Re: [Qemu-devel] [PATCH v3 13/16] target/arm: Decode aa32 armv8.3 3-same

[Peter Maydell]

On 28 February 2018 at 19:31, Richard Henderson
 wrote:
> Signed-off-by: Richard Henderson 
> ---
>  target/arm/translate.c | 68 
> ++
>  1 file changed, 68 insertions(+)

> @@ -8424,6 +8486,12 @@ static void disas_arm_insn(DisasContext *s, unsigned 
> int insn)
>  }
>  }
>  }
> +} else if ((insn & 0x0e000a00) == 0x0c000800
> +   && arm_dc_feature(s, ARM_FEATURE_V8)) {
> +if (disas_neon_insn_3same_ext(s, insn)) {
> +goto illegal_op;
> +}
> +return;
>  } else if ((insn & 0x0fe0) == 0x0c40) {
>  /* Coprocessor double register transfer.  */
>  ARCH(5TE);

Oh yes, shouldn't there be a similar fragment in disas_thumb2_insn()
to take care of the Thumb encodings of these insns ?

thanks
-- PMM

Re: [Qemu-devel] [PATCH v3 13/16] target/arm: Decode aa32 armv8.3 3-same

[Peter Maydell]

On 28 February 2018 at 19:31, Richard Henderson
 wrote:
> Signed-off-by: Richard Henderson 
> ---
>  target/arm/translate.c | 68 
> ++
>  1 file changed, 68 insertions(+)
>
> diff --git a/target/arm/translate.c b/target/arm/translate.c
> index 9169b6b367..45513c9d86 100644
> --- a/target/arm/translate.c
> +++ b/target/arm/translate.c
> @@ -7680,6 +7680,68 @@ static int disas_neon_data_insn(DisasContext *s, 
> uint32_t insn)
>  return 0;
>  }
>
> +/* Advanced SIMD three registers of the same length extension.
> + *  31   2523  2220   16   12  11   10   983 > 0
> + * 
> +---+-+---+-+++---++---++-++
> + * | 1 1 1 1 1 1 0 | op1 | D | op2 | Vn | Vd | 1 | o3 | 0 | o4 | N Q M U | 
> Vm |
> + * 
> +---+-+---+-+++---++---++-++
> + */
> +static int disas_neon_insn_3same_ext(DisasContext *s, uint32_t insn)
> +{
> +gen_helper_gvec_3_ptr *fn_gvec_ptr;
> +int rd, rn, rm, rot, size, opr_sz;
> +TCGv_ptr fpst;
> +bool q;
> +
> +q = extract32(insn, 6, 1);
> +VFP_DREG_D(rd, insn);
> +VFP_DREG_N(rn, insn);
> +VFP_DREG_M(rm, insn);
> +if ((rd | rn | rm) & q) {
> +return 1;
> +}
> +
> +if ((insn & 0xfe200f10) == 0xfc200800) {

Slightly confusing to be re-checking bits we've already decoded,
rather than just looking at op1/op2/op3/op4/Q/U.

Anyway
Reviewed-by: Peter Maydell 

thanks
-- PMM