From mboxrd@z Thu Jan 1 00:00:00 1970 From: axboe@fb.com (Jens Axboe) Date: Tue, 12 Jan 2016 13:33:56 -0700 Subject: [PATCHv3 4/5] NVMe: IO queue deletion re-write In-Reply-To: <20160106154414.GD15113@localhost.localdomain> References: <1451923859-14187-1-git-send-email-keith.busch@intel.com> <1451923859-14187-5-git-send-email-keith.busch@intel.com> <20160106130419.GA23245@infradead.org> <20160106154414.GD15113@localhost.localdomain> Message-ID: <56956334.90908@fb.com> On 01/06/2016 08:44 AM, Keith Busch wrote: > On Wed, Jan 06, 2016@05:04:19AM -0800, Christoph Hellwig wrote: >> This is my counter-suggestion to the requeue abuse: >> >> simply do two passes of your algorithm, one for the SQs, one for the >> CQs. Patch relative to the whole series below: > > Yes, thanks, this looks great! > > Just a nitpick, I think it'd look better semantically to have > nvme_disable_io_queues handle switching the opcodes instead of having > to call it twice, but this is fine as-is. > > Acked-by: Keith Busch Keith, are you going to re-post 4-5/5? -- Jens Axboe