On Wed, Oct 14, 2015 at 11:37 AM, Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> Yes, yes, it so _happens_ that "add_timer()" preferentially uses the
> current CPU etc, so in practice it may have happened to work. But
> there's absolutely zero reason to think it should always work that
> way.
Side note: even in practice, I think things like off-lining CPU's etc
(which some mobile environments seem to do as a power saving thing)
can end up moving timers to other CPU's even if they originally got
added on a particular cpu.
So I really think that the whole "schedule_delayed_work() ends up
running on the CPU" has actually never "really" been true. It has at
most been a "most of the time" thing, making it hard to see the
problem in practice.
Linus
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>