On Mon, Sep 12, 2022 at 3:56 PM Richard Henderson <richard.hender...@linaro.org> wrote:
> > +static void gen_SSE4a_I(DisasContext *s, CPUX86State *env, X86DecodedInsn > > *decode) > > +{ > > + TCGv_i32 length = tcg_const_i32(decode->immediate & 255); > > + TCGv_i32 index = tcg_const_i32(decode->immediate >> 8); > > + > > + if (s->prefix & PREFIX_DATA) { > > + gen_helper_extrq_i(cpu_env, s->ptr0, index, length); > > + } else { > > + gen_helper_insertq_i(cpu_env, s->ptr0, index, length); > > + } > > + tcg_temp_free_i32(length); > > + tcg_temp_free_i32(index); > > Again, why the choice of delayed decode? I guess it doesn't matter, but it's > odd. Mostly because I wasn't sure of which would be preferable so I tried different things. I think now I have a better picture. I will mostly switch to decode_by_prefix. Paolo