From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bokhan Artem Subject: Re: [PATCH #upstream-fixes] sata_mv: don't issue two DMA commands concurrently Date: Thu, 14 Aug 2008 00:37:32 +0700 Message-ID: <48A31BDC.8070203@ngs.ru> References: <489C19CE.6030708@ngs.ru> <489C4B6E.9070306@rtr.ca> <489C4F29.6020007@rtr.ca> <489C54D1.5080901@rtr.ca> <48A29E14.3090908@gmail.com> <48A2BBDC.2090201@ngs.ru> <48A2BC5D.5060801@gmail.com> <48A2C35C.7020400@ngs.ru> <48A2C460.6040504@gmail.com> <48A2C770.2040200@ngs.ru> <48A2CB1B.4000200@gmail.com> <48A2CE09.9090806@ngs.ru> <48A30920.5030803@rtr.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from smtpout1.ngs.ru ([195.93.186.195]:45999 "EHLO smtpout1.ngs.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751087AbYHMRhf (ORCPT ); Wed, 13 Aug 2008 13:37:35 -0400 In-Reply-To: <48A30920.5030803@rtr.ca> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Mark Lord , Tejun Heo , linux-ide@vger.kernel.org It's very strange for me that it is not runtime configurable... OK, thn= x. Also I'm not very close to ata protocols, so : 1. Why there is need to reset link if sector can not be read? 2. Why device is locked until hard reset? Aug 13 08:45:45 ubuntu kernel: [ 5634.080692] sd 4:0:0:0: [sde]=20 1465149168 512-byte hardware sectors (750156 MB) Aug 13 08:45:46 ubuntu kernel: [ 5635.088067] res=20 40/00:08:e0:a7:36/01:00:01:00:00/00 Emask 0x4 (timeout) Aug 13 08:45:46 ubuntu kernel: [ 5635.088136] ata5: hard resetting link Mark Lord =D0=BF=D0=B8=D1=88=D0=B5=D1=82: > Artem Bokhan wrote: >> When I run 'dd if=3D/dev/sda of=3D/dev/null skip=3D9700 bs=3D1M' and= it meets=20 >> bad block, it takes about=20 >> '/sys/devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:02.0/host= 4/target4:0:0/4:0:0:0/timeout=20 >> + about 10 seconds' to determine I/O error. Until that IO to device=20 >> is locked, as I understand. I want to reduce the time of lock as =20 >> much as possible. > .. > > Tell the SCSI layer to stop flogging the drive with unnecessary retri= es. > A quick and dirty way, is this patch below, but it really should be m= ore > intelligent than this for some situations: > > --- linux/include/scsi/sd.h.orig 2008-06-09 14:27:19.000000000 -04= 00 > +++ linux/include/scsi/sd.h 2008-08-13 12:16:01.000000000 -0400 > @@ -23,7 +23,7 @@ > /* > * Number of allowed retries > */ > -#define SD_MAX_RETRIES 5 > +#define SD_MAX_RETRIES 1 > #define SD_PASSTHROUGH_RETRIES 1 > > /*