From: "tang.junhui" <tang.junhui@xxxxxxxxxx> We'd better let the stacking driver inherit underlying queue limits, so, we can avoid the unnessary mergering in the top layer, and It leads a 10% performance promotion in large sequential IO reading testing as follows: before: [root@ceph192-9-9-151 ~]# dd if=/dev/bcache3p2 of=/dev/zero bs=1M count=20480 iflag=direct 20480+0 records in 20480+0 records out 21474836480 bytes (21 GB) copied, 173.424 s, 124 MB/s after: [root@ceph192-9-9-151 tjh]# dd if=/dev/bcache3p2 of=/dev/zero bs=1M count=20480 iflag=direct 20480+0 records in 20480+0 records out 21474836480 bytes (21 GB) copied, 151.107 s, 142 MB/s Signed-off-by: tang.junhui <tang.junhui@xxxxxxxxxx> --- drivers/md/bcache/super.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index 3a19cbc..615782c 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -1129,6 +1129,8 @@ static int cached_dev_init(struct cached_dev *dc, unsigned block_size) if (ret) return ret; + blk_queue_stack_limits(dc->disk.disk->queue, q); + set_capacity(dc->disk.disk, dc->bdev->bd_part->nr_sects - dc->sb.data_offset); -- 2.8.1.windows.1 -- To unsubscribe from this list: send the line "unsubscribe linux-bcache" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
