Skip to content

K
Kernel

openeuler / Kernel
大约 1 年 前同步成功

通知 5
Star 0
Fork 0
K
Kernel

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 04 12月, 2009 3 次提交
    • E
      tcp: connect() race with timewait reuse · 13475a30
      由 Eric Dumazet 提交于 
      Its currently possible that several threads issuing a connect() find
the same timewait socket and try to reuse it, leading to list
corruptions.

Condition for bug is that these threads bound their socket on same
address/port of to-be-find timewait socket, and connected to same
target. (SO_REUSEADDR needed)

To fix this problem, we could unhash timewait socket while holding
ehash lock, to make sure lookups/changes will be serialized. Only
first thread finds the timewait socket, other ones find the
established socket and return an EADDRNOTAVAIL error.

This second version takes into account Evgeniy's review and makes sure
inet_twsk_put() is called outside of locked sections.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      13475a30
    • E
      net: Batch inet_twsk_purge · b099ce26
      由 Eric W. Biederman 提交于 
      This function walks the whole hashtable so there is no point in
passing it a network namespace.  Instead I purge all timewait
sockets from dead network namespaces that I find.  If the namespace
is one of the once I am trying to purge I am guaranteed no new timewait
sockets can be formed so this will get them all.  If the namespace
is one I am not acting for it might form a few more but I will
call inet_twsk_purge again and  shortly to get rid of them.  In
any even if the network namespace is dead timewait sockets are
useless.

Move the calls of inet_twsk_purge into batch_exit routines so
that if I am killing a bunch of namespaces at once I will just
call inet_twsk_purge once and save a lot of redundant unnecessary
work.

My simple 4k network namespace exit test the cleanup time dropped from
roughly 8.2s to 1.6s.  While the time spent running inet_twsk_purge fell
to about 2ms.  1ms for ipv4 and 1ms for ipv6.
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      b099ce26
    • E
      net: Use rcu lookups in inet_twsk_purge. · 575f4cd5
      由 Eric W. Biederman 提交于 
      While we are looking up entries to free there is no reason to take
the lock in inet_twsk_purge.  We have to drop locks and restart
occassionally anyway so adding a few more in case we get on the
wrong list because of a timewait move is no big deal.  At the
same time not taking the lock for long periods of time is much
more polite to the rest of the users of the hash table.

In my test configuration of killing 4k network namespaces
this change causes 4k back to back runs of inet_twsk_purge on an
empty hash table to go from roughly 20.7s to 3.3s, and the total
time to destroy 4k network namespaces goes from roughly 44s to
3.3s.
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Acked-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      575f4cd5
  3. 19 10月, 2009 1 次提交
    • E
      inet: rename some inet_sock fields · c720c7e8
      由 Eric Dumazet 提交于 
      In order to have better cache layouts of struct sock (separate zones
for rx/tx paths), we need this preliminary patch.

Goal is to transfert fields used at lookup time in the first
read-mostly cache line (inside struct sock_common) and move sk_refcnt
to a separate cache line (only written by rx path)

This patch adds inet_ prefix to daddr, rcv_saddr, dport, num, saddr,
sport and id fields. This allows a future patch to define these
fields as macros, like sk_refcnt, without name clashes.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      c720c7e8
  5. 13 10月, 2009 1 次提交
  7. 29 8月, 2009 1 次提交
    • O
      tcp: fix premature termination of FIN_WAIT2 time-wait sockets · 80a1096b
      由 Octavian Purdila 提交于 
      There is a race condition in the time-wait sockets code that can lead
to premature termination of FIN_WAIT2 and, subsequently, to RST
generation when the FIN,ACK from the peer finally arrives:

Time     TCP header
0.000000 30755 > http [SYN] Seq=0 Win=2920 Len=0 MSS=1460 TSV=282912 TSER=0
0.000008 http > 30755 aSYN, ACK] Seq=0 Ack=1 Win=2896 Len=0 MSS=1460 TSV=...
0.136899 HEAD /1b.html?n1Lg=v1 HTTP/1.0 [Packet size limited during capture]
0.136934 HTTP/1.0 200 OK [Packet size limited during capture]
0.136945 http > 30755 [FIN, ACK] Seq=187 Ack=207 Win=2690 Len=0 TSV=270521...
0.136974 30755 > http [ACK] Seq=207 Ack=187 Win=2734 Len=0 TSV=283049 TSER=...
0.177983 30755 > http [ACK] Seq=207 Ack=188 Win=2733 Len=0 TSV=283089 TSER=...
0.238618 30755 > http [FIN, ACK] Seq=207 Ack=188 Win=2733 Len=0 TSV=283151...
0.238625 http > 30755 [RST] Seq=188 Win=0 Len=0

Say twdr->slot = 1 and we are running inet_twdr_hangman and in this
instance inet_twdr_do_twkill_work returns 1. At that point we will
mark slot 1 and schedule inet_twdr_twkill_work. We will also make
twdr->slot = 2.

Next, a connection is closed and tcp_time_wait(TCP_FIN_WAIT2, timeo)
is called which will create a new FIN_WAIT2 time-wait socket and will
place it in the last to be reached slot, i.e. twdr->slot = 1.

At this point say inet_twdr_twkill_work will run which will start
destroying the time-wait sockets in slot 1, including the just added
TCP_FIN_WAIT2 one.

To avoid this issue we increment the slot only if all entries in the
slot have been purged.

This change may delay the slots cleanup by a time-wait death row
period but only if the worker thread didn't had the time to run/purge
the current slot in the next period (6 seconds with default sysctl
settings). However, on such a busy system even without this change we
would probably see delays...
Signed-off-by: NOctavian Purdila <opurdila@ixiacom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      80a1096b
  9. 15 6月, 2009 1 次提交
  11. 07 5月, 2009 1 次提交
  13. 21 11月, 2008 1 次提交
  15. 17 11月, 2008 1 次提交
    • E
      net: Convert TCP & DCCP hash tables to use RCU / hlist_nulls · 3ab5aee7
      由 Eric Dumazet 提交于 
      RCU was added to UDP lookups, using a fast infrastructure :
- sockets kmem_cache use SLAB_DESTROY_BY_RCU and dont pay the
  price of call_rcu() at freeing time.
- hlist_nulls permits to use few memory barriers.

This patch uses same infrastructure for TCP/DCCP established
and timewait sockets.

Thanks to SLAB_DESTROY_BY_RCU, no slowdown for applications
using short lived TCP connections. A followup patch, converting
rwlocks to spinlocks will even speedup this case.

__inet_lookup_established() is pretty fast now we dont have to
dirty a contended cache line (read_lock/read_unlock)

Only established and timewait hashtable are converted to RCU
(bind table and listen table are still using traditional locking)
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      3ab5aee7
  17. 01 10月, 2008 1 次提交
  19. 09 9月, 2008 1 次提交
    • D
      netns : fix kernel panic in timewait socket destruction · d315492b
      由 Daniel Lezcano 提交于 
      How to reproduce ?
 - create a network namespace
 - use tcp protocol and get timewait socket
 - exit the network namespace
 - after a moment (when the timewait socket is destroyed), the kernel
   panics.

# BUG: unable to handle kernel NULL pointer dereference at
0000000000000007
IP: [<ffffffff821e394d>] inet_twdr_do_twkill_work+0x6e/0xb8
PGD 119985067 PUD 11c5c0067 PMD 0
Oops: 0000 [1] SMP
CPU 1
Modules linked in: ipv6 button battery ac loop dm_mod tg3 libphy ext3 jbd
edd fan thermal processor thermal_sys sg sata_svw libata dock serverworks
sd_mod scsi_mod ide_disk ide_core [last unloaded: freq_table]
Pid: 0, comm: swapper Not tainted 2.6.27-rc2 #3
RIP: 0010:[<ffffffff821e394d>] [<ffffffff821e394d>]
inet_twdr_do_twkill_work+0x6e/0xb8
RSP: 0018:ffff88011ff7fed0 EFLAGS: 00010246
RAX: ffffffffffffffff RBX: ffffffff82339420 RCX: ffff88011ff7ff30
RDX: 0000000000000001 RSI: ffff88011a4d03c0 RDI: ffff88011ac2fc00
RBP: ffffffff823392e0 R08: 0000000000000000 R09: ffff88002802a200
R10: ffff8800a5c4b000 R11: ffffffff823e4080 R12: ffff88011ac2fc00
R13: 0000000000000001 R14: 0000000000000001 R15: 0000000000000000
FS: 0000000041cbd940(0000) GS:ffff8800bff839c0(0000)
knlGS:0000000000000000
CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b
CR2: 0000000000000007 CR3: 00000000bd87c000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process swapper (pid: 0, threadinfo ffff8800bff9e000, task
ffff88011ff76690)
Stack: ffffffff823392e0 0000000000000100 ffffffff821e3a3a
0000000000000008
0000000000000000 ffffffff821e3a61 ffff8800bff7c000 ffffffff8203c7e7
ffff88011ff7ff10 ffff88011ff7ff10 0000000000000021 ffffffff82351108
Call Trace:
<IRQ> [<ffffffff821e3a3a>] ? inet_twdr_hangman+0x0/0x9e
[<ffffffff821e3a61>] ? inet_twdr_hangman+0x27/0x9e
[<ffffffff8203c7e7>] ? run_timer_softirq+0x12c/0x193
[<ffffffff820390d1>] ? __do_softirq+0x5e/0xcd
[<ffffffff8200d08c>] ? call_softirq+0x1c/0x28
[<ffffffff8200e611>] ? do_softirq+0x2c/0x68
[<ffffffff8201a055>] ? smp_apic_timer_interrupt+0x8e/0xa9
[<ffffffff8200cad6>] ? apic_timer_interrupt+0x66/0x70
<EOI> [<ffffffff82011f4c>] ? default_idle+0x27/0x3b
[<ffffffff8200abbd>] ? cpu_idle+0x5f/0x7d


Code: e8 01 00 00 4c 89 e7 41 ff c5 e8 8d fd ff ff 49 8b 44 24 38 4c 89 e7
65 8b 14 25 24 00 00 00 89 d2 48 8b 80 e8 00 00 00 48 f7 d0 <48> 8b 04 d0
48 ff 40 58 e8 fc fc ff ff 48 89 df e8 c0 5f 04 00
RIP [<ffffffff821e394d>] inet_twdr_do_twkill_work+0x6e/0xb8
RSP <ffff88011ff7fed0>
CR2: 0000000000000007

This patch provides a function to purge all timewait sockets related
to a network namespace. The timewait sockets life cycle is not tied with
the network namespace, that means the timewait sockets stay alive while
the network namespace dies. The timewait sockets are for avoiding to
receive a duplicate packet from the network, if the network namespace is
freed, the network stack is removed, so no chance to receive any packets
from the outside world. Furthermore, having a pending destruction timer
on these sockets with a network namespace freed is not safe and will lead
to an oops if the timer callback which try to access data belonging to 
the namespace like for example in:
	inet_twdr_do_twkill_work
		-> NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_TIMEWAITED);

Purging the timewait sockets at the network namespace destruction will:
 1) speed up memory freeing for the namespace
 2) fix kernel panic on asynchronous timewait destruction
Signed-off-by: NDaniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: NDenis V. Lunev <den@openvz.org>
Acked-by: NEric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d315492b
  21. 26 7月, 2008 1 次提交
  23. 17 7月, 2008 1 次提交
  25. 17 6月, 2008 1 次提交
  27. 16 4月, 2008 1 次提交
  29. 11 4月, 2008 1 次提交
  31. 01 4月, 2008 1 次提交
  33. 26 3月, 2008 1 次提交
  35. 21 3月, 2008 1 次提交
  37. 29 1月, 2008 3 次提交
  39. 07 11月, 2007 1 次提交
    • E
      [INET]: Remove per bucket rwlock in tcp/dccp ehash table. · 230140cf
      由 Eric Dumazet 提交于 
      As done two years ago on IP route cache table (commit
22c047cc) , we can avoid using one
lock per hash bucket for the huge TCP/DCCP hash tables.

On a typical x86_64 platform, this saves about 2MB or 4MB of ram, for
litle performance differences. (we hit a different cache line for the
rwlock, but then the bucket cache line have a better sharing factor
among cpus, since we dirty it less often). For netstat or ss commands
that want a full scan of hash table, we perform fewer memory accesses.

Using a 'small' table of hashed rwlocks should be more than enough to
provide correct SMP concurrency between different buckets, without
using too much memory. Sizing of this table depends on
num_possible_cpus() and various CONFIG settings.

This patch provides some locking abstraction that may ease a future
work using a different model for TCP/DCCP table.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Acked-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      230140cf
  41. 11 10月, 2007 1 次提交
  43. 15 7月, 2007 1 次提交
  45. 09 2月, 2007 1 次提交
    • E
      [NET]: change layout of ehash table · dbca9b27
      由 Eric Dumazet 提交于 
      ehash table layout is currently this one :

First half of this table is used by sockets not in TIME_WAIT state
Second half of it is used by sockets in TIME_WAIT state.

This is non optimal because of for a given hash or socket, the two chain heads 
are located in separate cache lines.
Moreover the locks of the second half are never used.

If instead of this halving, we use two list heads in inet_ehash_bucket instead 
of only one, we probably can avoid one cache miss, and reduce ram usage, 
particularly if sizeof(rwlock_t) is big (various CONFIG_DEBUG_SPINLOCK, 
CONFIG_DEBUG_LOCK_ALLOC settings). So we still halves the table but we keep 
together related chains to speedup lookups and socket state change.

In this patch I did not try to align struct inet_ehash_bucket, but a future 
patch could try to make this structure have a convenient size (a power of two 
or a multiple of L1_CACHE_SIZE).
I guess rwlock will just vanish as soon as RCU is plugged into ehash :) , so 
maybe we dont need to scratch our heads to align the bucket...

Note : In case struct inet_ehash_bucket is not a power of two, we could 
probably change alloc_large_system_hash() (in case it use __get_free_pages()) 
to free the unused space. It currently allocates a big zone, but the last 
quarter of it could be freed. Again, this should be a temporary 'problem'.

Patch tested on ipv4 tcp only, but should be OK for IPV6 and DCCP.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      dbca9b27
  47. 08 12月, 2006 1 次提交
  49. 07 12月, 2006 2 次提交
  51. 22 11月, 2006 1 次提交
    • D
      WorkStruct: Pass the work_struct pointer instead of context data · 65f27f38
      由 David Howells 提交于 
      Pass the work_struct pointer to the work function rather than context data.
The work function can use container_of() to work out the data.

For the cases where the container of the work_struct may go away the moment the
pending bit is cleared, it is made possible to defer the release of the
structure by deferring the clearing of the pending bit.

To make this work, an extra flag is introduced into the management side of the
work_struct.  This governs auto-release of the structure upon execution.

Ordinarily, the work queue executor would release the work_struct for further
scheduling or deallocation by clearing the pending bit prior to jumping to the
work function.  This means that, unless the driver makes some guarantee itself
that the work_struct won't go away, the work function may not access anything
else in the work_struct or its container lest they be deallocated..  This is a
problem if the auxiliary data is taken away (as done by the last patch).

However, if the pending bit is *not* cleared before jumping to the work
function, then the work function *may* access the work_struct and its container
with no problems.  But then the work function must itself release the
work_struct by calling work_release().

In most cases, automatic release is fine, so this is the default.  Special
initiators exist for the non-auto-release case (ending in _NAR).
Signed-Off-By: NDavid Howells <dhowells@redhat.com>
      65f27f38
  53. 01 7月, 2006 1 次提交
  55. 04 1月, 2006 1 次提交
  57. 11 10月, 2005 1 次提交
  59. 04 10月, 2005 1 次提交
    • E
      [INET]: speedup inet (tcp/dccp) lookups · 81c3d547
      由 Eric Dumazet 提交于 
      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: NEric Dumazet <dada1@cosmosbay.com>
Acked-by: NArnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      81c3d547
  61. 30 8月, 2005 5 次提交