From d5de7839d78b08c7bd14b03dac0413699b90da67 Mon Sep 17 00:00:00 2001 From: Bastian Koppelmann Date: Thu, 27 Nov 2014 14:30:33 +0000 Subject: target-tricore: Add instructions of RR opcode format, that have 0xb as the first opcode Add instructions of RR opcode format, that have 0xb as the first opcode. Add helper functions, for hword and byte arithmetics: * add_h_ssov/suov: Add two halfword and saturate on overflow. * sub_h_ssov/suov: Sub two halfword and saturate on overflow. * absdif_h_ssov: Compute absolute difference for halfwords and saturate on overflow. * abs_h_ssov/suov: Compute absolute value for two halfwords and saturate on overflow. * abs_b/h: Compute absolute value for four/two bytes/halfwords * absdif_b/h: Compute absolute difference for four/two bytes/halfwords * add_b/h: Add four/two bytes/halfwords. * sub_b/h: Sub four/two bytes/halfwords. * eq_b/h: Compare four/two bytes/halfwords with four/two bytes/halfwords on equality and set all bits of to either one ore zero. * eqany_b/h: Compare four/two bytes/halfwords with four/two bytes/halfwords on equality. * lt_b/bu/h/hu: Compare four/two bytes/halfwords with four/two bytes/halfwords on less than signed and unsigned. * max_b/bu/h/hu: Calculate max for four/two bytes/halfwords signed and unsigned. * min_b/bu/h/hu: Calculate min for four/two bytes/halfwords signed and unsigned. Add helper function abs_ssov, that computes the absolute value for a 32 bit integer and saturates on overflow. Add microcode generator functions: * gen_sub_CC: Caluclates sub and sets the carry bit. * gen_subc_CC: Caluclates sub and carry and sets the carry bit * gen_abs: Compute absolute value for a 32 bit integer. * gen_cond_w: Compares two 32 bit values on cond and sets result either zero or all bits one. OPC2_32_RR_MIN switched with OPC2_32_RR_MIN_U. Signed-off-by: Bastian Koppelmann Reviewed-by: Richard Henderson --- target-tricore/helper.h | 32 +++ target-tricore/op_helper.c | 525 +++++++++++++++++++++++++++++++++++++++ target-tricore/translate.c | 383 ++++++++++++++++++++++++++++ target-tricore/tricore-opcodes.h | 4 +- 4 files changed, 942 insertions(+), 2 deletions(-) diff --git a/target-tricore/helper.h b/target-tricore/helper.h index 6c07bd7f28..df87c0d2d7 100644 --- a/target-tricore/helper.h +++ b/target-tricore/helper.h @@ -18,8 +18,12 @@ /* Arithmetic */ DEF_HELPER_3(add_ssov, i32, env, i32, i32) DEF_HELPER_3(add_suov, i32, env, i32, i32) +DEF_HELPER_3(add_h_ssov, i32, env, i32, i32) +DEF_HELPER_3(add_h_suov, i32, env, i32, i32) DEF_HELPER_3(sub_ssov, i32, env, i32, i32) DEF_HELPER_3(sub_suov, i32, env, i32, i32) +DEF_HELPER_3(sub_h_ssov, i32, env, i32, i32) +DEF_HELPER_3(sub_h_suov, i32, env, i32, i32) DEF_HELPER_3(mul_ssov, i32, env, i32, i32) DEF_HELPER_3(mul_suov, i32, env, i32, i32) DEF_HELPER_3(sha_ssov, i32, env, i32, i32) @@ -32,6 +36,34 @@ DEF_HELPER_4(msub32_ssov, i32, env, i32, i32, i32) DEF_HELPER_4(msub32_suov, i32, env, i32, i32, i32) DEF_HELPER_4(msub64_ssov, i64, env, i32, i64, i32) DEF_HELPER_4(msub64_suov, i64, env, i32, i64, i32) +DEF_HELPER_3(absdif_h_ssov, i32, env, i32, i32) +DEF_HELPER_2(abs_ssov, i32, env, i32) +DEF_HELPER_2(abs_h_ssov, i32, env, i32) +/* hword/byte arithmetic */ +DEF_HELPER_2(abs_b, i32, env, i32) +DEF_HELPER_2(abs_h, i32, env, i32) +DEF_HELPER_3(absdif_b, i32, env, i32, i32) +DEF_HELPER_3(absdif_h, i32, env, i32, i32) +DEF_HELPER_3(add_b, i32, env, i32, i32) +DEF_HELPER_3(add_h, i32, env, i32, i32) +DEF_HELPER_3(sub_b, i32, env, i32, i32) +DEF_HELPER_3(sub_h, i32, env, i32, i32) +DEF_HELPER_FLAGS_2(eq_b, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(eq_h, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(eqany_b, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(eqany_h, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(lt_b, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(lt_bu, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(lt_h, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(lt_hu, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(max_b, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(max_bu, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(max_h, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(max_hu, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(min_b, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(min_bu, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(min_h, TCG_CALL_NO_RWG_SE, i32, i32, i32) +DEF_HELPER_FLAGS_2(min_hu, TCG_CALL_NO_RWG_SE, i32, i32, i32) /* CSA */ DEF_HELPER_2(call, void, env, i32) DEF_HELPER_1(ret, void, env) diff --git a/target-tricore/op_helper.c b/target-tricore/op_helper.c index f1a8d16d30..26e01ed1b9 100644 --- a/target-tricore/op_helper.c +++ b/target-tricore/op_helper.c @@ -103,6 +103,66 @@ static uint32_t suov32(CPUTriCoreState *env, int64_t arg) return ret; } +static uint32_t ssov16(CPUTriCoreState *env, int32_t hw0, int32_t hw1) +{ + int32_t max_pos = INT16_MAX; + int32_t max_neg = INT16_MIN; + int32_t av0, av1; + + env->PSW_USB_V = 0; + av0 = hw0 ^ hw0 * 2u; + if (hw0 > max_pos) { + env->PSW_USB_V = (1 << 31); + hw0 = max_pos; + } else if (hw0 < max_neg) { + env->PSW_USB_V = (1 << 31); + hw0 = max_neg; + } + + av1 = hw1 ^ hw1 * 2u; + if (hw1 > max_pos) { + env->PSW_USB_V = (1 << 31); + hw1 = max_pos; + } else if (hw1 < max_neg) { + env->PSW_USB_V = (1 << 31); + hw1 = max_neg; + } + + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = (av0 | av1) << 16; + env->PSW_USB_SAV |= env->PSW_USB_AV; + return (hw0 & 0xffff) | (hw1 << 16); +} + +static uint32_t suov16(CPUTriCoreState *env, int32_t hw0, int32_t hw1) +{ + int32_t max_pos = UINT16_MAX; + int32_t av0, av1; + + env->PSW_USB_V = 0; + av0 = hw0 ^ hw0 * 2u; + if (hw0 > max_pos) { + env->PSW_USB_V = (1 << 31); + hw0 = max_pos; + } else if (hw0 < 0) { + env->PSW_USB_V = (1 << 31); + hw0 = 0; + } + + av1 = hw1 ^ hw1 * 2u; + if (hw1 > max_pos) { + env->PSW_USB_V = (1 << 31); + hw1 = max_pos; + } else if (hw1 < 0) { + env->PSW_USB_V = (1 << 31); + hw1 = 0; + } + + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = (av0 | av1) << 16; + env->PSW_USB_SAV |= env->PSW_USB_AV; + return (hw0 & 0xffff) | (hw1 << 16); +} target_ulong helper_add_ssov(CPUTriCoreState *env, target_ulong r1, target_ulong r2) @@ -113,6 +173,16 @@ target_ulong helper_add_ssov(CPUTriCoreState *env, target_ulong r1, return ssov32(env, result); } +target_ulong helper_add_h_ssov(CPUTriCoreState *env, target_ulong r1, + target_ulong r2) +{ + int32_t ret_hw0, ret_hw1; + + ret_hw0 = sextract32(r1, 0, 16) + sextract32(r2, 0, 16); + ret_hw1 = sextract32(r1, 16, 16) + sextract32(r2, 16, 16); + return ssov16(env, ret_hw0, ret_hw1); +} + target_ulong helper_add_suov(CPUTriCoreState *env, target_ulong r1, target_ulong r2) { @@ -122,6 +192,16 @@ target_ulong helper_add_suov(CPUTriCoreState *env, target_ulong r1, return suov32(env, result); } +target_ulong helper_add_h_suov(CPUTriCoreState *env, target_ulong r1, + target_ulong r2) +{ + int32_t ret_hw0, ret_hw1; + + ret_hw0 = extract32(r1, 0, 16) + extract32(r2, 0, 16); + ret_hw1 = extract32(r1, 16, 16) + extract32(r2, 16, 16); + return suov16(env, ret_hw0, ret_hw1); +} + target_ulong helper_sub_ssov(CPUTriCoreState *env, target_ulong r1, target_ulong r2) { @@ -131,6 +211,16 @@ target_ulong helper_sub_ssov(CPUTriCoreState *env, target_ulong r1, return ssov32(env, result); } +target_ulong helper_sub_h_ssov(CPUTriCoreState *env, target_ulong r1, + target_ulong r2) +{ + int32_t ret_hw0, ret_hw1; + + ret_hw0 = sextract32(r1, 0, 16) - sextract32(r2, 0, 16); + ret_hw1 = sextract32(r1, 16, 16) - sextract32(r2, 16, 16); + return ssov16(env, ret_hw0, ret_hw1); +} + target_ulong helper_sub_suov(CPUTriCoreState *env, target_ulong r1, target_ulong r2) { @@ -140,6 +230,16 @@ target_ulong helper_sub_suov(CPUTriCoreState *env, target_ulong r1, return suov32(env, result); } +target_ulong helper_sub_h_suov(CPUTriCoreState *env, target_ulong r1, + target_ulong r2) +{ + int32_t ret_hw0, ret_hw1; + + ret_hw0 = extract32(r1, 0, 16) - extract32(r2, 0, 16); + ret_hw1 = extract32(r1, 16, 16) - extract32(r2, 16, 16); + return suov16(env, ret_hw0, ret_hw1); +} + target_ulong helper_mul_ssov(CPUTriCoreState *env, target_ulong r1, target_ulong r2) { @@ -174,6 +274,26 @@ target_ulong helper_sha_ssov(CPUTriCoreState *env, target_ulong r1, return ssov32(env, result); } +uint32_t helper_abs_ssov(CPUTriCoreState *env, target_ulong r1) +{ + target_ulong result; + result = ((int32_t)r1 >= 0) ? r1 : (0 - r1); + return ssov32(env, result); +} + +uint32_t helper_abs_h_ssov(CPUTriCoreState *env, target_ulong r1) +{ + int32_t ret_h0, ret_h1; + + ret_h0 = sextract32(r1, 0, 16); + ret_h0 = (ret_h0 >= 0) ? ret_h0 : (0 - ret_h0); + + ret_h1 = sextract32(r1, 16, 16); + ret_h1 = (ret_h1 >= 0) ? ret_h1 : (0 - ret_h1); + + return ssov16(env, ret_h0, ret_h1); +} + target_ulong helper_absdif_ssov(CPUTriCoreState *env, target_ulong r1, target_ulong r2) { @@ -189,6 +309,31 @@ target_ulong helper_absdif_ssov(CPUTriCoreState *env, target_ulong r1, return ssov32(env, result); } +uint32_t helper_absdif_h_ssov(CPUTriCoreState *env, target_ulong r1, + target_ulong r2) +{ + int32_t t1, t2; + int32_t ret_h0, ret_h1; + + t1 = sextract32(r1, 0, 16); + t2 = sextract32(r2, 0, 16); + if (t1 > t2) { + ret_h0 = t1 - t2; + } else { + ret_h0 = t2 - t1; + } + + t1 = sextract32(r1, 16, 16); + t2 = sextract32(r2, 16, 16); + if (t1 > t2) { + ret_h1 = t1 - t2; + } else { + ret_h1 = t2 - t1; + } + + return ssov16(env, ret_h0, ret_h1); +} + target_ulong helper_madd32_ssov(CPUTriCoreState *env, target_ulong r1, target_ulong r2, target_ulong r3) { @@ -348,6 +493,386 @@ uint64_t helper_msub64_suov(CPUTriCoreState *env, target_ulong r1, return ret; } +uint32_t helper_abs_b(CPUTriCoreState *env, target_ulong arg) +{ + int32_t b, i; + int32_t ovf = 0; + int32_t avf = 0; + int32_t ret = 0; + + for (i = 0; i < 4; i++) { + b = sextract32(arg, i * 8, 8); + b = (b >= 0) ? b : (0 - b); + ovf |= (b > 0x7F) || (b < -0x80); + avf |= b ^ b * 2u; + ret |= (b & 0xff) << (i * 8); + } + + env->PSW_USB_V = ovf << 31; + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = avf << 24; + env->PSW_USB_SAV |= env->PSW_USB_AV; + + return ret; +} + +uint32_t helper_abs_h(CPUTriCoreState *env, target_ulong arg) +{ + int32_t h, i; + int32_t ovf = 0; + int32_t avf = 0; + int32_t ret = 0; + + for (i = 0; i < 2; i++) { + h = sextract32(arg, i * 16, 16); + h = (h >= 0) ? h : (0 - h); + ovf |= (h > 0x7FFF) || (h < -0x8000); + avf |= h ^ h * 2u; + ret |= (h & 0xffff) << (i * 16); + } + + env->PSW_USB_V = ovf << 31; + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = avf << 16; + env->PSW_USB_SAV |= env->PSW_USB_AV; + + return ret; +} + +uint32_t helper_absdif_b(CPUTriCoreState *env, target_ulong r1, target_ulong r2) +{ + int32_t b, i; + int32_t extr_r2; + int32_t ovf = 0; + int32_t avf = 0; + int32_t ret = 0; + + for (i = 0; i < 4; i++) { + extr_r2 = sextract32(r2, i * 8, 8); + b = sextract32(r1, i * 8, 8); + b = (b > extr_r2) ? (b - extr_r2) : (extr_r2 - b); + ovf |= (b > 0x7F) || (b < -0x80); + avf |= b ^ b * 2u; + ret |= (b & 0xff) << (i * 8); + } + + env->PSW_USB_V = ovf << 31; + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = avf << 24; + env->PSW_USB_SAV |= env->PSW_USB_AV; + return ret; +} + +uint32_t helper_absdif_h(CPUTriCoreState *env, target_ulong r1, target_ulong r2) +{ + int32_t h, i; + int32_t extr_r2; + int32_t ovf = 0; + int32_t avf = 0; + int32_t ret = 0; + + for (i = 0; i < 2; i++) { + extr_r2 = sextract32(r2, i * 16, 16); + h = sextract32(r1, i * 16, 16); + h = (h > extr_r2) ? (h - extr_r2) : (extr_r2 - h); + ovf |= (h > 0x7FFF) || (h < -0x8000); + avf |= h ^ h * 2u; + ret |= (h & 0xffff) << (i * 16); + } + + env->PSW_USB_V = ovf << 31; + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = avf << 16; + env->PSW_USB_SAV |= env->PSW_USB_AV; + + return ret; +} + +uint32_t helper_add_b(CPUTriCoreState *env, target_ulong r1, target_ulong r2) +{ + int32_t b, i; + int32_t extr_r1, extr_r2; + int32_t ovf = 0; + int32_t avf = 0; + uint32_t ret = 0; + + for (i = 0; i < 4; i++) { + extr_r1 = sextract32(r1, i * 8, 8); + extr_r2 = sextract32(r2, i * 8, 8); + + b = extr_r1 + extr_r2; + ovf |= ((b > 0x7f) || (b < -0x80)); + avf |= b ^ b * 2u; + ret |= ((b & 0xff) << (i*8)); + } + + env->PSW_USB_V = (ovf << 31); + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = avf << 24; + env->PSW_USB_SAV |= env->PSW_USB_AV; + + return ret; +} + +uint32_t helper_add_h(CPUTriCoreState *env, target_ulong r1, target_ulong r2) +{ + int32_t h, i; + int32_t extr_r1, extr_r2; + int32_t ovf = 0; + int32_t avf = 0; + int32_t ret = 0; + + for (i = 0; i < 2; i++) { + extr_r1 = sextract32(r1, i * 16, 16); + extr_r2 = sextract32(r2, i * 16, 16); + h = extr_r1 + extr_r2; + ovf |= ((h > 0x7fff) || (h < -0x8000)); + avf |= h ^ h * 2u; + ret |= (h & 0xffff) << (i * 16); + } + + env->PSW_USB_V = (ovf << 31); + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = (avf << 16); + env->PSW_USB_SAV |= env->PSW_USB_AV; + + return ret; +} + +uint32_t helper_sub_b(CPUTriCoreState *env, target_ulong r1, target_ulong r2) +{ + int32_t b, i; + int32_t extr_r1, extr_r2; + int32_t ovf = 0; + int32_t avf = 0; + uint32_t ret = 0; + + for (i = 0; i < 4; i++) { + extr_r1 = sextract32(r1, i * 8, 8); + extr_r2 = sextract32(r2, i * 8, 8); + + b = extr_r1 - extr_r2; + ovf |= ((b > 0x7f) || (b < -0x80)); + avf |= b ^ b * 2u; + ret |= ((b & 0xff) << (i*8)); + } + + env->PSW_USB_V = (ovf << 31); + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = avf << 24; + env->PSW_USB_SAV |= env->PSW_USB_AV; + + return ret; +} + +uint32_t helper_sub_h(CPUTriCoreState *env, target_ulong r1, target_ulong r2) +{ + int32_t h, i; + int32_t extr_r1, extr_r2; + int32_t ovf = 0; + int32_t avf = 0; + int32_t ret = 0; + + for (i = 0; i < 2; i++) { + extr_r1 = sextract32(r1, i * 16, 16); + extr_r2 = sextract32(r2, i * 16, 16); + h = extr_r1 - extr_r2; + ovf |= ((h > 0x7fff) || (h < -0x8000)); + avf |= h ^ h * 2u; + ret |= (h & 0xffff) << (i * 16); + } + + env->PSW_USB_V = (ovf << 31); + env->PSW_USB_SV |= env->PSW_USB_V; + env->PSW_USB_AV = avf << 16; + env->PSW_USB_SAV |= env->PSW_USB_AV; + + return ret; +} + +uint32_t helper_eq_b(target_ulong r1, target_ulong r2) +{ + int32_t ret; + int32_t i, msk; + + ret = 0; + msk = 0xff; + for (i = 0; i < 4; i++) { + if ((r1 & msk) == (r2 & msk)) { + ret |= msk; + } + msk = msk << 8; + } + + return ret; +} + +uint32_t helper_eq_h(target_ulong r1, target_ulong r2) +{ + int32_t ret = 0; + + if ((r1 & 0xffff) == (r2 & 0xffff)) { + ret = 0xffff; + } + + if ((r1 & 0xffff0000) == (r2 & 0xffff0000)) { + ret |= 0xffff0000; + } + + return ret; +} + +uint32_t helper_eqany_b(target_ulong r1, target_ulong r2) +{ + int32_t i; + uint32_t ret = 0; + + for (i = 0; i < 4; i++) { + ret |= (sextract32(r1, i * 8, 8) == sextract32(r2, i * 8, 8)); + } + + return ret; +} + +uint32_t helper_eqany_h(target_ulong r1, target_ulong r2) +{ + uint32_t ret; + + ret = (sextract32(r1, 0, 16) == sextract32(r2, 0, 16)); + ret |= (sextract32(r1, 16, 16) == sextract32(r2, 16, 16)); + + return ret; +} + +uint32_t helper_lt_b(target_ulong r1, target_ulong r2) +{ + int32_t i; + uint32_t ret = 0; + + for (i = 0; i < 4; i++) { + if (sextract32(r1, i * 8, 8) < sextract32(r2, i * 8, 8)) { + ret |= (0xff << (i * 8)); + } + } + + return ret; +} + +uint32_t helper_lt_bu(target_ulong r1, target_ulong r2) +{ + int32_t i; + uint32_t ret = 0; + + for (i = 0; i < 4; i++) { + if (extract32(r1, i * 8, 8) < extract32(r2, i * 8, 8)) { + ret |= (0xff << (i * 8)); + } + } + + return ret; +} + +uint32_t helper_lt_h(target_ulong r1, target_ulong r2) +{ + uint32_t ret = 0; + + if (sextract32(r1, 0, 16) < sextract32(r2, 0, 16)) { + ret |= 0xffff; + } + + if (sextract32(r1, 16, 16) < sextract32(r2, 16, 16)) { + ret |= 0xffff0000; + } + + return ret; +} + +uint32_t helper_lt_hu(target_ulong r1, target_ulong r2) +{ + uint32_t ret = 0; + + if (extract32(r1, 0, 16) < extract32(r2, 0, 16)) { + ret |= 0xffff; + } + + if (extract32(r1, 16, 16) < extract32(r2, 16, 16)) { + ret |= 0xffff0000; + } + + return ret; +} + +#define EXTREMA_H_B(name, op) \ +uint32_t helper_##name ##_b(target_ulong r1, target_ulong r2) \ +{ \ + int32_t i, extr_r1, extr_r2; \ + uint32_t ret = 0; \ + \ + for (i = 0; i < 4; i++) { \ + extr_r1 = sextract32(r1, i * 8, 8); \ + extr_r2 = sextract32(r2, i * 8, 8); \ + extr_r1 = (extr_r1 op extr_r2) ? extr_r1 : extr_r2; \ + ret |= (extr_r1 & 0xff) << (i * 8); \ + } \ + return ret; \ +} \ + \ +uint32_t helper_##name ##_bu(target_ulong r1, target_ulong r2)\ +{ \ + int32_t i; \ + uint32_t extr_r1, extr_r2; \ + uint32_t ret = 0; \ + \ + for (i = 0; i < 4; i++) { \ + extr_r1 = extract32(r1, i * 8, 8); \ + extr_r2 = extract32(r2, i * 8, 8); \ + extr_r1 = (extr_r1 op extr_r2) ? extr_r1 : extr_r2; \ + ret |= (extr_r1 & 0xff) << (i * 8); \ + } \ + return ret; \ +} \ + \ +uint32_t helper_##name ##_h(target_ulong r1, target_ulong r2) \ +{ \ + int32_t extr_r1, extr_r2; \ + uint32_t ret = 0; \ + \ + extr_r1 = sextract32(r1, 0, 16); \ + extr_r2 = sextract32(r2, 0, 16); \ + ret = (extr_r1 op extr_r2) ? extr_r1 : extr_r2; \ + ret = ret & 0xffff; \ + \ + extr_r1 = sextract32(r1, 16, 16); \ + extr_r2 = sextract32(r2, 16, 16); \ + extr_r1 = (extr_r1 op extr_r2) ? extr_r1 : extr_r2; \ + ret |= extr_r1 << 16; \ + \ + return ret; \ +} \ + \ +uint32_t helper_##name ##_hu(target_ulong r1, target_ulong r2)\ +{ \ + uint32_t extr_r1, extr_r2; \ + uint32_t ret = 0; \ + \ + extr_r1 = extract32(r1, 0, 16); \ + extr_r2 = extract32(r2, 0, 16); \ + ret = (extr_r1 op extr_r2) ? extr_r1 : extr_r2; \ + ret = ret & 0xffff; \ + \ + extr_r1 = extract32(r1, 16, 16); \ + extr_r2 = extract32(r2, 16, 16); \ + extr_r1 = (extr_r1 op extr_r2) ? extr_r1 : extr_r2; \ + ret |= extr_r1 << (16); \ + \ + return ret; \ +} \ + +EXTREMA_H_B(max, >) +EXTREMA_H_B(min, <) + +#undef EXTREMA_H_B + /* context save area (CSA) related helpers */ static int cdc_increment(target_ulong *psw) diff --git a/target-tricore/translate.c b/target-tricore/translate.c index 8f9e13e085..7111b52dd7 100644 --- a/target-tricore/translate.c +++ b/target-tricore/translate.c @@ -728,6 +728,62 @@ static inline void gen_sub_d(TCGv ret, TCGv r1, TCGv r2) tcg_temp_free(result); } +static inline void gen_sub_CC(TCGv ret, TCGv r1, TCGv r2) +{ + TCGv result = tcg_temp_new(); + TCGv temp = tcg_temp_new(); + + tcg_gen_sub_tl(result, r1, r2); + /* calc C bit */ + tcg_gen_setcond_tl(TCG_COND_GEU, cpu_PSW_C, r1, r2); + /* calc V bit */ + tcg_gen_xor_tl(cpu_PSW_V, result, r1); + tcg_gen_xor_tl(temp, r1, r2); + tcg_gen_and_tl(cpu_PSW_V, cpu_PSW_V, temp); + /* calc SV bit */ + tcg_gen_or_tl(cpu_PSW_SV, cpu_PSW_SV, cpu_PSW_V); + /* Calc AV bit */ + tcg_gen_add_tl(cpu_PSW_AV, result, result); + tcg_gen_xor_tl(cpu_PSW_AV, result, cpu_PSW_AV); + /* calc SAV bit */ + tcg_gen_or_tl(cpu_PSW_SAV, cpu_PSW_SAV, cpu_PSW_AV); + /* write back result */ + tcg_gen_mov_tl(ret, result); + + tcg_temp_free(result); + tcg_temp_free(temp); +} + +static inline void gen_subc_CC(TCGv ret, TCGv r1, TCGv r2) +{ + TCGv temp = tcg_temp_new(); + tcg_gen_not_tl(temp, r2); + gen_addc_CC(ret, r1, temp); + tcg_temp_free(temp); +} + +static inline void gen_abs(TCGv ret, TCGv r1) +{ + TCGv temp = tcg_temp_new(); + TCGv t0 = tcg_const_i32(0); + + tcg_gen_neg_tl(temp, r1); + tcg_gen_movcond_tl(TCG_COND_GE, ret, r1, t0, r1, temp); + /* overflow can only happen, if r1 = 0x80000000 */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_PSW_V, r1, 0x80000000); + tcg_gen_shli_tl(cpu_PSW_V, cpu_PSW_V, 31); + /* calc SV bit */ + tcg_gen_or_tl(cpu_PSW_SV, cpu_PSW_SV, cpu_PSW_V); + /* Calc AV bit */ + tcg_gen_add_tl(cpu_PSW_AV, ret, ret); + tcg_gen_xor_tl(cpu_PSW_AV, ret, cpu_PSW_AV); + /* calc SAV bit */ + tcg_gen_or_tl(cpu_PSW_SAV, cpu_PSW_SAV, cpu_PSW_AV); + + tcg_temp_free(temp); + tcg_temp_free(t0); +} + static inline void gen_absdif(TCGv ret, TCGv r1, TCGv r2) { TCGv temp = tcg_temp_new_i32(); @@ -1217,6 +1273,13 @@ gen_accumulating_condi(int cond, TCGv ret, TCGv r1, int32_t con, tcg_temp_free(temp); } +/* ret = (r1 cond r2) ? 0xFFFFFFFF ? 0x00000000;*/ +static inline void gen_cond_w(TCGCond cond, TCGv ret, TCGv r1, TCGv r2) +{ + tcg_gen_setcond_tl(cond, ret, r1, r2); + tcg_gen_neg_tl(ret, ret); +} + static inline void gen_eqany_bi(TCGv ret, TCGv r1, int32_t con) { TCGv b0 = tcg_temp_new(); @@ -3817,6 +3880,322 @@ static void decode_rlc_opc(CPUTriCoreState *env, DisasContext *ctx, } } +/* RR format */ +static void decode_rr_accumulator(CPUTriCoreState *env, DisasContext *ctx) +{ + uint32_t op2; + int r3, r2, r1; + + r3 = MASK_OP_RR_D(ctx->opcode); + r2 = MASK_OP_RR_S2(ctx->opcode); + r1 = MASK_OP_RR_S1(ctx->opcode); + op2 = MASK_OP_RR_OP2(ctx->opcode); + + switch (op2) { + case OPC2_32_RR_ABS: + gen_abs(cpu_gpr_d[r3], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ABS_B: + gen_helper_abs_b(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ABS_H: + gen_helper_abs_h(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ABSDIF: + gen_absdif(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ABSDIF_B: + gen_helper_absdif_b(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ABSDIF_H: + gen_helper_absdif_h(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ABSDIFS: + gen_helper_absdif_ssov(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ABSDIFS_H: + gen_helper_absdif_h_ssov(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ABSS: + gen_helper_abs_ssov(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ABSS_H: + gen_helper_abs_h_ssov(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ADD: + gen_add_d(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ADD_B: + gen_helper_add_b(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ADD_H: + gen_helper_add_h(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ADDC: + gen_addc_CC(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ADDS: + gen_adds(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ADDS_H: + gen_helper_add_h_ssov(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ADDS_HU: + gen_helper_add_h_suov(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ADDS_U: + gen_helper_add_suov(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_ADDX: + gen_add_CC(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_AND_EQ: + gen_accumulating_cond(TCG_COND_EQ, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_and_tl); + break; + case OPC2_32_RR_AND_GE: + gen_accumulating_cond(TCG_COND_GE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_and_tl); + break; + case OPC2_32_RR_AND_GE_U: + gen_accumulating_cond(TCG_COND_GEU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_and_tl); + break; + case OPC2_32_RR_AND_LT: + gen_accumulating_cond(TCG_COND_LT, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_and_tl); + break; + case OPC2_32_RR_AND_LT_U: + gen_accumulating_cond(TCG_COND_LTU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_and_tl); + break; + case OPC2_32_RR_AND_NE: + gen_accumulating_cond(TCG_COND_NE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_and_tl); + break; + case OPC2_32_RR_EQ: + tcg_gen_setcond_tl(TCG_COND_EQ, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_EQ_B: + gen_helper_eq_b(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_EQ_H: + gen_helper_eq_h(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_EQ_W: + gen_cond_w(TCG_COND_EQ, cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_EQANY_B: + gen_helper_eqany_b(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_EQANY_H: + gen_helper_eqany_h(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_GE: + tcg_gen_setcond_tl(TCG_COND_GE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_GE_U: + tcg_gen_setcond_tl(TCG_COND_GEU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_LT: + tcg_gen_setcond_tl(TCG_COND_LT, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_LT_U: + tcg_gen_setcond_tl(TCG_COND_LTU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_LT_B: + gen_helper_lt_b(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_LT_BU: + gen_helper_lt_bu(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_LT_H: + gen_helper_lt_h(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_LT_HU: + gen_helper_lt_hu(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_LT_W: + gen_cond_w(TCG_COND_LT, cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_LT_WU: + gen_cond_w(TCG_COND_LTU, cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MAX: + tcg_gen_movcond_tl(TCG_COND_GT, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MAX_U: + tcg_gen_movcond_tl(TCG_COND_GTU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MAX_B: + gen_helper_max_b(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MAX_BU: + gen_helper_max_bu(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MAX_H: + gen_helper_max_h(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MAX_HU: + gen_helper_max_hu(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MIN: + tcg_gen_movcond_tl(TCG_COND_LT, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MIN_U: + tcg_gen_movcond_tl(TCG_COND_LTU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MIN_B: + gen_helper_min_b(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MIN_BU: + gen_helper_min_bu(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MIN_H: + gen_helper_min_h(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MIN_HU: + gen_helper_min_hu(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_MOV: + tcg_gen_mov_tl(cpu_gpr_d[r3], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_NE: + tcg_gen_setcond_tl(TCG_COND_NE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_OR_EQ: + gen_accumulating_cond(TCG_COND_EQ, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_or_tl); + break; + case OPC2_32_RR_OR_GE: + gen_accumulating_cond(TCG_COND_GE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_or_tl); + break; + case OPC2_32_RR_OR_GE_U: + gen_accumulating_cond(TCG_COND_GEU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_or_tl); + break; + case OPC2_32_RR_OR_LT: + gen_accumulating_cond(TCG_COND_LT, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_or_tl); + break; + case OPC2_32_RR_OR_LT_U: + gen_accumulating_cond(TCG_COND_LTU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_or_tl); + break; + case OPC2_32_RR_OR_NE: + gen_accumulating_cond(TCG_COND_NE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_or_tl); + break; + case OPC2_32_RR_SAT_B: + gen_saturate(cpu_gpr_d[r3], cpu_gpr_d[r1], 0x7f, -0x80); + break; + case OPC2_32_RR_SAT_BU: + gen_saturate_u(cpu_gpr_d[r3], cpu_gpr_d[r1], 0xff); + break; + case OPC2_32_RR_SAT_H: + gen_saturate(cpu_gpr_d[r3], cpu_gpr_d[r1], 0x7fff, -0x8000); + break; + case OPC2_32_RR_SAT_HU: + gen_saturate_u(cpu_gpr_d[r3], cpu_gpr_d[r1], 0xffff); + break; + case OPC2_32_RR_SH_EQ: + gen_sh_cond(TCG_COND_EQ, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SH_GE: + gen_sh_cond(TCG_COND_GE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SH_GE_U: + gen_sh_cond(TCG_COND_GEU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SH_LT: + gen_sh_cond(TCG_COND_LT, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SH_LT_U: + gen_sh_cond(TCG_COND_LTU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SH_NE: + gen_sh_cond(TCG_COND_NE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SUB: + gen_sub_d(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SUB_B: + gen_helper_sub_b(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SUB_H: + gen_helper_sub_h(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SUBC: + gen_subc_CC(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SUBS: + gen_subs(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SUBS_U: + gen_subsu(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SUBS_H: + gen_helper_sub_h_ssov(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SUBS_HU: + gen_helper_sub_h_suov(cpu_gpr_d[r3], cpu_env, cpu_gpr_d[r1], + cpu_gpr_d[r2]); + break; + case OPC2_32_RR_SUBX: + gen_sub_CC(cpu_gpr_d[r3], cpu_gpr_d[r1], cpu_gpr_d[r2]); + break; + case OPC2_32_RR_XOR_EQ: + gen_accumulating_cond(TCG_COND_EQ, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_xor_tl); + break; + case OPC2_32_RR_XOR_GE: + gen_accumulating_cond(TCG_COND_GE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_xor_tl); + break; + case OPC2_32_RR_XOR_GE_U: + gen_accumulating_cond(TCG_COND_GEU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_xor_tl); + break; + case OPC2_32_RR_XOR_LT: + gen_accumulating_cond(TCG_COND_LT, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_xor_tl); + break; + case OPC2_32_RR_XOR_LT_U: + gen_accumulating_cond(TCG_COND_LTU, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_xor_tl); + break; + case OPC2_32_RR_XOR_NE: + gen_accumulating_cond(TCG_COND_NE, cpu_gpr_d[r3], cpu_gpr_d[r1], + cpu_gpr_d[r2], &tcg_gen_xor_tl); + break; + } +} + static void decode_32Bit_opc(CPUTriCoreState *env, DisasContext *ctx) { int op1; @@ -4048,6 +4427,10 @@ static void decode_32Bit_opc(CPUTriCoreState *env, DisasContext *ctx) case OPC1_32_RLC_MTCR: decode_rlc_opc(env, ctx, op1); break; +/* RR Format */ + case OPCM_32_RR_ACCUMULATOR: + decode_rr_accumulator(env, ctx); + break; } } diff --git a/target-tricore/tricore-opcodes.h b/target-tricore/tricore-opcodes.h index a76a7e4c45..17d9ad950b 100644 --- a/target-tricore/tricore-opcodes.h +++ b/target-tricore/tricore-opcodes.h @@ -1034,8 +1034,8 @@ enum { OPC2_32_RR_MAX_BU = 0x5b, OPC2_32_RR_MAX_H = 0x7a, OPC2_32_RR_MAX_HU = 0x7b, - OPC2_32_RR_MIN = 0x19, - OPC2_32_RR_MIN_U = 0x18, + OPC2_32_RR_MIN = 0x18, + OPC2_32_RR_MIN_U = 0x19, OPC2_32_RR_MIN_B = 0x58, OPC2_32_RR_MIN_BU = 0x59, OPC2_32_RR_MIN_H = 0x78, -- cgit v1.2.3