From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sunset.davemloft.net (dsl027-180-168.sfo1.dsl.speakeasy.net [216.27.180.168]) by ozlabs.org (Postfix) with ESMTP id 85D3967BBC for ; Thu, 24 Aug 2006 08:03:13 +1000 (EST) Date: Wed, 23 Aug 2006 15:03:14 -0700 (PDT) Message-Id: <20060823.150314.41637066.davem@davemloft.net> To: linas@austin.ibm.com Subject: Re: [PATCH 2/6]: powerpc/cell spidernet low watermark patch. From: David Miller In-Reply-To: <20060823213642.GG4401@austin.ibm.com> References: <1156055509.5803.77.camel@localhost.localdomain> <200608201203.15645.arnd@arndb.de> <20060823213642.GG4401@austin.ibm.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Cc: arnd@arndb.de, jklewis@us.ibm.com, linux-kernel@vger.kernel.org, linuxppc-dev@ozlabs.org, netdev@vger.kernel.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , From: linas@austin.ibm.com (Linas Vepstas) Date: Wed, 23 Aug 2006 16:36:42 -0500 > I could create a searate patch to change struct descr {} to split > the u32 into several u8's; there's a dozen spots that get touched. > > Alternatel, I could do a cheesy cast to char[4] and access that way. > Opinions? The most portable scheme would be a "u32/u8[4]" union with appropriate endianness checks when determining which byte to access in the u8[] view.