From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263735AbUDGQd5 (ORCPT ); Wed, 7 Apr 2004 12:33:57 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263740AbUDGQd5 (ORCPT ); Wed, 7 Apr 2004 12:33:57 -0400 Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:45761 "EHLO www.linux.org.uk") by vger.kernel.org with ESMTP id S263735AbUDGQdz (ORCPT ); Wed, 7 Apr 2004 12:33:55 -0400 Message-ID: <40742D63.4000803@pobox.com> Date: Wed, 07 Apr 2004 12:33:39 -0400 From: Jeff Garzik User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030703 X-Accept-Language: en-us, en MIME-Version: 1.0 To: "Miller, Mike (OS Dev)" CC: alpm@odsl.org, axboe@suse.de, linux-kernel@vger.kernel.org Subject: Re: cciss updates for 2.6.6xxx [1/2] References: In-Reply-To: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Miller, Mike (OS Dev) wrote: > Yep, you're right. I just regurgitated the same code. I'll pull my head out and try again :( The easiest thing to do may be to take your patch #1, and then add code to clamp the per-queue outstanding-command (tag) depth to 1024 / n_arrays_found at initialization time. Or perhaps s/n_arrays_found/max_arrays_per_hba/ I bet that's just a few additional lines of code, and should work... Regards, Jeff