From mboxrd@z Thu Jan 1 00:00:00 1970 From: Phil Sutter Subject: Re: [PATCH] fix pata-rb532-cf Date: Sat, 1 Nov 2008 17:45:45 +0100 Message-ID: <20081101164545.GB10321@nuty> References: <20081101160930.GA10321@nuty> <1225555965-27557-1-git-send-email-n0-1@freewrt.org> <490C833A.7010702@ru.mvista.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from orbit.nwl.cc ([81.169.176.177]:32784 "EHLO mail.ifyouseekate.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751769AbYKAQof (ORCPT ); Sat, 1 Nov 2008 12:44:35 -0400 Received: from nuty (localhost [127.0.0.1]) by mail.ifyouseekate.net (Postfix) with ESMTP id C8944386DBBE for ; Sat, 1 Nov 2008 17:44:33 +0100 (CET) Content-Disposition: inline In-Reply-To: <490C833A.7010702@ru.mvista.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: linux-ide@vger.kernel.org Hi, On Sat, Nov 01, 2008 at 07:26:34PM +0300, Sergei Shtylyov wrote: > So, I didn't get what was wrong with using readsl() and writesl()? > Besides, using readl() and witel() this way would be wrong on BE mode > since the data is expected to be stored to memory in the LE order. Sorry, my fault. When having a glance at the macro definition, I messed up the parameter's meanings and assumed the target address would be incremented, which is normally the case when writing regular memory, not to IO addresses. I changed (and successfully tested) the code, patch follows. Thanks again, Phil