diff options
Diffstat (limited to 'tcg/ppc')
-rw-r--r-- | tcg/ppc/tcg-target-con-set.h | 12 | ||||
-rw-r--r-- | tcg/ppc/tcg-target-con-str.h | 1 | ||||
-rw-r--r-- | tcg/ppc/tcg-target-has.h | 59 | ||||
-rw-r--r-- | tcg/ppc/tcg-target.c.inc | 1948 |
4 files changed, 1078 insertions, 942 deletions
diff --git a/tcg/ppc/tcg-target-con-set.h b/tcg/ppc/tcg-target-con-set.h index 453abde..da7a383 100644 --- a/tcg/ppc/tcg-target-con-set.h +++ b/tcg/ppc/tcg-target-con-set.h @@ -15,29 +15,29 @@ C_O0_I2(r, rC) C_O0_I2(v, r) C_O0_I3(r, r, r) C_O0_I3(o, m, r) -C_O0_I4(r, r, ri, ri) +C_O0_I4(r, r, rU, rC) C_O0_I4(r, r, r, r) C_O1_I1(r, r) C_O1_I1(v, r) C_O1_I1(v, v) C_O1_I1(v, vr) C_O1_I2(r, 0, rZ) -C_O1_I2(r, rI, ri) -C_O1_I2(r, rI, rT) +C_O1_I2(r, rI, r) C_O1_I2(r, r, r) C_O1_I2(r, r, ri) C_O1_I2(r, r, rC) C_O1_I2(r, r, rI) C_O1_I2(r, r, rT) C_O1_I2(r, r, rU) +C_O1_I2(r, r, rZM) C_O1_I2(r, r, rZW) +C_O1_I2(r, rI, rN) +C_O1_I2(r, rZM, rZM) C_O1_I2(v, v, v) C_O1_I3(v, v, v, v) C_O1_I4(v, v, v, vZM, v) C_O1_I4(r, r, rC, rZ, rZ) -C_O1_I4(r, r, r, ri, ri) +C_O1_I4(r, r, r, rU, rC) C_O2_I1(r, r, r) C_N1O1_I1(o, m, r) C_O2_I2(r, r, r, r) -C_O2_I4(r, r, rI, rZM, r, r) -C_O2_I4(r, r, r, r, rI, rZM) diff --git a/tcg/ppc/tcg-target-con-str.h b/tcg/ppc/tcg-target-con-str.h index 16b6872..faf92da 100644 --- a/tcg/ppc/tcg-target-con-str.h +++ b/tcg/ppc/tcg-target-con-str.h @@ -19,6 +19,7 @@ REGS('v', ALL_VECTOR_REGS) CONST('C', TCG_CT_CONST_CMP) CONST('I', TCG_CT_CONST_S16) CONST('M', TCG_CT_CONST_MONE) +CONST('N', TCG_CT_CONST_N16) CONST('T', TCG_CT_CONST_S32) CONST('U', TCG_CT_CONST_U32) CONST('W', TCG_CT_CONST_WSZ) diff --git a/tcg/ppc/tcg-target-has.h b/tcg/ppc/tcg-target-has.h index 6db91f7..81ec5ae 100644 --- a/tcg/ppc/tcg-target-has.h +++ b/tcg/ppc/tcg-target-has.h @@ -16,68 +16,9 @@ #define have_altivec (cpuinfo & CPUINFO_ALTIVEC) #define have_vsx (cpuinfo & CPUINFO_VSX) -/* optional instructions automatically implemented */ -#define TCG_TARGET_HAS_ext8u_i32 0 /* andi */ -#define TCG_TARGET_HAS_ext16u_i32 0 - /* optional instructions */ -#define TCG_TARGET_HAS_div_i32 1 -#define TCG_TARGET_HAS_rem_i32 have_isa_3_00 -#define TCG_TARGET_HAS_rot_i32 1 -#define TCG_TARGET_HAS_ext8s_i32 1 -#define TCG_TARGET_HAS_ext16s_i32 1 -#define TCG_TARGET_HAS_bswap16_i32 1 -#define TCG_TARGET_HAS_bswap32_i32 1 -#define TCG_TARGET_HAS_not_i32 1 -#define TCG_TARGET_HAS_andc_i32 1 -#define TCG_TARGET_HAS_orc_i32 1 -#define TCG_TARGET_HAS_eqv_i32 1 -#define TCG_TARGET_HAS_nand_i32 1 -#define TCG_TARGET_HAS_nor_i32 1 -#define TCG_TARGET_HAS_clz_i32 1 -#define TCG_TARGET_HAS_ctz_i32 have_isa_3_00 -#define TCG_TARGET_HAS_ctpop_i32 have_isa_2_06 -#define TCG_TARGET_HAS_extract2_i32 0 -#define TCG_TARGET_HAS_negsetcond_i32 1 -#define TCG_TARGET_HAS_mulu2_i32 0 -#define TCG_TARGET_HAS_muls2_i32 0 -#define TCG_TARGET_HAS_muluh_i32 1 -#define TCG_TARGET_HAS_mulsh_i32 1 -#define TCG_TARGET_HAS_qemu_st8_i32 0 - #if TCG_TARGET_REG_BITS == 64 -#define TCG_TARGET_HAS_add2_i32 0 -#define TCG_TARGET_HAS_sub2_i32 0 #define TCG_TARGET_HAS_extr_i64_i32 0 -#define TCG_TARGET_HAS_div_i64 1 -#define TCG_TARGET_HAS_rem_i64 have_isa_3_00 -#define TCG_TARGET_HAS_rot_i64 1 -#define TCG_TARGET_HAS_ext8s_i64 1 -#define TCG_TARGET_HAS_ext16s_i64 1 -#define TCG_TARGET_HAS_ext32s_i64 1 -#define TCG_TARGET_HAS_ext8u_i64 0 -#define TCG_TARGET_HAS_ext16u_i64 0 -#define TCG_TARGET_HAS_ext32u_i64 0 -#define TCG_TARGET_HAS_bswap16_i64 1 -#define TCG_TARGET_HAS_bswap32_i64 1 -#define TCG_TARGET_HAS_bswap64_i64 1 -#define TCG_TARGET_HAS_not_i64 1 -#define TCG_TARGET_HAS_andc_i64 1 -#define TCG_TARGET_HAS_orc_i64 1 -#define TCG_TARGET_HAS_eqv_i64 1 -#define TCG_TARGET_HAS_nand_i64 1 -#define TCG_TARGET_HAS_nor_i64 1 -#define TCG_TARGET_HAS_clz_i64 1 -#define TCG_TARGET_HAS_ctz_i64 have_isa_3_00 -#define TCG_TARGET_HAS_ctpop_i64 have_isa_2_06 -#define TCG_TARGET_HAS_extract2_i64 0 -#define TCG_TARGET_HAS_negsetcond_i64 1 -#define TCG_TARGET_HAS_add2_i64 1 -#define TCG_TARGET_HAS_sub2_i64 1 -#define TCG_TARGET_HAS_mulu2_i64 0 -#define TCG_TARGET_HAS_muls2_i64 0 -#define TCG_TARGET_HAS_muluh_i64 1 -#define TCG_TARGET_HAS_mulsh_i64 1 #endif #define TCG_TARGET_HAS_qemu_ldst_i128 \ diff --git a/tcg/ppc/tcg-target.c.inc b/tcg/ppc/tcg-target.c.inc index 822925a..b8b23d4 100644 --- a/tcg/ppc/tcg-target.c.inc +++ b/tcg/ppc/tcg-target.c.inc @@ -89,14 +89,15 @@ /* Shorthand for size of a register. */ #define SZR (TCG_TARGET_REG_BITS / 8) -#define TCG_CT_CONST_S16 0x100 -#define TCG_CT_CONST_U16 0x200 -#define TCG_CT_CONST_S32 0x400 -#define TCG_CT_CONST_U32 0x800 -#define TCG_CT_CONST_ZERO 0x1000 -#define TCG_CT_CONST_MONE 0x2000 -#define TCG_CT_CONST_WSZ 0x4000 -#define TCG_CT_CONST_CMP 0x8000 +#define TCG_CT_CONST_S16 0x00100 +#define TCG_CT_CONST_U16 0x00200 +#define TCG_CT_CONST_N16 0x00400 +#define TCG_CT_CONST_S32 0x00800 +#define TCG_CT_CONST_U32 0x01000 +#define TCG_CT_CONST_ZERO 0x02000 +#define TCG_CT_CONST_MONE 0x04000 +#define TCG_CT_CONST_WSZ 0x08000 +#define TCG_CT_CONST_CMP 0x10000 #define ALL_GENERAL_REGS 0xffffffffu #define ALL_VECTOR_REGS 0xffffffff00000000ull @@ -342,6 +343,9 @@ static bool tcg_target_const_match(int64_t sval, int ct, if ((ct & TCG_CT_CONST_U16) && uval == (uint16_t)uval) { return 1; } + if ((ct & TCG_CT_CONST_N16) && -sval == (int16_t)-sval) { + return 1; + } if ((ct & TCG_CT_CONST_S32) && sval == (int32_t)sval) { return 1; } @@ -1012,111 +1016,6 @@ static void tcg_out_addpcis(TCGContext *s, TCGReg dst, intptr_t imm) tcg_out32(s, ADDPCIS | RT(dst) | (d1 << 16) | (d0 << 6) | d2); } -static void tcg_out_bswap16(TCGContext *s, TCGReg dst, TCGReg src, int flags) -{ - TCGReg tmp = dst == src ? TCG_REG_R0 : dst; - - if (have_isa_3_10) { - tcg_out32(s, BRH | RA(dst) | RS(src)); - if (flags & TCG_BSWAP_OS) { - tcg_out_ext16s(s, TCG_TYPE_REG, dst, dst); - } else if ((flags & (TCG_BSWAP_IZ | TCG_BSWAP_OZ)) == TCG_BSWAP_OZ) { - tcg_out_ext16u(s, dst, dst); - } - return; - } - - /* - * In the following, - * dep(a, b, m) -> (a & ~m) | (b & m) - * - * Begin with: src = xxxxabcd - */ - /* tmp = rol32(src, 24) & 0x000000ff = 0000000c */ - tcg_out_rlw(s, RLWINM, tmp, src, 24, 24, 31); - /* tmp = dep(tmp, rol32(src, 8), 0x0000ff00) = 000000dc */ - tcg_out_rlw(s, RLWIMI, tmp, src, 8, 16, 23); - - if (flags & TCG_BSWAP_OS) { - tcg_out_ext16s(s, TCG_TYPE_REG, dst, tmp); - } else { - tcg_out_mov(s, TCG_TYPE_REG, dst, tmp); - } -} - -static void tcg_out_bswap32(TCGContext *s, TCGReg dst, TCGReg src, int flags) -{ - TCGReg tmp = dst == src ? TCG_REG_R0 : dst; - - if (have_isa_3_10) { - tcg_out32(s, BRW | RA(dst) | RS(src)); - if (flags & TCG_BSWAP_OS) { - tcg_out_ext32s(s, dst, dst); - } else if ((flags & (TCG_BSWAP_IZ | TCG_BSWAP_OZ)) == TCG_BSWAP_OZ) { - tcg_out_ext32u(s, dst, dst); - } - return; - } - - /* - * Stolen from gcc's builtin_bswap32. - * In the following, - * dep(a, b, m) -> (a & ~m) | (b & m) - * - * Begin with: src = xxxxabcd - */ - /* tmp = rol32(src, 8) & 0xffffffff = 0000bcda */ - tcg_out_rlw(s, RLWINM, tmp, src, 8, 0, 31); - /* tmp = dep(tmp, rol32(src, 24), 0xff000000) = 0000dcda */ - tcg_out_rlw(s, RLWIMI, tmp, src, 24, 0, 7); - /* tmp = dep(tmp, rol32(src, 24), 0x0000ff00) = 0000dcba */ - tcg_out_rlw(s, RLWIMI, tmp, src, 24, 16, 23); - - if (flags & TCG_BSWAP_OS) { - tcg_out_ext32s(s, dst, tmp); - } else { - tcg_out_mov(s, TCG_TYPE_REG, dst, tmp); - } -} - -static void tcg_out_bswap64(TCGContext *s, TCGReg dst, TCGReg src) -{ - TCGReg t0 = dst == src ? TCG_REG_R0 : dst; - TCGReg t1 = dst == src ? dst : TCG_REG_R0; - - if (have_isa_3_10) { - tcg_out32(s, BRD | RA(dst) | RS(src)); - return; - } - - /* - * In the following, - * dep(a, b, m) -> (a & ~m) | (b & m) - * - * Begin with: src = abcdefgh - */ - /* t0 = rol32(src, 8) & 0xffffffff = 0000fghe */ - tcg_out_rlw(s, RLWINM, t0, src, 8, 0, 31); - /* t0 = dep(t0, rol32(src, 24), 0xff000000) = 0000hghe */ - tcg_out_rlw(s, RLWIMI, t0, src, 24, 0, 7); - /* t0 = dep(t0, rol32(src, 24), 0x0000ff00) = 0000hgfe */ - tcg_out_rlw(s, RLWIMI, t0, src, 24, 16, 23); - - /* t0 = rol64(t0, 32) = hgfe0000 */ - tcg_out_rld(s, RLDICL, t0, t0, 32, 0); - /* t1 = rol64(src, 32) = efghabcd */ - tcg_out_rld(s, RLDICL, t1, src, 32, 0); - - /* t0 = dep(t0, rol32(t1, 24), 0xffffffff) = hgfebcda */ - tcg_out_rlw(s, RLWIMI, t0, t1, 8, 0, 31); - /* t0 = dep(t0, rol32(t1, 24), 0xff000000) = hgfedcda */ - tcg_out_rlw(s, RLWIMI, t0, t1, 24, 0, 7); - /* t0 = dep(t0, rol32(t1, 24), 0x0000ff00) = hgfedcba */ - tcg_out_rlw(s, RLWIMI, t0, t1, 24, 16, 23); - - tcg_out_mov(s, TCG_TYPE_REG, dst, t0); -} - /* Emit a move into ret of arg, if it can be done in one insn. */ static bool tcg_out_movi_one(TCGContext *s, TCGReg ret, tcg_target_long arg) { @@ -1777,9 +1676,8 @@ static void tcg_out_test(TCGContext *s, TCGReg dest, TCGReg arg1, TCGArg arg2, } static void tcg_out_cmp(TCGContext *s, int cond, TCGArg arg1, TCGArg arg2, - int const_arg2, int cr, TCGType type) + bool const_arg2, int cr, TCGType type) { - int imm; uint32_t op; tcg_debug_assert(TCG_TARGET_REG_BITS == 64 || type == TCG_TYPE_I32); @@ -1796,18 +1694,15 @@ static void tcg_out_cmp(TCGContext *s, int cond, TCGArg arg1, TCGArg arg2, case TCG_COND_EQ: case TCG_COND_NE: if (const_arg2) { - if ((int16_t) arg2 == arg2) { + if ((int16_t)arg2 == arg2) { op = CMPI; - imm = 1; - break; - } else if ((uint16_t) arg2 == arg2) { - op = CMPLI; - imm = 1; break; } + tcg_debug_assert((uint16_t)arg2 == arg2); + op = CMPLI; + break; } op = CMPL; - imm = 0; break; case TCG_COND_TSTEQ: @@ -1821,14 +1716,11 @@ static void tcg_out_cmp(TCGContext *s, int cond, TCGArg arg1, TCGArg arg2, case TCG_COND_LE: case TCG_COND_GT: if (const_arg2) { - if ((int16_t) arg2 == arg2) { - op = CMPI; - imm = 1; - break; - } + tcg_debug_assert((int16_t)arg2 == arg2); + op = CMPI; + break; } op = CMP; - imm = 0; break; case TCG_COND_LTU: @@ -1836,30 +1728,20 @@ static void tcg_out_cmp(TCGContext *s, int cond, TCGArg arg1, TCGArg arg2, case TCG_COND_LEU: case TCG_COND_GTU: if (const_arg2) { - if ((uint16_t) arg2 == arg2) { - op = CMPLI; - imm = 1; - break; - } + tcg_debug_assert((uint16_t)arg2 == arg2); + op = CMPLI; + break; } op = CMPL; - imm = 0; break; default: g_assert_not_reached(); } op |= BF(cr) | ((type == TCG_TYPE_I64) << 21); - - if (imm) { - tcg_out32(s, op | RA(arg1) | (arg2 & 0xffff)); - } else { - if (const_arg2) { - tcg_out_movi(s, type, TCG_REG_R0, arg2); - arg2 = TCG_REG_R0; - } - tcg_out32(s, op | RA(arg1) | RB(arg2)); - } + op |= RA(arg1); + op |= const_arg2 ? arg2 & 0xffff : RB(arg2); + tcg_out32(s, op); } static void tcg_out_setcond_eq0(TCGContext *s, TCGType type, @@ -1926,8 +1808,8 @@ static TCGReg tcg_gen_setcond_xor(TCGContext *s, TCGReg arg1, TCGArg arg2, } static void tcg_out_setcond(TCGContext *s, TCGType type, TCGCond cond, - TCGArg arg0, TCGArg arg1, TCGArg arg2, - int const_arg2, bool neg) + TCGReg arg0, TCGReg arg1, TCGArg arg2, + bool const_arg2, bool neg) { int sh; bool inv; @@ -2072,6 +1954,54 @@ static void tcg_out_setcond(TCGContext *s, TCGType type, TCGCond cond, } } +static void tgen_setcond(TCGContext *s, TCGType type, TCGCond cond, + TCGReg dest, TCGReg arg1, TCGReg arg2) +{ + tcg_out_setcond(s, type, cond, dest, arg1, arg2, false, false); +} + +static void tgen_setcondi(TCGContext *s, TCGType type, TCGCond cond, + TCGReg dest, TCGReg arg1, tcg_target_long arg2) +{ + tcg_out_setcond(s, type, cond, dest, arg1, arg2, true, false); +} + +static const TCGOutOpSetcond outop_setcond = { + .base.static_constraint = C_O1_I2(r, r, rC), + .out_rrr = tgen_setcond, + .out_rri = tgen_setcondi, +}; + +static void tgen_negsetcond(TCGContext *s, TCGType type, TCGCond cond, + TCGReg dest, TCGReg arg1, TCGReg arg2) +{ + tcg_out_setcond(s, type, cond, dest, arg1, arg2, false, true); +} + +static void tgen_negsetcondi(TCGContext *s, TCGType type, TCGCond cond, + TCGReg dest, TCGReg arg1, tcg_target_long arg2) +{ + tcg_out_setcond(s, type, cond, dest, arg1, arg2, true, true); +} + +static const TCGOutOpSetcond outop_negsetcond = { + .base.static_constraint = C_O1_I2(r, r, rC), + .out_rrr = tgen_negsetcond, + .out_rri = tgen_negsetcondi, +}; + +void tcg_out_br(TCGContext *s, TCGLabel *l) +{ + uint32_t insn = B; + + if (l->has_value) { + insn |= reloc_pc24_val(tcg_splitwx_to_rx(s->code_ptr), l->u.value_ptr); + } else { + tcg_out_reloc(s, s->code_ptr, R_PPC_REL24, l, 0); + } + tcg_out32(s, insn); +} + static void tcg_out_bc(TCGContext *s, TCGCond cond, int bd) { tcg_out32(s, tcg_to_bc[cond] | bd); @@ -2088,17 +2018,29 @@ static void tcg_out_bc_lab(TCGContext *s, TCGCond cond, TCGLabel *l) tcg_out_bc(s, cond, bd); } -static void tcg_out_brcond(TCGContext *s, TCGCond cond, - TCGArg arg1, TCGArg arg2, int const_arg2, - TCGLabel *l, TCGType type) +static void tgen_brcond(TCGContext *s, TCGType type, TCGCond cond, + TCGReg arg1, TCGReg arg2, TCGLabel *l) +{ + tcg_out_cmp(s, cond, arg1, arg2, false, 0, type); + tcg_out_bc_lab(s, cond, l); +} + +static void tgen_brcondi(TCGContext *s, TCGType type, TCGCond cond, + TCGReg arg1, tcg_target_long arg2, TCGLabel *l) { - tcg_out_cmp(s, cond, arg1, arg2, const_arg2, 0, type); + tcg_out_cmp(s, cond, arg1, arg2, true, 0, type); tcg_out_bc_lab(s, cond, l); } -static void tcg_out_movcond(TCGContext *s, TCGType type, TCGCond cond, - TCGArg dest, TCGArg c1, TCGArg c2, TCGArg v1, - TCGArg v2, bool const_c2) +static const TCGOutOpBrcond outop_brcond = { + .base.static_constraint = C_O0_I2(r, rC), + .out_rr = tgen_brcond, + .out_ri = tgen_brcondi, +}; + +static void tgen_movcond(TCGContext *s, TCGType type, TCGCond cond, + TCGReg dest, TCGReg c1, TCGArg c2, bool const_c2, + TCGArg v1, bool const_v1, TCGArg v2, bool const_v2) { /* If for some reason both inputs are zero, don't produce bad code. */ if (v1 == 0 && v2 == 0) { @@ -2144,6 +2086,11 @@ static void tcg_out_movcond(TCGContext *s, TCGType type, TCGCond cond, } } +static const TCGOutOpMovcond outop_movcond = { + .base.static_constraint = C_O1_I4(r, r, rC, rZ, rZ), + .out = tgen_movcond, +}; + static void tcg_out_cntxz(TCGContext *s, TCGType type, uint32_t opc, TCGArg a0, TCGArg a1, TCGArg a2, bool const_a2) { @@ -2170,8 +2117,8 @@ static void tcg_out_cntxz(TCGContext *s, TCGType type, uint32_t opc, } } -static void tcg_out_cmp2(TCGContext *s, const TCGArg *args, - const int *const_args) +static void tcg_out_cmp2(TCGContext *s, TCGCond cond, TCGReg al, TCGReg ah, + TCGArg bl, bool blconst, TCGArg bh, bool bhconst) { static const struct { uint8_t bit1, bit2; } bits[] = { [TCG_COND_LT ] = { CR_LT, CR_LT }, @@ -2184,18 +2131,9 @@ static void tcg_out_cmp2(TCGContext *s, const TCGArg *args, [TCG_COND_GEU] = { CR_GT, CR_LT }, }; - TCGCond cond = args[4], cond2; - TCGArg al, ah, bl, bh; - int blconst, bhconst; + TCGCond cond2; int op, bit1, bit2; - al = args[0]; - ah = args[1]; - bl = args[2]; - bh = args[3]; - blconst = const_args[2]; - bhconst = const_args[3]; - switch (cond) { case TCG_COND_EQ: op = CRAND; @@ -2247,22 +2185,42 @@ static void tcg_out_cmp2(TCGContext *s, const TCGArg *args, } } -static void tcg_out_setcond2(TCGContext *s, const TCGArg *args, - const int *const_args) +static void tgen_setcond2(TCGContext *s, TCGCond cond, TCGReg ret, + TCGReg al, TCGReg ah, + TCGArg bl, bool const_bl, + TCGArg bh, bool const_bh) { - tcg_out_cmp2(s, args + 1, const_args + 1); + tcg_out_cmp2(s, cond, al, ah, bl, const_bl, bh, const_bh); tcg_out32(s, MFOCRF | RT(TCG_REG_R0) | FXM(0)); - tcg_out_rlw(s, RLWINM, args[0], TCG_REG_R0, CR_EQ + 0*4 + 1, 31, 31); + tcg_out_rlw(s, RLWINM, ret, TCG_REG_R0, CR_EQ + 0*4 + 1, 31, 31); } -static void tcg_out_brcond2(TCGContext *s, const TCGArg *args, - const int *const_args) +#if TCG_TARGET_REG_BITS != 32 +__attribute__((unused)) +#endif +static const TCGOutOpSetcond2 outop_setcond2 = { + .base.static_constraint = C_O1_I4(r, r, r, rU, rC), + .out = tgen_setcond2, +}; + +static void tgen_brcond2(TCGContext *s, TCGCond cond, TCGReg al, TCGReg ah, + TCGArg bl, bool const_bl, + TCGArg bh, bool const_bh, TCGLabel *l) { - tcg_out_cmp2(s, args, const_args); - tcg_out_bc_lab(s, TCG_COND_EQ, arg_label(args[5])); + assert(TCG_TARGET_REG_BITS == 32); + tcg_out_cmp2(s, cond, al, ah, bl, const_bl, bh, const_bh); + tcg_out_bc_lab(s, TCG_COND_EQ, l); } -static void tcg_out_mb(TCGContext *s, TCGArg a0) +#if TCG_TARGET_REG_BITS != 32 +__attribute__((unused)) +#endif +static const TCGOutOpBrcond2 outop_brcond2 = { + .base.static_constraint = C_O0_I4(r, r, rU, rC), + .out = tgen_brcond2, +}; + +static void tcg_out_mb(TCGContext *s, unsigned a0) { uint32_t insn; @@ -2482,10 +2440,10 @@ static TCGLabelQemuLdst *prepare_host_addr(TCGContext *s, HostAddress *h, /* Extract the page index, shifted into place for tlb index. */ if (TCG_TARGET_REG_BITS == 32) { tcg_out_shri32(s, TCG_REG_R0, addr, - s->page_bits - CPU_TLB_ENTRY_BITS); + TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS); } else { tcg_out_shri64(s, TCG_REG_R0, addr, - s->page_bits - CPU_TLB_ENTRY_BITS); + TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS); } tcg_out32(s, AND | SAB(TCG_REG_TMP1, TCG_REG_TMP1, TCG_REG_R0)); @@ -2522,7 +2480,7 @@ static TCGLabelQemuLdst *prepare_host_addr(TCGContext *s, HostAddress *h, a_bits = s_bits; } tcg_out_rlw(s, RLWINM, TCG_REG_R0, addr, 0, - (32 - a_bits) & 31, 31 - s->page_bits); + (32 - a_bits) & 31, 31 - TARGET_PAGE_BITS); } else { TCGReg t = addr; @@ -2543,13 +2501,13 @@ static TCGLabelQemuLdst *prepare_host_addr(TCGContext *s, HostAddress *h, /* Mask the address for the requested alignment. */ if (addr_type == TCG_TYPE_I32) { tcg_out_rlw(s, RLWINM, TCG_REG_R0, t, 0, - (32 - a_bits) & 31, 31 - s->page_bits); + (32 - a_bits) & 31, 31 - TARGET_PAGE_BITS); } else if (a_bits == 0) { - tcg_out_rld(s, RLDICR, TCG_REG_R0, t, 0, 63 - s->page_bits); + tcg_out_rld(s, RLDICR, TCG_REG_R0, t, 0, 63 - TARGET_PAGE_BITS); } else { tcg_out_rld(s, RLDICL, TCG_REG_R0, t, - 64 - s->page_bits, s->page_bits - a_bits); - tcg_out_rld(s, RLDICL, TCG_REG_R0, TCG_REG_R0, s->page_bits, 0); + 64 - TARGET_PAGE_BITS, TARGET_PAGE_BITS - a_bits); + tcg_out_rld(s, RLDICL, TCG_REG_R0, TCG_REG_R0, TARGET_PAGE_BITS, 0); } } @@ -2737,6 +2695,60 @@ static void tcg_out_qemu_ldst_i128(TCGContext *s, TCGReg datalo, TCGReg datahi, } } +static void tgen_qemu_ld(TCGContext *s, TCGType type, TCGReg data, + TCGReg addr, MemOpIdx oi) +{ + tcg_out_qemu_ld(s, data, -1, addr, oi, type); +} + +static const TCGOutOpQemuLdSt outop_qemu_ld = { + .base.static_constraint = C_O1_I1(r, r), + .out = tgen_qemu_ld, +}; + +static void tgen_qemu_ld2(TCGContext *s, TCGType type, TCGReg datalo, + TCGReg datahi, TCGReg addr, MemOpIdx oi) +{ + if (TCG_TARGET_REG_BITS == 32) { + tcg_out_qemu_ld(s, datalo, datahi, addr, oi, type); + } else { + tcg_out_qemu_ldst_i128(s, datalo, datahi, addr, oi, true); + } +} + +static const TCGOutOpQemuLdSt2 outop_qemu_ld2 = { + .base.static_constraint = + TCG_TARGET_REG_BITS == 64 ? C_N1O1_I1(o, m, r) : C_O2_I1(r, r, r), + .out = tgen_qemu_ld2, +}; + +static void tgen_qemu_st(TCGContext *s, TCGType type, TCGReg data, + TCGReg addr, MemOpIdx oi) +{ + tcg_out_qemu_st(s, data, -1, addr, oi, type); +} + +static const TCGOutOpQemuLdSt outop_qemu_st = { + .base.static_constraint = C_O0_I2(r, r), + .out = tgen_qemu_st, +}; + +static void tgen_qemu_st2(TCGContext *s, TCGType type, TCGReg datalo, + TCGReg datahi, TCGReg addr, MemOpIdx oi) +{ + if (TCG_TARGET_REG_BITS == 32) { + tcg_out_qemu_st(s, datalo, datahi, addr, oi, type); + } else { + tcg_out_qemu_ldst_i128(s, datalo, datahi, addr, oi, false); + } +} + +static const TCGOutOpQemuLdSt2 outop_qemu_st2 = { + .base.static_constraint = + TCG_TARGET_REG_BITS == 64 ? C_O0_I3(o, m, r) : C_O0_I3(r, r, r), + .out = tgen_qemu_st2, +}; + static void tcg_out_nop_fill(tcg_insn_unit *p, int count) { int i; @@ -2885,6 +2897,13 @@ static void tcg_out_goto_tb(TCGContext *s, int which) set_jmp_reset_offset(s, which); } +static void tcg_out_goto_ptr(TCGContext *s, TCGReg a0) +{ + tcg_out32(s, MTSPR | RS(a0) | CTR); + tcg_out32(s, ADDI | TAI(TCG_REG_R3, 0, 0)); + tcg_out32(s, BCCTR | BO_ALWAYS); +} + void tb_target_set_jmp_target(const TranslationBlock *tb, int n, uintptr_t jmp_rx, uintptr_t jmp_rw) { @@ -2902,595 +2921,913 @@ void tb_target_set_jmp_target(const TranslationBlock *tb, int n, flush_idcache_range(jmp_rx, jmp_rw, 4); } -static void tcg_out_op(TCGContext *s, TCGOpcode opc, TCGType type, - const TCGArg args[TCG_MAX_OP_ARGS], - const int const_args[TCG_MAX_OP_ARGS]) + +static void tgen_add(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) { - TCGArg a0, a1, a2; + tcg_out32(s, ADD | TAB(a0, a1, a2)); +} - switch (opc) { - case INDEX_op_goto_ptr: - tcg_out32(s, MTSPR | RS(args[0]) | CTR); - tcg_out32(s, ADDI | TAI(TCG_REG_R3, 0, 0)); - tcg_out32(s, BCCTR | BO_ALWAYS); - break; - case INDEX_op_br: - { - TCGLabel *l = arg_label(args[0]); - uint32_t insn = B; - - if (l->has_value) { - insn |= reloc_pc24_val(tcg_splitwx_to_rx(s->code_ptr), - l->u.value_ptr); - } else { - tcg_out_reloc(s, s->code_ptr, R_PPC_REL24, l, 0); - } - tcg_out32(s, insn); - } - break; - case INDEX_op_ld8u_i32: - case INDEX_op_ld8u_i64: - tcg_out_mem_long(s, LBZ, LBZX, args[0], args[1], args[2]); - break; - case INDEX_op_ld8s_i32: - case INDEX_op_ld8s_i64: - tcg_out_mem_long(s, LBZ, LBZX, args[0], args[1], args[2]); - tcg_out_ext8s(s, TCG_TYPE_REG, args[0], args[0]); - break; - case INDEX_op_ld16u_i32: - case INDEX_op_ld16u_i64: - tcg_out_mem_long(s, LHZ, LHZX, args[0], args[1], args[2]); - break; - case INDEX_op_ld16s_i32: - case INDEX_op_ld16s_i64: - tcg_out_mem_long(s, LHA, LHAX, args[0], args[1], args[2]); - break; - case INDEX_op_ld_i32: - case INDEX_op_ld32u_i64: - tcg_out_mem_long(s, LWZ, LWZX, args[0], args[1], args[2]); - break; - case INDEX_op_ld32s_i64: - tcg_out_mem_long(s, LWA, LWAX, args[0], args[1], args[2]); - break; - case INDEX_op_ld_i64: - tcg_out_mem_long(s, LD, LDX, args[0], args[1], args[2]); - break; - case INDEX_op_st8_i32: - case INDEX_op_st8_i64: - tcg_out_mem_long(s, STB, STBX, args[0], args[1], args[2]); - break; - case INDEX_op_st16_i32: - case INDEX_op_st16_i64: - tcg_out_mem_long(s, STH, STHX, args[0], args[1], args[2]); - break; - case INDEX_op_st_i32: - case INDEX_op_st32_i64: - tcg_out_mem_long(s, STW, STWX, args[0], args[1], args[2]); - break; - case INDEX_op_st_i64: - tcg_out_mem_long(s, STD, STDX, args[0], args[1], args[2]); - break; +static void tgen_addi(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + tcg_out_mem_long(s, ADDI, ADD, a0, a1, a2); +} - case INDEX_op_add_i32: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - do_addi_32: - tcg_out_mem_long(s, ADDI, ADD, a0, a1, (int32_t)a2); - } else { - tcg_out32(s, ADD | TAB(a0, a1, a2)); - } - break; - case INDEX_op_sub_i32: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[1]) { - if (const_args[2]) { - tcg_out_movi(s, TCG_TYPE_I32, a0, a1 - a2); - } else { - tcg_out32(s, SUBFIC | TAI(a0, a2, a1)); - } - } else if (const_args[2]) { - a2 = -a2; - goto do_addi_32; - } else { - tcg_out32(s, SUBF | TAB(a0, a2, a1)); - } - break; +static const TCGOutOpBinary outop_add = { + .base.static_constraint = C_O1_I2(r, r, rT), + .out_rrr = tgen_add, + .out_rri = tgen_addi, +}; - case INDEX_op_and_i32: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - tcg_out_andi32(s, a0, a1, a2); - } else { - tcg_out32(s, AND | SAB(a1, a0, a2)); - } - break; - case INDEX_op_and_i64: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - tcg_out_andi64(s, a0, a1, a2); - } else { - tcg_out32(s, AND | SAB(a1, a0, a2)); - } - break; - case INDEX_op_or_i64: - case INDEX_op_or_i32: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - tcg_out_ori32(s, a0, a1, a2); - } else { - tcg_out32(s, OR | SAB(a1, a0, a2)); - } - break; - case INDEX_op_xor_i64: - case INDEX_op_xor_i32: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - tcg_out_xori32(s, a0, a1, a2); - } else { - tcg_out32(s, XOR | SAB(a1, a0, a2)); - } - break; - case INDEX_op_andc_i32: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - tcg_out_andi32(s, a0, a1, ~a2); - } else { - tcg_out32(s, ANDC | SAB(a1, a0, a2)); - } - break; - case INDEX_op_andc_i64: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - tcg_out_andi64(s, a0, a1, ~a2); - } else { - tcg_out32(s, ANDC | SAB(a1, a0, a2)); - } - break; - case INDEX_op_orc_i32: - if (const_args[2]) { - tcg_out_ori32(s, args[0], args[1], ~args[2]); - break; - } - /* FALLTHRU */ - case INDEX_op_orc_i64: - tcg_out32(s, ORC | SAB(args[1], args[0], args[2])); - break; - case INDEX_op_eqv_i32: - if (const_args[2]) { - tcg_out_xori32(s, args[0], args[1], ~args[2]); - break; - } - /* FALLTHRU */ - case INDEX_op_eqv_i64: - tcg_out32(s, EQV | SAB(args[1], args[0], args[2])); - break; - case INDEX_op_nand_i32: - case INDEX_op_nand_i64: - tcg_out32(s, NAND | SAB(args[1], args[0], args[2])); - break; - case INDEX_op_nor_i32: - case INDEX_op_nor_i64: - tcg_out32(s, NOR | SAB(args[1], args[0], args[2])); - break; +static void tgen_addco_rrr(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, ADDC | TAB(a0, a1, a2)); +} - case INDEX_op_clz_i32: - tcg_out_cntxz(s, TCG_TYPE_I32, CNTLZW, args[0], args[1], - args[2], const_args[2]); - break; - case INDEX_op_ctz_i32: - tcg_out_cntxz(s, TCG_TYPE_I32, CNTTZW, args[0], args[1], - args[2], const_args[2]); - break; - case INDEX_op_ctpop_i32: - tcg_out32(s, CNTPOPW | SAB(args[1], args[0], 0)); - break; +static void tgen_addco_rri(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + tcg_out32(s, ADDIC | TAI(a0, a1, a2)); +} - case INDEX_op_clz_i64: - tcg_out_cntxz(s, TCG_TYPE_I64, CNTLZD, args[0], args[1], - args[2], const_args[2]); - break; - case INDEX_op_ctz_i64: - tcg_out_cntxz(s, TCG_TYPE_I64, CNTTZD, args[0], args[1], - args[2], const_args[2]); - break; - case INDEX_op_ctpop_i64: - tcg_out32(s, CNTPOPD | SAB(args[1], args[0], 0)); - break; +static TCGConstraintSetIndex cset_addco(TCGType type, unsigned flags) +{ + /* + * Note that the CA bit is defined based on the word size of the + * environment. So in 64-bit mode it's always carry-out of bit 63. + * The fallback code using deposit works just as well for TCG_TYPE_I32. + */ + return type == TCG_TYPE_REG ? C_O1_I2(r, r, rI) : C_NotImplemented; +} - case INDEX_op_mul_i32: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - tcg_out32(s, MULLI | TAI(a0, a1, a2)); - } else { - tcg_out32(s, MULLW | TAB(a0, a1, a2)); - } - break; +static const TCGOutOpBinary outop_addco = { + .base.static_constraint = C_Dynamic, + .base.dynamic_constraint = cset_addco, + .out_rrr = tgen_addco_rrr, + .out_rri = tgen_addco_rri, +}; - case INDEX_op_div_i32: - tcg_out32(s, DIVW | TAB(args[0], args[1], args[2])); - break; +static void tgen_addcio_rrr(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, ADDE | TAB(a0, a1, a2)); +} - case INDEX_op_divu_i32: - tcg_out32(s, DIVWU | TAB(args[0], args[1], args[2])); - break; +static void tgen_addcio_rri(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + tcg_out32(s, (a2 ? ADDME : ADDZE) | RT(a0) | RA(a1)); +} - case INDEX_op_rem_i32: - tcg_out32(s, MODSW | TAB(args[0], args[1], args[2])); - break; +static TCGConstraintSetIndex cset_addcio(TCGType type, unsigned flags) +{ + return type == TCG_TYPE_REG ? C_O1_I2(r, r, rZM) : C_NotImplemented; +} - case INDEX_op_remu_i32: - tcg_out32(s, MODUW | TAB(args[0], args[1], args[2])); - break; +static const TCGOutOpBinary outop_addcio = { + .base.static_constraint = C_Dynamic, + .base.dynamic_constraint = cset_addcio, + .out_rrr = tgen_addcio_rrr, + .out_rri = tgen_addcio_rri, +}; - case INDEX_op_shl_i32: - if (const_args[2]) { - /* Limit immediate shift count lest we create an illegal insn. */ - tcg_out_shli32(s, args[0], args[1], args[2] & 31); - } else { - tcg_out32(s, SLW | SAB(args[1], args[0], args[2])); - } - break; - case INDEX_op_shr_i32: - if (const_args[2]) { - /* Limit immediate shift count lest we create an illegal insn. */ - tcg_out_shri32(s, args[0], args[1], args[2] & 31); - } else { - tcg_out32(s, SRW | SAB(args[1], args[0], args[2])); - } - break; - case INDEX_op_sar_i32: - if (const_args[2]) { - tcg_out_sari32(s, args[0], args[1], args[2]); - } else { - tcg_out32(s, SRAW | SAB(args[1], args[0], args[2])); - } - break; - case INDEX_op_rotl_i32: - if (const_args[2]) { - tcg_out_rlw(s, RLWINM, args[0], args[1], args[2], 0, 31); - } else { - tcg_out32(s, RLWNM | SAB(args[1], args[0], args[2]) - | MB(0) | ME(31)); - } - break; - case INDEX_op_rotr_i32: - if (const_args[2]) { - tcg_out_rlw(s, RLWINM, args[0], args[1], 32 - args[2], 0, 31); - } else { - tcg_out32(s, SUBFIC | TAI(TCG_REG_R0, args[2], 32)); - tcg_out32(s, RLWNM | SAB(args[1], args[0], TCG_REG_R0) - | MB(0) | ME(31)); - } - break; +static const TCGOutOpAddSubCarry outop_addci = { + .base.static_constraint = C_Dynamic, + .base.dynamic_constraint = cset_addcio, + .out_rrr = tgen_addcio_rrr, + .out_rri = tgen_addcio_rri, +}; - case INDEX_op_brcond_i32: - tcg_out_brcond(s, args[2], args[0], args[1], const_args[1], - arg_label(args[3]), TCG_TYPE_I32); - break; - case INDEX_op_brcond_i64: - tcg_out_brcond(s, args[2], args[0], args[1], const_args[1], - arg_label(args[3]), TCG_TYPE_I64); - break; - case INDEX_op_brcond2_i32: - tcg_out_brcond2(s, args, const_args); - break; +static void tcg_out_set_carry(TCGContext *s) +{ + tcg_out32(s, SUBFC | TAB(TCG_REG_R0, TCG_REG_R0, TCG_REG_R0)); +} - case INDEX_op_neg_i32: - case INDEX_op_neg_i64: - tcg_out32(s, NEG | RT(args[0]) | RA(args[1])); - break; +static void tgen_and(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, AND | SAB(a1, a0, a2)); +} - case INDEX_op_not_i32: - case INDEX_op_not_i64: - tcg_out32(s, NOR | SAB(args[1], args[0], args[1])); - break; +static void tgen_andi(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + if (type == TCG_TYPE_I32) { + tcg_out_andi32(s, a0, a1, a2); + } else { + tcg_out_andi64(s, a0, a1, a2); + } +} - case INDEX_op_add_i64: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - do_addi_64: - tcg_out_mem_long(s, ADDI, ADD, a0, a1, a2); - } else { - tcg_out32(s, ADD | TAB(a0, a1, a2)); - } - break; - case INDEX_op_sub_i64: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[1]) { - if (const_args[2]) { - tcg_out_movi(s, TCG_TYPE_I64, a0, a1 - a2); - } else { - tcg_out32(s, SUBFIC | TAI(a0, a2, a1)); - } - } else if (const_args[2]) { - a2 = -a2; - goto do_addi_64; - } else { - tcg_out32(s, SUBF | TAB(a0, a2, a1)); - } - break; +static const TCGOutOpBinary outop_and = { + .base.static_constraint = C_O1_I2(r, r, ri), + .out_rrr = tgen_and, + .out_rri = tgen_andi, +}; - case INDEX_op_shl_i64: - if (const_args[2]) { - /* Limit immediate shift count lest we create an illegal insn. */ - tcg_out_shli64(s, args[0], args[1], args[2] & 63); - } else { - tcg_out32(s, SLD | SAB(args[1], args[0], args[2])); - } - break; - case INDEX_op_shr_i64: - if (const_args[2]) { - /* Limit immediate shift count lest we create an illegal insn. */ - tcg_out_shri64(s, args[0], args[1], args[2] & 63); - } else { - tcg_out32(s, SRD | SAB(args[1], args[0], args[2])); - } - break; - case INDEX_op_sar_i64: - if (const_args[2]) { - tcg_out_sari64(s, args[0], args[1], args[2]); - } else { - tcg_out32(s, SRAD | SAB(args[1], args[0], args[2])); - } - break; - case INDEX_op_rotl_i64: - if (const_args[2]) { - tcg_out_rld(s, RLDICL, args[0], args[1], args[2], 0); - } else { - tcg_out32(s, RLDCL | SAB(args[1], args[0], args[2]) | MB64(0)); - } - break; - case INDEX_op_rotr_i64: - if (const_args[2]) { - tcg_out_rld(s, RLDICL, args[0], args[1], 64 - args[2], 0); - } else { - tcg_out32(s, SUBFIC | TAI(TCG_REG_R0, args[2], 64)); - tcg_out32(s, RLDCL | SAB(args[1], args[0], TCG_REG_R0) | MB64(0)); - } - break; +static void tgen_andc(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, ANDC | SAB(a1, a0, a2)); +} - case INDEX_op_mul_i64: - a0 = args[0], a1 = args[1], a2 = args[2]; - if (const_args[2]) { - tcg_out32(s, MULLI | TAI(a0, a1, a2)); - } else { - tcg_out32(s, MULLD | TAB(a0, a1, a2)); - } - break; - case INDEX_op_div_i64: - tcg_out32(s, DIVD | TAB(args[0], args[1], args[2])); - break; - case INDEX_op_divu_i64: - tcg_out32(s, DIVDU | TAB(args[0], args[1], args[2])); - break; - case INDEX_op_rem_i64: - tcg_out32(s, MODSD | TAB(args[0], args[1], args[2])); - break; - case INDEX_op_remu_i64: - tcg_out32(s, MODUD | TAB(args[0], args[1], args[2])); - break; +static const TCGOutOpBinary outop_andc = { + .base.static_constraint = C_O1_I2(r, r, r), + .out_rrr = tgen_andc, +}; - case INDEX_op_qemu_ld_i32: - tcg_out_qemu_ld(s, args[0], -1, args[1], args[2], TCG_TYPE_I32); - break; - case INDEX_op_qemu_ld_i64: - if (TCG_TARGET_REG_BITS == 64) { - tcg_out_qemu_ld(s, args[0], -1, args[1], args[2], TCG_TYPE_I64); - } else { - tcg_out_qemu_ld(s, args[0], args[1], args[2], - args[3], TCG_TYPE_I64); - } - break; - case INDEX_op_qemu_ld_i128: - tcg_debug_assert(TCG_TARGET_REG_BITS == 64); - tcg_out_qemu_ldst_i128(s, args[0], args[1], args[2], args[3], true); - break; +static void tgen_clz(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? CNTLZW : CNTLZD; + tcg_out_cntxz(s, type, insn, a0, a1, a2, false); +} - case INDEX_op_qemu_st_i32: - tcg_out_qemu_st(s, args[0], -1, args[1], args[2], TCG_TYPE_I32); - break; - case INDEX_op_qemu_st_i64: - if (TCG_TARGET_REG_BITS == 64) { - tcg_out_qemu_st(s, args[0], -1, args[1], args[2], TCG_TYPE_I64); - } else { - tcg_out_qemu_st(s, args[0], args[1], args[2], - args[3], TCG_TYPE_I64); - } - break; - case INDEX_op_qemu_st_i128: - tcg_debug_assert(TCG_TARGET_REG_BITS == 64); - tcg_out_qemu_ldst_i128(s, args[0], args[1], args[2], args[3], false); - break; +static void tgen_clzi(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? CNTLZW : CNTLZD; + tcg_out_cntxz(s, type, insn, a0, a1, a2, true); +} - case INDEX_op_setcond_i32: - tcg_out_setcond(s, TCG_TYPE_I32, args[3], args[0], args[1], args[2], - const_args[2], false); - break; - case INDEX_op_setcond_i64: - tcg_out_setcond(s, TCG_TYPE_I64, args[3], args[0], args[1], args[2], - const_args[2], false); - break; - case INDEX_op_negsetcond_i32: - tcg_out_setcond(s, TCG_TYPE_I32, args[3], args[0], args[1], args[2], - const_args[2], true); - break; - case INDEX_op_negsetcond_i64: - tcg_out_setcond(s, TCG_TYPE_I64, args[3], args[0], args[1], args[2], - const_args[2], true); - break; - case INDEX_op_setcond2_i32: - tcg_out_setcond2(s, args, const_args); - break; +static const TCGOutOpBinary outop_clz = { + .base.static_constraint = C_O1_I2(r, r, rZW), + .out_rrr = tgen_clz, + .out_rri = tgen_clzi, +}; - case INDEX_op_bswap16_i32: - case INDEX_op_bswap16_i64: - tcg_out_bswap16(s, args[0], args[1], args[2]); - break; - case INDEX_op_bswap32_i32: - tcg_out_bswap32(s, args[0], args[1], 0); - break; - case INDEX_op_bswap32_i64: - tcg_out_bswap32(s, args[0], args[1], args[2]); - break; - case INDEX_op_bswap64_i64: - tcg_out_bswap64(s, args[0], args[1]); - break; +static void tgen_ctpop(TCGContext *s, TCGType type, TCGReg a0, TCGReg a1) +{ + uint32_t insn = type == TCG_TYPE_I32 ? CNTPOPW : CNTPOPD; + tcg_out32(s, insn | SAB(a1, a0, 0)); +} - case INDEX_op_deposit_i32: - if (const_args[2]) { - uint32_t mask = ((2u << (args[4] - 1)) - 1) << args[3]; - tcg_out_andi32(s, args[0], args[0], ~mask); - } else { - tcg_out_rlw(s, RLWIMI, args[0], args[2], args[3], - 32 - args[3] - args[4], 31 - args[3]); - } - break; - case INDEX_op_deposit_i64: - if (const_args[2]) { - uint64_t mask = ((2ull << (args[4] - 1)) - 1) << args[3]; - tcg_out_andi64(s, args[0], args[0], ~mask); - } else { - tcg_out_rld(s, RLDIMI, args[0], args[2], args[3], - 64 - args[3] - args[4]); - } - break; +static TCGConstraintSetIndex cset_ctpop(TCGType type, unsigned flags) +{ + return have_isa_2_06 ? C_O1_I1(r, r) : C_NotImplemented; +} - case INDEX_op_extract_i32: - if (args[2] == 0 && args[3] <= 16) { - tcg_out32(s, ANDI | SAI(args[1], args[0], (1 << args[3]) - 1)); - break; - } - tcg_out_rlw(s, RLWINM, args[0], args[1], - 32 - args[2], 32 - args[3], 31); - break; - case INDEX_op_extract_i64: - if (args[2] == 0 && args[3] <= 16) { - tcg_out32(s, ANDI | SAI(args[1], args[0], (1 << args[3]) - 1)); - break; - } - tcg_out_rld(s, RLDICL, args[0], args[1], 64 - args[2], 64 - args[3]); - break; +static const TCGOutOpUnary outop_ctpop = { + .base.static_constraint = C_Dynamic, + .base.dynamic_constraint = cset_ctpop, + .out_rr = tgen_ctpop, +}; - case INDEX_op_sextract_i64: - if (args[2] + args[3] == 32) { - if (args[2] == 0) { - tcg_out_ext32s(s, args[0], args[1]); - } else { - tcg_out_sari32(s, args[0], args[1], args[2]); - } - break; - } - /* FALLTHRU */ - case INDEX_op_sextract_i32: - if (args[2] == 0 && args[3] == 8) { - tcg_out_ext8s(s, TCG_TYPE_I32, args[0], args[1]); - } else if (args[2] == 0 && args[3] == 16) { - tcg_out_ext16s(s, TCG_TYPE_I32, args[0], args[1]); - } else { - g_assert_not_reached(); - } - break; +static void tgen_ctz(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? CNTTZW : CNTTZD; + tcg_out_cntxz(s, type, insn, a0, a1, a2, false); +} - case INDEX_op_movcond_i32: - tcg_out_movcond(s, TCG_TYPE_I32, args[5], args[0], args[1], args[2], - args[3], args[4], const_args[2]); - break; - case INDEX_op_movcond_i64: - tcg_out_movcond(s, TCG_TYPE_I64, args[5], args[0], args[1], args[2], - args[3], args[4], const_args[2]); - break; +static void tgen_ctzi(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? CNTTZW : CNTTZD; + tcg_out_cntxz(s, type, insn, a0, a1, a2, true); +} + +static TCGConstraintSetIndex cset_ctz(TCGType type, unsigned flags) +{ + return have_isa_3_00 ? C_O1_I2(r, r, rZW) : C_NotImplemented; +} + +static const TCGOutOpBinary outop_ctz = { + .base.static_constraint = C_Dynamic, + .base.dynamic_constraint = cset_ctz, + .out_rrr = tgen_ctz, + .out_rri = tgen_ctzi, +}; + +static void tgen_eqv(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, EQV | SAB(a1, a0, a2)); +} #if TCG_TARGET_REG_BITS == 64 - case INDEX_op_add2_i64: -#else - case INDEX_op_add2_i32: +static void tgen_extrh_i64_i32(TCGContext *s, TCGType t, TCGReg a0, TCGReg a1) +{ + tcg_out_shri64(s, a0, a1, 32); +} + +static const TCGOutOpUnary outop_extrh_i64_i32 = { + .base.static_constraint = C_O1_I1(r, r), + .out_rr = tgen_extrh_i64_i32, +}; #endif - /* Note that the CA bit is defined based on the word size of the - environment. So in 64-bit mode it's always carry-out of bit 63. - The fallback code using deposit works just as well for 32-bit. */ - a0 = args[0], a1 = args[1]; - if (a0 == args[3] || (!const_args[5] && a0 == args[5])) { - a0 = TCG_REG_R0; - } - if (const_args[4]) { - tcg_out32(s, ADDIC | TAI(a0, args[2], args[4])); - } else { - tcg_out32(s, ADDC | TAB(a0, args[2], args[4])); + +static void tgen_divs(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? DIVW : DIVD; + tcg_out32(s, insn | TAB(a0, a1, a2)); +} + +static const TCGOutOpBinary outop_divs = { + .base.static_constraint = C_O1_I2(r, r, r), + .out_rrr = tgen_divs, +}; + +static const TCGOutOpDivRem outop_divs2 = { + .base.static_constraint = C_NotImplemented, +}; + +static void tgen_divu(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? DIVWU : DIVDU; + tcg_out32(s, insn | TAB(a0, a1, a2)); +} + +static const TCGOutOpBinary outop_divu = { + .base.static_constraint = C_O1_I2(r, r, r), + .out_rrr = tgen_divu, +}; + +static const TCGOutOpDivRem outop_divu2 = { + .base.static_constraint = C_NotImplemented, +}; + +static const TCGOutOpBinary outop_eqv = { + .base.static_constraint = C_O1_I2(r, r, r), + .out_rrr = tgen_eqv, +}; + +static void tgen_mul(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? MULLW : MULLD; + tcg_out32(s, insn | TAB(a0, a1, a2)); +} + +static void tgen_muli(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + tcg_out32(s, MULLI | TAI(a0, a1, a2)); +} + +static const TCGOutOpBinary outop_mul = { + .base.static_constraint = C_O1_I2(r, r, rI), + .out_rrr = tgen_mul, + .out_rri = tgen_muli, +}; + +static const TCGOutOpMul2 outop_muls2 = { + .base.static_constraint = C_NotImplemented, +}; + +static void tgen_mulsh(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? MULHW : MULHD; + tcg_out32(s, insn | TAB(a0, a1, a2)); +} + +static const TCGOutOpBinary outop_mulsh = { + .base.static_constraint = C_O1_I2(r, r, r), + .out_rrr = tgen_mulsh, +}; + +static const TCGOutOpMul2 outop_mulu2 = { + .base.static_constraint = C_NotImplemented, +}; + +static void tgen_muluh(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? MULHWU : MULHDU; + tcg_out32(s, insn | TAB(a0, a1, a2)); +} + +static const TCGOutOpBinary outop_muluh = { + .base.static_constraint = C_O1_I2(r, r, r), + .out_rrr = tgen_muluh, +}; + +static void tgen_nand(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, NAND | SAB(a1, a0, a2)); +} + +static const TCGOutOpBinary outop_nand = { + .base.static_constraint = C_O1_I2(r, r, r), + .out_rrr = tgen_nand, +}; + +static void tgen_nor(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, NOR | SAB(a1, a0, a2)); +} + +static const TCGOutOpBinary outop_nor = { + .base.static_constraint = C_O1_I2(r, r, r), + .out_rrr = tgen_nor, +}; + +static void tgen_or(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, OR | SAB(a1, a0, a2)); +} + +static void tgen_ori(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + tcg_out_ori32(s, a0, a1, a2); +} + +static const TCGOutOpBinary outop_or = { + .base.static_constraint = C_O1_I2(r, r, rU), + .out_rrr = tgen_or, + .out_rri = tgen_ori, +}; + +static void tgen_orc(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, ORC | SAB(a1, a0, a2)); +} + +static const TCGOutOpBinary outop_orc = { + .base.static_constraint = C_O1_I2(r, r, r), + .out_rrr = tgen_orc, +}; + +static TCGConstraintSetIndex cset_mod(TCGType type, unsigned flags) +{ + return have_isa_3_00 ? C_O1_I2(r, r, r) : C_NotImplemented; +} + +static void tgen_rems(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? MODSW : MODSD; + tcg_out32(s, insn | TAB(a0, a1, a2)); +} + +static const TCGOutOpBinary outop_rems = { + .base.static_constraint = C_Dynamic, + .base.dynamic_constraint = cset_mod, + .out_rrr = tgen_rems, +}; + +static void tgen_remu(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? MODUW : MODUD; + tcg_out32(s, insn | TAB(a0, a1, a2)); +} + +static const TCGOutOpBinary outop_remu = { + .base.static_constraint = C_Dynamic, + .base.dynamic_constraint = cset_mod, + .out_rrr = tgen_remu, +}; + +static void tgen_rotl(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + if (type == TCG_TYPE_I32) { + tcg_out32(s, RLWNM | SAB(a1, a0, a2) | MB(0) | ME(31)); + } else { + tcg_out32(s, RLDCL | SAB(a1, a0, a2) | MB64(0)); + } +} + +static void tgen_rotli(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + if (type == TCG_TYPE_I32) { + tcg_out_rlw(s, RLWINM, a0, a1, a2, 0, 31); + } else { + tcg_out_rld(s, RLDICL, a0, a1, a2, 0); + } +} + +static const TCGOutOpBinary outop_rotl = { + .base.static_constraint = C_O1_I2(r, r, ri), + .out_rrr = tgen_rotl, + .out_rri = tgen_rotli, +}; + +static const TCGOutOpBinary outop_rotr = { + .base.static_constraint = C_NotImplemented, +}; + +static void tgen_sar(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? SRAW : SRAD; + tcg_out32(s, insn | SAB(a1, a0, a2)); +} + +static void tgen_sari(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + /* Limit immediate shift count lest we create an illegal insn. */ + if (type == TCG_TYPE_I32) { + tcg_out_sari32(s, a0, a1, a2 & 31); + } else { + tcg_out_sari64(s, a0, a1, a2 & 63); + } +} + +static const TCGOutOpBinary outop_sar = { + .base.static_constraint = C_O1_I2(r, r, ri), + .out_rrr = tgen_sar, + .out_rri = tgen_sari, +}; + +static void tgen_shl(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? SLW : SLD; + tcg_out32(s, insn | SAB(a1, a0, a2)); +} + +static void tgen_shli(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + /* Limit immediate shift count lest we create an illegal insn. */ + if (type == TCG_TYPE_I32) { + tcg_out_shli32(s, a0, a1, a2 & 31); + } else { + tcg_out_shli64(s, a0, a1, a2 & 63); + } +} + +static const TCGOutOpBinary outop_shl = { + .base.static_constraint = C_O1_I2(r, r, ri), + .out_rrr = tgen_shl, + .out_rri = tgen_shli, +}; + +static void tgen_shr(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + uint32_t insn = type == TCG_TYPE_I32 ? SRW : SRD; + tcg_out32(s, insn | SAB(a1, a0, a2)); +} + +static void tgen_shri(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + /* Limit immediate shift count lest we create an illegal insn. */ + if (type == TCG_TYPE_I32) { + tcg_out_shri32(s, a0, a1, a2 & 31); + } else { + tcg_out_shri64(s, a0, a1, a2 & 63); + } +} + +static const TCGOutOpBinary outop_shr = { + .base.static_constraint = C_O1_I2(r, r, ri), + .out_rrr = tgen_shr, + .out_rri = tgen_shri, +}; + +static void tgen_sub(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, SUBF | TAB(a0, a2, a1)); +} + +static void tgen_subfi(TCGContext *s, TCGType type, + TCGReg a0, tcg_target_long a1, TCGReg a2) +{ + tcg_out32(s, SUBFIC | TAI(a0, a2, a1)); +} + +static const TCGOutOpSubtract outop_sub = { + .base.static_constraint = C_O1_I2(r, rI, r), + .out_rrr = tgen_sub, + .out_rir = tgen_subfi, +}; + +static void tgen_subbo_rrr(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, SUBFC | TAB(a0, a2, a1)); +} + +static void tgen_subbo_rri(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + if (a2 == 0) { + tcg_out_movi(s, type, TCG_REG_R0, 0); + tgen_subbo_rrr(s, type, a0, a1, TCG_REG_R0); + } else { + tgen_addco_rri(s, type, a0, a1, -a2); + } +} + +/* The underlying insn for subfi is subfic. */ +#define tgen_subbo_rir tgen_subfi + +static void tgen_subbo_rii(TCGContext *s, TCGType type, + TCGReg a0, tcg_target_long a1, tcg_target_long a2) +{ + tcg_out_movi(s, type, TCG_REG_R0, a2); + tgen_subbo_rir(s, type, a0, a1, TCG_REG_R0); +} + +static TCGConstraintSetIndex cset_subbo(TCGType type, unsigned flags) +{ + /* Recall that the CA bit is defined based on the host word size. */ + return type == TCG_TYPE_REG ? C_O1_I2(r, rI, rN) : C_NotImplemented; +} + +static const TCGOutOpAddSubCarry outop_subbo = { + .base.static_constraint = C_Dynamic, + .base.dynamic_constraint = cset_subbo, + .out_rrr = tgen_subbo_rrr, + .out_rri = tgen_subbo_rri, + .out_rir = tgen_subbo_rir, + .out_rii = tgen_subbo_rii, +}; + +static void tgen_subbio_rrr(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, SUBFE | TAB(a0, a2, a1)); +} + +static void tgen_subbio_rri(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + tgen_addcio_rri(s, type, a0, a1, ~a2); +} + +static void tgen_subbio_rir(TCGContext *s, TCGType type, + TCGReg a0, tcg_target_long a1, TCGReg a2) +{ + tcg_debug_assert(a1 == 0 || a1 == -1); + tcg_out32(s, (a1 ? SUBFME : SUBFZE) | RT(a0) | RA(a2)); +} + +static void tgen_subbio_rii(TCGContext *s, TCGType type, + TCGReg a0, tcg_target_long a1, tcg_target_long a2) +{ + tcg_out_movi(s, type, TCG_REG_R0, a2); + tgen_subbio_rir(s, type, a0, a1, TCG_REG_R0); +} + +static TCGConstraintSetIndex cset_subbio(TCGType type, unsigned flags) +{ + return type == TCG_TYPE_REG ? C_O1_I2(r, rZM, rZM) : C_NotImplemented; +} + +static const TCGOutOpAddSubCarry outop_subbio = { + .base.static_constraint = C_Dynamic, + .base.dynamic_constraint = cset_subbio, + .out_rrr = tgen_subbio_rrr, + .out_rri = tgen_subbio_rri, + .out_rir = tgen_subbio_rir, + .out_rii = tgen_subbio_rii, +}; + +#define outop_subbi outop_subbio + +static void tcg_out_set_borrow(TCGContext *s) +{ + /* borrow = !carry */ + tcg_out32(s, ADDIC | TAI(TCG_REG_R0, TCG_REG_R0, 0)); +} + +static void tgen_xor(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, TCGReg a2) +{ + tcg_out32(s, XOR | SAB(a1, a0, a2)); +} + +static void tgen_xori(TCGContext *s, TCGType type, + TCGReg a0, TCGReg a1, tcg_target_long a2) +{ + tcg_out_xori32(s, a0, a1, a2); +} + +static const TCGOutOpBinary outop_xor = { + .base.static_constraint = C_O1_I2(r, r, rU), + .out_rrr = tgen_xor, + .out_rri = tgen_xori, +}; + +static void tgen_bswap16(TCGContext *s, TCGType type, + TCGReg dst, TCGReg src, unsigned flags) +{ + TCGReg tmp = dst == src ? TCG_REG_R0 : dst; + + if (have_isa_3_10) { + tcg_out32(s, BRH | RA(dst) | RS(src)); + if (flags & TCG_BSWAP_OS) { + tcg_out_ext16s(s, TCG_TYPE_REG, dst, dst); + } else if ((flags & (TCG_BSWAP_IZ | TCG_BSWAP_OZ)) == TCG_BSWAP_OZ) { + tcg_out_ext16u(s, dst, dst); } - if (const_args[5]) { - tcg_out32(s, (args[5] ? ADDME : ADDZE) | RT(a1) | RA(args[3])); - } else { - tcg_out32(s, ADDE | TAB(a1, args[3], args[5])); + return; + } + + /* + * In the following, + * dep(a, b, m) -> (a & ~m) | (b & m) + * + * Begin with: src = xxxxabcd + */ + /* tmp = rol32(src, 24) & 0x000000ff = 0000000c */ + tcg_out_rlw(s, RLWINM, tmp, src, 24, 24, 31); + /* tmp = dep(tmp, rol32(src, 8), 0x0000ff00) = 000000dc */ + tcg_out_rlw(s, RLWIMI, tmp, src, 8, 16, 23); + + if (flags & TCG_BSWAP_OS) { + tcg_out_ext16s(s, TCG_TYPE_REG, dst, tmp); + } else { + tcg_out_mov(s, TCG_TYPE_REG, dst, tmp); + } +} + +static const TCGOutOpBswap outop_bswap16 = { + .base.static_constraint = C_O1_I1(r, r), + .out_rr = tgen_bswap16, +}; + +static void tgen_bswap32(TCGContext *s, TCGType type, + TCGReg dst, TCGReg src, unsigned flags) +{ + TCGReg tmp = dst == src ? TCG_REG_R0 : dst; + + if (have_isa_3_10) { + tcg_out32(s, BRW | RA(dst) | RS(src)); + if (flags & TCG_BSWAP_OS) { + tcg_out_ext32s(s, dst, dst); + } else if ((flags & (TCG_BSWAP_IZ | TCG_BSWAP_OZ)) == TCG_BSWAP_OZ) { + tcg_out_ext32u(s, dst, dst); } - if (a0 != args[0]) { - tcg_out_mov(s, TCG_TYPE_REG, args[0], a0); + return; + } + + /* + * Stolen from gcc's builtin_bswap32. + * In the following, + * dep(a, b, m) -> (a & ~m) | (b & m) + * + * Begin with: src = xxxxabcd + */ + /* tmp = rol32(src, 8) & 0xffffffff = 0000bcda */ + tcg_out_rlw(s, RLWINM, tmp, src, 8, 0, 31); + /* tmp = dep(tmp, rol32(src, 24), 0xff000000) = 0000dcda */ + tcg_out_rlw(s, RLWIMI, tmp, src, 24, 0, 7); + /* tmp = dep(tmp, rol32(src, 24), 0x0000ff00) = 0000dcba */ + tcg_out_rlw(s, RLWIMI, tmp, src, 24, 16, 23); + + if (flags & TCG_BSWAP_OS) { + tcg_out_ext32s(s, dst, tmp); + } else { + tcg_out_mov(s, TCG_TYPE_REG, dst, tmp); + } +} + +static const TCGOutOpBswap outop_bswap32 = { + .base.static_constraint = C_O1_I1(r, r), + .out_rr = tgen_bswap32, +}; + +#if TCG_TARGET_REG_BITS == 64 +static void tgen_bswap64(TCGContext *s, TCGType type, TCGReg dst, TCGReg src) +{ + TCGReg t0 = dst == src ? TCG_REG_R0 : dst; + TCGReg t1 = dst == src ? dst : TCG_REG_R0; + + if (have_isa_3_10) { + tcg_out32(s, BRD | RA(dst) | RS(src)); + return; + } + + /* + * In the following, + * dep(a, b, m) -> (a & ~m) | (b & m) + * + * Begin with: src = abcdefgh + */ + /* t0 = rol32(src, 8) & 0xffffffff = 0000fghe */ + tcg_out_rlw(s, RLWINM, t0, src, 8, 0, 31); + /* t0 = dep(t0, rol32(src, 24), 0xff000000) = 0000hghe */ + tcg_out_rlw(s, RLWIMI, t0, src, 24, 0, 7); + /* t0 = dep(t0, rol32(src, 24), 0x0000ff00) = 0000hgfe */ + tcg_out_rlw(s, RLWIMI, t0, src, 24, 16, 23); + + /* t0 = rol64(t0, 32) = hgfe0000 */ + tcg_out_rld(s, RLDICL, t0, t0, 32, 0); + /* t1 = rol64(src, 32) = efghabcd */ + tcg_out_rld(s, RLDICL, t1, src, 32, 0); + + /* t0 = dep(t0, rol32(t1, 24), 0xffffffff) = hgfebcda */ + tcg_out_rlw(s, RLWIMI, t0, t1, 8, 0, 31); + /* t0 = dep(t0, rol32(t1, 24), 0xff000000) = hgfedcda */ + tcg_out_rlw(s, RLWIMI, t0, t1, 24, 0, 7); + /* t0 = dep(t0, rol32(t1, 24), 0x0000ff00) = hgfedcba */ + tcg_out_rlw(s, RLWIMI, t0, t1, 24, 16, 23); + + tcg_out_mov(s, TCG_TYPE_REG, dst, t0); +} + +static const TCGOutOpUnary outop_bswap64 = { + .base.static_constraint = C_O1_I1(r, r), + .out_rr = tgen_bswap64, +}; +#endif /* TCG_TARGET_REG_BITS == 64 */ + +static void tgen_neg(TCGContext *s, TCGType type, TCGReg a0, TCGReg a1) +{ + tcg_out32(s, NEG | RT(a0) | RA(a1)); +} + +static const TCGOutOpUnary outop_neg = { + .base.static_constraint = C_O1_I1(r, r), + .out_rr = tgen_neg, +}; + +static void tgen_not(TCGContext *s, TCGType type, TCGReg a0, TCGReg a1) +{ + tgen_nor(s, type, a0, a1, a1); +} + +static const TCGOutOpUnary outop_not = { + .base.static_constraint = C_O1_I1(r, r), + .out_rr = tgen_not, +}; + +static void tgen_deposit(TCGContext *s, TCGType type, TCGReg a0, TCGReg a1, + TCGReg a2, unsigned ofs, unsigned len) +{ + if (type == TCG_TYPE_I32) { + tcg_out_rlw(s, RLWIMI, a0, a2, ofs, 32 - ofs - len, 31 - ofs); + } else { + tcg_out_rld(s, RLDIMI, a0, a2, ofs, 64 - ofs - len); + } +} + +static void tgen_depositi(TCGContext *s, TCGType type, TCGReg a0, TCGReg a1, + tcg_target_long a2, unsigned ofs, unsigned len) +{ + tgen_andi(s, type, a0, a1, ~MAKE_64BIT_MASK(ofs, len)); +} + +static const TCGOutOpDeposit outop_deposit = { + .base.static_constraint = C_O1_I2(r, 0, rZ), + .out_rrr = tgen_deposit, + .out_rri = tgen_depositi, +}; + +static void tgen_extract(TCGContext *s, TCGType type, TCGReg a0, TCGReg a1, + unsigned ofs, unsigned len) +{ + if (ofs == 0 && len <= 16) { + tgen_andi(s, TCG_TYPE_I32, a0, a1, (1 << len) - 1); + } else if (type == TCG_TYPE_I32) { + tcg_out_rlw(s, RLWINM, a0, a1, 32 - ofs, 32 - len, 31); + } else { + tcg_out_rld(s, RLDICL, a0, a1, 64 - ofs, 64 - len); + } +} + +static const TCGOutOpExtract outop_extract = { + .base.static_constraint = C_O1_I1(r, r), + .out_rr = tgen_extract, +}; + +static void tgen_sextract(TCGContext *s, TCGType type, TCGReg a0, TCGReg a1, + unsigned ofs, unsigned len) +{ + if (ofs == 0) { + switch (len) { + case 8: + tcg_out_ext8s(s, type, a0, a1); + return; + case 16: + tcg_out_ext16s(s, type, a0, a1); + return; + case 32: + tcg_out_ext32s(s, a0, a1); + return; } - break; + } else if (ofs + len == 32) { + tcg_out_sari32(s, a0, a1, ofs); + return; + } + g_assert_not_reached(); +} + +static const TCGOutOpExtract outop_sextract = { + .base.static_constraint = C_O1_I1(r, r), + .out_rr = tgen_sextract, +}; + +static const TCGOutOpExtract2 outop_extract2 = { + .base.static_constraint = C_NotImplemented, +}; + +static void tgen_ld8u(TCGContext *s, TCGType type, TCGReg dest, + TCGReg base, ptrdiff_t offset) +{ + tcg_out_mem_long(s, LBZ, LBZX, dest, base, offset); +} + +static const TCGOutOpLoad outop_ld8u = { + .base.static_constraint = C_O1_I1(r, r), + .out = tgen_ld8u, +}; + +static void tgen_ld8s(TCGContext *s, TCGType type, TCGReg dest, + TCGReg base, ptrdiff_t offset) +{ + tgen_ld8u(s, type, dest, base, offset); + tcg_out_ext8s(s, type, dest, dest); +} + +static const TCGOutOpLoad outop_ld8s = { + .base.static_constraint = C_O1_I1(r, r), + .out = tgen_ld8s, +}; + +static void tgen_ld16u(TCGContext *s, TCGType type, TCGReg dest, + TCGReg base, ptrdiff_t offset) +{ + tcg_out_mem_long(s, LHZ, LHZX, dest, base, offset); +} + +static const TCGOutOpLoad outop_ld16u = { + .base.static_constraint = C_O1_I1(r, r), + .out = tgen_ld16u, +}; + +static void tgen_ld16s(TCGContext *s, TCGType type, TCGReg dest, + TCGReg base, ptrdiff_t offset) +{ + tcg_out_mem_long(s, LHA, LHAX, dest, base, offset); +} + +static const TCGOutOpLoad outop_ld16s = { + .base.static_constraint = C_O1_I1(r, r), + .out = tgen_ld16s, +}; #if TCG_TARGET_REG_BITS == 64 - case INDEX_op_sub2_i64: -#else - case INDEX_op_sub2_i32: +static void tgen_ld32u(TCGContext *s, TCGType type, TCGReg dest, + TCGReg base, ptrdiff_t offset) +{ + tcg_out_mem_long(s, LWZ, LWZX, dest, base, offset); +} + +static const TCGOutOpLoad outop_ld32u = { + .base.static_constraint = C_O1_I1(r, r), + .out = tgen_ld32u, +}; + +static void tgen_ld32s(TCGContext *s, TCGType type, TCGReg dest, + TCGReg base, ptrdiff_t offset) +{ + tcg_out_mem_long(s, LWA, LWAX, dest, base, offset); +} + +static const TCGOutOpLoad outop_ld32s = { + .base.static_constraint = C_O1_I1(r, r), + .out = tgen_ld32s, +}; #endif - a0 = args[0], a1 = args[1]; - if (a0 == args[5] || (!const_args[3] && a0 == args[3])) { - a0 = TCG_REG_R0; - } - if (const_args[2]) { - tcg_out32(s, SUBFIC | TAI(a0, args[4], args[2])); - } else { - tcg_out32(s, SUBFC | TAB(a0, args[4], args[2])); - } - if (const_args[3]) { - tcg_out32(s, (args[3] ? SUBFME : SUBFZE) | RT(a1) | RA(args[5])); - } else { - tcg_out32(s, SUBFE | TAB(a1, args[5], args[3])); - } - if (a0 != args[0]) { - tcg_out_mov(s, TCG_TYPE_REG, args[0], a0); - } - break; - case INDEX_op_muluh_i32: - tcg_out32(s, MULHWU | TAB(args[0], args[1], args[2])); - break; - case INDEX_op_mulsh_i32: - tcg_out32(s, MULHW | TAB(args[0], args[1], args[2])); - break; - case INDEX_op_muluh_i64: - tcg_out32(s, MULHDU | TAB(args[0], args[1], args[2])); - break; - case INDEX_op_mulsh_i64: - tcg_out32(s, MULHD | TAB(args[0], args[1], args[2])); - break; +static void tgen_st8(TCGContext *s, TCGType type, TCGReg data, + TCGReg base, ptrdiff_t offset) +{ + tcg_out_mem_long(s, STB, STBX, data, base, offset); +} - case INDEX_op_mb: - tcg_out_mb(s, args[0]); - break; +static const TCGOutOpStore outop_st8 = { + .base.static_constraint = C_O0_I2(r, r), + .out_r = tgen_st8, +}; - case INDEX_op_mov_i32: /* Always emitted via tcg_out_mov. */ - case INDEX_op_mov_i64: - case INDEX_op_call: /* Always emitted via tcg_out_call. */ - case INDEX_op_exit_tb: /* Always emitted via tcg_out_exit_tb. */ - case INDEX_op_goto_tb: /* Always emitted via tcg_out_goto_tb. */ - case INDEX_op_ext8s_i32: /* Always emitted via tcg_reg_alloc_op. */ - case INDEX_op_ext8s_i64: - case INDEX_op_ext8u_i32: - case INDEX_op_ext8u_i64: - case INDEX_op_ext16s_i32: - case INDEX_op_ext16s_i64: - case INDEX_op_ext16u_i32: - case INDEX_op_ext16u_i64: - case INDEX_op_ext32s_i64: - case INDEX_op_ext32u_i64: - case INDEX_op_ext_i32_i64: - case INDEX_op_extu_i32_i64: - case INDEX_op_extrl_i64_i32: - default: - g_assert_not_reached(); - } +static void tgen_st16(TCGContext *s, TCGType type, TCGReg data, + TCGReg base, ptrdiff_t offset) +{ + tcg_out_mem_long(s, STH, STHX, data, base, offset); } +static const TCGOutOpStore outop_st16 = { + .base.static_constraint = C_O0_I2(r, r), + .out_r = tgen_st16, +}; + +static const TCGOutOpStore outop_st = { + .base.static_constraint = C_O0_I2(r, r), + .out_r = tcg_out_st, +}; + + int tcg_can_emit_vec_op(TCGOpcode opc, TCGType type, unsigned vece) { switch (opc) { @@ -4098,154 +4435,11 @@ static TCGConstraintSetIndex tcg_target_op_def(TCGOpcode op, TCGType type, unsigned flags) { switch (op) { - case INDEX_op_goto_ptr: - return C_O0_I1(r); - - case INDEX_op_ld8u_i32: - case INDEX_op_ld8s_i32: - case INDEX_op_ld16u_i32: - case INDEX_op_ld16s_i32: - case INDEX_op_ld_i32: - case INDEX_op_ctpop_i32: - case INDEX_op_neg_i32: - case INDEX_op_not_i32: - case INDEX_op_ext8s_i32: - case INDEX_op_ext16s_i32: - case INDEX_op_bswap16_i32: - case INDEX_op_bswap32_i32: - case INDEX_op_extract_i32: - case INDEX_op_sextract_i32: - case INDEX_op_ld8u_i64: - case INDEX_op_ld8s_i64: - case INDEX_op_ld16u_i64: - case INDEX_op_ld16s_i64: - case INDEX_op_ld32u_i64: - case INDEX_op_ld32s_i64: - case INDEX_op_ld_i64: - case INDEX_op_ctpop_i64: - case INDEX_op_neg_i64: - case INDEX_op_not_i64: - case INDEX_op_ext8s_i64: - case INDEX_op_ext16s_i64: - case INDEX_op_ext32s_i64: - case INDEX_op_ext_i32_i64: - case INDEX_op_extu_i32_i64: - case INDEX_op_bswap16_i64: - case INDEX_op_bswap32_i64: - case INDEX_op_bswap64_i64: - case INDEX_op_extract_i64: - case INDEX_op_sextract_i64: - return C_O1_I1(r, r); - - case INDEX_op_st8_i32: - case INDEX_op_st16_i32: - case INDEX_op_st_i32: - case INDEX_op_st8_i64: - case INDEX_op_st16_i64: - case INDEX_op_st32_i64: - case INDEX_op_st_i64: + case INDEX_op_qemu_st: return C_O0_I2(r, r); - - case INDEX_op_add_i32: - case INDEX_op_and_i32: - case INDEX_op_or_i32: - case INDEX_op_xor_i32: - case INDEX_op_andc_i32: - case INDEX_op_orc_i32: - case INDEX_op_eqv_i32: - case INDEX_op_shl_i32: - case INDEX_op_shr_i32: - case INDEX_op_sar_i32: - case INDEX_op_rotl_i32: - case INDEX_op_rotr_i32: - case INDEX_op_and_i64: - case INDEX_op_andc_i64: - case INDEX_op_shl_i64: - case INDEX_op_shr_i64: - case INDEX_op_sar_i64: - case INDEX_op_rotl_i64: - case INDEX_op_rotr_i64: - return C_O1_I2(r, r, ri); - - case INDEX_op_mul_i32: - case INDEX_op_mul_i64: - return C_O1_I2(r, r, rI); - - case INDEX_op_div_i32: - case INDEX_op_divu_i32: - case INDEX_op_rem_i32: - case INDEX_op_remu_i32: - case INDEX_op_nand_i32: - case INDEX_op_nor_i32: - case INDEX_op_muluh_i32: - case INDEX_op_mulsh_i32: - case INDEX_op_orc_i64: - case INDEX_op_eqv_i64: - case INDEX_op_nand_i64: - case INDEX_op_nor_i64: - case INDEX_op_div_i64: - case INDEX_op_divu_i64: - case INDEX_op_rem_i64: - case INDEX_op_remu_i64: - case INDEX_op_mulsh_i64: - case INDEX_op_muluh_i64: - return C_O1_I2(r, r, r); - - case INDEX_op_sub_i32: - return C_O1_I2(r, rI, ri); - case INDEX_op_add_i64: - return C_O1_I2(r, r, rT); - case INDEX_op_or_i64: - case INDEX_op_xor_i64: - return C_O1_I2(r, r, rU); - case INDEX_op_sub_i64: - return C_O1_I2(r, rI, rT); - case INDEX_op_clz_i32: - case INDEX_op_ctz_i32: - case INDEX_op_clz_i64: - case INDEX_op_ctz_i64: - return C_O1_I2(r, r, rZW); - - case INDEX_op_brcond_i32: - case INDEX_op_brcond_i64: - return C_O0_I2(r, rC); - case INDEX_op_setcond_i32: - case INDEX_op_setcond_i64: - case INDEX_op_negsetcond_i32: - case INDEX_op_negsetcond_i64: - return C_O1_I2(r, r, rC); - case INDEX_op_movcond_i32: - case INDEX_op_movcond_i64: - return C_O1_I4(r, r, rC, rZ, rZ); - - case INDEX_op_deposit_i32: - case INDEX_op_deposit_i64: - return C_O1_I2(r, 0, rZ); - case INDEX_op_brcond2_i32: - return C_O0_I4(r, r, ri, ri); - case INDEX_op_setcond2_i32: - return C_O1_I4(r, r, r, ri, ri); - case INDEX_op_add2_i64: - case INDEX_op_add2_i32: - return C_O2_I4(r, r, r, r, rI, rZM); - case INDEX_op_sub2_i64: - case INDEX_op_sub2_i32: - return C_O2_I4(r, r, rI, rZM, r, r); - - case INDEX_op_qemu_ld_i32: - return C_O1_I1(r, r); - case INDEX_op_qemu_ld_i64: - return TCG_TARGET_REG_BITS == 64 ? C_O1_I1(r, r) : C_O2_I1(r, r, r); - - case INDEX_op_qemu_st_i32: - return C_O0_I2(r, r); - case INDEX_op_qemu_st_i64: - return TCG_TARGET_REG_BITS == 64 ? C_O0_I2(r, r) : C_O0_I3(r, r, r); - - case INDEX_op_qemu_ld_i128: - return C_N1O1_I1(o, m, r); - case INDEX_op_qemu_st_i128: - return C_O0_I3(o, m, r); + case INDEX_op_qemu_st2: + return TCG_TARGET_REG_BITS == 64 + ? C_O0_I3(o, m, r) : C_O0_I3(r, r, r); case INDEX_op_add_vec: case INDEX_op_sub_vec: |