[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PATCH 11/38] target-i386: Use mulu2 and muls2
From: |
Richard Henderson |
Subject: |
[Qemu-devel] [PATCH 11/38] target-i386: Use mulu2 and muls2 |
Date: |
Tue, 19 Feb 2013 23:51:59 -0800 |
These correspond very closely to the insns that we're emulating.
Signed-off-by: Richard Henderson <address@hidden>
---
target-i386/helper.h | 4 --
target-i386/int_helper.c | 40 ------------
target-i386/translate.c | 167 ++++++++++++++++-------------------------------
3 files changed, 56 insertions(+), 155 deletions(-)
diff --git a/target-i386/helper.h b/target-i386/helper.h
index 26a0cc8..d6974df 100644
--- a/target-i386/helper.h
+++ b/target-i386/helper.h
@@ -14,12 +14,8 @@ DEF_HELPER_2(idivw_AX, void, env, tl)
DEF_HELPER_2(divl_EAX, void, env, tl)
DEF_HELPER_2(idivl_EAX, void, env, tl)
#ifdef TARGET_X86_64
-DEF_HELPER_2(mulq_EAX_T0, void, env, tl)
-DEF_HELPER_2(imulq_EAX_T0, void, env, tl)
-DEF_HELPER_3(imulq_T0_T1, tl, env, tl, tl)
DEF_HELPER_2(divq_EAX, void, env, tl)
DEF_HELPER_2(idivq_EAX, void, env, tl)
-DEF_HELPER_FLAGS_2(umulh, TCG_CALL_NO_RWG_SE, tl, tl, tl)
#endif
DEF_HELPER_2(aam, void, env, int)
diff --git a/target-i386/int_helper.c b/target-i386/int_helper.c
index 3b56075..74c7c36 100644
--- a/target-i386/int_helper.c
+++ b/target-i386/int_helper.c
@@ -374,46 +374,6 @@ static int idiv64(uint64_t *plow, uint64_t *phigh, int64_t
b)
return 0;
}
-void helper_mulq_EAX_T0(CPUX86State *env, target_ulong t0)
-{
- uint64_t r0, r1;
-
- mulu64(&r0, &r1, EAX, t0);
- EAX = r0;
- EDX = r1;
- CC_DST = r0;
- CC_SRC = r1;
-}
-
-target_ulong helper_umulh(target_ulong t0, target_ulong t1)
-{
- uint64_t h, l;
- mulu64(&l, &h, t0, t1);
- return h;
-}
-
-void helper_imulq_EAX_T0(CPUX86State *env, target_ulong t0)
-{
- uint64_t r0, r1;
-
- muls64(&r0, &r1, EAX, t0);
- EAX = r0;
- EDX = r1;
- CC_DST = r0;
- CC_SRC = ((int64_t)r1 != ((int64_t)r0 >> 63));
-}
-
-target_ulong helper_imulq_T0_T1(CPUX86State *env, target_ulong t0,
- target_ulong t1)
-{
- uint64_t r0, r1;
-
- muls64(&r0, &r1, t0, t1);
- CC_DST = r0;
- CC_SRC = ((int64_t)r1 != ((int64_t)r0 >> 63));
- return r0;
-}
-
void helper_divq_EAX(CPUX86State *env, target_ulong t0)
{
uint64_t r0, r1;
diff --git a/target-i386/translate.c b/target-i386/translate.c
index 439d19e..1545e3f 100644
--- a/target-i386/translate.c
+++ b/target-i386/translate.c
@@ -4111,31 +4111,18 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
ot = s->dflag == 2 ? OT_QUAD : OT_LONG;
gen_ldst_modrm(env, s, modrm, ot, OR_TMP0, 0);
switch (ot) {
- TCGv_i64 t0, t1;
default:
- t0 = tcg_temp_new_i64();
- t1 = tcg_temp_new_i64();
-#ifdef TARGET_X86_64
- tcg_gen_ext32u_i64(t0, cpu_T[0]);
- tcg_gen_ext32u_i64(t1, cpu_regs[R_EDX]);
-#else
- tcg_gen_extu_i32_i64(t0, cpu_T[0]);
- tcg_gen_extu_i32_i64(t0, cpu_regs[R_EDX]);
-#endif
- tcg_gen_mul_i64(t0, t0, t1);
- tcg_gen_trunc_i64_tl(cpu_T[0], t0);
- tcg_gen_shri_i64(t0, t0, 32);
- tcg_gen_trunc_i64_tl(cpu_T[1], t0);
- tcg_temp_free_i64(t0);
- tcg_temp_free_i64(t1);
- gen_op_mov_reg_T0(OT_LONG, s->vex_v);
- gen_op_mov_reg_T1(OT_LONG, reg);
+ tcg_gen_trunc_tl_i32(cpu_tmp2_i32, cpu_T[0]);
+ tcg_gen_trunc_tl_i32(cpu_tmp3_i32, cpu_regs[R_EDX]);
+ tcg_gen_mulu2_i32(cpu_tmp2_i32, cpu_tmp3_i32,
+ cpu_tmp2_i32, cpu_tmp3_i32);
+ tcg_gen_extu_i32_tl(cpu_regs[s->vex_v], cpu_tmp2_i32);
+ tcg_gen_extu_i32_tl(cpu_regs[reg], cpu_tmp3_i32);
break;
#ifdef TARGET_X86_64
case OT_QUAD:
- tcg_gen_mov_tl(cpu_T[1], cpu_regs[R_EDX]);
- tcg_gen_mul_tl(cpu_regs[s->vex_v], cpu_T[0], cpu_T[1]);
- gen_helper_umulh(cpu_regs[reg], cpu_T[0], cpu_T[1]);
+ tcg_gen_mulu2_i64(cpu_regs[s->vex_v], cpu_regs[reg],
+ cpu_T[0], cpu_regs[R_EDX]);
break;
#endif
}
@@ -5034,39 +5021,22 @@ static target_ulong disas_insn(CPUX86State *env,
DisasContext *s,
break;
default:
case OT_LONG:
-#ifdef TARGET_X86_64
- gen_op_mov_TN_reg(OT_LONG, 1, R_EAX);
- tcg_gen_ext32u_tl(cpu_T[0], cpu_T[0]);
- tcg_gen_ext32u_tl(cpu_T[1], cpu_T[1]);
- tcg_gen_mul_tl(cpu_T[0], cpu_T[0], cpu_T[1]);
- gen_op_mov_reg_T0(OT_LONG, R_EAX);
- tcg_gen_mov_tl(cpu_cc_dst, cpu_T[0]);
- tcg_gen_shri_tl(cpu_T[0], cpu_T[0], 32);
- gen_op_mov_reg_T0(OT_LONG, R_EDX);
- tcg_gen_mov_tl(cpu_cc_src, cpu_T[0]);
-#else
- {
- TCGv_i64 t0, t1;
- t0 = tcg_temp_new_i64();
- t1 = tcg_temp_new_i64();
- gen_op_mov_TN_reg(OT_LONG, 1, R_EAX);
- tcg_gen_extu_i32_i64(t0, cpu_T[0]);
- tcg_gen_extu_i32_i64(t1, cpu_T[1]);
- tcg_gen_mul_i64(t0, t0, t1);
- tcg_gen_trunc_i64_i32(cpu_T[0], t0);
- gen_op_mov_reg_T0(OT_LONG, R_EAX);
- tcg_gen_mov_tl(cpu_cc_dst, cpu_T[0]);
- tcg_gen_shri_i64(t0, t0, 32);
- tcg_gen_trunc_i64_i32(cpu_T[0], t0);
- gen_op_mov_reg_T0(OT_LONG, R_EDX);
- tcg_gen_mov_tl(cpu_cc_src, cpu_T[0]);
- }
-#endif
+ tcg_gen_trunc_tl_i32(cpu_tmp2_i32, cpu_T[0]);
+ tcg_gen_trunc_tl_i32(cpu_tmp3_i32, cpu_regs[R_EAX]);
+ tcg_gen_mulu2_i32(cpu_tmp2_i32, cpu_tmp3_i32,
+ cpu_tmp2_i32, cpu_tmp3_i32);
+ tcg_gen_extu_i32_tl(cpu_regs[R_EAX], cpu_tmp2_i32);
+ tcg_gen_extu_i32_tl(cpu_regs[R_EDX], cpu_tmp3_i32);
+ tcg_gen_mov_tl(cpu_cc_dst, cpu_regs[R_EAX]);
+ tcg_gen_mov_tl(cpu_cc_src, cpu_regs[R_EDX]);
set_cc_op(s, CC_OP_MULL);
break;
#ifdef TARGET_X86_64
case OT_QUAD:
- gen_helper_mulq_EAX_T0(cpu_env, cpu_T[0]);
+ tcg_gen_mulu2_i64(cpu_regs[R_EAX], cpu_regs[R_EDX],
+ cpu_T[0], cpu_regs[R_EAX]);
+ tcg_gen_mov_tl(cpu_cc_dst, cpu_regs[R_EAX]);
+ tcg_gen_mov_tl(cpu_cc_src, cpu_regs[R_EDX]);
set_cc_op(s, CC_OP_MULQ);
break;
#endif
@@ -5102,41 +5072,25 @@ static target_ulong disas_insn(CPUX86State *env,
DisasContext *s,
break;
default:
case OT_LONG:
-#ifdef TARGET_X86_64
- gen_op_mov_TN_reg(OT_LONG, 1, R_EAX);
- tcg_gen_ext32s_tl(cpu_T[0], cpu_T[0]);
- tcg_gen_ext32s_tl(cpu_T[1], cpu_T[1]);
- tcg_gen_mul_tl(cpu_T[0], cpu_T[0], cpu_T[1]);
- gen_op_mov_reg_T0(OT_LONG, R_EAX);
- tcg_gen_mov_tl(cpu_cc_dst, cpu_T[0]);
- tcg_gen_ext32s_tl(cpu_tmp0, cpu_T[0]);
- tcg_gen_sub_tl(cpu_cc_src, cpu_T[0], cpu_tmp0);
- tcg_gen_shri_tl(cpu_T[0], cpu_T[0], 32);
- gen_op_mov_reg_T0(OT_LONG, R_EDX);
-#else
- {
- TCGv_i64 t0, t1;
- t0 = tcg_temp_new_i64();
- t1 = tcg_temp_new_i64();
- gen_op_mov_TN_reg(OT_LONG, 1, R_EAX);
- tcg_gen_ext_i32_i64(t0, cpu_T[0]);
- tcg_gen_ext_i32_i64(t1, cpu_T[1]);
- tcg_gen_mul_i64(t0, t0, t1);
- tcg_gen_trunc_i64_i32(cpu_T[0], t0);
- gen_op_mov_reg_T0(OT_LONG, R_EAX);
- tcg_gen_mov_tl(cpu_cc_dst, cpu_T[0]);
- tcg_gen_sari_tl(cpu_tmp0, cpu_T[0], 31);
- tcg_gen_shri_i64(t0, t0, 32);
- tcg_gen_trunc_i64_i32(cpu_T[0], t0);
- gen_op_mov_reg_T0(OT_LONG, R_EDX);
- tcg_gen_sub_tl(cpu_cc_src, cpu_T[0], cpu_tmp0);
- }
-#endif
+ tcg_gen_trunc_tl_i32(cpu_tmp2_i32, cpu_T[0]);
+ tcg_gen_trunc_tl_i32(cpu_tmp3_i32, cpu_regs[R_EAX]);
+ tcg_gen_muls2_i32(cpu_tmp2_i32, cpu_tmp3_i32,
+ cpu_tmp2_i32, cpu_tmp3_i32);
+ tcg_gen_extu_i32_tl(cpu_regs[R_EAX], cpu_tmp2_i32);
+ tcg_gen_extu_i32_tl(cpu_regs[R_EDX], cpu_tmp3_i32);
+ tcg_gen_sari_i32(cpu_tmp2_i32, cpu_tmp2_i32, 31);
+ tcg_gen_mov_tl(cpu_cc_dst, cpu_regs[R_EAX]);
+ tcg_gen_sub_i32(cpu_tmp2_i32, cpu_tmp2_i32, cpu_tmp3_i32);
+ tcg_gen_extu_i32_tl(cpu_cc_src, cpu_tmp2_i32);
set_cc_op(s, CC_OP_MULL);
break;
#ifdef TARGET_X86_64
case OT_QUAD:
- gen_helper_imulq_EAX_T0(cpu_env, cpu_T[0]);
+ tcg_gen_muls2_i64(cpu_regs[R_EAX], cpu_regs[R_EDX],
+ cpu_T[0], cpu_regs[R_EAX]);
+ tcg_gen_mov_tl(cpu_cc_dst, cpu_regs[R_EAX]);
+ tcg_gen_sari_tl(cpu_cc_src, cpu_regs[R_EAX], 63);
+ tcg_gen_sub_tl(cpu_cc_src, cpu_cc_src, cpu_regs[R_EDX]);
set_cc_op(s, CC_OP_MULQ);
break;
#endif
@@ -5391,37 +5345,27 @@ static target_ulong disas_insn(CPUX86State *env,
DisasContext *s,
} else {
gen_op_mov_TN_reg(ot, 1, reg);
}
-
-#ifdef TARGET_X86_64
- if (ot == OT_QUAD) {
- gen_helper_imulq_T0_T1(cpu_T[0], cpu_env, cpu_T[0], cpu_T[1]);
- } else
-#endif
- if (ot == OT_LONG) {
+ switch (ot) {
#ifdef TARGET_X86_64
- tcg_gen_ext32s_tl(cpu_T[0], cpu_T[0]);
- tcg_gen_ext32s_tl(cpu_T[1], cpu_T[1]);
- tcg_gen_mul_tl(cpu_T[0], cpu_T[0], cpu_T[1]);
- tcg_gen_mov_tl(cpu_cc_dst, cpu_T[0]);
- tcg_gen_ext32s_tl(cpu_tmp0, cpu_T[0]);
- tcg_gen_sub_tl(cpu_cc_src, cpu_T[0], cpu_tmp0);
-#else
- {
- TCGv_i64 t0, t1;
- t0 = tcg_temp_new_i64();
- t1 = tcg_temp_new_i64();
- tcg_gen_ext_i32_i64(t0, cpu_T[0]);
- tcg_gen_ext_i32_i64(t1, cpu_T[1]);
- tcg_gen_mul_i64(t0, t0, t1);
- tcg_gen_trunc_i64_i32(cpu_T[0], t0);
- tcg_gen_mov_tl(cpu_cc_dst, cpu_T[0]);
- tcg_gen_sari_tl(cpu_tmp0, cpu_T[0], 31);
- tcg_gen_shri_i64(t0, t0, 32);
- tcg_gen_trunc_i64_i32(cpu_T[1], t0);
- tcg_gen_sub_tl(cpu_cc_src, cpu_T[1], cpu_tmp0);
- }
+ case OT_QUAD:
+ tcg_gen_muls2_i64(cpu_regs[reg], cpu_T[1], cpu_T[0], cpu_T[1]);
+ tcg_gen_mov_tl(cpu_cc_dst, cpu_regs[reg]);
+ tcg_gen_sari_tl(cpu_cc_src, cpu_cc_dst, 63);
+ tcg_gen_sub_tl(cpu_cc_src, cpu_cc_src, cpu_T[1]);
+ break;
#endif
- } else {
+ case OT_LONG:
+ tcg_gen_trunc_tl_i32(cpu_tmp2_i32, cpu_T[0]);
+ tcg_gen_trunc_tl_i32(cpu_tmp3_i32, cpu_T[1]);
+ tcg_gen_muls2_i32(cpu_tmp2_i32, cpu_tmp3_i32,
+ cpu_tmp2_i32, cpu_tmp3_i32);
+ tcg_gen_extu_i32_tl(cpu_regs[reg], cpu_tmp2_i32);
+ tcg_gen_sari_i32(cpu_tmp2_i32, cpu_tmp2_i32, 31);
+ tcg_gen_mov_tl(cpu_cc_dst, cpu_regs[reg]);
+ tcg_gen_sub_i32(cpu_tmp2_i32, cpu_tmp2_i32, cpu_tmp3_i32);
+ tcg_gen_extu_i32_tl(cpu_cc_src, cpu_tmp2_i32);
+ break;
+ default:
tcg_gen_ext16s_tl(cpu_T[0], cpu_T[0]);
tcg_gen_ext16s_tl(cpu_T[1], cpu_T[1]);
/* XXX: use 32 bit mul which could be faster */
@@ -5429,8 +5373,9 @@ static target_ulong disas_insn(CPUX86State *env,
DisasContext *s,
tcg_gen_mov_tl(cpu_cc_dst, cpu_T[0]);
tcg_gen_ext16s_tl(cpu_tmp0, cpu_T[0]);
tcg_gen_sub_tl(cpu_cc_src, cpu_T[0], cpu_tmp0);
+ gen_op_mov_reg_T0(ot, reg);
+ break;
}
- gen_op_mov_reg_T0(ot, reg);
set_cc_op(s, CC_OP_MULB + ot);
break;
case 0x1c0:
--
1.8.1.2
- [Qemu-devel] [PATCH 03/38] tcg-sparc: Always implement 32-bit multiword ops, (continued)
- [Qemu-devel] [PATCH 03/38] tcg-sparc: Always implement 32-bit multiword ops, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 05/38] tcg: Add signed multiword multiplication operations, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 07/38] tcg: Implement multiword multiply helpers, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 06/38] tcg: Implement a 64-bit to 32-bit extraction helper, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 08/38] tcg: Implement multiword addition helpers, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 12/38] target-i386: Use add2 to implement the ADX extension, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 10/38] tcg-arm: Implement muls2_i32, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 09/38] tcg-i386: Implement multiword arithmetic ops, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 14/38] tcg: Apply life analysis to 64-bit multiword arithmetic ops, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 16/38] target-s390x: Use mulu2 for mlgr insn, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 11/38] target-i386: Use mulu2 and muls2,
Richard Henderson <=
- [Qemu-devel] [PATCH 13/38] tcg: Implement muls2 with mulu2, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 18/38] target-arm: Use mul[us]2 and add2 in umlal et al, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 15/38] target-alpha: Use mulu2 for umulh insn, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 28/38] target-ppc: Compute addition carry with setcond, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 32/38] target-ppc: Compute mullwo without branches, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 34/38] target-sparc: Use mul*2 for multiply, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 33/38] target-sparc: Use official add2/sub2 interfaces for addx/subx, Richard Henderson, 2013/02/20
- [Qemu-devel] [PATCH 30/38] target-ppc: Implement neg in terms of subf, Richard Henderson, 2013/02/20