add a bvec_phys helper v2

add a bvec_phys helper v2

[Christoph Hellwig]

Hi Jens,

this series adds a bvec_phys helper to get the physical address
of a bio_vec so that callers don't have to poke into bvec internals.
There aren't a whole lot of user of it yet, but with the new proposed
DMA mapping API we might grow a lot more soon.

Changes since v1:
 - reorder the two patches as suggested by Geert
 - fix a comment typo
 - use PFN_PHYS instead of open coding it
 - also pass a len argument to get_max_segment_size instead of open
   coding a min in both callers

Diffstat:
 arch/m68k/emu/nfblock.c |    2 +-
 block/bio.c             |    2 +-
 block/blk-merge.c       |   27 ++++++++++++---------------
 block/blk.h             |    4 ++--
 include/linux/bvec.h    |   14 ++++++++++++++
 5 files changed, 30 insertions(+), 19 deletions(-)

[PATCH 1/2] block: add a bvec_phys helper

[Christoph Hellwig]

Get callers out of poking into bvec internals a bit more.  Not a huge win
right now, but with the proposed new DMA mapping API we might end up with
a lot more of this otherwise.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 arch/m68k/emu/nfblock.c |  2 +-
 block/bio.c             |  2 +-
 block/blk.h             |  4 ++--
 include/linux/bvec.h    | 14 ++++++++++++++
 4 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/arch/m68k/emu/nfblock.c b/arch/m68k/emu/nfblock.c
index 8eea7ef9115146..874fe958877388 100644
--- a/arch/m68k/emu/nfblock.c
+++ b/arch/m68k/emu/nfblock.c
@@ -71,7 +71,7 @@ static void nfhd_submit_bio(struct bio *bio)
 		len = bvec.bv_len;
 		len >>= 9;
 		nfhd_read_write(dev->id, 0, dir, sec >> shift, len >> shift,
-				page_to_phys(bvec.bv_page) + bvec.bv_offset);
+				bvec_phys(&bvec));
 		sec += len;
 	}
 	bio_endio(bio);
diff --git a/block/bio.c b/block/bio.c
index e9e809a63c5975..a3b1b2266c50be 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -953,7 +953,7 @@ bool bvec_try_merge_hw_page(struct request_queue *q, struct bio_vec *bv,
 		bool *same_page)
 {
 	unsigned long mask = queue_segment_boundary(q);
-	phys_addr_t addr1 = page_to_phys(bv->bv_page) + bv->bv_offset;
+	phys_addr_t addr1 = bvec_phys(bv);
 	phys_addr_t addr2 = page_to_phys(page) + offset + len - 1;
 
 	if ((addr1 | mask) != (addr2 | mask))
diff --git a/block/blk.h b/block/blk.h
index 47dadd2439b1ca..8e8936e97307c6 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -98,8 +98,8 @@ static inline bool biovec_phys_mergeable(struct request_queue *q,
 		struct bio_vec *vec1, struct bio_vec *vec2)
 {
 	unsigned long mask = queue_segment_boundary(q);
-	phys_addr_t addr1 = page_to_phys(vec1->bv_page) + vec1->bv_offset;
-	phys_addr_t addr2 = page_to_phys(vec2->bv_page) + vec2->bv_offset;
+	phys_addr_t addr1 = bvec_phys(vec1);
+	phys_addr_t addr2 = bvec_phys(vec2);
 
 	/*
 	 * Merging adjacent physical pages may not work correctly under KMSAN
diff --git a/include/linux/bvec.h b/include/linux/bvec.h
index bd1e361b351c5a..f41c7f0ef91ed5 100644
--- a/include/linux/bvec.h
+++ b/include/linux/bvec.h
@@ -280,4 +280,18 @@ static inline void *bvec_virt(struct bio_vec *bvec)
 	return page_address(bvec->bv_page) + bvec->bv_offset;
 }
 
+/**
+ * bvec_phys - return the physical address for a bvec
+ * @bvec: bvec to return the physical address for
+ */
+static inline phys_addr_t bvec_phys(const struct bio_vec *bvec)
+{
+	/*
+	 * Note this open codes page_to_phys because page_to_phys is defined in
+	 * <asm/io.h>, which we don't want to pull in here.  If it ever moves to
+	 * a sensible place we should start using it.
+	 */
+	return PFN_PHYS(page_to_pfn(bvec->bv_page)) + bvec->bv_offset;
+}
+
 #endif /* __LINUX_BVEC_H */
-- 
2.43.0

[PATCH 2/2] block: pass a phys_addr_t to get_max_segment_size

[Christoph Hellwig]

Work on a single address to simplify the logic, and prepare the callers
from using better helpers.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 block/blk-merge.c | 27 ++++++++++++---------------
 1 file changed, 12 insertions(+), 15 deletions(-)

diff --git a/block/blk-merge.c b/block/blk-merge.c
index cff20bcc0252a7..b1e1b7a6933511 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -207,25 +207,24 @@ static inline unsigned get_max_io_size(struct bio *bio,
 }
 
 /**
- * get_max_segment_size() - maximum number of bytes to add as a single segment
+ * get_max_segment_size() - maximum number of bytes to add to a single segment
  * @lim: Request queue limits.
- * @start_page: See below.
- * @offset: Offset from @start_page where to add a segment.
+ * @paddr: address of the range to add
+ * @max_len: maximum length available to add at @paddr
  *
- * Returns the maximum number of bytes that can be added as a single segment.
+ * Returns the maximum number of bytes of the range starting at @paddr that can
+ * be added to a single segment.
  */
 static inline unsigned get_max_segment_size(const struct queue_limits *lim,
-		struct page *start_page, unsigned long offset)
+		phys_addr_t paddr, unsigned int len)
 {
-	unsigned long mask = lim->seg_boundary_mask;
-
-	offset = mask & (page_to_phys(start_page) + offset);
-
 	/*
 	 * Prevent an overflow if mask = ULONG_MAX and offset = 0 by adding 1
 	 * after having calculated the minimum.
 	 */
-	return min(mask - offset, (unsigned long)lim->max_segment_size - 1) + 1;
+	return min_t(unsigned long, len,
+		min(lim->seg_boundary_mask - (lim->seg_boundary_mask & paddr),
+		    (unsigned long)lim->max_segment_size - 1) + 1);
 }
 
 /**
@@ -258,9 +257,7 @@ static bool bvec_split_segs(const struct queue_limits *lim,
 	unsigned seg_size = 0;
 
 	while (len && *nsegs < max_segs) {
-		seg_size = get_max_segment_size(lim, bv->bv_page,
-						bv->bv_offset + total_len);
-		seg_size = min(seg_size, len);
+		seg_size = get_max_segment_size(lim, bvec_phys(bv) + total_len, len);
 
 		(*nsegs)++;
 		total_len += seg_size;
@@ -494,8 +491,8 @@ static unsigned blk_bvec_map_sg(struct request_queue *q,
 
 	while (nbytes > 0) {
 		unsigned offset = bvec->bv_offset + total;
-		unsigned len = min(get_max_segment_size(&q->limits,
-				   bvec->bv_page, offset), nbytes);
+		unsigned len = get_max_segment_size(&q->limits, bvec_phys(bvec),
+			nbytes);
 		struct page *page = bvec->bv_page;
 
 		/*
-- 
2.43.0

Re: [PATCH 2/2] block: pass a phys_addr_t to get_max_segment_size

[Jens Axboe]

On 7/5/24 6:32 AM, Christoph Hellwig wrote:
> Work on a single address to simplify the logic, and prepare the callers
> from using better helpers.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  block/blk-merge.c | 27 ++++++++++++---------------
>  1 file changed, 12 insertions(+), 15 deletions(-)
> 
> diff --git a/block/blk-merge.c b/block/blk-merge.c
> index cff20bcc0252a7..b1e1b7a6933511 100644
> --- a/block/blk-merge.c
> +++ b/block/blk-merge.c
> @@ -207,25 +207,24 @@ static inline unsigned get_max_io_size(struct bio *bio,
>  }
>  
>  /**
> - * get_max_segment_size() - maximum number of bytes to add as a single segment
> + * get_max_segment_size() - maximum number of bytes to add to a single segment

v1 had this change too, not sure why? The previous description seems
better than the changed one.

-- 
Jens Axboe

Re: [PATCH 2/2] block: pass a phys_addr_t to get_max_segment_size

[Christoph Hellwig]

On Sat, Jul 06, 2024 at 12:21:06AM -0600, Jens Axboe wrote:
> >  /**
> > - * get_max_segment_size() - maximum number of bytes to add as a single segment
> > + * get_max_segment_size() - maximum number of bytes to add to a single segment
> 
> v1 had this change too, not sure why? The previous description seems
> better than the changed one.

Because it is used to also append data from another bio_vec to a
pre-existing SG segment, not just to create new ones.

Re: [PATCH 2/2] block: pass a phys_addr_t to get_max_segment_size

[Jens Axboe]

On 7/6/24 12:22 AM, Christoph Hellwig wrote:
> On Sat, Jul 06, 2024 at 12:21:06AM -0600, Jens Axboe wrote:
>>>  /**
>>> - * get_max_segment_size() - maximum number of bytes to add as a single segment
>>> + * get_max_segment_size() - maximum number of bytes to add to a single segment
>>
>> v1 had this change too, not sure why? The previous description seems
>> better than the changed one.
> 
> Because it is used to also append data from another bio_vec to a
> pre-existing SG segment, not just to create new ones.

Sure, but then let's make it something that makes more sense at least.
Maximum size to consider as a single segment, or something like that.

-- 
Jens Axboe

add a bvec_phys helper v3

[Christoph Hellwig]

Hi Jens,

this series adds a bvec_phys helper to get the physical address
of a bio_vec so that callers don't have to poke into bvec internals.
There aren't a whole lot of user of it yet, but with the new proposed
DMA mapping API we might grow a lot more soon.

Changes since v2:
 - keep the existing (somewhat weird) description of get_max_segment_size

Changes since v1:
 - reorder the two patches as suggested by Geert
 - fix a comment typo
 - use PFN_PHYS instead of open coding it
 - also pass a len argument to get_max_segment_size instead of open
   coding a min in both callers

Diffstat:
 arch/m68k/emu/nfblock.c |    2 +-
 block/bio.c             |    2 +-
 block/blk-merge.c       |   25 +++++++++++--------------
 block/blk.h             |    4 ++--
 include/linux/bvec.h    |   14 ++++++++++++++
 5 files changed, 29 insertions(+), 18 deletions(-)

[PATCH 1/2] block: add a bvec_phys helper

[Christoph Hellwig]

Get callers out of poking into bvec internals a bit more.  Not a huge win
right now, but with the proposed new DMA mapping API we might end up with
a lot more of this otherwise.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 arch/m68k/emu/nfblock.c |  2 +-
 block/bio.c             |  2 +-
 block/blk.h             |  4 ++--
 include/linux/bvec.h    | 14 ++++++++++++++
 4 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/arch/m68k/emu/nfblock.c b/arch/m68k/emu/nfblock.c
index 8eea7ef9115146..874fe958877388 100644
--- a/arch/m68k/emu/nfblock.c
+++ b/arch/m68k/emu/nfblock.c
@@ -71,7 +71,7 @@ static void nfhd_submit_bio(struct bio *bio)
 		len = bvec.bv_len;
 		len >>= 9;
 		nfhd_read_write(dev->id, 0, dir, sec >> shift, len >> shift,
-				page_to_phys(bvec.bv_page) + bvec.bv_offset);
+				bvec_phys(&bvec));
 		sec += len;
 	}
 	bio_endio(bio);
diff --git a/block/bio.c b/block/bio.c
index e9e809a63c5975..a3b1b2266c50be 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -953,7 +953,7 @@ bool bvec_try_merge_hw_page(struct request_queue *q, struct bio_vec *bv,
 		bool *same_page)
 {
 	unsigned long mask = queue_segment_boundary(q);
-	phys_addr_t addr1 = page_to_phys(bv->bv_page) + bv->bv_offset;
+	phys_addr_t addr1 = bvec_phys(bv);
 	phys_addr_t addr2 = page_to_phys(page) + offset + len - 1;
 
 	if ((addr1 | mask) != (addr2 | mask))
diff --git a/block/blk.h b/block/blk.h
index 8c27d524303aa7..d099ef1df5f64a 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -109,8 +109,8 @@ static inline bool biovec_phys_mergeable(struct request_queue *q,
 		struct bio_vec *vec1, struct bio_vec *vec2)
 {
 	unsigned long mask = queue_segment_boundary(q);
-	phys_addr_t addr1 = page_to_phys(vec1->bv_page) + vec1->bv_offset;
-	phys_addr_t addr2 = page_to_phys(vec2->bv_page) + vec2->bv_offset;
+	phys_addr_t addr1 = bvec_phys(vec1);
+	phys_addr_t addr2 = bvec_phys(vec2);
 
 	/*
 	 * Merging adjacent physical pages may not work correctly under KMSAN
diff --git a/include/linux/bvec.h b/include/linux/bvec.h
index bd1e361b351c5a..f41c7f0ef91ed5 100644
--- a/include/linux/bvec.h
+++ b/include/linux/bvec.h
@@ -280,4 +280,18 @@ static inline void *bvec_virt(struct bio_vec *bvec)
 	return page_address(bvec->bv_page) + bvec->bv_offset;
 }
 
+/**
+ * bvec_phys - return the physical address for a bvec
+ * @bvec: bvec to return the physical address for
+ */
+static inline phys_addr_t bvec_phys(const struct bio_vec *bvec)
+{
+	/*
+	 * Note this open codes page_to_phys because page_to_phys is defined in
+	 * <asm/io.h>, which we don't want to pull in here.  If it ever moves to
+	 * a sensible place we should start using it.
+	 */
+	return PFN_PHYS(page_to_pfn(bvec->bv_page)) + bvec->bv_offset;
+}
+
 #endif /* __LINUX_BVEC_H */
-- 
2.43.0

Re: [PATCH 1/2] block: add a bvec_phys helper

[Chaitanya Kulkarni]

On 7/6/2024 12:52 AM, Christoph Hellwig wrote:
> Get callers out of poking into bvec internals a bit more.  Not a huge win
> right now, but with the proposed new DMA mapping API we might end up with
> a lot more of this otherwise.
> 
> Signed-off-by: Christoph Hellwig<hch@lst.de>

thanks for the adding open code comment, looks good.

Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>

-ck