linux-kernel - Re: [patch] make wake-affine a sysctl variable

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] [day] [month] [year] [list]

Message-Id: <1228467117.18899.30.camel@twins>
Date:	Fri, 05 Dec 2008 09:51:57 +0100
From:	Peter Zijlstra <a.p.zijlstra@...llo.nl>
To:	Ken Chen <kenchen@...gle.com>
Cc:	Ingo Molnar <mingo@...e.hu>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [patch] make wake-affine a sysctl variable

On Fri, 2008-12-05 at 00:44 -0800, Ken Chen wrote:
> On Thu, Dec 4, 2008 at 11:49 PM, Peter Zijlstra wrote:
> > This patch looks like utter rubbish, sorry.
> 
> Yeah, I think turning off wake-affine completely is a lot cleaner and
> is a preferred solution.  I proposed sysctl variable earlier so that
> it can give people a knob to adjust and is more flexible.

Hehe, right, I guess I deserved that.

What I meant was, sysctl_sched_features is a horrid user interface. If
you want to expose it as a sysctl - which I still don't think we want -
create a new one that goes on/off, like sysctl_sched_wake_affine.

Anyway, is this thing you're seeing the same as pgbench, if not, care to
cook a benchmark? I really think the 1:N waker issue is solvable (not
saying its easy though).

--
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/