From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <B07421@freescale.com>
Received: from VA3EHSOBE001.bigfish.com (va3ehsobe001.messaging.microsoft.com
	[216.32.180.11]) (using TLSv1 with cipher AES128-SHA (128/128 bits))
	(Client CN "mail.global.frontbridge.com",
	Issuer "Cybertrust SureServer Standard Validation CA" (verified OK))
	by ozlabs.org (Postfix) with ESMTPS id 3506CB6F59
	for <linuxppc-dev@lists.ozlabs.org>;
	Thu, 19 May 2011 08:27:33 +1000 (EST)
Date: Wed, 18 May 2011 17:27:20 -0500
From: Scott Wood <scottwood@freescale.com>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Subject: Re: [PATCH 6/7] powerpc/mm: 64-bit: tlb handler micro-optimization
Message-ID: <20110518172720.465e3789@schlenkerla.am.freescale.net>
In-Reply-To: <1305755688.7481.12.camel@pasglop>
References: <20110518210536.GE29524@schlenkerla.am.freescale.net>
	<1305754667.7481.5.camel@pasglop>
	<20110518165134.1023dccc@schlenkerla.am.freescale.net>
	<1305755688.7481.12.camel@pasglop>
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Cc: linuxppc-dev@lists.ozlabs.org
List-Id: Linux on PowerPC Developers Mail List <linuxppc-dev.lists.ozlabs.org>
List-Unsubscribe: <https://lists.ozlabs.org/options/linuxppc-dev>,
	<mailto:linuxppc-dev-request@lists.ozlabs.org?subject=unsubscribe>
List-Archive: <http://lists.ozlabs.org/pipermail/linuxppc-dev>
List-Post: <mailto:linuxppc-dev@lists.ozlabs.org>
List-Help: <mailto:linuxppc-dev-request@lists.ozlabs.org?subject=help>
List-Subscribe: <https://lists.ozlabs.org/listinfo/linuxppc-dev>,
	<mailto:linuxppc-dev-request@lists.ozlabs.org?subject=subscribe>

On Thu, 19 May 2011 07:54:48 +1000
Benjamin Herrenschmidt <benh@kernel.crashing.org> wrote:

> On Wed, 2011-05-18 at 16:51 -0500, Scott Wood wrote:
> > On Thu, 19 May 2011 07:37:47 +1000
> > Benjamin Herrenschmidt <benh@kernel.crashing.org> wrote:
> > 
> > > On Wed, 2011-05-18 at 16:05 -0500, Scott Wood wrote:
> > > > A little more speed up measured on e5500.
> > > > 
> > > > Setting of U0-3 is dropped as it is not used by Linux as far as I can
> > > > see.
> > > 
> > > Please keep them for now. If your core doesn't have them, make them an
> > > MMU feature.
> > 
> > We have them, it was just an attempt to clean out unused things to speed up
> > the miss handler.  I'll drop that part if you think we'll use it in the
> > future.
> 
> I never know for sure ... damn research people ... :-)
> 
> I'd rather keep them for now, does it make a significant difference ?

It was minor but measurable (wouldn't have been worthwhile except as part
of a series of small things that add up), but upon trying again I was able
to reorder slightly and fit it in without seeing an impact.

-Scott