Re: [PATCH v3] dm cache: parse invalidate_cblocks with kstrtouint()

[Ming Hung Tsai]

On Sun, Jul 12, 2026 at 5:39 AM Samuel Moelius
<sam.moelius@xxxxxxxxxxxxxxx> wrote:
>
> +       if (str) {
> +               blocknr = str;
> +
> +               r = kstrtouint(blocknr, 10, &e);
> +               if (r)
> +                       goto bad;
>
> -       if (r == 1) {
> -               result->begin = to_cblock(b);
> -               result->end = to_cblock(from_cblock(result->begin) + 1u);
> +               result->end = to_cblock(e);
>                 return 0;
>         }
>
> -       DMERR("%s: invalid cblock range '%s'", cache_device_name(cache), str);
> +       result->end = cblock_succ(result->begin);
> +       return 0;

A minor nit: the range vs. single-value cases are peer alternatives.
Rather than multiple "returns 0", using if-else makes the parallel
structure more explicit: both paths share the same prologue (parse
result->begin) and epilogue (return 0), differ only in how result->end
is computed.