linux/include
Kuniyuki Iwashima d9f21b3613 af_unix: Try to run GC async.
If more than 16000 inflight AF_UNIX sockets exist and the garbage
collector is not running, unix_(dgram|stream)_sendmsg() call unix_gc().
Also, they wait for unix_gc() to complete.

In unix_gc(), all inflight AF_UNIX sockets are traversed at least once,
and more if they are the GC candidate.  Thus, sendmsg() significantly
slows down with too many inflight AF_UNIX sockets.

However, if a process sends data with no AF_UNIX FD, the sendmsg() call
does not need to wait for GC.  After this change, only the process that
meets the condition below will be blocked under such a situation.

  1) cmsg contains AF_UNIX socket
  2) more than 32 AF_UNIX sent by the same user are still inflight

Note that even a sendmsg() call that does not meet the condition but has
AF_UNIX FD will be blocked later in unix_scm_to_skb() by the spinlock,
but we allow that as a bonus for sane users.

The results below are the time spent in unix_dgram_sendmsg() sending 1
byte of data with no FD 4096 times on a host where 32K inflight AF_UNIX
sockets exist.

Without series: the sane sendmsg() needs to wait gc unreasonably.

  $ sudo /usr/share/bcc/tools/funclatency -p 11165 unix_dgram_sendmsg
  Tracing 1 functions for "unix_dgram_sendmsg"... Hit Ctrl-C to end.
  ^C
       nsecs               : count     distribution
  [...]
      524288 -> 1048575    : 0        |                                        |
     1048576 -> 2097151    : 3881     |****************************************|
     2097152 -> 4194303    : 214      |**                                      |
     4194304 -> 8388607    : 1        |                                        |

  avg = 1825567 nsecs, total: 7477526027 nsecs, count: 4096

With series: the sane sendmsg() can finish much faster.

  $ sudo /usr/share/bcc/tools/funclatency -p 8702  unix_dgram_sendmsg
  Tracing 1 functions for "unix_dgram_sendmsg"... Hit Ctrl-C to end.
  ^C
       nsecs               : count     distribution
  [...]
         128 -> 255        : 0        |                                        |
         256 -> 511        : 4092     |****************************************|
         512 -> 1023       : 2        |                                        |
        1024 -> 2047       : 0        |                                        |
        2048 -> 4095       : 0        |                                        |
        4096 -> 8191       : 1        |                                        |
        8192 -> 16383      : 1        |                                        |

  avg = 410 nsecs, total: 1680510 nsecs, count: 4096

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20240123170856.41348-6-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-01-26 20:34:25 -08:00
..
acpi IOMMU Updates for Linux v6.8 2024-01-18 15:16:57 -08:00
asm-generic RISC-V Patches for the 6.8 Merge Window, Part 4 2024-01-20 11:06:04 -08:00
clocksource
crypto crypto: skcipher - remove excess kerneldoc members 2023-12-29 11:25:56 +08:00
drm drm-next for 6.8: 2024-01-12 11:32:19 -08:00
dt-bindings dmaengine fixes for v6.8-rc1 2024-01-20 15:03:25 -08:00
keys
kunit Networking changes for 6.8. 2024-01-11 10:07:29 -08:00
kvm
linux Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2024-01-25 14:20:08 -08:00
math-emu
media media: videobuf2: core: Rename min_buffers_needed field in vb2_queue 2023-12-13 17:31:27 +01:00
memory
misc
net af_unix: Try to run GC async. 2024-01-26 20:34:25 -08:00
pcmcia
ras
rdma RDMA/core: Fix umem iterator when PAGE_SIZE is greater then HCA pgsz 2023-12-04 20:02:41 -04:00
rv
scsi
soc - New Device Support 2024-01-17 15:21:21 -08:00
sound ASoC: Fixes for v6.8 2024-01-16 17:37:17 +01:00
target
trace afs: Fix error handling with lookup via FS.InlineBulkStatus 2024-01-22 22:30:14 +00:00
uapi for-6.8-rc1-tag 2024-01-22 13:29:42 -08:00
ufs scsi: ufs: core: Rename ufshcd_auto_hibern8_enable() and make it static 2023-12-18 20:51:14 -05:00
vdso
video video/sticore: Remove info field from STI struct 2024-01-12 12:38:37 +01:00
xen xen: update PV-device interface headers 2024-01-09 11:46:24 +01:00