https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114741
Wilco <wilco at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |wilco at gcc dot gnu.org
--- Comment #1 from Wilco <wilco at gcc dot gnu.org> ---
This example always goes wrong:
void foo2(unsigned *p)
{
*p &= 1;
}
Eg. with -mcpu=neoverse-v1:
ldr s31, [x0]
and z31.s, z31.s, #1
str s31, [x0]
ret
This doesn't make any sense since there are usually fewer vector units than
integer ALUs, and the typically have higher latency.
