From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755947Ab0IWXUz (ORCPT ); Thu, 23 Sep 2010 19:20:55 -0400 Received: from earthlight.etchedpixels.co.uk ([81.2.110.250]:54705 "EHLO www.etchedpixels.co.uk" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753413Ab0IWXUy (ORCPT ); Thu, 23 Sep 2010 19:20:54 -0400 Date: Fri, 24 Sep 2010 00:43:43 +0100 From: Alan Cox To: Chuck Ebbert Cc: linux-ide@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: pata_it821x driver reports many FLUSH CACHE command failures Message-ID: <20100924004343.0abebfad@lxorguk.ukuu.org.uk> In-Reply-To: <20100923172724.0da07fb1@katamari> References: <20100923172724.0da07fb1@katamari> X-Mailer: Claws Mail 3.7.6 (GTK+ 2.18.9; x86_64-redhat-linux-gnu) Face: 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 Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 23 Sep 2010 17:27:24 -0400 Chuck Ebbert wrote: > It looks like pata_it821x has never supported FLUSH_CACHE, but now That depends on the firmware you have I believe The ident it provides also doesn't advertise flush cache so something appears to be issuing a cache flush despite the drive reporting that it is not supported. > we're seeing repeated failures. They happen over and over again and > prevent installation. I stuck a dump_stack() in the failure path and > got this: Yes you'll crash some firmware versions if this occurs. See: static unsigned int it821x_read_id(struct ata_device *adev, struct ata_taskfile *tf, u16 *id) If this has stopped working then you probably want to get Jeff to trace back into the ATA and SCSI midlayer to see why it is issuing flushes for a device which is reporting that cache flush is not supported - because that is seriously broken.