On Tue, Jun 08, 2021 at 01:56:33PM -0700, Andrew Morton wrote:
> On Tue, 8 Jun 2021 11:39:55 -0700 Kees Cook <keescook@xxxxxxxxxxxx> wrote:
>
> > It turns out that SLUB redzoning ("slub_debug=Z") checks from
> > s->object_size rather than from s->inuse (which is normally bumped
> > to make room for the freelist pointer), so a cache created with an
> > object size less than 24 would have the freelist pointer written beyond
> > s->object_size, causing the redzone to be corrupted by the freelist
> > pointer. This was very visible with "slub_debug=ZF":
> >
> > BUG test (Tainted: G B ): Right Redzone overwritten
> > -----------------------------------------------------------------------------
> >
> > INFO: 0xffff957ead1c05de-0xffff957ead1c05df @offset=1502. First byte 0x1a instead of 0xbb
> > INFO: Slab 0xffffef3950b47000 objects=170 used=170 fp=0x0000000000000000 flags=0x8000000000000200
> > INFO: Object 0xffff957ead1c05d8 @offset=1496 fp=0xffff957ead1c0620
> >
> > Redzone (____ptrval____): bb bb bb bb bb bb bb bb ........
> > Object (____ptrval____): 00 00 00 00 00 f6 f4 a5 ........
> > Redzone (____ptrval____): 40 1d e8 1a aa @....
> > Padding (____ptrval____): 00 00 00 00 00 00 00 00 ........
> >
> > Adjust the offset to stay within s->object_size.
> >
> > (Note that no caches of in this size range are known to exist in the
> > kernel currently.)
>
> We already have
> https://lkml.kernel.org/r/6746FEEA-FD69-4792-8DDA-C78F5FE7DA02@xxxxxxx.
> Is this patch better?
Yes, I believe so, since it reduces code and corrects the size checking
more directly (and more clearly demonstrates the redzone calculation
problem in the commit log).
-Kees
>
> > --- a/mm/slub.c
> > +++ b/mm/slub.c
> > @@ -3689,7 +3689,6 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
> > {
> > slab_flags_t flags = s->flags;
> > unsigned int size = s->object_size;
> > - unsigned int freepointer_area;
> > unsigned int order;
> >
> > /*
> > @@ -3698,13 +3697,6 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
> > * the possible location of the free pointer.
> > */
> > size = ALIGN(size, sizeof(void *));
> > - /*
> > - * This is the area of the object where a freepointer can be
> > - * safely written. If redzoning adds more to the inuse size, we
> > - * can't use that portion for writing the freepointer, so
> > - * s->offset must be limited within this for the general case.
> > - */
> > - freepointer_area = size;
> >
> > #ifdef CONFIG_SLUB_DEBUG
> > /*
> > @@ -3730,7 +3722,7 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
> >
> > /*
> > * With that we have determined the number of bytes in actual use
> > - * by the object. This is the potential offset to the free pointer.
> > + * by the object and redzoning.
> > */
> > s->inuse = size;
> >
> > @@ -3753,13 +3745,13 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
> > */
> > s->offset = size;
> > size += sizeof(void *);
> > - } else if (freepointer_area > sizeof(void *)) {
> > + } else {
> > /*
> > * Store freelist pointer near middle of object to keep
> > * it away from the edges of the object to avoid small
> > * sized over/underflows from neighboring allocations.
> > */
> > - s->offset = ALIGN(freepointer_area / 2, sizeof(void *));
> > + s->offset = ALIGN_DOWN(s->object_size / 2, sizeof(void *));
> > }
> >
> > #ifdef CONFIG_SLUB_DEBUG
> > --
> > 2.25.1
--
Kees Cook
