Date: Fri, 4 Jul 2014 18:21:28 +1000
From: Dave Chinner <david@fromorbit.com>
To: Jason Low <jason.low2@hp.com>
Cc: Davidlohr Bueso <davidlohr@hp.com>, Peter Zijlstra <peterz@infradead.org>,
        Tim Chen <tim.c.chen@linux.intel.com>, Ingo Molnar <mingo@kernel.org>,
        linux-kernel@vger.kernel.org,
        Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: [regression, 3.16-rc] rwsem: optimistic spinning causing
 performance degradation
Message-ID: <20140704082128.GL9508@dastard>
References: <1404413420.8764.42.camel@j-VirtualBox>
 <20140704010147.GY4453@dastard>
 <1404438366.8764.121.camel@j-VirtualBox>
 <1404438890.8764.125.camel@j-VirtualBox>
 <20140704061306.GK9508@dastard>
 <1404457579.8764.161.camel@j-VirtualBox>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1404457579.8764.161.camel@j-VirtualBox>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org

On Fri, Jul 04, 2014 at 12:06:19AM -0700, Jason Low wrote:
> On Fri, 2014-07-04 at 16:13 +1000, Dave Chinner wrote:
> > On Thu, Jul 03, 2014 at 06:54:50PM -0700, Jason Low wrote:
> > > On Thu, 2014-07-03 at 18:46 -0700, Jason Low wrote:
> > > > On Fri, 2014-07-04 at 11:01 +1000, Dave Chinner wrote:
> > > 
> > > > > FWIW, the rwsems in the struct xfs_inode are often heavily
> > > > > read/write contended, so there are lots of IO related workloads that
> > > > > are going to regress on XFS without this optimisation...
> > > > > 
> > > > > Anyway, consider the patch:
> > > > > 
> > > > > Tested-by: Dave Chinner <dchinner@redhat.com>
> > > > 
> > > > Hi Dave,
> > > > 
> > > > Thanks for testing. I'll update the patch with an actual changelog.
> > > 
> > > ---
> > > Subject: [PATCH] rwsem: In rwsem_can_spin_on_owner(), return false if no owner
> > > 
> > > It was found that the rwsem optimistic spinning feature can potentially degrade
> > > performance when there are readers. Perf profiles indicate in some workloads
> > > that significant time can be spent spinning on !owner. This is because we don't
> > > set the lock owner when readers(s) obtain the rwsem.
> > 
> > I don't think you're being a little shifty with the truth here.
> > There's no "potentially degrade performance" here - I reported a
> > massive real world performance regression caused by optimistic
> > spinning.
> 
> Sure, though I mainly used the word "potentially" since there can be
> other workloads out there where spinning even when readers have the lock
> is a positive thing.
> 
> And agreed that the changelog can be modified to try reflecting more on
> it being a "regression fix" then a "new performance" addition.
> 
> So how about the following?

Looks good. Thanks!

-Dave.
-- 
Dave Chinner
david@fromorbit.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/