linux-kernel - Re: 20% performance drop on PostgreSQL 9.2 from kernel 3.5.3 to 3.6-rc5 on AMD chipsets

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 for Android: free password hash cracker in your pocket

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

Message-ID: <CA+55aFyTrcmx3Uf7UGU=-K=BwTym8aN15fTo==cV2bCK6so0Fg@mail.gmail.com>
Date:	Thu, 27 Sep 2012 11:19:46 -0700
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Borislav Petkov <bp@...en8.de>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Ingo Molnar <mingo@...nel.org>, Mike Galbraith <efault@....de>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Mel Gorman <mgorman@...e.de>,
	Nikolay Ulyanitsky <lystor@...il.com>,
	linux-kernel@...r.kernel.org,
	Andreas Herrmann <andreas.herrmann3@....com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Suresh Siddha <suresh.b.siddha@...el.com>
Subject: Re: 20% performance drop on PostgreSQL 9.2 from kernel 3.5.3 to
 3.6-rc5 on AMD chipsets - bisected

On Thu, Sep 27, 2012 at 11:05 AM, Borislav Petkov <bp@...en8.de> wrote:
> On Thu, Sep 27, 2012 at 10:44:26AM -0700, Linus Torvalds wrote:
>> Or could we just improve the heuristics. What happens if the
>> scheduling granularity is increased, for example? It's set to 1ms
>> right now, with a logarithmic scaling by number of cpus.
>
> /proc/sys/kernel/sched_wakeup_granularity_ns=10000000 (10ms)
> ------------------------------------------------------
> tps = 4994.730809 (including connections establishing)
> tps = 5000.260764 (excluding connections establishing)
>
> A bit better over the default NO_WAKEUP_PREEMPTION setting.

Ok, so this gives us something possible to actually play with.

For example, maybe SCHED_TUNABLESCALING_LINEAR is more appropriate
than SCHED_TUNABLESCALING_LOG. At least for WAKEUP_PREEMPTION. Hmm?

(Btw, "linear" right now looks like 1:1. That's linear, but it's a
very aggressive linearity. Something like "factor = (cpus+1)/2" would
also be linear, but by a less extreme factor.

              Linus
--
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/