To illustrate in which scenario code in tree-ssa-forwprop doesn't help
is binop-tor4.c
w/o this patch we get
foo (int a, int b, int c)
{
int e;
int d;
int D.2701;
_Bool D.2700;
_Bool D.2699;
_Bool D.2698;
_Bool D.2697;
_Bool D.2696;
int D.2695;
<bb 2>:
D.2695_3 = b_2(D) | a_1(D);
d_4 = D.2695_3 != 0;
D.2696_5 = a_1(D) == 0;
D.2697_6 = b_2(D) == 0;
D.2698_7 = D.2697_6 | D.2696_5;
D.2699_9 = c_8(D) != 0;
D.2700_10 = D.2698_7 | D.2699_9;
e_11 = (int) D.2700_10;
D.2701_12 = e_11 | d_4;
return D.2701_12;
}
Of interest is here D.2701_12, which doesn't have a type sinking.
This is caused by
D.2695_3 = b_2(D) | a_1(D);
d_4 = D.2695_3 != 0;
which is a comparison result with implicit integer cast. So maybe the
solution here could be to first doing boolification of comparison in
gimplifier. By this, the code for type-sinking in my patch could go
away.
Regards,
Kai
