Re: [PATCH v3 03/44] target/arm: Implement MVE VCLZ

[Peter Maydell]

On Thu, 17 Jun 2021 at 13:16, Peter Maydell <peter.maydell@linaro.org> wrote:
>
> Implement the MVE VCLZ insn (and the necessary machinery
> for MVE 1-input vector ops).
>
> Note that for non-load instructions predication is always performed
> at a byte level granularity regardless of element size (R_ZLSJ),
> and so the masking logic here differs from that used in the VLDR
> and VSTR helpers.
>
> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>


> +#define DO_1OP(OP, ESIZE, TYPE, FN)                                     \
> +    void HELPER(mve_##OP)(CPUARMState *env, void *vd, void *vm)         \
> +    {                                                                   \
> +        TYPE *d = vd, *m = vm;                                          \
> +        uint16_t mask = mve_element_mask(env);                          \
> +        unsigned e;                                                     \
> +        unsigned const esize = sizeof(TYPE);                            \
> +        for (e = 0; e < 16 / esize; e++, mask >>= esize) {              \
> +            mergemask(&d[H##ESIZE(e)], FN(m[H##ESIZE(e)]), mask);       \
> +        }                                                               \
> +        mve_advance_vpt(env);                                           \
> +    }

Just noticed I didn't quite get the refactoring from "use sizeof(TYPE)"
to "take ESIZE" right here, so we currently do both (correct behaviour,
but oddly written). I'm going to squash in this trivial fixup:

--- a/target/arm/mve_helper.c
+++ b/target/arm/mve_helper.c
@@ -265,8 +265,7 @@ static inline void unknown_mergemask_type(void *d,
uint64_t r, uint16_t mask)
         TYPE *d = vd, *m = vm;                                          \
         uint16_t mask = mve_element_mask(env);                          \
         unsigned e;                                                     \
-        unsigned const esize = sizeof(TYPE);                            \
-        for (e = 0; e < 16 / esize; e++, mask >>= esize) {              \
+        for (e = 0; e < 16 / ESIZE; e++, mask >>= ESIZE) {              \
             mergemask(&d[H##ESIZE(e)], FN(m[H##ESIZE(e)]), mask);       \
         }                                                               \
         mve_advance_vpt(env);                                           \

thanks
-- PMM