Jeff King <peff@xxxxxxxx> writes:
> On Wed, Aug 22, 2018 at 12:49:34PM -0400, Derrick Stolee wrote:
>
>> > Yes, that was what I meant. We actually did switch to that hand-rolled
>> > loop, but later we went back to memcmp in 0b006014c8 (hashcmp: use
>> > memcmp instead of open-coded loop, 2017-08-09).
>>
>> Looking at that commit, I'm surprised the old logic was just a for
>> loop, instead of a word-based approach, such as the following:
>> [...]
>> +struct object_id_20 {
>> + uint64_t data0;
>> + uint64_t data1;
>> + uint32_t data2;
>> +};
>> +
>> static inline int hashcmp(const unsigned char *sha1, const unsigned char
>> *sha2)
>> {
>> - return memcmp(sha1, sha2, the_hash_algo->rawsz);
>> + if (the_hash_algo->rawsz == 20) {
>> + struct object_id_20 *obj1 = (struct object_id_20 *)sha1;
>> + struct object_id_20 *obj2 = (struct object_id_20 *)sha2;
>
> I wonder if you're potentially running afoul of alignment requirements
> here.
Yup, and I think that all was discussed in that old thread ;-)
