From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: Need help understanding SATA error message. Date: Tue, 01 Apr 2008 09:26:15 +0900 Message-ID: <47F18127.8070509@gmail.com> References: <47ED136F.9010501@rtr.ca> <47ED15D2.3080403@rtr.ca> <47EDBC87.8090109@rtr.ca> <47F0376F.2090000@gmail.com> <47F0FAD3.9060001@rtr.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from rv-out-0910.google.com ([209.85.198.187]:33135 "EHLO rv-out-0910.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751251AbYDAA0V (ORCPT ); Mon, 31 Mar 2008 20:26:21 -0400 Received: by rv-out-0910.google.com with SMTP id k20so1053635rvb.1 for ; Mon, 31 Mar 2008 17:26:21 -0700 (PDT) In-Reply-To: <47F0FAD3.9060001@rtr.ca> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Mark Lord Cc: Tomas Lund , linux-ide@vger.kernel.org, Robert Hancock Mark Lord wrote: >> The response to an unrecoverable sector shouldn't be 51/04 if the >> flush fails, it should be 51/10 or 51/40. >> >> 51/04 would be the response if the FLUSH CACHE command was issued when >> there were still outstanding NCQ commands active. > .. > > Tejun: I see we have another thread as well with FLUSH errors. > I really doubt that these are bad drives. > There's very likely a bug in libata / LLD there someplace. Possibly. The only thing I can think of which can screw FLUSH is issuing it when NCQ phase is still in progress as was in the case for ADMA. FLUSH being a non-data command, it's pretty difficult to get it wrong otherwise. The thing is that sata_sil24 does its own command sequencing and even if libata slips there a bit, the silicon won't issue FLUSH if NCQ is in progress, so I'm a bit skeptical. Any other ideas? -- tejun