Re: [PATCH 4.4 33/35] nfit, libnvdimm: fix interleave set cookie calculation

[Dan Williams]

On Sun, Mar 19, 2017 at 9:38 AM, Ben Hutchings <ben@xxxxxxxxxxxxxxx> wrote:
> On Thu, 2017-03-16 at 23:29 +0900, Greg Kroah-Hartman wrote:
>> 4.4-stable review patch.  If anyone has any objections, please let me know.
>>
>> ------------------
>>
>> From: Dan Williams <dan.j.williams@xxxxxxxxx>
>>
>> commit 86ef58a4e35e8fa66afb5898cf6dec6a3bb29f67 upstream.
>>
>> The interleave-set cookie is a sum that sanity checks the composition of
>> an interleave set has not changed from when the namespace was initially
>> created.  The checksum is calculated by sorting the DIMMs by their
>> location in the interleave-set. The comparison for the sort must be
>> 64-bit wide, not byte-by-byte as performed by memcmp() in the broken
>> case.
> [...]
>> --- a/drivers/acpi/nfit.c
>> +++ b/drivers/acpi/nfit.c
> [...]
>> +static int cmp_map(const void *m0, const void *m1)
>> +{
>> +     const struct nfit_set_info_map *map0 = m0;
>> +     const struct nfit_set_info_map *map1 = m1;
>> +
>> +     return map0->region_offset - map1->region_offset;
>> +}
> [...]
>
> This is returning an int, thus it's effectively doing a 32-bit
> comparison and not the 64-bit comparison you say is needed.
>
> I think this function needs to do something like:
>
>         return (map0->region_offset < map1->region_offset) ? -1 :
>                 (map0->region_offset == map1->region_offset) ? 0 : 1;

Yes, you're right. We could end up with unexpected sign changes. Good catch.