On 3/3/22 7:28 PM, Pavel Begunkov wrote: > On 3/4/22 02:18, Jens Axboe wrote: >> On 3/3/22 6:49 PM, Pavel Begunkov wrote: >>> On 3/3/22 16:31, Jens Axboe wrote: >>>> On 3/3/22 7:40 AM, Jens Axboe wrote: >>>>> On 3/3/22 7:36 AM, Jens Axboe wrote: >>>>>> The only potential oddity here is that the fd passed back is not a >>>>>> legitimate fd. io_uring does support poll(2) on its file descriptor, so >>>>>> that could cause some confusion even if I don't think anyone actually >>>>>> does poll(2) on io_uring. >>>>> > [...] >>>> which is about a 15% improvement, pretty massive... >>> >>> Is the bench single threaded (including io-wq)? Because if it >>> is, get/put shouldn't do any atomics and I don't see where the >>> result comes from. >> >> Yes, it has a main thread and IO threads. Which is not uncommon, most >> things are multithreaded these days... > > They definitely are, just was confused by the bench as I can't > recall t/io_uring having >1 threads for nops and/or direct bdev I/O It has a main task that spawns a thread for each job. Even for nops that's true. The main task just checks progress of each IO job (thread) and prints it, sleeping most of the time. -- Jens Axboe
