sched: Replace spin_unlock_wait() with lock/unlock pair
authorPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Thu, 29 Jun 2017 19:08:26 +0000 (12:08 -0700)
committerPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Fri, 11 Aug 2017 20:09:14 +0000 (13:09 -0700)
commit23a9b748a3d27f67cdb078fcb891a920285e75d9
tree3939c14a9385cb193210f2fb20e0edf4b96cf55b
parentf274f1e72d7171c80c8c790040e47a23a74796b6
sched: Replace spin_unlock_wait() with lock/unlock pair

There is no agreed-upon definition of spin_unlock_wait()'s semantics,
and it appears that all callers could do just as well with a lock/unlock
pair.  This commit therefore replaces the spin_unlock_wait() call in
do_task_dead() with spin_lock() followed immediately by spin_unlock().
This should be safe from a performance perspective because the lock is
this tasks ->pi_lock, and this is called only after the task exits.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Andrea Parri <parri.andrea@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
[ paulmck: Drop smp_mb() based on Peter Zijlstra's analysis:
  http://lkml.kernel.org/r/20170811144150.26gowhxte7ri5fpk@hirez.programming.kicks-ass.net ]
kernel/sched/core.c