aio-posix: split poll check from ready handler

Adaptive polling measures the execution time of the polling check plus
handlers called when a polled event becomes ready. Handlers can take a
significant amount of time, making it look like polling was running for
a long time when in fact the event handler was running for a long time.

For example, on Linux the io_submit(2) syscall invoked when a virtio-blk
device's virtqueue becomes ready can take 10s of microseconds. This
can exceed the default polling interval (32 microseconds) and cause
adaptive polling to stop polling.

By excluding the handler's execution time from the polling check we make
the adaptive polling calculation more accurate. As a result, the event
loop now stays in polling mode where previously it would have fallen
back to file descriptor monitoring.

The following data was collected with virtio-blk num-queues=2
event_idx=off using an IOThread. Before:

168k IOPS, IOThread syscalls:

  9837.115 ( 0.020 ms): IO iothread1/620155 io_submit(ctx_id: 140512552468480, nr: 16, iocbpp: 0x7fcb9f937db0)    = 16
  9837.158 ( 0.002 ms): IO iothread1/620155 write(fd: 103, buf: 0x556a2ef71b88, count: 8)                         = 8
  9837.161 ( 0.001 ms): IO iothread1/620155 write(fd: 104, buf: 0x556a2ef71b88, count: 8)                         = 8
  9837.163 ( 0.001 ms): IO iothread1/620155 ppoll(ufds: 0x7fcb90002800, nfds: 4, tsp: 0x7fcb9f1342d0, sigsetsize: 8) = 3
  9837.164 ( 0.001 ms): IO iothread1/620155 read(fd: 107, buf: 0x7fcb9f939cc0, count: 512)                        = 8
  9837.174 ( 0.001 ms): IO iothread1/620155 read(fd: 105, buf: 0x7fcb9f939cc0, count: 512)                        = 8
  9837.176 ( 0.001 ms): IO iothread1/620155 read(fd: 106, buf: 0x7fcb9f939cc0, count: 512)                        = 8
  9837.209 ( 0.035 ms): IO iothread1/620155 io_submit(ctx_id: 140512552468480, nr: 32, iocbpp: 0x7fca7d0cebe0)    = 32

174k IOPS (+3.6%), IOThread syscalls:

  9809.566 ( 0.036 ms): IO iothread1/623061 io_submit(ctx_id: 140539805028352, nr: 32, iocbpp: 0x7fd0cdd62be0)    = 32
  9809.625 ( 0.001 ms): IO iothread1/623061 write(fd: 103, buf: 0x5647cfba5f58, count: 8)                         = 8
  9809.627 ( 0.002 ms): IO iothread1/623061 write(fd: 104, buf: 0x5647cfba5f58, count: 8)                         = 8
  9809.663 ( 0.036 ms): IO iothread1/623061 io_submit(ctx_id: 140539805028352, nr: 32, iocbpp: 0x7fd0d0388b50)    = 32

Notice that ppoll(2) and eventfd read(2) syscalls are eliminated because
the IOThread stays in polling mode instead of falling back to file
descriptor monitoring.

As usual, polling is not implemented on Windows so this patch ignores
the new io_poll_read() callback in aio-win32.c.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-id: 20211207132336.36627-2-stefanha@redhat.com

[Fixed up aio_set_event_notifier() calls in
tests/unit/test-fdmon-epoll.c added after this series was queued.
--Stefan]

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

1 files changed, 33 insertions, 18 deletions

diff --git a/block/nvme.c b/block/nvme.c
index fa360b9..dd20de3 100644
--- a/block/nvme.c
+++ b/block/nvme.c
@@ -605,10 +605,8 @@ out:
     return ret;
 }
 
-static bool nvme_poll_queue(NVMeQueuePair *q)
+static void nvme_poll_queue(NVMeQueuePair *q)
 {
-    bool progress = false;
-
     const size_t cqe_offset = q->cq.head * NVME_CQ_ENTRY_BYTES;
     NvmeCqe *cqe = (NvmeCqe *)&q->cq.queue[cqe_offset];
 
@@ -619,30 +617,23 @@ static bool nvme_poll_queue(NVMeQueuePair *q)
      * cannot race with itself.
      */
     if ((le16_to_cpu(cqe->status) & 0x1) == q->cq_phase) {
-        return false;
+        return;
     }
 
     qemu_mutex_lock(&q->lock);
     while (nvme_process_completion(q)) {
         /* Keep polling */
-        progress = true;
     }
     qemu_mutex_unlock(&q->lock);
-
-    return progress;
 }
 
-static bool nvme_poll_queues(BDRVNVMeState *s)
+static void nvme_poll_queues(BDRVNVMeState *s)
 {
-    bool progress = false;
     int i;
 
     for (i = 0; i < s->queue_count; i++) {
-        if (nvme_poll_queue(s->queues[i])) {
-            progress = true;
-        }
+        nvme_poll_queue(s->queues[i]);
     }
-    return progress;
 }
 
 static void nvme_handle_event(EventNotifier *n)
@@ -703,8 +694,30 @@ static bool nvme_poll_cb(void *opaque)
     EventNotifier *e = opaque;
     BDRVNVMeState *s = container_of(e, BDRVNVMeState,
                                     irq_notifier[MSIX_SHARED_IRQ_IDX]);
+    int i;
 
-    return nvme_poll_queues(s);
+    for (i = 0; i < s->queue_count; i++) {
+        NVMeQueuePair *q = s->queues[i];
+        const size_t cqe_offset = q->cq.head * NVME_CQ_ENTRY_BYTES;
+        NvmeCqe *cqe = (NvmeCqe *)&q->cq.queue[cqe_offset];
+
+        /*
+         * q->lock isn't needed because nvme_process_completion() only runs in
+         * the event loop thread and cannot race with itself.
+         */
+        if ((le16_to_cpu(cqe->status) & 0x1) != q->cq_phase) {
+            return true;
+        }
+    }
+    return false;
+}
+
+static void nvme_poll_ready(EventNotifier *e)
+{
+    BDRVNVMeState *s = container_of(e, BDRVNVMeState,
+                                    irq_notifier[MSIX_SHARED_IRQ_IDX]);
+
+    nvme_poll_queues(s);
 }
 
 static int nvme_init(BlockDriverState *bs, const char *device, int namespace,
@@ -839,7 +852,8 @@ static int nvme_init(BlockDriverState *bs, const char *device, int namespace,
     }
     aio_set_event_notifier(bdrv_get_aio_context(bs),
                            &s->irq_notifier[MSIX_SHARED_IRQ_IDX],
-                           false, nvme_handle_event, nvme_poll_cb);
+                           false, nvme_handle_event, nvme_poll_cb,
+                           nvme_poll_ready);
 
     if (!nvme_identify(bs, namespace, errp)) {
         ret = -EIO;
@@ -924,7 +938,7 @@ static void nvme_close(BlockDriverState *bs)
     g_free(s->queues);
     aio_set_event_notifier(bdrv_get_aio_context(bs),
                            &s->irq_notifier[MSIX_SHARED_IRQ_IDX],
-                           false, NULL, NULL);
+                           false, NULL, NULL, NULL);
     event_notifier_cleanup(&s->irq_notifier[MSIX_SHARED_IRQ_IDX]);
     qemu_vfio_pci_unmap_bar(s->vfio, 0, s->bar0_wo_map,
                             0, sizeof(NvmeBar) + NVME_DOORBELL_SIZE);
@@ -1520,7 +1534,7 @@ static void nvme_detach_aio_context(BlockDriverState *bs)
 
     aio_set_event_notifier(bdrv_get_aio_context(bs),
                            &s->irq_notifier[MSIX_SHARED_IRQ_IDX],
-                           false, NULL, NULL);
+                           false, NULL, NULL, NULL);
 }
 
 static void nvme_attach_aio_context(BlockDriverState *bs,
@@ -1530,7 +1544,8 @@ static void nvme_attach_aio_context(BlockDriverState *bs,
 
     s->aio_context = new_context;
     aio_set_event_notifier(new_context, &s->irq_notifier[MSIX_SHARED_IRQ_IDX],
-                           false, nvme_handle_event, nvme_poll_cb);
+                           false, nvme_handle_event, nvme_poll_cb,
+                           nvme_poll_ready);
 
     for (unsigned i = 0; i < s->queue_count; i++) {
         NVMeQueuePair *q = s->queues[i];