From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: [PATCH] tmscsim: scsi device attributes Date: 22 Oct 2004 09:00:11 -0400 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <1098450017.2017.19.camel@mulgrave> References: <20041022124007.GC4201@parcelfarce.linux.theplanet.co.uk> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from stat16.steeleye.com ([209.192.50.48]:29091 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S271304AbUJVNAZ (ORCPT ); Fri, 22 Oct 2004 09:00:25 -0400 In-Reply-To: <20041022124007.GC4201@parcelfarce.linux.theplanet.co.uk> List-Id: linux-scsi@vger.kernel.org To: Matthew Wilcox Cc: Guennadi Liakhovetski , SCSI Mailing List On Fri, 2004-10-22 at 08:40, Matthew Wilcox wrote: > How about something more along the following lines? I haven't converted > sym2 to the generic queue code yet, so it's untested (and is obviously > broken because nothing initialises max_queue_depth). Yes, I was considering something like this. The original queue code in 53c700 was designed as an example of how to do attribute overrides. However, there needs to be a callback into the LLD to tell it that the queue is being overriden (and allow a veto for depth increases). I'll look at adding the extra pieces. James