Ming Lei - 02.06.18, 06:54: > On Fri, Jun 01, 2018 at 10:31:13PM -0600, Jens Axboe wrote: > > On 6/1/18 8:36 PM, Ming Lei wrote: > > >> clear the bit when we need to, potentially even adding a debug > > >> check to blk_queue_enter_live() that complains if the ref was > > >> not already elevated. Though that would be expensive, compared > > >> to the percpu inc now. Not saying the bit is necessarily the > > >> best way forward, but I do like it a LOT more than adding an > > >> argument to generic_make_request.> > > > > We may introduce generic_make_request_with_ref() for > > > blk_queue_split(), what do you think of the following patch? > > > > Functionally it'll work as well. > > If you are fine with this one, I'm glad to prepare for a normal one, > so that we can save one precious bio flag, and the total number is > just 13. […] I am completely lost about what you are talking about. That is fine with me, but in case there is anything to test… let me know. I do not promise quick responses tough, as I have been quite a bit overwhelmed with stuff lately. Currently I run 4.17-rc7 without any blk-mq related patches and it appears to be free of hangs or stalls related to blk-mq. Thanks, -- Martin
