[PATCH v4 4/5] md/raid1: Atomic write support

John Garry posted 5 patches 1 week, 4 days ago
There is a newer version of this series
[PATCH v4 4/5] md/raid1: Atomic write support
Posted by John Garry 1 week, 4 days ago
Set BLK_FEAT_ATOMIC_WRITES_STACKED to enable atomic writes.

For an attempt to atomic write to a region which has bad blocks, error
the write as we just cannot do this. It is unlikely to find devices which
support atomic writes and bad blocks.

Signed-off-by: John Garry <john.g.garry@oracle.com>
---
 drivers/md/raid1.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index a5adf08ee174..cd44b4bebf49 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1571,7 +1571,15 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio,
 				continue;
 			}
 			if (is_bad) {
-				int good_sectors = first_bad - r1_bio->sector;
+				int good_sectors;
+
+				if (bio->bi_opf & REQ_ATOMIC) {
+					/* We just cannot atomically write this ... */
+					error = -EFAULT;
+					goto err_handle;
+				}
+
+				good_sectors = first_bad - r1_bio->sector;
 				if (good_sectors < max_sectors)
 					max_sectors = good_sectors;
 			}
@@ -1657,7 +1665,8 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio,
 
 		mbio->bi_iter.bi_sector	= (r1_bio->sector + rdev->data_offset);
 		mbio->bi_end_io	= raid1_end_write_request;
-		mbio->bi_opf = bio_op(bio) | (bio->bi_opf & (REQ_SYNC | REQ_FUA));
+		mbio->bi_opf = bio_op(bio) |
+			(bio->bi_opf & (REQ_SYNC | REQ_FUA | REQ_ATOMIC));
 		if (test_bit(FailFast, &rdev->flags) &&
 		    !test_bit(WriteMostly, &rdev->flags) &&
 		    conf->raid_disks - mddev->degraded > 1)
@@ -3224,6 +3233,7 @@ static int raid1_set_limits(struct mddev *mddev)
 
 	md_init_stacking_limits(&lim);
 	lim.max_write_zeroes_sectors = 0;
+	lim.features |= BLK_FEAT_ATOMIC_WRITES_STACKED;
 	err = mddev_stack_rdev_limits(mddev, &lim, MDDEV_STACK_INTEGRITY);
 	if (err) {
 		queue_limits_cancel_update(mddev->gendisk->queue);
-- 
2.31.1
Re: [PATCH v4 4/5] md/raid1: Atomic write support
Posted by Yu Kuai 1 week ago
在 2024/11/12 20:42, John Garry 写道:
> Set BLK_FEAT_ATOMIC_WRITES_STACKED to enable atomic writes.
> 
> For an attempt to atomic write to a region which has bad blocks, error
> the write as we just cannot do this. It is unlikely to find devices which
> support atomic writes and bad blocks.
> 
> Signed-off-by: John Garry <john.g.garry@oracle.com>
> ---
>   drivers/md/raid1.c | 14 ++++++++++++--
>   1 file changed, 12 insertions(+), 2 deletions(-)
> 

I review the patch 5 first, it's the same, so.

Reviewed-by: Yu Kuai <yukuai3@huawei.com>

> diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
> index a5adf08ee174..cd44b4bebf49 100644
> --- a/drivers/md/raid1.c
> +++ b/drivers/md/raid1.c
> @@ -1571,7 +1571,15 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio,
>   				continue;
>   			}
>   			if (is_bad) {
> -				int good_sectors = first_bad - r1_bio->sector;
> +				int good_sectors;
> +
> +				if (bio->bi_opf & REQ_ATOMIC) {
> +					/* We just cannot atomically write this ... */
> +					error = -EFAULT;
> +					goto err_handle;
> +				}
> +
> +				good_sectors = first_bad - r1_bio->sector;
>   				if (good_sectors < max_sectors)
>   					max_sectors = good_sectors;
>   			}
> @@ -1657,7 +1665,8 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio,
>   
>   		mbio->bi_iter.bi_sector	= (r1_bio->sector + rdev->data_offset);
>   		mbio->bi_end_io	= raid1_end_write_request;
> -		mbio->bi_opf = bio_op(bio) | (bio->bi_opf & (REQ_SYNC | REQ_FUA));
> +		mbio->bi_opf = bio_op(bio) |
> +			(bio->bi_opf & (REQ_SYNC | REQ_FUA | REQ_ATOMIC));
>   		if (test_bit(FailFast, &rdev->flags) &&
>   		    !test_bit(WriteMostly, &rdev->flags) &&
>   		    conf->raid_disks - mddev->degraded > 1)
> @@ -3224,6 +3233,7 @@ static int raid1_set_limits(struct mddev *mddev)
>   
>   	md_init_stacking_limits(&lim);
>   	lim.max_write_zeroes_sectors = 0;
> +	lim.features |= BLK_FEAT_ATOMIC_WRITES_STACKED;
>   	err = mddev_stack_rdev_limits(mddev, &lim, MDDEV_STACK_INTEGRITY);
>   	if (err) {
>   		queue_limits_cancel_update(mddev->gendisk->queue);
>