[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <8d5a0a41-0112-4c53-a7c4-67bfc5332ba9@kernel.dk>
Date: Tue, 3 Sep 2024 09:49:30 -0600
From: Jens Axboe <axboe@...nel.dk>
To: Luca Stefani <luca.stefani.ge1@...il.com>
Cc: linux-block@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] block: Export blk_alloc_discard_bio
On 9/3/24 1:39 AM, Luca Stefani wrote:
> The fs trim loops over ranges and sends discard requests, some ranges
> can be large so it's all transparently handled by blkdev_issue_discard()
> and processed in smaller chunks.
>
> To support cancellation (or suspend) requests we need to insert checks
> into the the loop, exporting the symbol allows to reimplement
> such loop with the desired behavior.
>
> Signed-off-by: Luca Stefani <luca.stefani.ge1@...il.com>
> ---
> block/blk-lib.c | 1 +
> include/linux/blkdev.h | 2 ++
> 2 files changed, 3 insertions(+)
>
> diff --git a/block/blk-lib.c b/block/blk-lib.c
> index 4c9f20a689f7..ebaef47d8ce7 100644
> --- a/block/blk-lib.c
> +++ b/block/blk-lib.c
> @@ -59,6 +59,7 @@ struct bio *blk_alloc_discard_bio(struct block_device *bdev,
> cond_resched();
> return bio;
> }
> +EXPORT_SYMBOL_GPL(blk_alloc_discard_bio);
>
> int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
> sector_t nr_sects, gfp_t gfp_mask, struct bio **biop)
> diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
> index b7664d593486..f3631044d905 100644
> --- a/include/linux/blkdev.h
> +++ b/include/linux/blkdev.h
> @@ -1088,6 +1088,8 @@ static inline long nr_blockdev_pages(void)
>
> extern void blk_io_schedule(void);
>
> +struct bio *blk_alloc_discard_bio(struct block_device *bdev,
> + sector_t *sector, sector_t *nr_sects, gfp_t gfp_mask);
> int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
> sector_t nr_sects, gfp_t gfp_mask);
> int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
Since blk_alloc_discard_bio() is already defined in a header (otherwise
it would've been static and your export symbol above would have failed
miserably), why add it to another header?
--
Jens Axboe
Powered by blists - more mailing lists