linux/include/net/sock.h

1602 lines
45 KiB
C
Raw Normal View History

/*
* INET An implementation of the TCP/IP protocol suite for the LINUX
* operating system. INET is implemented using the BSD Socket
* interface as the means of communication with the user level.
*
* Definitions for the AF_INET socket handler.
*
* Version: @(#)sock.h 1.0.4 05/13/93
*
* Authors: Ross Biro
* Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
* Corey Minyard <wf-rch!minyard@relay.EU.net>
* Florian La Roche <flla@stud.uni-sb.de>
*
* Fixes:
* Alan Cox : Volatiles in skbuff pointers. See
* skbuff comments. May be overdone,
* better to prove they can be removed
* than the reverse.
* Alan Cox : Added a zapped field for tcp to note
* a socket is reset and must stay shut up
* Alan Cox : New fields for options
* Pauline Middelink : identd support
* Alan Cox : Eliminate low level recv/recvfrom
* David S. Miller : New socket lookup architecture.
* Steve Whitehouse: Default routines for sock_ops
* Arnaldo C. Melo : removed net_pinfo, tp_pinfo and made
* protinfo be just a void pointer, as the
* protocol specific parts were moved to
* respective headers and ipv4/v6, etc now
* use private slabcaches for its socks
* Pedro Hortas : New flags field for socket options
*
*
* 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 _SOCK_H
#define _SOCK_H
#include <linux/kernel.h>
#include <linux/list.h>
#include <linux/list_nulls.h>
#include <linux/timer.h>
#include <linux/cache.h>
#include <linux/module.h>
#include <linux/lockdep.h>
#include <linux/netdevice.h>
#include <linux/skbuff.h> /* struct sk_buff */
#include <linux/mm.h>
#include <linux/security.h>
#include <linux/filter.h>
#include <linux/rculist_nulls.h>
net: adding memory barrier to the poll and receive callbacks Adding memory barrier after the poll_wait function, paired with receive callbacks. Adding fuctions sock_poll_wait and sk_has_sleeper to wrap the memory barrier. Without the memory barrier, following race can happen. The race fires, when following code paths meet, and the tp->rcv_nxt and __add_wait_queue updates stay in CPU caches. CPU1 CPU2 sys_select receive packet ... ... __add_wait_queue update tp->rcv_nxt ... ... tp->rcv_nxt check sock_def_readable ... { schedule ... if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) wake_up_interruptible(sk->sk_sleep) ... } If there was no cache the code would work ok, since the wait_queue and rcv_nxt are opposit to each other. Meaning that once tp->rcv_nxt is updated by CPU2, the CPU1 either already passed the tp->rcv_nxt check and sleeps, or will get the new value for tp->rcv_nxt and will return with new data mask. In both cases the process (CPU1) is being added to the wait queue, so the waitqueue_active (CPU2) call cannot miss and will wake up CPU1. The bad case is when the __add_wait_queue changes done by CPU1 stay in its cache, and so does the tp->rcv_nxt update on CPU2 side. The CPU1 will then endup calling schedule and sleep forever if there are no more data on the socket. Calls to poll_wait in following modules were ommited: net/bluetooth/af_bluetooth.c net/irda/af_irda.c net/irda/irnet/irnet_ppp.c net/mac80211/rc80211_pid_debugfs.c net/phonet/socket.c net/rds/af_rds.c net/rfkill/core.c net/sunrpc/cache.c net/sunrpc/rpc_pipe.c net/tipc/socket.c Signed-off-by: Jiri Olsa <jolsa@redhat.com> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-08 12:09:13 +00:00
#include <linux/poll.h>
#include <asm/atomic.h>
#include <net/dst.h>
#include <net/checksum.h>
/*
* This structure really needs to be cleaned up.
* Most of it is for TCP, and not used by any of
* the other protocols.
*/
/* Define this to get the SOCK_DBG debugging facility. */
#define SOCK_DEBUGGING
#ifdef SOCK_DEBUGGING
#define SOCK_DEBUG(sk, msg...) do { if ((sk) && sock_flag((sk), SOCK_DBG)) \
printk(KERN_DEBUG msg); } while (0)
#else
/* Validate arguments and do nothing */
static void inline int __attribute__ ((format (printf, 2, 3)))
SOCK_DEBUG(struct sock *sk, const char *msg, ...)
{
}
#endif
/* This is the per-socket lock. The spinlock provides a synchronization
* between user contexts and software interrupt processing, whereas the
* mini-semaphore synchronizes multiple users amongst themselves.
*/
typedef struct {
spinlock_t slock;
int owned;
wait_queue_head_t wq;
/*
* We express the mutex-alike socket_lock semantics
* to the lock validator by explicitly managing
* the slock as a lock variant (in addition to
* the slock itself):
*/
#ifdef CONFIG_DEBUG_LOCK_ALLOC
struct lockdep_map dep_map;
#endif
} socket_lock_t;
struct sock;
struct proto;
struct net;
/**
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* struct sock_common - minimal network layer representation of sockets
* @skc_node: main hash linkage for various protocol lookup tables
* @skc_nulls_node: main hash linkage for UDP/UDP-Lite protocol
* @skc_refcnt: reference count
* @skc_hash: hash value used with various protocol lookup tables
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* @skc_family: network address family
* @skc_state: Connection state
* @skc_reuse: %SO_REUSEADDR setting
* @skc_bound_dev_if: bound device index if != 0
* @skc_bind_node: bind hash linkage for various protocol lookup tables
* @skc_prot: protocol handlers inside a network family
* @skc_net: reference to the network namespace of this socket
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
*
* This is the minimal network layer representation of sockets, the header
* for struct sock and struct inet_timewait_sock.
*/
struct sock_common {
/*
* first fields are not copied in sock_copy()
*/
union {
struct hlist_node skc_node;
struct hlist_nulls_node skc_nulls_node;
};
atomic_t skc_refcnt;
[INET]: speedup inet (tcp/dccp) lookups Arnaldo and I agreed it could be applied now, because I have other pending patches depending on this one (Thank you Arnaldo) (The other important patch moves skc_refcnt in a separate cache line, so that the SMP/NUMA performance doesnt suffer from cache line ping pongs) 1) First some performance data : -------------------------------- tcp_v4_rcv() wastes a *lot* of time in __inet_lookup_established() The most time critical code is : sk_for_each(sk, node, &head->chain) { if (INET_MATCH(sk, acookie, saddr, daddr, ports, dif)) goto hit; /* You sunk my battleship! */ } The sk_for_each() does use prefetch() hints but only the begining of "struct sock" is prefetched. As INET_MATCH first comparison uses inet_sk(__sk)->daddr, wich is far away from the begining of "struct sock", it has to bring into CPU cache cold cache line. Each iteration has to use at least 2 cache lines. This can be problematic if some chains are very long. 2) The goal ----------- The idea I had is to change things so that INET_MATCH() may return FALSE in 99% of cases only using the data already in the CPU cache, using one cache line per iteration. 3) Description of the patch --------------------------- Adds a new 'unsigned int skc_hash' field in 'struct sock_common', filling a 32 bits hole on 64 bits platform. struct sock_common { unsigned short skc_family; volatile unsigned char skc_state; unsigned char skc_reuse; int skc_bound_dev_if; struct hlist_node skc_node; struct hlist_node skc_bind_node; atomic_t skc_refcnt; + unsigned int skc_hash; struct proto *skc_prot; }; Store in this 32 bits field the full hash, not masked by (ehash_size - 1) Using this full hash as the first comparison done in INET_MATCH permits us immediatly skip the element without touching a second cache line in case of a miss. Suppress the sk_hashent/tw_hashent fields since skc_hash (aliased to sk_hash and tw_hash) already contains the slot number if we mask with (ehash_size - 1) File include/net/inet_hashtables.h 64 bits platforms : #define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\ (((__sk)->sk_hash == (__hash)) ((*((__u64 *)&(inet_sk(__sk)->daddr)))== (__cookie)) && \ ((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) 32bits platforms: #define TCP_IPV4_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\ (((__sk)->sk_hash == (__hash)) && \ (inet_sk(__sk)->daddr == (__saddr)) && \ (inet_sk(__sk)->rcv_saddr == (__daddr)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) - Adds a prefetch(head->chain.first) in __inet_lookup_established()/__tcp_v4_check_established() and __inet6_lookup_established()/__tcp_v6_check_established() and __dccp_v4_check_established() to bring into cache the first element of the list, before the {read|write}_lock(&head->lock); Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-10-03 21:13:38 +00:00
unsigned int skc_hash;
unsigned short skc_family;
volatile unsigned char skc_state;
unsigned char skc_reuse;
int skc_bound_dev_if;
struct hlist_node skc_bind_node;
struct proto *skc_prot;
#ifdef CONFIG_NET_NS
struct net *skc_net;
#endif
};
/**
* struct sock - network layer representation of sockets
* @__sk_common: shared layout with inet_timewait_sock
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* @sk_shutdown: mask of %SEND_SHUTDOWN and/or %RCV_SHUTDOWN
* @sk_userlocks: %SO_SNDBUF and %SO_RCVBUF settings
* @sk_lock: synchronizer
* @sk_rcvbuf: size of receive buffer in bytes
* @sk_sleep: sock wait queue
* @sk_dst_cache: destination cache
* @sk_dst_lock: destination cache lock
* @sk_policy: flow policy
* @sk_rmem_alloc: receive queue bytes committed
* @sk_receive_queue: incoming packets
* @sk_wmem_alloc: transmit queue bytes committed
* @sk_write_queue: Packet sending queue
* @sk_async_wait_queue: DMA copied packets
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* @sk_omem_alloc: "o" is "option" or "other"
* @sk_wmem_queued: persistent queue size
* @sk_forward_alloc: space allocated forward
* @sk_allocation: allocation mode
* @sk_sndbuf: size of send buffer in bytes
* @sk_flags: %SO_LINGER (l_onoff), %SO_BROADCAST, %SO_KEEPALIVE,
* %SO_OOBINLINE settings, %SO_TIMESTAMPING settings
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* @sk_no_check: %SO_NO_CHECK setting, wether or not checkup packets
* @sk_route_caps: route capabilities (e.g. %NETIF_F_TSO)
* @sk_gso_type: GSO type (e.g. %SKB_GSO_TCPV4)
[NET]: Add per-connection option to set max TSO frame size Update: My mailer ate one of Jarek's feedback mails... Fixed the parameter in netif_set_gso_max_size() to be u32, not u16. Fixed the whitespace issue due to a patch import botch. Changed the types from u32 to unsigned int to be more consistent with other variables in the area. Also brought the patch up to the latest net-2.6.26 tree. Update: Made gso_max_size container 32 bits, not 16. Moved the location of gso_max_size within netdev to be less hotpath. Made more consistent names between the sock and netdev layers, and added a define for the max GSO size. Update: Respun for net-2.6.26 tree. Update: changed max_gso_frame_size and sk_gso_max_size from signed to unsigned - thanks Stephen! This patch adds the ability for device drivers to control the size of the TSO frames being sent to them, per TCP connection. By setting the netdevice's gso_max_size value, the socket layer will set the GSO frame size based on that value. This will propogate into the TCP layer, and send TSO's of that size to the hardware. This can be desirable to help tune the bursty nature of TSO on a per-adapter basis, where one may have 1 GbE and 10 GbE devices coexisting in a system, one running multiqueue and the other not, etc. This can also be desirable for devices that cannot support full 64 KB TSO's, but still want to benefit from some level of segmentation offloading. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-03-21 10:43:19 +00:00
* @sk_gso_max_size: Maximum GSO segment size to build
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* @sk_lingertime: %SO_LINGER l_linger setting
* @sk_backlog: always used with the per-socket spinlock held
* @sk_callback_lock: used with the callbacks in the end of this struct
* @sk_error_queue: rarely used
* @sk_prot_creator: sk_prot of original sock creator (see ipv6_setsockopt,
* IPV6_ADDRFORM for instance)
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* @sk_err: last error
* @sk_err_soft: errors that don't cause failure but are the cause of a
* persistent failure not just 'timed out'
* @sk_drops: raw/udp drops counter
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* @sk_ack_backlog: current listen backlog
* @sk_max_ack_backlog: listen backlog set in listen()
* @sk_priority: %SO_PRIORITY setting
* @sk_type: socket type (%SOCK_STREAM, etc)
* @sk_protocol: which protocol this socket belongs in this network family
* @sk_peercred: %SO_PEERCRED setting
* @sk_rcvlowat: %SO_RCVLOWAT setting
* @sk_rcvtimeo: %SO_RCVTIMEO setting
* @sk_sndtimeo: %SO_SNDTIMEO setting
* @sk_filter: socket filtering instructions
* @sk_protinfo: private area, net family specific, when not using slab
* @sk_timer: sock cleanup timer
* @sk_stamp: time stamp of last packet received
* @sk_socket: Identd and reporting IO signals
* @sk_user_data: RPC layer private data
* @sk_sndmsg_page: cached page for sendmsg
* @sk_sndmsg_off: cached offset for sendmsg
* @sk_send_head: front of stuff to transmit
* @sk_security: used by security modules
* @sk_mark: generic packet mark
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* @sk_write_pending: a write to stream socket waits to start
* @sk_state_change: callback to indicate change in the state of the sock
* @sk_data_ready: callback to indicate there is data to be processed
* @sk_write_space: callback to indicate there is bf sending space available
* @sk_error_report: callback to indicate errors (e.g. %MSG_ERRQUEUE)
* @sk_backlog_rcv: callback to process the backlog
* @sk_destruct: called at sock freeing time, i.e. when all refcnt == 0
*/
struct sock {
/*
* Now struct inet_timewait_sock also uses sock_common, so please just
* don't add nothing before this first member (__sk_common) --acme
*/
struct sock_common __sk_common;
#define sk_node __sk_common.skc_node
#define sk_nulls_node __sk_common.skc_nulls_node
#define sk_refcnt __sk_common.skc_refcnt
#define sk_copy_start __sk_common.skc_hash
#define sk_hash __sk_common.skc_hash
#define sk_family __sk_common.skc_family
#define sk_state __sk_common.skc_state
#define sk_reuse __sk_common.skc_reuse
#define sk_bound_dev_if __sk_common.skc_bound_dev_if
#define sk_bind_node __sk_common.skc_bind_node
#define sk_prot __sk_common.skc_prot
#define sk_net __sk_common.skc_net
net: annotate struct sock bitfield 2009/2/24 Ingo Molnar <mingo@elte.hu>: > ok, this is the last warning i have from today's overnight -tip > testruns - a 32-bit system warning in sock_init_data(): > > [ 2.610389] NET: Registered protocol family 16 > [ 2.616138] initcall netlink_proto_init+0x0/0x170 returned 0 after 7812 usecs > [ 2.620010] WARNING: kmemcheck: Caught 32-bit read from uninitialized memory (f642c184) > [ 2.624002] 010000000200000000000000604990c000000000000000000000000000000000 > [ 2.634076] i i i i i i u u i i i i i i i i i i i i i i i i i i i i i i i i > [ 2.641038] ^ > [ 2.643376] > [ 2.644004] Pid: 1, comm: swapper Not tainted (2.6.29-rc6-tip-01751-g4d1c22c-dirty #885) > [ 2.648003] EIP: 0060:[<c07141a1>] EFLAGS: 00010282 CPU: 0 > [ 2.652008] EIP is at sock_init_data+0xa1/0x190 > [ 2.656003] EAX: 0001a800 EBX: f6836c00 ECX: 00463000 EDX: c0e46fe0 > [ 2.660003] ESI: f642c180 EDI: c0b83088 EBP: f6863ed8 ESP: c0c412ec > [ 2.664003] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 > [ 2.668003] CR0: 8005003b CR2: f682c400 CR3: 00b91000 CR4: 000006f0 > [ 2.672003] DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 > [ 2.676003] DR6: ffff4ff0 DR7: 00000400 > [ 2.680002] [<c07423e5>] __netlink_create+0x35/0xa0 > [ 2.684002] [<c07443cc>] netlink_kernel_create+0x4c/0x140 > [ 2.688002] [<c072755e>] rtnetlink_net_init+0x1e/0x40 > [ 2.696002] [<c071b601>] register_pernet_operations+0x11/0x30 > [ 2.700002] [<c071b72c>] register_pernet_subsys+0x1c/0x30 > [ 2.704002] [<c0bf3c8c>] rtnetlink_init+0x4c/0x100 > [ 2.708002] [<c0bf4669>] netlink_proto_init+0x159/0x170 > [ 2.712002] [<c0101124>] do_one_initcall+0x24/0x150 > [ 2.716002] [<c0bbf3c7>] do_initcalls+0x27/0x40 > [ 2.723201] [<c0bbf3fc>] do_basic_setup+0x1c/0x20 > [ 2.728002] [<c0bbfb8a>] kernel_init+0x5a/0xa0 > [ 2.732002] [<c0103e47>] kernel_thread_helper+0x7/0x10 > [ 2.736002] [<ffffffff>] 0xffffffff We fix this false positive by annotating the bitfield in struct sock. Reported-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
2009-02-26 13:46:57 +00:00
kmemcheck_bitfield_begin(flags);
unsigned char sk_shutdown : 2,
sk_no_check : 2,
sk_userlocks : 4;
net: annotate struct sock bitfield 2009/2/24 Ingo Molnar <mingo@elte.hu>: > ok, this is the last warning i have from today's overnight -tip > testruns - a 32-bit system warning in sock_init_data(): > > [ 2.610389] NET: Registered protocol family 16 > [ 2.616138] initcall netlink_proto_init+0x0/0x170 returned 0 after 7812 usecs > [ 2.620010] WARNING: kmemcheck: Caught 32-bit read from uninitialized memory (f642c184) > [ 2.624002] 010000000200000000000000604990c000000000000000000000000000000000 > [ 2.634076] i i i i i i u u i i i i i i i i i i i i i i i i i i i i i i i i > [ 2.641038] ^ > [ 2.643376] > [ 2.644004] Pid: 1, comm: swapper Not tainted (2.6.29-rc6-tip-01751-g4d1c22c-dirty #885) > [ 2.648003] EIP: 0060:[<c07141a1>] EFLAGS: 00010282 CPU: 0 > [ 2.652008] EIP is at sock_init_data+0xa1/0x190 > [ 2.656003] EAX: 0001a800 EBX: f6836c00 ECX: 00463000 EDX: c0e46fe0 > [ 2.660003] ESI: f642c180 EDI: c0b83088 EBP: f6863ed8 ESP: c0c412ec > [ 2.664003] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 > [ 2.668003] CR0: 8005003b CR2: f682c400 CR3: 00b91000 CR4: 000006f0 > [ 2.672003] DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 > [ 2.676003] DR6: ffff4ff0 DR7: 00000400 > [ 2.680002] [<c07423e5>] __netlink_create+0x35/0xa0 > [ 2.684002] [<c07443cc>] netlink_kernel_create+0x4c/0x140 > [ 2.688002] [<c072755e>] rtnetlink_net_init+0x1e/0x40 > [ 2.696002] [<c071b601>] register_pernet_operations+0x11/0x30 > [ 2.700002] [<c071b72c>] register_pernet_subsys+0x1c/0x30 > [ 2.704002] [<c0bf3c8c>] rtnetlink_init+0x4c/0x100 > [ 2.708002] [<c0bf4669>] netlink_proto_init+0x159/0x170 > [ 2.712002] [<c0101124>] do_one_initcall+0x24/0x150 > [ 2.716002] [<c0bbf3c7>] do_initcalls+0x27/0x40 > [ 2.723201] [<c0bbf3fc>] do_basic_setup+0x1c/0x20 > [ 2.728002] [<c0bbfb8a>] kernel_init+0x5a/0xa0 > [ 2.732002] [<c0103e47>] kernel_thread_helper+0x7/0x10 > [ 2.736002] [<ffffffff>] 0xffffffff We fix this false positive by annotating the bitfield in struct sock. Reported-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
2009-02-26 13:46:57 +00:00
kmemcheck_bitfield_end(flags);
unsigned char sk_protocol;
unsigned short sk_type;
int sk_rcvbuf;
socket_lock_t sk_lock;
/*
* The backlog queue is special, it is always used with
* the per-socket spinlock held and requires low latency
* access. Therefore we special case it's implementation.
*/
struct {
struct sk_buff *head;
struct sk_buff *tail;
} sk_backlog;
wait_queue_head_t *sk_sleep;
struct dst_entry *sk_dst_cache;
#ifdef CONFIG_XFRM
struct xfrm_policy *sk_policy[2];
#endif
rwlock_t sk_dst_lock;
atomic_t sk_rmem_alloc;
atomic_t sk_wmem_alloc;
atomic_t sk_omem_alloc;
int sk_sndbuf;
struct sk_buff_head sk_receive_queue;
struct sk_buff_head sk_write_queue;
#ifdef CONFIG_NET_DMA
struct sk_buff_head sk_async_wait_queue;
#endif
int sk_wmem_queued;
int sk_forward_alloc;
gfp_t sk_allocation;
int sk_route_caps;
int sk_gso_type;
[NET]: Add per-connection option to set max TSO frame size Update: My mailer ate one of Jarek's feedback mails... Fixed the parameter in netif_set_gso_max_size() to be u32, not u16. Fixed the whitespace issue due to a patch import botch. Changed the types from u32 to unsigned int to be more consistent with other variables in the area. Also brought the patch up to the latest net-2.6.26 tree. Update: Made gso_max_size container 32 bits, not 16. Moved the location of gso_max_size within netdev to be less hotpath. Made more consistent names between the sock and netdev layers, and added a define for the max GSO size. Update: Respun for net-2.6.26 tree. Update: changed max_gso_frame_size and sk_gso_max_size from signed to unsigned - thanks Stephen! This patch adds the ability for device drivers to control the size of the TSO frames being sent to them, per TCP connection. By setting the netdevice's gso_max_size value, the socket layer will set the GSO frame size based on that value. This will propogate into the TCP layer, and send TSO's of that size to the hardware. This can be desirable to help tune the bursty nature of TSO on a per-adapter basis, where one may have 1 GbE and 10 GbE devices coexisting in a system, one running multiqueue and the other not, etc. This can also be desirable for devices that cannot support full 64 KB TSO's, but still want to benefit from some level of segmentation offloading. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-03-21 10:43:19 +00:00
unsigned int sk_gso_max_size;
int sk_rcvlowat;
unsigned long sk_flags;
unsigned long sk_lingertime;
struct sk_buff_head sk_error_queue;
struct proto *sk_prot_creator;
rwlock_t sk_callback_lock;
int sk_err,
sk_err_soft;
atomic_t sk_drops;
unsigned short sk_ack_backlog;
unsigned short sk_max_ack_backlog;
__u32 sk_priority;
struct ucred sk_peercred;
long sk_rcvtimeo;
long sk_sndtimeo;
struct sk_filter *sk_filter;
void *sk_protinfo;
struct timer_list sk_timer;
ktime_t sk_stamp;
struct socket *sk_socket;
void *sk_user_data;
struct page *sk_sndmsg_page;
struct sk_buff *sk_send_head;
__u32 sk_sndmsg_off;
int sk_write_pending;
#ifdef CONFIG_SECURITY
void *sk_security;
#endif
__u32 sk_mark;
/* XXX 4 bytes hole on 64 bit */
void (*sk_state_change)(struct sock *sk);
void (*sk_data_ready)(struct sock *sk, int bytes);
void (*sk_write_space)(struct sock *sk);
void (*sk_error_report)(struct sock *sk);
int (*sk_backlog_rcv)(struct sock *sk,
struct sk_buff *skb);
void (*sk_destruct)(struct sock *sk);
};
/*
* Hashed lists helper routines
*/
static inline struct sock *__sk_head(const struct hlist_head *head)
{
return hlist_entry(head->first, struct sock, sk_node);
}
static inline struct sock *sk_head(const struct hlist_head *head)
{
return hlist_empty(head) ? NULL : __sk_head(head);
}
static inline struct sock *__sk_nulls_head(const struct hlist_nulls_head *head)
{
return hlist_nulls_entry(head->first, struct sock, sk_nulls_node);
}
static inline struct sock *sk_nulls_head(const struct hlist_nulls_head *head)
{
return hlist_nulls_empty(head) ? NULL : __sk_nulls_head(head);
}
static inline struct sock *sk_next(const struct sock *sk)
{
return sk->sk_node.next ?
hlist_entry(sk->sk_node.next, struct sock, sk_node) : NULL;
}
static inline struct sock *sk_nulls_next(const struct sock *sk)
{
return (!is_a_nulls(sk->sk_nulls_node.next)) ?
hlist_nulls_entry(sk->sk_nulls_node.next,
struct sock, sk_nulls_node) :
NULL;
}
static inline int sk_unhashed(const struct sock *sk)
{
return hlist_unhashed(&sk->sk_node);
}
static inline int sk_hashed(const struct sock *sk)
{
return !sk_unhashed(sk);
}
static __inline__ void sk_node_init(struct hlist_node *node)
{
node->pprev = NULL;
}
static __inline__ void sk_nulls_node_init(struct hlist_nulls_node *node)
{
node->pprev = NULL;
}
static __inline__ void __sk_del_node(struct sock *sk)
{
__hlist_del(&sk->sk_node);
}
static __inline__ int __sk_del_node_init(struct sock *sk)
{
if (sk_hashed(sk)) {
__sk_del_node(sk);
sk_node_init(&sk->sk_node);
return 1;
}
return 0;
}
/* Grab socket reference count. This operation is valid only
when sk is ALREADY grabbed f.e. it is found in hash table
or a list and the lookup is made under lock preventing hash table
modifications.
*/
static inline void sock_hold(struct sock *sk)
{
atomic_inc(&sk->sk_refcnt);
}
/* Ungrab socket in the context, which assumes that socket refcnt
cannot hit zero, f.e. it is true in context of any socketcall.
*/
static inline void __sock_put(struct sock *sk)
{
atomic_dec(&sk->sk_refcnt);
}
static __inline__ int sk_del_node_init(struct sock *sk)
{
int rc = __sk_del_node_init(sk);
if (rc) {
/* paranoid for a while -acme */
WARN_ON(atomic_read(&sk->sk_refcnt) == 1);
__sock_put(sk);
}
return rc;
}
static __inline__ int __sk_nulls_del_node_init_rcu(struct sock *sk)
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
{
if (sk_hashed(sk)) {
hlist_nulls_del_init_rcu(&sk->sk_nulls_node);
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
return 1;
}
return 0;
}
static __inline__ int sk_nulls_del_node_init_rcu(struct sock *sk)
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
{
int rc = __sk_nulls_del_node_init_rcu(sk);
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
if (rc) {
/* paranoid for a while -acme */
WARN_ON(atomic_read(&sk->sk_refcnt) == 1);
__sock_put(sk);
}
return rc;
}
static __inline__ void __sk_add_node(struct sock *sk, struct hlist_head *list)
{
hlist_add_head(&sk->sk_node, list);
}
static __inline__ void sk_add_node(struct sock *sk, struct hlist_head *list)
{
sock_hold(sk);
__sk_add_node(sk, list);
}
static __inline__ void __sk_nulls_add_node_rcu(struct sock *sk, struct hlist_nulls_head *list)
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
{
hlist_nulls_add_head_rcu(&sk->sk_nulls_node, list);
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
}
static __inline__ void sk_nulls_add_node_rcu(struct sock *sk, struct hlist_nulls_head *list)
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
{
sock_hold(sk);
__sk_nulls_add_node_rcu(sk, list);
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
}
static __inline__ void __sk_del_bind_node(struct sock *sk)
{
__hlist_del(&sk->sk_bind_node);
}
static __inline__ void sk_add_bind_node(struct sock *sk,
struct hlist_head *list)
{
hlist_add_head(&sk->sk_bind_node, list);
}
#define sk_for_each(__sk, node, list) \
hlist_for_each_entry(__sk, node, list, sk_node)
#define sk_nulls_for_each(__sk, node, list) \
hlist_nulls_for_each_entry(__sk, node, list, sk_nulls_node)
#define sk_nulls_for_each_rcu(__sk, node, list) \
hlist_nulls_for_each_entry_rcu(__sk, node, list, sk_nulls_node)
#define sk_for_each_from(__sk, node) \
if (__sk && ({ node = &(__sk)->sk_node; 1; })) \
hlist_for_each_entry_from(__sk, node, sk_node)
#define sk_nulls_for_each_from(__sk, node) \
if (__sk && ({ node = &(__sk)->sk_nulls_node; 1; })) \
hlist_nulls_for_each_entry_from(__sk, node, sk_nulls_node)
#define sk_for_each_continue(__sk, node) \
if (__sk && ({ node = &(__sk)->sk_node; 1; })) \
hlist_for_each_entry_continue(__sk, node, sk_node)
#define sk_for_each_safe(__sk, node, tmp, list) \
hlist_for_each_entry_safe(__sk, node, tmp, list, sk_node)
#define sk_for_each_bound(__sk, node, list) \
hlist_for_each_entry(__sk, node, list, sk_bind_node)
/* Sock flags */
enum sock_flags {
SOCK_DEAD,
SOCK_DONE,
SOCK_URGINLINE,
SOCK_KEEPOPEN,
SOCK_LINGER,
SOCK_DESTROY,
SOCK_BROADCAST,
SOCK_TIMESTAMP,
SOCK_ZAPPED,
SOCK_USE_WRITE_QUEUE, /* whether to call sk->sk_write_space in sock_wfree */
SOCK_DBG, /* %SO_DEBUG setting */
SOCK_RCVTSTAMP, /* %SO_TIMESTAMP setting */
SOCK_RCVTSTAMPNS, /* %SO_TIMESTAMPNS setting */
SOCK_LOCALROUTE, /* route locally only, %SO_DONTROUTE setting */
SOCK_QUEUE_SHRUNK, /* write queue has been shrunk recently */
SOCK_TIMESTAMPING_TX_HARDWARE, /* %SOF_TIMESTAMPING_TX_HARDWARE */
SOCK_TIMESTAMPING_TX_SOFTWARE, /* %SOF_TIMESTAMPING_TX_SOFTWARE */
SOCK_TIMESTAMPING_RX_HARDWARE, /* %SOF_TIMESTAMPING_RX_HARDWARE */
SOCK_TIMESTAMPING_RX_SOFTWARE, /* %SOF_TIMESTAMPING_RX_SOFTWARE */
SOCK_TIMESTAMPING_SOFTWARE, /* %SOF_TIMESTAMPING_SOFTWARE */
SOCK_TIMESTAMPING_RAW_HARDWARE, /* %SOF_TIMESTAMPING_RAW_HARDWARE */
SOCK_TIMESTAMPING_SYS_HARDWARE, /* %SOF_TIMESTAMPING_SYS_HARDWARE */
};
static inline void sock_copy_flags(struct sock *nsk, struct sock *osk)
{
nsk->sk_flags = osk->sk_flags;
}
static inline void sock_set_flag(struct sock *sk, enum sock_flags flag)
{
__set_bit(flag, &sk->sk_flags);
}
static inline void sock_reset_flag(struct sock *sk, enum sock_flags flag)
{
__clear_bit(flag, &sk->sk_flags);
}
static inline int sock_flag(struct sock *sk, enum sock_flags flag)
{
return test_bit(flag, &sk->sk_flags);
}
static inline void sk_acceptq_removed(struct sock *sk)
{
sk->sk_ack_backlog--;
}
static inline void sk_acceptq_added(struct sock *sk)
{
sk->sk_ack_backlog++;
}
static inline int sk_acceptq_is_full(struct sock *sk)
{
return sk->sk_ack_backlog > sk->sk_max_ack_backlog;
}
/*
* Compute minimal free write space needed to queue new packets.
*/
static inline int sk_stream_min_wspace(struct sock *sk)
{
return sk->sk_wmem_queued >> 1;
}
static inline int sk_stream_wspace(struct sock *sk)
{
return sk->sk_sndbuf - sk->sk_wmem_queued;
}
extern void sk_stream_write_space(struct sock *sk);
static inline int sk_stream_memory_free(struct sock *sk)
{
return sk->sk_wmem_queued < sk->sk_sndbuf;
}
/* The per-socket spinlock must be held here. */
static inline void sk_add_backlog(struct sock *sk, struct sk_buff *skb)
{
if (!sk->sk_backlog.tail) {
sk->sk_backlog.head = sk->sk_backlog.tail = skb;
} else {
sk->sk_backlog.tail->next = skb;
sk->sk_backlog.tail = skb;
}
skb->next = NULL;
}
static inline int sk_backlog_rcv(struct sock *sk, struct sk_buff *skb)
{
return sk->sk_backlog_rcv(sk, skb);
}
#define sk_wait_event(__sk, __timeo, __condition) \
({ int __rc; \
release_sock(__sk); \
__rc = __condition; \
if (!__rc) { \
*(__timeo) = schedule_timeout(*(__timeo)); \
} \
lock_sock(__sk); \
__rc = __condition; \
__rc; \
})
extern int sk_stream_wait_connect(struct sock *sk, long *timeo_p);
extern int sk_stream_wait_memory(struct sock *sk, long *timeo_p);
extern void sk_stream_wait_close(struct sock *sk, long timeo_p);
extern int sk_stream_error(struct sock *sk, int flags, int err);
extern void sk_stream_kill_queues(struct sock *sk);
extern int sk_wait_data(struct sock *sk, long *timeo);
struct request_sock_ops;
struct timewait_sock_ops;
[SOCK] proto: Add hashinfo member to struct proto This way we can remove TCP and DCCP specific versions of sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port sk->sk_prot->hash: inet_hash is directly used, only v6 need a specific version to deal with mapped sockets sk->sk_prot->unhash: both v4 and v6 use inet_hash directly struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so that inet_csk_get_port can find the per family routine. Now only the lookup routines receive as a parameter a struct inet_hashtable. With this we further reuse code, reducing the difference among INET transport protocols. Eventually work has to be done on UDP and SCTP to make them share this infrastructure and get as a bonus inet_diag interfaces so that iproute can be used with these protocols. net-2.6/net/ipv4/inet_hashtables.c: struct proto | +8 struct inet_connection_sock_af_ops | +8 2 structs changed __inet_hash_nolisten | +18 __inet_hash | -210 inet_put_port | +8 inet_bind_bucket_create | +1 __inet_hash_connect | -8 5 functions changed, 27 bytes added, 218 bytes removed, diff: -191 net-2.6/net/core/sock.c: proto_seq_show | +3 1 function changed, 3 bytes added, diff: +3 net-2.6/net/ipv4/inet_connection_sock.c: inet_csk_get_port | +15 1 function changed, 15 bytes added, diff: +15 net-2.6/net/ipv4/tcp.c: tcp_set_state | -7 1 function changed, 7 bytes removed, diff: -7 net-2.6/net/ipv4/tcp_ipv4.c: tcp_v4_get_port | -31 tcp_v4_hash | -48 tcp_v4_destroy_sock | -7 tcp_v4_syn_recv_sock | -2 tcp_unhash | -179 5 functions changed, 267 bytes removed, diff: -267 net-2.6/net/ipv6/inet6_hashtables.c: __inet6_hash | +8 1 function changed, 8 bytes added, diff: +8 net-2.6/net/ipv4/inet_hashtables.c: inet_unhash | +190 inet_hash | +242 2 functions changed, 432 bytes added, diff: +432 vmlinux: 16 functions changed, 485 bytes added, 492 bytes removed, diff: -7 /home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c: tcp_v6_get_port | -31 tcp_v6_hash | -7 tcp_v6_syn_recv_sock | -9 3 functions changed, 47 bytes removed, diff: -47 /home/acme/git/net-2.6/net/dccp/proto.c: dccp_destroy_sock | -7 dccp_unhash | -179 dccp_hash | -49 dccp_set_state | -7 dccp_done | +1 5 functions changed, 1 bytes added, 242 bytes removed, diff: -241 /home/acme/git/net-2.6/net/dccp/ipv4.c: dccp_v4_get_port | -31 dccp_v4_request_recv_sock | -2 2 functions changed, 33 bytes removed, diff: -33 /home/acme/git/net-2.6/net/dccp/ipv6.c: dccp_v6_get_port | -31 dccp_v6_hash | -7 dccp_v6_request_recv_sock | +5 3 functions changed, 5 bytes added, 38 bytes removed, diff: -33 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 12:06:04 +00:00
struct inet_hashinfo;
struct raw_hashinfo;
/* Networking protocol blocks we attach to sockets.
* socket layer -> transport layer interface
* transport -> network interface is defined by struct inet_proto
*/
struct proto {
void (*close)(struct sock *sk,
long timeout);
int (*connect)(struct sock *sk,
struct sockaddr *uaddr,
int addr_len);
int (*disconnect)(struct sock *sk, int flags);
struct sock * (*accept) (struct sock *sk, int flags, int *err);
int (*ioctl)(struct sock *sk, int cmd,
unsigned long arg);
int (*init)(struct sock *sk);
void (*destroy)(struct sock *sk);
void (*shutdown)(struct sock *sk, int how);
int (*setsockopt)(struct sock *sk, int level,
int optname, char __user *optval,
int optlen);
int (*getsockopt)(struct sock *sk, int level,
int optname, char __user *optval,
int __user *option);
#ifdef CONFIG_COMPAT
int (*compat_setsockopt)(struct sock *sk,
int level,
int optname, char __user *optval,
int optlen);
int (*compat_getsockopt)(struct sock *sk,
int level,
int optname, char __user *optval,
int __user *option);
#endif
int (*sendmsg)(struct kiocb *iocb, struct sock *sk,
struct msghdr *msg, size_t len);
int (*recvmsg)(struct kiocb *iocb, struct sock *sk,
struct msghdr *msg,
size_t len, int noblock, int flags,
int *addr_len);
int (*sendpage)(struct sock *sk, struct page *page,
int offset, size_t size, int flags);
int (*bind)(struct sock *sk,
struct sockaddr *uaddr, int addr_len);
int (*backlog_rcv) (struct sock *sk,
struct sk_buff *skb);
/* Keeping track of sk's, looking them up, and port selection methods. */
void (*hash)(struct sock *sk);
void (*unhash)(struct sock *sk);
int (*get_port)(struct sock *sk, unsigned short snum);
/* Keeping track of sockets in use */
#ifdef CONFIG_PROC_FS
unsigned int inuse_idx;
#endif
/* Memory pressure */
void (*enter_memory_pressure)(struct sock *sk);
atomic_t *memory_allocated; /* Current allocated memory. */
struct percpu_counter *sockets_allocated; /* Current number of sockets. */
/*
* Pressure flag: try to collapse.
* Technical note: it is used by multiple contexts non atomically.
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
* All the __sk_mem_schedule() is of this nature: accounting
* is strict, actions are advisory and have some latency.
*/
int *memory_pressure;
int *sysctl_mem;
int *sysctl_wmem;
int *sysctl_rmem;
int max_header;
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
struct kmem_cache *slab;
unsigned int obj_size;
udp: RCU handling for Unicast packets. Goals are : 1) Optimizing handling of incoming Unicast UDP frames, so that no memory writes should happen in the fast path. Note: Multicasts and broadcasts still will need to take a lock, because doing a full lockless lookup in this case is difficult. 2) No expensive operations in the socket bind/unhash phases : - No expensive synchronize_rcu() calls. - No added rcu_head in socket structure, increasing memory needs, but more important, forcing us to use call_rcu() calls, that have the bad property of making sockets structure cold. (rcu grace period between socket freeing and its potential reuse make this socket being cold in CPU cache). David did a previous patch using call_rcu() and noticed a 20% impact on TCP connection rates. Quoting Cristopher Lameter : "Right. That results in cacheline cooldown. You'd want to recycle the object as they are cache hot on a per cpu basis. That is screwed up by the delayed regular rcu processing. We have seen multiple regressions due to cacheline cooldown. The only choice in cacheline hot sensitive areas is to deal with the complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU." - Because udp sockets are allocated from dedicated kmem_cache, use of SLAB_DESTROY_BY_RCU can help here. Theory of operation : --------------------- As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()), special attention must be taken by readers and writers. Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed, reused, inserted in a different chain or in worst case in the same chain while readers could do lookups in the same time. In order to avoid loops, a reader must check each socket found in a chain really belongs to the chain the reader was traversing. If it finds a mismatch, lookup must start again at the begining. This *restart* loop is the reason we had to use rdlock for the multicast case, because we dont want to send same message several times to the same socket. We use RCU only for fast path. Thus, /proc/net/udp still takes spinlocks. Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-29 09:11:14 +00:00
int slab_flags;
struct percpu_counter *orphan_count;
struct request_sock_ops *rsk_prot;
struct timewait_sock_ops *twsk_prot;
union {
struct inet_hashinfo *hashinfo;
struct udp_table *udp_table;
struct raw_hashinfo *raw_hash;
} h;
[SOCK] proto: Add hashinfo member to struct proto This way we can remove TCP and DCCP specific versions of sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port sk->sk_prot->hash: inet_hash is directly used, only v6 need a specific version to deal with mapped sockets sk->sk_prot->unhash: both v4 and v6 use inet_hash directly struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so that inet_csk_get_port can find the per family routine. Now only the lookup routines receive as a parameter a struct inet_hashtable. With this we further reuse code, reducing the difference among INET transport protocols. Eventually work has to be done on UDP and SCTP to make them share this infrastructure and get as a bonus inet_diag interfaces so that iproute can be used with these protocols. net-2.6/net/ipv4/inet_hashtables.c: struct proto | +8 struct inet_connection_sock_af_ops | +8 2 structs changed __inet_hash_nolisten | +18 __inet_hash | -210 inet_put_port | +8 inet_bind_bucket_create | +1 __inet_hash_connect | -8 5 functions changed, 27 bytes added, 218 bytes removed, diff: -191 net-2.6/net/core/sock.c: proto_seq_show | +3 1 function changed, 3 bytes added, diff: +3 net-2.6/net/ipv4/inet_connection_sock.c: inet_csk_get_port | +15 1 function changed, 15 bytes added, diff: +15 net-2.6/net/ipv4/tcp.c: tcp_set_state | -7 1 function changed, 7 bytes removed, diff: -7 net-2.6/net/ipv4/tcp_ipv4.c: tcp_v4_get_port | -31 tcp_v4_hash | -48 tcp_v4_destroy_sock | -7 tcp_v4_syn_recv_sock | -2 tcp_unhash | -179 5 functions changed, 267 bytes removed, diff: -267 net-2.6/net/ipv6/inet6_hashtables.c: __inet6_hash | +8 1 function changed, 8 bytes added, diff: +8 net-2.6/net/ipv4/inet_hashtables.c: inet_unhash | +190 inet_hash | +242 2 functions changed, 432 bytes added, diff: +432 vmlinux: 16 functions changed, 485 bytes added, 492 bytes removed, diff: -7 /home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c: tcp_v6_get_port | -31 tcp_v6_hash | -7 tcp_v6_syn_recv_sock | -9 3 functions changed, 47 bytes removed, diff: -47 /home/acme/git/net-2.6/net/dccp/proto.c: dccp_destroy_sock | -7 dccp_unhash | -179 dccp_hash | -49 dccp_set_state | -7 dccp_done | +1 5 functions changed, 1 bytes added, 242 bytes removed, diff: -241 /home/acme/git/net-2.6/net/dccp/ipv4.c: dccp_v4_get_port | -31 dccp_v4_request_recv_sock | -2 2 functions changed, 33 bytes removed, diff: -33 /home/acme/git/net-2.6/net/dccp/ipv6.c: dccp_v6_get_port | -31 dccp_v6_hash | -7 dccp_v6_request_recv_sock | +5 3 functions changed, 5 bytes added, 38 bytes removed, diff: -33 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 12:06:04 +00:00
struct module *owner;
char name[32];
struct list_head node;
#ifdef SOCK_REFCNT_DEBUG
atomic_t socks;
#endif
};
extern int proto_register(struct proto *prot, int alloc_slab);
extern void proto_unregister(struct proto *prot);
#ifdef SOCK_REFCNT_DEBUG
static inline void sk_refcnt_debug_inc(struct sock *sk)
{
atomic_inc(&sk->sk_prot->socks);
}
static inline void sk_refcnt_debug_dec(struct sock *sk)
{
atomic_dec(&sk->sk_prot->socks);
printk(KERN_DEBUG "%s socket %p released, %d are still alive\n",
sk->sk_prot->name, sk, atomic_read(&sk->sk_prot->socks));
}
static inline void sk_refcnt_debug_release(const struct sock *sk)
{
if (atomic_read(&sk->sk_refcnt) != 1)
printk(KERN_DEBUG "Destruction of the %s socket %p delayed, refcnt=%d\n",
sk->sk_prot->name, sk, atomic_read(&sk->sk_refcnt));
}
#else /* SOCK_REFCNT_DEBUG */
#define sk_refcnt_debug_inc(sk) do { } while (0)
#define sk_refcnt_debug_dec(sk) do { } while (0)
#define sk_refcnt_debug_release(sk) do { } while (0)
#endif /* SOCK_REFCNT_DEBUG */
#ifdef CONFIG_PROC_FS
/* Called with local bh disabled */
extern void sock_prot_inuse_add(struct net *net, struct proto *prot, int inc);
extern int sock_prot_inuse_get(struct net *net, struct proto *proto);
#else
static void inline sock_prot_inuse_add(struct net *net, struct proto *prot,
int inc)
{
}
#endif
/* With per-bucket locks this operation is not-atomic, so that
* this version is not worse.
*/
static inline void __sk_prot_rehash(struct sock *sk)
{
sk->sk_prot->unhash(sk);
sk->sk_prot->hash(sk);
}
/* About 10 seconds */
#define SOCK_DESTROY_TIME (10*HZ)
/* Sockets 0-1023 can't be bound to unless you are superuser */
#define PROT_SOCK 1024
#define SHUTDOWN_MASK 3
#define RCV_SHUTDOWN 1
#define SEND_SHUTDOWN 2
#define SOCK_SNDBUF_LOCK 1
#define SOCK_RCVBUF_LOCK 2
#define SOCK_BINDADDR_LOCK 4
#define SOCK_BINDPORT_LOCK 8
/* sock_iocb: used to kick off async processing of socket ios */
struct sock_iocb {
struct list_head list;
int flags;
int size;
struct socket *sock;
struct sock *sk;
struct scm_cookie *scm;
struct msghdr *msg, async_msg;
struct kiocb *kiocb;
};
static inline struct sock_iocb *kiocb_to_siocb(struct kiocb *iocb)
{
return (struct sock_iocb *)iocb->private;
}
static inline struct kiocb *siocb_to_kiocb(struct sock_iocb *si)
{
return si->kiocb;
}
struct socket_alloc {
struct socket socket;
struct inode vfs_inode;
};
static inline struct socket *SOCKET_I(struct inode *inode)
{
return &container_of(inode, struct socket_alloc, vfs_inode)->socket;
}
static inline struct inode *SOCK_INODE(struct socket *socket)
{
return &container_of(socket, struct socket_alloc, socket)->vfs_inode;
}
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
/*
* Functions for memory accounting
*/
extern int __sk_mem_schedule(struct sock *sk, int size, int kind);
extern void __sk_mem_reclaim(struct sock *sk);
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
#define SK_MEM_QUANTUM ((int)PAGE_SIZE)
#define SK_MEM_QUANTUM_SHIFT ilog2(SK_MEM_QUANTUM)
#define SK_MEM_SEND 0
#define SK_MEM_RECV 1
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
static inline int sk_mem_pages(int amt)
{
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
return (amt + SK_MEM_QUANTUM - 1) >> SK_MEM_QUANTUM_SHIFT;
}
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
static inline int sk_has_account(struct sock *sk)
{
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
/* return true if protocol supports memory accounting */
return !!sk->sk_prot->memory_allocated;
}
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
static inline int sk_wmem_schedule(struct sock *sk, int size)
{
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
if (!sk_has_account(sk))
return 1;
return size <= sk->sk_forward_alloc ||
__sk_mem_schedule(sk, size, SK_MEM_SEND);
}
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
static inline int sk_rmem_schedule(struct sock *sk, int size)
{
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
if (!sk_has_account(sk))
return 1;
return size <= sk->sk_forward_alloc ||
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
__sk_mem_schedule(sk, size, SK_MEM_RECV);
}
static inline void sk_mem_reclaim(struct sock *sk)
{
if (!sk_has_account(sk))
return;
if (sk->sk_forward_alloc >= SK_MEM_QUANTUM)
__sk_mem_reclaim(sk);
}
static inline void sk_mem_reclaim_partial(struct sock *sk)
{
if (!sk_has_account(sk))
return;
if (sk->sk_forward_alloc > SK_MEM_QUANTUM)
__sk_mem_reclaim(sk);
}
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
static inline void sk_mem_charge(struct sock *sk, int size)
{
if (!sk_has_account(sk))
return;
sk->sk_forward_alloc -= size;
}
static inline void sk_mem_uncharge(struct sock *sk, int size)
{
if (!sk_has_account(sk))
return;
sk->sk_forward_alloc += size;
}
static inline void sk_wmem_free_skb(struct sock *sk, struct sk_buff *skb)
{
sock_set_flag(sk, SOCK_QUEUE_SHRUNK);
sk->sk_wmem_queued -= skb->truesize;
sk_mem_uncharge(sk, skb->truesize);
__kfree_skb(skb);
}
/* Used by processes to "lock" a socket state, so that
* interrupts and bottom half handlers won't change it
* from under us. It essentially blocks any incoming
* packets, so that we won't get any new data or any
* packets that change the state of the socket.
*
* While locked, BH processing will add new packets to
* the backlog queue. This queue is processed by the
* owner of the socket lock right before it is released.
*
* Since ~2.3.5 it is also exclusive sleep lock serializing
* accesses from user process context.
*/
#define sock_owned_by_user(sk) ((sk)->sk_lock.owned)
/*
* Macro so as to not evaluate some arguments when
* lockdep is not enabled.
*
* Mark both the sk_lock and the sk_lock.slock as a
* per-address-family lock class.
*/
#define sock_lock_init_class_and_name(sk, sname, skey, name, key) \
do { \
sk->sk_lock.owned = 0; \
init_waitqueue_head(&sk->sk_lock.wq); \
spin_lock_init(&(sk)->sk_lock.slock); \
debug_check_no_locks_freed((void *)&(sk)->sk_lock, \
sizeof((sk)->sk_lock)); \
lockdep_set_class_and_name(&(sk)->sk_lock.slock, \
(skey), (sname)); \
lockdep_init_map(&(sk)->sk_lock.dep_map, (name), (key), 0); \
} while (0)
extern void lock_sock_nested(struct sock *sk, int subclass);
static inline void lock_sock(struct sock *sk)
{
lock_sock_nested(sk, 0);
}
extern void release_sock(struct sock *sk);
/* BH context may only use the following locking interface. */
#define bh_lock_sock(__sk) spin_lock(&((__sk)->sk_lock.slock))
#define bh_lock_sock_nested(__sk) \
spin_lock_nested(&((__sk)->sk_lock.slock), \
SINGLE_DEPTH_NESTING)
#define bh_unlock_sock(__sk) spin_unlock(&((__sk)->sk_lock.slock))
extern struct sock *sk_alloc(struct net *net, int family,
gfp_t priority,
struct proto *prot);
extern void sk_free(struct sock *sk);
extern void sk_release_kernel(struct sock *sk);
extern struct sock *sk_clone(const struct sock *sk,
const gfp_t priority);
extern struct sk_buff *sock_wmalloc(struct sock *sk,
unsigned long size, int force,
gfp_t priority);
extern struct sk_buff *sock_rmalloc(struct sock *sk,
unsigned long size, int force,
gfp_t priority);
extern void sock_wfree(struct sk_buff *skb);
extern void sock_rfree(struct sk_buff *skb);
extern int sock_setsockopt(struct socket *sock, int level,
int op, char __user *optval,
int optlen);
extern int sock_getsockopt(struct socket *sock, int level,
int op, char __user *optval,
int __user *optlen);
extern struct sk_buff *sock_alloc_send_skb(struct sock *sk,
unsigned long size,
int noblock,
int *errcode);
extern struct sk_buff *sock_alloc_send_pskb(struct sock *sk,
unsigned long header_len,
unsigned long data_len,
int noblock,
int *errcode);
extern void *sock_kmalloc(struct sock *sk, int size,
gfp_t priority);
extern void sock_kfree_s(struct sock *sk, void *mem, int size);
extern void sk_send_sigurg(struct sock *sk);
/*
* Functions to fill in entries in struct proto_ops when a protocol
* does not implement a particular function.
*/
extern int sock_no_bind(struct socket *,
struct sockaddr *, int);
extern int sock_no_connect(struct socket *,
struct sockaddr *, int, int);
extern int sock_no_socketpair(struct socket *,
struct socket *);
extern int sock_no_accept(struct socket *,
struct socket *, int);
extern int sock_no_getname(struct socket *,
struct sockaddr *, int *, int);
extern unsigned int sock_no_poll(struct file *, struct socket *,
struct poll_table_struct *);
extern int sock_no_ioctl(struct socket *, unsigned int,
unsigned long);
extern int sock_no_listen(struct socket *, int);
extern int sock_no_shutdown(struct socket *, int);
extern int sock_no_getsockopt(struct socket *, int , int,
char __user *, int __user *);
extern int sock_no_setsockopt(struct socket *, int, int,
char __user *, int);
extern int sock_no_sendmsg(struct kiocb *, struct socket *,
struct msghdr *, size_t);
extern int sock_no_recvmsg(struct kiocb *, struct socket *,
struct msghdr *, size_t, int);
extern int sock_no_mmap(struct file *file,
struct socket *sock,
struct vm_area_struct *vma);
extern ssize_t sock_no_sendpage(struct socket *sock,
struct page *page,
int offset, size_t size,
int flags);
/*
* Functions to fill in entries in struct proto_ops when a protocol
* uses the inet style.
*/
extern int sock_common_getsockopt(struct socket *sock, int level, int optname,
char __user *optval, int __user *optlen);
extern int sock_common_recvmsg(struct kiocb *iocb, struct socket *sock,
struct msghdr *msg, size_t size, int flags);
extern int sock_common_setsockopt(struct socket *sock, int level, int optname,
char __user *optval, int optlen);
extern int compat_sock_common_getsockopt(struct socket *sock, int level,
int optname, char __user *optval, int __user *optlen);
extern int compat_sock_common_setsockopt(struct socket *sock, int level,
int optname, char __user *optval, int optlen);
extern void sk_common_release(struct sock *sk);
/*
* Default socket callbacks and setup code
*/
/* Initialise core socket variables */
extern void sock_init_data(struct socket *sock, struct sock *sk);
/**
* sk_filter_release: Release a socket filter
* @fp: filter to remove
*
* Remove a filter from a socket and release its resources.
*/
static inline void sk_filter_release(struct sk_filter *fp)
{
if (atomic_dec_and_test(&fp->refcnt))
kfree(fp);
}
static inline void sk_filter_uncharge(struct sock *sk, struct sk_filter *fp)
{
unsigned int size = sk_filter_len(fp);
atomic_sub(size, &sk->sk_omem_alloc);
sk_filter_release(fp);
}
static inline void sk_filter_charge(struct sock *sk, struct sk_filter *fp)
{
atomic_inc(&fp->refcnt);
atomic_add(sk_filter_len(fp), &sk->sk_omem_alloc);
}
/*
* Socket reference counting postulates.
*
* * Each user of socket SHOULD hold a reference count.
* * Each access point to socket (an hash table bucket, reference from a list,
* running timer, skb in flight MUST hold a reference count.
* * When reference count hits 0, it means it will never increase back.
* * When reference count hits 0, it means that no references from
* outside exist to this socket and current process on current CPU
* is last user and may/should destroy this socket.
* * sk_free is called from any context: process, BH, IRQ. When
* it is called, socket has no references from outside -> sk_free
* may release descendant resources allocated by the socket, but
* to the time when it is called, socket is NOT referenced by any
* hash tables, lists etc.
* * Packets, delivered from outside (from network or from another process)
* and enqueued on receive/error queues SHOULD NOT grab reference count,
* when they sit in queue. Otherwise, packets will leak to hole, when
* socket is looked up by one cpu and unhasing is made by another CPU.
* It is true for udp/raw, netlink (leak to receive and error queues), tcp
* (leak to backlog). Packet socket does all the processing inside
* BR_NETPROTO_LOCK, so that it has not this race condition. UNIX sockets
* use separate SMP lock, so that they are prone too.
*/
/* Ungrab socket and destroy it, if it was the last reference. */
static inline void sock_put(struct sock *sk)
{
if (atomic_dec_and_test(&sk->sk_refcnt))
sk_free(sk);
}
extern int sk_receive_skb(struct sock *sk, struct sk_buff *skb,
const int nested);
static inline void sk_set_socket(struct sock *sk, struct socket *sock)
{
sk->sk_socket = sock;
}
/* Detach socket from process context.
* Announce socket dead, detach it from wait queue and inode.
* Note that parent inode held reference count on this struct sock,
* we do not release it in this function, because protocol
* probably wants some additional cleanups or even continuing
* to work with this socket (TCP).
*/
static inline void sock_orphan(struct sock *sk)
{
write_lock_bh(&sk->sk_callback_lock);
sock_set_flag(sk, SOCK_DEAD);
sk_set_socket(sk, NULL);
sk->sk_sleep = NULL;
write_unlock_bh(&sk->sk_callback_lock);
}
static inline void sock_graft(struct sock *sk, struct socket *parent)
{
write_lock_bh(&sk->sk_callback_lock);
sk->sk_sleep = &parent->wait;
parent->sk = sk;
sk_set_socket(sk, parent);
security_sock_graft(sk, parent);
write_unlock_bh(&sk->sk_callback_lock);
}
extern int sock_i_uid(struct sock *sk);
extern unsigned long sock_i_ino(struct sock *sk);
static inline struct dst_entry *
__sk_dst_get(struct sock *sk)
{
return sk->sk_dst_cache;
}
static inline struct dst_entry *
sk_dst_get(struct sock *sk)
{
struct dst_entry *dst;
read_lock(&sk->sk_dst_lock);
dst = sk->sk_dst_cache;
if (dst)
dst_hold(dst);
read_unlock(&sk->sk_dst_lock);
return dst;
}
static inline void
__sk_dst_set(struct sock *sk, struct dst_entry *dst)
{
struct dst_entry *old_dst;
old_dst = sk->sk_dst_cache;
sk->sk_dst_cache = dst;
dst_release(old_dst);
}
static inline void
sk_dst_set(struct sock *sk, struct dst_entry *dst)
{
write_lock(&sk->sk_dst_lock);
__sk_dst_set(sk, dst);
write_unlock(&sk->sk_dst_lock);
}
static inline void
__sk_dst_reset(struct sock *sk)
{
struct dst_entry *old_dst;
old_dst = sk->sk_dst_cache;
sk->sk_dst_cache = NULL;
dst_release(old_dst);
}
static inline void
sk_dst_reset(struct sock *sk)
{
write_lock(&sk->sk_dst_lock);
__sk_dst_reset(sk);
write_unlock(&sk->sk_dst_lock);
}
[NET]: deinline 200+ byte inlines in sock.h Sizes in bytes (allyesconfig, i386) and files where those inlines are used: 238 sock_queue_rcv_skb 2.6.16/net/x25/x25_in.o 238 sock_queue_rcv_skb 2.6.16/net/rose/rose_in.o 238 sock_queue_rcv_skb 2.6.16/net/packet/af_packet.o 238 sock_queue_rcv_skb 2.6.16/net/netrom/nr_in.o 238 sock_queue_rcv_skb 2.6.16/net/llc/llc_sap.o 238 sock_queue_rcv_skb 2.6.16/net/llc/llc_conn.o 238 sock_queue_rcv_skb 2.6.16/net/irda/af_irda.o 238 sock_queue_rcv_skb 2.6.16/net/ipx/af_ipx.o 238 sock_queue_rcv_skb 2.6.16/net/ipv6/udp.o 238 sock_queue_rcv_skb 2.6.16/net/ipv6/raw.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/udp.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/raw.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/ipmr.o 238 sock_queue_rcv_skb 2.6.16/net/econet/econet.o 238 sock_queue_rcv_skb 2.6.16/net/econet/af_econet.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/sco.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/l2cap.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/hci_sock.o 238 sock_queue_rcv_skb 2.6.16/net/ax25/ax25_in.o 238 sock_queue_rcv_skb 2.6.16/net/ax25/af_ax25.o 238 sock_queue_rcv_skb 2.6.16/net/appletalk/ddp.o 238 sock_queue_rcv_skb 2.6.16/drivers/net/pppoe.o 276 sk_receive_skb 2.6.16/net/decnet/dn_nsp_in.o 276 sk_receive_skb 2.6.16/net/dccp/ipv6.o 276 sk_receive_skb 2.6.16/net/dccp/ipv4.o 276 sk_receive_skb 2.6.16/net/dccp/dccp_ipv6.o 276 sk_receive_skb 2.6.16/drivers/net/pppoe.o 209 sk_dst_check 2.6.16/net/ipv6/ip6_output.o 209 sk_dst_check 2.6.16/net/ipv4/udp.o 209 sk_dst_check 2.6.16/net/decnet/dn_nsp_out.o Large inlines with multiple callers: Size Uses Wasted Name and definition ===== ==== ====== ================================================ 238 21 4360 sock_queue_rcv_skb include/net/sock.h 109 10 801 sock_recv_timestamp include/net/sock.h 276 4 768 sk_receive_skb include/net/sock.h 94 8 518 __sk_dst_check include/net/sock.h 209 3 378 sk_dst_check include/net/sock.h 131 4 333 sk_setup_caps include/net/sock.h 152 2 132 sk_stream_alloc_pskb include/net/sock.h 125 2 105 sk_stream_writequeue_purge include/net/sock.h Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-03-28 09:08:21 +00:00
extern struct dst_entry *__sk_dst_check(struct sock *sk, u32 cookie);
[NET]: deinline 200+ byte inlines in sock.h Sizes in bytes (allyesconfig, i386) and files where those inlines are used: 238 sock_queue_rcv_skb 2.6.16/net/x25/x25_in.o 238 sock_queue_rcv_skb 2.6.16/net/rose/rose_in.o 238 sock_queue_rcv_skb 2.6.16/net/packet/af_packet.o 238 sock_queue_rcv_skb 2.6.16/net/netrom/nr_in.o 238 sock_queue_rcv_skb 2.6.16/net/llc/llc_sap.o 238 sock_queue_rcv_skb 2.6.16/net/llc/llc_conn.o 238 sock_queue_rcv_skb 2.6.16/net/irda/af_irda.o 238 sock_queue_rcv_skb 2.6.16/net/ipx/af_ipx.o 238 sock_queue_rcv_skb 2.6.16/net/ipv6/udp.o 238 sock_queue_rcv_skb 2.6.16/net/ipv6/raw.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/udp.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/raw.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/ipmr.o 238 sock_queue_rcv_skb 2.6.16/net/econet/econet.o 238 sock_queue_rcv_skb 2.6.16/net/econet/af_econet.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/sco.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/l2cap.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/hci_sock.o 238 sock_queue_rcv_skb 2.6.16/net/ax25/ax25_in.o 238 sock_queue_rcv_skb 2.6.16/net/ax25/af_ax25.o 238 sock_queue_rcv_skb 2.6.16/net/appletalk/ddp.o 238 sock_queue_rcv_skb 2.6.16/drivers/net/pppoe.o 276 sk_receive_skb 2.6.16/net/decnet/dn_nsp_in.o 276 sk_receive_skb 2.6.16/net/dccp/ipv6.o 276 sk_receive_skb 2.6.16/net/dccp/ipv4.o 276 sk_receive_skb 2.6.16/net/dccp/dccp_ipv6.o 276 sk_receive_skb 2.6.16/drivers/net/pppoe.o 209 sk_dst_check 2.6.16/net/ipv6/ip6_output.o 209 sk_dst_check 2.6.16/net/ipv4/udp.o 209 sk_dst_check 2.6.16/net/decnet/dn_nsp_out.o Large inlines with multiple callers: Size Uses Wasted Name and definition ===== ==== ====== ================================================ 238 21 4360 sock_queue_rcv_skb include/net/sock.h 109 10 801 sock_recv_timestamp include/net/sock.h 276 4 768 sk_receive_skb include/net/sock.h 94 8 518 __sk_dst_check include/net/sock.h 209 3 378 sk_dst_check include/net/sock.h 131 4 333 sk_setup_caps include/net/sock.h 152 2 132 sk_stream_alloc_pskb include/net/sock.h 125 2 105 sk_stream_writequeue_purge include/net/sock.h Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-03-28 09:08:21 +00:00
extern struct dst_entry *sk_dst_check(struct sock *sk, u32 cookie);
static inline int sk_can_gso(const struct sock *sk)
{
return net_gso_ok(sk->sk_route_caps, sk->sk_gso_type);
}
extern void sk_setup_caps(struct sock *sk, struct dst_entry *dst);
static inline int skb_copy_to_page(struct sock *sk, char __user *from,
struct sk_buff *skb, struct page *page,
int off, int copy)
{
if (skb->ip_summed == CHECKSUM_NONE) {
int err = 0;
__wsum csum = csum_and_copy_from_user(from,
page_address(page) + off,
copy, 0, &err);
if (err)
return err;
skb->csum = csum_block_add(skb->csum, csum, skb->len);
} else if (copy_from_user(page_address(page) + off, from, copy))
return -EFAULT;
skb->len += copy;
skb->data_len += copy;
skb->truesize += copy;
sk->sk_wmem_queued += copy;
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
sk_mem_charge(sk, copy);
return 0;
}
/**
* sk_wmem_alloc_get - returns write allocations
* @sk: socket
*
* Returns sk_wmem_alloc minus initial offset of one
*/
static inline int sk_wmem_alloc_get(const struct sock *sk)
{
return atomic_read(&sk->sk_wmem_alloc) - 1;
}
/**
* sk_rmem_alloc_get - returns read allocations
* @sk: socket
*
* Returns sk_rmem_alloc
*/
static inline int sk_rmem_alloc_get(const struct sock *sk)
{
return atomic_read(&sk->sk_rmem_alloc);
}
/**
* sk_has_allocations - check if allocations are outstanding
* @sk: socket
*
* Returns true if socket has write or read allocations
*/
static inline int sk_has_allocations(const struct sock *sk)
{
return sk_wmem_alloc_get(sk) || sk_rmem_alloc_get(sk);
}
net: adding memory barrier to the poll and receive callbacks Adding memory barrier after the poll_wait function, paired with receive callbacks. Adding fuctions sock_poll_wait and sk_has_sleeper to wrap the memory barrier. Without the memory barrier, following race can happen. The race fires, when following code paths meet, and the tp->rcv_nxt and __add_wait_queue updates stay in CPU caches. CPU1 CPU2 sys_select receive packet ... ... __add_wait_queue update tp->rcv_nxt ... ... tp->rcv_nxt check sock_def_readable ... { schedule ... if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) wake_up_interruptible(sk->sk_sleep) ... } If there was no cache the code would work ok, since the wait_queue and rcv_nxt are opposit to each other. Meaning that once tp->rcv_nxt is updated by CPU2, the CPU1 either already passed the tp->rcv_nxt check and sleeps, or will get the new value for tp->rcv_nxt and will return with new data mask. In both cases the process (CPU1) is being added to the wait queue, so the waitqueue_active (CPU2) call cannot miss and will wake up CPU1. The bad case is when the __add_wait_queue changes done by CPU1 stay in its cache, and so does the tp->rcv_nxt update on CPU2 side. The CPU1 will then endup calling schedule and sleep forever if there are no more data on the socket. Calls to poll_wait in following modules were ommited: net/bluetooth/af_bluetooth.c net/irda/af_irda.c net/irda/irnet/irnet_ppp.c net/mac80211/rc80211_pid_debugfs.c net/phonet/socket.c net/rds/af_rds.c net/rfkill/core.c net/sunrpc/cache.c net/sunrpc/rpc_pipe.c net/tipc/socket.c Signed-off-by: Jiri Olsa <jolsa@redhat.com> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-08 12:09:13 +00:00
/**
* sk_has_sleeper - check if there are any waiting processes
* @sk: socket
*
* Returns true if socket has waiting processes
*
* The purpose of the sk_has_sleeper and sock_poll_wait is to wrap the memory
* barrier call. They were added due to the race found within the tcp code.
*
* Consider following tcp code paths:
*
* CPU1 CPU2
*
* sys_select receive packet
* ... ...
* __add_wait_queue update tp->rcv_nxt
* ... ...
* tp->rcv_nxt check sock_def_readable
* ... {
* schedule ...
* if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
* wake_up_interruptible(sk->sk_sleep)
* ...
* }
*
* The race for tcp fires when the __add_wait_queue changes done by CPU1 stay
* in its cache, and so does the tp->rcv_nxt update on CPU2 side. The CPU1
* could then endup calling schedule and sleep forever if there are no more
* data on the socket.
*
* The sk_has_sleeper is always called right after a call to read_lock, so we
* can use smp_mb__after_lock barrier.
net: adding memory barrier to the poll and receive callbacks Adding memory barrier after the poll_wait function, paired with receive callbacks. Adding fuctions sock_poll_wait and sk_has_sleeper to wrap the memory barrier. Without the memory barrier, following race can happen. The race fires, when following code paths meet, and the tp->rcv_nxt and __add_wait_queue updates stay in CPU caches. CPU1 CPU2 sys_select receive packet ... ... __add_wait_queue update tp->rcv_nxt ... ... tp->rcv_nxt check sock_def_readable ... { schedule ... if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) wake_up_interruptible(sk->sk_sleep) ... } If there was no cache the code would work ok, since the wait_queue and rcv_nxt are opposit to each other. Meaning that once tp->rcv_nxt is updated by CPU2, the CPU1 either already passed the tp->rcv_nxt check and sleeps, or will get the new value for tp->rcv_nxt and will return with new data mask. In both cases the process (CPU1) is being added to the wait queue, so the waitqueue_active (CPU2) call cannot miss and will wake up CPU1. The bad case is when the __add_wait_queue changes done by CPU1 stay in its cache, and so does the tp->rcv_nxt update on CPU2 side. The CPU1 will then endup calling schedule and sleep forever if there are no more data on the socket. Calls to poll_wait in following modules were ommited: net/bluetooth/af_bluetooth.c net/irda/af_irda.c net/irda/irnet/irnet_ppp.c net/mac80211/rc80211_pid_debugfs.c net/phonet/socket.c net/rds/af_rds.c net/rfkill/core.c net/sunrpc/cache.c net/sunrpc/rpc_pipe.c net/tipc/socket.c Signed-off-by: Jiri Olsa <jolsa@redhat.com> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-08 12:09:13 +00:00
*/
static inline int sk_has_sleeper(struct sock *sk)
{
/*
* We need to be sure we are in sync with the
* add_wait_queue modifications to the wait queue.
*
* This memory barrier is paired in the sock_poll_wait.
*/
smp_mb__after_lock();
net: adding memory barrier to the poll and receive callbacks Adding memory barrier after the poll_wait function, paired with receive callbacks. Adding fuctions sock_poll_wait and sk_has_sleeper to wrap the memory barrier. Without the memory barrier, following race can happen. The race fires, when following code paths meet, and the tp->rcv_nxt and __add_wait_queue updates stay in CPU caches. CPU1 CPU2 sys_select receive packet ... ... __add_wait_queue update tp->rcv_nxt ... ... tp->rcv_nxt check sock_def_readable ... { schedule ... if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) wake_up_interruptible(sk->sk_sleep) ... } If there was no cache the code would work ok, since the wait_queue and rcv_nxt are opposit to each other. Meaning that once tp->rcv_nxt is updated by CPU2, the CPU1 either already passed the tp->rcv_nxt check and sleeps, or will get the new value for tp->rcv_nxt and will return with new data mask. In both cases the process (CPU1) is being added to the wait queue, so the waitqueue_active (CPU2) call cannot miss and will wake up CPU1. The bad case is when the __add_wait_queue changes done by CPU1 stay in its cache, and so does the tp->rcv_nxt update on CPU2 side. The CPU1 will then endup calling schedule and sleep forever if there are no more data on the socket. Calls to poll_wait in following modules were ommited: net/bluetooth/af_bluetooth.c net/irda/af_irda.c net/irda/irnet/irnet_ppp.c net/mac80211/rc80211_pid_debugfs.c net/phonet/socket.c net/rds/af_rds.c net/rfkill/core.c net/sunrpc/cache.c net/sunrpc/rpc_pipe.c net/tipc/socket.c Signed-off-by: Jiri Olsa <jolsa@redhat.com> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-08 12:09:13 +00:00
return sk->sk_sleep && waitqueue_active(sk->sk_sleep);
}
/**
* sock_poll_wait - place memory barrier behind the poll_wait call.
* @filp: file
* @wait_address: socket wait queue
* @p: poll_table
*
* See the comments in the sk_has_sleeper function.
*/
static inline void sock_poll_wait(struct file *filp,
wait_queue_head_t *wait_address, poll_table *p)
{
if (p && wait_address) {
poll_wait(filp, wait_address, p);
/*
* We need to be sure we are in sync with the
* socket flags modification.
*
* This memory barrier is paired in the sk_has_sleeper.
*/
smp_mb();
}
}
/*
* Queue a received datagram if it will fit. Stream and sequenced
* protocols can't normally use this as they need to fit buffers in
* and play with them.
*
* Inlined as it's very short and called for pretty much every
* packet ever received.
*/
static inline void skb_set_owner_w(struct sk_buff *skb, struct sock *sk)
{
skb_orphan(skb);
skb->sk = sk;
skb->destructor = sock_wfree;
net: No more expensive sock_hold()/sock_put() on each tx One of the problem with sock memory accounting is it uses a pair of sock_hold()/sock_put() for each transmitted packet. This slows down bidirectional flows because the receive path also needs to take a refcount on socket and might use a different cpu than transmit path or transmit completion path. So these two atomic operations also trigger cache line bounces. We can see this in tx or tx/rx workloads (media gateways for example), where sock_wfree() can be in top five functions in profiles. We use this sock_hold()/sock_put() so that sock freeing is delayed until all tx packets are completed. As we also update sk_wmem_alloc, we could offset sk_wmem_alloc by one unit at init time, until sk_free() is called. Once sk_free() is called, we atomic_dec_and_test(sk_wmem_alloc) to decrement initial offset and atomicaly check if any packets are in flight. skb_set_owner_w() doesnt call sock_hold() anymore sock_wfree() doesnt call sock_put() anymore, but check if sk_wmem_alloc reached 0 to perform the final freeing. Drawback is that a skb->truesize error could lead to unfreeable sockets, or even worse, prematurely calling __sk_free() on a live socket. Nice speedups on SMP. tbench for example, going from 2691 MB/s to 2711 MB/s on my 8 cpu dev machine, even if tbench was not really hitting sk_refcnt contention point. 5 % speedup on a UDP transmit workload (depends on number of flows), lowering TX completion cpu usage. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-06-11 09:55:43 +00:00
/*
* We used to take a refcount on sk, but following operation
* is enough to guarantee sk_free() wont free this sock until
* all in-flight packets are completed
*/
atomic_add(skb->truesize, &sk->sk_wmem_alloc);
}
static inline void skb_set_owner_r(struct sk_buff *skb, struct sock *sk)
{
skb_orphan(skb);
skb->sk = sk;
skb->destructor = sock_rfree;
atomic_add(skb->truesize, &sk->sk_rmem_alloc);
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
sk_mem_charge(sk, skb->truesize);
}
extern void sk_reset_timer(struct sock *sk, struct timer_list* timer,
unsigned long expires);
extern void sk_stop_timer(struct sock *sk, struct timer_list* timer);
[NET]: deinline 200+ byte inlines in sock.h Sizes in bytes (allyesconfig, i386) and files where those inlines are used: 238 sock_queue_rcv_skb 2.6.16/net/x25/x25_in.o 238 sock_queue_rcv_skb 2.6.16/net/rose/rose_in.o 238 sock_queue_rcv_skb 2.6.16/net/packet/af_packet.o 238 sock_queue_rcv_skb 2.6.16/net/netrom/nr_in.o 238 sock_queue_rcv_skb 2.6.16/net/llc/llc_sap.o 238 sock_queue_rcv_skb 2.6.16/net/llc/llc_conn.o 238 sock_queue_rcv_skb 2.6.16/net/irda/af_irda.o 238 sock_queue_rcv_skb 2.6.16/net/ipx/af_ipx.o 238 sock_queue_rcv_skb 2.6.16/net/ipv6/udp.o 238 sock_queue_rcv_skb 2.6.16/net/ipv6/raw.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/udp.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/raw.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/ipmr.o 238 sock_queue_rcv_skb 2.6.16/net/econet/econet.o 238 sock_queue_rcv_skb 2.6.16/net/econet/af_econet.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/sco.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/l2cap.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/hci_sock.o 238 sock_queue_rcv_skb 2.6.16/net/ax25/ax25_in.o 238 sock_queue_rcv_skb 2.6.16/net/ax25/af_ax25.o 238 sock_queue_rcv_skb 2.6.16/net/appletalk/ddp.o 238 sock_queue_rcv_skb 2.6.16/drivers/net/pppoe.o 276 sk_receive_skb 2.6.16/net/decnet/dn_nsp_in.o 276 sk_receive_skb 2.6.16/net/dccp/ipv6.o 276 sk_receive_skb 2.6.16/net/dccp/ipv4.o 276 sk_receive_skb 2.6.16/net/dccp/dccp_ipv6.o 276 sk_receive_skb 2.6.16/drivers/net/pppoe.o 209 sk_dst_check 2.6.16/net/ipv6/ip6_output.o 209 sk_dst_check 2.6.16/net/ipv4/udp.o 209 sk_dst_check 2.6.16/net/decnet/dn_nsp_out.o Large inlines with multiple callers: Size Uses Wasted Name and definition ===== ==== ====== ================================================ 238 21 4360 sock_queue_rcv_skb include/net/sock.h 109 10 801 sock_recv_timestamp include/net/sock.h 276 4 768 sk_receive_skb include/net/sock.h 94 8 518 __sk_dst_check include/net/sock.h 209 3 378 sk_dst_check include/net/sock.h 131 4 333 sk_setup_caps include/net/sock.h 152 2 132 sk_stream_alloc_pskb include/net/sock.h 125 2 105 sk_stream_writequeue_purge include/net/sock.h Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-03-28 09:08:21 +00:00
extern int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);
static inline int sock_queue_err_skb(struct sock *sk, struct sk_buff *skb)
{
/* Cast skb->rcvbuf to unsigned... It's pointless, but reduces
number of warnings when compiling with -W --ANK
*/
if (atomic_read(&sk->sk_rmem_alloc) + skb->truesize >=
(unsigned)sk->sk_rcvbuf)
return -ENOMEM;
skb_set_owner_r(skb, sk);
skb_queue_tail(&sk->sk_error_queue, skb);
if (!sock_flag(sk, SOCK_DEAD))
sk->sk_data_ready(sk, skb->len);
return 0;
}
/*
* Recover an error report and clear atomically
*/
static inline int sock_error(struct sock *sk)
{
int err;
if (likely(!sk->sk_err))
return 0;
err = xchg(&sk->sk_err, 0);
return -err;
}
static inline unsigned long sock_wspace(struct sock *sk)
{
int amt = 0;
if (!(sk->sk_shutdown & SEND_SHUTDOWN)) {
amt = sk->sk_sndbuf - atomic_read(&sk->sk_wmem_alloc);
if (amt < 0)
amt = 0;
}
return amt;
}
static inline void sk_wake_async(struct sock *sk, int how, int band)
{
if (sk->sk_socket && sk->sk_socket->fasync_list)
sock_wake_async(sk->sk_socket, how, band);
}
#define SOCK_MIN_SNDBUF 2048
#define SOCK_MIN_RCVBUF 256
static inline void sk_stream_moderate_sndbuf(struct sock *sk)
{
if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK)) {
sk->sk_sndbuf = min(sk->sk_sndbuf, sk->sk_wmem_queued >> 1);
sk->sk_sndbuf = max(sk->sk_sndbuf, SOCK_MIN_SNDBUF);
}
}
struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp);
static inline struct page *sk_stream_alloc_page(struct sock *sk)
{
struct page *page = NULL;
page = alloc_pages(sk->sk_allocation, 0);
if (!page) {
sk->sk_prot->enter_memory_pressure(sk);
sk_stream_moderate_sndbuf(sk);
}
return page;
}
/*
* Default write policy as shown to user space via poll/select/SIGIO
*/
static inline int sock_writeable(const struct sock *sk)
{
return atomic_read(&sk->sk_wmem_alloc) < (sk->sk_sndbuf >> 1);
}
static inline gfp_t gfp_any(void)
{
return in_softirq() ? GFP_ATOMIC : GFP_KERNEL;
}
static inline long sock_rcvtimeo(const struct sock *sk, int noblock)
{
return noblock ? 0 : sk->sk_rcvtimeo;
}
static inline long sock_sndtimeo(const struct sock *sk, int noblock)
{
return noblock ? 0 : sk->sk_sndtimeo;
}
static inline int sock_rcvlowat(const struct sock *sk, int waitall, int len)
{
return (waitall ? len : min_t(int, sk->sk_rcvlowat, len)) ? : 1;
}
/* Alas, with timeout socket operations are not restartable.
* Compare this to poll().
*/
static inline int sock_intr_errno(long timeo)
{
return timeo == MAX_SCHEDULE_TIMEOUT ? -ERESTARTSYS : -EINTR;
}
extern void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk,
struct sk_buff *skb);
static __inline__ void
sock_recv_timestamp(struct msghdr *msg, struct sock *sk, struct sk_buff *skb)
{
ktime_t kt = skb->tstamp;
struct skb_shared_hwtstamps *hwtstamps = skb_hwtstamps(skb);
/*
* generate control messages if
* - receive time stamping in software requested (SOCK_RCVTSTAMP
* or SOCK_TIMESTAMPING_RX_SOFTWARE)
* - software time stamp available and wanted
* (SOCK_TIMESTAMPING_SOFTWARE)
* - hardware time stamps available and wanted
* (SOCK_TIMESTAMPING_SYS_HARDWARE or
* SOCK_TIMESTAMPING_RAW_HARDWARE)
*/
if (sock_flag(sk, SOCK_RCVTSTAMP) ||
sock_flag(sk, SOCK_TIMESTAMPING_RX_SOFTWARE) ||
(kt.tv64 && sock_flag(sk, SOCK_TIMESTAMPING_SOFTWARE)) ||
(hwtstamps->hwtstamp.tv64 &&
sock_flag(sk, SOCK_TIMESTAMPING_RAW_HARDWARE)) ||
(hwtstamps->syststamp.tv64 &&
sock_flag(sk, SOCK_TIMESTAMPING_SYS_HARDWARE)))
__sock_recv_timestamp(msg, sk, skb);
else
sk->sk_stamp = kt;
}
/**
* sock_tx_timestamp - checks whether the outgoing packet is to be time stamped
* @msg: outgoing packet
* @sk: socket sending this packet
* @shtx: filled with instructions for time stamping
*
* Currently only depends on SOCK_TIMESTAMPING* flags. Returns error code if
* parameters are invalid.
*/
extern int sock_tx_timestamp(struct msghdr *msg,
struct sock *sk,
union skb_shared_tx *shtx);
/**
* sk_eat_skb - Release a skb if it is no longer needed
[PATCH] DocBook: changes and extensions to the kernel documentation I have recompiled Linux kernel 2.6.11.5 documentation for me and our university students again. The documentation could be extended for more sources which are equipped by structured comments for recent 2.6 kernels. I have tried to proceed with that task. I have done that more times from 2.6.0 time and it gets boring to do same changes again and again. Linux kernel compiles after changes for i386 and ARM targets. I have added references to some more files into kernel-api book, I have added some section names as well. So please, check that changes do not break something and that categories are not too much skewed. I have changed kernel-doc to accept "fastcall" and "asmlinkage" words reserved by kernel convention. Most of the other changes are modifications in the comments to make kernel-doc happy, accept some parameters description and do not bail out on errors. Changed <pid> to @pid in the description, moved some #ifdef before comments to correct function to comments bindings, etc. You can see result of the modified documentation build at http://cmp.felk.cvut.cz/~pisa/linux/lkdb-2.6.11.tar.gz Some more sources are ready to be included into kernel-doc generated documentation. Sources has been added into kernel-api for now. Some more section names added and probably some more chaos introduced as result of quick cleanup work. Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by: Martin Waitz <tali@admingilde.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 15:59:25 +00:00
* @sk: socket to eat this skb from
* @skb: socket buffer to eat
* @copied_early: flag indicating whether DMA operations copied this data early
*
* This routine must be called with interrupts disabled or with the socket
* locked so that the sk_buff queue operation is ok.
*/
#ifdef CONFIG_NET_DMA
static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_early)
{
__skb_unlink(skb, &sk->sk_receive_queue);
if (!copied_early)
__kfree_skb(skb);
else
__skb_queue_tail(&sk->sk_async_wait_queue, skb);
}
#else
static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_early)
{
__skb_unlink(skb, &sk->sk_receive_queue);
__kfree_skb(skb);
}
#endif
static inline
struct net *sock_net(const struct sock *sk)
{
#ifdef CONFIG_NET_NS
return sk->sk_net;
#else
return &init_net;
#endif
}
static inline
void sock_net_set(struct sock *sk, struct net *net)
{
#ifdef CONFIG_NET_NS
sk->sk_net = net;
#endif
}
/*
* Kernel sockets, f.e. rtnl or icmp_socket, are a part of a namespace.
* They should not hold a referrence to a namespace in order to allow
* to stop it.
* Sockets after sk_change_net should be released using sk_release_kernel
*/
static inline void sk_change_net(struct sock *sk, struct net *net)
{
put_net(sock_net(sk));
sock_net_set(sk, hold_net(net));
}
static inline struct sock *skb_steal_sock(struct sk_buff *skb)
{
if (unlikely(skb->sk)) {
struct sock *sk = skb->sk;
skb->destructor = NULL;
skb->sk = NULL;
return sk;
}
return NULL;
}
extern void sock_enable_timestamp(struct sock *sk, int flag);
extern int sock_get_timestamp(struct sock *, struct timeval __user *);
extern int sock_get_timestampns(struct sock *, struct timespec __user *);
/*
* Enable debug/info messages
*/
extern int net_msg_warn;
#define NETDEBUG(fmt, args...) \
do { if (net_msg_warn) printk(fmt,##args); } while (0)
#define LIMIT_NETDEBUG(fmt, args...) \
do { if (net_msg_warn && net_ratelimit()) printk(fmt,##args); } while(0)
extern __u32 sysctl_wmem_max;
extern __u32 sysctl_rmem_max;
extern void sk_init(void);
extern int sysctl_optmem_max;
extern __u32 sysctl_wmem_default;
extern __u32 sysctl_rmem_default;
#endif /* _SOCK_H */