svn commit: r344926 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs

[Alexander Motin]

Author: mav
Date: Fri Mar  8 18:49:27 2019
New Revision: 344926
URL: https://svnweb.freebsd.org/changeset/base/344926

Log:
  MFV/ZoL: Fix zfs_vdev_aggregation_limit bounds checking
  
  Update the bounds checking for zfs_vdev_aggregation_limit so that
  it has a floor of zero and a maximum value of the supported block
  size for the pool.
  
  Additionally add an early return when zfs_vdev_aggregation_limit
  equals zero to disable aggregation.  For very fast solid state or
  memory devices it may be more expensive to perform the aggregation
  than to issue the IO immediately.
  
  Author: Brian Behlendorf <behlendorf1 at llnl.gov>
  zfsonlinux/zfs at a58df6f53687ac6d1dee21f60de41b2552a43201
  
  MFV/ZoL: Cap maximum aggregate IO size
  
  Commit 8542ef8 allowed optional IOs to be aggregated beyond
  the specified aggregation limit.  Since the aggregation limit
  was also used to enforce the maximum block size, setting
  `zfs_vdev_aggregation_limit=16777216` could result in an
  attempt to allocate an ABD larger than 16M.
  
  Author: Brian Behlendorf <behlendorf1 at llnl.gov>
  Reviewed by: Matthew Ahrens <mahrens at delphix.com>
  Reviewed-by: George Melikov <mail at gmelikov.ru>
  Reviewed-by: Giuseppe Di Natale <dinatale2 at llnl.gov>
  Signed-off-by: Brian Behlendorf <behlendorf1 at llnl.gov>
  Closes #6259
  Closes #6270
  zfsonlinux/zfs at 2d678f779aba26a93314c8ee1142c3985fa25cb6

Modified:
  head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/vdev_queue.c

Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/vdev_queue.c
==============================================================================
--- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/vdev_queue.c	Fri Mar  8 18:17:57 2019	(r344925)
+++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/vdev_queue.c	Fri Mar  8 18:49:27 2019	(r344926)
@@ -673,13 +673,18 @@ vdev_queue_aggregate(vdev_queue_t *vq, zio_t *zio)
 	zio_link_t *zl = NULL;
 	uint64_t maxgap = 0;
 	uint64_t size;
+	uint64_t limit;
+	int maxblocksize;
 	boolean_t stretch;
 	avl_tree_t *t;
 	enum zio_flag flags;
 
 	ASSERT(MUTEX_HELD(&vq->vq_lock));
 
-	if (zio->io_flags & ZIO_FLAG_DONT_AGGREGATE)
+	maxblocksize = spa_maxblocksize(vq->vq_vdev->vdev_spa);
+	limit = MAX(MIN(zfs_vdev_aggregation_limit, maxblocksize), 0);
+
+	if (zio->io_flags & ZIO_FLAG_DONT_AGGREGATE || limit == 0)
 		return (NULL);
 
 	first = last = zio;
@@ -710,7 +715,7 @@ vdev_queue_aggregate(vdev_queue_t *vq, zio_t *zio)
 	t = vdev_queue_type_tree(vq, zio->io_type);
 	while (t != NULL && (dio = AVL_PREV(t, first)) != NULL &&
 	    (dio->io_flags & ZIO_FLAG_AGG_INHERIT) == flags &&
-	    IO_SPAN(dio, last) <= zfs_vdev_aggregation_limit &&
+	    IO_SPAN(dio, last) <= limit &&
 	    IO_GAP(dio, first) <= maxgap &&
 	    dio->io_type == zio->io_type) {
 		first = dio;
@@ -734,8 +739,9 @@ vdev_queue_aggregate(vdev_queue_t *vq, zio_t *zio)
 	 */
 	while ((dio = AVL_NEXT(t, last)) != NULL &&
 	    (dio->io_flags & ZIO_FLAG_AGG_INHERIT) == flags &&
-	    (IO_SPAN(first, dio) <= zfs_vdev_aggregation_limit ||
+	    (IO_SPAN(first, dio) <= limit ||
 	    (dio->io_flags & ZIO_FLAG_OPTIONAL)) &&
+	    IO_SPAN(first, dio) <= maxblocksize &&
 	    IO_GAP(last, dio) <= maxgap &&
 	    dio->io_type == zio->io_type) {
 		last = dio;
@@ -789,7 +795,7 @@ vdev_queue_aggregate(vdev_queue_t *vq, zio_t *zio)
 		return (NULL);
 
 	size = IO_SPAN(first, last);
-	ASSERT3U(size, <=, SPA_MAXBLOCKSIZE);
+	ASSERT3U(size, <=, maxblocksize);
 
 	aio = zio_vdev_delegated_io(first->io_vd, first->io_offset,
 	    abd_alloc_for_io(size, B_TRUE), size, first->io_type,