Re: [Qemu-devel] [PATCH] Add new instruction for M68K family

[Thomas Huth]

On Fri, 31 Oct 2014 04:28:16 +0000
"Guo, Lei" <address@hidden> wrote:

> This patch aims to add new instructions (ABCD) for M68K family.
> The original TCG for M68k doesn’t include all instructions,so I intend to add 
> the unrealized instruction or fix wrong instructions step by step.
> This instruction is realized according to M68000RPM pdf.
> It’s my great hornor if someone can review my new code.
> Best Regards
> 
> Instruction Description:
> Adds the source operand to the destination operand along with the extend bit,
> and stores the result in the destination location. The addition is performed 
> using binary-coded decimal arithmetic. The operands, which are packed 
> binary-coded decimal
> numbers, can be addressed in two different ways:
> 1. Data Register to Data Register: The operands are contained in the data 
> regis-ters specified in the instruction.
> 2. Memory to Memory: The operands are addressed with the predecrement 
> ad-dressing mode using the address registers specified in the instruction.
> 
> igned-off-by: Guolei <address@hidden >
> ---
> target-m68k/helper.c    |   31 +++++++++++++++++++++++++++++++
> target-m68k/helper.h    |    2 ++
> target-m68k/translate.c |   27 +++++++++++++++++++++++++++
> 3 files changed, 60 insertions(+)
> 
> diff --git a/target-m68k/helper.c b/target-m68k/helper.c
> index 77225a2..e480ba4 100644
> --- a/target-m68k/helper.c
> +++ b/target-m68k/helper.c
> @@ -865,6 +865,37 @@ void HELPER(set_mac_extu)(CPUM68KState *env, uint32_t 
> val, uint32_t acc)
>      env->macc[acc + 1] = res;
> }
> +uint32_t HELPER(abcd_cc)(CPUM68KState *env, uint32_t src, uint32_t dest)
> +{
> +    uint16_t hi, lo;
> +    uint16_t res;
> +    uint32_t flags;
> +
> +    flags = env->cc_dest;
> +    flags &= ~(CCF_C|CCF_X);
> +    lo = (src & 0x0f) + (dest & 0x0f);
> +    if (env->cc_x)
> +        lo ++;
> +    hi = (src & 0xf0) + (dest & 0xf0);
> +    res = hi + lo;
> +    if (lo > 9)
> +        res += 0x06;
> +
> +    /* C and X flags: set if decimal carry, cleared otherwise */
> +    if ((res & 0x3F0) > 0x90) {
> +        res += 0x60;
> +        flags |= CCF_C|CCF_X;
> +    }
> +
> +    /* Z flag: cleared if nonzero */
> +    if (res & 0xff)
> +        flags &= ~CCF_Z;
> +    dest = (dest & 0xffffff00) | (res & 0xff);
> +    env->cc_x = (flags & CCF_X) != 0;
> +    env->cc_dest = flags;
> +    return dest;
> +}
> +
> void m68k_cpu_exec_enter(CPUState *cs)
> {
>      M68kCPU *cpu = M68K_CPU(cs);
> diff --git a/target-m68k/helper.h b/target-m68k/helper.h
> index f4e5fdf..21d31f2 100644
> --- a/target-m68k/helper.h
> +++ b/target-m68k/helper.h
> @@ -48,3 +48,5 @@ DEF_HELPER_3(set_mac_extu, void, env, i32, i32)
>  DEF_HELPER_2(flush_flags, void, env, i32)
> DEF_HELPER_2(raise_exception, void, env, i32)
> +
> +DEF_HELPER_3(abcd_cc, i32, env, i32, i32)
> diff --git a/target-m68k/translate.c b/target-m68k/translate.c
> index efd4cfc..df18a66 100644
> --- a/target-m68k/translate.c
> +++ b/target-m68k/translate.c
> @@ -872,6 +872,33 @@ static void gen_jmp_tb(DisasContext *s, int n, uint32_t 
> dest)
>      s->is_jmp = DISAS_TB_JUMP;
> }
> +DISAS_INSN(abcd_reg)
> +{
> +    TCGv src;
> +    TCGv dest;
> +
> +    src = DREG(insn, 0);
> +    dest = DREG(insn, 9);
> +    gen_helper_abcd_cc(dest, cpu_env, src, dest);
> +}
> +
> +DISAS_INSN(abcd_mem)
> +{
> +    TCGv src;
> +    TCGv addr_src;
> +    TCGv dest;
> +    TCGv addr_dest;
> +
> +    addr_src = AREG(insn, 0);
> +    tcg_gen_subi_i32(addr_src, addr_src, opsize_bytes(OS_BYTE));
> +    src = gen_load(s, OS_BYTE, addr_src, 0);
> +    addr_dest = AREG(insn, 9);
> +    tcg_gen_subi_i32(addr_dest, addr_dest, opsize_bytes(OS_BYTE));
> +    dest = gen_load(s, OS_BYTE, addr_dest, 0);
> +    gen_helper_abcd_cc(dest, cpu_env, src, dest);
> +    gen_store(s, OS_BYTE, addr_dest, dest);
> +}
> +
> DISAS_INSN(undef_mac)
> {
>      gen_exception(s, s->pc - 2, EXCP_LINEA);
> --
> 1.7.9.5

Déjà-vu :

https://gitorious.org/qemu-m68k/qemu-m68k/commit/17b1b9e1d82055c70fd1d38a9a6fb60cbdc989fe

Have you been in touch with the author of that original patch already?

 Thomas