linux/net/ipv4/inet_timewait_sock.c

304 lines
8.9 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* 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.
*
* Generic TIME_WAIT sockets functions
*
* From code orinally in TCP
*/
#include <linux/kernel.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 11:04:11 +03:00
#include <linux/slab.h>
#include <linux/module.h>
#include <net/inet_hashtables.h>
#include <net/inet_timewait_sock.h>
#include <net/ip.h>
/**
* inet_twsk_bind_unhash - unhash a timewait socket from bind hash
* @tw: timewait socket
* @hashinfo: hashinfo pointer
*
* unhash a timewait socket from bind hash, if hashed.
* bind hash lock must be held by caller.
* Returns 1 if caller should call inet_twsk_put() after lock release.
*/
void inet_twsk_bind_unhash(struct inet_timewait_sock *tw,
struct inet_hashinfo *hashinfo)
{
struct inet_bind_bucket *tb = tw->tw_tb;
if (!tb)
return;
__hlist_del(&tw->tw_bind_node);
tw->tw_tb = NULL;
inet_bind_bucket_destroy(hashinfo->bind_bucket_cachep, tb);
__sock_put((struct sock *)tw);
}
/* Must be called with locally disabled BHs. */
tcp/dccp: get rid of central timewait timer Using a timer wheel for timewait sockets was nice ~15 years ago when memory was expensive and machines had a single processor. This does not scale, code is ugly and source of huge latencies (Typically 30 ms have been seen, cpus spinning on death_lock spinlock.) We can afford to use an extra 64 bytes per timewait sock and spread timewait load to all cpus to have better behavior. Tested: On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1 on the target (lpaa24) Before patch : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 419594 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 437171 While test is running, we can observe 25 or even 33 ms latencies. lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2 lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2 After patch : About 90% increase of throughput : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 810442 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 800992 And latencies are kept to minimal values during this load, even if network utilization is 90% higher : lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-13 04:51:09 +03:00
static void inet_twsk_kill(struct inet_timewait_sock *tw)
{
tcp/dccp: get rid of central timewait timer Using a timer wheel for timewait sockets was nice ~15 years ago when memory was expensive and machines had a single processor. This does not scale, code is ugly and source of huge latencies (Typically 30 ms have been seen, cpus spinning on death_lock spinlock.) We can afford to use an extra 64 bytes per timewait sock and spread timewait load to all cpus to have better behavior. Tested: On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1 on the target (lpaa24) Before patch : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 419594 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 437171 While test is running, we can observe 25 or even 33 ms latencies. lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2 lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2 After patch : About 90% increase of throughput : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 810442 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 800992 And latencies are kept to minimal values during this load, even if network utilization is 90% higher : lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-13 04:51:09 +03:00
struct inet_hashinfo *hashinfo = tw->tw_dr->hashinfo;
spinlock_t *lock = inet_ehash_lockp(hashinfo, tw->tw_hash);
struct inet_bind_hashbucket *bhead;
spin_lock(lock);
sk_nulls_del_node_init_rcu((struct sock *)tw);
spin_unlock(lock);
/* Disassociate with bind bucket. */
bhead = &hashinfo->bhash[inet_bhashfn(twsk_net(tw), tw->tw_num,
hashinfo->bhash_size)];
spin_lock(&bhead->lock);
inet_twsk_bind_unhash(tw, hashinfo);
spin_unlock(&bhead->lock);
tcp: allocate tcp_death_row outside of struct netns_ipv4 I forgot tcp had per netns tracking of timewait sockets, and their sysctl to change the limit. After 0dad4087a86a ("tcp/dccp: get rid of inet_twsk_purge()"), whole struct net can be freed before last tw socket is freed. We need to allocate a separate struct inet_timewait_death_row object per netns. tw_count becomes a refcount and gains associated debugging infrastructure. BUG: KASAN: use-after-free in inet_twsk_kill+0x358/0x3c0 net/ipv4/inet_timewait_sock.c:46 Read of size 8 at addr ffff88807d5f9f40 by task kworker/1:7/3690 CPU: 1 PID: 3690 Comm: kworker/1:7 Not tainted 5.16.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Workqueue: events pwq_unbound_release_workfn Call Trace: <IRQ> __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106 print_address_description.constprop.0.cold+0x8d/0x336 mm/kasan/report.c:255 __kasan_report mm/kasan/report.c:442 [inline] kasan_report.cold+0x83/0xdf mm/kasan/report.c:459 inet_twsk_kill+0x358/0x3c0 net/ipv4/inet_timewait_sock.c:46 call_timer_fn+0x1a5/0x6b0 kernel/time/timer.c:1421 expire_timers kernel/time/timer.c:1466 [inline] __run_timers.part.0+0x67c/0xa30 kernel/time/timer.c:1734 __run_timers kernel/time/timer.c:1715 [inline] run_timer_softirq+0xb3/0x1d0 kernel/time/timer.c:1747 __do_softirq+0x29b/0x9c2 kernel/softirq.c:558 invoke_softirq kernel/softirq.c:432 [inline] __irq_exit_rcu+0x123/0x180 kernel/softirq.c:637 irq_exit_rcu+0x5/0x20 kernel/softirq.c:649 sysvec_apic_timer_interrupt+0x93/0xc0 arch/x86/kernel/apic/apic.c:1097 </IRQ> <TASK> asm_sysvec_apic_timer_interrupt+0x12/0x20 arch/x86/include/asm/idtentry.h:638 RIP: 0010:lockdep_unregister_key+0x1c9/0x250 kernel/locking/lockdep.c:6328 Code: 00 00 00 48 89 ee e8 46 fd ff ff 4c 89 f7 e8 5e c9 ff ff e8 09 cc ff ff 9c 58 f6 c4 02 75 26 41 f7 c4 00 02 00 00 74 01 fb 5b <5d> 41 5c 41 5d 41 5e 41 5f e9 19 4a 08 00 0f 0b 5b 5d 41 5c 41 5d RSP: 0018:ffffc90004077cb8 EFLAGS: 00000206 RAX: 0000000000000046 RBX: ffff88807b61b498 RCX: 0000000000000001 RDX: dffffc0000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: ffff888077027128 R08: 0000000000000001 R09: ffffffff8f1ea4fc R10: fffffbfff1ff93ee R11: 000000000000af1e R12: 0000000000000246 R13: 0000000000000000 R14: ffffffff8ffc89b8 R15: ffffffff90157fb0 wq_unregister_lockdep kernel/workqueue.c:3508 [inline] pwq_unbound_release_workfn+0x254/0x340 kernel/workqueue.c:3746 process_one_work+0x9ac/0x1650 kernel/workqueue.c:2307 worker_thread+0x657/0x1110 kernel/workqueue.c:2454 kthread+0x2e9/0x3a0 kernel/kthread.c:377 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295 </TASK> Allocated by task 3635: kasan_save_stack+0x1e/0x50 mm/kasan/common.c:38 kasan_set_track mm/kasan/common.c:46 [inline] set_alloc_info mm/kasan/common.c:437 [inline] __kasan_slab_alloc+0x90/0xc0 mm/kasan/common.c:470 kasan_slab_alloc include/linux/kasan.h:260 [inline] slab_post_alloc_hook mm/slab.h:732 [inline] slab_alloc_node mm/slub.c:3230 [inline] slab_alloc mm/slub.c:3238 [inline] kmem_cache_alloc+0x202/0x3a0 mm/slub.c:3243 kmem_cache_zalloc include/linux/slab.h:705 [inline] net_alloc net/core/net_namespace.c:407 [inline] copy_net_ns+0x125/0x760 net/core/net_namespace.c:462 create_new_namespaces+0x3f6/0xb20 kernel/nsproxy.c:110 unshare_nsproxy_namespaces+0xc1/0x1f0 kernel/nsproxy.c:226 ksys_unshare+0x445/0x920 kernel/fork.c:3048 __do_sys_unshare kernel/fork.c:3119 [inline] __se_sys_unshare kernel/fork.c:3117 [inline] __x64_sys_unshare+0x2d/0x40 kernel/fork.c:3117 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae The buggy address belongs to the object at ffff88807d5f9a80 which belongs to the cache net_namespace of size 6528 The buggy address is located 1216 bytes inside of 6528-byte region [ffff88807d5f9a80, ffff88807d5fb400) The buggy address belongs to the page: page:ffffea0001f57e00 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88807d5f9a80 pfn:0x7d5f8 head:ffffea0001f57e00 order:3 compound_mapcount:0 compound_pincount:0 memcg:ffff888070023001 flags: 0xfff00000010200(slab|head|node=0|zone=1|lastcpupid=0x7ff) raw: 00fff00000010200 ffff888010dd4f48 ffffea0001404e08 ffff8880118fd000 raw: ffff88807d5f9a80 0000000000040002 00000001ffffffff ffff888070023001 page dumped because: kasan: bad access detected page_owner tracks the page as allocated page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 3634, ts 119694798460, free_ts 119693556950 prep_new_page mm/page_alloc.c:2434 [inline] get_page_from_freelist+0xa72/0x2f50 mm/page_alloc.c:4165 __alloc_pages+0x1b2/0x500 mm/page_alloc.c:5389 alloc_pages+0x1aa/0x310 mm/mempolicy.c:2271 alloc_slab_page mm/slub.c:1799 [inline] allocate_slab mm/slub.c:1944 [inline] new_slab+0x28a/0x3b0 mm/slub.c:2004 ___slab_alloc+0x87c/0xe90 mm/slub.c:3018 __slab_alloc.constprop.0+0x4d/0xa0 mm/slub.c:3105 slab_alloc_node mm/slub.c:3196 [inline] slab_alloc mm/slub.c:3238 [inline] kmem_cache_alloc+0x35c/0x3a0 mm/slub.c:3243 kmem_cache_zalloc include/linux/slab.h:705 [inline] net_alloc net/core/net_namespace.c:407 [inline] copy_net_ns+0x125/0x760 net/core/net_namespace.c:462 create_new_namespaces+0x3f6/0xb20 kernel/nsproxy.c:110 unshare_nsproxy_namespaces+0xc1/0x1f0 kernel/nsproxy.c:226 ksys_unshare+0x445/0x920 kernel/fork.c:3048 __do_sys_unshare kernel/fork.c:3119 [inline] __se_sys_unshare kernel/fork.c:3117 [inline] __x64_sys_unshare+0x2d/0x40 kernel/fork.c:3117 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae page last free stack trace: reset_page_owner include/linux/page_owner.h:24 [inline] free_pages_prepare mm/page_alloc.c:1352 [inline] free_pcp_prepare+0x374/0x870 mm/page_alloc.c:1404 free_unref_page_prepare mm/page_alloc.c:3325 [inline] free_unref_page+0x19/0x690 mm/page_alloc.c:3404 skb_free_head net/core/skbuff.c:655 [inline] skb_release_data+0x65d/0x790 net/core/skbuff.c:677 skb_release_all net/core/skbuff.c:742 [inline] __kfree_skb net/core/skbuff.c:756 [inline] consume_skb net/core/skbuff.c:914 [inline] consume_skb+0xc2/0x160 net/core/skbuff.c:908 skb_free_datagram+0x1b/0x1f0 net/core/datagram.c:325 netlink_recvmsg+0x636/0xea0 net/netlink/af_netlink.c:1998 sock_recvmsg_nosec net/socket.c:948 [inline] sock_recvmsg net/socket.c:966 [inline] sock_recvmsg net/socket.c:962 [inline] ____sys_recvmsg+0x2c4/0x600 net/socket.c:2632 ___sys_recvmsg+0x127/0x200 net/socket.c:2674 __sys_recvmsg+0xe2/0x1a0 net/socket.c:2704 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae Memory state around the buggy address: ffff88807d5f9e00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88807d5f9e80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb >ffff88807d5f9f00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ^ ffff88807d5f9f80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88807d5fa000: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb Fixes: 0dad4087a86a ("tcp/dccp: get rid of inet_twsk_purge()") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot <syzkaller@googlegroups.com> Reported-by: Paolo Abeni <pabeni@redhat.com> Tested-by: Paolo Abeni <pabeni@redhat.com> Link: https://lore.kernel.org/r/20220126180714.845362-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-01-26 21:07:14 +03:00
if (refcount_dec_and_test(&tw->tw_dr->tw_refcount))
kfree(tw->tw_dr);
tcp/dccp: get rid of central timewait timer Using a timer wheel for timewait sockets was nice ~15 years ago when memory was expensive and machines had a single processor. This does not scale, code is ugly and source of huge latencies (Typically 30 ms have been seen, cpus spinning on death_lock spinlock.) We can afford to use an extra 64 bytes per timewait sock and spread timewait load to all cpus to have better behavior. Tested: On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1 on the target (lpaa24) Before patch : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 419594 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 437171 While test is running, we can observe 25 or even 33 ms latencies. lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2 lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2 After patch : About 90% increase of throughput : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 810442 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 800992 And latencies are kept to minimal values during this load, even if network utilization is 90% higher : lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-13 04:51:09 +03:00
inet_twsk_put(tw);
}
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 11:22:02 +04:00
void inet_twsk_free(struct inet_timewait_sock *tw)
{
struct module *owner = tw->tw_prot->owner;
twsk_destructor((struct sock *)tw);
#ifdef SOCK_REFCNT_DEBUG
pr_debug("%s timewait_sock %p released\n", tw->tw_prot->name, tw);
#endif
kmem_cache_free(tw->tw_prot->twsk_prot->twsk_slab, tw);
module_put(owner);
}
void inet_twsk_put(struct inet_timewait_sock *tw)
{
if (refcount_dec_and_test(&tw->tw_refcnt))
inet_twsk_free(tw);
}
EXPORT_SYMBOL_GPL(inet_twsk_put);
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 11:22:02 +04:00
static void inet_twsk_add_node_rcu(struct inet_timewait_sock *tw,
struct hlist_nulls_head *list)
{
hlist_nulls_add_head_rcu(&tw->tw_node, list);
}
static void inet_twsk_add_bind_node(struct inet_timewait_sock *tw,
struct hlist_head *list)
{
hlist_add_head(&tw->tw_bind_node, list);
}
/*
* Enter the time wait state. This is called with locally disabled BH.
* Essentially we whip up a timewait bucket, copy the relevant info into it
* from the SK, and mess with hash chains and list linkage.
*/
void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
struct inet_hashinfo *hashinfo)
{
const struct inet_sock *inet = inet_sk(sk);
const struct inet_connection_sock *icsk = inet_csk(sk);
[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>
2005-10-04 01:13:38 +04:00
struct inet_ehash_bucket *ehead = inet_ehash_bucket(hashinfo, sk->sk_hash);
spinlock_t *lock = inet_ehash_lockp(hashinfo, sk->sk_hash);
struct inet_bind_hashbucket *bhead;
/* Step 1: Put TW into bind hash. Original socket stays there too.
Note, that any socket with inet->num != 0 MUST be bound in
binding cache, even if it is closed.
*/
bhead = &hashinfo->bhash[inet_bhashfn(twsk_net(tw), inet->inet_num,
hashinfo->bhash_size)];
spin_lock(&bhead->lock);
tw->tw_tb = icsk->icsk_bind_hash;
WARN_ON(!icsk->icsk_bind_hash);
inet_twsk_add_bind_node(tw, &tw->tw_tb->owners);
spin_unlock(&bhead->lock);
spin_lock(lock);
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 11:22:02 +04:00
inet_twsk_add_node_rcu(tw, &ehead->chain);
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 11:22:02 +04:00
/* Step 3: Remove SK from hash chain */
if (__sk_nulls_del_node_init_rcu(sk))
sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
spin_unlock(lock);
/* tw_refcnt is set to 3 because we have :
* - one reference for bhash chain.
* - one reference for ehash chain.
* - one reference for timer.
* We can use atomic_set() because prior spin_lock()/spin_unlock()
* committed into memory all tw fields.
* Also note that after this point, we lost our implicit reference
* so we are not allowed to use tw anymore.
*/
refcount_set(&tw->tw_refcnt, 3);
}
EXPORT_SYMBOL_GPL(inet_twsk_hashdance);
static void tw_timer_handler(struct timer_list *t)
{
struct inet_timewait_sock *tw = from_timer(tw, t, tw_timer);
tcp/dccp: get rid of central timewait timer Using a timer wheel for timewait sockets was nice ~15 years ago when memory was expensive and machines had a single processor. This does not scale, code is ugly and source of huge latencies (Typically 30 ms have been seen, cpus spinning on death_lock spinlock.) We can afford to use an extra 64 bytes per timewait sock and spread timewait load to all cpus to have better behavior. Tested: On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1 on the target (lpaa24) Before patch : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 419594 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 437171 While test is running, we can observe 25 or even 33 ms latencies. lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2 lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2 After patch : About 90% increase of throughput : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 810442 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 800992 And latencies are kept to minimal values during this load, even if network utilization is 90% higher : lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-13 04:51:09 +03:00
inet_twsk_kill(tw);
}
struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk,
struct inet_timewait_death_row *dr,
const int state)
{
struct inet_timewait_sock *tw;
tcp: allocate tcp_death_row outside of struct netns_ipv4 I forgot tcp had per netns tracking of timewait sockets, and their sysctl to change the limit. After 0dad4087a86a ("tcp/dccp: get rid of inet_twsk_purge()"), whole struct net can be freed before last tw socket is freed. We need to allocate a separate struct inet_timewait_death_row object per netns. tw_count becomes a refcount and gains associated debugging infrastructure. BUG: KASAN: use-after-free in inet_twsk_kill+0x358/0x3c0 net/ipv4/inet_timewait_sock.c:46 Read of size 8 at addr ffff88807d5f9f40 by task kworker/1:7/3690 CPU: 1 PID: 3690 Comm: kworker/1:7 Not tainted 5.16.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Workqueue: events pwq_unbound_release_workfn Call Trace: <IRQ> __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106 print_address_description.constprop.0.cold+0x8d/0x336 mm/kasan/report.c:255 __kasan_report mm/kasan/report.c:442 [inline] kasan_report.cold+0x83/0xdf mm/kasan/report.c:459 inet_twsk_kill+0x358/0x3c0 net/ipv4/inet_timewait_sock.c:46 call_timer_fn+0x1a5/0x6b0 kernel/time/timer.c:1421 expire_timers kernel/time/timer.c:1466 [inline] __run_timers.part.0+0x67c/0xa30 kernel/time/timer.c:1734 __run_timers kernel/time/timer.c:1715 [inline] run_timer_softirq+0xb3/0x1d0 kernel/time/timer.c:1747 __do_softirq+0x29b/0x9c2 kernel/softirq.c:558 invoke_softirq kernel/softirq.c:432 [inline] __irq_exit_rcu+0x123/0x180 kernel/softirq.c:637 irq_exit_rcu+0x5/0x20 kernel/softirq.c:649 sysvec_apic_timer_interrupt+0x93/0xc0 arch/x86/kernel/apic/apic.c:1097 </IRQ> <TASK> asm_sysvec_apic_timer_interrupt+0x12/0x20 arch/x86/include/asm/idtentry.h:638 RIP: 0010:lockdep_unregister_key+0x1c9/0x250 kernel/locking/lockdep.c:6328 Code: 00 00 00 48 89 ee e8 46 fd ff ff 4c 89 f7 e8 5e c9 ff ff e8 09 cc ff ff 9c 58 f6 c4 02 75 26 41 f7 c4 00 02 00 00 74 01 fb 5b <5d> 41 5c 41 5d 41 5e 41 5f e9 19 4a 08 00 0f 0b 5b 5d 41 5c 41 5d RSP: 0018:ffffc90004077cb8 EFLAGS: 00000206 RAX: 0000000000000046 RBX: ffff88807b61b498 RCX: 0000000000000001 RDX: dffffc0000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: ffff888077027128 R08: 0000000000000001 R09: ffffffff8f1ea4fc R10: fffffbfff1ff93ee R11: 000000000000af1e R12: 0000000000000246 R13: 0000000000000000 R14: ffffffff8ffc89b8 R15: ffffffff90157fb0 wq_unregister_lockdep kernel/workqueue.c:3508 [inline] pwq_unbound_release_workfn+0x254/0x340 kernel/workqueue.c:3746 process_one_work+0x9ac/0x1650 kernel/workqueue.c:2307 worker_thread+0x657/0x1110 kernel/workqueue.c:2454 kthread+0x2e9/0x3a0 kernel/kthread.c:377 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295 </TASK> Allocated by task 3635: kasan_save_stack+0x1e/0x50 mm/kasan/common.c:38 kasan_set_track mm/kasan/common.c:46 [inline] set_alloc_info mm/kasan/common.c:437 [inline] __kasan_slab_alloc+0x90/0xc0 mm/kasan/common.c:470 kasan_slab_alloc include/linux/kasan.h:260 [inline] slab_post_alloc_hook mm/slab.h:732 [inline] slab_alloc_node mm/slub.c:3230 [inline] slab_alloc mm/slub.c:3238 [inline] kmem_cache_alloc+0x202/0x3a0 mm/slub.c:3243 kmem_cache_zalloc include/linux/slab.h:705 [inline] net_alloc net/core/net_namespace.c:407 [inline] copy_net_ns+0x125/0x760 net/core/net_namespace.c:462 create_new_namespaces+0x3f6/0xb20 kernel/nsproxy.c:110 unshare_nsproxy_namespaces+0xc1/0x1f0 kernel/nsproxy.c:226 ksys_unshare+0x445/0x920 kernel/fork.c:3048 __do_sys_unshare kernel/fork.c:3119 [inline] __se_sys_unshare kernel/fork.c:3117 [inline] __x64_sys_unshare+0x2d/0x40 kernel/fork.c:3117 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae The buggy address belongs to the object at ffff88807d5f9a80 which belongs to the cache net_namespace of size 6528 The buggy address is located 1216 bytes inside of 6528-byte region [ffff88807d5f9a80, ffff88807d5fb400) The buggy address belongs to the page: page:ffffea0001f57e00 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88807d5f9a80 pfn:0x7d5f8 head:ffffea0001f57e00 order:3 compound_mapcount:0 compound_pincount:0 memcg:ffff888070023001 flags: 0xfff00000010200(slab|head|node=0|zone=1|lastcpupid=0x7ff) raw: 00fff00000010200 ffff888010dd4f48 ffffea0001404e08 ffff8880118fd000 raw: ffff88807d5f9a80 0000000000040002 00000001ffffffff ffff888070023001 page dumped because: kasan: bad access detected page_owner tracks the page as allocated page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 3634, ts 119694798460, free_ts 119693556950 prep_new_page mm/page_alloc.c:2434 [inline] get_page_from_freelist+0xa72/0x2f50 mm/page_alloc.c:4165 __alloc_pages+0x1b2/0x500 mm/page_alloc.c:5389 alloc_pages+0x1aa/0x310 mm/mempolicy.c:2271 alloc_slab_page mm/slub.c:1799 [inline] allocate_slab mm/slub.c:1944 [inline] new_slab+0x28a/0x3b0 mm/slub.c:2004 ___slab_alloc+0x87c/0xe90 mm/slub.c:3018 __slab_alloc.constprop.0+0x4d/0xa0 mm/slub.c:3105 slab_alloc_node mm/slub.c:3196 [inline] slab_alloc mm/slub.c:3238 [inline] kmem_cache_alloc+0x35c/0x3a0 mm/slub.c:3243 kmem_cache_zalloc include/linux/slab.h:705 [inline] net_alloc net/core/net_namespace.c:407 [inline] copy_net_ns+0x125/0x760 net/core/net_namespace.c:462 create_new_namespaces+0x3f6/0xb20 kernel/nsproxy.c:110 unshare_nsproxy_namespaces+0xc1/0x1f0 kernel/nsproxy.c:226 ksys_unshare+0x445/0x920 kernel/fork.c:3048 __do_sys_unshare kernel/fork.c:3119 [inline] __se_sys_unshare kernel/fork.c:3117 [inline] __x64_sys_unshare+0x2d/0x40 kernel/fork.c:3117 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae page last free stack trace: reset_page_owner include/linux/page_owner.h:24 [inline] free_pages_prepare mm/page_alloc.c:1352 [inline] free_pcp_prepare+0x374/0x870 mm/page_alloc.c:1404 free_unref_page_prepare mm/page_alloc.c:3325 [inline] free_unref_page+0x19/0x690 mm/page_alloc.c:3404 skb_free_head net/core/skbuff.c:655 [inline] skb_release_data+0x65d/0x790 net/core/skbuff.c:677 skb_release_all net/core/skbuff.c:742 [inline] __kfree_skb net/core/skbuff.c:756 [inline] consume_skb net/core/skbuff.c:914 [inline] consume_skb+0xc2/0x160 net/core/skbuff.c:908 skb_free_datagram+0x1b/0x1f0 net/core/datagram.c:325 netlink_recvmsg+0x636/0xea0 net/netlink/af_netlink.c:1998 sock_recvmsg_nosec net/socket.c:948 [inline] sock_recvmsg net/socket.c:966 [inline] sock_recvmsg net/socket.c:962 [inline] ____sys_recvmsg+0x2c4/0x600 net/socket.c:2632 ___sys_recvmsg+0x127/0x200 net/socket.c:2674 __sys_recvmsg+0xe2/0x1a0 net/socket.c:2704 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae Memory state around the buggy address: ffff88807d5f9e00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88807d5f9e80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb >ffff88807d5f9f00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ^ ffff88807d5f9f80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88807d5fa000: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb Fixes: 0dad4087a86a ("tcp/dccp: get rid of inet_twsk_purge()") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot <syzkaller@googlegroups.com> Reported-by: Paolo Abeni <pabeni@redhat.com> Tested-by: Paolo Abeni <pabeni@redhat.com> Link: https://lore.kernel.org/r/20220126180714.845362-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-01-26 21:07:14 +03:00
if (refcount_read(&dr->tw_refcount) - 1 >= dr->sysctl_max_tw_buckets)
tcp/dccp: get rid of central timewait timer Using a timer wheel for timewait sockets was nice ~15 years ago when memory was expensive and machines had a single processor. This does not scale, code is ugly and source of huge latencies (Typically 30 ms have been seen, cpus spinning on death_lock spinlock.) We can afford to use an extra 64 bytes per timewait sock and spread timewait load to all cpus to have better behavior. Tested: On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1 on the target (lpaa24) Before patch : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 419594 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 437171 While test is running, we can observe 25 or even 33 ms latencies. lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2 lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2 After patch : About 90% increase of throughput : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 810442 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 800992 And latencies are kept to minimal values during this load, even if network utilization is 90% higher : lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-13 04:51:09 +03:00
return NULL;
tw = kmem_cache_alloc(sk->sk_prot_creator->twsk_prot->twsk_slab,
GFP_ATOMIC);
if (tw) {
const struct inet_sock *inet = inet_sk(sk);
tcp/dccp: get rid of central timewait timer Using a timer wheel for timewait sockets was nice ~15 years ago when memory was expensive and machines had a single processor. This does not scale, code is ugly and source of huge latencies (Typically 30 ms have been seen, cpus spinning on death_lock spinlock.) We can afford to use an extra 64 bytes per timewait sock and spread timewait load to all cpus to have better behavior. Tested: On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1 on the target (lpaa24) Before patch : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 419594 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 437171 While test is running, we can observe 25 or even 33 ms latencies. lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2 lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2 After patch : About 90% increase of throughput : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 810442 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 800992 And latencies are kept to minimal values during this load, even if network utilization is 90% higher : lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-13 04:51:09 +03:00
tw->tw_dr = dr;
/* Give us an identity. */
tw->tw_daddr = inet->inet_daddr;
tw->tw_rcv_saddr = inet->inet_rcv_saddr;
tw->tw_bound_dev_if = sk->sk_bound_dev_if;
tw->tw_tos = inet->tos;
tw->tw_num = inet->inet_num;
tw->tw_state = TCP_TIME_WAIT;
tw->tw_substate = state;
tw->tw_sport = inet->inet_sport;
tw->tw_dport = inet->inet_dport;
tw->tw_family = sk->sk_family;
tw->tw_reuse = sk->sk_reuse;
soreuseport: initialise timewait reuseport field syzbot reported an uninit-value in inet_csk_bind_conflict() [1] It turns out we never propagated sk->sk_reuseport into timewait socket. [1] BUG: KMSAN: uninit-value in inet_csk_bind_conflict+0x5f9/0x990 net/ipv4/inet_connection_sock.c:151 CPU: 1 PID: 3589 Comm: syzkaller008242 Not tainted 4.16.0+ #82 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:17 [inline] dump_stack+0x185/0x1d0 lib/dump_stack.c:53 kmsan_report+0x142/0x240 mm/kmsan/kmsan.c:1067 __msan_warning_32+0x6c/0xb0 mm/kmsan/kmsan_instr.c:676 inet_csk_bind_conflict+0x5f9/0x990 net/ipv4/inet_connection_sock.c:151 inet_csk_get_port+0x1d28/0x1e40 net/ipv4/inet_connection_sock.c:320 inet6_bind+0x121c/0x1820 net/ipv6/af_inet6.c:399 SYSC_bind+0x3f2/0x4b0 net/socket.c:1474 SyS_bind+0x54/0x80 net/socket.c:1460 do_syscall_64+0x309/0x430 arch/x86/entry/common.c:287 entry_SYSCALL_64_after_hwframe+0x3d/0xa2 RIP: 0033:0x4416e9 RSP: 002b:00007ffce6d15c88 EFLAGS: 00000217 ORIG_RAX: 0000000000000031 RAX: ffffffffffffffda RBX: 0100000000000000 RCX: 00000000004416e9 RDX: 000000000000001c RSI: 0000000020402000 RDI: 0000000000000004 RBP: 0000000000000000 R08: 00000000e6d15e08 R09: 00000000e6d15e08 R10: 0000000000000004 R11: 0000000000000217 R12: 0000000000009478 R13: 00000000006cd448 R14: 0000000000000000 R15: 0000000000000000 Uninit was stored to memory at: kmsan_save_stack_with_flags mm/kmsan/kmsan.c:278 [inline] kmsan_save_stack mm/kmsan/kmsan.c:293 [inline] kmsan_internal_chain_origin+0x12b/0x210 mm/kmsan/kmsan.c:684 __msan_chain_origin+0x69/0xc0 mm/kmsan/kmsan_instr.c:521 tcp_time_wait+0xf17/0xf50 net/ipv4/tcp_minisocks.c:283 tcp_rcv_state_process+0xebe/0x6490 net/ipv4/tcp_input.c:6003 tcp_v6_do_rcv+0x11dd/0x1d90 net/ipv6/tcp_ipv6.c:1331 sk_backlog_rcv include/net/sock.h:908 [inline] __release_sock+0x2d6/0x680 net/core/sock.c:2271 release_sock+0x97/0x2a0 net/core/sock.c:2786 tcp_close+0x277/0x18f0 net/ipv4/tcp.c:2269 inet_release+0x240/0x2a0 net/ipv4/af_inet.c:427 inet6_release+0xaf/0x100 net/ipv6/af_inet6.c:435 sock_release net/socket.c:595 [inline] sock_close+0xe0/0x300 net/socket.c:1149 __fput+0x49e/0xa10 fs/file_table.c:209 ____fput+0x37/0x40 fs/file_table.c:243 task_work_run+0x243/0x2c0 kernel/task_work.c:113 exit_task_work include/linux/task_work.h:22 [inline] do_exit+0x10e1/0x38d0 kernel/exit.c:867 do_group_exit+0x1a0/0x360 kernel/exit.c:970 SYSC_exit_group+0x21/0x30 kernel/exit.c:981 SyS_exit_group+0x25/0x30 kernel/exit.c:979 do_syscall_64+0x309/0x430 arch/x86/entry/common.c:287 entry_SYSCALL_64_after_hwframe+0x3d/0xa2 Uninit was stored to memory at: kmsan_save_stack_with_flags mm/kmsan/kmsan.c:278 [inline] kmsan_save_stack mm/kmsan/kmsan.c:293 [inline] kmsan_internal_chain_origin+0x12b/0x210 mm/kmsan/kmsan.c:684 __msan_chain_origin+0x69/0xc0 mm/kmsan/kmsan_instr.c:521 inet_twsk_alloc+0xaef/0xc00 net/ipv4/inet_timewait_sock.c:182 tcp_time_wait+0xd9/0xf50 net/ipv4/tcp_minisocks.c:258 tcp_rcv_state_process+0xebe/0x6490 net/ipv4/tcp_input.c:6003 tcp_v6_do_rcv+0x11dd/0x1d90 net/ipv6/tcp_ipv6.c:1331 sk_backlog_rcv include/net/sock.h:908 [inline] __release_sock+0x2d6/0x680 net/core/sock.c:2271 release_sock+0x97/0x2a0 net/core/sock.c:2786 tcp_close+0x277/0x18f0 net/ipv4/tcp.c:2269 inet_release+0x240/0x2a0 net/ipv4/af_inet.c:427 inet6_release+0xaf/0x100 net/ipv6/af_inet6.c:435 sock_release net/socket.c:595 [inline] sock_close+0xe0/0x300 net/socket.c:1149 __fput+0x49e/0xa10 fs/file_table.c:209 ____fput+0x37/0x40 fs/file_table.c:243 task_work_run+0x243/0x2c0 kernel/task_work.c:113 exit_task_work include/linux/task_work.h:22 [inline] do_exit+0x10e1/0x38d0 kernel/exit.c:867 do_group_exit+0x1a0/0x360 kernel/exit.c:970 SYSC_exit_group+0x21/0x30 kernel/exit.c:981 SyS_exit_group+0x25/0x30 kernel/exit.c:979 do_syscall_64+0x309/0x430 arch/x86/entry/common.c:287 entry_SYSCALL_64_after_hwframe+0x3d/0xa2 Uninit was created at: kmsan_save_stack_with_flags mm/kmsan/kmsan.c:278 [inline] kmsan_internal_poison_shadow+0xb8/0x1b0 mm/kmsan/kmsan.c:188 kmsan_kmalloc+0x94/0x100 mm/kmsan/kmsan.c:314 kmem_cache_alloc+0xaab/0xb90 mm/slub.c:2756 inet_twsk_alloc+0x13b/0xc00 net/ipv4/inet_timewait_sock.c:163 tcp_time_wait+0xd9/0xf50 net/ipv4/tcp_minisocks.c:258 tcp_rcv_state_process+0xebe/0x6490 net/ipv4/tcp_input.c:6003 tcp_v6_do_rcv+0x11dd/0x1d90 net/ipv6/tcp_ipv6.c:1331 sk_backlog_rcv include/net/sock.h:908 [inline] __release_sock+0x2d6/0x680 net/core/sock.c:2271 release_sock+0x97/0x2a0 net/core/sock.c:2786 tcp_close+0x277/0x18f0 net/ipv4/tcp.c:2269 inet_release+0x240/0x2a0 net/ipv4/af_inet.c:427 inet6_release+0xaf/0x100 net/ipv6/af_inet6.c:435 sock_release net/socket.c:595 [inline] sock_close+0xe0/0x300 net/socket.c:1149 __fput+0x49e/0xa10 fs/file_table.c:209 ____fput+0x37/0x40 fs/file_table.c:243 task_work_run+0x243/0x2c0 kernel/task_work.c:113 exit_task_work include/linux/task_work.h:22 [inline] do_exit+0x10e1/0x38d0 kernel/exit.c:867 do_group_exit+0x1a0/0x360 kernel/exit.c:970 SYSC_exit_group+0x21/0x30 kernel/exit.c:981 SyS_exit_group+0x25/0x30 kernel/exit.c:979 do_syscall_64+0x309/0x430 arch/x86/entry/common.c:287 entry_SYSCALL_64_after_hwframe+0x3d/0xa2 Fixes: da5e36308d9f ("soreuseport: TCP/IPv4 implementation") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot <syzkaller@googlegroups.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-07 23:42:43 +03:00
tw->tw_reuseport = sk->sk_reuseport;
[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>
2005-10-04 01:13:38 +04:00
tw->tw_hash = sk->sk_hash;
tw->tw_ipv6only = 0;
tw->tw_transparent = inet->transparent;
tw->tw_prot = sk->sk_prot_creator;
atomic64_set(&tw->tw_cookie, atomic64_read(&sk->sk_cookie));
twsk_net_set(tw, sock_net(sk));
timer_setup(&tw->tw_timer, tw_timer_handler, TIMER_PINNED);
/*
* Because we use RCU lookups, we should not set tw_refcnt
* to a non null value before everything is setup for this
* timewait socket.
*/
refcount_set(&tw->tw_refcnt, 0);
tcp/dccp: get rid of central timewait timer Using a timer wheel for timewait sockets was nice ~15 years ago when memory was expensive and machines had a single processor. This does not scale, code is ugly and source of huge latencies (Typically 30 ms have been seen, cpus spinning on death_lock spinlock.) We can afford to use an extra 64 bytes per timewait sock and spread timewait load to all cpus to have better behavior. Tested: On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1 on the target (lpaa24) Before patch : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 419594 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 437171 While test is running, we can observe 25 or even 33 ms latencies. lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2 lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2 After patch : About 90% increase of throughput : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 810442 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 800992 And latencies are kept to minimal values during this load, even if network utilization is 90% higher : lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-13 04:51:09 +03:00
__module_get(tw->tw_prot->owner);
}
return tw;
}
EXPORT_SYMBOL_GPL(inet_twsk_alloc);
/* These are always called from BH context. See callers in
* tcp_input.c to verify this.
*/
/* This is for handling early-kills of TIME_WAIT sockets.
* Warning : consume reference.
* Caller should not access tw anymore.
*/
void inet_twsk_deschedule_put(struct inet_timewait_sock *tw)
{
tcp/dccp: get rid of central timewait timer Using a timer wheel for timewait sockets was nice ~15 years ago when memory was expensive and machines had a single processor. This does not scale, code is ugly and source of huge latencies (Typically 30 ms have been seen, cpus spinning on death_lock spinlock.) We can afford to use an extra 64 bytes per timewait sock and spread timewait load to all cpus to have better behavior. Tested: On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1 on the target (lpaa24) Before patch : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 419594 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 437171 While test is running, we can observe 25 or even 33 ms latencies. lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2 lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2 After patch : About 90% increase of throughput : lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 810442 lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0 800992 And latencies are kept to minimal values during this load, even if network utilization is 90% higher : lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23 ... 1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-13 04:51:09 +03:00
if (del_timer_sync(&tw->tw_timer))
inet_twsk_kill(tw);
inet_twsk_put(tw);
}
EXPORT_SYMBOL(inet_twsk_deschedule_put);
void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo, bool rearm)
{
/* timeout := RTO * 3.5
*
* 3.5 = 1+2+0.5 to wait for two retransmits.
*
* RATIONALE: if FIN arrived and we entered TIME-WAIT state,
* our ACK acking that FIN can be lost. If N subsequent retransmitted
* FINs (or previous seqments) are lost (probability of such event
* is p^(N+1), where p is probability to lose single packet and
* time to detect the loss is about RTO*(2^N - 1) with exponential
* backoff). Normal timewait length is calculated so, that we
* waited at least for one retransmitted FIN (maximal RTO is 120sec).
* [ BTW Linux. following BSD, violates this requirement waiting
* only for 60sec, we should wait at least for 240 secs.
* Well, 240 consumes too much of resources 8)
* ]
* This interval is not reduced to catch old duplicate and
* responces to our wandering segments living for two MSLs.
* However, if we use PAWS to detect
* old duplicates, we can reduce the interval to bounds required
* by RTO, rather than MSL. So, if peer understands PAWS, we
* kill tw bucket after 3.5*RTO (it is important that this number
* is greater than TS tick!) and detect old duplicates with help
* of PAWS.
*/
if (!rearm) {
bool kill = timeo <= 4*HZ;
__NET_INC_STATS(twsk_net(tw), kill ? LINUX_MIB_TIMEWAITKILLED :
LINUX_MIB_TIMEWAITED);
BUG_ON(mod_timer(&tw->tw_timer, jiffies + timeo));
tcp: allocate tcp_death_row outside of struct netns_ipv4 I forgot tcp had per netns tracking of timewait sockets, and their sysctl to change the limit. After 0dad4087a86a ("tcp/dccp: get rid of inet_twsk_purge()"), whole struct net can be freed before last tw socket is freed. We need to allocate a separate struct inet_timewait_death_row object per netns. tw_count becomes a refcount and gains associated debugging infrastructure. BUG: KASAN: use-after-free in inet_twsk_kill+0x358/0x3c0 net/ipv4/inet_timewait_sock.c:46 Read of size 8 at addr ffff88807d5f9f40 by task kworker/1:7/3690 CPU: 1 PID: 3690 Comm: kworker/1:7 Not tainted 5.16.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Workqueue: events pwq_unbound_release_workfn Call Trace: <IRQ> __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106 print_address_description.constprop.0.cold+0x8d/0x336 mm/kasan/report.c:255 __kasan_report mm/kasan/report.c:442 [inline] kasan_report.cold+0x83/0xdf mm/kasan/report.c:459 inet_twsk_kill+0x358/0x3c0 net/ipv4/inet_timewait_sock.c:46 call_timer_fn+0x1a5/0x6b0 kernel/time/timer.c:1421 expire_timers kernel/time/timer.c:1466 [inline] __run_timers.part.0+0x67c/0xa30 kernel/time/timer.c:1734 __run_timers kernel/time/timer.c:1715 [inline] run_timer_softirq+0xb3/0x1d0 kernel/time/timer.c:1747 __do_softirq+0x29b/0x9c2 kernel/softirq.c:558 invoke_softirq kernel/softirq.c:432 [inline] __irq_exit_rcu+0x123/0x180 kernel/softirq.c:637 irq_exit_rcu+0x5/0x20 kernel/softirq.c:649 sysvec_apic_timer_interrupt+0x93/0xc0 arch/x86/kernel/apic/apic.c:1097 </IRQ> <TASK> asm_sysvec_apic_timer_interrupt+0x12/0x20 arch/x86/include/asm/idtentry.h:638 RIP: 0010:lockdep_unregister_key+0x1c9/0x250 kernel/locking/lockdep.c:6328 Code: 00 00 00 48 89 ee e8 46 fd ff ff 4c 89 f7 e8 5e c9 ff ff e8 09 cc ff ff 9c 58 f6 c4 02 75 26 41 f7 c4 00 02 00 00 74 01 fb 5b <5d> 41 5c 41 5d 41 5e 41 5f e9 19 4a 08 00 0f 0b 5b 5d 41 5c 41 5d RSP: 0018:ffffc90004077cb8 EFLAGS: 00000206 RAX: 0000000000000046 RBX: ffff88807b61b498 RCX: 0000000000000001 RDX: dffffc0000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: ffff888077027128 R08: 0000000000000001 R09: ffffffff8f1ea4fc R10: fffffbfff1ff93ee R11: 000000000000af1e R12: 0000000000000246 R13: 0000000000000000 R14: ffffffff8ffc89b8 R15: ffffffff90157fb0 wq_unregister_lockdep kernel/workqueue.c:3508 [inline] pwq_unbound_release_workfn+0x254/0x340 kernel/workqueue.c:3746 process_one_work+0x9ac/0x1650 kernel/workqueue.c:2307 worker_thread+0x657/0x1110 kernel/workqueue.c:2454 kthread+0x2e9/0x3a0 kernel/kthread.c:377 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295 </TASK> Allocated by task 3635: kasan_save_stack+0x1e/0x50 mm/kasan/common.c:38 kasan_set_track mm/kasan/common.c:46 [inline] set_alloc_info mm/kasan/common.c:437 [inline] __kasan_slab_alloc+0x90/0xc0 mm/kasan/common.c:470 kasan_slab_alloc include/linux/kasan.h:260 [inline] slab_post_alloc_hook mm/slab.h:732 [inline] slab_alloc_node mm/slub.c:3230 [inline] slab_alloc mm/slub.c:3238 [inline] kmem_cache_alloc+0x202/0x3a0 mm/slub.c:3243 kmem_cache_zalloc include/linux/slab.h:705 [inline] net_alloc net/core/net_namespace.c:407 [inline] copy_net_ns+0x125/0x760 net/core/net_namespace.c:462 create_new_namespaces+0x3f6/0xb20 kernel/nsproxy.c:110 unshare_nsproxy_namespaces+0xc1/0x1f0 kernel/nsproxy.c:226 ksys_unshare+0x445/0x920 kernel/fork.c:3048 __do_sys_unshare kernel/fork.c:3119 [inline] __se_sys_unshare kernel/fork.c:3117 [inline] __x64_sys_unshare+0x2d/0x40 kernel/fork.c:3117 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae The buggy address belongs to the object at ffff88807d5f9a80 which belongs to the cache net_namespace of size 6528 The buggy address is located 1216 bytes inside of 6528-byte region [ffff88807d5f9a80, ffff88807d5fb400) The buggy address belongs to the page: page:ffffea0001f57e00 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88807d5f9a80 pfn:0x7d5f8 head:ffffea0001f57e00 order:3 compound_mapcount:0 compound_pincount:0 memcg:ffff888070023001 flags: 0xfff00000010200(slab|head|node=0|zone=1|lastcpupid=0x7ff) raw: 00fff00000010200 ffff888010dd4f48 ffffea0001404e08 ffff8880118fd000 raw: ffff88807d5f9a80 0000000000040002 00000001ffffffff ffff888070023001 page dumped because: kasan: bad access detected page_owner tracks the page as allocated page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 3634, ts 119694798460, free_ts 119693556950 prep_new_page mm/page_alloc.c:2434 [inline] get_page_from_freelist+0xa72/0x2f50 mm/page_alloc.c:4165 __alloc_pages+0x1b2/0x500 mm/page_alloc.c:5389 alloc_pages+0x1aa/0x310 mm/mempolicy.c:2271 alloc_slab_page mm/slub.c:1799 [inline] allocate_slab mm/slub.c:1944 [inline] new_slab+0x28a/0x3b0 mm/slub.c:2004 ___slab_alloc+0x87c/0xe90 mm/slub.c:3018 __slab_alloc.constprop.0+0x4d/0xa0 mm/slub.c:3105 slab_alloc_node mm/slub.c:3196 [inline] slab_alloc mm/slub.c:3238 [inline] kmem_cache_alloc+0x35c/0x3a0 mm/slub.c:3243 kmem_cache_zalloc include/linux/slab.h:705 [inline] net_alloc net/core/net_namespace.c:407 [inline] copy_net_ns+0x125/0x760 net/core/net_namespace.c:462 create_new_namespaces+0x3f6/0xb20 kernel/nsproxy.c:110 unshare_nsproxy_namespaces+0xc1/0x1f0 kernel/nsproxy.c:226 ksys_unshare+0x445/0x920 kernel/fork.c:3048 __do_sys_unshare kernel/fork.c:3119 [inline] __se_sys_unshare kernel/fork.c:3117 [inline] __x64_sys_unshare+0x2d/0x40 kernel/fork.c:3117 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae page last free stack trace: reset_page_owner include/linux/page_owner.h:24 [inline] free_pages_prepare mm/page_alloc.c:1352 [inline] free_pcp_prepare+0x374/0x870 mm/page_alloc.c:1404 free_unref_page_prepare mm/page_alloc.c:3325 [inline] free_unref_page+0x19/0x690 mm/page_alloc.c:3404 skb_free_head net/core/skbuff.c:655 [inline] skb_release_data+0x65d/0x790 net/core/skbuff.c:677 skb_release_all net/core/skbuff.c:742 [inline] __kfree_skb net/core/skbuff.c:756 [inline] consume_skb net/core/skbuff.c:914 [inline] consume_skb+0xc2/0x160 net/core/skbuff.c:908 skb_free_datagram+0x1b/0x1f0 net/core/datagram.c:325 netlink_recvmsg+0x636/0xea0 net/netlink/af_netlink.c:1998 sock_recvmsg_nosec net/socket.c:948 [inline] sock_recvmsg net/socket.c:966 [inline] sock_recvmsg net/socket.c:962 [inline] ____sys_recvmsg+0x2c4/0x600 net/socket.c:2632 ___sys_recvmsg+0x127/0x200 net/socket.c:2674 __sys_recvmsg+0xe2/0x1a0 net/socket.c:2704 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae Memory state around the buggy address: ffff88807d5f9e00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88807d5f9e80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb >ffff88807d5f9f00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ^ ffff88807d5f9f80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88807d5fa000: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb Fixes: 0dad4087a86a ("tcp/dccp: get rid of inet_twsk_purge()") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot <syzkaller@googlegroups.com> Reported-by: Paolo Abeni <pabeni@redhat.com> Tested-by: Paolo Abeni <pabeni@redhat.com> Link: https://lore.kernel.org/r/20220126180714.845362-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-01-26 21:07:14 +03:00
refcount_inc(&tw->tw_dr->tw_refcount);
} else {
mod_timer_pending(&tw->tw_timer, jiffies + timeo);
}
}
EXPORT_SYMBOL_GPL(__inet_twsk_schedule);
void inet_twsk_purge(struct inet_hashinfo *hashinfo, int family)
{
struct inet_timewait_sock *tw;
struct sock *sk;
struct hlist_nulls_node *node;
unsigned int slot;
for (slot = 0; slot <= hashinfo->ehash_mask; slot++) {
struct inet_ehash_bucket *head = &hashinfo->ehash[slot];
restart_rcu:
cond_resched();
rcu_read_lock();
restart:
sk_nulls_for_each_rcu(sk, node, &head->chain) {
if (sk->sk_state != TCP_TIME_WAIT)
continue;
tw = inet_twsk(sk);
if ((tw->tw_family != family) ||
refcount_read(&twsk_net(tw)->ns.count))
continue;
if (unlikely(!refcount_inc_not_zero(&tw->tw_refcnt)))
continue;
if (unlikely((tw->tw_family != family) ||
refcount_read(&twsk_net(tw)->ns.count))) {
inet_twsk_put(tw);
goto restart;
}
rcu_read_unlock();
local_bh_disable();
inet_twsk_deschedule_put(tw);
local_bh_enable();
goto restart_rcu;
}
/* If the nulls value we got at the end of this lookup is
* not the expected one, we must restart lookup.
* We probably met an item that was moved to another chain.
*/
if (get_nulls_value(node) != slot)
goto restart;
rcu_read_unlock();
}
}
EXPORT_SYMBOL_GPL(inet_twsk_purge);