From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:48580) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fGpdH-00057W-4Z for qemu-devel@nongnu.org; Thu, 10 May 2018 13:45:36 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fGpdF-0000uW-GS for qemu-devel@nongnu.org; Thu, 10 May 2018 13:45:35 -0400 Received: from orth.archaic.org.uk ([2001:8b0:1d0::2]:41604) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1fGpdF-0000tL-7C for qemu-devel@nongnu.org; Thu, 10 May 2018 13:45:33 -0400 Received: from pm215 by orth.archaic.org.uk with local (Exim 4.89) (envelope-from ) id 1fGpdE-0003ZJ-0W for qemu-devel@nongnu.org; Thu, 10 May 2018 18:45:32 +0100 From: Peter Maydell Date: Thu, 10 May 2018 18:45:12 +0100 Message-Id: <20180510174519.11264-15-peter.maydell@linaro.org> In-Reply-To: <20180510174519.11264-1-peter.maydell@linaro.org> References: <20180510174519.11264-1-peter.maydell@linaro.org> Subject: [Qemu-devel] [PULL 14/21] target/arm: Introduce ARM_FEATURE_V8_ATOMICS and initial decode List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org From: Richard Henderson The insns in the ARMv8.1-Atomics are added to the existing load/store exclusive and load/store reg opcode spaces. Rearrange the top-level decoders for these to accomodate. The Atomics insns themselves still generate Unallocated. Signed-off-by: Richard Henderson Message-id: 20180508151437.4232-8-richard.henderson@linaro.org [PMM: Drop the ARM_FEATURE_V8_1 feature flag] Reviewed-by: Peter Maydell Signed-off-by: Peter Maydell --- target/arm/cpu.h | 1 + linux-user/elfload.c | 1 + target/arm/translate-a64.c | 182 +++++++++++++++++++++++++++---------- 3 files changed, 138 insertions(+), 46 deletions(-) diff --git a/target/arm/cpu.h b/target/arm/cpu.h index 44e6b77151..3b086be570 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -1449,6 +1449,7 @@ enum arm_features { ARM_FEATURE_V8_SHA3, /* implements SHA3 part of v8 Crypto Extensions */ ARM_FEATURE_V8_SM3, /* implements SM3 part of v8 Crypto Extensions */ ARM_FEATURE_V8_SM4, /* implements SM4 part of v8 Crypto Extensions */ + ARM_FEATURE_V8_ATOMICS, /* ARMv8.1-Atomics feature */ ARM_FEATURE_V8_RDM, /* implements v8.1 simd round multiply */ ARM_FEATURE_V8_FP16, /* implements v8.2 half-precision float */ ARM_FEATURE_V8_FCMA, /* has complex number part of v8.3 extensions. */ diff --git a/linux-user/elfload.c b/linux-user/elfload.c index 36d52194bc..13bc78d0c8 100644 --- a/linux-user/elfload.c +++ b/linux-user/elfload.c @@ -581,6 +581,7 @@ static uint32_t get_elf_hwcap(void) GET_FEATURE(ARM_FEATURE_V8_SHA512, ARM_HWCAP_A64_SHA512); GET_FEATURE(ARM_FEATURE_V8_FP16, ARM_HWCAP_A64_FPHP | ARM_HWCAP_A64_ASIMDHP); + GET_FEATURE(ARM_FEATURE_V8_ATOMICS, ARM_HWCAP_A64_ATOMICS); GET_FEATURE(ARM_FEATURE_V8_RDM, ARM_HWCAP_A64_ASIMDRDM); GET_FEATURE(ARM_FEATURE_V8_FCMA, ARM_HWCAP_A64_FCMA); #undef GET_FEATURE diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index 60d104cc8a..bb8a176f9a 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -2147,62 +2147,98 @@ static void disas_ldst_excl(DisasContext *s, uint32_t insn) int rt = extract32(insn, 0, 5); int rn = extract32(insn, 5, 5); int rt2 = extract32(insn, 10, 5); - int is_lasr = extract32(insn, 15, 1); int rs = extract32(insn, 16, 5); - int is_pair = extract32(insn, 21, 1); - int is_store = !extract32(insn, 22, 1); - int is_excl = !extract32(insn, 23, 1); + int is_lasr = extract32(insn, 15, 1); + int o2_L_o1_o0 = extract32(insn, 21, 3) * 2 | is_lasr; int size = extract32(insn, 30, 2); TCGv_i64 tcg_addr; - if ((!is_excl && !is_pair && !is_lasr) || - (!is_excl && is_pair) || - (is_pair && size < 2)) { - unallocated_encoding(s); + switch (o2_L_o1_o0) { + case 0x0: /* STXR */ + case 0x1: /* STLXR */ + if (rn == 31) { + gen_check_sp_alignment(s); + } + if (is_lasr) { + tcg_gen_mb(TCG_MO_ALL | TCG_BAR_STRL); + } + tcg_addr = read_cpu_reg_sp(s, rn, 1); + gen_store_exclusive(s, rs, rt, rt2, tcg_addr, size, false); return; - } - if (rn == 31) { - gen_check_sp_alignment(s); - } - tcg_addr = read_cpu_reg_sp(s, rn, 1); - - /* Note that since TCG is single threaded load-acquire/store-release - * semantics require no extra if (is_lasr) { ... } handling. - */ - - if (is_excl) { - if (!is_store) { - s->is_ldex = true; - gen_load_exclusive(s, rt, rt2, tcg_addr, size, is_pair); - if (is_lasr) { - tcg_gen_mb(TCG_MO_ALL | TCG_BAR_LDAQ); - } - } else { - if (is_lasr) { - tcg_gen_mb(TCG_MO_ALL | TCG_BAR_STRL); - } - gen_store_exclusive(s, rs, rt, rt2, tcg_addr, size, is_pair); + case 0x4: /* LDXR */ + case 0x5: /* LDAXR */ + if (rn == 31) { + gen_check_sp_alignment(s); } - } else { - TCGv_i64 tcg_rt = cpu_reg(s, rt); - bool iss_sf = disas_ldst_compute_iss_sf(size, false, 0); + tcg_addr = read_cpu_reg_sp(s, rn, 1); + s->is_ldex = true; + gen_load_exclusive(s, rt, rt2, tcg_addr, size, false); + if (is_lasr) { + tcg_gen_mb(TCG_MO_ALL | TCG_BAR_LDAQ); + } + return; + case 0x9: /* STLR */ /* Generate ISS for non-exclusive accesses including LASR. */ - if (is_store) { + if (rn == 31) { + gen_check_sp_alignment(s); + } + tcg_gen_mb(TCG_MO_ALL | TCG_BAR_STRL); + tcg_addr = read_cpu_reg_sp(s, rn, 1); + do_gpr_st(s, cpu_reg(s, rt), tcg_addr, size, true, rt, + disas_ldst_compute_iss_sf(size, false, 0), is_lasr); + return; + + case 0xd: /* LDAR */ + /* Generate ISS for non-exclusive accesses including LASR. */ + if (rn == 31) { + gen_check_sp_alignment(s); + } + tcg_addr = read_cpu_reg_sp(s, rn, 1); + do_gpr_ld(s, cpu_reg(s, rt), tcg_addr, size, false, false, true, rt, + disas_ldst_compute_iss_sf(size, false, 0), is_lasr); + tcg_gen_mb(TCG_MO_ALL | TCG_BAR_LDAQ); + return; + + case 0x2: case 0x3: /* CASP / STXP */ + if (size & 2) { /* STXP / STLXP */ + if (rn == 31) { + gen_check_sp_alignment(s); + } if (is_lasr) { tcg_gen_mb(TCG_MO_ALL | TCG_BAR_STRL); } - do_gpr_st(s, tcg_rt, tcg_addr, size, - true, rt, iss_sf, is_lasr); - } else { - do_gpr_ld(s, tcg_rt, tcg_addr, size, false, false, - true, rt, iss_sf, is_lasr); + tcg_addr = read_cpu_reg_sp(s, rn, 1); + gen_store_exclusive(s, rs, rt, rt2, tcg_addr, size, true); + return; + } + /* CASP / CASPL */ + break; + + case 0x6: case 0x7: /* CASP / LDXP */ + if (size & 2) { /* LDXP / LDAXP */ + if (rn == 31) { + gen_check_sp_alignment(s); + } + tcg_addr = read_cpu_reg_sp(s, rn, 1); + s->is_ldex = true; + gen_load_exclusive(s, rt, rt2, tcg_addr, size, true); if (is_lasr) { tcg_gen_mb(TCG_MO_ALL | TCG_BAR_LDAQ); } + return; } + /* CASPA / CASPAL */ + break; + + case 0xa: /* CAS */ + case 0xb: /* CASL */ + case 0xe: /* CASA */ + case 0xf: /* CASAL */ + break; } + unallocated_encoding(s); } /* @@ -2715,6 +2751,55 @@ static void disas_ldst_reg_unsigned_imm(DisasContext *s, uint32_t insn, } } +/* Atomic memory operations + * + * 31 30 27 26 24 22 21 16 15 12 10 5 0 + * +------+-------+---+-----+-----+---+----+----+-----+-----+----+-----+ + * | size | 1 1 1 | V | 0 0 | A R | 1 | Rs | o3 | opc | 0 0 | Rn | Rt | + * +------+-------+---+-----+-----+--------+----+-----+-----+----+-----+ + * + * Rt: the result register + * Rn: base address or SP + * Rs: the source register for the operation + * V: vector flag (always 0 as of v8.3) + * A: acquire flag + * R: release flag + */ +static void disas_ldst_atomic(DisasContext *s, uint32_t insn, + int size, int rt, bool is_vector) +{ + int rs = extract32(insn, 16, 5); + int rn = extract32(insn, 5, 5); + int o3_opc = extract32(insn, 12, 4); + int feature = ARM_FEATURE_V8_ATOMICS; + + if (is_vector) { + unallocated_encoding(s); + return; + } + switch (o3_opc) { + case 000: /* LDADD */ + case 001: /* LDCLR */ + case 002: /* LDEOR */ + case 003: /* LDSET */ + case 004: /* LDSMAX */ + case 005: /* LDSMIN */ + case 006: /* LDUMAX */ + case 007: /* LDUMIN */ + case 010: /* SWP */ + default: + unallocated_encoding(s); + return; + } + if (!arm_dc_feature(s, feature)) { + unallocated_encoding(s); + return; + } + + (void)rs; + (void)rn; +} + /* Load/store register (all forms) */ static void disas_ldst_reg(DisasContext *s, uint32_t insn) { @@ -2725,23 +2810,28 @@ static void disas_ldst_reg(DisasContext *s, uint32_t insn) switch (extract32(insn, 24, 2)) { case 0: - if (extract32(insn, 21, 1) == 1 && extract32(insn, 10, 2) == 2) { - disas_ldst_reg_roffset(s, insn, opc, size, rt, is_vector); - } else { + if (extract32(insn, 21, 1) == 0) { /* Load/store register (unscaled immediate) * Load/store immediate pre/post-indexed * Load/store register unprivileged */ disas_ldst_reg_imm9(s, insn, opc, size, rt, is_vector); + return; + } + switch (extract32(insn, 10, 2)) { + case 0: + disas_ldst_atomic(s, insn, size, rt, is_vector); + return; + case 2: + disas_ldst_reg_roffset(s, insn, opc, size, rt, is_vector); + return; } break; case 1: disas_ldst_reg_unsigned_imm(s, insn, opc, size, rt, is_vector); - break; - default: - unallocated_encoding(s); - break; + return; } + unallocated_encoding(s); } /* AdvSIMD load/store multiple structures -- 2.17.0