From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [PATCH 05/14] libata: make ->data_xfer return the number of consumed bytes Date: Thu, 29 Nov 2007 14:37:03 -0500 Message-ID: <474F14DF.8050201@garzik.org> References: <1196346817387-git-send-email-htejun@gmail.com> <11963468181417-git-send-email-htejun@gmail.com> <20071129155546.694c6c45@the-village.bc.nu> <474EE394.6020604@gmail.com> <20071129174259.6a1edc88@the-village.bc.nu> <474EFE58.701@gmail.com> <20071129182409.2fcd66d6@the-village.bc.nu> <474F0B7F.8090209@rtr.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from srv5.dvmed.net ([207.36.208.214]:55640 "EHLO mail.dvmed.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759717AbXK2ThG (ORCPT ); Thu, 29 Nov 2007 14:37:06 -0500 In-Reply-To: <474F0B7F.8090209@rtr.ca> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Mark Lord Cc: Alan Cox , Tejun Heo , linux-ide@vger.kernel.org, albertl@mail.com, jens.axboe@oracle.com Mark Lord wrote: > Alan Cox wrote: >>> DMA alignment is host restriction so I think it belongs to ata_host if >>> we ever need it. Do you know of any controller which require such >>> thing? No need to add complexity when it's not necessary. >> >> If we ever get the blasted inic162x working then that appears to have >> some alignment limits. At least the docs say the DMA buffers must be quad >> word aligned and sized (although it doesn't describe what occurs if the >> total length of xfer disagrees with the buffers) > ... > > If it's an ADMA device, then they may support bits in the CPB > to direct what should happen for various overrun/underrun conditions. Yeah most devices that are not strictly SFF (i.e. have a real register space) are like that. Jeff