diff options
Diffstat (limited to 'target/i386/tcg/emit.c.inc')
| -rw-r--r-- | target/i386/tcg/emit.c.inc | 202 |
1 files changed, 202 insertions, 0 deletions
diff --git a/target/i386/tcg/emit.c.inc b/target/i386/tcg/emit.c.inc index 947deacd37..8dbacc21ed 100644 --- a/target/i386/tcg/emit.c.inc +++ b/target/i386/tcg/emit.c.inc @@ -71,6 +71,56 @@ static inline int xmm_offset(MemOp ot) } } +static int vector_reg_offset(X86DecodedOp *op) +{ + assert(op->unit == X86_OP_MMX || op->unit == X86_OP_SSE); + + if (op->unit == X86_OP_MMX) { + return op->offset - mmx_offset(op->ot); + } else { + return op->offset - xmm_offset(op->ot); + } +} + +static int vector_elem_offset(X86DecodedOp *op, MemOp ot, int n) +{ + int base_ofs = vector_reg_offset(op); + switch(ot) { + case MO_8: + if (op->unit == X86_OP_MMX) { + return base_ofs + offsetof(MMXReg, MMX_B(n)); + } else { + return base_ofs + offsetof(ZMMReg, ZMM_B(n)); + } + case MO_16: + if (op->unit == X86_OP_MMX) { + return base_ofs + offsetof(MMXReg, MMX_W(n)); + } else { + return base_ofs + offsetof(ZMMReg, ZMM_W(n)); + } + case MO_32: + if (op->unit == X86_OP_MMX) { + return base_ofs + offsetof(MMXReg, MMX_L(n)); + } else { + return base_ofs + offsetof(ZMMReg, ZMM_L(n)); + } + case MO_64: + if (op->unit == X86_OP_MMX) { + return base_ofs; + } else { + return base_ofs + offsetof(ZMMReg, ZMM_Q(n)); + } + case MO_128: + assert(op->unit == X86_OP_SSE); + return base_ofs + offsetof(ZMMReg, ZMM_X(n)); + case MO_256: + assert(op->unit == X86_OP_SSE); + return base_ofs + offsetof(ZMMReg, ZMM_Y(n)); + default: + g_assert_not_reached(); + } +} + static void compute_mmx_offset(X86DecodedOp *op) { if (!op->has_ea) { @@ -183,6 +233,23 @@ static void gen_load(DisasContext *s, X86DecodedInsn *decode, int opn, TCGv v) } } +static TCGv_ptr op_ptr(X86DecodedInsn *decode, int opn) +{ + X86DecodedOp *op = &decode->op[opn]; + if (op->v_ptr) { + return op->v_ptr; + } + op->v_ptr = tcg_temp_new_ptr(); + + /* The temporary points to the MMXReg or ZMMReg. */ + tcg_gen_addi_ptr(op->v_ptr, cpu_env, vector_reg_offset(op)); + return op->v_ptr; +} + +#define OP_PTR0 op_ptr(decode, 0) +#define OP_PTR1 op_ptr(decode, 1) +#define OP_PTR2 op_ptr(decode, 2) + static void gen_writeback(DisasContext *s, X86DecodedInsn *decode, int opn, TCGv v) { X86DecodedOp *op = &decode->op[opn]; @@ -216,6 +283,114 @@ static void gen_writeback(DisasContext *s, X86DecodedInsn *decode, int opn, TCGv } } +static inline int vector_len(DisasContext *s, X86DecodedInsn *decode) +{ + if (decode->e.special == X86_SPECIAL_MMX && + !(s->prefix & (PREFIX_DATA | PREFIX_REPZ | PREFIX_REPNZ))) { + return 8; + } + return s->vex_l ? 32 : 16; +} + +static void gen_store_sse(DisasContext *s, X86DecodedInsn *decode, int src_ofs) +{ + MemOp ot = decode->op[0].ot; + int vec_len = vector_len(s, decode); + bool aligned = sse_needs_alignment(s, decode, ot); + + if (!decode->op[0].has_ea) { + tcg_gen_gvec_mov(MO_64, decode->op[0].offset, src_ofs, vec_len, vec_len); + return; + } + + switch (ot) { + case MO_64: + gen_stq_env_A0(s, src_ofs); + break; + case MO_128: + gen_sto_env_A0(s, src_ofs, aligned); + break; + case MO_256: + gen_sty_env_A0(s, src_ofs, aligned); + break; + default: + g_assert_not_reached(); + } +} + +#define BINARY_INT_GVEC(uname, func, ...) \ +static void gen_##uname(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) \ +{ \ + int vec_len = vector_len(s, decode); \ + \ + func(__VA_ARGS__, \ + decode->op[0].offset, decode->op[1].offset, \ + decode->op[2].offset, vec_len, vec_len); \ +} + +BINARY_INT_GVEC(PCMPGTB, tcg_gen_gvec_cmp, TCG_COND_GT, MO_8) +BINARY_INT_GVEC(PCMPGTW, tcg_gen_gvec_cmp, TCG_COND_GT, MO_16) +BINARY_INT_GVEC(PCMPGTD, tcg_gen_gvec_cmp, TCG_COND_GT, MO_32) + + +/* + * 00 = p* Pq, Qq (if mmx not NULL; no VEX) + * 66 = vp* Vx, Hx, Wx + * + * These are really the same encoding, because 1) V is the same as P when VEX.V + * is not present 2) P and Q are the same as H and W apart from MM/XMM + */ +static inline void gen_binary_int_sse(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode, + SSEFunc_0_eppp mmx, SSEFunc_0_eppp xmm, SSEFunc_0_eppp ymm) +{ + assert(!!mmx == !!(decode->e.special == X86_SPECIAL_MMX)); + + if (mmx && (s->prefix & PREFIX_VEX) && !(s->prefix & PREFIX_DATA)) { + /* VEX encoding is not applicable to MMX instructions. */ + gen_illegal_opcode(s); + return; + } + if (!(s->prefix & PREFIX_DATA)) { + mmx(cpu_env, OP_PTR0, OP_PTR1, OP_PTR2); + } else if (!s->vex_l) { + xmm(cpu_env, OP_PTR0, OP_PTR1, OP_PTR2); + } else { + ymm(cpu_env, OP_PTR0, OP_PTR1, OP_PTR2); + } +} + + +#define BINARY_INT_MMX(uname, lname) \ +static void gen_##uname(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) \ +{ \ + gen_binary_int_sse(s, env, decode, \ + gen_helper_##lname##_mmx, \ + gen_helper_##lname##_xmm, \ + gen_helper_##lname##_ymm); \ +} +BINARY_INT_MMX(PUNPCKLBW, punpcklbw) +BINARY_INT_MMX(PUNPCKLWD, punpcklwd) +BINARY_INT_MMX(PUNPCKLDQ, punpckldq) +BINARY_INT_MMX(PACKSSWB, packsswb) +BINARY_INT_MMX(PACKUSWB, packuswb) +BINARY_INT_MMX(PUNPCKHBW, punpckhbw) +BINARY_INT_MMX(PUNPCKHWD, punpckhwd) +BINARY_INT_MMX(PUNPCKHDQ, punpckhdq) +BINARY_INT_MMX(PACKSSDW, packssdw) + +/* Instructions with no MMX equivalent. */ +#define BINARY_INT_SSE(uname, lname) \ +static void gen_##uname(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) \ +{ \ + gen_binary_int_sse(s, env, decode, \ + NULL, \ + gen_helper_##lname##_xmm, \ + gen_helper_##lname##_ymm); \ +} + +BINARY_INT_SSE(PUNPCKLQDQ, punpcklqdq) +BINARY_INT_SSE(PUNPCKHQDQ, punpckhqdq) + static void gen_ADCOX(DisasContext *s, CPUX86State *env, MemOp ot, int cc_op) { TCGv carry_in = NULL; @@ -383,6 +558,33 @@ static void gen_MOVBE(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) } } +static void gen_MOVD_to(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) +{ + MemOp ot = decode->op[2].ot; + int vec_len = vector_len(s, decode); + int lo_ofs = vector_elem_offset(&decode->op[0], ot, 0); + + tcg_gen_gvec_dup_imm(MO_64, decode->op[0].offset, vec_len, vec_len, 0); + + switch (ot) { + case MO_32: +#ifdef TARGET_X86_64 + tcg_gen_st32_tl(s->T1, cpu_env, lo_ofs); + break; + case MO_64: +#endif + tcg_gen_st_tl(s->T1, cpu_env, lo_ofs); + break; + default: + g_assert_not_reached(); + } +} + +static void gen_MOVDQ(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) +{ + gen_store_sse(s, decode, decode->op[2].offset); +} + static void gen_MULX(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) { MemOp ot = decode->op[0].ot; |