From d99f160ac53e51090f015a8f0617cea25f81a191 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 5 Nov 2006 23:52:12 -0800 Subject: [PATCH] sysctl: allow a zero ctl_name in the middle of a sysctl table Since it is becoming clear that there are just enough users of the binary sysctl interface that completely removing the binary interface from the kernel will not be an option for foreseeable future, we need to find a way to address the sysctl maintenance issues. The basic problem is that sysctl requires one central authority to allocate sysctl numbers, or else conflicts and ABI breakage occur. The proc interface to sysctl does not have that problem, as names are not densely allocated. By not terminating a sysctl table until I have neither a ctl_name nor a procname, it becomes simple to add sysctl entries that don't show up in the binary sysctl interface. Which allows people to avoid allocating a binary sysctl value when not needed. I have audited the kernel code and in my reading I have not found a single sysctl table that wasn't terminated by a completely zero filled entry. So this change in behavior should not affect anything. I think this mechanism eases the pain enough that combined with a little disciple we can solve the reoccurring sysctl ABI breakage. Signed-off-by: Eric W. Biederman Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sysctl.h | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'include/linux/sysctl.h') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 1b24bd45e080..c184732a70fc 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -961,8 +961,8 @@ extern ctl_handler sysctl_ms_jiffies; /* * Register a set of sysctl names by calling register_sysctl_table * with an initialised array of ctl_table's. An entry with zero - * ctl_name terminates the table. table->de will be set up by the - * registration and need not be initialised in advance. + * ctl_name and NULL procname terminates the table. table->de will be + * set up by the registration and need not be initialised in advance. * * sysctl names can be mirrored automatically under /proc/sys. The * procname supplied controls /proc naming. @@ -973,7 +973,10 @@ extern ctl_handler sysctl_ms_jiffies; * Leaf nodes in the sysctl tree will be represented by a single file * under /proc; non-leaf nodes will be represented by directories. A * null procname disables /proc mirroring at this node. - * + * + * sysctl entries with a zero ctl_name will not be available through + * the binary sysctl interface. + * * sysctl(2) can automatically manage read and write requests through * the sysctl table. The data and maxlen fields of the ctl_table * struct enable minimal validation of the values being written to be -- cgit v1.2.3 From 7cc13edc139108bb527b692f0548dce6bc648572 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 5 Nov 2006 23:52:13 -0800 Subject: [PATCH] sysctl: implement CTL_UNNUMBERED This patch takes the CTL_UNNUMBERD concept from NFS and makes it available to all new sysctl users. At the same time the sysctl binary interface maintenance documentation is updated to mention and to describe what is needed to successfully maintain the sysctl binary interface. Signed-off-by: Eric W. Biederman Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/lockd/svc.c | 3 --- fs/nfs/sysctl.c | 5 ----- include/linux/sysctl.h | 14 +++++++++++--- 3 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include/linux/sysctl.h') diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 634139232aaf..8ca18085e68d 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -353,9 +353,6 @@ EXPORT_SYMBOL(lockd_down); * Sysctl parameters (same as module parameters, different interface). */ -/* Something that isn't CTL_ANY, CTL_NONE or a value that may clash. */ -#define CTL_UNNUMBERED -2 - static ctl_table nlm_sysctls[] = { { .ctl_name = CTL_UNNUMBERED, diff --git a/fs/nfs/sysctl.c b/fs/nfs/sysctl.c index 2fe3403c2409..3ea50ac64820 100644 --- a/fs/nfs/sysctl.c +++ b/fs/nfs/sysctl.c @@ -18,11 +18,6 @@ static const int nfs_set_port_min = 0; static const int nfs_set_port_max = 65535; static struct ctl_table_header *nfs_callback_sysctl_table; -/* - * Something that isn't CTL_ANY, CTL_NONE or a value that may clash. - * Use the same values as fs/lockd/svc.c - */ -#define CTL_UNNUMBERED -2 static ctl_table nfs_cb_sysctls[] = { #ifdef CONFIG_NFS_V4 diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index c184732a70fc..d98562f1df76 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -6,10 +6,17 @@ **************************************************************** **************************************************************** ** + ** WARNING: ** The values in this file are exported to user space via - ** the sysctl() binary interface. However this interface - ** is unstable and deprecated and will be removed in the future. - ** For a stable interface use /proc/sys. + ** the sysctl() binary interface. Do *NOT* change the + ** numbering of any existing values here, and do not change + ** any numbers within any one set of values. If you have to + ** have to redefine an existing interface, use a new number for it. + ** The kernel will then return -ENOTDIR to any application using + ** the old binary interface. + ** + ** For new interfaces unless you really need a binary number + ** please use CTL_UNNUMBERED. ** **************************************************************** **************************************************************** @@ -48,6 +55,7 @@ struct __sysctl_args { #ifdef __KERNEL__ #define CTL_ANY -1 /* Matches any name */ #define CTL_NONE 0 +#define CTL_UNNUMBERED CTL_NONE /* sysctl without a binary number */ #endif enum -- cgit v1.2.3 From 3ff825b28d3345ef381eceae22bf9d92231f23dc Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 9 Nov 2006 16:32:06 -0800 Subject: [TCP]: Add tcp_available_congestion_control sysctl. Create /proc/sys/net/ipv4/tcp_available_congestion_control that reflects currently available TCP choices. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 6 ++++++ include/linux/sysctl.h | 1 + include/net/tcp.h | 4 ++++ net/ipv4/sysctl_net_ipv4.c | 24 ++++++++++++++++++++++++ net/ipv4/tcp_cong.c | 16 ++++++++++++++++ 5 files changed, 51 insertions(+) (limited to 'include/linux/sysctl.h') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index fd3c0c012351..db4280856588 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -351,10 +351,16 @@ tcp_frto - BOOLEAN where packet loss is typically due to random radio interference rather than intermediate router congestion. +tcp_available_congestion_control - STRING + Shows the available congestion control choices that are registered. + More congestion control algorithms may be available as modules, + but not loaded. + tcp_congestion_control - STRING Set the congestion control algorithm to be used for new connections. The algorithm "reno" is always available, but additional choices may be available based on kernel configuration. + Default is set as part of kernel configuration. somaxconn - INTEGER Limit of socket listen() backlog, known in userspace as SOMAXCONN. diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index d98562f1df76..28a48279654d 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -426,6 +426,7 @@ enum NET_CIPSOV4_CACHE_BUCKET_SIZE=119, NET_CIPSOV4_RBM_OPTFMT=120, NET_CIPSOV4_RBM_STRICTVALID=121, + NET_TCP_AVAIL_CONG_CONTROL=122, }; enum { diff --git a/include/net/tcp.h b/include/net/tcp.h index 246916c2321e..6af4baf5b769 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -620,6 +620,9 @@ enum tcp_ca_event { * Interface for adding new TCP congestion control handlers */ #define TCP_CA_NAME_MAX 16 +#define TCP_CA_MAX 128 +#define TCP_CA_BUF_MAX (TCP_CA_NAME_MAX*TCP_CA_MAX) + struct tcp_congestion_ops { struct list_head list; @@ -659,6 +662,7 @@ extern void tcp_init_congestion_control(struct sock *sk); extern void tcp_cleanup_congestion_control(struct sock *sk); extern int tcp_set_default_congestion_control(const char *name); extern void tcp_get_default_congestion_control(char *name); +extern void tcp_get_available_congestion_control(char *buf, size_t len); extern int tcp_set_congestion_control(struct sock *sk, const char *name); extern void tcp_slow_start(struct tcp_sock *tp); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 15061b314411..2e770f45d829 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -129,6 +129,23 @@ static int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, return ret; } +static int proc_tcp_available_congestion_control(ctl_table *ctl, + int write, struct file * filp, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX, }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + tcp_get_available_congestion_control(tbl.data, TCP_CA_BUF_MAX); + ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos); + kfree(tbl.data); + return ret; +} + ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_TCP_TIMESTAMPS, @@ -731,6 +748,13 @@ ctl_table ipv4_table[] = { .proc_handler = &proc_dointvec, }, #endif /* CONFIG_NETLABEL */ + { + .ctl_name = NET_TCP_AVAIL_CONG_CONTROL, + .procname = "tcp_available_congestion_control", + .maxlen = TCP_CA_BUF_MAX, + .mode = 0444, + .proc_handler = &proc_tcp_available_congestion_control, + }, { .ctl_name = 0 } }; diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 1e2982f4acd4..d846d7b95e1f 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -139,6 +139,22 @@ static int __init tcp_congestion_default(void) late_initcall(tcp_congestion_default); +/* Build string with list of available congestion control values */ +void tcp_get_available_congestion_control(char *buf, size_t maxlen) +{ + struct tcp_congestion_ops *ca; + size_t offs = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(ca, &tcp_cong_list, list) { + offs += snprintf(buf + offs, maxlen - offs, + "%s%s", + offs == 0 ? "" : " ", ca->name); + + } + rcu_read_unlock(); +} + /* Get current default congestion control */ void tcp_get_default_congestion_control(char *name) { -- cgit v1.2.3 From ce7bc3bf15cbf5dc5a5587ccb6b04c5b4dde4336 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 9 Nov 2006 16:35:15 -0800 Subject: [TCP]: Restrict congestion control choices. Allow normal users to only choose among a restricted set of congestion control choices. The default is reno and what ever has been configured as default. But the policy can be changed by administrator at any time. For example, to allow any choice: cp /proc/sys/net/ipv4/tcp_available_congestion_control \ /proc/sys/net/ipv4/tcp_allowed_congestion_control Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 6 ++++ include/linux/sysctl.h | 1 + include/net/tcp.h | 3 ++ net/ipv4/sysctl_net_ipv4.c | 52 ++++++++++++++++++++++++++++ net/ipv4/tcp_cong.c | 63 ++++++++++++++++++++++++++++++++++ 5 files changed, 125 insertions(+) (limited to 'include/linux/sysctl.h') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index db4280856588..bbcc8deda172 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -351,6 +351,12 @@ tcp_frto - BOOLEAN where packet loss is typically due to random radio interference rather than intermediate router congestion. +tcp_allowed_congestion_control - STRING + Show/set the congestion control choices available to non-privileged + processes. The list is a subset of those listed in + tcp_available_congestion_control. + Default is "reno" and the default setting (tcp_congestion_control). + tcp_available_congestion_control - STRING Shows the available congestion control choices that are registered. More congestion control algorithms may be available as modules, diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 28a48279654d..0725441621d0 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -427,6 +427,7 @@ enum NET_CIPSOV4_RBM_OPTFMT=120, NET_CIPSOV4_RBM_STRICTVALID=121, NET_TCP_AVAIL_CONG_CONTROL=122, + NET_TCP_ALLOWED_CONG_CONTROL=123, }; enum { diff --git a/include/net/tcp.h b/include/net/tcp.h index 6af4baf5b769..e1a5d29d0a1f 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -625,6 +625,7 @@ enum tcp_ca_event { struct tcp_congestion_ops { struct list_head list; + int non_restricted; /* initialize private data (optional) */ void (*init)(struct sock *sk); @@ -663,6 +664,8 @@ extern void tcp_cleanup_congestion_control(struct sock *sk); extern int tcp_set_default_congestion_control(const char *name); extern void tcp_get_default_congestion_control(char *name); extern void tcp_get_available_congestion_control(char *buf, size_t len); +extern void tcp_get_allowed_congestion_control(char *buf, size_t len); +extern int tcp_set_allowed_congestion_control(char *allowed); extern int tcp_set_congestion_control(struct sock *sk, const char *name); extern void tcp_slow_start(struct tcp_sock *tp); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 2e770f45d829..dfcf47f10f88 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -146,6 +146,50 @@ static int proc_tcp_available_congestion_control(ctl_table *ctl, return ret; } +static int proc_allowed_congestion_control(ctl_table *ctl, + int write, struct file * filp, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + + tcp_get_allowed_congestion_control(tbl.data, tbl.maxlen); + ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos); + if (write && ret == 0) + ret = tcp_set_allowed_congestion_control(tbl.data); + kfree(tbl.data); + return ret; +} + +static int strategy_allowed_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) +{ + ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + + tcp_get_available_congestion_control(tbl.data, tbl.maxlen); + ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen, + context); + if (ret == 0 && newval && newlen) + ret = tcp_set_allowed_congestion_control(tbl.data); + kfree(tbl.data); + + return ret; + +} + ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_TCP_TIMESTAMPS, @@ -755,6 +799,14 @@ ctl_table ipv4_table[] = { .mode = 0444, .proc_handler = &proc_tcp_available_congestion_control, }, + { + .ctl_name = NET_TCP_ALLOWED_CONG_CONTROL, + .procname = "tcp_allowed_congestion_control", + .maxlen = TCP_CA_BUF_MAX, + .mode = 0644, + .proc_handler = &proc_allowed_congestion_control, + .strategy = &strategy_allowed_congestion_control, + }, { .ctl_name = 0 } }; diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index d846d7b95e1f..343d6197c92e 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -123,6 +123,7 @@ int tcp_set_default_congestion_control(const char *name) #endif if (ca) { + ca->non_restricted = 1; /* default is always allowed */ list_move(&ca->list, &tcp_cong_list); ret = 0; } @@ -168,6 +169,64 @@ void tcp_get_default_congestion_control(char *name) rcu_read_unlock(); } +/* Built list of non-restricted congestion control values */ +void tcp_get_allowed_congestion_control(char *buf, size_t maxlen) +{ + struct tcp_congestion_ops *ca; + size_t offs = 0; + + *buf = '\0'; + rcu_read_lock(); + list_for_each_entry_rcu(ca, &tcp_cong_list, list) { + if (!ca->non_restricted) + continue; + offs += snprintf(buf + offs, maxlen - offs, + "%s%s", + offs == 0 ? "" : " ", ca->name); + + } + rcu_read_unlock(); +} + +/* Change list of non-restricted congestion control */ +int tcp_set_allowed_congestion_control(char *val) +{ + struct tcp_congestion_ops *ca; + char *clone, *name; + int ret = 0; + + clone = kstrdup(val, GFP_USER); + if (!clone) + return -ENOMEM; + + spin_lock(&tcp_cong_list_lock); + /* pass 1 check for bad entries */ + while ((name = strsep(&clone, " ")) && *name) { + ca = tcp_ca_find(name); + if (!ca) { + ret = -ENOENT; + goto out; + } + } + + /* pass 2 clear */ + list_for_each_entry_rcu(ca, &tcp_cong_list, list) + ca->non_restricted = 0; + + /* pass 3 mark as allowed */ + while ((name = strsep(&val, " ")) && *name) { + ca = tcp_ca_find(name); + WARN_ON(!ca); + if (ca) + ca->non_restricted = 1; + } +out: + spin_unlock(&tcp_cong_list_lock); + + return ret; +} + + /* Change congestion control for socket */ int tcp_set_congestion_control(struct sock *sk, const char *name) { @@ -183,6 +242,9 @@ int tcp_set_congestion_control(struct sock *sk, const char *name) if (!ca) err = -ENOENT; + else if (!(ca->non_restricted || capable(CAP_NET_ADMIN))) + err = -EPERM; + else if (!try_module_get(ca->owner)) err = -EBUSY; @@ -284,6 +346,7 @@ EXPORT_SYMBOL_GPL(tcp_reno_min_cwnd); struct tcp_congestion_ops tcp_reno = { .name = "reno", + .non_restricted = 1, .owner = THIS_MODULE, .ssthresh = tcp_reno_ssthresh, .cong_avoid = tcp_reno_cong_avoid, -- cgit v1.2.3 From 2e2e9e92bd723244ea20fa488b1780111f2b05e1 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 13 Nov 2006 13:23:52 -0200 Subject: [DCCP]: Add sysctls to control retransmission behaviour This adds 3 sysctls which govern the retransmission behaviour of DCCP control packets (3way handshake, feature negotiation). It removes 4 FIXMEs from the code. The close resemblance of sysctl variables to their TCP analogues is emphasised not only by their name, but also by giving them the same initial values. This is useful since there is not much practical experience with DCCP yet. Furthermore, with regard to the previous patch, it is now possible to limit the number of keepalive-Responses by setting net.dccp.default.request_retries (also a bit like in TCP). Lastly, added documentation of all existing DCCP sysctls. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- Documentation/networking/dccp.txt | 41 +++++++++++++++++++++++++++++++++++++++ include/linux/sysctl.h | 3 +++ net/dccp/dccp.h | 11 +++++++++++ net/dccp/feat.h | 7 ------- net/dccp/proto.c | 1 + net/dccp/sysctl.c | 25 ++++++++++++++++++++++++ net/dccp/timer.c | 16 +++++++++------ 7 files changed, 91 insertions(+), 13 deletions(-) (limited to 'include/linux/sysctl.h') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index a8142a81038a..c2328c862e98 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -63,6 +63,47 @@ DCCP_SOCKOPT_SEND_CSCOV is for the receiver and has a different meaning: it coverage value are also acceptable. The higher the number, the more restrictive this setting (see [RFC 4340, sec. 9.2.1]). +Sysctl variables +================ +Several DCCP default parameters can be managed by the following sysctls +(sysctl net.dccp.default or /proc/sys/net/dccp/default): + +request_retries + The number of active connection initiation retries (the number of + Requests minus one) before timing out. In addition, it also governs + the behaviour of the other, passive side: this variable also sets + the number of times DCCP repeats sending a Response when the initial + handshake does not progress from RESPOND to OPEN (i.e. when no Ack + is received after the initial Request). This value should be greater + than 0, suggested is less than 10. Analogue of tcp_syn_retries. + +retries1 + How often a DCCP Response is retransmitted until the listening DCCP + side considers its connecting peer dead. Analogue of tcp_retries1. + +retries2 + The number of times a general DCCP packet is retransmitted. This has + importance for retransmitted acknowledgments and feature negotiation, + data packets are never retransmitted. Analogue of tcp_retries2. + +send_ndp = 1 + Whether or not to send NDP count options (sec. 7.7.2). + +send_ackvec = 1 + Whether or not to send Ack Vector options (sec. 11.5). + +ack_ratio = 2 + The default Ack Ratio (sec. 11.3) to use. + +tx_ccid = 2 + Default CCID for the sender-receiver half-connection. + +rx_ccid = 2 + Default CCID for the receiver-sender half-connection. + +seq_window = 100 + The initial sequence window (sec. 7.5.2). + Notes ===== diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 0725441621d0..2e8c5ad82793 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -614,6 +614,9 @@ enum { NET_DCCP_DEFAULT_ACK_RATIO = 4, NET_DCCP_DEFAULT_SEND_ACKVEC = 5, NET_DCCP_DEFAULT_SEND_NDP = 6, + NET_DCCP_DEFAULT_REQ_RETRIES = 7, + NET_DCCP_DEFAULT_RETRIES1 = 8, + NET_DCCP_DEFAULT_RETRIES2 = 9, }; /* /proc/sys/net/ipx */ diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index d5c414bf7819..e7b96074a1b1 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -64,6 +64,17 @@ extern void dccp_time_wait(struct sock *sk, int state, int timeo); #define DCCP_XMIT_TIMEO 30000 /* Time/msecs for blocking transmit per packet */ +/* sysctl variables for DCCP */ +extern int sysctl_dccp_request_retries; +extern int sysctl_dccp_retries1; +extern int sysctl_dccp_retries2; +extern int dccp_feat_default_sequence_window; +extern int dccp_feat_default_rx_ccid; +extern int dccp_feat_default_tx_ccid; +extern int dccp_feat_default_ack_ratio; +extern int dccp_feat_default_send_ack_vector; +extern int dccp_feat_default_send_ndp_count; + /* is seq1 < seq2 ? */ static inline int before48(const u64 seq1, const u64 seq2) { diff --git a/net/dccp/feat.h b/net/dccp/feat.h index cee553d416ca..6048373c7186 100644 --- a/net/dccp/feat.h +++ b/net/dccp/feat.h @@ -26,11 +26,4 @@ extern void dccp_feat_clean(struct dccp_minisock *dmsk); extern int dccp_feat_clone(struct sock *oldsk, struct sock *newsk); extern int dccp_feat_init(struct dccp_minisock *dmsk); -extern int dccp_feat_default_sequence_window; -extern int dccp_feat_default_rx_ccid; -extern int dccp_feat_default_tx_ccid; -extern int dccp_feat_default_ack_ratio; -extern int dccp_feat_default_send_ack_vector; -extern int dccp_feat_default_send_ndp_count; - #endif /* _DCCP_FEAT_H */ diff --git a/net/dccp/proto.c b/net/dccp/proto.c index db54e557eff1..9c9c08cffdaf 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -212,6 +212,7 @@ int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized) dccp_init_xmit_timers(sk); icsk->icsk_rto = DCCP_TIMEOUT_INIT; + icsk->icsk_syn_retries = sysctl_dccp_request_retries; sk->sk_state = DCCP_CLOSED; sk->sk_write_space = dccp_write_space; icsk->icsk_sync_mss = dccp_sync_mss; diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 38bc157876f3..7b09f2179985 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -11,6 +11,7 @@ #include #include +#include "dccp.h" #include "feat.h" #ifndef CONFIG_SYSCTL @@ -66,6 +67,30 @@ static struct ctl_table dccp_default_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .ctl_name = NET_DCCP_DEFAULT_REQ_RETRIES, + .procname = "request_retries", + .data = &sysctl_dccp_request_retries, + .maxlen = sizeof(sysctl_dccp_request_retries), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .ctl_name = NET_DCCP_DEFAULT_RETRIES1, + .procname = "retries1", + .data = &sysctl_dccp_retries1, + .maxlen = sizeof(sysctl_dccp_retries1), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .ctl_name = NET_DCCP_DEFAULT_RETRIES2, + .procname = "retries2", + .data = &sysctl_dccp_retries2, + .maxlen = sizeof(sysctl_dccp_retries2), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { .ctl_name = 0, } }; diff --git a/net/dccp/timer.c b/net/dccp/timer.c index bda0af639ae4..7b3f16e29a97 100644 --- a/net/dccp/timer.c +++ b/net/dccp/timer.c @@ -15,6 +15,11 @@ #include "dccp.h" +/* sysctl variables governing numbers of retransmission attempts */ +int sysctl_dccp_request_retries __read_mostly = TCP_SYN_RETRIES; +int sysctl_dccp_retries1 __read_mostly = TCP_RETR1; +int sysctl_dccp_retries2 __read_mostly = TCP_RETR2; + static void dccp_write_timer(unsigned long data); static void dccp_keepalive_timer(unsigned long data); static void dccp_delack_timer(unsigned long data); @@ -44,11 +49,10 @@ static int dccp_write_timeout(struct sock *sk) if (sk->sk_state == DCCP_REQUESTING || sk->sk_state == DCCP_PARTOPEN) { if (icsk->icsk_retransmits != 0) dst_negative_advice(&sk->sk_dst_cache); - retry_until = icsk->icsk_syn_retries ? : - /* FIXME! */ 3 /* FIXME! sysctl_tcp_syn_retries */; + retry_until = icsk->icsk_syn_retries ? + : sysctl_dccp_request_retries; } else { - if (icsk->icsk_retransmits >= - /* FIXME! sysctl_tcp_retries1 */ 5 /* FIXME! */) { + if (icsk->icsk_retransmits >= sysctl_dccp_retries1) { /* NOTE. draft-ietf-tcpimpl-pmtud-01.txt requires pmtu black hole detection. :-( @@ -72,7 +76,7 @@ static int dccp_write_timeout(struct sock *sk) dst_negative_advice(&sk->sk_dst_cache); } - retry_until = /* FIXME! */ 15 /* FIXME! sysctl_tcp_retries2 */; + retry_until = sysctl_dccp_retries2; /* * FIXME: see tcp_write_timout and tcp_out_of_resources */ @@ -196,7 +200,7 @@ backoff: icsk->icsk_rto = min(icsk->icsk_rto << 1, DCCP_RTO_MAX); inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, DCCP_RTO_MAX); - if (icsk->icsk_retransmits > 3 /* FIXME: sysctl_dccp_retries1 */) + if (icsk->icsk_retransmits > sysctl_dccp_retries1) __sk_dst_reset(sk); out:; } -- cgit v1.2.3 From 82e3ab9dbeebd5c8d5402ad1607d22086271a56d Mon Sep 17 00:00:00 2001 From: Ian McDonald Date: Mon, 20 Nov 2006 19:19:32 -0200 Subject: [DCCP]: Adds the tx buffer sysctls This one got lost on the way from Ian to Gerrit to me, fix it. Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- Documentation/networking/dccp.txt | 4 ++++ include/linux/sysctl.h | 1 + net/dccp/sysctl.c | 9 +++++++++ 3 files changed, 14 insertions(+) (limited to 'include/linux/sysctl.h') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index 1910d097a0ad..dda15886bcb5 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -100,6 +100,10 @@ rx_ccid = 2 seq_window = 100 The initial sequence window (sec. 7.5.2). +tx_qlen = 5 + The size of the transmit buffer in packets. A value of 0 corresponds + to an unbounded transmit buffer. + Notes ===== diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 2e8c5ad82793..61dd99c6c2a5 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -617,6 +617,7 @@ enum { NET_DCCP_DEFAULT_REQ_RETRIES = 7, NET_DCCP_DEFAULT_RETRIES1 = 8, NET_DCCP_DEFAULT_RETRIES2 = 9, + NET_DCCP_DEFAULT_TX_QLEN = 10, }; /* /proc/sys/net/ipx */ diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 8b62061e5701..4775ba3faa04 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -91,6 +91,15 @@ static struct ctl_table dccp_default_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .ctl_name = NET_DCCP_DEFAULT_TX_QLEN, + .procname = "tx_qlen", + .data = &sysctl_dccp_tx_qlen, + .maxlen = sizeof(sysctl_dccp_tx_qlen), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { .ctl_name = 0, } }; -- cgit v1.2.3 From 4384260443efe90a2ec0d907568dbc58ae792cd0 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 28 Nov 2006 18:14:10 -0200 Subject: [DCCP]: Remove allocation of sysctl numbers This is in response to a request sent earlier by Eric W. Biederman and replaces all sysctl numbers for net.dccp.default with CTL_UNNUMBERED. It has been tested to compile and to work. Commiter note: I've removed the use of CTL_UNNUMBERED, not setting .ctl_name sets it to 0, that is the what CTL_UNNUMBERED is, reason is to avoid unneeded source code cluttering. Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/sysctl.h | 14 -------------- net/dccp/sysctl.c | 10 ---------- 2 files changed, 24 deletions(-) (limited to 'include/linux/sysctl.h') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 61dd99c6c2a5..94316a98e0d0 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -606,20 +606,6 @@ enum { NET_DCCP_DEFAULT=1, }; -/* /proc/sys/net/dccp/default */ -enum { - NET_DCCP_DEFAULT_SEQ_WINDOW = 1, - NET_DCCP_DEFAULT_RX_CCID = 2, - NET_DCCP_DEFAULT_TX_CCID = 3, - NET_DCCP_DEFAULT_ACK_RATIO = 4, - NET_DCCP_DEFAULT_SEND_ACKVEC = 5, - NET_DCCP_DEFAULT_SEND_NDP = 6, - NET_DCCP_DEFAULT_REQ_RETRIES = 7, - NET_DCCP_DEFAULT_RETRIES1 = 8, - NET_DCCP_DEFAULT_RETRIES2 = 9, - NET_DCCP_DEFAULT_TX_QLEN = 10, -}; - /* /proc/sys/net/ipx */ enum { NET_IPX_PPROP_BROADCASTING=1, diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 4775ba3faa04..fdcfca3e9208 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -20,7 +20,6 @@ static struct ctl_table dccp_default_table[] = { { - .ctl_name = NET_DCCP_DEFAULT_SEQ_WINDOW, .procname = "seq_window", .data = &sysctl_dccp_feat_sequence_window, .maxlen = sizeof(sysctl_dccp_feat_sequence_window), @@ -28,7 +27,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_RX_CCID, .procname = "rx_ccid", .data = &sysctl_dccp_feat_rx_ccid, .maxlen = sizeof(sysctl_dccp_feat_rx_ccid), @@ -36,7 +34,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_TX_CCID, .procname = "tx_ccid", .data = &sysctl_dccp_feat_tx_ccid, .maxlen = sizeof(sysctl_dccp_feat_tx_ccid), @@ -44,7 +41,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_ACK_RATIO, .procname = "ack_ratio", .data = &sysctl_dccp_feat_ack_ratio, .maxlen = sizeof(sysctl_dccp_feat_ack_ratio), @@ -52,7 +48,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_SEND_ACKVEC, .procname = "send_ackvec", .data = &sysctl_dccp_feat_send_ack_vector, .maxlen = sizeof(sysctl_dccp_feat_send_ack_vector), @@ -60,7 +55,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_SEND_NDP, .procname = "send_ndp", .data = &sysctl_dccp_feat_send_ndp_count, .maxlen = sizeof(sysctl_dccp_feat_send_ndp_count), @@ -68,7 +62,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_REQ_RETRIES, .procname = "request_retries", .data = &sysctl_dccp_request_retries, .maxlen = sizeof(sysctl_dccp_request_retries), @@ -76,7 +69,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_RETRIES1, .procname = "retries1", .data = &sysctl_dccp_retries1, .maxlen = sizeof(sysctl_dccp_retries1), @@ -84,7 +76,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_RETRIES2, .procname = "retries2", .data = &sysctl_dccp_retries2, .maxlen = sizeof(sysctl_dccp_retries2), @@ -92,7 +83,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_TX_QLEN, .procname = "tx_qlen", .data = &sysctl_dccp_tx_qlen, .maxlen = sizeof(sysctl_dccp_tx_qlen), -- cgit v1.2.3