[PATCH] block: save page offset gaps in cloned bio

[PATCH] block: save page offset gaps in cloned bio

[Keith Busch]

From: Keith Busch <kbusch@kernel.org>

The cloned bio needs to inherit the accumulated gaps between vectors so
that we can know if this bio can subscribe to the iova coalescing
optimization. The split remainder meanwhile can reset its gaps to 0
since it advanced beyond it and will start accounting from there.

Fixes: 2f6b2565d43c ("block: accumulate memory segment gaps per bio")
Reported-by: Eric Auger <eauger@redhat.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
---
 block/bio.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/block/bio.c b/block/bio.c
index 898b2f5ef8c81..33aa439c1bae2 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -859,6 +859,7 @@ static int __bio_clone(struct bio *bio, struct bio *bio_src, gfp_t gfp)
 	bio->bi_ioprio = bio_src->bi_ioprio;
 	bio->bi_write_hint = bio_src->bi_write_hint;
 	bio->bi_write_stream = bio_src->bi_write_stream;
+	bio->bi_bvec_gap_bit = bio_src->bi_bvec_gap_bit;
 	bio->bi_iter = bio_src->bi_iter;
 	bio->bi_io_vec = bio_src->bi_io_vec;
 
@@ -1971,6 +1972,7 @@ struct bio *bio_split(struct bio *bio, int sectors,
 		bio_integrity_trim(split);
 
 	bio_advance(bio, split->bi_iter.bi_size);
+	bio->bi_bvec_gap_bit = 0;
 
 	if (bio_flagged(bio, BIO_TRACE_COMPLETION))
 		bio_set_flag(split, BIO_TRACE_COMPLETION);
-- 
2.53.0-Meta

Re: [PATCH] block: save page offset gaps in cloned bio

[Eric Auger]

Hi Keith,

On 8/17/26 6:59 PM, Keith Busch wrote:
> From: Keith Busch <kbusch@kernel.org>
> 
> The cloned bio needs to inherit the accumulated gaps between vectors so
> that we can know if this bio can subscribe to the iova coalescing
> optimization. The split remainder meanwhile can reset its gaps to 0
> since it advanced beyond it and will start accounting from there.
> 
> Fixes: 2f6b2565d43c ("block: accumulate memory segment gaps per bio")
> Reported-by: Eric Auger <eauger@redhat.com>
> Signed-off-by: Keith Busch <kbusch@kernel.org>

Tested-by: Eric Auger <eric.auger@redhat.com>

Many Thanks!

Eric

> ---
>  block/bio.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/block/bio.c b/block/bio.c
> index 898b2f5ef8c81..33aa439c1bae2 100644
> --- a/block/bio.c
> +++ b/block/bio.c
> @@ -859,6 +859,7 @@ static int __bio_clone(struct bio *bio, struct bio *bio_src, gfp_t gfp)
>  	bio->bi_ioprio = bio_src->bi_ioprio;
>  	bio->bi_write_hint = bio_src->bi_write_hint;
>  	bio->bi_write_stream = bio_src->bi_write_stream;
> +	bio->bi_bvec_gap_bit = bio_src->bi_bvec_gap_bit;
>  	bio->bi_iter = bio_src->bi_iter;
>  	bio->bi_io_vec = bio_src->bi_io_vec;
>  
> @@ -1971,6 +1972,7 @@ struct bio *bio_split(struct bio *bio, int sectors,
>  		bio_integrity_trim(split);
>  
>  	bio_advance(bio, split->bi_iter.bi_size);
> +	bio->bi_bvec_gap_bit = 0;
>  
>  	if (bio_flagged(bio, BIO_TRACE_COMPLETION))
>  		bio_set_flag(split, BIO_TRACE_COMPLETION);

Re: [PATCH] block: save page offset gaps in cloned bio

[Christoph Hellwig]

On Mon, Aug 17, 2026 at 09:59:08AM -0700, Keith Busch wrote:
> From: Keith Busch <kbusch@kernel.org>
> 
> The cloned bio needs to inherit the accumulated gaps between vectors so
> that we can know if this bio can subscribe to the iova coalescing
> optimization.

Makes sense.

> The split remainder meanwhile can reset its gaps to 0
> since it advanced beyond it and will start accounting from there.

I don't really understand this part, though.  As in I don't even
understand what you mean here..

Re: [PATCH] block: save page offset gaps in cloned bio

[Keith Busch]

On Wed, Aug 19, 2026 at 07:23:06AM +0200, Christoph Hellwig wrote:
> On Mon, Aug 17, 2026 at 09:59:08AM -0700, Keith Busch wrote:
> > From: Keith Busch <kbusch@kernel.org>
> > 
> > The cloned bio needs to inherit the accumulated gaps between vectors so
> > that we can know if this bio can subscribe to the iova coalescing
> > optimization.
> 
> Makes sense.
> 
> > The split remainder meanwhile can reset its gaps to 0
> > since it advanced beyond it and will start accounting from there.
> 
> I don't really understand this part, though.  As in I don't even
> understand what you mean here..

The accounting for the vector gaps happens at the same time we consider
a bio for splitting. When we do split, the gap value applies only to the
front bio because that's as far as been processed. The remaining split
bio should reset its gap value to 0, and it will be recalculated when it
continues through the split consideration.

Re: [PATCH] block: save page offset gaps in cloned bio

[Christoph Hellwig]

On Wed, Aug 19, 2026 at 08:04:43AM -0600, Keith Busch wrote:
> > > The split remainder meanwhile can reset its gaps to 0
> > > since it advanced beyond it and will start accounting from there.
> > 
> > I don't really understand this part, though.  As in I don't even
> > understand what you mean here..
> 
> The accounting for the vector gaps happens at the same time we consider
> a bio for splitting. When we do split, the gap value applies only to the
> front bio because that's as far as been processed. The remaining split
> bio should reset its gap value to 0, and it will be recalculated when it
> continues through the split consideration.

Ḣ. but bio_split is used for a lot of different thing.  Then again the
bi_bvec_gap_bit is only used by the low-level blk-mq code, so it might
not matter.  But this for sure warrants a big fat comment explaining
the rationale for the clearing.