projects / linux.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 5585263) | patch
task_work: add a scheduling point in task_work_run()
authorEric Dumazet <edumazet@google.com>
Tue, 21 Aug 2012 13:05:14 +0000 (15:05 +0200)
committerAl Viro <viro@zeniv.linux.org.uk>
Wed, 22 Aug 2012 14:31:05 +0000 (10:31 -0400)
commit88ec2789d856056344161aa20420dd37e893b0fe
tree0e944baca39ae432b11ca105fa5c8d0cdd6a47a5tree | snapshot
parent55852635a8e2803cbc22d0e143d727813f0fcdb5commit | diff
task_work: add a scheduling point in task_work_run()

It seems commit 4a9d4b02 (switch fput to task_work_add) reintroduced
the problem addressed in commit 944be0b2 (close_files(): add scheduling
point)

If a server process with a lot of files (say 2 million tcp sockets)
is killed, we can spend a lot of time in task_work_run() and trigger
a soft lockup.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
kernel/task_work.c diff | blob | history
ep93xx device tree rework repo: git://git.maquefel.me/linux.git