bio_discard_limit() uses round_up() and round_down(), which require a
power-of-two alignment. However, discard_granularity need not be a power
of two. For example, dm-thin supports data block sizes that are arbitrary
multiples of 64 KiB and advertises the block size as its granularity.
With a 192 KiB granularity, discarding one complete block at offset
384 KiB is split into 128 KiB and 64 KiB bios. Neither covers a complete
thin block, so dm-thin ignores both and the mapping remains allocated.
Use div_u64_rem() to find the starting sector's remainder without
modifying the sector. Round the maximum bio size down with rounddown()
and subtract the remainder so that the next bio starts on a granularity
boundary. This handles non-power-of-two granularities and avoids a small
initial bio when a large discard starts at an unaligned sector.
When discard is unsupported or disabled, discard_granularity can be zero.
Return BIO_MAX_SECTORS in that case so the existing bio submission checks
can handle unsupported discard requests without dividing by zero.
Fixes: 9b15d109a6b2 ("block: improve discard bio alignment in
__blkdev_issue_discard()")
Reported-by: Ming-Hung Tsai <[email protected]>
Link:
https://lore.kernel.org/r/CALjSBEvjbiGDVewuapWKhGDtaxZwcn4=22rauibeoyj22ry...@mail.gmail.com
Suggested-by: Christoph Hellwig <[email protected]>
Cc: [email protected]
Signed-off-by: Hungyu Lin <[email protected]>
---
Changes in v3:
- Use div_u64_rem() instead of sector_div(), as suggested by Christoph
Hellwig, so the input sector is not modified.
- Subtract the starting-sector remainder from the rounded maximum size
instead of emitting a short initial bio, following Christoph's proposal.
- Include <linux/math64.h> explicitly and retain v2's zero-granularity guard.
Changes in v2:
- Handle zero discard_granularity before dividing, as pointed out by
Sashiko. Leave unsupported discard handling to the bio submission path.
v2: https://lore.kernel.org/r/[email protected]
block/blk-lib.c | 28 ++++++++++++++--------------
1 file changed, 14 insertions(+), 14 deletions(-)
diff --git a/block/blk-lib.c b/block/blk-lib.c
index 688bc67cbf7..e98aa483707 100644
--- a/block/blk-lib.c
+++ b/block/blk-lib.c
@@ -3,6 +3,7 @@
* Functions related to generic helpers functions
*/
#include <linux/kernel.h>
+#include <linux/math64.h>
#include <linux/module.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
@@ -13,26 +14,25 @@
static sector_t bio_discard_limit(struct block_device *bdev, sector_t sector)
{
unsigned int discard_granularity = bdev_discard_granularity(bdev);
- sector_t granularity_aligned_sector;
+ u32 unaligned_sectors;
+
+ if (!discard_granularity)
+ return BIO_MAX_SECTORS;
if (bdev_is_partition(bdev))
sector += bdev->bd_start_sect;
- granularity_aligned_sector =
- round_up(sector, discard_granularity >> SECTOR_SHIFT);
-
- /*
- * Make sure subsequent bios start aligned to the discard granularity if
- * it needs to be split.
- */
- if (granularity_aligned_sector != sector)
- return granularity_aligned_sector - sector;
-
/*
- * Align the bio size to the discard granularity to make splitting the
bio
- * at discard granularity boundaries easier in the driver if needed.
+ * Align the bio size to the discard granularity to make splitting the
+ * bio at discard granularity boundaries easier in the driver if needed.
+ *
+ * If the current sector is not aligned, shorten the range so that the
+ * next one is aligned.
*/
- return round_down(BIO_MAX_SIZE, discard_granularity) >> SECTOR_SHIFT;
+ div_u64_rem(sector, discard_granularity >> SECTOR_SHIFT,
+ &unaligned_sectors);
+ return (rounddown(BIO_MAX_SIZE, discard_granularity) >> SECTOR_SHIFT) -
+ unaligned_sectors;
}
struct bio *blk_alloc_discard_bio(struct block_device *bdev,
base-commit: 6628e7d4c6b25c8d18f1345e3a20d21829aae7e4
--
2.50.1 (Apple Git-155)
