linux-kernel - Re: [RFC] sched: make callers check lock contention for cond_resched

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: <4FA34244.4000405@linux.vnet.ibm.com>
Date:	Fri, 04 May 2012 10:43:16 +0800
From:	Michael Wang <wangyun@...ux.vnet.ibm.com>
To:	Takuya Yoshikawa <takuya.yoshikawa@...il.com>
CC:	Peter Zijlstra <peterz@...radead.org>, mingo@...e.hu,
	linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org,
	kvm@...r.kernel.org, avi@...hat.com, mtosatti@...hat.com,
	yoshikawa.takuya@....ntt.co.jp
Subject: Re: [RFC] sched: make callers check lock contention for cond_resched_lock()

On 05/03/2012 09:00 PM, Takuya Yoshikawa wrote:

> On Thu, 03 May 2012 14:29:10 +0200
> Peter Zijlstra <peterz@...radead.org> wrote:
> 
>> On Thu, 2012-05-03 at 21:22 +0900, Takuya Yoshikawa wrote:
>>> Although the real use case is out of this RFC patch, we are now discussing
>>> a case in which we may hold a spin_lock for long time, ms order, depending
>>> on workload;  and in that case, other threads -- VCPU threads -- should be
>>> given higher priority for that problematic lock. 
>>
>> Firstly, if you can hold a lock that long, it shouldn't be a spinlock,
> 
> I agree with you in principle, but isn't cond_resched_lock() there for that?
> 
>> secondly why isn't TIF_RESCHED being set if its running that long? That
>> should still make cond_resched_lock() break.
> 
> I see.
> 
> I did some tests using spin_is_contended() and need_resched() and saw
> that need_resched() was called as often as spin_is_contended(), so
> experimentally I understand your point.
> 
> But as I could not see why spin_needbreak() was differently implemented
> depending on CONFIG_PREEMPT, I wanted to understand the meaning.


I think enable CONFIG_PREEMPT means allow preemption in kernel, so if
disabled, we can't reschedule a task if it is running in kernel not the
user space at a given time.

As the cond_resched_lock() was invoked in kernel, and looks like
cpu_relax() will give up cpu(I'm not sure whether this will invoke
schedule on some arch, just because that name...), so we can't do break
if CONFIG_PREEMPT disabled, because that will cause kernel preemption
while not allowed.

May be that's the reason why we need to consider CONFIG_PREEMPT in
spin_needbreak().

Regards,
Michael Wang

> 
> Thanks,
> 	Takuya
> --
> 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/
> 


--
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/