Eric Schnoebelen wrote:
> Simon Burge says:
> 
> This looks like it happened in netinet/tcp_sack.c at:
> 
>         for (i = 0; i < num_sack_blks; i++, lp += 2) {
>                 memcpy(&left, lp, sizeof(*lp));
>                 memcpy(&right, lp + 1, sizeof(*lp));
> --->            left = ntohl(left);
>                 right = ntohl(right);

> I think that it looks like gcc is optimising the memcpy out and doing an
> unaligned load directly.  We probably need some sort of qualifier on a
> variable somewhere?

What about replacing memcpy() + ntohl() with endian-independent code? Does
this generate proper assembler code?

static inline u_int32_t
peek_be32(const void *p)
{
   const u_int8_t *u8 = p;
   return ((u_int32_t) p[0] << 24) |
	  ((u_int32_t) p[1] << 16) |
	  ((u_int32_t) p[2] << 8) |
	  ((u_int32_t) p[3]);
}

left = peek_be32(lp);
right = peek_be32(lp + 1);
 
-- 
Christian