sched: avoid stack var in move_task_off_dead_cpu, fix
authorRusty Russell <rusty@rustcorp.com.au>
Mon, 24 Nov 2008 23:29:20 +0000 (09:59 +1030)
committerIngo Molnar <mingo@elte.hu>
Wed, 26 Nov 2008 06:58:29 +0000 (07:58 +0100)
commit1224e376f2a7e3c7ab19ef37099a78597978a696
tree244bbf8c345ac70487ddcb6283c7895f6cbdb7dd
parent3d8cbdf8650f44d95333ca645d950832a0653f35
sched: avoid stack var in move_task_off_dead_cpu, fix

Impact: locking fix

We can't call cpuset_cpus_allowed_locked() with the rq lock held.
However, the rq lock merely protects us from (1) cpu_online_mask changing
and (2) someone else changing p->cpus_allowed.

The first can't happen because we're being called from a cpu hotplug
notifier.  The second doesn't really matter: we are forcing the task off
a CPU it was affine to, so we're not doing very well anyway.

So we remove the rq lock from this path, and all is good.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/sched.c
This page took 0.026719 seconds and 5 git commands to generate.