* [PATCH v2 2/3] block: Constify most queue limits pointers
2022-10-25 19:17 [PATCH 0/3] Block layer cleanup patches Bart Van Assche
2022-10-25 19:17 ` [PATCH v2 1/3] block: Remove request.write_hint Bart Van Assche
@ 2022-10-25 19:17 ` Bart Van Assche
2022-10-25 19:17 ` [PATCH v2 3/3] block: Micro-optimize get_max_segment_size() Bart Van Assche
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Bart Van Assche @ 2022-10-25 19:17 UTC (permalink / raw)
To: Jens Axboe
Cc: linux-block, Christoph Hellwig, Bart Van Assche, Ming Lei, Keith Busch
Document which functions do not modify the queue limits.
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Keith Busch <kbusch@kernel.org>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
block/blk-map.c | 2 +-
block/blk-merge.c | 29 ++++++++++++++++-------------
block/blk-settings.c | 6 +++---
block/blk.h | 11 ++++++-----
4 files changed, 26 insertions(+), 22 deletions(-)
diff --git a/block/blk-map.c b/block/blk-map.c
index 34735626b00f..46688e70b141 100644
--- a/block/blk-map.c
+++ b/block/blk-map.c
@@ -555,7 +555,7 @@ static int blk_rq_map_user_bvec(struct request *rq, const struct iov_iter *iter)
size_t nr_iter = iov_iter_count(iter);
size_t nr_segs = iter->nr_segs;
struct bio_vec *bvecs, *bvprvp = NULL;
- struct queue_limits *lim = &q->limits;
+ const struct queue_limits *lim = &q->limits;
unsigned int nsegs = 0, bytes = 0;
struct bio *bio;
size_t i;
diff --git a/block/blk-merge.c b/block/blk-merge.c
index ff04e9290715..58fdc3f8905b 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -100,13 +100,14 @@ static inline bool req_gap_front_merge(struct request *req, struct bio *bio)
* is defined as 'unsigned int', meantime it has to be aligned to with the
* logical block size, which is the minimum accepted unit by hardware.
*/
-static unsigned int bio_allowed_max_sectors(struct queue_limits *lim)
+static unsigned int bio_allowed_max_sectors(const struct queue_limits *lim)
{
return round_down(UINT_MAX, lim->logical_block_size) >> SECTOR_SHIFT;
}
-static struct bio *bio_split_discard(struct bio *bio, struct queue_limits *lim,
- unsigned *nsegs, struct bio_set *bs)
+static struct bio *bio_split_discard(struct bio *bio,
+ const struct queue_limits *lim,
+ unsigned *nsegs, struct bio_set *bs)
{
unsigned int max_discard_sectors, granularity;
sector_t tmp;
@@ -146,7 +147,8 @@ static struct bio *bio_split_discard(struct bio *bio, struct queue_limits *lim,
}
static struct bio *bio_split_write_zeroes(struct bio *bio,
- struct queue_limits *lim, unsigned *nsegs, struct bio_set *bs)
+ const struct queue_limits *lim,
+ unsigned *nsegs, struct bio_set *bs)
{
*nsegs = 0;
if (!lim->max_write_zeroes_sectors)
@@ -165,7 +167,7 @@ static struct bio *bio_split_write_zeroes(struct bio *bio,
* aligned to a physical block boundary.
*/
static inline unsigned get_max_io_size(struct bio *bio,
- struct queue_limits *lim)
+ const struct queue_limits *lim)
{
unsigned pbs = lim->physical_block_size >> SECTOR_SHIFT;
unsigned lbs = lim->logical_block_size >> SECTOR_SHIFT;
@@ -184,7 +186,7 @@ static inline unsigned get_max_io_size(struct bio *bio,
return max_sectors & ~(lbs - 1);
}
-static inline unsigned get_max_segment_size(struct queue_limits *lim,
+static inline unsigned get_max_segment_size(const struct queue_limits *lim,
struct page *start_page, unsigned long offset)
{
unsigned long mask = lim->seg_boundary_mask;
@@ -219,9 +221,9 @@ static inline unsigned get_max_segment_size(struct queue_limits *lim,
* *@nsegs segments and *@sectors sectors would make that bio unacceptable for
* the block driver.
*/
-static bool bvec_split_segs(struct queue_limits *lim, const struct bio_vec *bv,
- unsigned *nsegs, unsigned *bytes, unsigned max_segs,
- unsigned max_bytes)
+static bool bvec_split_segs(const struct queue_limits *lim,
+ const struct bio_vec *bv, unsigned *nsegs, unsigned *bytes,
+ unsigned max_segs, unsigned max_bytes)
{
unsigned max_len = min(max_bytes, UINT_MAX) - *bytes;
unsigned len = min(bv->bv_len, max_len);
@@ -267,7 +269,7 @@ static bool bvec_split_segs(struct queue_limits *lim, const struct bio_vec *bv,
* responsible for ensuring that @bs is only destroyed after processing of the
* split bio has finished.
*/
-static struct bio *bio_split_rw(struct bio *bio, struct queue_limits *lim,
+static struct bio *bio_split_rw(struct bio *bio, const struct queue_limits *lim,
unsigned *segs, struct bio_set *bs, unsigned max_bytes)
{
struct bio_vec bv, bvprv, *bvprvp = NULL;
@@ -331,8 +333,9 @@ static struct bio *bio_split_rw(struct bio *bio, struct queue_limits *lim,
* The split bio is allocated from @q->bio_split, which is provided by the
* block layer.
*/
-struct bio *__bio_split_to_limits(struct bio *bio, struct queue_limits *lim,
- unsigned int *nr_segs)
+struct bio *__bio_split_to_limits(struct bio *bio,
+ const struct queue_limits *lim,
+ unsigned int *nr_segs)
{
struct bio_set *bs = &bio->bi_bdev->bd_disk->bio_split;
struct bio *split;
@@ -377,7 +380,7 @@ struct bio *__bio_split_to_limits(struct bio *bio, struct queue_limits *lim,
*/
struct bio *bio_split_to_limits(struct bio *bio)
{
- struct queue_limits *lim = &bdev_get_queue(bio->bi_bdev)->limits;
+ const struct queue_limits *lim = &bdev_get_queue(bio->bi_bdev)->limits;
unsigned int nr_segs;
if (bio_may_exceed_limits(bio, lim))
diff --git a/block/blk-settings.c b/block/blk-settings.c
index 8bb9eef5310e..1cba5c2a2796 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -481,7 +481,7 @@ void blk_queue_io_opt(struct request_queue *q, unsigned int opt)
}
EXPORT_SYMBOL(blk_queue_io_opt);
-static int queue_limit_alignment_offset(struct queue_limits *lim,
+static int queue_limit_alignment_offset(const struct queue_limits *lim,
sector_t sector)
{
unsigned int granularity = max(lim->physical_block_size, lim->io_min);
@@ -491,8 +491,8 @@ static int queue_limit_alignment_offset(struct queue_limits *lim,
return (granularity + lim->alignment_offset - alignment) % granularity;
}
-static unsigned int queue_limit_discard_alignment(struct queue_limits *lim,
- sector_t sector)
+static unsigned int queue_limit_discard_alignment(
+ const struct queue_limits *lim, sector_t sector)
{
unsigned int alignment, granularity, offset;
diff --git a/block/blk.h b/block/blk.h
index d6ea0d1a6db0..7f9e089ab1f7 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -104,7 +104,7 @@ static inline bool biovec_phys_mergeable(struct request_queue *q,
return true;
}
-static inline bool __bvec_gap_to_prev(struct queue_limits *lim,
+static inline bool __bvec_gap_to_prev(const struct queue_limits *lim,
struct bio_vec *bprv, unsigned int offset)
{
return (offset & lim->virt_boundary_mask) ||
@@ -115,7 +115,7 @@ static inline bool __bvec_gap_to_prev(struct queue_limits *lim,
* Check if adding a bio_vec after bprv with offset would create a gap in
* the SG list. Most drivers don't care about this, but some do.
*/
-static inline bool bvec_gap_to_prev(struct queue_limits *lim,
+static inline bool bvec_gap_to_prev(const struct queue_limits *lim,
struct bio_vec *bprv, unsigned int offset)
{
if (!lim->virt_boundary_mask)
@@ -297,7 +297,7 @@ ssize_t part_timeout_store(struct device *, struct device_attribute *,
const char *, size_t);
static inline bool bio_may_exceed_limits(struct bio *bio,
- struct queue_limits *lim)
+ const struct queue_limits *lim)
{
switch (bio_op(bio)) {
case REQ_OP_DISCARD:
@@ -320,8 +320,9 @@ static inline bool bio_may_exceed_limits(struct bio *bio,
bio->bi_io_vec->bv_len + bio->bi_io_vec->bv_offset > PAGE_SIZE;
}
-struct bio *__bio_split_to_limits(struct bio *bio, struct queue_limits *lim,
- unsigned int *nr_segs);
+struct bio *__bio_split_to_limits(struct bio *bio,
+ const struct queue_limits *lim,
+ unsigned int *nr_segs);
int ll_back_merge_fn(struct request *req, struct bio *bio,
unsigned int nr_segs);
bool blk_attempt_req_merge(struct request_queue *q, struct request *rq,
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH v2 3/3] block: Micro-optimize get_max_segment_size()
2022-10-25 19:17 [PATCH 0/3] Block layer cleanup patches Bart Van Assche
2022-10-25 19:17 ` [PATCH v2 1/3] block: Remove request.write_hint Bart Van Assche
2022-10-25 19:17 ` [PATCH v2 2/3] block: Constify most queue limits pointers Bart Van Assche
@ 2022-10-25 19:17 ` Bart Van Assche
2022-10-25 19:40 ` [PATCH 0/3] Block layer cleanup patches Keith Busch
2022-10-25 19:42 ` Jens Axboe
4 siblings, 0 replies; 6+ messages in thread
From: Bart Van Assche @ 2022-10-25 19:17 UTC (permalink / raw)
To: Jens Axboe
Cc: linux-block, Christoph Hellwig, Bart Van Assche, Ming Lei,
Keith Busch, Steven Rostedt, Guenter Roeck
This patch removes a conditional jump from get_max_segment_size(). The
x86-64 assembler code for this function without this patch is as follows:
206 return min_not_zero(mask - offset + 1,
0x0000000000000118 <+72>: not %rax
0x000000000000011b <+75>: and 0x8(%r10),%rax
0x000000000000011f <+79>: add $0x1,%rax
0x0000000000000123 <+83>: je 0x138 <bvec_split_segs+104>
0x0000000000000125 <+85>: cmp %rdx,%rax
0x0000000000000128 <+88>: mov %rdx,%r12
0x000000000000012b <+91>: cmovbe %rax,%r12
0x000000000000012f <+95>: test %rdx,%rdx
0x0000000000000132 <+98>: mov %eax,%edx
0x0000000000000134 <+100>: cmovne %r12d,%edx
With this patch applied:
206 return min(mask - offset, (unsigned long)lim->max_segment_size - 1) + 1;
0x000000000000003f <+63>: mov 0x28(%rdi),%ebp
0x0000000000000042 <+66>: not %rax
0x0000000000000045 <+69>: and 0x8(%rdi),%rax
0x0000000000000049 <+73>: sub $0x1,%rbp
0x000000000000004d <+77>: cmp %rbp,%rax
0x0000000000000050 <+80>: cmova %rbp,%rax
0x0000000000000054 <+84>: add $0x1,%eax
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Keith Busch <kbusch@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
block/blk-merge.c | 15 +++++++++++----
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 58fdc3f8905b..35a8f75cc45d 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -186,6 +186,14 @@ static inline unsigned get_max_io_size(struct bio *bio,
return max_sectors & ~(lbs - 1);
}
+/**
+ * get_max_segment_size() - maximum number of bytes to add as a single segment
+ * @lim: Request queue limits.
+ * @start_page: See below.
+ * @offset: Offset from @start_page where to add a segment.
+ *
+ * Returns the maximum number of bytes that can be added as a single segment.
+ */
static inline unsigned get_max_segment_size(const struct queue_limits *lim,
struct page *start_page, unsigned long offset)
{
@@ -194,11 +202,10 @@ static inline unsigned get_max_segment_size(const struct queue_limits *lim,
offset = mask & (page_to_phys(start_page) + offset);
/*
- * overflow may be triggered in case of zero page physical address
- * on 32bit arch, use queue's max segment size when that happens.
+ * Prevent an overflow if mask = ULONG_MAX and offset = 0 by adding 1
+ * after having calculated the minimum.
*/
- return min_not_zero(mask - offset + 1,
- (unsigned long)lim->max_segment_size);
+ return min(mask - offset, (unsigned long)lim->max_segment_size - 1) + 1;
}
/**
^ permalink raw reply related [flat|nested] 6+ messages in thread