Re: [PATCH v3 34/48] tcg/optimize: Split out fold_to_not
On 10/25/21 11:13 AM, Luis Fernando Fujita Pires wrote:
From: Richard Henderson
static bool fold_eqv(OptContext *ctx, TCGOp *op) {
-return fold_const2(ctx, op);
+if (fold_const2(ctx, op) ||
+fold_xi_to_not(ctx, op, 0)) {
Should be fold_ix_to_not (not fold xi_to_not).
No, because for eqv we expect the second operand to be the constant -- eqv is
commutative.
Ah, got it! The previous code was wrong, and I failed to notice that eqv
would've had its arguments swapped to have the constant as second.
Ah! I failed to notice that the previous code was wrong. ;-)
r~
RE: [PATCH v3 34/48] tcg/optimize: Split out fold_to_not
From: Richard Henderson
> >> static bool fold_eqv(OptContext *ctx, TCGOp *op) {
> >> -return fold_const2(ctx, op);
> >> +if (fold_const2(ctx, op) ||
> >> +fold_xi_to_not(ctx, op, 0)) {
> >
> > Should be fold_ix_to_not (not fold xi_to_not).
>
> No, because for eqv we expect the second operand to be the constant -- eqv is
> commutative.
Ah, got it! The previous code was wrong, and I failed to notice that eqv
would've had its arguments swapped to have the constant as second.
--
Luis Pires
Instituto de Pesquisas ELDORADO
Aviso Legal - Disclaimer
Re: [PATCH v3 34/48] tcg/optimize: Split out fold_to_not
On 10/25/21 7:17 AM, Luis Fernando Fujita Pires wrote:
From: Richard Henderson
Split out the conditional conversion from a more complex logical operation to a
simple NOT. Create a couple more helpers to make this easy for the outer-most
logical operations.
Signed-off-by: Richard Henderson
---
tcg/optimize.c | 154 +++--
1 file changed, 86 insertions(+), 68 deletions(-)
static bool fold_eqv(OptContext *ctx, TCGOp *op) {
-return fold_const2(ctx, op);
+if (fold_const2(ctx, op) ||
+fold_xi_to_not(ctx, op, 0)) {
Should be fold_ix_to_not (not fold xi_to_not).
No, because for eqv we expect the second operand to be the constant -- eqv is
commutative.
static bool fold_orc(OptContext *ctx, TCGOp *op) {
-return fold_const2(ctx, op);
+if (fold_const2(ctx, op) ||
+fold_xi_to_not(ctx, op, 0)) {
But for orc you are correct. Thanks.
r~
RE: [PATCH v3 34/48] tcg/optimize: Split out fold_to_not
From: Richard Henderson
> Split out the conditional conversion from a more complex logical operation to
> a
> simple NOT. Create a couple more helpers to make this easy for the outer-most
> logical operations.
>
> Signed-off-by: Richard Henderson
> ---
> tcg/optimize.c | 154 +++--
> 1 file changed, 86 insertions(+), 68 deletions(-)
> static bool fold_eqv(OptContext *ctx, TCGOp *op) {
> -return fold_const2(ctx, op);
> +if (fold_const2(ctx, op) ||
> +fold_xi_to_not(ctx, op, 0)) {
Should be fold_ix_to_not (not fold xi_to_not).
> static bool fold_orc(OptContext *ctx, TCGOp *op) {
> -return fold_const2(ctx, op);
> +if (fold_const2(ctx, op) ||
> +fold_xi_to_not(ctx, op, 0)) {
Same here.
--
Luis Pires
Instituto de Pesquisas ELDORADO
Aviso Legal - Disclaimer
