From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ingo Molnar <mingo@elte.hu>
Subject: Re: [GIT PULL] adaptive spinning mutexes
Date: Wed, 14 Jan 2009 21:14:35 +0100
Message-ID: <20090114201435.GA6519@elte.hu>
References: <alpine.LFD.2.00.0901130812590.6528@localhost.localdomain> <1231863710.7141.3.camel@twins> <1231864854.7141.8.camel@twins> <alpine.LFD.2.00.0901130846320.6528@localhost.localdomain> <1231867314.7141.16.camel@twins> <1231952436.14825.28.camel@laptop> <20090114183319.GA18630@elte.hu> <20090114105300.66bd014d.akpm@linux-foundation.org> <20090114190008.GA13203@elte.hu> <20090114113638.c818fcf8.akpm@linux-foundation.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: torvalds@linux-foundation.org, a.p.zijlstra@chello.nl,
	paulmck@linux.vnet.ibm.com, ghaskins@novell.com, matthew@wil.cx,
	andi@firstfloor.org, chris.mason@oracle.com,
	linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	linux-btrfs@vger.kernel.org, tglx@linutronix.de, npiggin@suse.de,
	pmorreale@novell.com, SDietrich@novell.com,
	dmitry.adamushko@gmail.com, hannes@cmpxchg.org
To: Andrew Morton <akpm@linux-foundation.org>
Return-path: <linux-fsdevel-owner@vger.kernel.org>
In-Reply-To: <20090114113638.c818fcf8.akpm@linux-foundation.org>
List-ID: <linux-btrfs.vger.kernel.org>


* Andrew Morton <akpm@linux-foundation.org> wrote:

> On Wed, 14 Jan 2009 20:00:08 +0100
> Ingo Molnar <mingo@elte.hu> wrote:
> 
> > 
> > * Andrew Morton <akpm@linux-foundation.org> wrote:
> > 
> > > On Wed, 14 Jan 2009 19:33:19 +0100 Ingo Molnar <mingo@elte.hu> wrote:
> > > 
> > > > Please pull the adaptive-mutexes-for-linus git tree
> > > 
> > > <fear>
> > > 
> > > - It seems a major shortcoming that the feature is disabled if
> > >   CONFIG_DEBUG_MUTEXES=y.  It means that lots of people won't test it.
> 
> ^^^?
>  
> > > - When people hit performance/latency oddities, it would be nice if
> > >   they had a /proc knob with which they could disable this feature at
> > >   runtime.
> > > 
> > >   This would also be useful for comparative performance testing.
> > 
> > Yeah. From my other mail:
> > 
> > > > We still have the /sys/debug/sched_features tunable under 
> > > > CONFIG_SCHED_DEBUG=y, so should this cause any performance regressions 
> > > > somewhere, it can be pinned down and blamed back on this change 
> > > > easily, without bisection and without rebooting the box.
> > 
> > This kind of easy knob was included early on - this is how all those spin 
> > versus no-spin numbers were done.
> 
> Do people enable CONFIG_SCHED_DEBUG?

If they suspect performance problems and want to analyze them?

Note that CONFIG_SCHED_DEBUG=y is also the default.

> CONFIG_DEBUG_MUTEXES=n, CONFIG_SCHED_DEBUG=y is getting to be a pretty 
> small subset?

Those two are the default config settings actually, so i'd expect it to be 
the most commonly occuring combinations.

	Ingo