Move write aggregation memory copy out of vq_lock

Memory copy is too heavy operation to do under the congested lock.
Moving it out reduces congestion by many times to almost invisible.
Since the original zio removed from the queue, and the child zio is
not executed yet, I don't see why would the copy need protection.
My guess it just remained like this from the time when lock was not
dropped here, which was added later to fix lock ordering issue.

Multi-threaded sequential write tests with both HDD and SSD pools
with ZVOL block sizes of 4KB, 16KB, 64KB and 128KB all show major
reduction of lock congestion, saving from 15% to 35% of CPU time
and increasing throughput from 10% to 40%.

Reviewed-by: Richard Yao <ryao@gentoo.org>
Reviewed-by: Matt Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by:  Alexander Motin <mav@FreeBSD.org>
Closes #8890

diff --git a/module/zfs/vdev_queue.c b/module/zfs/vdev_queue.c
index e74df76b753083761fd2a6e416b2a3f9b1595cdc..86b20f134834e650edfcb1802acd1d68b66818e4 100644 (file)
--- a/module/zfs/vdev_queue.c
+++ b/module/zfs/vdev_queue.c
@@ -709,6 +709,18 @@ vdev_queue_aggregate(vdev_queue_t *vq, zio_t *zio)
        do {
                dio = nio;
                nio = AVL_NEXT(t, dio);
+               zio_add_child(dio, aio);
+               vdev_queue_io_remove(vq, dio);
+       } while (dio != last);
+
+       /*
+        * We need to drop the vdev queue's lock during zio_execute() to
+        * avoid a deadlock that we could encounter due to lock order
+        * reversal between vq_lock and io_lock in zio_change_priority().
+        * Use the dropped lock to do memory copy without congestion.
+        */
+       mutex_exit(&vq->vq_lock);
+       while ((dio = zio_walk_parents(aio, &zl)) != NULL) {
                ASSERT3U(dio->io_type, ==, aio->io_type);
 
                if (dio->io_flags & ZIO_FLAG_NODATA) {
@@ -720,16 +732,6 @@ vdev_queue_aggregate(vdev_queue_t *vq, zio_t *zio)
                            dio->io_offset - aio->io_offset, 0, dio->io_size);
                }
 
-               zio_add_child(dio, aio);
-               vdev_queue_io_remove(vq, dio);
-       } while (dio != last);
-
-       /*
-        * We need to drop the vdev queue's lock to avoid a deadlock that we
-        * could encounter since this I/O will complete immediately.
-        */
-       mutex_exit(&vq->vq_lock);
-       while ((dio = zio_walk_parents(aio, &zl)) != NULL) {
                zio_vdev_io_bypass(dio);
                zio_execute(dio);
        }