From mboxrd@z Thu Jan  1 00:00:00 1970
From: "H. Peter Anvin" <hpa@zytor.com>
Subject: Re: [RFC PATCH] kernel: add byteorder macros with alignment fixups
Date: Thu, 20 Mar 2008 13:08:00 -0700
Message-ID: <47E2C420.5050008@zytor.com>
References: <1206034454.17059.4.camel@brick>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: Andrew Morton <akpm@linux-foundation.org>,
	LKML <linux-kernel@vger.kernel.org>,
	linux-netdev <netdev@vger.kernel.org>
To: Harvey Harrison <harvey.harrison@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from terminus.zytor.com ([198.137.202.10]:39296 "EHLO
	terminus.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1755028AbYCTUII (ORCPT
	<rfc822;netdev@vger.kernel.org>); Thu, 20 Mar 2008 16:08:08 -0400
In-Reply-To: <1206034454.17059.4.camel@brick>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

Harvey Harrison wrote:
> A common pattern in the kernel (especially networking) is:
> 
> le32_to_cpu(get_unaligned((__le32 *)x));
> 
> Repeat for various combinations of le/be and 64/32/16 bit.  Add
> a variant that operates on possibly unaligned pointers to
> byteorder/generic.h
> 

This should go in <asm-generic/*> since on some architectures this can 
be done significantly faster than the chained pattern, since you can get 
the reordering for "free".  Thus, architectures should be able to 
override the generic.

	-hpa