linux-kernel - Re: [Query] Preemption (hogging) of the work handler

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]

Date:	Tue, 12 Jul 2016 07:12:18 -0700
From:	Viresh Kumar <viresh.kumar@...aro.org>
To:	Sergey Senozhatsky <sergey.senozhatsky@...il.com>
Cc:	Petr Mladek <pmladek@...e.com>,
	Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>,
	Jan Kara <jack@...e.cz>, Tejun Heo <tj@...nel.org>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	vlevenetz@...sol.com, vaibhav.hiremath@...aro.org,
	alex.elder@...aro.org, johan@...nel.org, akpm@...ux-foundation.org,
	rostedt@...dmis.org
Subject: Re: [Query] Preemption (hogging) of the work handler

Hi Sergey,

On 12-07-16, 23:03, Sergey Senozhatsky wrote:
> Thanks, Petr.
> 
> 
> so, I'm looking at this thing now:

Thanks. Though just to mention, I have seen this only once yet :)

> : [   12.874909] sched: RT throttling activated for rt_rq ffffffc0ac13fcd0 (cpu 0)
> : [   12.874909] potential CPU hogs:
> : [   12.874909]  printk (292)
> 
> so it's either cond_resched() does not reshed, keeping printk kthread
> active, which, however, upsets the sched and triggers throttling (umm, what);
> 
> or we, somehow, have `console_may_schedule == 0' in this final console_unlock(),
> so cond_resched() never happens.
> 
> I'm looking at mainline 3.10, tho.

My setup currently looks like this:
- 3.10
- + all printk patches from 3.10 to mainline
- + your patches
- Attaching my printk.c, so you can see the current state.

> Viresh, can you verify if we can do cond_resched() from console_unlock()
> (console_may_schedule != 0) ?

I will try that today.

-- 
viresh

View attachment "printk.c" of type "text/x-csrc" (88107 bytes)