high_memory address in /proc/dri/*/vma

[Kees Cook]

On Mon, Dec 19, 2011 at 6:18 PM, Ben Hutchings <ben at decadent.org.uk> wrote:
> On Mon, 2011-12-19 at 16:14 -0800, Kees Cook wrote:
>> On Mon, Dec 19, 2011 at 4:09 PM, Ben Hutchings <ben at decadent.org.uk> wrote:
>> > Kees, in commit 01e2f533a234dc62d16c0d3d4fb9d71cf1ce50c3 ("drm: do not
>> > leak kernel addresses via /proc/dri/*/vma") you changed the logging of
>> > high_memory:
>> >
>> > -       seq_printf(m, "vma use count: %d, high_memory = %p, 0x%08llx\n",
>> > +       seq_printf(m, "vma use count: %d, high_memory = %pK, 0x%pK\n",
>> >                   atomic_read(&dev->vma_count),
>> > -                  high_memory, (u64)virt_to_phys(high_memory));
>> > +                  high_memory, (void *)virt_to_phys(high_memory));
>> >
>> > This doesn't make sense because the physical address may be truncated
>> > (in theory at least).
>>
>> Leaking even a truncated address is still a problem, IMO. Or do you
>> mean there is some side-effect causing a problem?
>
> I mean that this the conversion to void * can be a narrowing conversion,
> so when printing of kernel pointers is enabled the full physical address
> may not be displayed.

Ah-ha, okay, I see what you mean now. This is, as you say, only a
problem "in theory". Is it worth fixing currently? If so, we probably
need to add the "K" option to %x in some fashion.

-Kees

-- 
Kees Cook
ChromeOS Security