diff options
Diffstat (limited to 'target/i386/tcg/emit.c.inc')
| -rw-r--r-- | target/i386/tcg/emit.c.inc | 220 |
1 files changed, 173 insertions, 47 deletions
diff --git a/target/i386/tcg/emit.c.inc b/target/i386/tcg/emit.c.inc index 82da5488d4..6bcf88ecd7 100644 --- a/target/i386/tcg/emit.c.inc +++ b/target/i386/tcg/emit.c.inc @@ -55,11 +55,6 @@ static void gen_NM_exception(DisasContext *s) gen_exception(s, EXCP07_PREX); } -static void gen_illegal(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) -{ - gen_illegal_opcode(s); -} - static void gen_load_ea(DisasContext *s, AddressParts *mem, bool is_vsib) { TCGv ea = gen_lea_modrm_1(s, *mem, is_vsib); @@ -237,9 +232,30 @@ static void gen_load(DisasContext *s, X86DecodedInsn *decode, int opn, TCGv v) break; case X86_OP_INT: if (op->has_ea) { - gen_op_ld_v(s, op->ot, v, s->A0); + if (v == s->T0 && decode->e.special == X86_SPECIAL_SExtT0) { + gen_op_ld_v(s, op->ot | MO_SIGN, v, s->A0); + } else { + gen_op_ld_v(s, op->ot, v, s->A0); + } + + } else if (op->ot == MO_8 && byte_reg_is_xH(s, op->n)) { + if (v == s->T0 && decode->e.special == X86_SPECIAL_SExtT0) { + tcg_gen_sextract_tl(v, cpu_regs[op->n - 4], 8, 8); + } else { + tcg_gen_extract_tl(v, cpu_regs[op->n - 4], 8, 8); + } + + } else if (op->ot < MO_TL && v == s->T0 && + (decode->e.special == X86_SPECIAL_SExtT0 || + decode->e.special == X86_SPECIAL_ZExtT0)) { + if (decode->e.special == X86_SPECIAL_SExtT0) { + tcg_gen_ext_tl(v, cpu_regs[op->n], op->ot | MO_SIGN); + } else { + tcg_gen_ext_tl(v, cpu_regs[op->n], op->ot); + } + } else { - gen_op_mov_v_reg(s, op->ot, v, op->n); + tcg_gen_mov_tl(v, cpu_regs[op->n]); } break; case X86_OP_IMM: @@ -323,6 +339,19 @@ static inline int vector_len(DisasContext *s, X86DecodedInsn *decode) return s->vex_l ? 32 : 16; } +static void prepare_update1_cc(X86DecodedInsn *decode, DisasContext *s, CCOp op) +{ + decode->cc_dst = s->T0; + decode->cc_op = op; +} + +static void prepare_update2_cc(X86DecodedInsn *decode, DisasContext *s, CCOp op) +{ + decode->cc_src = s->T1; + decode->cc_dst = s->T0; + decode->cc_op = op; +} + static void gen_store_sse(DisasContext *s, X86DecodedInsn *decode, int src_ofs) { MemOp ot = decode->op[0].ot; @@ -1011,6 +1040,7 @@ static void gen_##uname(DisasContext *s, CPUX86State *env, X86DecodedInsn *decod VSIB_AVX(VPGATHERD, vpgatherd) VSIB_AVX(VPGATHERQ, vpgatherq) +/* ADCX/ADOX do not have memory operands and can use set_cc_op. */ static void gen_ADCOX(DisasContext *s, CPUX86State *env, MemOp ot, int cc_op) { int opposite_cc_op; @@ -1073,8 +1103,7 @@ static void gen_ANDN(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) MemOp ot = decode->op[0].ot; tcg_gen_andc_tl(s->T0, s->T1, s->T0); - gen_op_update1_cc(s); - set_cc_op(s, CC_OP_LOGICB + ot); + prepare_update1_cc(decode, s, CC_OP_LOGICB + ot); } static void gen_BEXTR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) @@ -1089,9 +1118,6 @@ static void gen_BEXTR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) * Shifts larger than operand size get zeros. */ tcg_gen_ext8u_tl(s->A0, s->T1); - if (TARGET_LONG_BITS == 64 && ot == MO_32) { - tcg_gen_ext32u_tl(s->T0, s->T0); - } tcg_gen_shr_tl(s->T0, s->T0, s->A0); tcg_gen_movcond_tl(TCG_COND_LEU, s->T0, s->A0, bound, s->T0, zero); @@ -1105,10 +1131,10 @@ static void gen_BEXTR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) tcg_gen_movcond_tl(TCG_COND_LEU, s->T1, s->A0, bound, s->T1, zero); tcg_gen_andc_tl(s->T0, s->T0, s->T1); - gen_op_update1_cc(s); - set_cc_op(s, CC_OP_LOGICB + ot); + prepare_update1_cc(decode, s, CC_OP_LOGICB + ot); } +/* BLSI do not have memory operands and can use set_cc_op. */ static void gen_BLSI(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) { MemOp ot = decode->op[0].ot; @@ -1120,6 +1146,7 @@ static void gen_BLSI(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) set_cc_op(s, CC_OP_BMILGB + ot); } +/* BLSMSK do not have memory operands and can use set_cc_op. */ static void gen_BLSMSK(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) { MemOp ot = decode->op[0].ot; @@ -1131,6 +1158,7 @@ static void gen_BLSMSK(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode set_cc_op(s, CC_OP_BMILGB + ot); } +/* BLSR do not have memory operands and can use set_cc_op. */ static void gen_BLSR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) { MemOp ot = decode->op[0].ot; @@ -1151,18 +1179,119 @@ static void gen_BZHI(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) tcg_gen_ext8u_tl(s->T1, s->T1); + tcg_gen_shl_tl(s->A0, mone, s->T1); + tcg_gen_movcond_tl(TCG_COND_LEU, s->A0, s->T1, bound, s->A0, zero); + tcg_gen_andc_tl(s->T0, s->T0, s->A0); /* * Note that since we're using BMILG (in order to get O * cleared) we need to store the inverse into C. */ - tcg_gen_setcond_tl(TCG_COND_LEU, cpu_cc_src, s->T1, bound); + tcg_gen_setcond_tl(TCG_COND_LEU, s->T1, s->T1, bound); + prepare_update2_cc(decode, s, CC_OP_BMILGB + ot); +} + +static void gen_CMPccXADD(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) +{ + TCGLabel *label_top = gen_new_label(); + TCGLabel *label_bottom = gen_new_label(); + TCGv oldv = tcg_temp_new(); + TCGv newv = tcg_temp_new(); + TCGv cmpv = tcg_temp_new(); + TCGCond cond; + + TCGv cmp_lhs, cmp_rhs; + MemOp ot, ot_full; + + int jcc_op = (decode->b >> 1) & 7; + static const TCGCond cond_table[8] = { + [JCC_O] = TCG_COND_LT, /* test sign bit by comparing against 0 */ + [JCC_B] = TCG_COND_LTU, + [JCC_Z] = TCG_COND_EQ, + [JCC_BE] = TCG_COND_LEU, + [JCC_S] = TCG_COND_LT, /* test sign bit by comparing against 0 */ + [JCC_P] = TCG_COND_EQ, /* even parity - tests low bit of popcount */ + [JCC_L] = TCG_COND_LT, + [JCC_LE] = TCG_COND_LE, + }; - tcg_gen_shl_tl(s->A0, mone, s->T1); - tcg_gen_movcond_tl(TCG_COND_LEU, s->A0, s->T1, bound, s->A0, zero); - tcg_gen_andc_tl(s->T0, s->T0, s->A0); + cond = cond_table[jcc_op]; + if (decode->b & 1) { + cond = tcg_invert_cond(cond); + } - gen_op_update1_cc(s); - set_cc_op(s, CC_OP_BMILGB + ot); + ot = decode->op[0].ot; + ot_full = ot | MO_LE; + if (jcc_op >= JCC_S) { + /* + * Sign-extend values before subtracting for S, P (zero/sign extension + * does not matter there) L, LE and their inverses. + */ + ot_full |= MO_SIGN; + } + + /* + * cmpv will be moved to cc_src *after* cpu_regs[] is written back, so use + * tcg_gen_ext_tl instead of gen_ext_tl. + */ + tcg_gen_ext_tl(cmpv, cpu_regs[decode->op[1].n], ot_full); + + /* + * Cmpxchg loop starts here. + * - s->T1: addition operand (from decoder) + * - s->A0: dest address (from decoder) + * - s->cc_srcT: memory operand (lhs for comparison) + * - cmpv: rhs for comparison + */ + gen_set_label(label_top); + gen_op_ld_v(s, ot_full, s->cc_srcT, s->A0); + tcg_gen_sub_tl(s->T0, s->cc_srcT, cmpv); + + /* Compute the comparison result by hand, to avoid clobbering cc_*. */ + switch (jcc_op) { + case JCC_O: + /* (src1 ^ src2) & (src1 ^ dst). newv is only used here for a moment */ + tcg_gen_xor_tl(newv, s->cc_srcT, s->T0); + tcg_gen_xor_tl(s->tmp0, s->cc_srcT, cmpv); + tcg_gen_and_tl(s->tmp0, s->tmp0, newv); + tcg_gen_sextract_tl(s->tmp0, s->tmp0, 0, 8 << ot); + cmp_lhs = s->tmp0, cmp_rhs = tcg_constant_tl(0); + break; + + case JCC_P: + tcg_gen_ext8u_tl(s->tmp0, s->T0); + tcg_gen_ctpop_tl(s->tmp0, s->tmp0); + tcg_gen_andi_tl(s->tmp0, s->tmp0, 1); + cmp_lhs = s->tmp0, cmp_rhs = tcg_constant_tl(0); + break; + + case JCC_S: + tcg_gen_sextract_tl(s->tmp0, s->T0, 0, 8 << ot); + cmp_lhs = s->tmp0, cmp_rhs = tcg_constant_tl(0); + break; + + default: + cmp_lhs = s->cc_srcT, cmp_rhs = cmpv; + break; + } + + /* Compute new value: if condition does not hold, just store back s->cc_srcT */ + tcg_gen_add_tl(newv, s->cc_srcT, s->T1); + tcg_gen_movcond_tl(cond, newv, cmp_lhs, cmp_rhs, newv, s->cc_srcT); + tcg_gen_atomic_cmpxchg_tl(oldv, s->A0, s->cc_srcT, newv, s->mem_index, ot_full); + + /* Exit unconditionally if cmpxchg succeeded. */ + tcg_gen_brcond_tl(TCG_COND_EQ, oldv, s->cc_srcT, label_bottom); + + /* Try again if there was actually a store to make. */ + tcg_gen_brcond_tl(cond, cmp_lhs, cmp_rhs, label_top); + gen_set_label(label_bottom); + + /* Store old value to registers only after a successful store. */ + gen_writeback(s, decode, 1, s->cc_srcT); + + decode->cc_dst = s->T0; + decode->cc_src = cmpv; + decode->cc_op = CC_OP_SUBB + ot; } static void gen_CRC32(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) @@ -1242,9 +1371,7 @@ static void gen_LDMXCSR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decod static void gen_MASKMOV(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) { - tcg_gen_mov_tl(s->A0, cpu_regs[R_EDI]); - gen_extu(s->aflag, s->A0); - gen_add_A0_ds_seg(s); + gen_lea_v_seg(s, s->aflag, cpu_regs[R_EDI], R_DS, s->override); if (s->prefix & PREFIX_DATA) { gen_helper_maskmov_xmm(tcg_env, OP_PTR1, OP_PTR2, s->A0); @@ -1355,7 +1482,8 @@ static void gen_MULX(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) /* low part of result in VEX.vvvv, high in MODRM */ switch (ot) { - default: + case MO_32: +#ifdef TARGET_X86_64 tcg_gen_trunc_tl_i32(s->tmp2_i32, s->T0); tcg_gen_trunc_tl_i32(s->tmp3_i32, s->T1); tcg_gen_mulu2_i32(s->tmp2_i32, s->tmp3_i32, @@ -1363,13 +1491,15 @@ static void gen_MULX(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) tcg_gen_extu_i32_tl(cpu_regs[s->vex_v], s->tmp2_i32); tcg_gen_extu_i32_tl(s->T0, s->tmp3_i32); break; -#ifdef TARGET_X86_64 + case MO_64: - tcg_gen_mulu2_i64(cpu_regs[s->vex_v], s->T0, s->T0, s->T1); - break; #endif - } + tcg_gen_mulu2_tl(cpu_regs[s->vex_v], s->T0, s->T0, s->T1); + break; + default: + g_assert_not_reached(); + } } static void gen_PALIGNR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) @@ -1432,19 +1562,11 @@ static void gen_PCMPISTRM(DisasContext *s, CPUX86State *env, X86DecodedInsn *dec static void gen_PDEP(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) { - MemOp ot = decode->op[1].ot; - if (ot < MO_64) { - tcg_gen_ext32u_tl(s->T0, s->T0); - } gen_helper_pdep(s->T0, s->T0, s->T1); } static void gen_PEXT(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) { - MemOp ot = decode->op[1].ot; - if (ot < MO_64) { - tcg_gen_ext32u_tl(s->T0, s->T0); - } gen_helper_pext(s->T0, s->T0, s->T1); } @@ -1772,14 +1894,24 @@ static void gen_PSLLDQ_i(DisasContext *s, CPUX86State *env, X86DecodedInsn *deco static void gen_RORX(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) { MemOp ot = decode->op[0].ot; - int b = decode->immediate; + int mask = ot == MO_64 ? 63 : 31; + int b = decode->immediate & mask; - if (ot == MO_64) { - tcg_gen_rotri_tl(s->T0, s->T0, b & 63); - } else { + switch (ot) { + case MO_32: +#ifdef TARGET_X86_64 tcg_gen_trunc_tl_i32(s->tmp2_i32, s->T0); - tcg_gen_rotri_i32(s->tmp2_i32, s->tmp2_i32, b & 31); + tcg_gen_rotri_i32(s->tmp2_i32, s->tmp2_i32, b); tcg_gen_extu_i32_tl(s->T0, s->tmp2_i32); + break; + + case MO_64: +#endif + tcg_gen_rotri_tl(s->T0, s->T0, b); + break; + + default: + g_assert_not_reached(); } } @@ -1790,9 +1922,6 @@ static void gen_SARX(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) mask = ot == MO_64 ? 63 : 31; tcg_gen_andi_tl(s->T1, s->T1, mask); - if (ot != MO_64) { - tcg_gen_ext32s_tl(s->T0, s->T0); - } tcg_gen_sar_tl(s->T0, s->T0, s->T1); } @@ -1867,9 +1996,6 @@ static void gen_SHRX(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode) mask = ot == MO_64 ? 63 : 31; tcg_gen_andi_tl(s->T1, s->T1, mask); - if (ot != MO_64) { - tcg_gen_ext32u_tl(s->T0, s->T0); - } tcg_gen_shr_tl(s->T0, s->T0, s->T1); } |