On 11/13/18 8:52 AM, Keith Busch wrote: > On Tue, Nov 13, 2018 at 08:42:28AM -0700, Jens Axboe wrote: >> If we're polling for IO on a device that doesn't use interrupts, then >> IO completion loop (and wake of task) is done by submitting task itself. >> If that is the case, then we don't need to enter the wake_up_process() >> function, we can simply mark ourselves as TASK_RUNNING. >> >> Signed-off-by: Jens Axboe <axboe@xxxxxxxxx> >> --- >> fs/block_dev.c | 6 ++---- >> fs/iomap.c | 3 +-- >> include/linux/blkdev.h | 19 +++++++++++++++++++ >> 3 files changed, 22 insertions(+), 6 deletions(-) > > One more for swap read: > > --- > diff --git a/mm/page_io.c b/mm/page_io.c > index d4d1c89bcddd..57572ff46016 100644 > --- a/mm/page_io.c > +++ b/mm/page_io.c > @@ -140,7 +140,7 @@ static void end_swap_bio_read(struct bio *bio) > unlock_page(page); > WRITE_ONCE(bio->bi_private, NULL); > bio_put(bio); > - wake_up_process(waiter); > + blk_wake_io_task(waiter); > put_task_struct(waiter); > } > Indeed, thanks Keith! I'll fold it in. -- Jens Axboe
