Commit 6a190051 authored Nov 15, 2015 by Oleg Nesterov Committed by Ingo Molnar Nov 23, 2015

stop_machine: Don't disable preemption in stop_two_cpus()



Now that stop_two_cpus() path does not check cpu_active() we can remove
preempt_disable(), it was only needed to ensure that stop_machine() can
not be called after we observe cpu_active() == T and before we queue the
new work.

Also, turn the pointless and confusing ->executed check into WARN_ON().
We know that both works must be executed, otherwise we have a bug. And
in fact I think that done->executed should die, see the next changes.

Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Milos Vyletel <milos@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20151115193314.GA8249@redhat.com


Signed-off-by: Ingo Molnar <mingo@kernel.org>

parent 64038f29

kernel/stop_machine.c

+3 −8

Original line number	Diff line number	Diff line
		@@ -258,7 +258,6 @@ int stop_two_cpus(unsigned int cpu1, unsigned int cpu2, cpu_stop_fn_t fn, void *
		struct cpu_stop_work work1, work2;
		struct multi_stop_data msdata;

		preempt_disable();
		msdata = (struct multi_stop_data){
		.fn = fn,
		.data = arg,
		@@ -277,16 +276,12 @@ int stop_two_cpus(unsigned int cpu1, unsigned int cpu2, cpu_stop_fn_t fn, void *

		if (cpu1 > cpu2)
		swap(cpu1, cpu2);
		if (cpu_stop_queue_two_works(cpu1, &work1, cpu2, &work2)) {
		preempt_enable();
		if (cpu_stop_queue_two_works(cpu1, &work1, cpu2, &work2))
		return -ENOENT;
		}

		preempt_enable();

		wait_for_completion(&done.completion);

		return done.executed ? done.ret : -ENOENT;
		WARN_ON(!done.executed);
		return done.ret;
		}

		/**