Implement x86 SSSE3 instructions.

git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@5315 c046a42c-6fe2-441c-8c8c-71466251a162

3 files changed, 272 insertions, 4 deletions

diff --git a/target-i386/ops_sse.h b/target-i386/ops_sse.h
index 7568681835..4fa8e06624 100644
--- a/target-i386/ops_sse.h
+++ b/target-i386/ops_sse.h
@@ -1,5 +1,5 @@
 /*
- *  MMX/3DNow!/SSE/SSE2/SSE3/PNI support
+ *  MMX/3DNow!/SSE/SSE2/SSE3/SSSE3/PNI support
  *
  *  Copyright (c) 2005 Fabrice Bellard
  *
@@ -1275,6 +1275,151 @@ void helper_pswapd(MMXReg *d, MMXReg *s)
 }
 #endif
 
+/* SSSE3 op helpers */
+void glue(helper_pshufb, SUFFIX) (Reg *d, Reg *s)
+{
+    int i;
+    Reg r;
+
+    for (i = 0; i < (8 << SHIFT); i++)
+        r.B(i) = (s->B(i) & 0x80) ? 0 : (d->B(s->B(i) & ((8 << SHIFT) - 1)));
+
+    *d = r;
+}
+
+void glue(helper_phaddw, SUFFIX) (Reg *d, Reg *s)
+{
+    d->W(0) = (int16_t)d->W(0) + (int16_t)d->W(1);
+    d->W(1) = (int16_t)d->W(2) + (int16_t)d->W(3);
+    XMM_ONLY(d->W(2) = (int16_t)d->W(4) + (int16_t)d->W(5));
+    XMM_ONLY(d->W(3) = (int16_t)d->W(6) + (int16_t)d->W(7));
+    d->W((2 << SHIFT) + 0) = (int16_t)s->W(0) + (int16_t)s->W(1);
+    d->W((2 << SHIFT) + 1) = (int16_t)s->W(2) + (int16_t)s->W(3);
+    XMM_ONLY(d->W(6) = (int16_t)s->W(4) + (int16_t)s->W(5));
+    XMM_ONLY(d->W(7) = (int16_t)s->W(6) + (int16_t)s->W(7));
+}
+
+void glue(helper_phaddd, SUFFIX) (Reg *d, Reg *s)
+{
+    d->L(0) = (int32_t)d->L(0) + (int32_t)d->L(1);
+    XMM_ONLY(d->L(1) = (int32_t)d->L(2) + (int32_t)d->L(3));
+    d->L((1 << SHIFT) + 0) = (int32_t)s->L(0) + (int32_t)s->L(1);
+    XMM_ONLY(d->L(3) = (int32_t)s->L(2) + (int32_t)s->L(3));
+}
+
+void glue(helper_phaddsw, SUFFIX) (Reg *d, Reg *s)
+{
+    d->W(0) = satsw((int16_t)d->W(0) + (int16_t)d->W(1));
+    d->W(1) = satsw((int16_t)d->W(2) + (int16_t)d->W(3));
+    XMM_ONLY(d->W(2) = satsw((int16_t)d->W(4) + (int16_t)d->W(5)));
+    XMM_ONLY(d->W(3) = satsw((int16_t)d->W(6) + (int16_t)d->W(7)));
+    d->W((2 << SHIFT) + 0) = satsw((int16_t)s->W(0) + (int16_t)s->W(1));
+    d->W((2 << SHIFT) + 1) = satsw((int16_t)s->W(2) + (int16_t)s->W(3));
+    XMM_ONLY(d->W(6) = satsw((int16_t)s->W(4) + (int16_t)s->W(5)));
+    XMM_ONLY(d->W(7) = satsw((int16_t)s->W(6) + (int16_t)s->W(7)));
+}
+
+void glue(helper_pmaddubsw, SUFFIX) (Reg *d, Reg *s)
+{
+    d->W(0) = satsw((int8_t)s->B( 0) * (uint8_t)d->B( 0) +
+                    (int8_t)s->B( 1) * (uint8_t)d->B( 1));
+    d->W(1) = satsw((int8_t)s->B( 2) * (uint8_t)d->B( 2) +
+                    (int8_t)s->B( 3) * (uint8_t)d->B( 3));
+    d->W(2) = satsw((int8_t)s->B( 4) * (uint8_t)d->B( 4) +
+                    (int8_t)s->B( 5) * (uint8_t)d->B( 5));
+    d->W(3) = satsw((int8_t)s->B( 6) * (uint8_t)d->B( 6) +
+                    (int8_t)s->B( 7) * (uint8_t)d->B( 7));
+#if SHIFT == 1
+    d->W(4) = satsw((int8_t)s->B( 8) * (uint8_t)d->B( 8) +
+                    (int8_t)s->B( 9) * (uint8_t)d->B( 9));
+    d->W(5) = satsw((int8_t)s->B(10) * (uint8_t)d->B(10) +
+                    (int8_t)s->B(11) * (uint8_t)d->B(11));
+    d->W(6) = satsw((int8_t)s->B(12) * (uint8_t)d->B(12) +
+                    (int8_t)s->B(13) * (uint8_t)d->B(13));
+    d->W(7) = satsw((int8_t)s->B(14) * (uint8_t)d->B(14) +
+                    (int8_t)s->B(15) * (uint8_t)d->B(15));
+#endif
+}
+
+void glue(helper_phsubw, SUFFIX) (Reg *d, Reg *s)
+{
+    d->W(0) = (int16_t)d->W(0) - (int16_t)d->W(1);
+    d->W(1) = (int16_t)d->W(2) - (int16_t)d->W(3);
+    XMM_ONLY(d->W(2) = (int16_t)d->W(4) - (int16_t)d->W(5));
+    XMM_ONLY(d->W(3) = (int16_t)d->W(6) - (int16_t)d->W(7));
+    d->W((2 << SHIFT) + 0) = (int16_t)s->W(0) - (int16_t)s->W(1);
+    d->W((2 << SHIFT) + 1) = (int16_t)s->W(2) - (int16_t)s->W(3);
+    XMM_ONLY(d->W(6) = (int16_t)s->W(4) - (int16_t)s->W(5));
+    XMM_ONLY(d->W(7) = (int16_t)s->W(6) - (int16_t)s->W(7));
+}
+
+void glue(helper_phsubd, SUFFIX) (Reg *d, Reg *s)
+{
+    d->L(0) = (int32_t)d->L(0) - (int32_t)d->L(1);
+    XMM_ONLY(d->L(1) = (int32_t)d->L(2) - (int32_t)d->L(3));
+    d->L((1 << SHIFT) + 0) = (int32_t)s->L(0) - (int32_t)s->L(1);
+    XMM_ONLY(d->L(3) = (int32_t)s->L(2) - (int32_t)s->L(3));
+}
+
+void glue(helper_phsubsw, SUFFIX) (Reg *d, Reg *s)
+{
+    d->W(0) = satsw((int16_t)d->W(0) - (int16_t)d->W(1));
+    d->W(1) = satsw((int16_t)d->W(2) - (int16_t)d->W(3));
+    XMM_ONLY(d->W(2) = satsw((int16_t)d->W(4) - (int16_t)d->W(5)));
+    XMM_ONLY(d->W(3) = satsw((int16_t)d->W(6) - (int16_t)d->W(7)));
+    d->W((2 << SHIFT) + 0) = satsw((int16_t)s->W(0) - (int16_t)s->W(1));
+    d->W((2 << SHIFT) + 1) = satsw((int16_t)s->W(2) - (int16_t)s->W(3));
+    XMM_ONLY(d->W(6) = satsw((int16_t)s->W(4) - (int16_t)s->W(5)));
+    XMM_ONLY(d->W(7) = satsw((int16_t)s->W(6) - (int16_t)s->W(7)));
+}
+
+#define FABSB(_, x) x > INT8_MAX  ? -(int8_t ) x : x
+#define FABSW(_, x) x > INT16_MAX ? -(int16_t) x : x
+#define FABSL(_, x) x > INT32_MAX ? -(int32_t) x : x
+SSE_HELPER_B(helper_pabsb, FABSB)
+SSE_HELPER_W(helper_pabsw, FABSW)
+SSE_HELPER_L(helper_pabsd, FABSL)
+
+#define FMULHRSW(d, s) ((int16_t) d * (int16_t) s + 0x4000) >> 15
+SSE_HELPER_W(helper_pmulhrsw, FMULHRSW)
+
+#define FSIGNB(d, s) s <= INT8_MAX  ? s ? d : 0 : -(int8_t ) d
+#define FSIGNW(d, s) s <= INT16_MAX ? s ? d : 0 : -(int16_t) d
+#define FSIGNL(d, s) s <= INT32_MAX ? s ? d : 0 : -(int32_t) d
+SSE_HELPER_B(helper_psignb, FSIGNB)
+SSE_HELPER_W(helper_psignw, FSIGNW)
+SSE_HELPER_L(helper_psignd, FSIGNL)
+
+void glue(helper_palignr, SUFFIX) (Reg *d, Reg *s, int32_t shift)
+{
+    Reg r;
+
+    /* XXX could be checked during translation */
+    if (shift >= (16 << SHIFT)) {
+        r.Q(0) = 0;
+        XMM_ONLY(r.Q(1) = 0);
+    } else {
+        shift <<= 3;
+#define SHR(v, i) (i < 64 && i > -64 ? i > 0 ? v >> (i) : (v << -(i)) : 0)
+#if SHIFT == 0
+        r.Q(0) = SHR(s->Q(0), shift -   0) |
+                 SHR(d->Q(0), shift -  64);
+#else
+        r.Q(0) = SHR(s->Q(0), shift -   0) |
+                 SHR(s->Q(1), shift -  64) |
+                 SHR(d->Q(0), shift - 128) |
+                 SHR(d->Q(1), shift - 192);
+        r.Q(1) = SHR(s->Q(0), shift +  64) |
+                 SHR(s->Q(1), shift -   0) |
+                 SHR(d->Q(0), shift -  64) |
+                 SHR(d->Q(1), shift - 128);
+#endif
+#undef SHR
+    }
+
+    *d = r;
+}
+
 #undef SHIFT
 #undef XMM_ONLY
 #undef Reg
diff --git a/target-i386/ops_sse_header.h b/target-i386/ops_sse_header.h
index 442e12cd42..22b77b7a79 100644
--- a/target-i386/ops_sse_header.h
+++ b/target-i386/ops_sse_header.h
@@ -1,5 +1,5 @@
 /*
- *  MMX/3DNow!/SSE/SSE2/SSE3/PNI support
+ *  MMX/3DNow!/SSE/SSE2/SSE3/SSSE3/PNI support
  *
  *  Copyright (c) 2005 Fabrice Bellard
  *
@@ -251,6 +251,24 @@ DEF_HELPER(void, helper_pfsubr, (MMXReg *d, MMXReg *s))
 DEF_HELPER(void, helper_pswapd, (MMXReg *d, MMXReg *s))
 #endif
 
+/* SSSE3 op helpers */
+DEF_HELPER(void, glue(helper_phaddw, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_phaddd, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_phaddsw, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_phsubw, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_phsubd, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_phsubsw, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_pabsb, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_pabsw, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_pabsd, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_pmaddubsw, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_pmulhrsw, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_pshufb, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_psignb, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_psignw, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_psignd, SUFFIX), (Reg *d, Reg *s))
+DEF_HELPER(void, glue(helper_palignr, SUFFIX), (Reg *d, Reg *s, int32_t shift))
+
 #undef SHIFT
 #undef Reg
 #undef SUFFIX
diff --git a/target-i386/translate.c b/target-i386/translate.c
index c4a119543a..c739d690ad 100644
--- a/target-i386/translate.c
+++ b/target-i386/translate.c
@@ -2770,6 +2770,9 @@ static void *sse_op_table1[256][4] = {
     [0xc2] = SSE_FOP(cmpeq),
     [0xc6] = { helper_shufps, helper_shufpd },
 
+    [0x38] = { SSE_SPECIAL, SSE_SPECIAL },  /* SSSE3 */
+    [0x3a] = { SSE_SPECIAL, SSE_SPECIAL },  /* SSSE3 */
+
     /* MMX ops and their SSE extensions */
     [0x60] = MMX_OP2(punpcklbw),
     [0x61] = MMX_OP2(punpcklwd),
@@ -2921,6 +2924,28 @@ static void *sse_op_table5[256] = {
     [0xbf] = helper_pavgb_mmx /* pavgusb */
 };
 
+static void *sse_op_table6[256][2] = {
+    [0x00] = MMX_OP2(pshufb),
+    [0x01] = MMX_OP2(phaddw),
+    [0x02] = MMX_OP2(phaddd),
+    [0x03] = MMX_OP2(phaddsw),
+    [0x04] = MMX_OP2(pmaddubsw),
+    [0x05] = MMX_OP2(phsubw),
+    [0x06] = MMX_OP2(phsubd),
+    [0x07] = MMX_OP2(phsubsw),
+    [0x08] = MMX_OP2(psignb),
+    [0x09] = MMX_OP2(psignw),
+    [0x0a] = MMX_OP2(psignd),
+    [0x0b] = MMX_OP2(pmulhrsw),
+    [0x1c] = MMX_OP2(pabsb),
+    [0x1d] = MMX_OP2(pabsw),
+    [0x1e] = MMX_OP2(pabsd),
+};
+
+static void *sse_op_table7[256][2] = {
+    [0x0f] = MMX_OP2(palignr),
+};
+
 static void gen_sse(DisasContext *s, int b, target_ulong pc_start, int rex_r)
 {
     int b1, op1_offset, op2_offset, is_xmm, val, ot;
@@ -2960,7 +2985,8 @@ static void gen_sse(DisasContext *s, int b, target_ulong pc_start, int rex_r)
         return;
     }
     if (is_xmm && !(s->flags & HF_OSFXSR_MASK))
-        goto illegal_op;
+        if ((b != 0x38 && b != 0x3a) || (s->prefix & PREFIX_DATA))
+            goto illegal_op;
     if (b == 0x0e) {
         if (!(s->cpuid_ext2_features & CPUID_EXT2_3DNOW))
             goto illegal_op;
@@ -3482,6 +3508,84 @@ static void gen_sse(DisasContext *s, int b, target_ulong pc_start, int rex_r)
             reg = ((modrm >> 3) & 7) | rex_r;
             gen_op_mov_reg_T0(OT_LONG, reg);
             break;
+        case 0x038:
+        case 0x138:
+            if (!(s->cpuid_ext_features & CPUID_EXT_SSSE3))
+                goto illegal_op;
+
+            b = modrm;
+            modrm = ldub_code(s->pc++);
+            rm = modrm & 7;
+            reg = ((modrm >> 3) & 7) | rex_r;
+            mod = (modrm >> 6) & 3;
+
+            sse_op2 = sse_op_table6[b][b1];
+            if (!sse_op2)
+                goto illegal_op;
+
+            if (b1) {
+                op1_offset = offsetof(CPUX86State,xmm_regs[reg]);
+                if (mod == 3) {
+                    op2_offset = offsetof(CPUX86State,xmm_regs[rm | REX_B(s)]);
+                } else {
+                    op2_offset = offsetof(CPUX86State,xmm_t0);
+                    gen_lea_modrm(s, modrm, &reg_addr, &offset_addr);
+                    gen_ldo_env_A0(s->mem_index, op2_offset);
+                }
+            } else {
+                op1_offset = offsetof(CPUX86State,fpregs[reg].mmx);
+                if (mod == 3) {
+                    op2_offset = offsetof(CPUX86State,fpregs[rm].mmx);
+                } else {
+                    op2_offset = offsetof(CPUX86State,mmx_t0);
+                    gen_lea_modrm(s, modrm, &reg_addr, &offset_addr);
+                    gen_ldq_env_A0(s->mem_index, op2_offset);
+                }
+            }
+            tcg_gen_addi_ptr(cpu_ptr0, cpu_env, op1_offset);
+            tcg_gen_addi_ptr(cpu_ptr1, cpu_env, op2_offset);
+            tcg_gen_helper_0_2(sse_op2, cpu_ptr0, cpu_ptr1);
+            break;
+        case 0x03a:
+        case 0x13a:
+            if (!(s->cpuid_ext_features & CPUID_EXT_SSSE3))
+                goto illegal_op;
+
+            b = modrm;
+            modrm = ldub_code(s->pc++);
+            rm = modrm & 7;
+            reg = ((modrm >> 3) & 7) | rex_r;
+            mod = (modrm >> 6) & 3;
+
+            sse_op2 = sse_op_table7[b][b1];
+            if (!sse_op2)
+                goto illegal_op;
+
+            if (b1) {
+                op1_offset = offsetof(CPUX86State,xmm_regs[reg]);
+                if (mod == 3) {
+                    op2_offset = offsetof(CPUX86State,xmm_regs[rm | REX_B(s)]);
+                } else {
+                    op2_offset = offsetof(CPUX86State,xmm_t0);
+                    gen_lea_modrm(s, modrm, &reg_addr, &offset_addr);
+                    gen_ldo_env_A0(s->mem_index, op2_offset);
+                }
+            } else {
+                op1_offset = offsetof(CPUX86State,fpregs[reg].mmx);
+                if (mod == 3) {
+                    op2_offset = offsetof(CPUX86State,fpregs[rm].mmx);
+                } else {
+                    op2_offset = offsetof(CPUX86State,mmx_t0);
+                    gen_lea_modrm(s, modrm, &reg_addr, &offset_addr);
+                    gen_ldq_env_A0(s->mem_index, op2_offset);
+                }
+            }
+            val = ldub_code(s->pc++);
+
+            tcg_gen_addi_ptr(cpu_ptr0, cpu_env, op1_offset);
+            tcg_gen_addi_ptr(cpu_ptr1, cpu_env, op2_offset);
+            tcg_gen_helper_0_3(sse_op2, cpu_ptr0, cpu_ptr1, tcg_const_i32(val));
+            break;
         default:
             goto illegal_op;
         }
@@ -6987,7 +7091,7 @@ static target_ulong disas_insn(DisasContext *s, target_ulong pc_start)
             gen_eob(s);
         }
         break;
-    /* MMX/3DNow!/SSE/SSE2/SSE3 support */
+    /* MMX/3DNow!/SSE/SSE2/SSE3/SSSE3 support */
     case 0x1c3: /* MOVNTI reg, mem */
         if (!(s->cpuid_features & CPUID_SSE2))
             goto illegal_op;
@@ -7100,6 +7204,7 @@ static target_ulong disas_insn(DisasContext *s, target_ulong pc_start)
         s->prefix &= ~(PREFIX_REPZ | PREFIX_REPNZ | PREFIX_DATA);
     case 0x110 ... 0x117:
     case 0x128 ... 0x12f:
+    case 0x138 ... 0x13a:
     case 0x150 ... 0x177:
     case 0x17c ... 0x17f:
     case 0x1c2: