On September 23, 2015 3:00:51 PM GMT+02:00, Richard Biener
wrote:
>*** copy_reference_ops_from_ref (tree ref, v
>*** 816,826
>--- 828,846
> temp.op0 = TREE_OPERAND (ref, 1);
> if (tree_fits_shwi_p (TREE_OPERAND (ref, 1)))
> temp.off = tree_to_shwi
I noticed we don't handle secondary effects of restrict in FRE when
looking at another testcase from PR48885:
int
f (int *__restrict__ &__restrict__ p, int *p2)
{
*p = 1;
*p2 = 2;
return *p;
}
with the previously posted patch to improve the handling for p2
we should be able to optimize the
