From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner+w=401wt.eu-S933239AbXGMNeu@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S933239AbXGMNeu (ORCPT <rfc822;w@1wt.eu>);
	Fri, 13 Jul 2007 09:34:50 -0400
Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932837AbXGMNed
	(ORCPT <rfc822;linux-kernel-outgoing>);
	Fri, 13 Jul 2007 09:34:33 -0400
Received: from hancock.steeleye.com ([71.30.118.248]:38411 "EHLO
	hancock.sc.steeleye.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org
	with ESMTP id S932781AbXGMNeb (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Fri, 13 Jul 2007 09:34:31 -0400
Subject: Re: [patch 5/6] ps3: BD/DVD/CD-ROM Storage Driver
From: James Bottomley <James.Bottomley@SteelEye.com>
To: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Cc: Jens Axboe <jens.axboe@oracle.com>, Paul Mackerras <paulus@samba.org>,
       Alessandro Rubini <rubini@vision.unipv.it>, linuxppc-dev@ozlabs.org,
       linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org,
       Geoff Levand <geoffrey.levand@am.sony.com>
In-Reply-To: <Pine.LNX.4.62.0707131518120.18802@pademelon.sonytel.be>
References: <20070704132212.726923000@pademelon.sonytel.be>
	 <20070704132346.591348000@pademelon.sonytel.be>
	 <1184331741.3402.13.camel@localhost.localdomain>
	 <20070713130508.GQ5328@kernel.dk>
	 <Pine.LNX.4.62.0707131518120.18802@pademelon.sonytel.be>
Content-Type: text/plain
Date: Fri, 13 Jul 2007 09:34:30 -0400
Message-Id: <1184333670.3402.17.camel@localhost.localdomain>
Mime-Version: 1.0
X-Mailer: Evolution 2.10.2 (2.10.2-3.fc7) 
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
X-Mailing-List: linux-kernel@vger.kernel.org

On Fri, 2007-07-13 at 15:25 +0200, Geert Uytterhoeven wrote:
> kmap() just returns page_address() on ppc64, as there's no highmem.
> kunmap() is a no-op.

> So technically I could just use page_address() directly, but Christoph
> wanted
> me to keep the kmap()/kunmap() sequence because it's considered a good
> practice.

The point isn't what kmap and kunmap do ... it's the addresses they
return.  By and large, a kernel virtual address for a page is different
from the user virtual address.  If the cache is virtually indexed you
get different cache lines for the same page ... and that sets you up
with aliases you need to resolve.  parisc is the same ... our
kmap/kunmap are nops as well, but our kernel virtual addresses are still
different from the user virtual ones.

James