linux-kernel - Re: [PATCH -next 3/8] raid5: fix missing io accounting in raid5_align

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 20 Jun 2023 16:52:35 +0800
From:   Xiao Ni <xni@...hat.com>
To:     Yu Kuai <yukuai1@...weicloud.com>
Cc:     song@...nel.org, linux-raid@...r.kernel.org,
        linux-kernel@...r.kernel.org, yukuai3@...wei.com,
        yi.zhang@...wei.com, yangerkun@...wei.com
Subject: Re: [PATCH -next 3/8] raid5: fix missing io accounting in raid5_align_endio()

On Mon, Jun 19, 2023 at 8:50 PM Yu Kuai <yukuai1@...weicloud.com> wrote:
>
> From: Yu Kuai <yukuai3@...wei.com>
>
> Io will only be accounted as done from raid5_align_endio() if the io
> succeed, and io inflight counter will be leaked if such io failed.
>
> Fix this problem by switching to use md_account_bio() for io accounting.
>
> Signed-off-by: Yu Kuai <yukuai3@...wei.com>
> ---
>  drivers/md/raid5.c | 29 ++++++++---------------------
>  1 file changed, 8 insertions(+), 21 deletions(-)
>
> diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
> index cef0b400b2ee..4cdb35e54251 100644
> --- a/drivers/md/raid5.c
> +++ b/drivers/md/raid5.c
> @@ -5468,26 +5468,17 @@ static struct bio *remove_bio_from_retry(struct r5conf *conf,
>   */
>  static void raid5_align_endio(struct bio *bi)
>  {
> -       struct md_io_clone *md_io_clone = bi->bi_private;
> -       struct bio *raid_bi = md_io_clone->orig_bio;
> -       struct mddev *mddev;
> -       struct r5conf *conf;
> -       struct md_rdev *rdev;
> +       struct bio *raid_bi = bi->bi_private;
> +       struct md_rdev *rdev = (void *)raid_bi->bi_next;
> +       struct mddev *mddev = rdev->mddev;
> +       struct r5conf *conf = mddev->private;
>         blk_status_t error = bi->bi_status;
> -       unsigned long start_time = md_io_clone->start_time;
>
>         bio_put(bi);
> -
> -       rdev = (void*)raid_bi->bi_next;
>         raid_bi->bi_next = NULL;
> -       mddev = rdev->mddev;
> -       conf = mddev->private;
> -
>         rdev_dec_pending(rdev, conf->mddev);
>
>         if (!error) {
> -               if (blk_queue_io_stat(raid_bi->bi_bdev->bd_disk->queue))
> -                       bio_end_io_acct(raid_bi, start_time);
>                 bio_endio(raid_bi);
>                 if (atomic_dec_and_test(&conf->active_aligned_reads))
>                         wake_up(&conf->wait_for_quiescent);
> @@ -5506,7 +5497,6 @@ static int raid5_read_one_chunk(struct mddev *mddev, struct bio *raid_bio)
>         struct md_rdev *rdev;
>         sector_t sector, end_sector, first_bad;
>         int bad_sectors, dd_idx;
> -       struct md_io_clone *md_io_clone;
>         bool did_inc;
>
>         if (!in_chunk_boundary(mddev, raid_bio)) {
> @@ -5543,16 +5533,13 @@ static int raid5_read_one_chunk(struct mddev *mddev, struct bio *raid_bio)
>                 return 0;
>         }
>
> -       align_bio = bio_alloc_clone(rdev->bdev, raid_bio, GFP_NOIO,
> -                                   &mddev->io_clone_set);
> -       md_io_clone = container_of(align_bio, struct md_io_clone, bio_clone);
> +       md_account_bio(mddev, &raid_bio);
>         raid_bio->bi_next = (void *)rdev;
> -       if (blk_queue_io_stat(raid_bio->bi_bdev->bd_disk->queue))
> -               md_io_clone->start_time = bio_start_io_acct(raid_bio);
> -       md_io_clone->orig_bio = raid_bio;
>
> +       align_bio = bio_alloc_clone(rdev->bdev, raid_bio, GFP_NOIO,
> +                                   &mddev->bio_set);
>         align_bio->bi_end_io = raid5_align_endio;
> -       align_bio->bi_private = md_io_clone;
> +       align_bio->bi_private = raid_bio;
>         align_bio->bi_iter.bi_sector = sector;
>
>         /* No reshape active, so we can trust rdev->data_offset */
> --
> 2.39.2
>

Reviewed-by: Xiao Ni <xni@...hat.com>