From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: [PATCH] aic7xxx: upport all sequencer and core fixes from adaptec version 6.3.9 Date: Fri, 05 Aug 2005 13:16:12 -0500 Message-ID: <1123265773.5003.50.camel@mulgrave> References: <1123194802.5026.75.camel@mulgrave> <42F368DD.504@vger.kernel.org> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from stat16.steeleye.com ([209.192.50.48]:15312 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S263036AbVHESQR (ORCPT ); Fri, 5 Aug 2005 14:16:17 -0400 Received: from midgard.sc.steeleye.com (midgard.sc.steeleye.com [172.17.6.40]) by hancock.sc.steeleye.com (8.11.6/8.11.6) with ESMTP id j75IGEA00418 for ; Fri, 5 Aug 2005 14:16:14 -0400 In-Reply-To: <42F368DD.504@vger.kernel.org> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: SCSI Mailing List On Fri, 2005-08-05 at 09:25 -0400, SCSI Mailing List wrote: > Tremendously good work! Thanks. There's unfortunately still an outstanding problem I can't fix. It seems that after a certain number of repetitions of a failing command, the entire card locks up and the sequencer always returns SAM_STAT_BUSY to any new commands. This is a bit of a problem for the scsi mid-layer because it blindly always retries SAM_STAT_BUSY and SAM_STAT_QUEUE_FULL. The only way to recover from this is to issue a bus reset. This condition, I think, is what causes some of the hang reports about this driver. However, I think I see the code in the sequencer which is causing it (target_cmdphase) but I don't have enough information about the sequencer to fix this. James