linux-kernel - Re: 2.6.32 / RH 6.1 scheduler bloated

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: <8AF29618-6746-484D-B773-AED31FBA714D@gmail.com>
Date:	Sun, 9 Oct 2011 10:26:22 -0400
From:	Robert Kubrick <robertkubrick@...il.com>
To:	Mike Galbraith <efault@....de>
Cc:	linux-kernel@...r.kernel.org
Subject: Re: 2.6.32 / RH 6.1 scheduler bloated

Will try this, but I have older kernel version running on the same  
hardware, with the same boot parameters, and showing a more  
reasonable 2us wakeup latency. Still high, but coming out of idle can  
be slow as you say.
But 10us??

On Oct 9, 2011, at 2:03 AM, Mike Galbraith wrote:

> On Sat, 2011-10-08 at 20:26 -0400, Robert Kubrick wrote:
>> A simple ftrace wakeup test shows 10us latency during the idle-to-run
>> transition:
>
> If you boot nohz=off idle=poll, and measure with cyclictest  
> (measure in
> userspace for free), you'll likely see a remarkable difference.
>
> Coming out of idle can be slow, but that has more to do with silicon
> than scheduler bloat (not that bloat doesn't exist mind you).  My  
> Q6600
> box _sucks_ at getting off it's duff, even intel_idle.max_cstate=1 is
> horrid. E5620 gets moving > order of magnitude more quickly with the
> same kernel/config/params.
>
> 	-Mike
>

--
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/