From mboxrd@z Thu Jan  1 00:00:00 1970
From: Chris Wilson <chris@chris-wilson.co.uk>
Subject: Re: [PATCH 1/2] drm/i915: Prefault all pages for pread
	and pwrite
Date: Sun, 10 Jul 2011 09:08:50 +0100
Message-ID: <aefc95$okvdr@orsmga001.jf.intel.com>
References: <1310200731-18086-1-git-send-email-chris@chris-wilson.co.uk>
	<20110710014031.GA8410@cloud01>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org>
Received: from mga02.intel.com (mga02.intel.com [134.134.136.20])
	by gabe.freedesktop.org (Postfix) with ESMTP id A5E859E7E1
	for <intel-gfx@lists.freedesktop.org>;
	Sun, 10 Jul 2011 06:33:23 -0700 (PDT)
In-Reply-To: <20110710014031.GA8410@cloud01>
List-Unsubscribe: <http://lists.freedesktop.org/mailman/options/intel-gfx>,
	<mailto:intel-gfx-request@lists.freedesktop.org?subject=unsubscribe>
List-Archive: <http://lists.freedesktop.org/archives/intel-gfx>
List-Post: <mailto:intel-gfx@lists.freedesktop.org>
List-Help: <mailto:intel-gfx-request@lists.freedesktop.org?subject=help>
List-Subscribe: <http://lists.freedesktop.org/mailman/listinfo/intel-gfx>,
	<mailto:intel-gfx-request@lists.freedesktop.org?subject=subscribe>
Sender: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org
Errors-To: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org
To: Ben Widawsky <ben@bwidawsk.net>
Cc: intel-gfx@lists.freedesktop.org
List-Id: intel-gfx@lists.freedesktop.org

On Sun, 10 Jul 2011 01:40:31 +0000, Ben Widawsky <ben@bwidawsk.net> wrote:
> On Sat, Jul 09, 2011 at 09:38:50AM +0100, Chris Wilson wrote:
> > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> > index 4fc9738..2fce620 100644
> > --- a/drivers/gpu/drm/i915/i915_gem.c
> > +++ b/drivers/gpu/drm/i915/i915_gem.c
> > @@ -503,6 +503,19 @@ out:
> >  	return ret;
> >  }
> >  
> > +static int prefault_writeable(unsigned long uaddr, unsigned long len)
> > +{
> > +	int ret = 0;
> > +
> > +	len += uaddr;
> > +	while (uaddr < len) {
> > +		ret |= __put_user(0, (char __user *)uaddr);
> > +		uaddr += 4096;
> > +	}
> > +
> > +	return ret ? -EFAULT : 0;
> > +}
> > +
> >  /**
> >   * Reads data from the object referenced by handle.
> >   *
> 
> What's the reason for not breaking out of the loop on the first failing
> fault? It seems like you could incur a bunch of latency for a call which
> you know will ending up failin anyway. Although TBH I'm not clear how
> it could actually fail if you've called access_ok().

Because encountering a fault in that loop is an extremely rare event, and
the branch inside the loop appeared on the profiles.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre