From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benjamin Herrenschmidt Subject: Re: libata and DMA quirks Date: Fri, 27 Jun 2008 18:51:43 +1000 Message-ID: <1214556703.8011.531.camel@pasglop> References: <1214548895.8011.526.camel@pasglop> <4864A6C9.2050709@ru.mvista.com> Reply-To: benh@kernel.crashing.org Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from gate.crashing.org ([63.228.1.57]:40010 "EHLO gate.crashing.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750704AbYF0JCW (ORCPT ); Fri, 27 Jun 2008 05:02:22 -0400 In-Reply-To: <4864A6C9.2050709@ru.mvista.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Sergei Shtylyov Cc: linux-ide@vger.kernel.org, Jeff Garzik , Tejun Heo On Fri, 2008-06-27 at 12:37 +0400, Sergei Shtylyov wrote: > Hello. >=20 > Benjamin Herrenschmidt wrote: > > The main one is that one their latest cell, they do the following > > "workarounds" which I never implemented in drivers/ide/ppc/pmac.c b= ut > > I'd like to implement in the libata driver, unless you believe that= is > > unnecessary: > > > > a) =EF=BB=BFFor any ATAPI DMA, If the transfer size is not a multip= le of 16 > > bytes, switch to PIO for this command. > > =20 >=20 > No need to do this in old driver anyway since the *IDE* core only=20 > uses DMA for block commands like READ and WRITE. > Not sure about libata. It doesn't use DMA to transfer the bulk of ATAPI ? I know it uses PIO for the CDB but the rest of the transfer isn't done with DMA ? Ben.