Re: [PATCH] find: Do not read beyond variable boundaries on small sizes

[Yury Norov]

On Fri, Dec 03, 2021 at 02:30:35PM +0200, Andy Shevchenko wrote:
> On Fri, Dec 03, 2021 at 02:08:46AM -0800, Kees Cook wrote:
> > It's common practice to cast small variable arguments to the find_*_bit()

Not that common - I found 19 examples of this cast, and most of them
are in drivers. The only non-driver case is kernel/trace/pid_list.c:

        static inline bool upper_empty(union upper_chunk *chunk)
        {
                /*
                 * If chunk->data has no lower chunks, it will be the same
                 * as a zeroed bitmask. Use find_first_bit() to test it
                 * and if it doesn't find any bits set, then the array
                 * is empty.
                 */
                int bit = find_first_bit((unsigned long *)chunk->data,
                                         sizeof(chunk->data) * 8);
                return bit >= sizeof(chunk->data) * 8;
        }

And it's OK wrt type conversion because chunk->data is:

        union lower_chunk {
                union lower_chunk               *next;
                unsigned long                   data[LOWER_SIZE]; // 2K in size
        };

Although, this function should use bitmap_empty(), probably like this:

        static inline bool upper_empty(union upper_chunk *chunk)
        {
                return bitmap_empty(chunk->data->data[0], BITS_PER_LONG);
        }

Steven, can you comment on this?

> It's a bad practice and should be fixed accordingly, no?

Yes.

> > helpers to unsigned long and then use a size argument smaller than
> > sizeof(unsigned long):
> > 
> > 	unsigned int bits;
> > 	...
> > 	out = find_first_bit((unsigned long *)&bits, 32);
> > 
> > This leads to the find helper dereferencing a full unsigned long,
> > regardless of the size of the actual variable. The unwanted bits
> > get masked away, but strictly speaking, a read beyond the end of
> > the target variable happens. Builds under -Warray-bounds complain
> > about this situation, for example:
> > 
> > In file included from ./include/linux/bitmap.h:9,
> >                  from drivers/iommu/intel/iommu.c:17:
> > drivers/iommu/intel/iommu.c: In function 'domain_context_mapping_one':
> > ./include/linux/find.h:119:37: error: array subscript 'long unsigned int[0]' is partly outside array bounds of 'int[1]' [-Werror=array-bounds]
> >   119 |                 unsigned long val = *addr & GENMASK(size - 1, 0);
> >       |                                     ^~~~~
> > drivers/iommu/intel/iommu.c:2115:18: note: while referencing 'max_pde'
> >  2115 |         int pds, max_pde;
> >       |                  ^~~~~~~

The driver should be fixed. I would suggest using one of ffs/fls/ffz from
include/asm/bitops.h

Thanks,
Yury

> > Instead, just carefully read the correct variable size, all of which
> > happens at compile time since small_const_nbits(size) has already
> > determined that arguments are constant expressions.
> 
> What is the performance impact?
> 
> -- 
> With Best Regards,
> Andy Shevchenko
>