On Mon, Jan 13, 2020 at 3:31 PM Amir Goldstein <amir73il@xxxxxxxxx> wrote:
> OK, I understand why you don't like it and I think is makes sense to
> have an xino_state.
> However, xino_state and xino_bits are quite redundant.
>
> If we change:
> unsigned int xino_bits;
> to:
> int xino_mode;
>
> It can take the values:
> -1 /* not same dev */
> 0 /* same fs */
> 1..32 /* xino_bits */
>
> And:
>
> static inline unsigned int ovl_xino_bits(struct super_block *sb)
> {
> return OVL_FS(sb)->xino_mode > 0 : OVL_FS(sb)->xino_mode : 0;
> }
Okay.
Thanks,
Miklos
