diff options
author | 2012-08-21 15:05:14 +0200 | |
---|---|---|
committer | 2012-08-22 10:31:05 -0400 | |
commit | 88ec2789d856056344161aa20420dd37e893b0fe (patch) | |
tree | 0e944baca39ae432b11ca105fa5c8d0cdd6a47a5 /lib/mpi/mpi-bit.c | |
parent | fs: fix fs/namei.c kernel-doc warnings (diff) | |
download | linux-dev-88ec2789d856056344161aa20420dd37e893b0fe.tar.xz linux-dev-88ec2789d856056344161aa20420dd37e893b0fe.zip |
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>
Diffstat (limited to 'lib/mpi/mpi-bit.c')
0 files changed, 0 insertions, 0 deletions