From: Jeff King <peff@peff.net>
Date: Wed, 22 Aug 2018 16:59:23 GMT
Subject: Re: [ANNOUNCE] Git v2.19.0-rc0
Message-ID: <20180822165923.GA14878@sigill.intra.peff.net>
In-Reply-To: <7ea416cf-b043-1274-e161-85a8780b8e1c@gmail.com>

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.

-Peff