From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Subject: Re: [PATCH 9/9] nvme: add support for streams and directives To: Christoph Hellwig Cc: linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, adilger@dilger.ca, martin.petersen@oracle.com, linux-nvme@lists.infradead.org References: <1498004526-4543-1-git-send-email-axboe@kernel.dk> <1498004526-4543-10-git-send-email-axboe@kernel.dk> <20170626095907.GG14057@infradead.org> <20170627141115.GA2541@infradead.org> From: Jens Axboe Message-ID: Date: Tue, 27 Jun 2017 08:16:49 -0600 MIME-Version: 1.0 In-Reply-To: <20170627141115.GA2541@infradead.org> Content-Type: text/plain; charset=utf-8 List-ID: On 06/27/2017 08:11 AM, Christoph Hellwig wrote: > On Mon, Jun 26, 2017 at 07:56:22AM -0600, Jens Axboe wrote: >>> - do we even need the < 4 streams fallback now that they are global >>> instead of per-ns instead of just disabling the feature for now? >> >> Maybe the device only supports 2? or 3? > > My crystal ball indicates that those are unlikely too see the > light. IFF we need to handle them we can still add code for it. But we have to handle it, not doing so would be fragile. So our options are: 1) Keep the stream_mappings[] array. It's simple, and it'll work for any number of streams. 2) Kill stream_mappings[] and just do the MOD again. I'd strongly lean towards #1. I don't have a lot of faith in crystal balls. -- Jens Axboe From mboxrd@z Thu Jan 1 00:00:00 1970 From: axboe@kernel.dk (Jens Axboe) Date: Tue, 27 Jun 2017 08:16:49 -0600 Subject: [PATCH 9/9] nvme: add support for streams and directives In-Reply-To: <20170627141115.GA2541@infradead.org> References: <1498004526-4543-1-git-send-email-axboe@kernel.dk> <1498004526-4543-10-git-send-email-axboe@kernel.dk> <20170626095907.GG14057@infradead.org> <20170627141115.GA2541@infradead.org> Message-ID: On 06/27/2017 08:11 AM, Christoph Hellwig wrote: > On Mon, Jun 26, 2017@07:56:22AM -0600, Jens Axboe wrote: >>> - do we even need the < 4 streams fallback now that they are global >>> instead of per-ns instead of just disabling the feature for now? >> >> Maybe the device only supports 2? or 3? > > My crystal ball indicates that those are unlikely too see the > light. IFF we need to handle them we can still add code for it. But we have to handle it, not doing so would be fragile. So our options are: 1) Keep the stream_mappings[] array. It's simple, and it'll work for any number of streams. 2) Kill stream_mappings[] and just do the MOD again. I'd strongly lean towards #1. I don't have a lot of faith in crystal balls. -- Jens Axboe