On Fri, 18 Jul 2025 at 18:46, Richard Henderson
<richard.hender...@linaro.org> wrote:
>
> There is no such thing as vector extract.
>
> Fixes: 932522a9ddc1 ("tcg/optimize: Fold and to extract during optimize")
> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/3036
> Signed-off-by: Richard Henderson <richard.hender...@linaro.org>
> ---
> tcg/optimize.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tcg/optimize.c b/tcg/optimize.c
> index 62a128bc9b..3638ab9fea 100644
> --- a/tcg/optimize.c
> +++ b/tcg/optimize.c
> @@ -1454,7 +1454,7 @@ static bool fold_and(OptContext *ctx, TCGOp *op)
> a_mask = t1->z_mask & ~t2->o_mask;
>
> if (!fold_masks_zosa_int(ctx, op, z_mask, o_mask, s_mask, a_mask)) {
> - if (ti_is_const(t2)) {
> + if (op->opc == INDEX_op_and && ti_is_const(t2)) {
> /*
> * Canonicalize on extract, if valid. This aids x86 with its
> * 2 operand MOVZBL and 2 operand AND, selecting the TCGOpcod
How does the fold_masks_zosa stuff work for vector operations here?
The masks are only 64 bits but the value we're working with is
wider than that, right?
thanks
-- PMM
