From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mike Snitzer Subject: Re: [for-4.16 PATCH v2 2/5] nvme: use request_queue's failover_rq_fn callback for multipath failover Date: Fri, 29 Dec 2017 15:22:00 -0500 Message-ID: <20171229202200.GB8876@redhat.com> References: <20171227032257.8182-1-snitzer@redhat.com> <20171227032257.8182-3-snitzer@redhat.com> <20171229101119.GB25333@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20171229101119.GB25333@lst.de> Sender: linux-block-owner@vger.kernel.org To: Christoph Hellwig Cc: axboe@kernel.dk, keith.busch@intel.com, emilne@redhat.com, james.smart@broadcom.com, hare@suse.de, Bart.VanAssche@wdc.com, linux-block@vger.kernel.org, linux-nvme@lists.infradead.org, dm-devel@redhat.com List-Id: dm-devel.ids On Fri, Dec 29 2017 at 5:11am -0500, Christoph Hellwig wrote: > Independent on any policy question as raised in reply to the previous > patch: until the ANA (ALUA equivalent for NVMe) spec has been finalized > and our code for it has landed there is no way we're going to expose > anything from this code based as it will be highly in flux. Please stop hiding behind specs like they are governing this aspect of Linux's NVMe implementation. Please just own the need for NVMe to provide DM multipathing compatibility. Thanks. From mboxrd@z Thu Jan 1 00:00:00 1970 From: snitzer@redhat.com (Mike Snitzer) Date: Fri, 29 Dec 2017 15:22:00 -0500 Subject: [for-4.16 PATCH v2 2/5] nvme: use request_queue's failover_rq_fn callback for multipath failover In-Reply-To: <20171229101119.GB25333@lst.de> References: <20171227032257.8182-1-snitzer@redhat.com> <20171227032257.8182-3-snitzer@redhat.com> <20171229101119.GB25333@lst.de> Message-ID: <20171229202200.GB8876@redhat.com> On Fri, Dec 29 2017 at 5:11am -0500, Christoph Hellwig wrote: > Independent on any policy question as raised in reply to the previous > patch: until the ANA (ALUA equivalent for NVMe) spec has been finalized > and our code for it has landed there is no way we're going to expose > anything from this code based as it will be highly in flux. Please stop hiding behind specs like they are governing this aspect of Linux's NVMe implementation. Please just own the need for NVMe to provide DM multipathing compatibility. Thanks.