On 11/04/2017 08:18 AM, Jens Axboe wrote: > On 11/03/2017 10:17 PM, Ming Lei wrote: >> On Thu, Nov 02, 2017 at 11:24:31PM +0800, Ming Lei wrote: >>> Hi Jens, >>> >>> This patchset avoids to allocate driver tag beforehand for flush rq >>> in case of I/O scheduler, then flush rq isn't treated specially >>> wrt. get/put driver tag, code gets cleanup much, such as, >>> reorder_tags_to_front() is removed, and we needn't to worry >>> about request order in dispatch list for avoiding I/O deadlock. >>> >>> 'dbench -t 30 -s 64' has been run on different devices(shared tag, >>> multi-queue, singele queue, ...), and no issues are observed, >>> even very low queue depth test are run, debench still works well. >>> >>> Please consider it for V4.15, thanks! >> >> Hi Jens, >> >> As we discussed before, this patch is a good cleanup on handling flush >> request, could you share your opinion on V3? > > It looks fine to me. But I'd really like to have the potential hang > in the current 4.15 branch ironed out, before we pile more stuff on > top. Meanwhile, I'll see if this passes my testing. Passes testing. I applied it, but a heads-up that I rewrote most of your changelogs and also changed formatting. Please at least follow the 72 char line rule for commit messages, otherwise git log ends up looking like a mess really quick. -- Jens Axboe
