diff options
author | Stefan Hajnoczi <stefanha@redhat.com> | 2023-09-13 16:00:44 -0400 |
---|---|---|
committer | Kevin Wolf <kwolf@redhat.com> | 2023-10-31 15:42:14 +0100 |
commit | 84d61e5f36a73ed24742b7c7cf7b811e456dd024 (patch) | |
tree | 420c313770c60a0f9aaa2487b296c6883535b4b0 /block | |
parent | 433fcea40c31ff355f84da22a46977c2a1b596c3 (diff) |
virtio: use defer_call() in virtio_irqfd_notify()
virtio-blk and virtio-scsi invoke virtio_irqfd_notify() to send Used
Buffer Notifications from an IOThread. This involves an eventfd
write(2) syscall. Calling this repeatedly when completing multiple I/O
requests in a row is wasteful.
Use the defer_call() API to batch together virtio_irqfd_notify() calls
made during thread pool (aio=threads), Linux AIO (aio=native), and
io_uring (aio=io_uring) completion processing.
Behavior is unchanged for emulated devices that do not use
defer_call_begin()/defer_call_end() since defer_call() immediately
invokes the callback when called outside a
defer_call_begin()/defer_call_end() region.
fio rw=randread bs=4k iodepth=64 numjobs=8 IOPS increases by ~9% with a
single IOThread and 8 vCPUs. iodepth=1 decreases by ~1% but this could
be noise. Detailed performance data and configuration specifics are
available here:
https://gitlab.com/stefanha/virt-playbooks/-/tree/blk_io_plug-irqfd
This duplicates the BH that virtio-blk uses for batching. The next
commit will remove it.
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230913200045.1024233-4-stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Diffstat (limited to 'block')
-rw-r--r-- | block/io_uring.c | 6 | ||||
-rw-r--r-- | block/linux-aio.c | 4 |
2 files changed, 10 insertions, 0 deletions
diff --git a/block/io_uring.c b/block/io_uring.c index 3a1e1f45b3..7cdd00e9f1 100644 --- a/block/io_uring.c +++ b/block/io_uring.c @@ -125,6 +125,9 @@ static void luring_process_completions(LuringState *s) { struct io_uring_cqe *cqes; int total_bytes; + + defer_call_begin(); + /* * Request completion callbacks can run the nested event loop. * Schedule ourselves so the nested event loop will "see" remaining @@ -217,7 +220,10 @@ end: aio_co_wake(luringcb->co); } } + qemu_bh_cancel(s->completion_bh); + + defer_call_end(); } static int ioq_submit(LuringState *s) diff --git a/block/linux-aio.c b/block/linux-aio.c index a2670b3e46..ec05d946f3 100644 --- a/block/linux-aio.c +++ b/block/linux-aio.c @@ -205,6 +205,8 @@ static void qemu_laio_process_completions(LinuxAioState *s) { struct io_event *events; + defer_call_begin(); + /* Reschedule so nested event loops see currently pending completions */ qemu_bh_schedule(s->completion_bh); @@ -231,6 +233,8 @@ static void qemu_laio_process_completions(LinuxAioState *s) * own `for` loop. If we are the last all counters dropped to zero. */ s->event_max = 0; s->event_idx = 0; + + defer_call_end(); } static void qemu_laio_process_completions_and_submit(LinuxAioState *s) |