On 11.09.19 16:54, Richard Henderson wrote:
> On 9/6/19 3:57 AM, David Hildenbrand wrote:
>> if (dest == src + 1) {
>> fast_memset(env, dest, cpu_ldub_data_ra(env, src, ra), l, ra);
>> - } else if (dest < src || src + l <= dest) {
>> + } else if (!is_destructive_overlap(env, dest, src, l)) {
>> fast_memmove(env, dest, src, l, ra);
>> } else {
>> for (i = 0; i < l; i++) {
>
> I suppose, though last time I checked fast_memmove didn't support wrapping.
Yes, that's fixed by access_prepare(), access_memmove(). Wrapping is/was
broken in most mem handlers ...
>
> Reviewed-by: Richard Henderson <richard.hender...@linaro.org>
>
> r~
>
--
Thanks,
David / dhildenb
