account for ashift when gathering buffers to be written to l2arc device

The change that introduced the L2ARC compression support also introduced
a bug where the on-disk size of the selected buffers could end up larger
than the target size if the ashift is greater than 9.  This was because
the buffer selection could did not take into account the fact that
on-disk size could be larger than the in-memory buffer size due to
the alignment requirements.

At the moment b_asize is a misnomer as it does not always represent the
allocated size: if a buffer is compressed, then the compressed size is
properly rounded (on FreeBSD), but if the compression fails or it is not
applied, then the original size is kept and it could be smaller than what
ashift requires.

For the same reasons arcstat_l2_asize and the reported used space
on the cache device could be smaller than the actual allocated size
if ashift > 9.  That problem is not fixed by this change.

This change only ensures that l2ad_hand is not advanced by more
than target_sz.  Otherwise we would overwrite active (unevicted)
L2ARC buffers.  That problem is manifested as growing l2_cksum_bad
and l2_io_error counters.

This change also changes 'p' prefix to 'a' prefix in a few places
where variables represent allocated rather than physical size.

The resolved problem could also result in the reported allocated size
being greater than the cache device's capacity, because of the
overwritten buffers (more than one buffer claiming the same disk
space).

This change is already in ZFS-on-Linux:
zfsonlinux/zfs@ef56b0780c80ebb0b1e637b8b8c79530a8ab3201

PR:		198242
PR:		195746 (possibly related)
Reviewed by:	mahrens (https://reviews.csiden.org/r/229/)
Tested by:	gkontos@aicom.gr (most recently)
MFC after:	15 days
X-MFC note:	patch does not apply as is at the moment
Relnotes:	yes
Sponsored by:	ClusterHQ
Differential Revision:	https://reviews.freebsd.org/D2764
Reviewed by:	noone (@FreeBSD.org)

1 files changed, 33 insertions, 13 deletions

diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c
index e44c04391f58..e078bb30fa55 100644
--- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c
+++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c
@@ -6182,8 +6182,7 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
     boolean_t *headroom_boost)
 {
 	arc_buf_hdr_t *hdr, *hdr_prev, *head;
-	uint64_t write_asize, write_psize, write_sz, headroom,
-	    buf_compress_minsz;
+	uint64_t write_asize, write_sz, headroom, buf_compress_minsz;
 	void *buf_data;
 	boolean_t full;
 	l2arc_write_callback_t *cb;
@@ -6198,7 +6197,7 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
 	*headroom_boost = B_FALSE;
 
 	pio = NULL;
-	write_sz = write_asize = write_psize = 0;
+	write_sz = write_asize = 0;
 	full = B_FALSE;
 	head = kmem_cache_alloc(hdr_l2only_cache, KM_PUSHPAGE);
 	head->b_flags |= ARC_FLAG_L2_WRITE_HEAD;
@@ -6240,6 +6239,7 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
 		for (; hdr; hdr = hdr_prev) {
 			kmutex_t *hash_lock;
 			uint64_t buf_sz;
+			uint64_t buf_a_sz;
 
 			if (arc_warm == B_FALSE)
 				hdr_prev = multilist_sublist_next(mls, hdr);
@@ -6271,7 +6271,15 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
 				continue;
 			}
 
-			if ((write_sz + hdr->b_size) > target_sz) {
+			/*
+			 * Assume that the buffer is not going to be compressed
+			 * and could take more space on disk because of a larger
+			 * disk block size.
+			 */
+			buf_sz = hdr->b_size;
+			buf_a_sz = vdev_psize_to_asize(dev->l2ad_vdev, buf_sz);
+
+			if ((write_asize + buf_a_sz) > target_sz) {
 				full = B_TRUE;
 				mutex_exit(hash_lock);
 				ARCSTAT_BUMP(arcstat_l2_write_full);
@@ -6336,8 +6344,6 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
 			 * using it to denote the header's state change.
 			 */
 			hdr->b_l2hdr.b_daddr = L2ARC_ADDR_UNSET;
-
-			buf_sz = hdr->b_size;
 			hdr->b_flags |= ARC_FLAG_HAS_L2HDR;
 
 			mutex_enter(&dev->l2ad_mtx);
@@ -6354,6 +6360,7 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
 			mutex_exit(hash_lock);
 
 			write_sz += buf_sz;
+			write_asize += buf_a_sz;
 		}
 
 		multilist_sublist_unlock(mls);
@@ -6373,6 +6380,19 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
 	mutex_enter(&dev->l2ad_mtx);
 
 	/*
+	 * Note that elsewhere in this file arcstat_l2_asize
+	 * and the used space on l2ad_vdev are updated using b_asize,
+	 * which is not necessarily rounded up to the device block size.
+	 * Too keep accounting consistent we do the same here as well:
+	 * stats_size accumulates the sum of b_asize of the written buffers,
+	 * while write_asize accumulates the sum of b_asize rounded up
+	 * to the device block size.
+	 * The latter sum is used only to validate the corectness of the code.
+	 */
+	uint64_t stats_size = 0;
+	write_asize = 0;
+
+	/*
 	 * Now start writing the buffers. We're starting at the write head
 	 * and work backwards, retracing the course of the buffer selector
 	 * loop above.
@@ -6425,7 +6445,7 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
 
 		/* Compression may have squashed the buffer to zero length. */
 		if (buf_sz != 0) {
-			uint64_t buf_p_sz;
+			uint64_t buf_a_sz;
 
 			wzio = zio_write_phys(pio, dev->l2ad_vdev,
 			    dev->l2ad_hand, buf_sz, buf_data, ZIO_CHECKSUM_OFF,
@@ -6436,14 +6456,14 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
 			    zio_t *, wzio);
 			(void) zio_nowait(wzio);
 
-			write_asize += buf_sz;
+			stats_size += buf_sz;
 
 			/*
 			 * Keep the clock hand suitably device-aligned.
 			 */
-			buf_p_sz = vdev_psize_to_asize(dev->l2ad_vdev, buf_sz);
-			write_psize += buf_p_sz;
-			dev->l2ad_hand += buf_p_sz;
+			buf_a_sz = vdev_psize_to_asize(dev->l2ad_vdev, buf_sz);
+			write_asize += buf_a_sz;
+			dev->l2ad_hand += buf_a_sz;
 		}
 	}
 
@@ -6453,8 +6473,8 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz,
 	ARCSTAT_BUMP(arcstat_l2_writes_sent);
 	ARCSTAT_INCR(arcstat_l2_write_bytes, write_asize);
 	ARCSTAT_INCR(arcstat_l2_size, write_sz);
-	ARCSTAT_INCR(arcstat_l2_asize, write_asize);
-	vdev_space_update(dev->l2ad_vdev, write_asize, 0, 0);
+	ARCSTAT_INCR(arcstat_l2_asize, stats_size);
+	vdev_space_update(dev->l2ad_vdev, stats_size, 0, 0);
 
 	/*
 	 * Bump device hand to the device start if it is approaching the end.