Re: [PATCH] fs: btrfs: fix potential overflow

[Brian Norris <computersforpeace@gmail.com>]

On Sun, Aug 24, 2014 at 12:44:11PM +0300, Timofey Titovets wrote:
> 2014-08-24 8:41 GMT+03:00 Brian Norris <computeaboversforpeace@gmail.com>:
> > --- a/fs/btrfs/scrub.c
> > +++ b/fs/btrfs/scrub.c
> > @@ -2320,26 +2320,26 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
> >         do_div(nstripes, map->stripe_len);
> >         if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
> >                 offset = map->stripe_len * num;
> > -               increment = map->stripe_len * map->num_stripes;
> > +               increment *= map->num_stripes;
> >                 mirror_num = 1;
> >         } else if (map->type & BTRFS_BLOCK_GROUP_RAID10) {
> >                 int factor = map->num_stripes / map->sub_stripes;
> >                 offset = map->stripe_len * (num / map->sub_stripes);
> > -               increment = map->stripe_len * factor;
> > +               increment *= factor;
> >                 mirror_num = num % map->sub_stripes + 1;
> >         } else if (map->type & BTRFS_BLOCK_GROUP_RAID1) {
> > -               increment = map->stripe_len;
> > +               increment *= 1;
> Why? I think multiple by one is useless or i miss something?

A misguided attempt to keep the code structure intact. I'll drop these
for v2.

[snip]

Thanks,
Brian