On 2/4/21 7:56 AM, Pavel Begunkov wrote: > On 04/02/2021 14:52, Jens Axboe wrote: >> On 2/4/21 6:52 AM, Pavel Begunkov wrote: >>> Saving one lock/unlock for io-wq is not super important, but adds some >>> ugliness in the code. More important, atomic decs not turning it to zero >>> for some archs won't give the right ordering/barriers so the >>> io_steal_work() may pretty easily get subtly and completely broken. >>> >>> Return back 2-step io-wq work exchange and clean it up. >> >> IIRC, this wasn't done to skip the lock/unlock exchange, which I agree >> doesn't matter, but to ensure that a link would not need another io-wq >> punt. And that is a big deal, it's much faster to run it from that >> same thread, rather than needing a new async queue and new thread grab >> to get there. > > Right, we just refer to different patches and moments. This one is fine > in that regard, it just moves returning link from ->do_work() to > ->free_work(). Got it, looks good then. -- Jens Axboe
