linux-kernel - Re: [PATCH] sched: resched proper CPU on yield

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <AANLkTikL28vknFkp1NBkqQq3wpZK8A-C5WABKiYsqR2o@mail.gmail.com>
Date:	Tue, 1 Mar 2011 19:37:07 -0800
From:	Venkatesh Pallipadi <venki@...gle.com>
To:	Rik van Riel <riel@...hat.com>
Cc:	Mike Galbraith <efault@....de>, a.p.zijlstra@...llo.nl,
	mingo@...hat.com, hpa@...or.com, linux-kernel@...r.kernel.org,
	mtosatti@...hat.com, tglx@...utronix.de, mingo@...e.hu
Subject: Re: [PATCH] sched: resched proper CPU on yield_to

On Tue, Mar 1, 2011 at 7:33 PM, Rik van Riel <riel@...hat.com> wrote:
> On 03/01/2011 07:28 PM, Venkatesh Pallipadi wrote:
>>
>> yield_to_task_fair() has code to resched the CPU of yielding task when the
>> intention is to resched the CPU of the task that is being yielded to.
>>
>> Change here fixes the problem and also makes the resched conditional on
>> rq != p_rq.
>
> That would result in not rescheduling when current and p are
> on the same runqueue, in effect making yield_to a noop for the
> easiest case...
>
> When rq != p_rq, we need to ensure both get rescheduled.
>

Yes. There is a schedule() right after this change which should take
care or resched on current CPU. Thats the reason I thought of skipping
resched on current CPU. No?

Thanks,
Venki

> We want to have current not run right now (because it is waiting
> on a resource that's not available), and we do want p to run.
>
> I'm about to fall over, so I'll go to sleep now.
>
> I can send a patch tomorrow morning, unless you beat me to it :)
>
> --
> All rights reversed
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/