lkml.org 
[lkml]   [2014]   [Dec]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH 2/2] sched: Pull resched loop to __schedule() callers
From
On Mon, Dec 15, 2014 at 8:32 AM, Frederic Weisbecker <fweisbec@gmail.com> wrote:
>
> But it may have an impact on some context checkers that rely on in_atomic*()
> which ignore the PREEMPT_ACTIVE value. It shouldn't ignore that though but I
> guess it's a hack for some specific situation.

I think we should remove it. The only reason for it is the scheduler
itself, which used to have the in_atomic() check (ok, still has, it's
just called "in_atomic_preempt_off()").

But yes, if we keep the "mask off PREEMPT_ACTIVE" in in_atomic(), then
we do need to update the counts with "PREEMPT_ACTIVE+1" instead. Or
something like that.

Linus


\
 
 \ /
  Last update: 2014-12-15 20:41    [W:0.049 / U:0.120 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site