https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126705
--- Comment #2 from Drea Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Drea Pinski from comment #1)
> This is an example of what something which is not optimized on aarch64:
> ```
> int f12(unsigned t)
> {
> unsigned t1 = t *3;
> if (t1)
> return __builtin_ctz(t*3);
> return 32;
> }
> ```
```
int f(unsigned t)
{
t *= 3;
if (t) // gets optimized to using the original t.
return __builtin_ctz(t);
return 32;
}
```
