diff options
Diffstat (limited to 'fs/cifs')
-rw-r--r-- | fs/cifs/cifs_debug.c | 12 | ||||
-rw-r--r-- | fs/cifs/cifsfs.c | 10 | ||||
-rw-r--r-- | fs/cifs/cifsfs.h | 2 | ||||
-rw-r--r-- | fs/cifs/cifsglob.h | 70 | ||||
-rw-r--r-- | fs/cifs/cifsproto.h | 7 | ||||
-rw-r--r-- | fs/cifs/connect.c | 63 | ||||
-rw-r--r-- | fs/cifs/file.c | 8 | ||||
-rw-r--r-- | fs/cifs/fscache.c | 8 | ||||
-rw-r--r-- | fs/cifs/fscache.h | 10 | ||||
-rw-r--r-- | fs/cifs/inode.c | 4 | ||||
-rw-r--r-- | fs/cifs/misc.c | 40 | ||||
-rw-r--r-- | fs/cifs/sess.c | 174 | ||||
-rw-r--r-- | fs/cifs/smb2ops.c | 181 | ||||
-rw-r--r-- | fs/cifs/smb2pdu.c | 26 | ||||
-rw-r--r-- | fs/cifs/trace.h | 38 |
15 files changed, 467 insertions, 186 deletions
diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 1dd995efd5b8..2cfbac8bb965 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -162,6 +162,8 @@ cifs_dump_iface(struct seq_file *m, struct cifs_server_iface *iface) seq_printf(m, "\t\tIPv4: %pI4\n", &ipv4->sin_addr); else if (iface->sockaddr.ss_family == AF_INET6) seq_printf(m, "\t\tIPv6: %pI6\n", &ipv6->sin6_addr); + if (!iface->is_active) + seq_puts(m, "\t\t[for-cleanup]\n"); } static int cifs_debug_files_proc_show(struct seq_file *m, void *v) @@ -221,6 +223,7 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) struct TCP_Server_Info *server; struct cifs_ses *ses; struct cifs_tcon *tcon; + struct cifs_server_iface *iface; int c, i, j; seq_puts(m, @@ -456,11 +459,10 @@ skip_rdma: if (ses->iface_count) seq_printf(m, "\n\n\tServer interfaces: %zu", ses->iface_count); - for (j = 0; j < ses->iface_count; j++) { - struct cifs_server_iface *iface; - - iface = &ses->iface_list[j]; - seq_printf(m, "\n\t%d)", j+1); + j = 0; + list_for_each_entry(iface, &ses->iface_list, + iface_head) { + seq_printf(m, "\n\t%d)", ++j); cifs_dump_iface(m, iface); if (is_ses_using_iface(ses, iface)) seq_puts(m, "\t\t[CONNECTED]\n"); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 12c872800326..8f2e003e0590 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -377,7 +377,7 @@ cifs_alloc_inode(struct super_block *sb) cifs_inode->flags = 0; spin_lock_init(&cifs_inode->writers_lock); cifs_inode->writers = 0; - cifs_inode->vfs_inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */ + cifs_inode->netfs.inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */ cifs_inode->server_eof = 0; cifs_inode->uniqueid = 0; cifs_inode->createtime = 0; @@ -389,12 +389,12 @@ cifs_alloc_inode(struct super_block *sb) * Can not set i_flags here - they get immediately overwritten to zero * by the VFS. */ - /* cifs_inode->vfs_inode.i_flags = S_NOATIME | S_NOCMTIME; */ + /* cifs_inode->netfs.inode.i_flags = S_NOATIME | S_NOCMTIME; */ INIT_LIST_HEAD(&cifs_inode->openFileList); INIT_LIST_HEAD(&cifs_inode->llist); INIT_LIST_HEAD(&cifs_inode->deferred_closes); spin_lock_init(&cifs_inode->deferred_lock); - return &cifs_inode->vfs_inode; + return &cifs_inode->netfs.inode; } static void @@ -1086,7 +1086,7 @@ struct file_system_type cifs_fs_type = { }; MODULE_ALIAS_FS("cifs"); -static struct file_system_type smb3_fs_type = { +struct file_system_type smb3_fs_type = { .owner = THIS_MODULE, .name = "smb3", .init_fs_context = smb3_init_fs_context, @@ -1418,7 +1418,7 @@ cifs_init_once(void *inode) { struct cifsInodeInfo *cifsi = inode; - inode_init_once(&cifsi->vfs_inode); + inode_init_once(&cifsi->netfs.inode); init_rwsem(&cifsi->lock_sem); } diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index dd7e070ca243..b17be47a8e59 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -38,7 +38,7 @@ static inline unsigned long cifs_get_time(struct dentry *dentry) return (unsigned long) dentry->d_fsdata; } -extern struct file_system_type cifs_fs_type; +extern struct file_system_type cifs_fs_type, smb3_fs_type; extern const struct address_space_operations cifs_addr_ops; extern const struct address_space_operations cifs_addr_ops_smallbuf; diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index f873379066c7..a643c84ff1e9 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -80,6 +80,9 @@ #define SMB_DNS_RESOLVE_INTERVAL_MIN 120 #define SMB_DNS_RESOLVE_INTERVAL_DEFAULT 600 +/* smb multichannel query server interfaces interval in seconds */ +#define SMB_INTERFACE_POLL_INTERVAL 600 + /* maximum number of PDUs in one compound */ #define MAX_COMPOUND 5 @@ -933,15 +936,67 @@ static inline void cifs_set_net_ns(struct TCP_Server_Info *srv, struct net *net) #endif struct cifs_server_iface { + struct list_head iface_head; + struct kref refcount; size_t speed; unsigned int rdma_capable : 1; unsigned int rss_capable : 1; + unsigned int is_active : 1; /* unset if non existent */ struct sockaddr_storage sockaddr; }; +/* release iface when last ref is dropped */ +static inline void +release_iface(struct kref *ref) +{ + struct cifs_server_iface *iface = container_of(ref, + struct cifs_server_iface, + refcount); + list_del_init(&iface->iface_head); + kfree(iface); +} + +/* + * compare two interfaces a and b + * return 0 if everything matches. + * return 1 if a has higher link speed, or rdma capable, or rss capable + * return -1 otherwise. + */ +static inline int +iface_cmp(struct cifs_server_iface *a, struct cifs_server_iface *b) +{ + int cmp_ret = 0; + + WARN_ON(!a || !b); + if (a->speed == b->speed) { + if (a->rdma_capable == b->rdma_capable) { + if (a->rss_capable == b->rss_capable) { + cmp_ret = memcmp(&a->sockaddr, &b->sockaddr, + sizeof(a->sockaddr)); + if (!cmp_ret) + return 0; + else if (cmp_ret > 0) + return 1; + else + return -1; + } else if (a->rss_capable > b->rss_capable) + return 1; + else + return -1; + } else if (a->rdma_capable > b->rdma_capable) + return 1; + else + return -1; + } else if (a->speed > b->speed) + return 1; + else + return -1; +} + struct cifs_chan { unsigned int in_reconnect : 1; /* if session setup in progress for this channel */ struct TCP_Server_Info *server; + struct cifs_server_iface *iface; /* interface in use */ __u8 signkey[SMB3_SIGN_KEY_SIZE]; }; @@ -993,7 +1048,7 @@ struct cifs_ses { */ spinlock_t iface_lock; /* ========= begin: protected by iface_lock ======== */ - struct cifs_server_iface *iface_list; + struct list_head iface_list; size_t iface_count; unsigned long iface_last_update; /* jiffies */ /* ========= end: protected by iface_lock ======== */ @@ -1203,6 +1258,7 @@ struct cifs_tcon { #ifdef CONFIG_CIFS_DFS_UPCALL struct list_head ulist; /* cache update list */ #endif + struct delayed_work query_interfaces; /* query interfaces workqueue job */ }; /* @@ -1479,20 +1535,16 @@ void cifsFileInfo_put(struct cifsFileInfo *cifs_file); #define CIFS_CACHE_RW_FLG (CIFS_CACHE_READ_FLG | CIFS_CACHE_WRITE_FLG) #define CIFS_CACHE_RHW_FLG (CIFS_CACHE_RW_FLG | CIFS_CACHE_HANDLE_FLG) -#define CIFS_CACHE_READ(cinode) ((cinode->oplock & CIFS_CACHE_READ_FLG) || (CIFS_SB(cinode->vfs_inode.i_sb)->mnt_cifs_flags & CIFS_MOUNT_RO_CACHE)) +#define CIFS_CACHE_READ(cinode) ((cinode->oplock & CIFS_CACHE_READ_FLG) || (CIFS_SB(cinode->netfs.inode.i_sb)->mnt_cifs_flags & CIFS_MOUNT_RO_CACHE)) #define CIFS_CACHE_HANDLE(cinode) (cinode->oplock & CIFS_CACHE_HANDLE_FLG) -#define CIFS_CACHE_WRITE(cinode) ((cinode->oplock & CIFS_CACHE_WRITE_FLG) || (CIFS_SB(cinode->vfs_inode.i_sb)->mnt_cifs_flags & CIFS_MOUNT_RW_CACHE)) +#define CIFS_CACHE_WRITE(cinode) ((cinode->oplock & CIFS_CACHE_WRITE_FLG) || (CIFS_SB(cinode->netfs.inode.i_sb)->mnt_cifs_flags & CIFS_MOUNT_RW_CACHE)) /* * One of these for each file inode */ struct cifsInodeInfo { - struct { - /* These must be contiguous */ - struct inode vfs_inode; /* the VFS's inode record */ - struct netfs_i_context netfs_ctx; /* Netfslib context */ - }; + struct netfs_inode netfs; /* Netfslib context and vfs inode */ bool can_cache_brlcks; struct list_head llist; /* locks helb by this inode */ /* @@ -1531,7 +1583,7 @@ struct cifsInodeInfo { static inline struct cifsInodeInfo * CIFS_I(struct inode *inode) { - return container_of(inode, struct cifsInodeInfo, vfs_inode); + return container_of(inode, struct cifsInodeInfo, netfs.inode); } static inline struct cifs_sb_info * diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 3b7366ec03c7..d59aebefa71c 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -636,6 +636,13 @@ cifs_chan_clear_need_reconnect(struct cifs_ses *ses, bool cifs_chan_needs_reconnect(struct cifs_ses *ses, struct TCP_Server_Info *server); +bool +cifs_chan_is_iface_active(struct cifs_ses *ses, + struct TCP_Server_Info *server); +int +cifs_chan_update_iface(struct cifs_ses *ses, struct TCP_Server_Info *server); +int +SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon); void extract_unc_hostname(const char *unc, const char **h, size_t *len); int copy_path_name(char *dst, const char *src); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index d46702f5a663..fa29c9aae24b 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -97,6 +97,10 @@ static int reconn_set_ipaddr_from_hostname(struct TCP_Server_Info *server) if (!server->hostname) return -EINVAL; + /* if server hostname isn't populated, there's nothing to do here */ + if (server->hostname[0] == '\0') + return 0; + len = strlen(server->hostname) + 3; unc = kmalloc(len, GFP_KERNEL); @@ -141,6 +145,25 @@ requeue_resolve: return rc; } +static void smb2_query_server_interfaces(struct work_struct *work) +{ + int rc; + struct cifs_tcon *tcon = container_of(work, + struct cifs_tcon, + query_interfaces.work); + + /* + * query server network interfaces, in case they change + */ + rc = SMB3_request_interfaces(0, tcon); + if (rc) { + cifs_dbg(FYI, "%s: failed to query server interfaces: %d\n", + __func__, rc); + } + + queue_delayed_work(cifsiod_wq, &tcon->query_interfaces, + (SMB_INTERFACE_POLL_INTERVAL * HZ)); +} static void cifs_resolve_server(struct work_struct *work) { @@ -213,7 +236,7 @@ cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server, bool mark_smb_session) { struct TCP_Server_Info *pserver; - struct cifs_ses *ses; + struct cifs_ses *ses, *nses; struct cifs_tcon *tcon; /* @@ -227,7 +250,20 @@ cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server, spin_lock(&cifs_tcp_ses_lock); - list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { + list_for_each_entry_safe(ses, nses, &pserver->smb_ses_list, smb_ses_list) { + /* check if iface is still active */ + if (!cifs_chan_is_iface_active(ses, server)) { + /* + * HACK: drop the lock before calling + * cifs_chan_update_iface to avoid deadlock + */ + ses->ses_count++; + spin_unlock(&cifs_tcp_ses_lock); + cifs_chan_update_iface(ses, server); + spin_lock(&cifs_tcp_ses_lock); + ses->ses_count--; + } + spin_lock(&ses->chan_lock); if (!mark_smb_session && cifs_chan_needs_reconnect(ses, server)) goto next_session; @@ -1890,9 +1926,11 @@ void cifs_put_smb_ses(struct cifs_ses *ses) int i; for (i = 1; i < chan_count; i++) { - spin_unlock(&ses->chan_lock); + if (ses->chans[i].iface) { + kref_put(&ses->chans[i].iface->refcount, release_iface); + ses->chans[i].iface = NULL; + } cifs_put_tcp_session(ses->chans[i].server, 0); - spin_lock(&ses->chan_lock); ses->chans[i].server = NULL; } } @@ -2266,6 +2304,9 @@ cifs_put_tcon(struct cifs_tcon *tcon) list_del_init(&tcon->tcon_list); spin_unlock(&cifs_tcp_ses_lock); + /* cancel polling of interfaces */ + cancel_delayed_work_sync(&tcon->query_interfaces); + if (tcon->use_witness) { int rc; @@ -2503,6 +2544,12 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) tcon->local_lease = ctx->local_lease; INIT_LIST_HEAD(&tcon->pending_opens); + /* schedule query interfaces poll */ + INIT_DELAYED_WORK(&tcon->query_interfaces, + smb2_query_server_interfaces); + queue_delayed_work(cifsiod_wq, &tcon->query_interfaces, + (SMB_INTERFACE_POLL_INTERVAL * HZ)); + spin_lock(&cifs_tcp_ses_lock); list_add(&tcon->tcon_list, &ses->tcon_list); spin_unlock(&cifs_tcp_ses_lock); @@ -3978,10 +4025,16 @@ cifs_setup_session(const unsigned int xid, struct cifs_ses *ses, struct nls_table *nls_info) { int rc = -ENOSYS; + struct sockaddr_in6 *addr6 = (struct sockaddr_in6 *)&server->dstaddr; + struct sockaddr_in *addr = (struct sockaddr_in *)&server->dstaddr; bool is_binding = false; - spin_lock(&cifs_tcp_ses_lock); + if (server->dstaddr.ss_family == AF_INET6) + scnprintf(ses->ip_addr, sizeof(ses->ip_addr), "%pI6", &addr6->sin6_addr); + else + scnprintf(ses->ip_addr, sizeof(ses->ip_addr), "%pI4", &addr->sin_addr); + if (ses->ses_status != SES_GOOD && ses->ses_status != SES_NEW && ses->ses_status != SES_NEED_RECON) { diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 1618e0537d58..e64cda7a7610 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2004,7 +2004,7 @@ struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only) { struct cifsFileInfo *open_file = NULL; - struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_inode->netfs.inode.i_sb); /* only filter by fsuid on multiuser mounts */ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)) @@ -2060,7 +2060,7 @@ cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, int flags, return rc; } - cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb); + cifs_sb = CIFS_SB(cifs_inode->netfs.inode.i_sb); /* only filter by fsuid on multiuser mounts */ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)) @@ -4669,14 +4669,14 @@ bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 end_of_file) /* This inode is open for write at least once */ struct cifs_sb_info *cifs_sb; - cifs_sb = CIFS_SB(cifsInode->vfs_inode.i_sb); + cifs_sb = CIFS_SB(cifsInode->netfs.inode.i_sb); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { /* since no page cache to corrupt on directio we can change size safely */ return true; } - if (i_size_read(&cifsInode->vfs_inode) < end_of_file) + if (i_size_read(&cifsInode->netfs.inode) < end_of_file) return true; return false; diff --git a/fs/cifs/fscache.c b/fs/cifs/fscache.c index a638b29e9062..23ef56f55ce5 100644 --- a/fs/cifs/fscache.c +++ b/fs/cifs/fscache.c @@ -101,13 +101,13 @@ void cifs_fscache_get_inode_cookie(struct inode *inode) struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); - cifs_fscache_fill_coherency(&cifsi->vfs_inode, &cd); + cifs_fscache_fill_coherency(&cifsi->netfs.inode, &cd); - cifsi->netfs_ctx.cache = + cifsi->netfs.cache = fscache_acquire_cookie(tcon->fscache, 0, &cifsi->uniqueid, sizeof(cifsi->uniqueid), &cd, sizeof(cd), - i_size_read(&cifsi->vfs_inode)); + i_size_read(&cifsi->netfs.inode)); } void cifs_fscache_unuse_inode_cookie(struct inode *inode, bool update) @@ -131,7 +131,7 @@ void cifs_fscache_release_inode_cookie(struct inode *inode) if (cookie) { cifs_dbg(FYI, "%s: (0x%p)\n", __func__, cookie); fscache_relinquish_cookie(cookie, false); - cifsi->netfs_ctx.cache = NULL; + cifsi->netfs.cache = NULL; } } diff --git a/fs/cifs/fscache.h b/fs/cifs/fscache.h index 52355c0912ae..aa3b941a5555 100644 --- a/fs/cifs/fscache.h +++ b/fs/cifs/fscache.h @@ -52,16 +52,16 @@ void cifs_fscache_fill_coherency(struct inode *inode, struct cifsInodeInfo *cifsi = CIFS_I(inode); memset(cd, 0, sizeof(*cd)); - cd->last_write_time_sec = cpu_to_le64(cifsi->vfs_inode.i_mtime.tv_sec); - cd->last_write_time_nsec = cpu_to_le32(cifsi->vfs_inode.i_mtime.tv_nsec); - cd->last_change_time_sec = cpu_to_le64(cifsi->vfs_inode.i_ctime.tv_sec); - cd->last_change_time_nsec = cpu_to_le32(cifsi->vfs_inode.i_ctime.tv_nsec); + cd->last_write_time_sec = cpu_to_le64(cifsi->netfs.inode.i_mtime.tv_sec); + cd->last_write_time_nsec = cpu_to_le32(cifsi->netfs.inode.i_mtime.tv_nsec); + cd->last_change_time_sec = cpu_to_le64(cifsi->netfs.inode.i_ctime.tv_sec); + cd->last_change_time_nsec = cpu_to_le32(cifsi->netfs.inode.i_ctime.tv_nsec); } static inline struct fscache_cookie *cifs_inode_cookie(struct inode *inode) { - return netfs_i_cookie(inode); + return netfs_i_cookie(&CIFS_I(inode)->netfs); } static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags) diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 2f9e7d2f81b6..81da81e18553 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -115,7 +115,7 @@ cifs_revalidate_cache(struct inode *inode, struct cifs_fattr *fattr) __func__, cifs_i->uniqueid); set_bit(CIFS_INO_INVALID_MAPPING, &cifs_i->flags); /* Invalidate fscache cookie */ - cifs_fscache_fill_coherency(&cifs_i->vfs_inode, &cd); + cifs_fscache_fill_coherency(&cifs_i->netfs.inode, &cd); fscache_invalidate(cifs_inode_cookie(inode), &cd, i_size_read(inode), 0); } @@ -2499,7 +2499,7 @@ int cifs_fiemap(struct inode *inode, struct fiemap_extent_info *fei, u64 start, u64 len) { struct cifsInodeInfo *cifs_i = CIFS_I(inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_i->vfs_inode.i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_i->netfs.inode.i_sb); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); struct TCP_Server_Info *server = tcon->ses->server; struct cifsFileInfo *cfile; diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 35962a1a23b9..0e84e6fcf8ab 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -75,6 +75,7 @@ sesInfoAlloc(void) INIT_LIST_HEAD(&ret_buf->tcon_list); mutex_init(&ret_buf->session_mutex); spin_lock_init(&ret_buf->iface_lock); + INIT_LIST_HEAD(&ret_buf->iface_list); spin_lock_init(&ret_buf->chan_lock); } return ret_buf; @@ -83,6 +84,8 @@ sesInfoAlloc(void) void sesInfoFree(struct cifs_ses *buf_to_free) { + struct cifs_server_iface *iface = NULL, *niface = NULL; + if (buf_to_free == NULL) { cifs_dbg(FYI, "Null buffer passed to sesInfoFree\n"); return; @@ -96,7 +99,11 @@ sesInfoFree(struct cifs_ses *buf_to_free) kfree(buf_to_free->user_name); kfree(buf_to_free->domainName); kfree_sensitive(buf_to_free->auth_key.response); - kfree(buf_to_free->iface_list); + spin_lock(&buf_to_free->iface_lock); + list_for_each_entry_safe(iface, niface, &buf_to_free->iface_list, + iface_head) + kref_put(&iface->refcount, release_iface); + spin_unlock(&buf_to_free->iface_lock); kfree_sensitive(buf_to_free); } @@ -537,11 +544,11 @@ void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock) if (oplock == OPLOCK_EXCLUSIVE) { cinode->oplock = CIFS_CACHE_WRITE_FLG | CIFS_CACHE_READ_FLG; cifs_dbg(FYI, "Exclusive Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else if (oplock == OPLOCK_READ) { cinode->oplock = CIFS_CACHE_READ_FLG; cifs_dbg(FYI, "Level II Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else cinode->oplock = 0; } @@ -1211,18 +1218,23 @@ static struct super_block *__cifs_get_super(void (*f)(struct super_block *, void .data = data, .sb = NULL, }; + struct file_system_type **fs_type = (struct file_system_type *[]) { + &cifs_fs_type, &smb3_fs_type, NULL, + }; - iterate_supers_type(&cifs_fs_type, f, &sd); - - if (!sd.sb) - return ERR_PTR(-EINVAL); - /* - * Grab an active reference in order to prevent automounts (DFS links) - * of expiring and then freeing up our cifs superblock pointer while - * we're doing failover. - */ - cifs_sb_active(sd.sb); - return sd.sb; + for (; *fs_type; fs_type++) { + iterate_supers_type(*fs_type, f, &sd); + if (sd.sb) { + /* + * Grab an active reference in order to prevent automounts (DFS links) + * of expiring and then freeing up our cifs superblock pointer while + * we're doing failover. + */ + cifs_sb_active(sd.sb); + return sd.sb; + } + } + return ERR_PTR(-EINVAL); } static void __cifs_put_super(struct super_block *sb) diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 3b7915af1f62..b85718f32b53 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -58,7 +58,7 @@ bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface) spin_lock(&ses->chan_lock); for (i = 0; i < ses->chan_count; i++) { - if (is_server_using_iface(ses->chans[i].server, iface)) { + if (ses->chans[i].iface == iface) { spin_unlock(&ses->chan_lock); return true; } @@ -81,6 +81,9 @@ cifs_ses_get_chan_index(struct cifs_ses *ses, } /* If we didn't find the channel, it is likely a bug */ + if (server) + cifs_dbg(VFS, "unable to get chan index for server: 0x%llx", + server->conn_id); WARN_ON(1); return 0; } @@ -143,16 +146,24 @@ cifs_chan_needs_reconnect(struct cifs_ses *ses, return CIFS_CHAN_NEEDS_RECONNECT(ses, chan_index); } +bool +cifs_chan_is_iface_active(struct cifs_ses *ses, + struct TCP_Server_Info *server) +{ + unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + + return ses->chans[chan_index].iface && + ses->chans[chan_index].iface->is_active; +} + /* returns number of channels added */ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) { int old_chan_count, new_chan_count; int left; - int i = 0; int rc = 0; int tries = 0; - struct cifs_server_iface *ifaces = NULL; - size_t iface_count; + struct cifs_server_iface *iface = NULL, *niface = NULL; spin_lock(&ses->chan_lock); @@ -182,32 +193,16 @@ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) spin_unlock(&ses->chan_lock); /* - * Make a copy of the iface list at the time and use that - * instead so as to not hold the iface spinlock for opening - * channels - */ - spin_lock(&ses->iface_lock); - iface_count = ses->iface_count; - if (iface_count <= 0) { - spin_unlock(&ses->iface_lock); - cifs_dbg(VFS, "no iface list available to open channels\n"); - return 0; - } - ifaces = kmemdup(ses->iface_list, iface_count*sizeof(*ifaces), - GFP_ATOMIC); - if (!ifaces) { - spin_unlock(&ses->iface_lock); - return 0; - } - spin_unlock(&ses->iface_lock); - - /* * Keep connecting to same, fastest, iface for all channels as * long as its RSS. Try next fastest one if not RSS or channel * creation fails. */ + spin_lock(&ses->iface_lock); + iface = list_first_entry(&ses->iface_list, struct cifs_server_iface, + iface_head); + spin_unlock(&ses->iface_lock); + while (left > 0) { - struct cifs_server_iface *iface; tries++; if (tries > 3*ses->chan_max) { @@ -216,31 +211,128 @@ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) break; } - iface = &ifaces[i]; - if (is_ses_using_iface(ses, iface) && !iface->rss_capable) { - i = (i+1) % iface_count; - continue; + spin_lock(&ses->iface_lock); + if (!ses->iface_count) { + spin_unlock(&ses->iface_lock); + break; } - rc = cifs_ses_add_channel(cifs_sb, ses, iface); - if (rc) { - cifs_dbg(FYI, "failed to open extra channel on iface#%d rc=%d\n", - i, rc); - i = (i+1) % iface_count; - continue; + list_for_each_entry_safe_from(iface, niface, &ses->iface_list, + iface_head) { + /* skip ifaces that are unusable */ + if (!iface->is_active || + (is_ses_using_iface(ses, iface) && + !iface->rss_capable)) { + continue; + } + + /* take ref before unlock */ + kref_get(&iface->refcount); + + spin_unlock(&ses->iface_lock); + rc = cifs_ses_add_channel(cifs_sb, ses, iface); + spin_lock(&ses->iface_lock); + + if (rc) { + cifs_dbg(VFS, "failed to open extra channel on iface:%pIS rc=%d\n", + &iface->sockaddr, + rc); + kref_put(&iface->refcount, release_iface); + continue; + } + + cifs_dbg(FYI, "successfully opened new channel on iface:%pIS\n", + &iface->sockaddr); + break; } + spin_unlock(&ses->iface_lock); - cifs_dbg(FYI, "successfully opened new channel on iface#%d\n", - i); left--; new_chan_count++; } - kfree(ifaces); return new_chan_count - old_chan_count; } /* + * update the iface for the channel if necessary. + * will return 0 when iface is updated, 1 if removed, 2 otherwise + * Must be called with chan_lock held. + */ +int +cifs_chan_update_iface(struct cifs_ses *ses, struct TCP_Server_Info *server) +{ + unsigned int chan_index; + struct cifs_server_iface *iface = NULL; + struct cifs_server_iface *old_iface = NULL; + int rc = 0; + + spin_lock(&ses->chan_lock); + chan_index = cifs_ses_get_chan_index(ses, server); + if (!chan_index) { + spin_unlock(&ses->chan_lock); + return 0; + } + + if (ses->chans[chan_index].iface) { + old_iface = ses->chans[chan_index].iface; + if (old_iface->is_active) { + spin_unlock(&ses->chan_lock); + return 1; + } + } + spin_unlock(&ses->chan_lock); + + spin_lock(&ses->iface_lock); + /* then look for a new one */ + list_for_each_entry(iface, &ses->iface_list, iface_head) { + if (!iface->is_active || + (is_ses_using_iface(ses, iface) && + !iface->rss_capable)) { + continue; + } + kref_get(&iface->refcount); + } + + if (!list_entry_is_head(iface, &ses->iface_list, iface_head)) { + rc = 1; + iface = NULL; + cifs_dbg(FYI, "unable to find a suitable iface\n"); + } + + /* now drop the ref to the current iface */ + if (old_iface && iface) { + kref_put(&old_iface->refcount, release_iface); + cifs_dbg(FYI, "replacing iface: %pIS with %pIS\n", + &old_iface->sockaddr, + &iface->sockaddr); + } else if (old_iface) { + kref_put(&old_iface->refcount, release_iface); + cifs_dbg(FYI, "releasing ref to iface: %pIS\n", + &old_iface->sockaddr); + } else { + WARN_ON(!iface); + cifs_dbg(FYI, "adding new iface: %pIS\n", &iface->sockaddr); + } + spin_unlock(&ses->iface_lock); + + spin_lock(&ses->chan_lock); + chan_index = cifs_ses_get_chan_index(ses, server); + ses->chans[chan_index].iface = iface; + + /* No iface is found. if secondary chan, drop connection */ + if (!iface && CIFS_SERVER_IS_CHAN(server)) + ses->chans[chan_index].server = NULL; + + spin_unlock(&ses->chan_lock); + + if (!iface && CIFS_SERVER_IS_CHAN(server)) + cifs_put_tcp_session(server, false); + + return rc; +} + +/* * If server is a channel of ses, return the corresponding enclosing * cifs_chan otherwise return NULL. */ @@ -301,7 +393,10 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, /* Auth */ ctx.domainauto = ses->domainAuto; ctx.domainname = ses->domainName; - ctx.server_hostname = ses->server->hostname; + + /* no hostname for extra channels */ + ctx.server_hostname = ""; + ctx.username = ses->user_name; ctx.password = ses->password; ctx.sectype = ses->sectype; @@ -349,6 +444,7 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, spin_unlock(&ses->chan_lock); goto out; } + chan->iface = iface; ses->chan_count++; atomic_set(&ses->chan_seq, 0); diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index 98a76fa791c0..8802995b2d3d 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -512,73 +512,41 @@ smb3_negotiate_rsize(struct cifs_tcon *tcon, struct smb3_fs_context *ctx) static int parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, size_t buf_len, - struct cifs_server_iface **iface_list, - size_t *iface_count) + struct cifs_ses *ses) { struct network_interface_info_ioctl_rsp *p; struct sockaddr_in *addr4; struct sockaddr_in6 *addr6; struct iface_info_ipv4 *p4; struct iface_info_ipv6 *p6; - struct cifs_server_iface *info; + struct cifs_server_iface *info = NULL, *iface = NULL, *niface = NULL; + struct cifs_server_iface tmp_iface; ssize_t bytes_left; size_t next = 0; int nb_iface = 0; - int rc = 0; - - *iface_list = NULL; - *iface_count = 0; - - /* - * Fist pass: count and sanity check - */ + int rc = 0, ret = 0; bytes_left = buf_len; p = buf; - while (bytes_left >= sizeof(*p)) { - nb_iface++; - next = le32_to_cpu(p->Next); - if (!next) { - bytes_left -= sizeof(*p); - break; - } - p = (struct network_interface_info_ioctl_rsp *)((u8 *)p+next); - bytes_left -= next; - } - - if (!nb_iface) { - cifs_dbg(VFS, "%s: malformed interface info\n", __func__); - rc = -EINVAL; - goto out; - } - - /* Azure rounds the buffer size up 8, to a 16 byte boundary */ - if ((bytes_left > 8) || p->Next) - cifs_dbg(VFS, "%s: incomplete interface info\n", __func__); - + spin_lock(&ses->iface_lock); /* - * Second pass: extract info to internal structure + * Go through iface_list and do kref_put to remove + * any unused ifaces. ifaces in use will be removed + * when the last user calls a kref_put on it */ - - *iface_list = kcalloc(nb_iface, sizeof(**iface_list), GFP_KERNEL); - if (!*iface_list) { - rc = -ENOMEM; - goto out; + list_for_each_entry_safe(iface, niface, &ses->iface_list, + iface_head) { + iface->is_active = 0; + kref_put(&iface->refcount, release_iface); } + spin_unlock(&ses->iface_lock); - info = *iface_list; - bytes_left = buf_len; - p = buf; while (bytes_left >= sizeof(*p)) { - info->speed = le64_to_cpu(p->LinkSpeed); - info->rdma_capable = le32_to_cpu(p->Capability & RDMA_CAPABLE) ? 1 : 0; - info->rss_capable = le32_to_cpu(p->Capability & RSS_CAPABLE) ? 1 : 0; - - cifs_dbg(FYI, "%s: adding iface %zu\n", __func__, *iface_count); - cifs_dbg(FYI, "%s: speed %zu bps\n", __func__, info->speed); - cifs_dbg(FYI, "%s: capabilities 0x%08x\n", __func__, - le32_to_cpu(p->Capability)); + memset(&tmp_iface, 0, sizeof(tmp_iface)); + tmp_iface.speed = le64_to_cpu(p->LinkSpeed); + tmp_iface.rdma_capable = le32_to_cpu(p->Capability & RDMA_CAPABLE) ? 1 : 0; + tmp_iface.rss_capable = le32_to_cpu(p->Capability & RSS_CAPABLE) ? 1 : 0; switch (p->Family) { /* @@ -587,7 +555,7 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, * conversion explicit in case either one changes. */ case INTERNETWORK: - addr4 = (struct sockaddr_in *)&info->sockaddr; + addr4 = (struct sockaddr_in *)&tmp_iface.sockaddr; p4 = (struct iface_info_ipv4 *)p->Buffer; addr4->sin_family = AF_INET; memcpy(&addr4->sin_addr, &p4->IPv4Address, 4); @@ -599,7 +567,7 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, &addr4->sin_addr); break; case INTERNETWORKV6: - addr6 = (struct sockaddr_in6 *)&info->sockaddr; + addr6 = (struct sockaddr_in6 *)&tmp_iface.sockaddr; p6 = (struct iface_info_ipv6 *)p->Buffer; addr6->sin6_family = AF_INET6; memcpy(&addr6->sin6_addr, &p6->IPv6Address, 16); @@ -619,46 +587,96 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, goto next_iface; } - (*iface_count)++; - info++; + /* + * The iface_list is assumed to be sorted by speed. + * Check if the new interface exists in that list. + * NEVER change iface. it could be in use. + * Add a new one instead + */ + spin_lock(&ses->iface_lock); + iface = niface = NULL; + list_for_each_entry_safe(iface, niface, &ses->iface_list, + iface_head) { + ret = iface_cmp(iface, &tmp_iface); + if (!ret) { + /* just get a ref so that it doesn't get picked/freed */ + iface->is_active = 1; + kref_get(&iface->refcount); + spin_unlock(&ses->iface_lock); + goto next_iface; + } else if (ret < 0) { + /* all remaining ifaces are slower */ + kref_get(&iface->refcount); + break; + } + } + spin_unlock(&ses->iface_lock); + + /* no match. insert the entry in the list */ + info = kmalloc(sizeof(struct cifs_server_iface), + GFP_KERNEL); + if (!info) { + rc = -ENOMEM; + goto out; + } + memcpy(info, &tmp_iface, sizeof(tmp_iface)); + + /* add this new entry to the list */ + kref_init(&info->refcount); + info->is_active = 1; + + cifs_dbg(FYI, "%s: adding iface %zu\n", __func__, ses->iface_count); + cifs_dbg(FYI, "%s: speed %zu bps\n", __func__, info->speed); + cifs_dbg(FYI, "%s: capabilities 0x%08x\n", __func__, + le32_to_cpu(p->Capability)); + + spin_lock(&ses->iface_lock); + if (!list_entry_is_head(iface, &ses->iface_list, iface_head)) { + list_add_tail(&info->iface_head, &iface->iface_head); + kref_put(&iface->refcount, release_iface); + } else + list_add_tail(&info->iface_head, &ses->iface_list); + spin_unlock(&ses->iface_lock); + + ses->iface_count++; + ses->iface_last_update = jiffies; next_iface: + nb_iface++; next = le32_to_cpu(p->Next); - if (!next) + if (!next) { + bytes_left -= sizeof(*p); break; + } p = (struct network_interface_info_ioctl_rsp *)((u8 *)p+next); bytes_left -= next; } - if (!*iface_count) { + if (!nb_iface) { + cifs_dbg(VFS, "%s: malformed interface info\n", __func__); rc = -EINVAL; goto out; } -out: - if (rc) { - kfree(*iface_list); - *iface_count = 0; - *iface_list = NULL; - } - return rc; -} + /* Azure rounds the buffer size up 8, to a 16 byte boundary */ + if ((bytes_left > 8) || p->Next) + cifs_dbg(VFS, "%s: incomplete interface info\n", __func__); -static int compare_iface(const void *ia, const void *ib) -{ - const struct cifs_server_iface *a = (struct cifs_server_iface *)ia; - const struct cifs_server_iface *b = (struct cifs_server_iface *)ib; - return a->speed == b->speed ? 0 : (a->speed > b->speed ? -1 : 1); + if (!ses->iface_count) { + rc = -EINVAL; + goto out; + } + +out: + return rc; } -static int +int SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon) { int rc; unsigned int ret_data_len = 0; struct network_interface_info_ioctl_rsp *out_buf = NULL; - struct cifs_server_iface *iface_list; - size_t iface_count; struct cifs_ses *ses = tcon->ses; rc = SMB2_ioctl(xid, tcon, NO_FILE_ID, NO_FILE_ID, @@ -674,21 +692,10 @@ SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon) goto out; } - rc = parse_server_interfaces(out_buf, ret_data_len, - &iface_list, &iface_count); + rc = parse_server_interfaces(out_buf, ret_data_len, ses); if (rc) goto out; - /* sort interfaces from fastest to slowest */ - sort(iface_list, iface_count, sizeof(*iface_list), compare_iface, NULL); - - spin_lock(&ses->iface_lock); - kfree(ses->iface_list); - ses->iface_list = iface_list; - ses->iface_count = iface_count; - ses->iface_last_update = jiffies; - spin_unlock(&ses->iface_lock); - out: kfree(out_buf); return rc; @@ -4260,15 +4267,15 @@ smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, if (oplock == SMB2_OPLOCK_LEVEL_BATCH) { cinode->oplock = CIFS_CACHE_RHW_FLG; cifs_dbg(FYI, "Batch Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else if (oplock == SMB2_OPLOCK_LEVEL_EXCLUSIVE) { cinode->oplock = CIFS_CACHE_RW_FLG; cifs_dbg(FYI, "Exclusive Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else if (oplock == SMB2_OPLOCK_LEVEL_II) { cinode->oplock = CIFS_CACHE_READ_FLG; cifs_dbg(FYI, "Level II Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else cinode->oplock = 0; } @@ -4307,7 +4314,7 @@ smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, cinode->oplock = new_oplock; cifs_dbg(FYI, "%s Lease granted on inode %p\n", message, - &cinode->vfs_inode); + &cinode->netfs.inode); } static void diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 0e8c85249579..12b4dddaedb0 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -288,6 +288,9 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, mutex_unlock(&ses->session_mutex); rc = -EHOSTDOWN; goto failed; + } else if (rc) { + mutex_unlock(&ses->session_mutex); + goto out; } } else { mutex_unlock(&ses->session_mutex); @@ -540,6 +543,7 @@ assemble_neg_contexts(struct smb2_negotiate_req *req, struct TCP_Server_Info *server, unsigned int *total_len) { char *pneg_ctxt; + char *hostname = NULL; unsigned int ctxt_len, neg_context_count; if (*total_len > 200) { @@ -567,16 +571,24 @@ assemble_neg_contexts(struct smb2_negotiate_req *req, *total_len += ctxt_len; pneg_ctxt += ctxt_len; - ctxt_len = build_netname_ctxt((struct smb2_netname_neg_context *)pneg_ctxt, - server->hostname); - *total_len += ctxt_len; - pneg_ctxt += ctxt_len; - build_posix_ctxt((struct smb2_posix_neg_context *)pneg_ctxt); *total_len += sizeof(struct smb2_posix_neg_context); pneg_ctxt += sizeof(struct smb2_posix_neg_context); - neg_context_count = 4; + /* + * secondary channels don't have the hostname field populated + * use the hostname field in the primary channel instead + */ + hostname = CIFS_SERVER_IS_CHAN(server) ? + server->primary_server->hostname : server->hostname; + if (hostname && (hostname[0] != 0)) { + ctxt_len = build_netname_ctxt((struct smb2_netname_neg_context *)pneg_ctxt, + hostname); + *total_len += ctxt_len; + pneg_ctxt += ctxt_len; + neg_context_count = 4; + } else /* second channels do not have a hostname */ + neg_context_count = 3; if (server->compress_algorithm) { build_compression_ctxt((struct smb2_compression_capabilities_context *) @@ -5151,6 +5163,8 @@ SMB2_set_eof(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, data = &info; size = sizeof(struct smb2_file_eof_info); + trace_smb3_set_eof(xid, persistent_fid, tcon->tid, tcon->ses->Suid, le64_to_cpu(*eof)); + return send_set_info(xid, tcon, persistent_fid, volatile_fid, pid, FILE_END_OF_FILE_INFORMATION, SMB2_O_INFO_FILE, 0, 1, &data, &size); diff --git a/fs/cifs/trace.h b/fs/cifs/trace.h index 2be5e0c8564d..6b88dc2e364f 100644 --- a/fs/cifs/trace.h +++ b/fs/cifs/trace.h @@ -121,6 +121,44 @@ DEFINE_SMB3_RW_DONE_EVENT(query_dir_done); DEFINE_SMB3_RW_DONE_EVENT(zero_done); DEFINE_SMB3_RW_DONE_EVENT(falloc_done); +/* For logging successful set EOF (truncate) */ +DECLARE_EVENT_CLASS(smb3_eof_class, + TP_PROTO(unsigned int xid, + __u64 fid, + __u32 tid, + __u64 sesid, + __u64 offset), + TP_ARGS(xid, fid, tid, sesid, offset), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(__u64, fid) + __field(__u32, tid) + __field(__u64, sesid) + __field(__u64, offset) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->fid = fid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->offset = offset; + ), + TP_printk("xid=%u sid=0x%llx tid=0x%x fid=0x%llx offset=0x%llx", + __entry->xid, __entry->sesid, __entry->tid, __entry->fid, + __entry->offset) +) + +#define DEFINE_SMB3_EOF_EVENT(name) \ +DEFINE_EVENT(smb3_eof_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + __u64 fid, \ + __u32 tid, \ + __u64 sesid, \ + __u64 offset), \ + TP_ARGS(xid, fid, tid, sesid, offset)) + +DEFINE_SMB3_EOF_EVENT(set_eof); + /* * For handle based calls other than read and write, and get/set info */ |