Skip to content

cloud-kernel
cloud-kernel

openanolis / cloud-kernel
大约 1 年 前同步成功

通知 158
Star 36
Fork 7
cloud-kernel
cloud-kernel

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 27 4月, 2016 1 次提交
  3. 22 4月, 2016 1 次提交
  5. 16 4月, 2016 6 次提交
    • X
      sctp: fix some rhashtable functions using in sctp proc/diag · 53fa1036
      由 Xin Long 提交于 
      When rhashtable_walk_init return err, no release function should be
called, and when rhashtable_walk_start return err, we should only invoke
rhashtable_walk_exit to release the source.

But now when sctp_transport_walk_start return err, we just call
rhashtable_walk_stop/exit, and never care about if rhashtable_walk_init
or start return err, which is so bad.

We will fix it by calling rhashtable_walk_exit if rhashtable_walk_start
return err in sctp_transport_walk_start, and if sctp_transport_walk_start
return err, we do not need to call sctp_transport_walk_stop any more.

For sctp proc, we will use 'iter->start_fail' to decide if we will call
rhashtable_walk_stop/exit.
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      53fa1036
    • X
      sctp: merge the seq_start/next/exits in remaddrs and assocs · b5e2f4e6
      由 Xin Long 提交于 
      In sctp proc, these three functions in remaddrs and assocs are the
same. we should merge them into one.
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      b5e2f4e6
    • X
      sctp: add the sctp_diag.c file · 8f840e47
      由 Xin Long 提交于 
      This one will implement all the interface of inet_diag, inet_diag_handler.
which includes sctp_diag_dump, sctp_diag_dump_one and sctp_diag_get_info.

It will work as a module, and register inet_diag_handler when loading.

v2->v3:
- fix the mistake in inet_assoc_attr_size().

- change inet_diag_msg_laddrs_fill() name to inet_diag_msg_sctpladdrs_fill.

- change inet_diag_msg_paddrs_fill() name to inet_diag_msg_sctpaddrs_fill.

- add inet_diag_msg_sctpinfo_fill() to make asoc/ep fill code clearer.

- add inet_diag_msg_sctpasoc_fill() to make asoc fill code clearer.

- merge inet_asoc_diag_fill() and inet_ep_diag_fill() to
  inet_sctp_diag_fill().

- call sctp_diag_get_info() directly, instead by handler, cause the caller
  is in the same file with it.

- call lock_sock in sctp_tsp_dump_one() to make sure we call get sctp info
  safely.

- after lock_sock(sk), we should check sk != assoc->base.sk.

- change mem[SK_MEMINFO_WMEM_ALLOC] to asoc->sndbuf_used for asoc dump when
  asoc->ep->sndbuf_policy is set. don't use INET_DIAG_MEMINFO attr any more.
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      8f840e47
    • X
      sctp: export some apis or variables for sctp_diag and reuse some for proc · 626d16f5
      由 Xin Long 提交于 
      For some main variables in sctp.ko, we couldn't export it to other modules,
so we have to define some api to access them.

It will include sctp transport and endpoint's traversal.

There are some transport traversal functions for sctp_diag, we can also
use it for sctp_proc. cause they have the similar situation to traversal
transport.

v2->v3:
- rhashtable_walk_init need the parameter gfp, because of recent upstrem
  update
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      626d16f5
    • X
      sctp: add sctp_info dump api for sctp_diag · 52c52a61
      由 Xin Long 提交于 
      sctp_diag will dump some important details of sctp's assoc or ep, we use
sctp_info to describe them,  sctp_get_sctp_info to get them, and export
it to sctp_diag.ko.

v2->v3:
- we will not use list_for_each_safe in sctp_get_sctp_info, cause
  all the callers of it will use lock_sock.

- fix the holes in struct sctp_info with __reserved* field.
  because sctp_diag is a new feature, and sctp_info is just for now,
  it may be changed in the future.
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      52c52a61
    • M
      sctp: simplify sk_receive_queue locking · 311b2177
      由 Marcelo Ricardo Leitner 提交于 
      SCTP already serializes access to rcvbuf through its sock lock:
sctp_recvmsg takes it right in the start and release at the end, while
rx path will also take the lock before doing any socket processing. On
sctp_rcv() it will check if there is an user using the socket and, if
there is, it will queue incoming packets to the backlog. The backlog
processing will do the same. Even timers will do such check and
re-schedule if an user is using the socket.

Simplifying this will allow us to remove sctp_skb_list_tail and get ride
of some expensive lockings.  The lists that it is used on are also
mangled with functions like __skb_queue_tail and __skb_unlink in the
same context, like on sctp_ulpq_tail_event() and sctp_clear_pd().
sctp_close() will also purge those while using only the sock lock.

Therefore the lockings performed by sctp_skb_list_tail() are not
necessary. This patch removes this function and replaces its calls with
just skb_queue_splice_tail_init() instead.

The biggest gain is at sctp_ulpq_tail_event(), because the events always
contain a list, even if it's queueing a single skb and this was
triggering expensive calls to spin_lock_irqsave/_irqrestore for every
data chunk received.

As SCTP will deliver each data chunk on a corresponding recvmsg, the
more effective the change will be.
Before this patch, with chunks with 30 bytes:
netperf -t SCTP_STREAM -H 192.168.1.2 -cC -l 60 -- -m 30 -S 400000
400000 -s 400000 400000
on a 10Gbit link with 1500 MTU:

SCTP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB   us/KB

425984 425984     30    60.00       137.45   7.34     7.36     52.504  52.608

With it:

SCTP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB   us/KB

425984 425984     30    60.00       179.10   7.97     6.70     43.740  36.788
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      311b2177
  7. 15 4月, 2016 1 次提交
  9. 14 4月, 2016 1 次提交
  11. 11 4月, 2016 1 次提交
    • M
      sctp: avoid refreshing heartbeat timer too often · ba6f5e33
      由 Marcelo Ricardo Leitner 提交于 
      Currently on high rate SCTP streams the heartbeat timer refresh can
consume quite a lot of resources as timer updates are costly and it
contains a random factor, which a) is also costly and b) invalidates
mod_timer() optimization for not editing a timer to the same value.
It may even cause the timer to be slightly advanced, for no good reason.

As suggested by David Laight this patch now removes this timer update
from hot path by leaving the timer on and re-evaluating upon its
expiration if the heartbeat is still needed or not, similarly to what is
done for TCP. If it's not needed anymore the timer is re-scheduled to
the new timeout, considering the time already elapsed.

For this, we now record the last tx timestamp per transport, updated in
the same spots as hb timer was restarted on tx. Also split up
sctp_transport_reset_timers into sctp_transport_reset_t3_rtx and
sctp_transport_reset_hb_timer, so we can re-arm T3 without re-arming the
heartbeat one.

On loopback with MTU of 65535 and data chunks with 1636, so that we
have a considerable amount of chunks without stressing system calls,
netperf -t SCTP_STREAM -l 30, perf looked like this before:

Samples: 103K of event 'cpu-clock', Event count (approx.): 25833000000
  Overhead  Command  Shared Object      Symbol
+    6,15%  netperf  [kernel.vmlinux]   [k] copy_user_enhanced_fast_string
-    5,43%  netperf  [kernel.vmlinux]   [k] _raw_write_unlock_irqrestore
   - _raw_write_unlock_irqrestore
      - 96,54% _raw_spin_unlock_irqrestore
         - 36,14% mod_timer
            + 97,24% sctp_transport_reset_timers
            + 2,76% sctp_do_sm
         + 33,65% __wake_up_sync_key
         + 28,77% sctp_ulpq_tail_event
         + 1,40% del_timer
      - 1,84% mod_timer
         + 99,03% sctp_transport_reset_timers
         + 0,97% sctp_do_sm
      + 1,50% sctp_ulpq_tail_event

And after this patch, now with netperf -l 60:

Samples: 230K of event 'cpu-clock', Event count (approx.): 57707250000
  Overhead  Command  Shared Object      Symbol
+    5,65%  netperf  [kernel.vmlinux]   [k] memcpy_erms
+    5,59%  netperf  [kernel.vmlinux]   [k] copy_user_enhanced_fast_string
-    5,05%  netperf  [kernel.vmlinux]   [k] _raw_spin_unlock_irqrestore
   - _raw_spin_unlock_irqrestore
      + 49,89% __wake_up_sync_key
      + 45,68% sctp_ulpq_tail_event
      - 2,85% mod_timer
         + 76,51% sctp_transport_reset_t3_rtx
         + 23,49% sctp_do_sm
      + 1,55% del_timer
+    2,50%  netperf  [sctp]             [k] sctp_datamsg_from_user
+    2,26%  netperf  [sctp]             [k] sctp_sendmsg

Throughput-wise, from 6800mbps without the patch to 7050mbps with it,
~3.7%.
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      ba6f5e33
  13. 06 4月, 2016 1 次提交
  15. 05 4月, 2016 1 次提交
  17. 31 3月, 2016 1 次提交
  19. 23 3月, 2016 1 次提交
  21. 21 3月, 2016 3 次提交
    • M
      sctp: align MTU to a word · 3822a5ff
      由 Marcelo Ricardo Leitner 提交于 
      SCTP is a protocol that is aligned to a word (4 bytes). Thus using bare
MTU can sometimes return values that are not aligned, like for loopback,
which is 65536 but ipv4_mtu() limits that to 65535. This mis-alignment
will cause the last non-aligned bytes to never be used and can cause
issues with congestion control.

So it's better to just consider a lower MTU and keep congestion control
calcs saner as they are based on PMTU.

Same applies to icmp frag needed messages, which is also fixed by this
patch.

One other effect of this is the inability to send MTU-sized packet
without queueing or fragmentation and without hitting Nagle. As the
check performed at sctp_packet_can_append_data():

if (chunk->skb->len + q->out_qlen >= transport->pathmtu - packet->overhead)
	/* Enough data queued to fill a packet */
	return SCTP_XMIT_OK;

with the above example of MTU, if there are no other messages queued,
one cannot send a packet that just fits one packet (65532 bytes) and
without causing DATA chunk fragmentation or a delay.

v2:
 - Added WORD_TRUNC macro
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      3822a5ff
    • M
      sctp: do not leak chunks that are sent to unconfirmed paths · 31b055ef
      由 Marcelo Ricardo Leitner 提交于 
      Currently, if a chunk is scheduled to be sent through a transport that
is currently unconfirmed, it will be leaked as it is dequeued from outq
and is not re-queued nor freed.

As I'm not aware of any situation that may lead to this situation, I'm
fixing this by freeing the chunk and also logging a trace so that we can
fix the other bug if it ever happens.
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      31b055ef
    • M
      sctp: do not update a_rwnd if we are not issuing a sack · 07b4d6a1
      由 Marcelo Ricardo Leitner 提交于 
      The SACK can be lost pretty much elsewhere, but if its allocation fail,
we know we are not sending it, so it is better to revert a_rwnd to its
previous value as this may give it a chance to issue a window update
later.
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      07b4d6a1
  23. 17 3月, 2016 1 次提交
  25. 14 3月, 2016 2 次提交
    • M
      sctp: allow sctp_transmit_packet and others to use gfp · cea8768f
      由 Marcelo Ricardo Leitner 提交于 
      Currently sctp_sendmsg() triggers some calls that will allocate memory
with GFP_ATOMIC even when not necessary. In the case of
sctp_packet_transmit it will allocate a linear skb that will be used to
construct the packet and this may cause sends to fail due to ENOMEM more
often than anticipated specially with big MTUs.

This patch thus allows it to inherit gfp flags from upper calls so that
it can use GFP_KERNEL if it was triggered by a sctp_sendmsg call or
similar. All others, like retransmits or flushes started from BH, are
still allocated using GFP_ATOMIC.

In netperf tests this didn't result in any performance drawbacks when
memory is not too fragmented and made it trigger ENOMEM way less often.
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      cea8768f
    • X
      sctp: fix the transports round robin issue when init is retransmitted · 39d2adeb
      由 Xin Long 提交于 
      prior to this patch, at the beginning if we have two paths in one assoc,
they may have the same params other than the last_time_heard, it will try
the paths like this:

1st cycle
  try trans1 fail.
  then trans2 is selected.(cause it's last_time_heard is after trans1).

2nd cycle:
  try  trans2 fail
  then trans2 is selected.(cause it's last_time_heard is after trans1).

3rd cycle:
  try  trans2 fail
  then trans2 is selected.(cause it's last_time_heard is after trans1).

....

trans1 will never have change to be selected, which is not what we expect.
we should keeping round robin all the paths if they are just added at the
beginning.

So at first every tranport's last_time_heard should be initialized 0, so
that we ensure they have the same value at the beginning, only by this,
all the transports could get equal chance to be selected.

Then for sctp_trans_elect_best, it should return the trans_next one when
*trans == *trans_next, so that we can try next if it fails,  but now it
always return trans. so we can fix it by exchanging these two params when
we calls sctp_trans_elect_tie().

Fixes: 4c47af4d ('net: sctp: rework multihoming retransmission path selection to rfc4960')
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      39d2adeb
  27. 09 3月, 2016 1 次提交
  29. 02 3月, 2016 3 次提交
  31. 22 2月, 2016 1 次提交
    • N
      sctp: Fix port hash table size computation · d9749fb5
      由 Neil Horman 提交于 
      Dmitry Vyukov noted recently that the sctp_port_hashtable had an error in
its size computation, observing that the current method never guaranteed
that the hashsize (measured in number of entries) would be a power of two,
which the input hash function for that table requires.  The root cause of
the problem is that two values need to be computed (one, the allocation
order of the storage requries, as passed to __get_free_pages, and two the
number of entries for the hash table).  Both need to be ^2, but for
different reasons, and the existing code is simply computing one order
value, and using it as the basis for both, which is wrong (i.e. it assumes
that ((1<<order)*PAGE_SIZE)/sizeof(bucket) is still ^2 when its not).

To fix this, we change the logic slightly.  We start by computing a goal
allocation order (which is limited by the maximum size hash table we want
to support.  Then we attempt to allocate that size table, decreasing the
order until a successful allocation is made.  Then, with the resultant
successful order we compute the number of buckets that hash table supports,
which we then round down to the nearest power of two, giving us the number
of entries the table actually supports.

I've tested this locally here, using non-debug and spinlock-debug kernels,
and the number of entries in the hashtable consistently work out to be
powers of two in all cases.
Signed-off-by: NNeil Horman <nhorman@tuxdriver.com>
Reported-by: NDmitry Vyukov <dvyukov@google.com>
CC: Dmitry Vyukov <dvyukov@google.com>
CC: Vladislav Yasevich <vyasevich@gmail.com>
CC: "David S. Miller" <davem@davemloft.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d9749fb5
  33. 18 2月, 2016 3 次提交
  35. 11 2月, 2016 1 次提交
  37. 09 2月, 2016 1 次提交
  39. 29 1月, 2016 3 次提交
  41. 27 1月, 2016 1 次提交
  43. 25 1月, 2016 1 次提交
  45. 18 1月, 2016 1 次提交
  47. 16 1月, 2016 2 次提交
  49. 12 1月, 2016 1 次提交