From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [RFC] hpsa: work in progress "lockless monster" patches Date: Sat, 26 Jul 2014 08:32:32 -0700 Message-ID: <20140726153232.GE1854@infradead.org> References: <20140725192859.GU14599@beardog.cce.hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from bombadil.infradead.org ([198.137.202.9]:43390 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751317AbaGZPce (ORCPT ); Sat, 26 Jul 2014 11:32:34 -0400 Content-Disposition: inline In-Reply-To: <20140725192859.GU14599@beardog.cce.hp.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: scameron@beardog.cce.hp.com Cc: linux-scsi@vger.kernel.org, jbottomley@parallels.com, martin.petersen@oracle.com, thenzl@redhat.com, hare@suse.de, hch@infradead.org, bvanassche@acm.org, axboe@kernel.dk, elliott@hp.com, scott.teel@hp.com, webbnh@hp.com, joseph.t.handzik@hp.com, justin.lindley@hp.com > Christoph Hellwig (1): > reserve block tags in scsi host So you found this one useful. This begs the question how we should move forward with it, as it will only work with the blk-mq path in it's current form. I can see three ways: - we implement equivalent functionality in the old block tagging code and make it available for any driver - we require drivers specific workarounds for the !mq path - we allow drivers to force the use of the blk-mq path if they want to use advanced features like this. This might become nessecary anyway when we expose actual multiqueue support to LLDDs.