Re: [RFC 6/8] target/riscv: delete redundant check for zcd instructions in decode_opc

[Alistair Francis]

On Fri, Sep 30, 2022 at 11:28 AM Weiwei Li <liwei...@iscas.ac.cn> wrote:
>
> All the check for Zcd instructions have been done in their trans function
>
> Signed-off-by: Weiwei Li <liwei...@iscas.ac.cn>
> Signed-off-by: Junqiang Wang <wangjunqi...@iscas.ac.cn>
> ---
>  target/riscv/translate.c | 7 -------
>  1 file changed, 7 deletions(-)
>
> diff --git a/target/riscv/translate.c b/target/riscv/translate.c
> index 347bc913eb..a55b4a7849 100644
> --- a/target/riscv/translate.c
> +++ b/target/riscv/translate.c
> @@ -1087,13 +1087,6 @@ static void decode_opc(CPURISCVState *env, 
> DisasContext *ctx, uint16_t opcode)
>               ((opcode & 0xe003) == 0xe000) ||
>               ((opcode & 0xe003) == 0xe002))) {
>              gen_exception_illegal(ctx);
> -        } else if (!(has_ext(ctx, RVC) || ctx->cfg_ptr->ext_zcd ||
> -                     ctx->cfg_ptr->ext_zcmp || ctx->cfg_ptr->ext_zcmt) &&
> -                   (((opcode & 0xe003) == 0x2000) ||
> -                    ((opcode & 0xe003) == 0x2002) ||
> -                    ((opcode & 0xe003) == 0xa000) ||
> -                    ((opcode & 0xe003) == 0xa002))) {
> -            gen_exception_illegal(ctx);

It's probably best to never add this in the first place.

Remember that the extension can't be enabled until the last patch, so
it's ok if we don't support it all in one go

Alistair

>          } else {
>              ctx->opcode = opcode;
>              ctx->pc_succ_insn = ctx->base.pc_next + 2;
> --
> 2.25.1
>
>