From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthew Wilcox Subject: Re: [PATCH 1/11] mpt2sas: mpt2sas_base sources Date: Tue, 3 Mar 2009 20:54:47 -0700 Message-ID: <20090304035447.GA1884@parisc-linux.org> References: <660360F4F2570145BD872F298951B17A7689E415@cosmail03.lsi.com> <49ADF402.4050504@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from palinux.external.hp.com ([192.25.206.14]:44192 "EHLO mail.parisc-linux.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751618AbZCDDzG (ORCPT ); Tue, 3 Mar 2009 22:55:06 -0500 Content-Disposition: inline In-Reply-To: <49ADF402.4050504@redhat.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Rob Evers Cc: "Moore, Eric" , James Bottomley , Grant Grundler , "linux-scsi@vger.kernel.org" On Tue, Mar 03, 2009 at 10:22:42PM -0500, Rob Evers wrote: > >(1) Add readl following writel from the function that disables interrupts > > Is a memory barrier required after the writel and before the readl to > ensure that the writel executes before the readl on processors other than > x86_64 and ia32? No. Architectures such as PowerPC and Altix which have the potential to issue writes out of order wrt io memory are required to implement barriers in their writel() implementation. -- Matthew Wilcox Intel Open Source Technology Centre "Bill, look, we understand that you're interested in selling us this operating system, but compare it to ours. We can't possibly take such a retrograde step."