From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: [example PATCH - not for applying] exclude certain commands Date: 26 Apr 2003 23:32:58 -0500 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <1051418077.4089.175.camel@mulgrave> References: Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from nat9.steeleye.com ([65.114.3.137]:64007 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S263354AbTD0EWg (ORCPT ); Sun, 27 Apr 2003 00:22:36 -0400 In-Reply-To: List-Id: linux-scsi@vger.kernel.org To: Andries.Brouwer@cwi.nl Cc: greg@kroah.com, SCSI Mailing List , linux-usb-devel@lists.sourceforge.net, mdharm-scsi@one-eyed-alien.net On Sat, 2003-04-26 at 21:29, Andries.Brouwer@cwi.nl wrote: > All docs say that this routine should call scsi_adjust_queue_depth. > Is that really true? When reading the docs one gets the impression > that our code is > if (slave_configure) > slave_configure(); > else > scsi_adjust_queue_depth(default); > But I read in scsi_scan.c: > if(sdev->host->hostt->slave_configure) > sdev->host->hostt->slave_configure(sdev); > Well, its design purpose was for queue depth adjustment, but it's ideally placed to modify any parameters about the device we choose, hence it could set the appropriate mode sense flag for USB storage. The rework of the SCSI code means that you don't get tag command queueing without calling adjust_queue_depth from the slave_configure routine. The default is not to do it. However, the slab allocation of commands rather took the teeth out of it for queue depth adjustment. James