to a different socket, but more often than not, we would first load up our sibling core, or fill two cores of a single remote socket before selecting a new one. This gives large improvements to STREAM on such systems. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-x86_64/topology.h | 2 + include/linux/sched.h | 10 +-- include/linux/topology.h | 2 + kernel/sched.c | 164 ++++++++++++++++++++++++++++-------------- 4 files changed, 119 insertions(+), 59 deletions(-) (limited to 'include/linux') diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h index 9cb7459ce722..802d09b9c99f 100644 --- a/include/asm-x86_64/topology.h +++ b/include/asm-x86_64/topology.h @@ -44,9 +44,11 @@ extern int __node_distance(int, int); .idle_idx = 2, \ .newidle_idx = 1, \ .wake_idx = 1, \ + .forkexec_idx = 1, \ .per_cpu_gain = 100, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ + | SD_BALANCE_FORK \ | SD_BALANCE_EXEC \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ diff --git a/include/linux/sched.h b/include/linux/sched.h index 664981ac1fb6..613491d3a875 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -460,10 +460,11 @@ enum idle_type #define SD_LOAD_BALANCE 1 /* Do load balancing on this domain. */ #define SD_BALANCE_NEWIDLE 2 /* Balance when about to become idle */ #define SD_BALANCE_EXEC 4 /* Balance on exec */ -#define SD_WAKE_IDLE 8 /* Wake to idle CPU on task wakeup */ -#define SD_WAKE_AFFINE 16 /* Wake task to waking CPU */ -#define SD_WAKE_BALANCE 32 /* Perform balancing at task wakeup */ -#define SD_SHARE_CPUPOWER 64 /* Domain members share cpu power */ +#define SD_BALANCE_FORK 8 /* Balance on fork, clone */ +#define SD_WAKE_IDLE 16 /* Wake to idle CPU on task wakeup */ +#define SD_WAKE_AFFINE 32 /* Wake task to waking CPU */ +#define SD_WAKE_BALANCE 64 /* Perform balancing at task wakeup */ +#define SD_SHARE_CPUPOWER 128 /* Domain members share cpu power */ struct sched_group { struct sched_group *next; /* Must be a circular list */ @@ -492,6 +493,7 @@ struct sched_domain { unsigned int idle_idx; unsigned int newidle_idx; unsigned int wake_idx; + unsigned int forkexec_idx; int flags; /* See SD_* */ /* Runtime fields. */ diff --git a/include/linux/topology.h b/include/linux/topology.h index b23ec64df7f1..665597207def 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -93,6 +93,7 @@ .idle_idx = 0, \ .newidle_idx = 0, \ .wake_idx = 0, \ + .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ @@ -123,6 +124,7 @@ .idle_idx = 0, \ .newidle_idx = 1, \ .wake_idx = 1, \ + .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ diff --git a/kernel/sched.c b/kernel/sched.c index 396724a2519f..7ecc237e2aab 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -893,6 +893,79 @@ static inline unsigned long target_load(int cpu, int type) return max(rq->cpu_load[type-1], load_now); } +/* + * find_idlest_group finds and returns the least busy CPU group within the + * domain. + */ +static struct sched_group * +find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) +{ + struct sched_group *idlest = NULL, *this = NULL, *group = sd->groups; + unsigned long min_load = ULONG_MAX, this_load = 0; + int load_idx = sd->forkexec_idx; + int imbalance = 100 + (sd->imbalance_pct-100)/2; + + do { + unsigned long load, avg_load; + int local_group; + int i; + + local_group = cpu_isset(this_cpu, group->cpumask); + /* XXX: put a cpus allowed check */ + + /* Tally up the load of all CPUs in the group */ + avg_load = 0; + + for_each_cpu_mask(i, group->cpumask) { + /* Bias balancing toward cpus of our domain */ + if (local_group) + load = source_load(i, load_idx); + else + load = target_load(i, load_idx); + + avg_load += load; + } + + /* Adjust by relative CPU power of the group */ + avg_load = (avg_load * SCHED_LOAD_SCALE) / group->cpu_power; + + if (local_group) { + this_load = avg_load; + this = group; + } else if (avg_load < min_load) { + min_load = avg_load; + idlest = group; + } + group = group->next; + } while (group != sd->groups); + + if (!idlest || 100*this_load < imbalance*min_load) + return NULL; + return idlest; +} + +/* + * find_idlest_queue - find the idlest runqueue among the cpus in group. + */ +static int find_idlest_cpu(struct sched_group *group, int this_cpu) +{ + unsigned long load, min_load = ULONG_MAX; + int idlest = -1; + int i; + + for_each_cpu_mask(i, group->cpumask) { + load = source_load(i, 0); + + if (load < min_load || (load == min_load && i == this_cpu)) { + min_load = load; + idlest = i; + } + } + + return idlest; +} + + #endif /* @@ -1107,11 +1180,6 @@ int fastcall wake_up_state(task_t *p, unsigned int state) return try_to_wake_up(p, state, 0); } -#ifdef CONFIG_SMP -static int find_idlest_cpu(struct task_struct *p, int this_cpu, - struct sched_domain *sd); -#endif - /* * Perform scheduler related setup for a newly forked process p. * p is forked by current. @@ -1181,12 +1249,38 @@ void fastcall wake_up_new_task(task_t * p, unsigned long clone_flags) unsigned long flags; int this_cpu, cpu; runqueue_t *rq, *this_rq; +#ifdef CONFIG_SMP + struct sched_domain *tmp, *sd = NULL; +#endif rq = task_rq_lock(p, &flags); - cpu = task_cpu(p); + BUG_ON(p->state != TASK_RUNNING); this_cpu = smp_processor_id(); + cpu = task_cpu(p); - BUG_ON(p->state != TASK_RUNNING); +#ifdef CONFIG_SMP + for_each_domain(cpu, tmp) + if (tmp->flags & SD_BALANCE_FORK) + sd = tmp; + + if (sd) { + struct sched_group *group; + + cpu = task_cpu(p); + group = find_idlest_group(sd, p, cpu); + if (group) { + int new_cpu; + new_cpu = find_idlest_cpu(group, cpu); + if (new_cpu != -1 && new_cpu != cpu && + cpu_isset(new_cpu, p->cpus_allowed)) { + set_task_cpu(p, new_cpu); + task_rq_unlock(rq, &flags); + rq = task_rq_lock(p, &flags); + cpu = task_cpu(p); + } + } + } +#endif /* * We decrease the sleep average of forking parents @@ -1480,51 +1574,6 @@ static void double_lock_balance(runqueue_t *this_rq, runqueue_t *busiest) } } -/* - * find_idlest_cpu - find the least busy runqueue. - */ -static int find_idlest_cpu(struct task_struct *p, int this_cpu, - struct sched_domain *sd) -{ - unsigned long load, min_load, this_load; - int i, min_cpu; - cpumask_t mask; - - min_cpu = UINT_MAX; - min_load = ULONG_MAX; - - cpus_and(mask, sd->span, p->cpus_allowed); - - for_each_cpu_mask(i, mask) { - load = target_load(i, sd->wake_idx); - - if (load < min_load) { - min_cpu = i; - min_load = load; - - /* break out early on an idle CPU: */ - if (!min_load) - break; - } - } - - /* add +1 to account for the new task */ - this_load = source_load(this_cpu, sd->wake_idx) + SCHED_LOAD_SCALE; - - /* - * Would with the addition of the new task to the - * current CPU there be an imbalance between this - * CPU and the idlest CPU? - * - * Use half of the balancing threshold - new-context is - * a good opportunity to balance. - */ - if (min_load*(100 + (sd->imbalance_pct-100)/2) < this_load*100) - return min_cpu; - - return this_cpu; -} - /* * If dest_cpu is allowed for this process, migrate the task to it. * This is accomplished by forcing the cpu_allowed mask to only @@ -1578,8 +1627,15 @@ void sched_exec(void) sd = tmp; if (sd) { + struct sched_group *group; schedstat_inc(sd, sbe_attempts); - new_cpu = find_idlest_cpu(current, this_cpu, sd); + group = find_idlest_group(sd, current, this_cpu); + if (!group) + goto out; + new_cpu = find_idlest_cpu(group, this_cpu); + if (new_cpu == -1) + goto out; + if (new_cpu != this_cpu) { schedstat_inc(sd, sbe_pushed); put_cpu(); @@ -1792,12 +1848,10 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, if (local_group) { this_load = avg_load; this = group; - goto nextgroup; } else if (avg_load > max_load) { max_load = avg_load; busiest = group; } -nextgroup: group = group->next; } while (group != sd->groups); -- cgit v1.2.3 From 68767a0ae428801649d510d9a65bb71feed44dd1 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:20 -0700 Subject: [PATCH] sched: schedstats update for balance on fork Add SCHEDSTAT statistics for sched-balance-fork. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 10 ++++++-- kernel/sched.c | 63 +++++++++++++++++++++++++++++---------------------- 2 files changed, 44 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 613491d3a875..36a10781c3f3 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -517,10 +517,16 @@ struct sched_domain { unsigned long alb_failed; unsigned long alb_pushed; - /* sched_balance_exec() stats */ - unsigned long sbe_attempts; + /* SD_BALANCE_EXEC stats */ + unsigned long sbe_cnt; + unsigned long sbe_balanced; unsigned long sbe_pushed; + /* SD_BALANCE_FORK stats */ + unsigned long sbf_cnt; + unsigned long sbf_balanced; + unsigned long sbf_pushed; + /* try_to_wake_up() stats */ unsigned long ttwu_wake_remote; unsigned long ttwu_move_affine; diff --git a/kernel/sched.c b/kernel/sched.c index 7ecc237e2aab..2711130cd973 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -309,7 +309,7 @@ static inline void task_rq_unlock(runqueue_t *rq, unsigned long *flags) * bump this up when changing the output format or the meaning of an existing * format, so that tools can adapt (or abort) */ -#define SCHEDSTAT_VERSION 11 +#define SCHEDSTAT_VERSION 12 static int show_schedstat(struct seq_file *seq, void *v) { @@ -356,9 +356,10 @@ static int show_schedstat(struct seq_file *seq, void *v) sd->lb_nobusyq[itype], sd->lb_nobusyg[itype]); } - seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu\n", + seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu\n", sd->alb_cnt, sd->alb_failed, sd->alb_pushed, - sd->sbe_pushed, sd->sbe_attempts, + sd->sbe_cnt, sd->sbe_balanced, sd->sbe_pushed, + sd->sbf_cnt, sd->sbf_balanced, sd->sbf_pushed, sd->ttwu_wake_remote, sd->ttwu_move_affine, sd->ttwu_move_balance); } #endif @@ -1264,24 +1265,34 @@ void fastcall wake_up_new_task(task_t * p, unsigned long clone_flags) sd = tmp; if (sd) { + int new_cpu; struct sched_group *group; + schedstat_inc(sd, sbf_cnt); cpu = task_cpu(p); group = find_idlest_group(sd, p, cpu); - if (group) { - int new_cpu; - new_cpu = find_idlest_cpu(group, cpu); - if (new_cpu != -1 && new_cpu != cpu && - cpu_isset(new_cpu, p->cpus_allowed)) { - set_task_cpu(p, new_cpu); - task_rq_unlock(rq, &flags); - rq = task_rq_lock(p, &flags); - cpu = task_cpu(p); - } + if (!group) { + schedstat_inc(sd, sbf_balanced); + goto no_forkbalance; + } + + new_cpu = find_idlest_cpu(group, cpu); + if (new_cpu == -1 || new_cpu == cpu) { + schedstat_inc(sd, sbf_balanced); + goto no_forkbalance; + } + + if (cpu_isset(new_cpu, p->cpus_allowed)) { + schedstat_inc(sd, sbf_pushed); + set_task_cpu(p, new_cpu); + task_rq_unlock(rq, &flags); + rq = task_rq_lock(p, &flags); + cpu = task_cpu(p); } } -#endif +no_forkbalance: +#endif /* * We decrease the sleep average of forking parents * and children as well, to keep max-interactive tasks @@ -1618,30 +1629,28 @@ void sched_exec(void) struct sched_domain *tmp, *sd = NULL; int new_cpu, this_cpu = get_cpu(); - /* Prefer the current CPU if there's only this task running */ - if (this_rq()->nr_running <= 1) - goto out; - for_each_domain(this_cpu, tmp) if (tmp->flags & SD_BALANCE_EXEC) sd = tmp; if (sd) { struct sched_group *group; - schedstat_inc(sd, sbe_attempts); + schedstat_inc(sd, sbe_cnt); group = find_idlest_group(sd, current, this_cpu); - if (!group) + if (!group) { + schedstat_inc(sd, sbe_balanced); goto out; + } new_cpu = find_idlest_cpu(group, this_cpu); - if (new_cpu == -1) + if (new_cpu == -1 || new_cpu == this_cpu) { + schedstat_inc(sd, sbe_balanced); goto out; - - if (new_cpu != this_cpu) { - schedstat_inc(sd, sbe_pushed); - put_cpu(); - sched_migrate_task(current, new_cpu); - return; } + + schedstat_inc(sd, sbe_pushed); + put_cpu(); + sched_migrate_task(current, new_cpu); + return; } out: put_cpu(); -- cgit v1.2.3 From 687f1661d302bc70ce906594a6d3f615ef075a50 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:21 -0700 Subject: [PATCH] sched: sched tuning Do some basic initial tuning. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/domain.c | 2 +- include/asm-i386/topology.h | 2 +- include/asm-ia64/topology.h | 61 +++++++++++++++++++++++++++++++++---------- include/asm-x86_64/topology.h | 3 +-- include/linux/topology.h | 11 ++++---- 5 files changed, 55 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/arch/ia64/kernel/domain.c b/arch/ia64/kernel/domain.c index fe532c970438..afbde79c3b3d 100644 --- a/arch/ia64/kernel/domain.c +++ b/arch/ia64/kernel/domain.c @@ -14,7 +14,7 @@ #include #include -#define SD_NODES_PER_DOMAIN 6 +#define SD_NODES_PER_DOMAIN 16 #ifdef CONFIG_NUMA /** diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h index 5eb6f61dcefc..2461b731781e 100644 --- a/include/asm-i386/topology.h +++ b/include/asm-i386/topology.h @@ -81,7 +81,7 @@ static inline int node_to_first_cpu(int node) .per_cpu_gain = 100, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_EXEC \ - | SD_BALANCE_NEWIDLE \ + | SD_BALANCE_FORK \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 1, \ diff --git a/include/asm-ia64/topology.h b/include/asm-ia64/topology.h index 21cf351fd05c..4e64c2a6b369 100644 --- a/include/asm-ia64/topology.h +++ b/include/asm-ia64/topology.h @@ -42,25 +42,54 @@ void build_cpu_to_node_map(void); +#define SD_CPU_INIT (struct sched_domain) { \ + .span = CPU_MASK_NONE, \ + .parent = NULL, \ + .groups = NULL, \ + .min_interval = 1, \ + .max_interval = 4, \ + .busy_factor = 64, \ + .imbalance_pct = 125, \ + .cache_hot_time = (10*1000000), \ + .per_cpu_gain = 100, \ + .cache_nice_tries = 2, \ + .busy_idx = 2, \ + .idle_idx = 1, \ + .newidle_idx = 2, \ + .wake_idx = 1, \ + .forkexec_idx = 1, \ + .flags = SD_LOAD_BALANCE \ + | SD_BALANCE_NEWIDLE \ + | SD_BALANCE_EXEC \ + | SD_WAKE_AFFINE, \ + .last_balance = jiffies, \ + .balance_interval = 1, \ + .nr_balance_failed = 0, \ +} + /* sched_domains SD_NODE_INIT for IA64 NUMA machines */ #define SD_NODE_INIT (struct sched_domain) { \ .span = CPU_MASK_NONE, \ .parent = NULL, \ .groups = NULL, \ - .min_interval = 80, \ - .max_interval = 320, \ - .busy_factor = 320, \ + .min_interval = 8, \ + .max_interval = 8*(min(num_online_cpus(), 32)), \ + .busy_factor = 64, \ .imbalance_pct = 125, \ .cache_hot_time = (10*1000000), \ - .cache_nice_tries = 1, \ + .cache_nice_tries = 2, \ + .busy_idx = 3, \ + .idle_idx = 2, \ + .newidle_idx = 0, /* unused */ \ + .wake_idx = 1, \ + .forkexec_idx = 1, \ .per_cpu_gain = 100, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_EXEC \ - | SD_BALANCE_NEWIDLE \ - | SD_WAKE_IDLE \ + | SD_BALANCE_FORK \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ - .balance_interval = 1, \ + .balance_interval = 64, \ .nr_balance_failed = 0, \ } @@ -69,17 +98,21 @@ void build_cpu_to_node_map(void); .span = CPU_MASK_NONE, \ .parent = NULL, \ .groups = NULL, \ - .min_interval = 80, \ - .max_interval = 320, \ - .busy_factor = 320, \ - .imbalance_pct = 125, \ + .min_interval = 64, \ + .max_interval = 64*num_online_cpus(), \ + .busy_factor = 128, \ + .imbalance_pct = 133, \ .cache_hot_time = (10*1000000), \ .cache_nice_tries = 1, \ + .busy_idx = 3, \ + .idle_idx = 3, \ + .newidle_idx = 0, /* unused */ \ + .wake_idx = 0, /* unused */ \ + .forkexec_idx = 0, /* unused */ \ .per_cpu_gain = 100, \ - .flags = SD_LOAD_BALANCE \ - | SD_BALANCE_EXEC, \ + .flags = SD_LOAD_BALANCE, \ .last_balance = jiffies, \ - .balance_interval = 100*(63+num_online_cpus())/64, \ + .balance_interval = 64, \ .nr_balance_failed = 0, \ } diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h index 802d09b9c99f..c1bc3fad482e 100644 --- a/include/asm-x86_64/topology.h +++ b/include/asm-x86_64/topology.h @@ -42,12 +42,11 @@ extern int __node_distance(int, int); .cache_nice_tries = 2, \ .busy_idx = 3, \ .idle_idx = 2, \ - .newidle_idx = 1, \ + .newidle_idx = 0, \ .wake_idx = 1, \ .forkexec_idx = 1, \ .per_cpu_gain = 100, \ .flags = SD_LOAD_BALANCE \ - | SD_BALANCE_NEWIDLE \ | SD_BALANCE_FORK \ | SD_BALANCE_EXEC \ | SD_WAKE_BALANCE, \ diff --git a/include/linux/topology.h b/include/linux/topology.h index 665597207def..0320225e96da 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -91,7 +91,7 @@ .per_cpu_gain = 25, \ .busy_idx = 0, \ .idle_idx = 0, \ - .newidle_idx = 0, \ + .newidle_idx = 1, \ .wake_idx = 0, \ .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ @@ -121,15 +121,14 @@ .cache_nice_tries = 1, \ .per_cpu_gain = 100, \ .busy_idx = 2, \ - .idle_idx = 0, \ - .newidle_idx = 1, \ + .idle_idx = 1, \ + .newidle_idx = 2, \ .wake_idx = 1, \ - .forkexec_idx = 0, \ + .forkexec_idx = 1, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ - | SD_WAKE_AFFINE \ - | SD_WAKE_BALANCE, \ + | SD_WAKE_AFFINE, \ .last_balance = jiffies, \ .balance_interval = 1, \ .nr_balance_failed = 0, \ -- cgit v1.2.3 From 4866cde064afbb6c2a488c265e696879de616daa Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:23 -0700 Subject: [PATCH] sched: cleanup context switch locking Instead of requiring architecture code to interact with the scheduler's locking implementation, provide a couple of defines that can be used by the architecture to request runqueue unlocked context switches, and ask for interrupts to be enabled over the context switch. Also replaces the "switch_lock" used by these architectures with an oncpu flag (note, not a potentially slow bitflag). This eliminates one bus locked memory operation when context switching, and simplifies the task_running function. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-arm/system.h | 30 ++-------- include/asm-ia64/system.h | 10 +--- include/asm-mips/system.h | 10 +--- include/asm-s390/system.h | 17 +----- include/asm-sparc/system.h | 4 +- include/asm-sparc64/system.h | 14 ++--- include/linux/init_task.h | 1 - include/linux/sched.h | 10 +++- kernel/sched.c | 132 +++++++++++++++++++++++++++++++++++-------- 9 files changed, 131 insertions(+), 97 deletions(-) (limited to 'include/linux') diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h index 39dd7008013c..3d0d2860b6db 100644 --- a/include/asm-arm/system.h +++ b/include/asm-arm/system.h @@ -145,34 +145,12 @@ extern unsigned int user_debug; #define set_wmb(var, value) do { var = value; wmb(); } while (0) #define nop() __asm__ __volatile__("mov\tr0,r0\t@ nop\n\t"); -#ifdef CONFIG_SMP /* - * Define our own context switch locking. This allows us to enable - * interrupts over the context switch, otherwise we end up with high - * interrupt latency. The real problem area is switch_mm() which may - * do a full cache flush. + * switch_mm() may do a full cache flush over the context switch, + * so enable interrupts over the context switch to avoid high + * latency. */ -#define prepare_arch_switch(rq,next) \ -do { \ - spin_lock(&(next)->switch_lock); \ - spin_unlock_irq(&(rq)->lock); \ -} while (0) - -#define finish_arch_switch(rq,prev) \ - spin_unlock(&(prev)->switch_lock) - -#define task_running(rq,p) \ - ((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock)) -#else -/* - * Our UP-case is more simple, but we assume knowledge of how - * spin_unlock_irq() and friends are implemented. This avoids - * us needlessly decrementing and incrementing the preempt count. - */ -#define prepare_arch_switch(rq,next) local_irq_enable() -#define finish_arch_switch(rq,prev) spin_unlock(&(rq)->lock) -#define task_running(rq,p) ((rq)->curr == (p)) -#endif +#define __ARCH_WANT_INTERRUPTS_ON_CTXSW /* * switch_to(prev, next) should switch from task `prev' to `next' diff --git a/include/asm-ia64/system.h b/include/asm-ia64/system.h index 6f516e76d1f0..cd2cf76b2db1 100644 --- a/include/asm-ia64/system.h +++ b/include/asm-ia64/system.h @@ -183,8 +183,6 @@ do { \ #ifdef __KERNEL__ -#define prepare_to_switch() do { } while(0) - #ifdef CONFIG_IA32_SUPPORT # define IS_IA32_PROCESS(regs) (ia64_psr(regs)->is != 0) #else @@ -274,13 +272,7 @@ extern void ia64_load_extra (struct task_struct *task); * of that CPU which will not be released, because there we wait for the * tasklist_lock to become available. */ -#define prepare_arch_switch(rq, next) \ -do { \ - spin_lock(&(next)->switch_lock); \ - spin_unlock(&(rq)->lock); \ -} while (0) -#define finish_arch_switch(rq, prev) spin_unlock_irq(&(prev)->switch_lock) -#define task_running(rq, p) ((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock)) +#define __ARCH_WANT_UNLOCKED_CTXSW #define ia64_platform_is(x) (strcmp(x, platform_name) == 0) diff --git a/include/asm-mips/system.h b/include/asm-mips/system.h index 888fd8908467..169f3d4265b1 100644 --- a/include/asm-mips/system.h +++ b/include/asm-mips/system.h @@ -422,16 +422,10 @@ extern void __die_if_kernel(const char *, struct pt_regs *, const char *file, extern int stop_a_enabled; /* - * Taken from include/asm-ia64/system.h; prevents deadlock on SMP + * See include/asm-ia64/system.h; prevents deadlock on SMP * systems. */ -#define prepare_arch_switch(rq, next) \ -do { \ - spin_lock(&(next)->switch_lock); \ - spin_unlock(&(rq)->lock); \ -} while (0) -#define finish_arch_switch(rq, prev) spin_unlock_irq(&(prev)->switch_lock) -#define task_running(rq, p) ((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock)) +#define __ARCH_WANT_UNLOCKED_CTXSW #define arch_align_stack(x) (x) diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h index e3cb3ce1d24a..b4a9f05a93d6 100644 --- a/include/asm-s390/system.h +++ b/include/asm-s390/system.h @@ -104,29 +104,18 @@ static inline void restore_access_regs(unsigned int *acrs) prev = __switch_to(prev,next); \ } while (0) -#define prepare_arch_switch(rq, next) do { } while(0) -#define task_running(rq, p) ((rq)->curr == (p)) - #ifdef CONFIG_VIRT_CPU_ACCOUNTING extern void account_user_vtime(struct task_struct *); extern void account_system_vtime(struct task_struct *); - -#define finish_arch_switch(rq, prev) do { \ - set_fs(current->thread.mm_segment); \ - spin_unlock(&(rq)->lock); \ - account_system_vtime(prev); \ - local_irq_enable(); \ -} while (0) - #else +#define account_system_vtime(prev) do { } while (0) +#endif #define finish_arch_switch(rq, prev) do { \ set_fs(current->thread.mm_segment); \ - spin_unlock_irq(&(rq)->lock); \ + account_system_vtime(prev); \ } while (0) -#endif - #define nop() __asm__ __volatile__ ("nop") #define xchg(ptr,x) \ diff --git a/include/asm-sparc/system.h b/include/asm-sparc/system.h index 80cf20cfaee1..898562ebe94c 100644 --- a/include/asm-sparc/system.h +++ b/include/asm-sparc/system.h @@ -101,7 +101,7 @@ extern void fpsave(unsigned long *fpregs, unsigned long *fsr, * SWITCH_ENTER and SWITH_DO_LAZY_FPU do not work yet (e.g. SMP does not work) * XXX WTF is the above comment? Found in late teen 2.4.x. */ -#define prepare_arch_switch(rq, next) do { \ +#define prepare_arch_switch(next) do { \ __asm__ __volatile__( \ ".globl\tflush_patch_switch\nflush_patch_switch:\n\t" \ "save %sp, -0x40, %sp; save %sp, -0x40, %sp; save %sp, -0x40, %sp\n\t" \ @@ -109,8 +109,6 @@ extern void fpsave(unsigned long *fpregs, unsigned long *fsr, "save %sp, -0x40, %sp\n\t" \ "restore; restore; restore; restore; restore; restore; restore"); \ } while(0) -#define finish_arch_switch(rq, next) spin_unlock_irq(&(rq)->lock) -#define task_running(rq, p) ((rq)->curr == (p)) /* Much care has gone into this code, do not touch it. * diff --git a/include/asm-sparc64/system.h b/include/asm-sparc64/system.h index fd12ca386f48..f9be2c5b4dc9 100644 --- a/include/asm-sparc64/system.h +++ b/include/asm-sparc64/system.h @@ -139,19 +139,13 @@ extern void __flushw_user(void); #define flush_user_windows flushw_user #define flush_register_windows flushw_all -#define prepare_arch_switch(rq, next) \ -do { spin_lock(&(next)->switch_lock); \ - spin_unlock(&(rq)->lock); \ +/* Don't hold the runqueue lock over context switch */ +#define __ARCH_WANT_UNLOCKED_CTXSW +#define prepare_arch_switch(next) \ +do { \ flushw_all(); \ } while (0) -#define finish_arch_switch(rq, prev) \ -do { spin_unlock_irq(&(prev)->switch_lock); \ -} while (0) - -#define task_running(rq, p) \ - ((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock)) - /* See what happens when you design the chip correctly? * * We tell gcc we clobber all non-fixed-usage registers except diff --git a/include/linux/init_task.h b/include/linux/init_task.h index a6a8c1a38d5e..03206a425d7a 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -108,7 +108,6 @@ extern struct group_info init_groups; .blocked = {{0}}, \ .alloc_lock = SPIN_LOCK_UNLOCKED, \ .proc_lock = SPIN_LOCK_UNLOCKED, \ - .switch_lock = SPIN_LOCK_UNLOCKED, \ .journal_info = NULL, \ .cpu_timers = INIT_CPU_TIMERS(tsk.cpu_timers), \ } diff --git a/include/linux/sched.h b/include/linux/sched.h index 36a10781c3f3..d27be9337425 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -368,6 +368,11 @@ struct signal_struct { #endif }; +/* Context switch must be unlocked if interrupts are to be enabled */ +#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW +# define __ARCH_WANT_UNLOCKED_CTXSW +#endif + /* * Bits in flags field of signal_struct. */ @@ -594,6 +599,9 @@ struct task_struct { int lock_depth; /* BKL lock depth */ +#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) + int oncpu; +#endif int prio, static_prio; struct list_head run_list; prio_array_t *array; @@ -716,8 +724,6 @@ struct task_struct { spinlock_t alloc_lock; /* Protection of proc_dentry: nesting proc_lock, dcache_lock, write_lock_irq(&tasklist_lock); */ spinlock_t proc_lock; -/* context-switch lock */ - spinlock_t switch_lock; /* journalling filesystem info */ void *journal_info; diff --git a/kernel/sched.c b/kernel/sched.c index 98bf1c091da5..b1410577f9a8 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -268,14 +268,71 @@ static DEFINE_PER_CPU(struct runqueue, runqueues); #define task_rq(p) cpu_rq(task_cpu(p)) #define cpu_curr(cpu) (cpu_rq(cpu)->curr) -/* - * Default context-switch locking: - */ #ifndef prepare_arch_switch -# define prepare_arch_switch(rq, next) do { } while (0) -# define finish_arch_switch(rq, next) spin_unlock_irq(&(rq)->lock) -# define task_running(rq, p) ((rq)->curr == (p)) +# define prepare_arch_switch(next) do { } while (0) +#endif +#ifndef finish_arch_switch +# define finish_arch_switch(prev) do { } while (0) +#endif + +#ifndef __ARCH_WANT_UNLOCKED_CTXSW +static inline int task_running(runqueue_t *rq, task_t *p) +{ + return rq->curr == p; +} + +static inline void prepare_lock_switch(runqueue_t *rq, task_t *next) +{ +} + +static inline void finish_lock_switch(runqueue_t *rq, task_t *prev) +{ + spin_unlock_irq(&rq->lock); +} + +#else /* __ARCH_WANT_UNLOCKED_CTXSW */ +static inline int task_running(runqueue_t *rq, task_t *p) +{ +#ifdef CONFIG_SMP + return p->oncpu; +#else + return rq->curr == p; +#endif +} + +static inline void prepare_lock_switch(runqueue_t *rq, task_t *next) +{ +#ifdef CONFIG_SMP + /* + * We can optimise this out completely for !SMP, because the + * SMP rebalancing from interrupt is the only thing that cares + * here. + */ + next->oncpu = 1; +#endif +#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW + spin_unlock_irq(&rq->lock); +#else + spin_unlock(&rq->lock); #endif +} + +static inline void finish_lock_switch(runqueue_t *rq, task_t *prev) +{ +#ifdef CONFIG_SMP + /* + * After ->oncpu is cleared, the task can be moved to a different CPU. + * We must ensure this doesn't happen until the switch is completely + * finished. + */ + smp_wmb(); + prev->oncpu = 0; +#endif +#ifndef __ARCH_WANT_INTERRUPTS_ON_CTXSW + local_irq_enable(); +#endif +} +#endif /* __ARCH_WANT_UNLOCKED_CTXSW */ /* * task_rq_lock - lock the runqueue a given task resides on and disable @@ -1196,17 +1253,14 @@ void fastcall sched_fork(task_t *p) p->state = TASK_RUNNING; INIT_LIST_HEAD(&p->run_list); p->array = NULL; - spin_lock_init(&p->switch_lock); #ifdef CONFIG_SCHEDSTATS memset(&p->sched_info, 0, sizeof(p->sched_info)); #endif +#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) + p->oncpu = 0; +#endif #ifdef CONFIG_PREEMPT - /* - * During context-switch we hold precisely one spinlock, which - * schedule_tail drops. (in the common case it's this_rq()->lock, - * but it also can be p->switch_lock.) So we compensate with a count - * of 1. Also, we want to start with kernel preemption disabled. - */ + /* Want to start with kernel preemption disabled. */ p->thread_info->preempt_count = 1; #endif /* @@ -1387,23 +1441,41 @@ void fastcall sched_exit(task_t * p) task_rq_unlock(rq, &flags); } +/** + * prepare_task_switch - prepare to switch tasks + * @rq: the runqueue preparing to switch + * @next: the task we are going to switch to. + * + * This is called with the rq lock held and interrupts off. It must + * be paired with a subsequent finish_task_switch after the context + * switch. + * + * prepare_task_switch sets up locking and calls architecture specific + * hooks. + */ +static inline void prepare_task_switch(runqueue_t *rq, task_t *next) +{ + prepare_lock_switch(rq, next); + prepare_arch_switch(next); +} + /** * finish_task_switch - clean up after a task-switch * @prev: the thread we just switched away from. * - * We enter this with the runqueue still locked, and finish_arch_switch() - * will unlock it along with doing any other architecture-specific cleanup - * actions. + * finish_task_switch must be called after the context switch, paired + * with a prepare_task_switch call before the context switch. + * finish_task_switch will reconcile locking set up by prepare_task_switch, + * and do any other architecture-specific cleanup actions. * * Note that we may have delayed dropping an mm in context_switch(). If * so, we finish that here outside of the runqueue lock. (Doing it * with the lock held can cause deadlocks; see schedule() for * details.) */ -static inline void finish_task_switch(task_t *prev) +static inline void finish_task_switch(runqueue_t *rq, task_t *prev) __releases(rq->lock) { - runqueue_t *rq = this_rq(); struct mm_struct *mm = rq->prev_mm; unsigned long prev_task_flags; @@ -1421,7 +1493,8 @@ static inline void finish_task_switch(task_t *prev) * Manfred Spraul */ prev_task_flags = prev->flags; - finish_arch_switch(rq, prev); + finish_arch_switch(prev); + finish_lock_switch(rq, prev); if (mm) mmdrop(mm); if (unlikely(prev_task_flags & PF_DEAD)) @@ -1435,8 +1508,12 @@ static inline void finish_task_switch(task_t *prev) asmlinkage void schedule_tail(task_t *prev) __releases(rq->lock) { - finish_task_switch(prev); - + runqueue_t *rq = this_rq(); + finish_task_switch(rq, prev); +#ifdef __ARCH_WANT_UNLOCKED_CTXSW + /* In this case, finish_task_switch does not reenable preemption */ + preempt_enable(); +#endif if (current->set_child_tid) put_user(current->pid, current->set_child_tid); } @@ -2816,11 +2893,15 @@ switch_tasks: rq->curr = next; ++*switch_count; - prepare_arch_switch(rq, next); + prepare_task_switch(rq, next); prev = context_switch(rq, prev, next); barrier(); - - finish_task_switch(prev); + /* + * this_rq must be evaluated again because prev may have moved + * CPUs since it called schedule(), thus the 'rq' on its stack + * frame will be invalid. + */ + finish_task_switch(this_rq(), prev); } else spin_unlock_irq(&rq->lock); @@ -4085,6 +4166,9 @@ void __devinit init_idle(task_t *idle, int cpu) spin_lock_irqsave(&rq->lock, flags); rq->curr = rq->idle = idle; +#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) + idle->oncpu = 1; +#endif set_tsk_need_resched(idle); spin_unlock_irqrestore(&rq->lock, flags); -- cgit v1.2.3 From 476d139c218e44e045e4bc6d4cc02b010b343939 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:29 -0700 Subject: [PATCH] sched: consolidate sbe sbf Consolidate balance-on-exec with balance-on-fork. This is made easy by the sched-domains RCU patches. As well as the general goodness of code reduction, this allows the runqueues to be unlocked during balance-on-fork. schedstats is a problem. Maybe just have balance-on-event instead of distinguishing fork and exec? Signed-off-by: Nick Piggin Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 2 +- kernel/fork.c | 21 +++--- kernel/sched.c | 174 ++++++++++++++++++++------------------------------ 3 files changed, 81 insertions(+), 116 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index d27be9337425..edb2c69a8873 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -930,7 +930,7 @@ extern void FASTCALL(wake_up_new_task(struct task_struct * tsk, #else static inline void kick_process(struct task_struct *tsk) { } #endif -extern void FASTCALL(sched_fork(task_t * p)); +extern void FASTCALL(sched_fork(task_t * p, int clone_flags)); extern void FASTCALL(sched_exit(task_t * p)); extern int in_group_p(gid_t); diff --git a/kernel/fork.c b/kernel/fork.c index a28d11e10877..2c7806873bfd 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1003,9 +1003,6 @@ static task_t *copy_process(unsigned long clone_flags, p->pdeath_signal = 0; p->exit_state = 0; - /* Perform scheduler related setup */ - sched_fork(p); - /* * Ok, make it visible to the rest of the system. * We dont wake it up yet. @@ -1014,18 +1011,24 @@ static task_t *copy_process(unsigned long clone_flags, INIT_LIST_HEAD(&p->ptrace_children); INIT_LIST_HEAD(&p->ptrace_list); + /* Perform scheduler related setup. Assign this task to a CPU. */ + sched_fork(p, clone_flags); + /* Need tasklist lock for parent etc handling! */ write_lock_irq(&tasklist_lock); /* - * The task hasn't been attached yet, so cpus_allowed mask cannot - * have changed. The cpus_allowed mask of the parent may have - * changed after it was copied first time, and it may then move to - * another CPU - so we re-copy it here and set the child's CPU to - * the parent's CPU. This avoids alot of nasty races. + * The task hasn't been attached yet, so its cpus_allowed mask will + * not be changed, nor will its assigned CPU. + * + * The cpus_allowed mask of the parent may have changed after it was + * copied first time - so re-copy it here, then check the child's CPU + * to ensure it is on a valid CPU (and if not, just force it back to + * parent's CPU). This avoids alot of nasty races. */ p->cpus_allowed = current->cpus_allowed; - set_task_cpu(p, smp_processor_id()); + if (unlikely(!cpu_isset(task_cpu(p), p->cpus_allowed))) + set_task_cpu(p, smp_processor_id()); /* * Check for pending SIGKILL! The new thread should not be allowed diff --git a/kernel/sched.c b/kernel/sched.c index 54ce787b6207..579da278e72f 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1021,8 +1021,59 @@ static int find_idlest_cpu(struct sched_group *group, int this_cpu) return idlest; } +/* + * sched_balance_self: balance the current task (running on cpu) in domains + * that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and + * SD_BALANCE_EXEC. + * + * Balance, ie. select the least loaded group. + * + * Returns the target CPU number, or the same CPU if no balancing is needed. + * + * preempt must be disabled. + */ +static int sched_balance_self(int cpu, int flag) +{ + struct task_struct *t = current; + struct sched_domain *tmp, *sd = NULL; -#endif + for_each_domain(cpu, tmp) + if (tmp->flags & flag) + sd = tmp; + + while (sd) { + cpumask_t span; + struct sched_group *group; + int new_cpu; + int weight; + + span = sd->span; + group = find_idlest_group(sd, t, cpu); + if (!group) + goto nextlevel; + + new_cpu = find_idlest_cpu(group, cpu); + if (new_cpu == -1 || new_cpu == cpu) + goto nextlevel; + + /* Now try balancing at a lower domain level */ + cpu = new_cpu; +nextlevel: + sd = NULL; + weight = cpus_weight(span); + for_each_domain(cpu, tmp) { + if (weight <= cpus_weight(tmp->span)) + break; + if (tmp->flags & flag) + sd = tmp; + } + /* while loop will break here if sd == NULL */ + } + + return cpu; +} + +#endif /* CONFIG_SMP */ /* * wake_idle() will wake a task on an idle cpu if task->cpu is @@ -1240,8 +1291,15 @@ int fastcall wake_up_state(task_t *p, unsigned int state) * Perform scheduler related setup for a newly forked process p. * p is forked by current. */ -void fastcall sched_fork(task_t *p) +void fastcall sched_fork(task_t *p, int clone_flags) { + int cpu = get_cpu(); + +#ifdef CONFIG_SMP + cpu = sched_balance_self(cpu, SD_BALANCE_FORK); +#endif + set_task_cpu(p, cpu); + /* * We mark the process as running here, but have not actually * inserted it onto the runqueue yet. This guarantees that @@ -1282,12 +1340,10 @@ void fastcall sched_fork(task_t *p) * runqueue lock is not a problem. */ current->time_slice = 1; - preempt_disable(); scheduler_tick(); - local_irq_enable(); - preempt_enable(); - } else - local_irq_enable(); + } + local_irq_enable(); + put_cpu(); } /* @@ -1302,64 +1358,12 @@ void fastcall wake_up_new_task(task_t * p, unsigned long clone_flags) unsigned long flags; int this_cpu, cpu; runqueue_t *rq, *this_rq; -#ifdef CONFIG_SMP - struct sched_domain *tmp, *sd = NULL; -#endif rq = task_rq_lock(p, &flags); BUG_ON(p->state != TASK_RUNNING); this_cpu = smp_processor_id(); cpu = task_cpu(p); -#ifdef CONFIG_SMP - for_each_domain(cpu, tmp) - if (tmp->flags & SD_BALANCE_FORK) - sd = tmp; - - if (sd) { - cpumask_t span; - int new_cpu; - struct sched_group *group; - -again: - schedstat_inc(sd, sbf_cnt); - span = sd->span; - cpu = task_cpu(p); - group = find_idlest_group(sd, p, cpu); - if (!group) { - schedstat_inc(sd, sbf_balanced); - goto nextlevel; - } - - new_cpu = find_idlest_cpu(group, cpu); - if (new_cpu == -1 || new_cpu == cpu) { - schedstat_inc(sd, sbf_balanced); - goto nextlevel; - } - - if (cpu_isset(new_cpu, p->cpus_allowed)) { - schedstat_inc(sd, sbf_pushed); - set_task_cpu(p, new_cpu); - task_rq_unlock(rq, &flags); - rq = task_rq_lock(p, &flags); - cpu = task_cpu(p); - } - - /* Now try balancing at a lower domain level */ -nextlevel: - sd = NULL; - for_each_domain(cpu, tmp) { - if (cpus_subset(span, tmp->span)) - break; - if (tmp->flags & SD_BALANCE_FORK) - sd = tmp; - } - - if (sd) - goto again; - } - -#endif /* * We decrease the sleep average of forking parents * and children as well, to keep max-interactive tasks @@ -1708,58 +1712,16 @@ out: } /* - * sched_exec(): find the highest-level, exec-balance-capable - * domain and try to migrate the task to the least loaded CPU. - * - * execve() is a valuable balancing opportunity, because at this point - * the task has the smallest effective memory and cache footprint. + * sched_exec - execve() is a valuable balancing opportunity, because at + * this point the task has the smallest effective memory and cache footprint. */ void sched_exec(void) { - struct sched_domain *tmp, *sd = NULL; int new_cpu, this_cpu = get_cpu(); - - for_each_domain(this_cpu, tmp) - if (tmp->flags & SD_BALANCE_EXEC) - sd = tmp; - - if (sd) { - cpumask_t span; - struct sched_group *group; -again: - schedstat_inc(sd, sbe_cnt); - span = sd->span; - group = find_idlest_group(sd, current, this_cpu); - if (!group) { - schedstat_inc(sd, sbe_balanced); - goto nextlevel; - } - new_cpu = find_idlest_cpu(group, this_cpu); - if (new_cpu == -1 || new_cpu == this_cpu) { - schedstat_inc(sd, sbe_balanced); - goto nextlevel; - } - - schedstat_inc(sd, sbe_pushed); - put_cpu(); - sched_migrate_task(current, new_cpu); - - /* Now try balancing at a lower domain level */ - this_cpu = get_cpu(); -nextlevel: - sd = NULL; - for_each_domain(this_cpu, tmp) { - if (cpus_subset(span, tmp->span)) - break; - if (tmp->flags & SD_BALANCE_EXEC) - sd = tmp; - } - - if (sd) - goto again; - } - + new_cpu = sched_balance_self(this_cpu, SD_BALANCE_EXEC); put_cpu(); + if (new_cpu != this_cpu) + sched_migrate_task(current, new_cpu); } /* -- cgit v1.2.3 From 1a20ff27ef75d866730ee796acd811a925af762f Mon Sep 17 00:00:00 2001 From: Dinakar Guniguntala Date: Sat, 25 Jun 2005 14:57:33 -0700 Subject: [PATCH] Dynamic sched domains: sched changes The following patches add dynamic sched domains functionality that was extensively discussed on lkml and lse-tech. I would like to see this added to -mm o The main advantage with this feature is that it ensures that the scheduler load balacing code only balances against the cpus that are in the sched domain as defined by an exclusive cpuset and not all of the cpus in the system. This removes any overhead due to load balancing code trying to pull tasks outside of the cpu exclusive cpuset only to be prevented by the tasks' cpus_allowed mask. o cpu exclusive cpusets are useful for servers running orthogonal workloads such as RT applications requiring low latency and HPC applications that are throughput sensitive o It provides a new API partition_sched_domains in sched.c that makes dynamic sched domains possible. o cpu_exclusive cpusets sets are now associated with a sched domain. Which means that the users can dynamically modify the sched domains through the cpuset file system interface o ia64 sched domain code has been updated to support this feature as well o Currently, this does not support hotplug. (However some of my tests indicate hotplug+preempt is currently broken) o I have tested it extensively on x86. o This should have very minimal impact on performance as none of the fast paths are affected Signed-off-by: Dinakar Guniguntala Acked-by: Paul Jackson Acked-by: Nick Piggin Acked-by: Matthew Dobson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 2 + kernel/sched.c | 132 ++++++++++++++++++++++++++++++++------------------ 2 files changed, 88 insertions(+), 46 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index edb2c69a8873..98c109e4f43d 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -539,6 +539,8 @@ struct sched_domain { #endif }; +extern void partition_sched_domains(cpumask_t *partition1, + cpumask_t *partition2); #ifdef ARCH_HAS_SCHED_DOMAIN /* Useful helpers that arch setup code may use. Defined in kernel/sched.c */ extern cpumask_t cpu_isolated_map; diff --git a/kernel/sched.c b/kernel/sched.c index d3d81b82e378..dee96b22635e 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -262,7 +262,7 @@ static DEFINE_PER_CPU(struct runqueue, runqueues); /* * The domain tree (rq->sd) is protected by RCU's quiescent state transition. - * See update_sched_domains: synchronize_kernel for details. + * See detach_destroy_domains: synchronize_sched for details. * * The domain tree of any CPU may only be accessed from within * preempt-disabled sections. @@ -4624,7 +4624,7 @@ int __init migration_init(void) #endif #ifdef CONFIG_SMP -#define SCHED_DOMAIN_DEBUG +#undef SCHED_DOMAIN_DEBUG #ifdef SCHED_DOMAIN_DEBUG static void sched_domain_debug(struct sched_domain *sd, int cpu) { @@ -4717,7 +4717,7 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu) #define sched_domain_debug(sd, cpu) {} #endif -static int __devinit sd_degenerate(struct sched_domain *sd) +static int sd_degenerate(struct sched_domain *sd) { if (cpus_weight(sd->span) == 1) return 1; @@ -4740,7 +4740,7 @@ static int __devinit sd_degenerate(struct sched_domain *sd) return 1; } -static int __devinit sd_parent_degenerate(struct sched_domain *sd, +static int sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) { unsigned long cflags = sd->flags, pflags = parent->flags; @@ -4772,7 +4772,7 @@ static int __devinit sd_parent_degenerate(struct sched_domain *sd, * Attach the domain 'sd' to 'cpu' as its base domain. Callers must * hold the hotplug lock. */ -void __devinit cpu_attach_domain(struct sched_domain *sd, int cpu) +void cpu_attach_domain(struct sched_domain *sd, int cpu) { runqueue_t *rq = cpu_rq(cpu); struct sched_domain *tmp; @@ -4823,7 +4823,7 @@ __setup ("isolcpus=", isolated_cpu_setup); * covered by the given span, and will set each group's ->cpumask correctly, * and ->cpu_power to 0. */ -void __devinit init_sched_build_groups(struct sched_group groups[], +void init_sched_build_groups(struct sched_group groups[], cpumask_t span, int (*group_fn)(int cpu)) { struct sched_group *first = NULL, *last = NULL; @@ -4859,13 +4859,14 @@ void __devinit init_sched_build_groups(struct sched_group groups[], #ifdef ARCH_HAS_SCHED_DOMAIN -extern void __devinit arch_init_sched_domains(void); -extern void __devinit arch_destroy_sched_domains(void); +extern void build_sched_domains(const cpumask_t *cpu_map); +extern void arch_init_sched_domains(const cpumask_t *cpu_map); +extern void arch_destroy_sched_domains(const cpumask_t *cpu_map); #else #ifdef CONFIG_SCHED_SMT static DEFINE_PER_CPU(struct sched_domain, cpu_domains); static struct sched_group sched_group_cpus[NR_CPUS]; -static int __devinit cpu_to_cpu_group(int cpu) +static int cpu_to_cpu_group(int cpu) { return cpu; } @@ -4873,7 +4874,7 @@ static int __devinit cpu_to_cpu_group(int cpu) static DEFINE_PER_CPU(struct sched_domain, phys_domains); static struct sched_group sched_group_phys[NR_CPUS]; -static int __devinit cpu_to_phys_group(int cpu) +static int cpu_to_phys_group(int cpu) { #ifdef CONFIG_SCHED_SMT return first_cpu(cpu_sibling_map[cpu]); @@ -4886,7 +4887,7 @@ static int __devinit cpu_to_phys_group(int cpu) static DEFINE_PER_CPU(struct sched_domain, node_domains); static struct sched_group sched_group_nodes[MAX_NUMNODES]; -static int __devinit cpu_to_node_group(int cpu) +static int cpu_to_node_group(int cpu) { return cpu_to_node(cpu); } @@ -4917,39 +4918,28 @@ static void check_sibling_maps(void) #endif /* - * Set up scheduler domains and groups. Callers must hold the hotplug lock. + * Build sched domains for a given set of cpus and attach the sched domains + * to the individual cpus */ -static void __devinit arch_init_sched_domains(void) +static void build_sched_domains(const cpumask_t *cpu_map) { int i; - cpumask_t cpu_default_map; - -#if defined(CONFIG_SCHED_SMT) && defined(CONFIG_NUMA) - check_sibling_maps(); -#endif - /* - * Setup mask for cpus without special case scheduling requirements. - * For now this just excludes isolated cpus, but could be used to - * exclude other special cases in the future. - */ - cpus_complement(cpu_default_map, cpu_isolated_map); - cpus_and(cpu_default_map, cpu_default_map, cpu_online_map); /* - * Set up domains. Isolated domains just stay on the NULL domain. + * Set up domains for cpus specified by the cpu_map. */ - for_each_cpu_mask(i, cpu_default_map) { + for_each_cpu_mask(i, *cpu_map) { int group; struct sched_domain *sd = NULL, *p; cpumask_t nodemask = node_to_cpumask(cpu_to_node(i)); - cpus_and(nodemask, nodemask, cpu_default_map); + cpus_and(nodemask, nodemask, *cpu_map); #ifdef CONFIG_NUMA sd = &per_cpu(node_domains, i); group = cpu_to_node_group(i); *sd = SD_NODE_INIT; - sd->span = cpu_default_map; + sd->span = *cpu_map; sd->groups = &sched_group_nodes[group]; #endif @@ -4967,7 +4957,7 @@ static void __devinit arch_init_sched_domains(void) group = cpu_to_cpu_group(i); *sd = SD_SIBLING_INIT; sd->span = cpu_sibling_map[i]; - cpus_and(sd->span, sd->span, cpu_default_map); + cpus_and(sd->span, sd->span, *cpu_map); sd->parent = p; sd->groups = &sched_group_cpus[group]; #endif @@ -4977,7 +4967,7 @@ static void __devinit arch_init_sched_domains(void) /* Set up CPU (sibling) groups */ for_each_online_cpu(i) { cpumask_t this_sibling_map = cpu_sibling_map[i]; - cpus_and(this_sibling_map, this_sibling_map, cpu_default_map); + cpus_and(this_sibling_map, this_sibling_map, *cpu_map); if (i != first_cpu(this_sibling_map)) continue; @@ -4990,7 +4980,7 @@ static void __devinit arch_init_sched_domains(void) for (i = 0; i < MAX_NUMNODES; i++) { cpumask_t nodemask = node_to_cpumask(i); - cpus_and(nodemask, nodemask, cpu_default_map); + cpus_and(nodemask, nodemask, *cpu_map); if (cpus_empty(nodemask)) continue; @@ -5000,12 +4990,12 @@ static void __devinit arch_init_sched_domains(void) #ifdef CONFIG_NUMA /* Set up node groups */ - init_sched_build_groups(sched_group_nodes, cpu_default_map, + init_sched_build_groups(sched_group_nodes, *cpu_map, &cpu_to_node_group); #endif /* Calculate CPU power for physical packages and nodes */ - for_each_cpu_mask(i, cpu_default_map) { + for_each_cpu_mask(i, *cpu_map) { int power; struct sched_domain *sd; #ifdef CONFIG_SCHED_SMT @@ -5029,7 +5019,7 @@ static void __devinit arch_init_sched_domains(void) } /* Attach the domains */ - for_each_online_cpu(i) { + for_each_cpu_mask(i, *cpu_map) { struct sched_domain *sd; #ifdef CONFIG_SCHED_SMT sd = &per_cpu(cpu_domains, i); @@ -5039,16 +5029,71 @@ static void __devinit arch_init_sched_domains(void) cpu_attach_domain(sd, i); } } +/* + * Set up scheduler domains and groups. Callers must hold the hotplug lock. + */ +static void arch_init_sched_domains(cpumask_t *cpu_map) +{ + cpumask_t cpu_default_map; -#ifdef CONFIG_HOTPLUG_CPU -static void __devinit arch_destroy_sched_domains(void) +#if defined(CONFIG_SCHED_SMT) && defined(CONFIG_NUMA) + check_sibling_maps(); +#endif + /* + * Setup mask for cpus without special case scheduling requirements. + * For now this just excludes isolated cpus, but could be used to + * exclude other special cases in the future. + */ + cpus_andnot(cpu_default_map, *cpu_map, cpu_isolated_map); + + build_sched_domains(&cpu_default_map); +} + +static void arch_destroy_sched_domains(const cpumask_t *cpu_map) { /* Do nothing: everything is statically allocated. */ } -#endif #endif /* ARCH_HAS_SCHED_DOMAIN */ +/* + * Detach sched domains from a group of cpus specified in cpu_map + * These cpus will now be attached to the NULL domain + */ +static inline void detach_destroy_domains(const cpumask_t *cpu_map) +{ + int i; + + for_each_cpu_mask(i, *cpu_map) + cpu_attach_domain(NULL, i); + synchronize_sched(); + arch_destroy_sched_domains(cpu_map); +} + +/* + * Partition sched domains as specified by the cpumasks below. + * This attaches all cpus from the cpumasks to the NULL domain, + * waits for a RCU quiescent period, recalculates sched + * domain information and then attaches them back to the + * correct sched domains + * Call with hotplug lock held + */ +void partition_sched_domains(cpumask_t *partition1, cpumask_t *partition2) +{ + cpumask_t change_map; + + cpus_and(*partition1, *partition1, cpu_online_map); + cpus_and(*partition2, *partition2, cpu_online_map); + cpus_or(change_map, *partition1, *partition2); + + /* Detach sched domains from all of the affected cpus */ + detach_destroy_domains(&change_map); + if (!cpus_empty(*partition1)) + build_sched_domains(partition1); + if (!cpus_empty(*partition2)) + build_sched_domains(partition2); +} + #ifdef CONFIG_HOTPLUG_CPU /* * Force a reinitialization of the sched domains hierarchy. The domains @@ -5059,15 +5104,10 @@ static void __devinit arch_destroy_sched_domains(void) static int update_sched_domains(struct notifier_block *nfb, unsigned long action, void *hcpu) { - int i; - switch (action) { case CPU_UP_PREPARE: case CPU_DOWN_PREPARE: - for_each_online_cpu(i) - cpu_attach_domain(NULL, i); - synchronize_kernel(); - arch_destroy_sched_domains(); + detach_destroy_domains(&cpu_online_map); return NOTIFY_OK; case CPU_UP_CANCELED: @@ -5083,7 +5123,7 @@ static int update_sched_domains(struct notifier_block *nfb, } /* The hotplug lock is already held by cpu_up/cpu_down */ - arch_init_sched_domains(); + arch_init_sched_domains(&cpu_online_map); return NOTIFY_OK; } @@ -5092,7 +5132,7 @@ static int update_sched_domains(struct notifier_block *nfb, void __init sched_init_smp(void) { lock_cpu_hotplug(); - arch_init_sched_domains(); + arch_init_sched_domains(&cpu_online_map); unlock_cpu_hotplug(); /* XXX: Theoretical race here - CPU may be hotplugged now */ hotcpu_notifier(update_sched_domains, 0); -- cgit v1.2.3 From f8cbd99bd3a023db8d6356d19a5f6f539d367327 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sat, 25 Jun 2005 14:57:39 -0700 Subject: [PATCH] sched: voluntary kernel preemption This patch adds a new preemption model: 'Voluntary Kernel Preemption'. The 3 models can be selected from a new menu: (X) No Forced Preemption (Server) ( ) Voluntary Kernel Preemption (Desktop) ( ) Preemptible Kernel (Low-Latency Desktop) we still default to the stock (Server) preemption model. Voluntary preemption works by adding a cond_resched() (reschedule-if-needed) call to every might_sleep() check. It is lighter than CONFIG_PREEMPT - at the cost of not having as tight latencies. It represents a different latency/complexity/overhead tradeoff. It has no runtime impact at all if disabled. Here are size stats that show how the various preemption models impact the kernel's size: text data bss dec hex filename 3618774 547184 179896 4345854 424ffe vmlinux.stock 3626406 547184 179896 4353486 426dce vmlinux.voluntary +0.2% 3748414 548640 179896 4476950 445016 vmlinux.preempt +3.5% voluntary-preempt is +0.2% of .text, preempt is +3.5%. This feature has been tested for many months by lots of people (and it's also included in the RHEL4 distribution and earlier variants were in Fedora as well), and it's intended for users and distributions who dont want to use full-blown CONFIG_PREEMPT for one reason or another. Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 18 +++++++++++----- kernel/Kconfig.preempt | 57 +++++++++++++++++++++++++++++++++++++++++++------- 2 files changed, 62 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index e25b97062ce1..687ba8c9973d 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -58,15 +58,23 @@ struct completion; * be biten later when the calling function happens to sleep when it is not * supposed to. */ +#ifdef CONFIG_PREEMPT_VOLUNTARY +extern int cond_resched(void); +# define might_resched() cond_resched() +#else +# define might_resched() do { } while (0) +#endif + #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP -#define might_sleep() __might_sleep(__FILE__, __LINE__) -#define might_sleep_if(cond) do { if (unlikely(cond)) might_sleep(); } while (0) -void __might_sleep(char *file, int line); + void __might_sleep(char *file, int line); +# define might_sleep() \ + do { __might_sleep(__FILE__, __LINE__); might_resched(); } while (0) #else -#define might_sleep() do {} while(0) -#define might_sleep_if(cond) do {} while (0) +# define might_sleep() do { might_resched(); } while (0) #endif +#define might_sleep_if(cond) do { if (unlikely(cond)) might_sleep(); } while (0) + #define abs(x) ({ \ int __x = (x); \ (__x < 0) ? -__x : __x; \ diff --git a/kernel/Kconfig.preempt b/kernel/Kconfig.preempt index 34c631221aa3..0b46a5dff4c0 100644 --- a/kernel/Kconfig.preempt +++ b/kernel/Kconfig.preempt @@ -1,15 +1,56 @@ -config PREEMPT - bool "Preemptible Kernel" +choice + prompt "Preemption Model" + default PREEMPT_NONE + +config PREEMPT_NONE + bool "No Forced Preemption (Server)" + help + This is the traditional Linux preemption model, geared towards + throughput. It will still provide good latencies most of the + time, but there are no guarantees and occasional longer delays + are possible. + + Select this option if you are building a kernel for a server or + scientific/computation system, or if you want to maximize the + raw processing power of the kernel, irrespective of scheduling + latencies. + +config PREEMPT_VOLUNTARY + bool "Voluntary Kernel Preemption (Desktop)" help - This option reduces the latency of the kernel when reacting to - real-time or interactive events by allowing a low priority process to - be preempted even if it is in kernel mode executing a system call. - This allows applications to run more reliably even when the system is + This option reduces the latency of the kernel by adding more + "explicit preemption points" to the kernel code. These new + preemption points have been selected to reduce the maximum + latency of rescheduling, providing faster application reactions, + at the cost of slighly lower throughput. + + This allows reaction to interactive events by allowing a + low priority process to voluntarily preempt itself even if it + is in kernel mode executing a system call. This allows + applications to run more 'smoothly' even when the system is under load. - Say Y here if you are building a kernel for a desktop, embedded - or real-time system. Say N if you are unsure. + Select this if you are building a kernel for a desktop system. + +config PREEMPT + bool "Preemptible Kernel (Low-Latency Desktop)" + help + This option reduces the latency of the kernel by making + all kernel code (that is not executing in a critical section) + preemptible. This allows reaction to interactive events by + permitting a low priority process to be preempted involuntarily + even if it is in kernel mode executing a system call and would + otherwise not be about to reach a natural preemption point. + This allows applications to run more 'smoothly' even when the + system is under load, at the cost of slighly lower throughput + and a slight runtime overhead to kernel code. + + Select this if you are building a kernel for a desktop or + embedded system with latency requirements in the milliseconds + range. + +endchoice config PREEMPT_BKL bool "Preempt The Big Kernel Lock" -- cgit v1.2.3 From dc009d92435f99498cbc579ce76bf28e837e2c14 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sat, 25 Jun 2005 14:57:52 -0700 Subject: [PATCH] kexec: add kexec syscalls This patch introduces the architecture independent implementation the sys_kexec_load, the compat_sys_kexec_load system calls. Kexec on panic support has been integrated into the core patch and is relatively clean. In addition the hopefully architecture independent option crashkernel=size@location has been docuemented. It's purpose is to reserve space for the panic kernel to live, and where no DMA transfer will ever be setup to access. Signed-off-by: Eric Biederman Signed-off-by: Alexander Nyberg Signed-off-by: Adrian Bunk Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 4 + MAINTAINERS | 10 + include/linux/kexec.h | 127 +++++ include/linux/reboot.h | 3 + include/linux/syscalls.h | 5 +- kernel/Makefile | 1 + kernel/kexec.c | 1036 +++++++++++++++++++++++++++++++++++ kernel/panic.c | 23 +- kernel/sys.c | 20 + kernel/sys_ni.c | 2 + 10 files changed, 1227 insertions(+), 4 deletions(-) create mode 100644 include/linux/kexec.h create mode 100644 kernel/kexec.c (limited to 'include/linux') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 86db43fd6b0f..560ff5ae3fd9 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -358,6 +358,10 @@ running once the system is up. cpia_pp= [HW,PPT] Format: { parport | auto | none } + crashkernel=nn[KMG]@ss[KMG] + [KNL] Reserve a chunk of physical memory to + hold a kernel to switch to with kexec on panic. + cs4232= [HW,OSS] Format: ,,,,, diff --git a/MAINTAINERS b/MAINTAINERS index dbdd8494b2e6..81728572799e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1330,6 +1330,16 @@ M: rml@novell.com L: linux-kernel@vger.kernel.org S: Maintained +KEXEC +P: Eric Biederman +P: Randy Dunlap +M: ebiederm@xmission.com +M: rddunlap@osdl.org +W: http://www.xmission.com/~ebiederm/files/kexec/ +L: linux-kernel@vger.kernel.org +L: fastboot@osdl.org +S: Maintained + LANMEDIA WAN CARD DRIVER P: Andrew Stanley-Jones M: asj@lanmedia.com diff --git a/include/linux/kexec.h b/include/linux/kexec.h new file mode 100644 index 000000000000..e3fc35f4e35f --- /dev/null +++ b/include/linux/kexec.h @@ -0,0 +1,127 @@ +#ifndef LINUX_KEXEC_H +#define LINUX_KEXEC_H + +#ifdef CONFIG_KEXEC +#include +#include +#include +#include +#include + +/* Verify architecture specific macros are defined */ + +#ifndef KEXEC_SOURCE_MEMORY_LIMIT +#error KEXEC_SOURCE_MEMORY_LIMIT not defined +#endif + +#ifndef KEXEC_DESTINATION_MEMORY_LIMIT +#error KEXEC_DESTINATION_MEMORY_LIMIT not defined +#endif + +#ifndef KEXEC_CONTROL_MEMORY_LIMIT +#error KEXEC_CONTROL_MEMORY_LIMIT not defined +#endif + +#ifndef KEXEC_CONTROL_CODE_SIZE +#error KEXEC_CONTROL_CODE_SIZE not defined +#endif + +#ifndef KEXEC_ARCH +#error KEXEC_ARCH not defined +#endif + +/* + * This structure is used to hold the arguments that are used when loading + * kernel binaries. + */ + +typedef unsigned long kimage_entry_t; +#define IND_DESTINATION 0x1 +#define IND_INDIRECTION 0x2 +#define IND_DONE 0x4 +#define IND_SOURCE 0x8 + +#define KEXEC_SEGMENT_MAX 8 +struct kexec_segment { + void __user *buf; + size_t bufsz; + unsigned long mem; /* User space sees this as a (void *) ... */ + size_t memsz; +}; + +#ifdef CONFIG_COMPAT +struct compat_kexec_segment { + compat_uptr_t buf; + compat_size_t bufsz; + compat_ulong_t mem; /* User space sees this as a (void *) ... */ + compat_size_t memsz; +}; +#endif + +struct kimage { + kimage_entry_t head; + kimage_entry_t *entry; + kimage_entry_t *last_entry; + + unsigned long destination; + + unsigned long start; + struct page *control_code_page; + + unsigned long nr_segments; + struct kexec_segment segment[KEXEC_SEGMENT_MAX]; + + struct list_head control_pages; + struct list_head dest_pages; + struct list_head unuseable_pages; + + /* Address of next control page to allocate for crash kernels. */ + unsigned long control_page; + + /* Flags to indicate special processing */ + unsigned int type : 1; +#define KEXEC_TYPE_DEFAULT 0 +#define KEXEC_TYPE_CRASH 1 +}; + + + +/* kexec interface functions */ +extern NORET_TYPE void machine_kexec(struct kimage *image) ATTRIB_NORET; +extern int machine_kexec_prepare(struct kimage *image); +extern void machine_kexec_cleanup(struct kimage *image); +extern asmlinkage long sys_kexec_load(unsigned long entry, + unsigned long nr_segments, struct kexec_segment __user *segments, + unsigned long flags); +#ifdef CONFIG_COMPAT +extern asmlinkage long compat_sys_kexec_load(unsigned long entry, + unsigned long nr_segments, struct compat_kexec_segment __user *segments, + unsigned long flags); +#endif +extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order); +extern void crash_kexec(void); +extern struct kimage *kexec_image; + +#define KEXEC_ON_CRASH 0x00000001 +#define KEXEC_ARCH_MASK 0xffff0000 + +/* These values match the ELF architecture values. + * Unless there is a good reason that should continue to be the case. + */ +#define KEXEC_ARCH_DEFAULT ( 0 << 16) +#define KEXEC_ARCH_386 ( 3 << 16) +#define KEXEC_ARCH_X86_64 (62 << 16) +#define KEXEC_ARCH_PPC (20 << 16) +#define KEXEC_ARCH_PPC64 (21 << 16) +#define KEXEC_ARCH_IA_64 (50 << 16) + +#define KEXEC_FLAGS (KEXEC_ON_CRASH) /* List of defined/legal kexec flags */ + +/* Location of a reserved region to hold the crash kernel. + */ +extern struct resource crashk_res; + +#else /* !CONFIG_KEXEC */ +static inline void crash_kexec(void) { } +#endif /* CONFIG_KEXEC */ +#endif /* LINUX_KEXEC_H */ diff --git a/include/linux/reboot.h b/include/linux/reboot.h index d60fafc8bdc5..c5a05e16edb2 100644 --- a/include/linux/reboot.h +++ b/include/linux/reboot.h @@ -51,6 +51,9 @@ extern void machine_restart(char *cmd); extern void machine_halt(void); extern void machine_power_off(void); +extern void machine_shutdown(void); +extern void machine_crash_shutdown(void); + #endif #endif /* _LINUX_REBOOT_H */ diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index c39f6f72cbbc..7ba8f8f747aa 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -159,8 +159,9 @@ asmlinkage long sys_shutdown(int, int); asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user *arg); asmlinkage long sys_restart_syscall(void); -asmlinkage long sys_kexec_load(void *entry, unsigned long nr_segments, - struct kexec_segment *segments, unsigned long flags); +asmlinkage long sys_kexec_load(unsigned long entry, + unsigned long nr_segments, struct kexec_segment __user *segments, + unsigned long flags); asmlinkage long sys_exit(int error_code); asmlinkage void sys_exit_group(int error_code); diff --git a/kernel/Makefile b/kernel/Makefile index b01d26fe8db7..cfc8b0dea950 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -17,6 +17,7 @@ obj-$(CONFIG_MODULES) += module.o obj-$(CONFIG_KALLSYMS) += kallsyms.o obj-$(CONFIG_PM) += power/ obj-$(CONFIG_BSD_PROCESS_ACCT) += acct.o +obj-$(CONFIG_KEXEC) += kexec.o obj-$(CONFIG_COMPAT) += compat.o obj-$(CONFIG_CPUSETS) += cpuset.o obj-$(CONFIG_IKCONFIG) += configs.o diff --git a/kernel/kexec.c b/kernel/kexec.c new file mode 100644 index 000000000000..def9c73ec9a6 --- /dev/null +++ b/kernel/kexec.c @@ -0,0 +1,1036 @@ +/* + * kexec.c - kexec system call + * Copyright (C) 2002-2004 Eric Biederman + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* Location of the reserved area for the crash kernel */ +struct resource crashk_res = { + .name = "Crash kernel", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_MEM +}; + +/* + * When kexec transitions to the new kernel there is a one-to-one + * mapping between physical and virtual addresses. On processors + * where you can disable the MMU this is trivial, and easy. For + * others it is still a simple predictable page table to setup. + * + * In that environment kexec copies the new kernel to its final + * resting place. This means I can only support memory whose + * physical address can fit in an unsigned long. In particular + * addresses where (pfn << PAGE_SHIFT) > ULONG_MAX cannot be handled. + * If the assembly stub has more restrictive requirements + * KEXEC_SOURCE_MEMORY_LIMIT and KEXEC_DEST_MEMORY_LIMIT can be + * defined more restrictively in . + * + * The code for the transition from the current kernel to the + * the new kernel is placed in the control_code_buffer, whose size + * is given by KEXEC_CONTROL_CODE_SIZE. In the best case only a single + * page of memory is necessary, but some architectures require more. + * Because this memory must be identity mapped in the transition from + * virtual to physical addresses it must live in the range + * 0 - TASK_SIZE, as only the user space mappings are arbitrarily + * modifiable. + * + * The assembly stub in the control code buffer is passed a linked list + * of descriptor pages detailing the source pages of the new kernel, + * and the destination addresses of those source pages. As this data + * structure is not used in the context of the current OS, it must + * be self-contained. + * + * The code has been made to work with highmem pages and will use a + * destination page in its final resting place (if it happens + * to allocate it). The end product of this is that most of the + * physical address space, and most of RAM can be used. + * + * Future directions include: + * - allocating a page table with the control code buffer identity + * mapped, to simplify machine_kexec and make kexec_on_panic more + * reliable. + */ + +/* + * KIMAGE_NO_DEST is an impossible destination address..., for + * allocating pages whose destination address we do not care about. + */ +#define KIMAGE_NO_DEST (-1UL) + +static int kimage_is_destination_range( + struct kimage *image, unsigned long start, unsigned long end); +static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mask, unsigned long dest); + +static int do_kimage_alloc(struct kimage **rimage, unsigned long entry, + unsigned long nr_segments, struct kexec_segment __user *segments) +{ + size_t segment_bytes; + struct kimage *image; + unsigned long i; + int result; + + /* Allocate a controlling structure */ + result = -ENOMEM; + image = kmalloc(sizeof(*image), GFP_KERNEL); + if (!image) { + goto out; + } + memset(image, 0, sizeof(*image)); + image->head = 0; + image->entry = &image->head; + image->last_entry = &image->head; + image->control_page = ~0; /* By default this does not apply */ + image->start = entry; + image->type = KEXEC_TYPE_DEFAULT; + + /* Initialize the list of control pages */ + INIT_LIST_HEAD(&image->control_pages); + + /* Initialize the list of destination pages */ + INIT_LIST_HEAD(&image->dest_pages); + + /* Initialize the list of unuseable pages */ + INIT_LIST_HEAD(&image->unuseable_pages); + + /* Read in the segments */ + image->nr_segments = nr_segments; + segment_bytes = nr_segments * sizeof(*segments); + result = copy_from_user(image->segment, segments, segment_bytes); + if (result) + goto out; + + /* + * Verify we have good destination addresses. The caller is + * responsible for making certain we don't attempt to load + * the new image into invalid or reserved areas of RAM. This + * just verifies it is an address we can use. + * + * Since the kernel does everything in page size chunks ensure + * the destination addreses are page aligned. Too many + * special cases crop of when we don't do this. The most + * insidious is getting overlapping destination addresses + * simply because addresses are changed to page size + * granularity. + */ + result = -EADDRNOTAVAIL; + for (i = 0; i < nr_segments; i++) { + unsigned long mstart, mend; + mstart = image->segment[i].mem; + mend = mstart + image->segment[i].memsz; + if ((mstart & ~PAGE_MASK) || (mend & ~PAGE_MASK)) + goto out; + if (mend >= KEXEC_DESTINATION_MEMORY_LIMIT) + goto out; + } + + /* Verify our destination addresses do not overlap. + * If we alloed overlapping destination addresses + * through very weird things can happen with no + * easy explanation as one segment stops on another. + */ + result = -EINVAL; + for(i = 0; i < nr_segments; i++) { + unsigned long mstart, mend; + unsigned long j; + mstart = image->segment[i].mem; + mend = mstart + image->segment[i].memsz; + for(j = 0; j < i; j++) { + unsigned long pstart, pend; + pstart = image->segment[j].mem; + pend = pstart + image->segment[j].memsz; + /* Do the segments overlap ? */ + if ((mend > pstart) && (mstart < pend)) + goto out; + } + } + + /* Ensure our buffer sizes are strictly less than + * our memory sizes. This should always be the case, + * and it is easier to check up front than to be surprised + * later on. + */ + result = -EINVAL; + for(i = 0; i < nr_segments; i++) { + if (image->segment[i].bufsz > image->segment[i].memsz) + goto out; + } + + + result = 0; + out: + if (result == 0) { + *rimage = image; + } else { + kfree(image); + } + return result; + +} + +static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry, + unsigned long nr_segments, struct kexec_segment __user *segments) +{ + int result; + struct kimage *image; + + /* Allocate and initialize a controlling structure */ + image = NULL; + result = do_kimage_alloc(&image, entry, nr_segments, segments); + if (result) { + goto out; + } + *rimage = image; + + /* + * Find a location for the control code buffer, and add it + * the vector of segments so that it's pages will also be + * counted as destination pages. + */ + result = -ENOMEM; + image->control_code_page = kimage_alloc_control_pages(image, + get_order(KEXEC_CONTROL_CODE_SIZE)); + if (!image->control_code_page) { + printk(KERN_ERR "Could not allocate control_code_buffer\n"); + goto out; + } + + result = 0; + out: + if (result == 0) { + *rimage = image; + } else { + kfree(image); + } + return result; +} + +static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry, + unsigned long nr_segments, struct kexec_segment *segments) +{ + int result; + struct kimage *image; + unsigned long i; + + image = NULL; + /* Verify we have a valid entry point */ + if ((entry < crashk_res.start) || (entry > crashk_res.end)) { + result = -EADDRNOTAVAIL; + goto out; + } + + /* Allocate and initialize a controlling structure */ + result = do_kimage_alloc(&image, entry, nr_segments, segments); + if (result) { + goto out; + } + + /* Enable the special crash kernel control page + * allocation policy. + */ + image->control_page = crashk_res.start; + image->type = KEXEC_TYPE_CRASH; + + /* + * Verify we have good destination addresses. Normally + * the caller is responsible for making certain we don't + * attempt to load the new image into invalid or reserved + * areas of RAM. But crash kernels are preloaded into a + * reserved area of ram. We must ensure the addresses + * are in the reserved area otherwise preloading the + * kernel could corrupt things. + */ + result = -EADDRNOTAVAIL; + for (i = 0; i < nr_segments; i++) { + unsigned long mstart, mend; + mstart = image->segment[i].mem; + mend = mstart + image->segment[i].memsz; + /* Ensure we are within the crash kernel limits */ + if ((mstart < crashk_res.start) || (mend > crashk_res.end)) + goto out; + } + + + /* + * Find a location for the control code buffer, and add + * the vector of segments so that it's pages will also be + * counted as destination pages. + */ + result = -ENOMEM; + image->control_code_page = kimage_alloc_control_pages(image, + get_order(KEXEC_CONTROL_CODE_SIZE)); + if (!image->control_code_page) { + printk(KERN_ERR "Could not allocate control_code_buffer\n"); + goto out; + } + + result = 0; + out: + if (result == 0) { + *rimage = image; + } else { + kfree(image); + } + return result; +} + +static int kimage_is_destination_range( + struct kimage *image, unsigned long start, unsigned long end) +{ + unsigned long i; + + for (i = 0; i < image->nr_segments; i++) { + unsigned long mstart, mend; + mstart = image->segment[i].mem; + mend = mstart + image->segment[i].memsz; + if ((end > mstart) && (start < mend)) { + return 1; + } + } + return 0; +} + +static struct page *kimage_alloc_pages(unsigned int gfp_mask, unsigned int order) +{ + struct page *pages; + pages = alloc_pages(gfp_mask, order); + if (pages) { + unsigned int count, i; + pages->mapping = NULL; + pages->private = order; + count = 1 << order; + for(i = 0; i < count; i++) { + SetPageReserved(pages + i); + } + } + return pages; +} + +static void kimage_free_pages(struct page *page) +{ + unsigned int order, count, i; + order = page->private; + count = 1 << order; + for(i = 0; i < count; i++) { + ClearPageReserved(page + i); + } + __free_pages(page, order); +} + +static void kimage_free_page_list(struct list_head *list) +{ + struct list_head *pos, *next; + list_for_each_safe(pos, next, list) { + struct page *page; + + page = list_entry(pos, struct page, lru); + list_del(&page->lru); + + kimage_free_pages(page); + } +} + +static struct page *kimage_alloc_normal_control_pages( + struct kimage *image, unsigned int order) +{ + /* Control pages are special, they are the intermediaries + * that are needed while we copy the rest of the pages + * to their final resting place. As such they must + * not conflict with either the destination addresses + * or memory the kernel is already using. + * + * The only case where we really need more than one of + * these are for architectures where we cannot disable + * the MMU and must instead generate an identity mapped + * page table for all of the memory. + * + * At worst this runs in O(N) of the image size. + */ + struct list_head extra_pages; + struct page *pages; + unsigned int count; + + count = 1 << order; + INIT_LIST_HEAD(&extra_pages); + + /* Loop while I can allocate a page and the page allocated + * is a destination page. + */ + do { + unsigned long pfn, epfn, addr, eaddr; + pages = kimage_alloc_pages(GFP_KERNEL, order); + if (!pages) + break; + pfn = page_to_pfn(pages); + epfn = pfn + count; + addr = pfn << PAGE_SHIFT; + eaddr = epfn << PAGE_SHIFT; + if ((epfn >= (KEXEC_CONTROL_MEMORY_LIMIT >> PAGE_SHIFT)) || + kimage_is_destination_range(image, addr, eaddr)) + { + list_add(&pages->lru, &extra_pages); + pages = NULL; + } + } while(!pages); + if (pages) { + /* Remember the allocated page... */ + list_add(&pages->lru, &image->control_pages); + + /* Because the page is already in it's destination + * location we will never allocate another page at + * that address. Therefore kimage_alloc_pages + * will not return it (again) and we don't need + * to give it an entry in image->segment[]. + */ + } + /* Deal with the destination pages I have inadvertently allocated. + * + * Ideally I would convert multi-page allocations into single + * page allocations, and add everyting to image->dest_pages. + * + * For now it is simpler to just free the pages. + */ + kimage_free_page_list(&extra_pages); + return pages; + +} + +static struct page *kimage_alloc_crash_control_pages( + struct kimage *image, unsigned int order) +{ + /* Control pages are special, they are the intermediaries + * that are needed while we copy the rest of the pages + * to their final resting place. As such they must + * not conflict with either the destination addresses + * or memory the kernel is already using. + * + * Control pages are also the only pags we must allocate + * when loading a crash kernel. All of the other pages + * are specified by the segments and we just memcpy + * into them directly. + * + * The only case where we really need more than one of + * these are for architectures where we cannot disable + * the MMU and must instead generate an identity mapped + * page table for all of the memory. + * + * Given the low demand this implements a very simple + * allocator that finds the first hole of the appropriate + * size in the reserved memory region, and allocates all + * of the memory up to and including the hole. + */ + unsigned long hole_start, hole_end, size; + struct page *pages; + pages = NULL; + size = (1 << order) << PAGE_SHIFT; + hole_start = (image->control_page + (size - 1)) & ~(size - 1); + hole_end = hole_start + size - 1; + while(hole_end <= crashk_res.end) { + unsigned long i; + if (hole_end > KEXEC_CONTROL_MEMORY_LIMIT) { + break; + } + if (hole_end > crashk_res.end) { + break; + } + /* See if I overlap any of the segments */ + for(i = 0; i < image->nr_segments; i++) { + unsigned long mstart, mend; + mstart = image->segment[i].mem; + mend = mstart + image->segment[i].memsz - 1; + if ((hole_end >= mstart) && (hole_start <= mend)) { + /* Advance the hole to the end of the segment */ + hole_start = (mend + (size - 1)) & ~(size - 1); + hole_end = hole_start + size - 1; + break; + } + } + /* If I don't overlap any segments I have found my hole! */ + if (i == image->nr_segments) { + pages = pfn_to_page(hole_start >> PAGE_SHIFT); + break; + } + } + if (pages) { + image->control_page = hole_end; + } + return pages; +} + + +struct page *kimage_alloc_control_pages( + struct kimage *image, unsigned int order) +{ + struct page *pages = NULL; + switch(image->type) { + case KEXEC_TYPE_DEFAULT: + pages = kimage_alloc_normal_control_pages(image, order); + break; + case KEXEC_TYPE_CRASH: + pages = kimage_alloc_crash_control_pages(image, order); + break; + } + return pages; +} + +static int kimage_add_entry(struct kimage *image, kimage_entry_t entry) +{ + if (*image->entry != 0) { + image->entry++; + } + if (image->entry == image->last_entry) { + kimage_entry_t *ind_page; + struct page *page; + page = kimage_alloc_page(image, GFP_KERNEL, KIMAGE_NO_DEST); + if (!page) { + return -ENOMEM; + } + ind_page = page_address(page); + *image->entry = virt_to_phys(ind_page) | IND_INDIRECTION; + image->entry = ind_page; + image->last_entry = + ind_page + ((PAGE_SIZE/sizeof(kimage_entry_t)) - 1); + } + *image->entry = entry; + image->entry++; + *image->entry = 0; + return 0; +} + +static int kimage_set_destination( + struct kimage *image, unsigned long destination) +{ + int result; + + destination &= PAGE_MASK; + result = kimage_add_entry(image, destination | IND_DESTINATION); + if (result == 0) { + image->destination = destination; + } + return result; +} + + +static int kimage_add_page(struct kimage *image, unsigned long page) +{ + int result; + + page &= PAGE_MASK; + result = kimage_add_entry(image, page | IND_SOURCE); + if (result == 0) { + image->destination += PAGE_SIZE; + } + return result; +} + + +static void kimage_free_extra_pages(struct kimage *image) +{ + /* Walk through and free any extra destination pages I may have */ + kimage_free_page_list(&image->dest_pages); + + /* Walk through and free any unuseable pages I have cached */ + kimage_free_page_list(&image->unuseable_pages); + +} +static int kimage_terminate(struct kimage *image) +{ + if (*image->entry != 0) { + image->entry++; + } + *image->entry = IND_DONE; + return 0; +} + +#define for_each_kimage_entry(image, ptr, entry) \ + for (ptr = &image->head; (entry = *ptr) && !(entry & IND_DONE); \ + ptr = (entry & IND_INDIRECTION)? \ + phys_to_virt((entry & PAGE_MASK)): ptr +1) + +static void kimage_free_entry(kimage_entry_t entry) +{ + struct page *page; + + page = pfn_to_page(entry >> PAGE_SHIFT); + kimage_free_pages(page); +} + +static void kimage_free(struct kimage *image) +{ + kimage_entry_t *ptr, entry; + kimage_entry_t ind = 0; + + if (!image) + return; + kimage_free_extra_pages(image); + for_each_kimage_entry(image, ptr, entry) { + if (entry & IND_INDIRECTION) { + /* Free the previous indirection page */ + if (ind & IND_INDIRECTION) { + kimage_free_entry(ind); + } + /* Save this indirection page until we are + * done with it. + */ + ind = entry; + } + else if (entry & IND_SOURCE) { + kimage_free_entry(entry); + } + } + /* Free the final indirection page */ + if (ind & IND_INDIRECTION) { + kimage_free_entry(ind); + } + + /* Handle any machine specific cleanup */ + machine_kexec_cleanup(image); + + /* Free the kexec control pages... */ + kimage_free_page_list(&image->control_pages); + kfree(image); +} + +static kimage_entry_t *kimage_dst_used(struct kimage *image, unsigned long page) +{ + kimage_entry_t *ptr, entry; + unsigned long destination = 0; + + for_each_kimage_entry(image, ptr, entry) { + if (entry & IND_DESTINATION) { + destination = entry & PAGE_MASK; + } + else if (entry & IND_SOURCE) { + if (page == destination) { + return ptr; + } + destination += PAGE_SIZE; + } + } + return 0; +} + +static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mask, unsigned long destination) +{ + /* + * Here we implement safeguards to ensure that a source page + * is not copied to its destination page before the data on + * the destination page is no longer useful. + * + * To do this we maintain the invariant that a source page is + * either its own destination page, or it is not a + * destination page at all. + * + * That is slightly stronger than required, but the proof + * that no problems will not occur is trivial, and the + * implementation is simply to verify. + * + * When allocating all pages normally this algorithm will run + * in O(N) time, but in the worst case it will run in O(N^2) + * time. If the runtime is a problem the data structures can + * be fixed. + */ + struct page *page; + unsigned long addr; + + /* + * Walk through the list of destination pages, and see if I + * have a match. + */ + list_for_each_entry(page, &image->dest_pages, lru) { + addr = page_to_pfn(page) << PAGE_SHIFT; + if (addr == destination) { + list_del(&page->lru); + return page; + } + } + page = NULL; + while (1) { + kimage_entry_t *old; + + /* Allocate a page, if we run out of memory give up */ + page = kimage_alloc_pages(gfp_mask, 0); + if (!page) { + return 0; + } + /* If the page cannot be used file it away */ + if (page_to_pfn(page) > (KEXEC_SOURCE_MEMORY_LIMIT >> PAGE_SHIFT)) { + list_add(&page->lru, &image->unuseable_pages); + continue; + } + addr = page_to_pfn(page) << PAGE_SHIFT; + + /* If it is the destination page we want use it */ + if (addr == destination) + break; + + /* If the page is not a destination page use it */ + if (!kimage_is_destination_range(image, addr, addr + PAGE_SIZE)) + break; + + /* + * I know that the page is someones destination page. + * See if there is already a source page for this + * destination page. And if so swap the source pages. + */ + old = kimage_dst_used(image, addr); + if (old) { + /* If so move it */ + unsigned long old_addr; + struct page *old_page; + + old_addr = *old & PAGE_MASK; + old_page = pfn_to_page(old_addr >> PAGE_SHIFT); + copy_highpage(page, old_page); + *old = addr | (*old & ~PAGE_MASK); + + /* The old page I have found cannot be a + * destination page, so return it. + */ + addr = old_addr; + page = old_page; + break; + } + else { + /* Place the page on the destination list I + * will use it later. + */ + list_add(&page->lru, &image->dest_pages); + } + } + return page; +} + +static int kimage_load_normal_segment(struct kimage *image, + struct kexec_segment *segment) +{ + unsigned long maddr; + unsigned long ubytes, mbytes; + int result; + unsigned char *buf; + + result = 0; + buf = segment->buf; + ubytes = segment->bufsz; + mbytes = segment->memsz; + maddr = segment->mem; + + result = kimage_set_destination(image, maddr); + if (result < 0) { + goto out; + } + while(mbytes) { + struct page *page; + char *ptr; + size_t uchunk, mchunk; + page = kimage_alloc_page(image, GFP_HIGHUSER, maddr); + if (page == 0) { + result = -ENOMEM; + goto out; + } + result = kimage_add_page(image, page_to_pfn(page) << PAGE_SHIFT); + if (result < 0) { + goto out; + } + ptr = kmap(page); + /* Start with a clear page */ + memset(ptr, 0, PAGE_SIZE); + ptr += maddr & ~PAGE_MASK; + mchunk = PAGE_SIZE - (maddr & ~PAGE_MASK); + if (mchunk > mbytes) { + mchunk = mbytes; + } + uchunk = mchunk; + if (uchunk > ubytes) { + uchunk = ubytes; + } + result = copy_from_user(ptr, buf, uchunk); + kunmap(page); + if (result) { + result = (result < 0) ? result : -EIO; + goto out; + } + ubytes -= uchunk; + maddr += mchunk; + buf += mchunk; + mbytes -= mchunk; + } + out: + return result; +} + +static int kimage_load_crash_segment(struct kimage *image, + struct kexec_segment *segment) +{ + /* For crash dumps kernels we simply copy the data from + * user space to it's destination. + * We do things a page at a time for the sake of kmap. + */ + unsigned long maddr; + unsigned long ubytes, mbytes; + int result; + unsigned char *buf; + + result = 0; + buf = segment->buf; + ubytes = segment->bufsz; + mbytes = segment->memsz; + maddr = segment->mem; + while(mbytes) { + struct page *page; + char *ptr; + size_t uchunk, mchunk; + page = pfn_to_page(maddr >> PAGE_SHIFT); + if (page == 0) { + result = -ENOMEM; + goto out; + } + ptr = kmap(page); + ptr += maddr & ~PAGE_MASK; + mchunk = PAGE_SIZE - (maddr & ~PAGE_MASK); + if (mchunk > mbytes) { + mchunk = mbytes; + } + uchunk = mchunk; + if (uchunk > ubytes) { + uchunk = ubytes; + /* Zero the trailing part of the page */ + memset(ptr + uchunk, 0, mchunk - uchunk); + } + result = copy_from_user(ptr, buf, uchunk); + kunmap(page); + if (result) { + result = (result < 0) ? result : -EIO; + goto out; + } + ubytes -= uchunk; + maddr += mchunk; + buf += mchunk; + mbytes -= mchunk; + } + out: + return result; +} + +static int kimage_load_segment(struct kimage *image, + struct kexec_segment *segment) +{ + int result = -ENOMEM; + switch(image->type) { + case KEXEC_TYPE_DEFAULT: + result = kimage_load_normal_segment(image, segment); + break; + case KEXEC_TYPE_CRASH: + result = kimage_load_crash_segment(image, segment); + break; + } + return result; +} + +/* + * Exec Kernel system call: for obvious reasons only root may call it. + * + * This call breaks up into three pieces. + * - A generic part which loads the new kernel from the current + * address space, and very carefully places the data in the + * allocated pages. + * + * - A generic part that interacts with the kernel and tells all of + * the devices to shut down. Preventing on-going dmas, and placing + * the devices in a consistent state so a later kernel can + * reinitialize them. + * + * - A machine specific part that includes the syscall number + * and the copies the image to it's final destination. And + * jumps into the image at entry. + * + * kexec does not sync, or unmount filesystems so if you need + * that to happen you need to do that yourself. + */ +struct kimage *kexec_image = NULL; +static struct kimage *kexec_crash_image = NULL; +/* + * A home grown binary mutex. + * Nothing can wait so this mutex is safe to use + * in interrupt context :) + */ +static int kexec_lock = 0; + +asmlinkage long sys_kexec_load(unsigned long entry, + unsigned long nr_segments, struct kexec_segment __user *segments, + unsigned long flags) +{ + struct kimage **dest_image, *image; + int locked; + int result; + + /* We only trust the superuser with rebooting the system. */ + if (!capable(CAP_SYS_BOOT)) + return -EPERM; + + /* + * Verify we have a legal set of flags + * This leaves us room for future extensions. + */ + if ((flags & KEXEC_FLAGS) != (flags & ~KEXEC_ARCH_MASK)) + return -EINVAL; + + /* Verify we are on the appropriate architecture */ + if (((flags & KEXEC_ARCH_MASK) != KEXEC_ARCH) && + ((flags & KEXEC_ARCH_MASK) != KEXEC_ARCH_DEFAULT)) + { + return -EINVAL; + } + + /* Put an artificial cap on the number + * of segments passed to kexec_load. + */ + if (nr_segments > KEXEC_SEGMENT_MAX) + return -EINVAL; + + image = NULL; + result = 0; + + /* Because we write directly to the reserved memory + * region when loading crash kernels we need a mutex here to + * prevent multiple crash kernels from attempting to load + * simultaneously, and to prevent a crash kernel from loading + * over the top of a in use crash kernel. + * + * KISS: always take the mutex. + */ + locked = xchg(&kexec_lock, 1); + if (locked) { + return -EBUSY; + } + dest_image = &kexec_image; + if (flags & KEXEC_ON_CRASH) { + dest_image = &kexec_crash_image; + } + if (nr_segments > 0) { + unsigned long i; + /* Loading another kernel to reboot into */ + if ((flags & KEXEC_ON_CRASH) == 0) { + result = kimage_normal_alloc(&image, entry, nr_segments, segments); + } + /* Loading another kernel to switch to if this one crashes */ + else if (flags & KEXEC_ON_CRASH) { + /* Free any current crash dump kernel before + * we corrupt it. + */ + kimage_free(xchg(&kexec_crash_image, NULL)); + result = kimage_crash_alloc(&image, entry, nr_segments, segments); + } + if (result) { + goto out; + } + result = machine_kexec_prepare(image); + if (result) { + goto out; + } + for(i = 0; i < nr_segments; i++) { + result = kimage_load_segment(image, &image->segment[i]); + if (result) { + goto out; + } + } + result = kimage_terminate(image); + if (result) { + goto out; + } + } + /* Install the new kernel, and Uninstall the old */ + image = xchg(dest_image, image); + + out: + xchg(&kexec_lock, 0); /* Release the mutex */ + kimage_free(image); + return result; +} + +#ifdef CONFIG_COMPAT +asmlinkage long compat_sys_kexec_load(unsigned long entry, + unsigned long nr_segments, struct compat_kexec_segment __user *segments, + unsigned long flags) +{ + struct compat_kexec_segment in; + struct kexec_segment out, __user *ksegments; + unsigned long i, result; + + /* Don't allow clients that don't understand the native + * architecture to do anything. + */ + if ((flags & KEXEC_ARCH_MASK) == KEXEC_ARCH_DEFAULT) { + return -EINVAL; + } + + if (nr_segments > KEXEC_SEGMENT_MAX) { + return -EINVAL; + } + + ksegments = compat_alloc_user_space(nr_segments * sizeof(out)); + for (i=0; i < nr_segments; i++) { + result = copy_from_user(&in, &segments[i], sizeof(in)); + if (result) { + return -EFAULT; + } + + out.buf = compat_ptr(in.buf); + out.bufsz = in.bufsz; + out.mem = in.mem; + out.memsz = in.memsz; + + result = copy_to_user(&ksegments[i], &out, sizeof(out)); + if (result) { + return -EFAULT; + } + } + + return sys_kexec_load(entry, nr_segments, ksegments, flags); +} +#endif + +void crash_kexec(void) +{ + struct kimage *image; + int locked; + + + /* Take the kexec_lock here to prevent sys_kexec_load + * running on one cpu from replacing the crash kernel + * we are using after a panic on a different cpu. + * + * If the crash kernel was not located in a fixed area + * of memory the xchg(&kexec_crash_image) would be + * sufficient. But since I reuse the memory... + */ + locked = xchg(&kexec_lock, 1); + if (!locked) { + image = xchg(&kexec_crash_image, NULL); + if (image) { + machine_crash_shutdown(); + machine_kexec(image); + } + xchg(&kexec_lock, 0); + } +} diff --git a/kernel/panic.c b/kernel/panic.c index 081f7465fc8d..66f43d33cd80 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -18,6 +18,7 @@ #include #include #include +#include int panic_timeout; int panic_on_oops; @@ -63,6 +64,13 @@ NORET_TYPE void panic(const char * fmt, ...) unsigned long caller = (unsigned long) __builtin_return_address(0); #endif + /* + * It's possible to come here directly from a panic-assertion and not + * have preempt disabled. Some functions called from here want + * preempt to be disabled. No point enabling it later though... + */ + preempt_disable(); + bust_spinlocks(1); va_start(args, fmt); vsnprintf(buf, sizeof(buf), fmt, args); @@ -70,7 +78,19 @@ NORET_TYPE void panic(const char * fmt, ...) printk(KERN_EMERG "Kernel panic - not syncing: %s\n",buf); bust_spinlocks(0); + /* + * If we have crashed and we have a crash kernel loaded let it handle + * everything else. + * Do we want to call this before we try to display a message? + */ + crash_kexec(); + #ifdef CONFIG_SMP + /* + * Note smp_send_stop is the usual smp shutdown function, which + * unfortunately means it may not be hardened to work in a panic + * situation. + */ smp_send_stop(); #endif @@ -79,8 +99,7 @@ NORET_TYPE void panic(const char * fmt, ...) if (!panic_blink) panic_blink = no_blink; - if (panic_timeout > 0) - { + if (panic_timeout > 0) { /* * Delay timeout seconds before rebooting the machine. * We can't use the "normal" timers since we just panicked.. diff --git a/kernel/sys.c b/kernel/sys.c index dac10161ca23..9a24374c23bc 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -16,6 +16,8 @@ #include #include #include +#include +#include #include #include #include @@ -439,6 +441,24 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user machine_restart(buffer); break; +#ifdef CONFIG_KEXEC + case LINUX_REBOOT_CMD_KEXEC: + { + struct kimage *image; + image = xchg(&kexec_image, 0); + if (!image) { + unlock_kernel(); + return -EINVAL; + } + notifier_call_chain(&reboot_notifier_list, SYS_RESTART, NULL); + system_state = SYSTEM_RESTART; + device_shutdown(); + printk(KERN_EMERG "Starting new kernel\n"); + machine_shutdown(); + machine_kexec(image); + break; + } +#endif #ifdef CONFIG_SOFTWARE_SUSPEND case LINUX_REBOOT_CMD_SW_SUSPEND: { diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 6f15bea7d1a8..29196ce9b40f 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -18,6 +18,8 @@ cond_syscall(sys_acct); cond_syscall(sys_lookup_dcookie); cond_syscall(sys_swapon); cond_syscall(sys_swapoff); +cond_syscall(sys_kexec_load); +cond_syscall(compat_sys_kexec_load); cond_syscall(sys_init_module); cond_syscall(sys_delete_module); cond_syscall(sys_socketpair); -- cgit v1.2.3 From cf13f0eaffa31bf6a145c53c589654b11c72ddc7 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Sat, 25 Jun 2005 14:58:11 -0700 Subject: [PATCH] kexec: s390 support Add kexec support for s390 architecture. From: Milton Miller - Fix passing of first argument to relocate_kernel assembly. - Fix Kconfig description. - Remove wrong comment and comments that describe obvious things. - Allow only KEXEC_TYPE_DEFAULT as image type -> dump not supported. Acked-by: Martin Schwidefsky Signed-off-by: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/Kconfig | 8 +++ arch/s390/kernel/Makefile | 10 ++++ arch/s390/kernel/compat_wrapper.S | 8 +++ arch/s390/kernel/crash.c | 17 +++++++ arch/s390/kernel/machine_kexec.c | 98 ++++++++++++++++++++++++++++++++++++ arch/s390/kernel/relocate_kernel.S | 81 +++++++++++++++++++++++++++++ arch/s390/kernel/relocate_kernel64.S | 82 ++++++++++++++++++++++++++++++ arch/s390/kernel/syscalls.S | 2 +- include/asm-s390/kexec.h | 42 ++++++++++++++++ include/asm-s390/unistd.h | 2 +- include/linux/kexec.h | 1 + 11 files changed, 349 insertions(+), 2 deletions(-) create mode 100644 arch/s390/kernel/crash.c create mode 100644 arch/s390/kernel/machine_kexec.c create mode 100644 arch/s390/kernel/relocate_kernel.S create mode 100644 arch/s390/kernel/relocate_kernel64.S create mode 100644 include/asm-s390/kexec.h (limited to 'include/linux') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 32696c1d9280..6600ee87f896 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -455,6 +455,14 @@ config NO_IDLE_HZ_INIT The HZ timer is switched off in idle by default. That means the HZ timer is already disabled at boot time. +config KEXEC + bool "kexec system call (EXPERIMENTAL)" + depends on EXPERIMENTAL + help + kexec is a system call that implements the ability to shutdown your + current kernel, and to start another kernel. It is like a reboot + but is independent of hardware/microcode support. + endmenu config PCMCIA diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index b41e0e199a7c..ab1e49d2e518 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -25,6 +25,16 @@ obj-$(CONFIG_ARCH_S390X) += entry64.o reipl64.o obj-$(CONFIG_VIRT_TIMER) += vtime.o +# Kexec part +S390_KEXEC_OBJS := machine_kexec.o crash.o +ifeq ($(CONFIG_ARCH_S390X),y) +S390_KEXEC_OBJS += relocate_kernel64.o +else +S390_KEXEC_OBJS += relocate_kernel.o +endif +obj-$(CONFIG_KEXEC) += $(S390_KEXEC_OBJS) + + # # This is just to get the dependencies... # diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index 7a607b1d0380..bf529739c8ab 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -1441,3 +1441,11 @@ compat_sys_waitid_wrapper: lgfr %r5,%r5 # int llgtr %r6,%r6 # struct rusage_emu31 * jg compat_sys_waitid + + .globl compat_sys_kexec_load_wrapper +compat_sys_kexec_load_wrapper: + llgfr %r2,%r2 # unsigned long + llgfr %r3,%r3 # unsigned long + llgtr %r4,%r4 # struct kexec_segment * + llgfr %r5,%r5 # unsigned long + jg compat_sys_kexec_load diff --git a/arch/s390/kernel/crash.c b/arch/s390/kernel/crash.c new file mode 100644 index 000000000000..db38283c1f27 --- /dev/null +++ b/arch/s390/kernel/crash.c @@ -0,0 +1,17 @@ +/* + * arch/s390/kernel/crash.c + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Heiko Carstens + * + */ + +#include +#include + +note_buf_t crash_notes[NR_CPUS]; + +void machine_crash_shutdown(void) +{ +} diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c new file mode 100644 index 000000000000..7a94db76df46 --- /dev/null +++ b/arch/s390/kernel/machine_kexec.c @@ -0,0 +1,98 @@ +/* + * arch/s390/kernel/machine_kexec.c + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger + * + */ + +/* + * s390_machine_kexec.c - handle the transition of Linux booting another kernel + * on the S390 architecture. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static void kexec_halt_all_cpus(void *); + +typedef void (*relocate_kernel_t) (kimage_entry_t *, unsigned long); + +const extern unsigned char relocate_kernel[]; +const extern unsigned long long relocate_kernel_len; + +int +machine_kexec_prepare(struct kimage *image) +{ + unsigned long reboot_code_buffer; + + /* We don't support anything but the default image type for now. */ + if (image->type != KEXEC_TYPE_DEFAULT) + return -EINVAL; + + /* Get the destination where the assembler code should be copied to.*/ + reboot_code_buffer = page_to_pfn(image->control_code_page)<control_code_page) << PAGE_SHIFT); + + /* Call the moving routine */ + (*data_mover) (&image->head, image->start); +} diff --git a/arch/s390/kernel/relocate_kernel.S b/arch/s390/kernel/relocate_kernel.S new file mode 100644 index 000000000000..d5e4a62fbb79 --- /dev/null +++ b/arch/s390/kernel/relocate_kernel.S @@ -0,0 +1,81 @@ +/* + * arch/s390/kernel/relocate_kernel.S + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger + * + */ + +/* + * moves the new kernel to its destination... + * %r2 = pointer to first kimage_entry_t + * %r3 = start address - where to jump to after the job is done... + * + * %r5 will be used as temp. storage + * %r6 holds the destination address + * %r7 = PAGE_SIZE + * %r8 holds the source address + * %r9 = PAGE_SIZE + * %r10 is a page mask + */ + + .text + .globl relocate_kernel + relocate_kernel: + basr %r13,0 #base address + .base: + spx zero64-.base(%r13) #absolute addressing mode + stnsm sys_msk-.base(%r13),0xf8 #disable DAT and IRQ (external) + lhi %r10,-1 #preparing the mask + sll %r10,12 #shift it such that it becomes 0xf000 + .top: + lhi %r7,4096 #load PAGE_SIZE in r7 + lhi %r9,4096 #load PAGE_SIZE in r9 + l %r5,0(%r2) #read another word for indirection page + ahi %r2,4 #increment pointer + tml %r5,0x1 #is it a destination page? + je .indir_check #NO, goto "indir_check" + lr %r6,%r5 #r6 = r5 + nr %r6,%r10 #mask it out and... + j .top #...next iteration + .indir_check: + tml %r5,0x2 #is it a indirection page? + je .done_test #NO, goto "done_test" + nr %r5,%r10 #YES, mask out, + lr %r2,%r5 #move it into the right register, + j .top #and read next... + .done_test: + tml %r5,0x4 #is it the done indicator? + je .source_test #NO! Well, then it should be the source indicator... + j .done #ok, lets finish it here... + .source_test: + tml %r5,0x8 #it should be a source indicator... + je .top #NO, ignore it... + lr %r8,%r5 #r8 = r5 + nr %r8,%r10 #masking + 0: mvcle %r6,%r8,0x0 #copy PAGE_SIZE bytes from r8 to r6 - pad with 0 + jo 0b + j .top + .done: + sr %r0,%r0 #clear register r0 + la %r4,load_psw-.base(%r13) #load psw-address into the register + o %r3,4(%r4) #or load address into psw + st %r3,4(%r4) + mvc 0(8,%r0),0(%r4) #copy psw to absolute address 0 + sr %r1,%r1 #clear %r1 + sr %r2,%r2 #clear %r2 + sigp %r1,%r2,0x12 #set cpuid to zero + lpsw 0 #hopefully start new kernel... + + .align 8 + zero64: + .quad 0 + load_psw: + .long 0x00080000,0x80000000 + sys_msk: + .quad 0 + relocate_kernel_end: + .globl relocate_kernel_len + relocate_kernel_len: + .quad relocate_kernel_end - relocate_kernel diff --git a/arch/s390/kernel/relocate_kernel64.S b/arch/s390/kernel/relocate_kernel64.S new file mode 100644 index 000000000000..96290cc4eb3c --- /dev/null +++ b/arch/s390/kernel/relocate_kernel64.S @@ -0,0 +1,82 @@ +/* + * arch/s390/kernel/relocate_kernel64.S + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger + * + */ + +/* + * moves the new kernel to its destination... + * %r2 = pointer to first kimage_entry_t + * %r3 = start address - where to jump to after the job is done... + * + * %r5 will be used as temp. storage + * %r6 holds the destination address + * %r7 = PAGE_SIZE + * %r8 holds the source address + * %r9 = PAGE_SIZE + * + * 0xf000 is a page_mask + */ + + .text + .globl relocate_kernel + relocate_kernel: + basr %r13,0 #base address + .base: + spx zero64-.base(%r13) #absolute addressing mode + stnsm sys_msk-.base(%r13),0xf8 #disable DAT and IRQ (external) + .top: + lghi %r7,4096 #load PAGE_SIZE in r7 + lghi %r9,4096 #load PAGE_SIZE in r9 + lg %r5,0(%r2) #read another word for indirection page + aghi %r2,8 #increment pointer + tml %r5,0x1 #is it a destination page? + je .indir_check #NO, goto "indir_check" + lgr %r6,%r5 #r6 = r5 + nill %r6,0xf000 #mask it out and... + j .top #...next iteration + .indir_check: + tml %r5,0x2 #is it a indirection page? + je .done_test #NO, goto "done_test" + nill %r5,0xf000 #YES, mask out, + lgr %r2,%r5 #move it into the right register, + j .top #and read next... + .done_test: + tml %r5,0x4 #is it the done indicator? + je .source_test #NO! Well, then it should be the source indicator... + j .done #ok, lets finish it here... + .source_test: + tml %r5,0x8 #it should be a source indicator... + je .top #NO, ignore it... + lgr %r8,%r5 #r8 = r5 + nill %r8,0xf000 #masking + 0: mvcle %r6,%r8,0x0 #copy PAGE_SIZE bytes from r8 to r6 - pad with 0 + jo 0b + j .top + .done: + sgr %r0,%r0 #clear register r0 + la %r4,load_psw-.base(%r13) #load psw-address into the register + o %r3,4(%r4) #or load address into psw + st %r3,4(%r4) + mvc 0(8,%r0),0(%r4) #copy psw to absolute address 0 + sam31 #31 bit mode + sr %r1,%r1 #erase register r1 + sr %r2,%r2 #erase register r2 + sigp %r1,%r2,0x12 #set cpuid to zero + lpsw 0 #hopefully start new kernel... + + .align 8 + zero64: + .quad 0 + load_psw: + .long 0x00080000,0x80000000 + sys_msk: + .quad 0 + relocate_kernel_end: + .globl relocate_kernel_len + relocate_kernel_len: + .quad relocate_kernel_end - relocate_kernel + diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index 515938628f82..a8668afb5f87 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -285,7 +285,7 @@ SYSCALL(sys_mq_timedsend,sys_mq_timedsend,compat_sys_mq_timedsend_wrapper) SYSCALL(sys_mq_timedreceive,sys_mq_timedreceive,compat_sys_mq_timedreceive_wrapper) SYSCALL(sys_mq_notify,sys_mq_notify,compat_sys_mq_notify_wrapper) /* 275 */ SYSCALL(sys_mq_getsetattr,sys_mq_getsetattr,compat_sys_mq_getsetattr_wrapper) -NI_SYSCALL /* reserved for kexec */ +SYSCALL(sys_kexec_load,sys_kexec_load,compat_sys_kexec_load_wrapper) SYSCALL(sys_add_key,sys_add_key,compat_sys_add_key_wrapper) SYSCALL(sys_request_key,sys_request_key,compat_sys_request_key_wrapper) SYSCALL(sys_keyctl,sys_keyctl,compat_sys_keyctl) /* 280 */ diff --git a/include/asm-s390/kexec.h b/include/asm-s390/kexec.h new file mode 100644 index 000000000000..54cf7d9f251c --- /dev/null +++ b/include/asm-s390/kexec.h @@ -0,0 +1,42 @@ +/* + * include/asm-s390/kexec.h + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger + * + */ + +#ifndef _S390_KEXEC_H +#define _S390_KEXEC_H + +#include +#include +/* + * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return. + * I.e. Maximum page that is mapped directly into kernel memory, + * and kmap is not required. + */ + +/* Maximum physical address we can use pages from */ +#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL) + +/* Maximum address we can reach in physical address mode */ +#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL) + +/* Maximum address we can use for the control pages */ +/* Not more than 2GB */ +#define KEXEC_CONTROL_MEMORY_LIMIT (1<<31) + +/* Allocate one page for the pdp and the second for the code */ +#define KEXEC_CONTROL_CODE_SIZE 4096 + +/* The native architecture */ +#define KEXEC_ARCH KEXEC_ARCH_S390 + +#define MAX_NOTE_BYTES 1024 +typedef u32 note_buf_t[MAX_NOTE_BYTES/4]; + +extern note_buf_t crash_notes[]; + +#endif /*_S390_KEXEC_H */ diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h index f1a204f7c0f0..363db45f8d07 100644 --- a/include/asm-s390/unistd.h +++ b/include/asm-s390/unistd.h @@ -269,7 +269,7 @@ #define __NR_mq_timedreceive 274 #define __NR_mq_notify 275 #define __NR_mq_getsetattr 276 -/* Number 277 is reserved for new sys_kexec_load */ +#define __NR_kexec_load 277 #define __NR_add_key 278 #define __NR_request_key 279 #define __NR_keyctl 280 diff --git a/include/linux/kexec.h b/include/linux/kexec.h index e3fc35f4e35f..0653a27c3d72 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -114,6 +114,7 @@ extern struct kimage *kexec_image; #define KEXEC_ARCH_PPC (20 << 16) #define KEXEC_ARCH_PPC64 (21 << 16) #define KEXEC_ARCH_IA_64 (50 << 16) +#define KEXEC_ARCH_S390 (22 << 16) #define KEXEC_FLAGS (KEXEC_ON_CRASH) /* List of defined/legal kexec flags */ -- cgit v1.2.3 From 92aa63a5a1bf2e7b0c79e6716d24b76dbbdcf951 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 25 Jun 2005 14:58:18 -0700 Subject: [PATCH] kdump: Retrieve saved max pfn This patch retrieves the max_pfn being used by previous kernel and stores it in a safe location (saved_max_pfn) before it is overwritten due to user defined memory map. This pfn is used to make sure that user does not try to read the physical memory beyond saved_max_pfn. Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/setup.c | 12 ++++++++++++ include/linux/bootmem.h | 4 ++++ mm/bootmem.c | 8 ++++++++ 3 files changed, 24 insertions(+) (limited to 'include/linux') diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index d88ebdfa6ccd..8d58a053e12e 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -59,6 +59,9 @@ #include "setup_arch_pre.h" #include +/* Forward Declaration. */ +void __init find_max_pfn(void); + /* This value is set up by the early boot code to point to the value immediately after the boot time page tables. It contains a *physical* address, and must not be in the .bss segment! */ @@ -736,6 +739,15 @@ static void __init parse_cmdline_early (char ** cmdline_p) if (to != command_line) to--; if (!memcmp(from+7, "exactmap", 8)) { +#ifdef CONFIG_CRASH_DUMP + /* If we are doing a crash dump, we + * still need to know the real mem + * size before original memory map is + * reset. + */ + find_max_pfn(); + saved_max_pfn = max_pfn; +#endif from += 8+7; e820.nr_map = 0; userdef = 1; diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index 500f451ce0c0..82bd8842d11c 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -22,6 +22,10 @@ extern unsigned long min_low_pfn; */ extern unsigned long max_pfn; +#ifdef CONFIG_CRASH_DUMP +extern unsigned long saved_max_pfn; +#endif + /* * node_bootmem_map is a map pointer - the bits represent all physical * memory pages (including holes) on the node. diff --git a/mm/bootmem.c b/mm/bootmem.c index f82f7aebbee3..45275f1f8947 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -33,6 +33,14 @@ EXPORT_SYMBOL(max_pfn); /* This is exported so * dma_get_required_mask(), which uses * it, can be an inline function */ +#ifdef CONFIG_CRASH_DUMP +/* + * If we have booted due to a crash, max_pfn will be a very low value. We need + * to know the amount of memory that the previous kernel used. + */ +unsigned long saved_max_pfn; +#endif + /* return the number of _pages_ that will be allocated for the boot bitmap */ unsigned long __init bootmem_bootmap_pages (unsigned long pages) { -- cgit v1.2.3 From 60e64d46a58236e3c718074372cab6a5b56a3b15 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 25 Jun 2005 14:58:19 -0700 Subject: [PATCH] kdump: Routines for copying dump pages This patch provides the interfaces necessary to read the dump contents, treating it as a high memory device. Signed off by Hariprasad Nellitheertha Signed-off-by: Eric Biederman Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/highmem.c | 18 +++++++++++++++++ include/asm-i386/highmem.h | 1 + include/linux/crash_dump.h | 13 ++++++++++++ include/linux/highmem.h | 1 + kernel/Makefile | 1 + kernel/crash_dump.c | 49 ++++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 83 insertions(+) create mode 100644 include/linux/crash_dump.h create mode 100644 kernel/crash_dump.c (limited to 'include/linux') diff --git a/arch/i386/mm/highmem.c b/arch/i386/mm/highmem.c index 4b7aaf99d7ea..b6eb4dcb8777 100644 --- a/arch/i386/mm/highmem.c +++ b/arch/i386/mm/highmem.c @@ -75,6 +75,24 @@ void kunmap_atomic(void *kvaddr, enum km_type type) preempt_check_resched(); } +/* This is the same as kmap_atomic() but can map memory that doesn't + * have a struct page associated with it. + */ +void *kmap_atomic_pfn(unsigned long pfn, enum km_type type) +{ + enum fixed_addresses idx; + unsigned long vaddr; + + inc_preempt_count(); + + idx = type + KM_TYPE_NR*smp_processor_id(); + vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); + set_pte(kmap_pte-idx, pfn_pte(pfn, kmap_prot)); + __flush_tlb_one(vaddr); + + return (void*) vaddr; +} + struct page *kmap_atomic_to_page(void *ptr) { unsigned long idx, vaddr = (unsigned long)ptr; diff --git a/include/asm-i386/highmem.h b/include/asm-i386/highmem.h index 1df42bf347df..0fd331306b60 100644 --- a/include/asm-i386/highmem.h +++ b/include/asm-i386/highmem.h @@ -70,6 +70,7 @@ void *kmap(struct page *page); void kunmap(struct page *page); void *kmap_atomic(struct page *page, enum km_type type); void kunmap_atomic(void *kvaddr, enum km_type type); +void *kmap_atomic_pfn(unsigned long pfn, enum km_type type); struct page *kmap_atomic_to_page(void *ptr); #define flush_cache_kmaps() do { } while (0) diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h new file mode 100644 index 000000000000..7d983b817429 --- /dev/null +++ b/include/linux/crash_dump.h @@ -0,0 +1,13 @@ +#ifndef LINUX_CRASH_DUMP_H +#define LINUX_CRASH_DUMP_H + +#ifdef CONFIG_CRASH_DUMP +#include +#include +#include +#include + +extern ssize_t copy_oldmem_page(unsigned long, char *, size_t, + unsigned long, int); +#endif /* CONFIG_CRASH_DUMP */ +#endif /* LINUX_CRASHDUMP_H */ diff --git a/include/linux/highmem.h b/include/linux/highmem.h index 2a7e6c65c882..6bece9280eb7 100644 --- a/include/linux/highmem.h +++ b/include/linux/highmem.h @@ -28,6 +28,7 @@ static inline void *kmap(struct page *page) #define kmap_atomic(page, idx) page_address(page) #define kunmap_atomic(addr, idx) do { } while (0) +#define kmap_atomic_pfn(pfn, idx) page_address(pfn_to_page(pfn)) #define kmap_atomic_to_page(ptr) virt_to_page(ptr) #endif /* CONFIG_HIGHMEM */ diff --git a/kernel/Makefile b/kernel/Makefile index cfc8b0dea950..cb05cd05d237 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -28,6 +28,7 @@ obj-$(CONFIG_AUDITSYSCALL) += auditsc.o obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_SYSFS) += ksysfs.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ +obj-$(CONFIG_CRASH_DUMP) += crash_dump.o obj-$(CONFIG_SECCOMP) += seccomp.o ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y) diff --git a/kernel/crash_dump.c b/kernel/crash_dump.c new file mode 100644 index 000000000000..5a1e6d5d203e --- /dev/null +++ b/kernel/crash_dump.c @@ -0,0 +1,49 @@ +/* + * kernel/crash_dump.c - Memory preserving reboot related code. + * + * Created by: Hariprasad Nellitheertha (hari@in.ibm.com) + * Copyright (C) IBM Corporation, 2004. All rights reserved + */ + +#include +#include +#include +#include +#include +#include + +#include +#include + +/* + * Copy a page from "oldmem". For this page, there is no pte mapped + * in the current kernel. We stitch up a pte, similar to kmap_atomic. + */ +ssize_t copy_oldmem_page(unsigned long pfn, char *buf, + size_t csize, unsigned long offset, int userbuf) +{ + void *page, *vaddr; + + if (!csize) + return 0; + + page = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!page) + return -ENOMEM; + + vaddr = kmap_atomic_pfn(pfn, KM_PTE0); + copy_page(page, vaddr); + kunmap_atomic(vaddr, KM_PTE0); + + if (userbuf) { + if (copy_to_user(buf, (page + offset), csize)) { + kfree(page); + return -EFAULT; + } + } else { + memcpy(buf, (page + offset), csize); + } + + kfree(page); + return csize; +} -- cgit v1.2.3 From 2030eae52b416a9a9f0ffda74c982b7f1e19496d Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 25 Jun 2005 14:58:20 -0700 Subject: [PATCH] Retrieve elfcorehdr address from command line This patch adds support for retrieving the address of elf core header if one is passed in command line. Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 4 ++++ arch/i386/kernel/setup.c | 8 ++++++++ include/linux/crash_dump.h | 1 + kernel/crash_dump.c | 3 +++ 4 files changed, 16 insertions(+) (limited to 'include/linux') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 560ff5ae3fd9..f44bb5567c5b 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -451,6 +451,10 @@ running once the system is up. Format: {"as"|"cfq"|"deadline"|"noop"} See Documentation/block/as-iosched.txt and Documentation/block/deadline-iosched.txt for details. + elfcorehdr= [IA-32] + Specifies physical address of start of kernel core image + elf header. + See Documentation/kdump.txt for details. enforcing [SELINUX] Set initial enforcing status. Format: {"0" | "1"} diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 8d58a053e12e..7306353c520e 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -44,6 +44,7 @@ #include #include #include +#include #include