sched: replace PF_THREAD_BOUND with PF_NO_SETAFFINITY
authorTejun Heo <tj@kernel.org>
Tue, 19 Mar 2013 20:45:20 +0000 (13:45 -0700)
committerTejun Heo <tj@kernel.org>
Tue, 19 Mar 2013 20:45:20 +0000 (13:45 -0700)
commit14a40ffccd6163bbcd1d6f32b28a88ffe6149fc6
treeeb61e5bf7b64c3e67f3e33fe6b07fde4ee1d4d43
parent2e109a2855bf6cf675a8b74dbd89b6492e8def42
sched: replace PF_THREAD_BOUND with PF_NO_SETAFFINITY

PF_THREAD_BOUND was originally used to mark kernel threads which were
bound to a specific CPU using kthread_bind() and a task with the flag
set allows cpus_allowed modifications only to itself.  Workqueue is
currently abusing it to prevent userland from meddling with
cpus_allowed of workqueue workers.

What we need is a flag to prevent userland from messing with
cpus_allowed of certain kernel tasks.  In kernel, anyone can
(incorrectly) squash the flag, and, for worker-type usages,
restricting cpus_allowed modification to the task itself doesn't
provide meaningful extra proection as other tasks can inject work
items to the task anyway.

This patch replaces PF_THREAD_BOUND with PF_NO_SETAFFINITY.
sched_setaffinity() checks the flag and return -EINVAL if set.
set_cpus_allowed_ptr() is no longer affected by the flag.

This will allow simplifying workqueue worker CPU affinity management.

Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
include/linux/sched.h
kernel/cgroup.c
kernel/cpuset.c
kernel/kthread.c
kernel/sched/core.c
kernel/workqueue.c