From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH] scsi: break from queue depth adjusting loops when device found Date: Sun, 6 Jul 2014 11:49:37 +0200 Message-ID: <20140706094937.GA29045@lst.de> References: <20140703150557.21608.37072.stgit@beardog.cce.hp.com> <20140703171106.GA16788@lst.de> <53B6878C.502@suse.de> <94D0CD8314A33A4D9D801C0FE68B402958B86659@G9W0745.americas.hpqcorp.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from verein.lst.de ([213.95.11.211]:45790 "EHLO newverein.lst.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750822AbaGFJtm (ORCPT ); Sun, 6 Jul 2014 05:49:42 -0400 Content-Disposition: inline In-Reply-To: <94D0CD8314A33A4D9D801C0FE68B402958B86659@G9W0745.americas.hpqcorp.net> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: "Elliott, Robert (Server Storage)" Cc: Hannes Reinecke , Christoph Hellwig , "Stephen M. Cameron" , "james.bottomley@parallels.com" , "stephenmcameron@gmail.com" , "linux-scsi@vger.kernel.org" , Vasu Dev , Mike Christie On Fri, Jul 04, 2014 at 07:21:52PM +0000, Elliott, Robert (Server Storage) wrote: > Perhaps the SCSI midlayer should keep track of both > SCSI target port and logical unit queue depths, parse > the status qualifier if present, and let the host > template advise on the policy to assume if the > status qualifier is not present. We already keep track of the target queue depth, although with the scsi-mq series this is not optional and only done for hosts that declare they have a per-target queue limit by setting ->can_queue in the scsi_target structurue. Only a small number of iSCSI offload HBAs does this. I need to look at the SAM scope defintions, but it doesn't sound too hard to implement. > There is some code for a scsi_target structure that I > don't understand and have just been ignoring: target_busy, > target_blocked, etc. Does that represent the SCSI target > port over multiple logical units, or does that relate to > target-mode where the system is acting as the SCSI target > and presenting logical units itself? It represents the target port, and is not related to target mode support at all. The target mode code uses completely different data structures than the initiator side mid layer.