* [PATCH] block: Correct comment on bio_alloc_clone()
@ 2026-02-20 14:32 John Garry
2026-02-20 15:10 ` Christoph Hellwig
0 siblings, 1 reply; 6+ messages in thread
From: John Garry @ 2026-02-20 14:32 UTC (permalink / raw)
To: axboe; +Cc: linux-block, hch, John Garry
Correct the comment that the source bio must not be freed before the clone.
Signed-off-by: John Garry <john.g.garry@oracle.com>
diff --git a/block/bio.c b/block/bio.c
index 8203bb7455a9f..9d9191fe19b50 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -900,7 +900,7 @@ static int __bio_clone(struct bio *bio, struct bio *bio_src, gfp_t gfp)
* Allocate a new bio that is a clone of @bio_src. The caller owns the returned
* bio, but not the actual data it points to.
*
- * The caller must ensure that the return bio is not freed before @bio_src.
+ * The caller must ensure that @bio_src is not freed before the returned bio.
*/
struct bio *bio_alloc_clone(struct block_device *bdev, struct bio *bio_src,
gfp_t gfp, struct bio_set *bs)
--
2.43.5
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] block: Correct comment on bio_alloc_clone()
2026-02-20 14:32 [PATCH] block: Correct comment on bio_alloc_clone() John Garry
@ 2026-02-20 15:10 ` Christoph Hellwig
2026-02-20 15:36 ` John Garry
0 siblings, 1 reply; 6+ messages in thread
From: Christoph Hellwig @ 2026-02-20 15:10 UTC (permalink / raw)
To: John Garry; +Cc: axboe, linux-block, hch
On Fri, Feb 20, 2026 at 02:32:33PM +0000, John Garry wrote:
> Correct the comment that the source bio must not be freed before the clone.
Note that while this version is better, it's not actually true either.
The bi_io_vec used by the clone must be freed after the clone, but it
might might not be owned by the passed in bio_src.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] block: Correct comment on bio_alloc_clone()
2026-02-20 15:10 ` Christoph Hellwig
@ 2026-02-20 15:36 ` John Garry
2026-02-20 15:43 ` Christoph Hellwig
0 siblings, 1 reply; 6+ messages in thread
From: John Garry @ 2026-02-20 15:36 UTC (permalink / raw)
To: Christoph Hellwig; +Cc: axboe, linux-block
On 20/02/2026 15:10, Christoph Hellwig wrote:
> On Fri, Feb 20, 2026 at 02:32:33PM +0000, John Garry wrote:
>> Correct the comment that the source bio must not be freed before the clone.
>
> Note that while this version is better, it's not actually true either.
> The bi_io_vec used by the clone must be freed after the clone, but it
> might might not be owned by the passed in bio_src.
>
This is the complete function comment today:
* Allocate a new bio that is a clone of @bio_src. The caller owns the
returned
* bio, but not the actual data it points to.
*
* The caller must ensure that the return bio is not freed before @bio_src.
So I can have this:
* Allocate a new bio that is a clone of @bio_src. The caller owns the
returned
* bio, but not the actual data it points to, bi_io_vec.
*
* The caller must ensure that the returned bio is freed before bi_io_vec.
ok?
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] block: Correct comment on bio_alloc_clone()
2026-02-20 15:36 ` John Garry
@ 2026-02-20 15:43 ` Christoph Hellwig
2026-02-20 15:55 ` John Garry
0 siblings, 1 reply; 6+ messages in thread
From: Christoph Hellwig @ 2026-02-20 15:43 UTC (permalink / raw)
To: John Garry; +Cc: Christoph Hellwig, axboe, linux-block
On Fri, Feb 20, 2026 at 03:36:41PM +0000, John Garry wrote:
>
> So I can have this:
>
> * Allocate a new bio that is a clone of @bio_src. The caller owns the
> returned
> * bio, but not the actual data it points to, bi_io_vec.
Well, bi_io_vec isn't really the data, but the indirection pointing
to the data. I guess we need to come up with consistent terminology
first before we can make good sense of this.
Allocate a new bio that is a clone of @bio_src. This reuses the bio_vecs
pointed to by @bio_src->bi_io_vec, and clones the iterator pointing to
the current position in it. The caller owns the returned bio, but not
the bio_vecs, and must ensure the bio is freed before the memory
pointed to by @bio_Src->bi_io_vecs.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] block: Correct comment on bio_alloc_clone()
2026-02-20 15:43 ` Christoph Hellwig
@ 2026-02-20 15:55 ` John Garry
2026-02-20 15:57 ` Christoph Hellwig
0 siblings, 1 reply; 6+ messages in thread
From: John Garry @ 2026-02-20 15:55 UTC (permalink / raw)
To: Christoph Hellwig; +Cc: axboe, linux-block
On 20/02/2026 15:43, Christoph Hellwig wrote:
> Well, bi_io_vec isn't really the data, but the indirection pointing
> to the data. I guess we need to come up with consistent terminology
> first before we can make good sense of this.
>
> Allocate a new bio that is a clone of @bio_src. This reuses the bio_vecs
> pointed to by @bio_src->bi_io_vec, and clones the iterator pointing to
> the current position in it. The caller owns the returned bio, but not
> the bio_vecs, and must ensure the bio is freed before the memory
> pointed to by @bio_Src->bi_io_vecs.
ok, fine.
BTW, the comment for bio_init_clone() was much the same as
bio_alloc_clone(), modulo what I tried to originally remedy. I can just
mention that that the same bio_src->bi_io_vec lifecycle rules apply
there (as bio_alloc_clone())
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] block: Correct comment on bio_alloc_clone()
2026-02-20 15:55 ` John Garry
@ 2026-02-20 15:57 ` Christoph Hellwig
0 siblings, 0 replies; 6+ messages in thread
From: Christoph Hellwig @ 2026-02-20 15:57 UTC (permalink / raw)
To: John Garry; +Cc: Christoph Hellwig, axboe, linux-block
On Fri, Feb 20, 2026 at 03:55:41PM +0000, John Garry wrote:
> BTW, the comment for bio_init_clone() was much the same as
> bio_alloc_clone(), modulo what I tried to originally remedy. I can just
> mention that that the same bio_src->bi_io_vec lifecycle rules apply there
> (as bio_alloc_clone())
Sounds good.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2026-02-20 15:57 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-02-20 14:32 [PATCH] block: Correct comment on bio_alloc_clone() John Garry
2026-02-20 15:10 ` Christoph Hellwig
2026-02-20 15:36 ` John Garry
2026-02-20 15:43 ` Christoph Hellwig
2026-02-20 15:55 ` John Garry
2026-02-20 15:57 ` Christoph Hellwig
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox