Some drive manufacturers export a very large supported max discard size.
However, when the operating system sends I/O of the max size to the
device, extreme I/O latency can often be encountered. Since hardware
does not provide an optimal discard value in addition to the max, and
there is no way to foreshadow how well a drive handles the large size,
take the method from max_sectors setting, and use BLK_DEF_MAX_SECTORS to
set a more reasonable default discard max. This should avoid the extreme
latency while still allowing the user to increase the value for specific
needs.
Signed-off-by: John Pittman <jpittman@xxxxxxxxxx>
Suggested-by: David Jeffery <djeffery@xxxxxxxxxx>
---
Documentation/ABI/stable/sysfs-block | 4 +++-
block/blk-settings.c | 3 ++-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/Documentation/ABI/stable/sysfs-block b/Documentation/ABI/stable/sysfs-block
index c57e5b7cb532..158a1e6f1f6d 100644
--- a/Documentation/ABI/stable/sysfs-block
+++ b/Documentation/ABI/stable/sysfs-block
@@ -235,7 +235,9 @@ Description:
large latencies when large discards are issued, setting this
value lower will make Linux issue smaller discards and
potentially help reduce latencies induced by large discard
- operations.
+ operations. For this reason, the max is currently defaulted to
+ four times BLK_DEF_MAX_SECTORS, but can be increased via sysfs
+ as needed.
What: /sys/block/<disk>/queue/discard_max_hw_bytes
diff --git a/block/blk-settings.c b/block/blk-settings.c
index 4dd59059b788..4401c0b8477e 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -179,7 +179,8 @@ void blk_queue_max_discard_sectors(struct request_queue *q,
unsigned int max_discard_sectors)
{
q->limits.max_hw_discard_sectors = max_discard_sectors;
- q->limits.max_discard_sectors = max_discard_sectors;
+ q->limits.max_discard_sectors = min(max_discard_sectors,
+ BLK_DEF_MAX_SECTORS * 4);
}
EXPORT_SYMBOL(blk_queue_max_discard_sectors);
--
2.38.1
