On 8/17/20 1:49 AM, frank.ch...@sifive.com wrote:
> +static bool vrgatherei16_vv_check(DisasContext *s, arg_rmrr *a)
> +{
> + int8_t emul = 4 - (s->sew + 3) + s->lmul;
Similarly for EEW elsewhere. This should be
int emul = MO_16 - s->sew + s->lmul;
> @@ -3339,7 +3353,8 @@ static bool trans_vrgather_vx(DisasContext *s, arg_rmrr
> *a)
> }
>
> if (a->vm && s->vl_eq_vlmax) {
> - int vlmax = s->vlen;
> + int scale = s->lmul - (s->sew + 3);
> + int vlmax = scale < 0 ? s->vlen >> -scale : s->vlen << scale;
Isn't this what s->vlen is already computed as?
> +#define GEN_VEXT_VRGATHER_VV(NAME, TS1, TS2, HS1, HS2) \
> void HELPER(NAME)(void *vd, void *v0, void *vs1, void *vs2, \
> CPURISCVState *env, uint32_t desc) \
> { \
> - uint32_t vlmax = vext_max_elems(desc, ctzl(sizeof(ETYPE))); \
> + uint32_t vlmax = vext_max_elems(desc, ctzl(sizeof(TS1))); \
Surely you can't just use TS1? For vrgatherei16_vv_[wd], the maximum number of
elements is still bound by TS2.
r~
