Re: [PATCH 2/2] md/raid1: move the exceed_read_errors condition out of fix_read_error

On Fri, May 29, 2026 at 1:45 PM Christoph Hellwig <[email protected]> wrote:
>
> This condition much better fits into the only caller, limiting
> fix_read_error to actually fix up data devices after a read error.
>
> Signed-off-by: Christoph Hellwig <[email protected]>
> ---
>  drivers/md/raid1.c | 10 ++++------
>  1 file changed, 4 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
> index 8fad1692cf66..e510ad7eef32 100644
> --- a/drivers/md/raid1.c
> +++ b/drivers/md/raid1.c
> @@ -2411,11 +2411,6 @@ static void fix_read_error(struct r1conf *conf, struct r1bio *r1_bio)
>         struct mddev *mddev = conf->mddev;
>         struct md_rdev *rdev = conf->mirrors[read_disk].rdev;
>
> -       if (exceed_read_errors(mddev, rdev)) {
> -               r1_bio->bios[r1_bio->read_disk] = IO_BLOCKED;
> -               return;
> -       }
> -
>         while(sectors) {
>                 int s = sectors;
>                 int d = read_disk;
> @@ -2652,7 +2647,10 @@ static void handle_read_error(struct r1conf *conf, struct r1bio *r1_bio)
>                 md_error(mddev, rdev);
>         } else {
>                 freeze_array(conf, 1);
> -               fix_read_error(conf, r1_bio);
> +               if (exceed_read_errors(mddev, rdev))
> +                       r1_bio->bios[r1_bio->read_disk] = IO_BLOCKED;
> +               else
> +                       fix_read_error(conf, r1_bio);
>                 unfreeze_array(conf);
>         }
>
> --
> 2.53.0
>
>
This patch looks good to me.
Reviewed-by: Xiao Ni <[email protected]>