Hi; here's the latest round of arm patches. I have included also my patchset for the RTC devices to avoid keeping time_t and time_t diffs in 32-bit variables. thanks -- PMM The following changes since commit 156618d9ea67f2f2e31d9dedd97f2dcccbe6808c: Merge tag 'block-pull-request' of https://gitlab.com/stefanha/qemu into staging (2023-08-30 09:20:27 -0400) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20230831 for you to fetch changes up to e73b8bb8a3e9a162f70e9ffbf922d4fafc96bbfb: hw/arm: Set number of MPU regions correctly for an505, an521, an524 (2023-08-31 11:07:02 +0100) ---------------------------------------------------------------- target-arm queue: * Some of the preliminary patches for Cortex-A710 support * i.MX7 and i.MX6UL refactoring * Implement SRC device for i.MX7 * Catch illegal-exception-return from EL3 with bad NSE/NS * Use 64-bit offsets for holding time_t differences in RTC devices * Model correct number of MPU regions for an505, an521, an524 boards ---------------------------------------------------------------- Alex Bennée (1): target/arm: properly document FEAT_CRC32 Jean-Christophe Dubois (6): Remove i.MX7 IOMUX GPR device from i.MX6UL Refactor i.MX6UL processor code Add i.MX6UL missing devices. Refactor i.MX7 processor code Add i.MX7 missing TZ devices and memory regions Add i.MX7 SRC device implementation Peter Maydell (8): target/arm: Catch illegal-exception-return from EL3 with bad NSE/NS hw/rtc/m48t59: Use 64-bit arithmetic in set_alarm() hw/rtc/twl92230: Use int64_t for sec_offset and alm_sec hw/rtc/aspeed_rtc: Use 64-bit offset for holding time_t difference rtc: Use time_t for passing and returning time offsets target/arm: Do all "ARM_FEATURE_X implies Y" checks in post_init hw/arm/armv7m: Add mpu-ns-regions and mpu-s-regions properties hw/arm: Set number of MPU regions correctly for an505, an521, an524 Richard Henderson (9): target/arm: Reduce dcz_blocksize to uint8_t target/arm: Allow cpu to configure GM blocksize target/arm: Support more GM blocksizes target/arm: When tag memory is not present, set MTE=1 target/arm: Introduce make_ccsidr64 target/arm: Apply access checks to neoverse-n1 special registers target/arm: Apply access checks to neoverse-v1 special registers target/arm: Suppress FEAT_TRBE (Trace Buffer Extension) target/arm: Implement FEAT_HPDS2 as a no-op docs/system/arm/emulation.rst | 2 + include/hw/arm/armsse.h | 5 + include/hw/arm/armv7m.h | 8 + include/hw/arm/fsl-imx6ul.h | 158 ++++++++++++++++--- include/hw/arm/fsl-imx7.h | 338 ++++++++++++++++++++++++++++++----------- include/hw/misc/imx7_src.h | 66 ++++++++ include/hw/rtc/aspeed_rtc.h | 2 +- include/sysemu/rtc.h | 4 +- target/arm/cpregs.h | 2 + target/arm/cpu.h | 5 +- target/arm/internals.h | 6 - target/arm/tcg/translate.h | 2 + hw/arm/armsse.c | 16 ++ hw/arm/armv7m.c | 21 +++ hw/arm/fsl-imx6ul.c | 174 +++++++++++++-------- hw/arm/fsl-imx7.c | 201 +++++++++++++++++++----- hw/arm/mps2-tz.c | 29 ++++ hw/misc/imx7_src.c | 276 +++++++++++++++++++++++++++++++++ hw/rtc/aspeed_rtc.c | 5 +- hw/rtc/m48t59.c | 2 +- hw/rtc/twl92230.c | 4 +- softmmu/rtc.c | 4 +- target/arm/cpu.c | 207 ++++++++++++++----------- target/arm/helper.c | 15 +- target/arm/tcg/cpu32.c | 2 +- target/arm/tcg/cpu64.c | 102 +++++++++---- target/arm/tcg/helper-a64.c | 9 ++ target/arm/tcg/mte_helper.c | 90 ++++++++--- target/arm/tcg/translate-a64.c | 5 +- hw/misc/meson.build | 1 + hw/misc/trace-events | 4 + 31 files changed, 1393 insertions(+), 372 deletions(-) create mode 100644 include/hw/misc/imx7_src.h create mode 100644 hw/misc/imx7_src.c From: Richard Henderson <richard.henderson@linaro.org> This value is only 4 bits wide. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-id: 20230811214031.171020-2-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/target/arm/cpu.h b/target/arm/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -XXX,XX +XXX,XX @@ struct ArchCPU { bool prop_lpa2; /* DCZ blocksize, in log_2(words), ie low 4 bits of DCZID_EL0 */ - uint32_t dcz_blocksize; + uint8_t dcz_blocksize; + uint64_t rvbar_prop; /* Property/input signals. */ /* Configurable aspects of GIC cpu interface (which is part of the CPU) */ -- 2.34.1 From: Richard Henderson <richard.henderson@linaro.org> Previously we hard-coded the blocksize with GMID_EL1_BS. But the value we choose for -cpu max does not match the value that cortex-a710 uses. Mirror the way we handle dcz_blocksize. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230811214031.171020-3-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 2 ++ target/arm/internals.h | 6 ----- target/arm/tcg/translate.h | 2 ++ target/arm/helper.c | 11 +++++--- target/arm/tcg/cpu64.c | 1 + target/arm/tcg/mte_helper.c | 46 ++++++++++++++++++++++------------ target/arm/tcg/translate-a64.c | 5 ++-- 7 files changed, 45 insertions(+), 28 deletions(-) diff --git a/target/arm/cpu.h b/target/arm/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -XXX,XX +XXX,XX @@ struct ArchCPU { /* DCZ blocksize, in log_2(words), ie low 4 bits of DCZID_EL0 */ uint8_t dcz_blocksize; + /* GM blocksize, in log_2(words), ie low 4 bits of GMID_EL0 */ + uint8_t gm_blocksize; uint64_t rvbar_prop; /* Property/input signals. */ diff --git a/target/arm/internals.h b/target/arm/internals.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/internals.h +++ b/target/arm/internals.h @@ -XXX,XX +XXX,XX @@ void arm_log_exception(CPUState *cs); #endif /* !CONFIG_USER_ONLY */ -/* - * The log2 of the words in the tag block, for GMID_EL1.BS. - * The is the maximum, 256 bytes, which manipulates 64-bits of tags. - */ -#define GMID_EL1_BS 6 - /* * SVE predicates are 1/8 the size of SVE vectors, and cannot use * the same simd_desc() encoding due to restrictions on size. diff --git a/target/arm/tcg/translate.h b/target/arm/tcg/translate.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate.h +++ b/target/arm/tcg/translate.h @@ -XXX,XX +XXX,XX @@ typedef struct DisasContext { int8_t btype; /* A copy of cpu->dcz_blocksize. */ uint8_t dcz_blocksize; + /* A copy of cpu->gm_blocksize. */ + uint8_t gm_blocksize; /* True if this page is guarded. */ bool guarded_page; /* Bottom two bits of XScale c15_cpar coprocessor access control reg */ diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo mte_reginfo[] = { .opc0 = 3, .opc1 = 0, .crn = 1, .crm = 0, .opc2 = 6, .access = PL1_RW, .accessfn = access_mte, .fieldoffset = offsetof(CPUARMState, cp15.gcr_el1) }, - { .name = "GMID_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 0, .crm = 0, .opc2 = 4, - .access = PL1_R, .accessfn = access_aa64_tid5, - .type = ARM_CP_CONST, .resetvalue = GMID_EL1_BS }, { .name = "TCO", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 3, .crn = 4, .crm = 2, .opc2 = 7, .type = ARM_CP_NO_RAW, @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) * then define only a RAZ/WI version of PSTATE.TCO. */ if (cpu_isar_feature(aa64_mte, cpu)) { + ARMCPRegInfo gmid_reginfo = { + .name = "GMID_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 0, .crm = 0, .opc2 = 4, + .access = PL1_R, .accessfn = access_aa64_tid5, + .type = ARM_CP_CONST, .resetvalue = cpu->gm_blocksize, + }; + define_one_arm_cp_reg(cpu, &gmid_reginfo); define_arm_cp_regs(cpu, mte_reginfo); define_arm_cp_regs(cpu, mte_el0_cacheop_reginfo); } else if (cpu_isar_feature(aa64_mte_insn_reg, cpu)) { diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu64.c +++ b/target/arm/tcg/cpu64.c @@ -XXX,XX +XXX,XX @@ void aarch64_max_tcg_initfn(Object *obj) cpu->ctr = 0x80038003; /* 32 byte I and D cacheline size, VIPT icache */ cpu->dcz_blocksize = 7; /* 512 bytes */ #endif + cpu->gm_blocksize = 6; /* 256 bytes */ cpu->sve_vq.supported = MAKE_64BIT_MASK(0, ARM_MAX_VQ); cpu->sme_vq.supported = SVE_VQ_POW2_MAP; diff --git a/target/arm/tcg/mte_helper.c b/target/arm/tcg/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/mte_helper.c +++ b/target/arm/tcg/mte_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(st2g_stub)(CPUARMState *env, uint64_t ptr) } } -#define LDGM_STGM_SIZE (4 << GMID_EL1_BS) - uint64_t HELPER(ldgm)(CPUARMState *env, uint64_t ptr) { int mmu_idx = cpu_mmu_index(env, false); uintptr_t ra = GETPC(); + int gm_bs = env_archcpu(env)->gm_blocksize; + int gm_bs_bytes = 4 << gm_bs; void *tag_mem; - ptr = QEMU_ALIGN_DOWN(ptr, LDGM_STGM_SIZE); + ptr = QEMU_ALIGN_DOWN(ptr, gm_bs_bytes); /* Trap if accessing an invalid page. */ tag_mem = allocation_tag_mem(env, mmu_idx, ptr, MMU_DATA_LOAD, - LDGM_STGM_SIZE, MMU_DATA_LOAD, - LDGM_STGM_SIZE / (2 * TAG_GRANULE), ra); + gm_bs_bytes, MMU_DATA_LOAD, + gm_bs_bytes / (2 * TAG_GRANULE), ra); /* The tag is squashed to zero if the page does not support tags. */ if (!tag_mem) { return 0; } - QEMU_BUILD_BUG_ON(GMID_EL1_BS != 6); /* - * We are loading 64-bits worth of tags. The ordering of elements - * within the word corresponds to a 64-bit little-endian operation. + * The ordering of elements within the word corresponds to + * a little-endian operation. */ - return ldq_le_p(tag_mem); + switch (gm_bs) { + case 6: + /* 256 bytes -> 16 tags -> 64 result bits */ + return ldq_le_p(tag_mem); + default: + /* cpu configured with unsupported gm blocksize. */ + g_assert_not_reached(); + } } void HELPER(stgm)(CPUARMState *env, uint64_t ptr, uint64_t val) { int mmu_idx = cpu_mmu_index(env, false); uintptr_t ra = GETPC(); + int gm_bs = env_archcpu(env)->gm_blocksize; + int gm_bs_bytes = 4 << gm_bs; void *tag_mem; - ptr = QEMU_ALIGN_DOWN(ptr, LDGM_STGM_SIZE); + ptr = QEMU_ALIGN_DOWN(ptr, gm_bs_bytes); /* Trap if accessing an invalid page. */ tag_mem = allocation_tag_mem(env, mmu_idx, ptr, MMU_DATA_STORE, - LDGM_STGM_SIZE, MMU_DATA_LOAD, - LDGM_STGM_SIZE / (2 * TAG_GRANULE), ra); + gm_bs_bytes, MMU_DATA_LOAD, + gm_bs_bytes / (2 * TAG_GRANULE), ra); /* * Tag store only happens if the page support tags, @@ -XXX,XX +XXX,XX @@ void HELPER(stgm)(CPUARMState *env, uint64_t ptr, uint64_t val) return; } - QEMU_BUILD_BUG_ON(GMID_EL1_BS != 6); /* - * We are storing 64-bits worth of tags. The ordering of elements - * within the word corresponds to a 64-bit little-endian operation. + * The ordering of elements within the word corresponds to + * a little-endian operation. */ - stq_le_p(tag_mem, val); + switch (gm_bs) { + case 6: + stq_le_p(tag_mem, val); + break; + default: + /* cpu configured with unsupported gm blocksize. */ + g_assert_not_reached(); + } } void HELPER(stzgm_tags)(CPUARMState *env, uint64_t ptr, uint64_t val) diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ static bool trans_STGM(DisasContext *s, arg_ldst_tag *a) gen_helper_stgm(cpu_env, addr, tcg_rt); } else { MMUAccessType acc = MMU_DATA_STORE; - int size = 4 << GMID_EL1_BS; + int size = 4 << s->gm_blocksize; clean_addr = clean_data_tbi(s, addr); tcg_gen_andi_i64(clean_addr, clean_addr, -size); @@ -XXX,XX +XXX,XX @@ static bool trans_LDGM(DisasContext *s, arg_ldst_tag *a) gen_helper_ldgm(tcg_rt, cpu_env, addr); } else { MMUAccessType acc = MMU_DATA_LOAD; - int size = 4 << GMID_EL1_BS; + int size = 4 << s->gm_blocksize; clean_addr = clean_data_tbi(s, addr); tcg_gen_andi_i64(clean_addr, clean_addr, -size); @@ -XXX,XX +XXX,XX @@ static void aarch64_tr_init_disas_context(DisasContextBase *dcbase, dc->cp_regs = arm_cpu->cp_regs; dc->features = env->features; dc->dcz_blocksize = arm_cpu->dcz_blocksize; + dc->gm_blocksize = arm_cpu->gm_blocksize; #ifdef CONFIG_USER_ONLY /* In sve_probe_page, we assume TBI is enabled. */ -- 2.34.1 From: Richard Henderson <richard.henderson@linaro.org> Support all of the easy GM block sizes. Use direct memory operations, since the pointers are aligned. While BS=2 (16 bytes, 1 tag) is a legal setting, that requires an atomic store of one nibble. This is not difficult, but there is also no point in supporting it until required. Note that cortex-a710 sets GM blocksize to match its cacheline size of 64 bytes. I expect many implementations will also match the cacheline, which makes 16 bytes very unlikely. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20230811214031.171020-4-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.c | 18 +++++++++--- target/arm/tcg/mte_helper.c | 56 +++++++++++++++++++++++++++++++------ 2 files changed, 62 insertions(+), 12 deletions(-) diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) ID_PFR1, VIRTUALIZATION, 0); } + if (cpu_isar_feature(aa64_mte, cpu)) { + /* + * The architectural range of GM blocksize is 2-6, however qemu + * doesn't support blocksize of 2 (see HELPER(ldgm)). + */ + if (tcg_enabled()) { + assert(cpu->gm_blocksize >= 3 && cpu->gm_blocksize <= 6); + } + #ifndef CONFIG_USER_ONLY - if (cpu->tag_memory == NULL && cpu_isar_feature(aa64_mte, cpu)) { /* * Disable the MTE feature bits if we do not have tag-memory * provided by the machine. */ - cpu->isar.id_aa64pfr1 = - FIELD_DP64(cpu->isar.id_aa64pfr1, ID_AA64PFR1, MTE, 0); - } + if (cpu->tag_memory == NULL) { + cpu->isar.id_aa64pfr1 = + FIELD_DP64(cpu->isar.id_aa64pfr1, ID_AA64PFR1, MTE, 0); + } #endif + } if (tcg_enabled()) { /* diff --git a/target/arm/tcg/mte_helper.c b/target/arm/tcg/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/mte_helper.c +++ b/target/arm/tcg/mte_helper.c @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(ldgm)(CPUARMState *env, uint64_t ptr) int gm_bs = env_archcpu(env)->gm_blocksize; int gm_bs_bytes = 4 << gm_bs; void *tag_mem; + uint64_t ret; + int shift; ptr = QEMU_ALIGN_DOWN(ptr, gm_bs_bytes); @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(ldgm)(CPUARMState *env, uint64_t ptr) /* * The ordering of elements within the word corresponds to - * a little-endian operation. + * a little-endian operation. Computation of shift comes from + * + * index = address<LOG2_TAG_GRANULE+3:LOG2_TAG_GRANULE> + * data<index*4+3:index*4> = tag + * + * Because of the alignment of ptr above, BS=6 has shift=0. + * All memory operations are aligned. Defer support for BS=2, + * requiring insertion or extraction of a nibble, until we + * support a cpu that requires it. */ switch (gm_bs) { + case 3: + /* 32 bytes -> 2 tags -> 8 result bits */ + ret = *(uint8_t *)tag_mem; + break; + case 4: + /* 64 bytes -> 4 tags -> 16 result bits */ + ret = cpu_to_le16(*(uint16_t *)tag_mem); + break; + case 5: + /* 128 bytes -> 8 tags -> 32 result bits */ + ret = cpu_to_le32(*(uint32_t *)tag_mem); + break; case 6: /* 256 bytes -> 16 tags -> 64 result bits */ - return ldq_le_p(tag_mem); + return cpu_to_le64(*(uint64_t *)tag_mem); default: - /* cpu configured with unsupported gm blocksize. */ + /* + * CPU configured with unsupported/invalid gm blocksize. + * This is detected early in arm_cpu_realizefn. + */ g_assert_not_reached(); } + shift = extract64(ptr, LOG2_TAG_GRANULE, 4) * 4; + return ret << shift; } void HELPER(stgm)(CPUARMState *env, uint64_t ptr, uint64_t val) @@ -XXX,XX +XXX,XX @@ void HELPER(stgm)(CPUARMState *env, uint64_t ptr, uint64_t val) int gm_bs = env_archcpu(env)->gm_blocksize; int gm_bs_bytes = 4 << gm_bs; void *tag_mem; + int shift; ptr = QEMU_ALIGN_DOWN(ptr, gm_bs_bytes); @@ -XXX,XX +XXX,XX @@ void HELPER(stgm)(CPUARMState *env, uint64_t ptr, uint64_t val) return; } - /* - * The ordering of elements within the word corresponds to - * a little-endian operation. - */ + /* See LDGM for comments on BS and on shift. */ + shift = extract64(ptr, LOG2_TAG_GRANULE, 4) * 4; + val >>= shift; switch (gm_bs) { + case 3: + /* 32 bytes -> 2 tags -> 8 result bits */ + *(uint8_t *)tag_mem = val; + break; + case 4: + /* 64 bytes -> 4 tags -> 16 result bits */ + *(uint16_t *)tag_mem = cpu_to_le16(val); + break; + case 5: + /* 128 bytes -> 8 tags -> 32 result bits */ + *(uint32_t *)tag_mem = cpu_to_le32(val); + break; case 6: - stq_le_p(tag_mem, val); + /* 256 bytes -> 16 tags -> 64 result bits */ + *(uint64_t *)tag_mem = cpu_to_le64(val); break; default: /* cpu configured with unsupported gm blocksize. */ -- 2.34.1 From: Richard Henderson <richard.henderson@linaro.org> When the cpu support MTE, but the system does not, reduce cpu support to user instructions at EL0 instead of completely disabling MTE. If we encounter a cpu implementation which does something else, we can revisit this setting. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20230811214031.171020-5-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) #ifndef CONFIG_USER_ONLY /* - * Disable the MTE feature bits if we do not have tag-memory - * provided by the machine. + * If we do not have tag-memory provided by the machine, + * reduce MTE support to instructions enabled at EL0. + * This matches Cortex-A710 BROADCASTMTE input being LOW. */ if (cpu->tag_memory == NULL) { cpu->isar.id_aa64pfr1 = - FIELD_DP64(cpu->isar.id_aa64pfr1, ID_AA64PFR1, MTE, 0); + FIELD_DP64(cpu->isar.id_aa64pfr1, ID_AA64PFR1, MTE, 1); } #endif } -- 2.34.1 From: Richard Henderson <richard.henderson@linaro.org> Do not hard-code the constants for Neoverse V1. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20230811214031.171020-6-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/tcg/cpu64.c | 48 ++++++++++++++++++++++++++++-------------- 1 file changed, 32 insertions(+), 16 deletions(-) diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu64.c +++ b/target/arm/tcg/cpu64.c @@ -XXX,XX +XXX,XX @@ #include "qemu/module.h" #include "qapi/visitor.h" #include "hw/qdev-properties.h" +#include "qemu/units.h" #include "internals.h" #include "cpregs.h" +static uint64_t make_ccsidr64(unsigned assoc, unsigned linesize, + unsigned cachesize) +{ + unsigned lg_linesize = ctz32(linesize); + unsigned sets; + + /* + * The 64-bit CCSIDR_EL1 format is: + * [55:32] number of sets - 1 + * [23:3] associativity - 1 + * [2:0] log2(linesize) - 4 + * so 0 == 16 bytes, 1 == 32 bytes, 2 == 64 bytes, etc + */ + assert(assoc != 0); + assert(is_power_of_2(linesize)); + assert(lg_linesize >= 4 && lg_linesize <= 7 + 4); + + /* sets * associativity * linesize == cachesize. */ + sets = cachesize / (assoc * linesize); + assert(cachesize % (assoc * linesize) == 0); + + return ((uint64_t)(sets - 1) << 32) + | ((assoc - 1) << 3) + | (lg_linesize - 4); +} + static void aarch64_a35_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); @@ -XXX,XX +XXX,XX @@ static void aarch64_neoverse_v1_initfn(Object *obj) * The Neoverse-V1 r1p2 TRM lists 32-bit format CCSIDR_EL1 values, * but also says it implements CCIDX, which means they should be * 64-bit format. So we here use values which are based on the textual - * information in chapter 2 of the TRM (and on the fact that - * sets * associativity * linesize == cachesize). - * - * The 64-bit CCSIDR_EL1 format is: - * [55:32] number of sets - 1 - * [23:3] associativity - 1 - * [2:0] log2(linesize) - 4 - * so 0 == 16 bytes, 1 == 32 bytes, 2 == 64 bytes, etc - * - * L1: 4-way set associative 64-byte line size, total size 64K, - * so sets is 256. + * information in chapter 2 of the TRM: * + * L1: 4-way set associative 64-byte line size, total size 64K. * L2: 8-way set associative, 64 byte line size, either 512K or 1MB. - * We pick 1MB, so this has 2048 sets. - * * L3: No L3 (this matches the CLIDR_EL1 value). */ - cpu->ccsidr[0] = 0x000000ff0000001aull; /* 64KB L1 dcache */ - cpu->ccsidr[1] = 0x000000ff0000001aull; /* 64KB L1 icache */ - cpu->ccsidr[2] = 0x000007ff0000003aull; /* 1MB L2 cache */ + cpu->ccsidr[0] = make_ccsidr64(4, 64, 64 * KiB); /* L1 dcache */ + cpu->ccsidr[1] = cpu->ccsidr[0]; /* L1 icache */ + cpu->ccsidr[2] = make_ccsidr64(8, 64, 1 * MiB); /* L2 cache */ /* From 3.2.115 SCTLR_EL3 */ cpu->reset_sctlr = 0x30c50838; -- 2.34.1 From: Richard Henderson <richard.henderson@linaro.org> Access to many of the special registers is enabled or disabled by ACTLR_EL[23], which we implement as constant 0, which means that all writes outside EL3 should trap. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20230811214031.171020-7-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpregs.h | 2 ++ target/arm/helper.c | 4 ++-- target/arm/tcg/cpu64.c | 46 +++++++++++++++++++++++++++++++++--------- 3 files changed, 41 insertions(+), 11 deletions(-) diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpregs.h +++ b/target/arm/cpregs.h @@ -XXX,XX +XXX,XX @@ static inline void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu) { } void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu); #endif +CPAccessResult access_tvm_trvm(CPUARMState *, const ARMCPRegInfo *, bool); + #endif /* TARGET_ARM_CPREGS_H */ diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ static CPAccessResult access_tpm(CPUARMState *env, const ARMCPRegInfo *ri, } /* Check for traps from EL1 due to HCR_EL2.TVM and HCR_EL2.TRVM. */ -static CPAccessResult access_tvm_trvm(CPUARMState *env, const ARMCPRegInfo *ri, - bool isread) +CPAccessResult access_tvm_trvm(CPUARMState *env, const ARMCPRegInfo *ri, + bool isread) { if (arm_current_el(env) == 1) { uint64_t trap = isread ? HCR_TRVM : HCR_TVM; diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu64.c +++ b/target/arm/tcg/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_a64fx_initfn(Object *obj) /* TODO: Add A64FX specific HPC extension registers */ } +static CPAccessResult access_actlr_w(CPUARMState *env, const ARMCPRegInfo *r, + bool read) +{ + if (!read) { + int el = arm_current_el(env); + + /* Because ACTLR_EL2 is constant 0, writes below EL2 trap to EL2. */ + if (el < 2 && arm_is_el2_enabled(env)) { + return CP_ACCESS_TRAP_EL2; + } + /* Because ACTLR_EL3 is constant 0, writes below EL3 trap to EL3. */ + if (el < 3 && arm_feature(env, ARM_FEATURE_EL3)) { + return CP_ACCESS_TRAP_EL3; + } + } + return CP_ACCESS_OK; +} + static const ARMCPRegInfo neoverse_n1_cp_reginfo[] = { { .name = "ATCR_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 7, .opc2 = 0, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0, + /* Traps and enables are the same as for TCR_EL1. */ + .accessfn = access_tvm_trvm, .fgt = FGT_TCR_EL1, }, { .name = "ATCR_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 15, .crm = 7, .opc2 = 0, .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo neoverse_n1_cp_reginfo[] = { .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, { .name = "CPUACTLR_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 1, .opc2 = 0, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0, + .accessfn = access_actlr_w }, { .name = "CPUACTLR2_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 1, .opc2 = 1, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0, + .accessfn = access_actlr_w }, { .name = "CPUACTLR3_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 1, .opc2 = 2, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0, + .accessfn = access_actlr_w }, /* * Report CPUCFR_EL1.SCU as 1, as we do not implement the DSU * (and in particular its system registers). @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo neoverse_n1_cp_reginfo[] = { .access = PL1_R, .type = ARM_CP_CONST, .resetvalue = 4 }, { .name = "CPUECTLR_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 1, .opc2 = 4, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0x961563010 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0x961563010, + .accessfn = access_actlr_w }, { .name = "CPUPCR_EL3", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 6, .crn = 15, .crm = 8, .opc2 = 1, .access = PL3_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo neoverse_n1_cp_reginfo[] = { .access = PL3_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, { .name = "CPUPWRCTLR_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 2, .opc2 = 7, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0, + .accessfn = access_actlr_w }, { .name = "ERXPFGCDN_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 2, .opc2 = 2, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0, + .accessfn = access_actlr_w }, { .name = "ERXPFGCTL_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 2, .opc2 = 1, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0, + .accessfn = access_actlr_w }, { .name = "ERXPFGF_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 2, .opc2 = 0, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0, + .accessfn = access_actlr_w }, }; static void define_neoverse_n1_cp_reginfo(ARMCPU *cpu) -- 2.34.1 From: Richard Henderson <richard.henderson@linaro.org> There is only one additional EL1 register modeled, which also needs to use access_actlr_w. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20230811214031.171020-8-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/tcg/cpu64.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu64.c +++ b/target/arm/tcg/cpu64.c @@ -XXX,XX +XXX,XX @@ static void define_neoverse_n1_cp_reginfo(ARMCPU *cpu) static const ARMCPRegInfo neoverse_v1_cp_reginfo[] = { { .name = "CPUECTLR2_EL1", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 0, .crn = 15, .crm = 1, .opc2 = 5, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0, + .accessfn = access_actlr_w }, { .name = "CPUPPMCR_EL3", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 6, .crn = 15, .crm = 2, .opc2 = 0, .access = PL3_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, -- 2.34.1 From: Richard Henderson <richard.henderson@linaro.org> Like FEAT_TRF (Self-hosted Trace Extension), suppress tracing external to the cpu, which is out of scope for QEMU. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20230811214031.171020-10-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) /* FEAT_SPE (Statistical Profiling Extension) */ cpu->isar.id_aa64dfr0 = FIELD_DP64(cpu->isar.id_aa64dfr0, ID_AA64DFR0, PMSVER, 0); + /* FEAT_TRBE (Trace Buffer Extension) */ + cpu->isar.id_aa64dfr0 = + FIELD_DP64(cpu->isar.id_aa64dfr0, ID_AA64DFR0, TRACEBUFFER, 0); /* FEAT_TRF (Self-hosted Trace Extension) */ cpu->isar.id_aa64dfr0 = FIELD_DP64(cpu->isar.id_aa64dfr0, ID_AA64DFR0, TRACEFILT, 0); -- 2.34.1 From: Richard Henderson <richard.henderson@linaro.org> This feature allows the operating system to set TCR_ELx.HWU* to allow the implementation to use the PBHA bits from the block and page descriptors for for IMPLEMENTATION DEFINED purposes. Since QEMU has no need to use these bits, we may simply ignore them. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20230811214031.171020-11-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/tcg/cpu32.c | 2 +- target/arm/tcg/cpu64.c | 2 +- 3 files changed, 3 insertions(+), 2 deletions(-) diff --git a/docs/system/arm/emulation.rst b/docs/system/arm/emulation.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/emulation.rst +++ b/docs/system/arm/emulation.rst @@ -XXX,XX +XXX,XX @@ the following architecture extensions: - FEAT_HAFDBS (Hardware management of the access flag and dirty bit state) - FEAT_HCX (Support for the HCRX_EL2 register) - FEAT_HPDS (Hierarchical permission disables) +- FEAT_HPDS2 (Translation table page-based hardware attributes) - FEAT_I8MM (AArch64 Int8 matrix multiplication instructions) - FEAT_IDST (ID space trap handling) - FEAT_IESB (Implicit error synchronization event) diff --git a/target/arm/tcg/cpu32.c b/target/arm/tcg/cpu32.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu32.c +++ b/target/arm/tcg/cpu32.c @@ -XXX,XX +XXX,XX @@ void aa32_max_features(ARMCPU *cpu) cpu->isar.id_mmfr3 = t; t = cpu->isar.id_mmfr4; - t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* FEAT_AA32HPD */ + t = FIELD_DP32(t, ID_MMFR4, HPDS, 2); /* FEAT_HPDS2 */ t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* FEAT_TTCNP */ t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* FEAT_XNX */ diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu64.c +++ b/target/arm/tcg/cpu64.c @@ -XXX,XX +XXX,XX @@ void aarch64_max_tcg_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64MMFR1, HAFDBS, 2); /* FEAT_HAFDBS */ t = FIELD_DP64(t, ID_AA64MMFR1, VMIDBITS, 2); /* FEAT_VMID16 */ t = FIELD_DP64(t, ID_AA64MMFR1, VH, 1); /* FEAT_VHE */ - t = FIELD_DP64(t, ID_AA64MMFR1, HPDS, 1); /* FEAT_HPDS */ + t = FIELD_DP64(t, ID_AA64MMFR1, HPDS, 2); /* FEAT_HPDS2 */ t = FIELD_DP64(t, ID_AA64MMFR1, LO, 1); /* FEAT_LOR */ t = FIELD_DP64(t, ID_AA64MMFR1, PAN, 3); /* FEAT_PAN3 */ t = FIELD_DP64(t, ID_AA64MMFR1, XNX, 1); /* FEAT_XNX */ -- 2.34.1 From: Alex Bennée <alex.bennee@linaro.org> This is a mandatory feature for Armv8.1 architectures but we don't state the feature clearly in our emulation list. Also include FEAT_CRC32 comment in aarch64_max_tcg_initfn for ease of grepping. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20230824075406.1515566-1-alex.bennee@linaro.org Cc: qemu-stable@nongnu.org Message-Id: <20230222110104.3996971-1-alex.bennee@linaro.org> [PMM: pluralize 'instructions' in docs] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/tcg/cpu64.c | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/docs/system/arm/emulation.rst b/docs/system/arm/emulation.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/emulation.rst +++ b/docs/system/arm/emulation.rst @@ -XXX,XX +XXX,XX @@ the following architecture extensions: - FEAT_BBM at level 2 (Translation table break-before-make levels) - FEAT_BF16 (AArch64 BFloat16 instructions) - FEAT_BTI (Branch Target Identification) +- FEAT_CRC32 (CRC32 instructions) - FEAT_CSV2 (Cache speculation variant 2) - FEAT_CSV2_1p1 (Cache speculation variant 2, version 1.1) - FEAT_CSV2_1p2 (Cache speculation variant 2, version 1.2) diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu64.c +++ b/target/arm/tcg/cpu64.c @@ -XXX,XX +XXX,XX @@ void aarch64_max_tcg_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64ISAR0, AES, 2); /* FEAT_PMULL */ t = FIELD_DP64(t, ID_AA64ISAR0, SHA1, 1); /* FEAT_SHA1 */ t = FIELD_DP64(t, ID_AA64ISAR0, SHA2, 2); /* FEAT_SHA512 */ - t = FIELD_DP64(t, ID_AA64ISAR0, CRC32, 1); + t = FIELD_DP64(t, ID_AA64ISAR0, CRC32, 1); /* FEAT_CRC32 */ t = FIELD_DP64(t, ID_AA64ISAR0, ATOMIC, 2); /* FEAT_LSE */ t = FIELD_DP64(t, ID_AA64ISAR0, RDM, 1); /* FEAT_RDM */ t = FIELD_DP64(t, ID_AA64ISAR0, SHA3, 1); /* FEAT_SHA3 */ -- 2.34.1 From: Jean-Christophe Dubois <jcd@tribudubois.net> i.MX7 IOMUX GPR device is not equivalent to i.MX6UL IOMUXC GPR device. In particular, register 22 is not present on i.MX6UL and this is actualy The only register that is really emulated in the i.MX7 IOMUX GPR device. Note: The i.MX6UL code is actually also implementing the IOMUX GPR device as an unimplemented device at the same bus adress and the 2 instantiations were actualy colliding. So we go back to the unimplemented device for now. Signed-off-by: Jean-Christophe Dubois <jcd@tribudubois.net> Message-id: 48681bf51ee97646479bb261bee19abebbc8074e.1692964892.git.jcd@tribudubois.net Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/fsl-imx6ul.h | 2 -- hw/arm/fsl-imx6ul.c | 11 ----------- 2 files changed, 13 deletions(-) diff --git a/include/hw/arm/fsl-imx6ul.h b/include/hw/arm/fsl-imx6ul.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/fsl-imx6ul.h +++ b/include/hw/arm/fsl-imx6ul.h @@ -XXX,XX +XXX,XX @@ #include "hw/misc/imx6ul_ccm.h" #include "hw/misc/imx6_src.h" #include "hw/misc/imx7_snvs.h" -#include "hw/misc/imx7_gpr.h" #include "hw/intc/imx_gpcv2.h" #include "hw/watchdog/wdt_imx2.h" #include "hw/gpio/imx_gpio.h" @@ -XXX,XX +XXX,XX @@ struct FslIMX6ULState { IMX6SRCState src; IMX7SNVSState snvs; IMXGPCv2State gpcv2; - IMX7GPRState gpr; IMXSPIState spi[FSL_IMX6UL_NUM_ECSPIS]; IMXI2CState i2c[FSL_IMX6UL_NUM_I2CS]; IMXSerialState uart[FSL_IMX6UL_NUM_UARTS]; diff --git a/hw/arm/fsl-imx6ul.c b/hw/arm/fsl-imx6ul.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/fsl-imx6ul.c +++ b/hw/arm/fsl-imx6ul.c @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_init(Object *obj) */ object_initialize_child(obj, "snvs", &s->snvs, TYPE_IMX7_SNVS); - /* - * GPR - */ - object_initialize_child(obj, "gpr", &s->gpr, TYPE_IMX7_GPR); - /* * GPIOs 1 to 5 */ @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) FSL_IMX6UL_WDOGn_IRQ[i])); } - /* - * GPR - */ - sysbus_realize(SYS_BUS_DEVICE(&s->gpr), &error_abort); - sysbus_mmio_map(SYS_BUS_DEVICE(&s->gpr), 0, FSL_IMX6UL_IOMUXC_GPR_ADDR); - /* * SDMA */ -- 2.34.1 From: Jean-Christophe Dubois <jcd@tribudubois.net> * Add Addr and size definition for most i.MX6UL devices in i.MX6UL header file. * Use those newly defined named constants whenever possible. * Standardize the way we init a familly of unimplemented devices - SAI - PWM - CAN * Add/rework few comments Signed-off-by: Jean-Christophe Dubois <jcd@tribudubois.net> Message-id: d579043fbd4e4b490370783fda43fc02c8e9be75.1692964892.git.jcd@tribudubois.net Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/fsl-imx6ul.h | 156 +++++++++++++++++++++++++++++++----- hw/arm/fsl-imx6ul.c | 147 ++++++++++++++++++++++----------- 2 files changed, 232 insertions(+), 71 deletions(-) diff --git a/include/hw/arm/fsl-imx6ul.h b/include/hw/arm/fsl-imx6ul.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/fsl-imx6ul.h +++ b/include/hw/arm/fsl-imx6ul.h @@ -XXX,XX +XXX,XX @@ #include "exec/memory.h" #include "cpu.h" #include "qom/object.h" +#include "qemu/units.h" #define TYPE_FSL_IMX6UL "fsl-imx6ul" OBJECT_DECLARE_SIMPLE_TYPE(FslIMX6ULState, FSL_IMX6UL) @@ -XXX,XX +XXX,XX @@ enum FslIMX6ULConfiguration { FSL_IMX6UL_NUM_ADCS = 2, FSL_IMX6UL_NUM_USB_PHYS = 2, FSL_IMX6UL_NUM_USBS = 2, + FSL_IMX6UL_NUM_SAIS = 3, + FSL_IMX6UL_NUM_CANS = 2, + FSL_IMX6UL_NUM_PWMS = 4, }; struct FslIMX6ULState { @@ -XXX,XX +XXX,XX @@ struct FslIMX6ULState { enum FslIMX6ULMemoryMap { FSL_IMX6UL_MMDC_ADDR = 0x80000000, - FSL_IMX6UL_MMDC_SIZE = 2 * 1024 * 1024 * 1024UL, + FSL_IMX6UL_MMDC_SIZE = (2 * GiB), FSL_IMX6UL_QSPI1_MEM_ADDR = 0x60000000, - FSL_IMX6UL_EIM_ALIAS_ADDR = 0x58000000, - FSL_IMX6UL_EIM_CS_ADDR = 0x50000000, - FSL_IMX6UL_AES_ENCRYPT_ADDR = 0x10000000, - FSL_IMX6UL_QSPI1_RX_ADDR = 0x0C000000, + FSL_IMX6UL_QSPI1_MEM_SIZE = (256 * MiB), - /* AIPS-2 */ + FSL_IMX6UL_EIM_ALIAS_ADDR = 0x58000000, + FSL_IMX6UL_EIM_ALIAS_SIZE = (128 * MiB), + + FSL_IMX6UL_EIM_CS_ADDR = 0x50000000, + FSL_IMX6UL_EIM_CS_SIZE = (128 * MiB), + + FSL_IMX6UL_AES_ENCRYPT_ADDR = 0x10000000, + FSL_IMX6UL_AES_ENCRYPT_SIZE = (1 * MiB), + + FSL_IMX6UL_QSPI1_RX_ADDR = 0x0C000000, + FSL_IMX6UL_QSPI1_RX_SIZE = (32 * MiB), + + /* AIPS-2 Begin */ FSL_IMX6UL_UART6_ADDR = 0x021FC000, + FSL_IMX6UL_I2C4_ADDR = 0x021F8000, + FSL_IMX6UL_UART5_ADDR = 0x021F4000, FSL_IMX6UL_UART4_ADDR = 0x021F0000, FSL_IMX6UL_UART3_ADDR = 0x021EC000, FSL_IMX6UL_UART2_ADDR = 0x021E8000, + FSL_IMX6UL_WDOG3_ADDR = 0x021E4000, + FSL_IMX6UL_QSPI_ADDR = 0x021E0000, + FSL_IMX6UL_QSPI_SIZE = 0x500, + FSL_IMX6UL_SYS_CNT_CTRL_ADDR = 0x021DC000, + FSL_IMX6UL_SYS_CNT_CTRL_SIZE = (16 * KiB), + FSL_IMX6UL_SYS_CNT_CMP_ADDR = 0x021D8000, + FSL_IMX6UL_SYS_CNT_CMP_SIZE = (16 * KiB), + FSL_IMX6UL_SYS_CNT_RD_ADDR = 0x021D4000, + FSL_IMX6UL_SYS_CNT_RD_SIZE = (16 * KiB), + FSL_IMX6UL_TZASC_ADDR = 0x021D0000, + FSL_IMX6UL_TZASC_SIZE = (16 * KiB), + FSL_IMX6UL_PXP_ADDR = 0x021CC000, + FSL_IMX6UL_PXP_SIZE = (16 * KiB), + FSL_IMX6UL_LCDIF_ADDR = 0x021C8000, + FSL_IMX6UL_LCDIF_SIZE = 0x100, + FSL_IMX6UL_CSI_ADDR = 0x021C4000, + FSL_IMX6UL_CSI_SIZE = 0x100, + FSL_IMX6UL_CSU_ADDR = 0x021C0000, + FSL_IMX6UL_CSU_SIZE = (16 * KiB), + FSL_IMX6UL_OCOTP_CTRL_ADDR = 0x021BC000, + FSL_IMX6UL_OCOTP_CTRL_SIZE = (4 * KiB), + FSL_IMX6UL_EIM_ADDR = 0x021B8000, + FSL_IMX6UL_EIM_SIZE = 0x100, + FSL_IMX6UL_SIM2_ADDR = 0x021B4000, + FSL_IMX6UL_MMDC_CFG_ADDR = 0x021B0000, + FSL_IMX6UL_MMDC_CFG_SIZE = (4 * KiB), + FSL_IMX6UL_ROMCP_ADDR = 0x021AC000, + FSL_IMX6UL_ROMCP_SIZE = 0x300, + FSL_IMX6UL_I2C3_ADDR = 0x021A8000, FSL_IMX6UL_I2C2_ADDR = 0x021A4000, FSL_IMX6UL_I2C1_ADDR = 0x021A0000, + FSL_IMX6UL_ADC2_ADDR = 0x0219C000, FSL_IMX6UL_ADC1_ADDR = 0x02198000, + FSL_IMX6UL_ADCn_SIZE = 0x100, + FSL_IMX6UL_USDHC2_ADDR = 0x02194000, FSL_IMX6UL_USDHC1_ADDR = 0x02190000, - FSL_IMX6UL_SIM1_ADDR = 0x0218C000, - FSL_IMX6UL_ENET1_ADDR = 0x02188000, - FSL_IMX6UL_USBO2_USBMISC_ADDR = 0x02184800, - FSL_IMX6UL_USBO2_USB_ADDR = 0x02184000, - FSL_IMX6UL_USBO2_PL301_ADDR = 0x02180000, - FSL_IMX6UL_AIPS2_CFG_ADDR = 0x0217C000, - FSL_IMX6UL_CAAM_ADDR = 0x02140000, - FSL_IMX6UL_A7MPCORE_DAP_ADDR = 0x02100000, - /* AIPS-1 */ + FSL_IMX6UL_SIM1_ADDR = 0x0218C000, + FSL_IMX6UL_SIMn_SIZE = (16 * KiB), + + FSL_IMX6UL_ENET1_ADDR = 0x02188000, + + FSL_IMX6UL_USBO2_USBMISC_ADDR = 0x02184800, + FSL_IMX6UL_USBO2_USB1_ADDR = 0x02184000, + FSL_IMX6UL_USBO2_USB2_ADDR = 0x02184200, + + FSL_IMX6UL_USBO2_PL301_ADDR = 0x02180000, + FSL_IMX6UL_USBO2_PL301_SIZE = (16 * KiB), + + FSL_IMX6UL_AIPS2_CFG_ADDR = 0x0217C000, + FSL_IMX6UL_AIPS2_CFG_SIZE = 0x100, + + FSL_IMX6UL_CAAM_ADDR = 0x02140000, + FSL_IMX6UL_CAAM_SIZE = (16 * KiB), + + FSL_IMX6UL_A7MPCORE_DAP_ADDR = 0x02100000, + FSL_IMX6UL_A7MPCORE_DAP_SIZE = (4 * KiB), + /* AIPS-2 End */ + + /* AIPS-1 Begin */ FSL_IMX6UL_PWM8_ADDR = 0x020FC000, FSL_IMX6UL_PWM7_ADDR = 0x020F8000, FSL_IMX6UL_PWM6_ADDR = 0x020F4000, FSL_IMX6UL_PWM5_ADDR = 0x020F0000, + FSL_IMX6UL_SDMA_ADDR = 0x020EC000, + FSL_IMX6UL_SDMA_SIZE = 0x300, + FSL_IMX6UL_GPT2_ADDR = 0x020E8000, + FSL_IMX6UL_IOMUXC_GPR_ADDR = 0x020E4000, + FSL_IMX6UL_IOMUXC_GPR_SIZE = 0x40, + FSL_IMX6UL_IOMUXC_ADDR = 0x020E0000, + FSL_IMX6UL_IOMUXC_SIZE = 0x700, + FSL_IMX6UL_GPC_ADDR = 0x020DC000, + FSL_IMX6UL_SRC_ADDR = 0x020D8000, + FSL_IMX6UL_EPIT2_ADDR = 0x020D4000, FSL_IMX6UL_EPIT1_ADDR = 0x020D0000, + FSL_IMX6UL_SNVS_HP_ADDR = 0x020CC000, + FSL_IMX6UL_USBPHY2_ADDR = 0x020CA000, - FSL_IMX6UL_USBPHY2_SIZE = (4 * 1024), FSL_IMX6UL_USBPHY1_ADDR = 0x020C9000, - FSL_IMX6UL_USBPHY1_SIZE = (4 * 1024), + FSL_IMX6UL_ANALOG_ADDR = 0x020C8000, + FSL_IMX6UL_ANALOG_SIZE = 0x300, + FSL_IMX6UL_CCM_ADDR = 0x020C4000, + FSL_IMX6UL_WDOG2_ADDR = 0x020C0000, FSL_IMX6UL_WDOG1_ADDR = 0x020BC000, + FSL_IMX6UL_KPP_ADDR = 0x020B8000, + FSL_IMX6UL_KPP_SIZE = 0x10, + FSL_IMX6UL_ENET2_ADDR = 0x020B4000, + FSL_IMX6UL_SNVS_LP_ADDR = 0x020B0000, + FSL_IMX6UL_SNVS_LP_SIZE = (16 * KiB), + FSL_IMX6UL_GPIO5_ADDR = 0x020AC000, FSL_IMX6UL_GPIO4_ADDR = 0x020A8000, FSL_IMX6UL_GPIO3_ADDR = 0x020A4000, FSL_IMX6UL_GPIO2_ADDR = 0x020A0000, FSL_IMX6UL_GPIO1_ADDR = 0x0209C000, + FSL_IMX6UL_GPT1_ADDR = 0x02098000, + FSL_IMX6UL_CAN2_ADDR = 0x02094000, FSL_IMX6UL_CAN1_ADDR = 0x02090000, + FSL_IMX6UL_CANn_SIZE = (4 * KiB), + FSL_IMX6UL_PWM4_ADDR = 0x0208C000, FSL_IMX6UL_PWM3_ADDR = 0x02088000, FSL_IMX6UL_PWM2_ADDR = 0x02084000, FSL_IMX6UL_PWM1_ADDR = 0x02080000, + FSL_IMX6UL_PWMn_SIZE = 0x20, + FSL_IMX6UL_AIPS1_CFG_ADDR = 0x0207C000, + FSL_IMX6UL_AIPS1_CFG_SIZE = (16 * KiB), + FSL_IMX6UL_BEE_ADDR = 0x02044000, + FSL_IMX6UL_BEE_SIZE = (16 * KiB), + FSL_IMX6UL_TOUCH_CTRL_ADDR = 0x02040000, + FSL_IMX6UL_TOUCH_CTRL_SIZE = 0x100, + FSL_IMX6UL_SPBA_ADDR = 0x0203C000, + FSL_IMX6UL_SPBA_SIZE = 0x100, + FSL_IMX6UL_ASRC_ADDR = 0x02034000, + FSL_IMX6UL_ASRC_SIZE = 0x100, + FSL_IMX6UL_SAI3_ADDR = 0x02030000, FSL_IMX6UL_SAI2_ADDR = 0x0202C000, FSL_IMX6UL_SAI1_ADDR = 0x02028000, + FSL_IMX6UL_SAIn_SIZE = 0x200, + FSL_IMX6UL_UART8_ADDR = 0x02024000, FSL_IMX6UL_UART1_ADDR = 0x02020000, FSL_IMX6UL_UART7_ADDR = 0x02018000, + FSL_IMX6UL_ECSPI4_ADDR = 0x02014000, FSL_IMX6UL_ECSPI3_ADDR = 0x02010000, FSL_IMX6UL_ECSPI2_ADDR = 0x0200C000, FSL_IMX6UL_ECSPI1_ADDR = 0x02008000, + FSL_IMX6UL_SPDIF_ADDR = 0x02004000, + FSL_IMX6UL_SPDIF_SIZE = 0x100, + /* AIPS-1 End */ + + FSL_IMX6UL_BCH_ADDR = 0x01808000, + FSL_IMX6UL_BCH_SIZE = 0x200, + + FSL_IMX6UL_GPMI_ADDR = 0x01806000, + FSL_IMX6UL_GPMI_SIZE = 0x200, FSL_IMX6UL_APBH_DMA_ADDR = 0x01804000, - FSL_IMX6UL_APBH_DMA_SIZE = (32 * 1024), + FSL_IMX6UL_APBH_DMA_SIZE = (4 * KiB), FSL_IMX6UL_A7MPCORE_ADDR = 0x00A00000, FSL_IMX6UL_OCRAM_ALIAS_ADDR = 0x00920000, - FSL_IMX6UL_OCRAM_ALIAS_SIZE = 0x00060000, + FSL_IMX6UL_OCRAM_ALIAS_SIZE = (384 * KiB), + FSL_IMX6UL_OCRAM_MEM_ADDR = 0x00900000, - FSL_IMX6UL_OCRAM_MEM_SIZE = 0x00020000, + FSL_IMX6UL_OCRAM_MEM_SIZE = (128 * KiB), + FSL_IMX6UL_CAAM_MEM_ADDR = 0x00100000, - FSL_IMX6UL_CAAM_MEM_SIZE = 0x00008000, + FSL_IMX6UL_CAAM_MEM_SIZE = (32 * KiB), + FSL_IMX6UL_ROM_ADDR = 0x00000000, - FSL_IMX6UL_ROM_SIZE = 0x00018000, + FSL_IMX6UL_ROM_SIZE = (96 * KiB), }; enum FslIMX6ULIRQs { diff --git a/hw/arm/fsl-imx6ul.c b/hw/arm/fsl-imx6ul.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/fsl-imx6ul.c +++ b/hw/arm/fsl-imx6ul.c @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_init(Object *obj) object_initialize_child(obj, "snvs", &s->snvs, TYPE_IMX7_SNVS); /* - * GPIOs 1 to 5 + * GPIOs */ for (i = 0; i < FSL_IMX6UL_NUM_GPIOS; i++) { snprintf(name, NAME_SIZE, "gpio%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_init(Object *obj) } /* - * GPT 1, 2 + * GPTs */ for (i = 0; i < FSL_IMX6UL_NUM_GPTS; i++) { snprintf(name, NAME_SIZE, "gpt%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_init(Object *obj) } /* - * EPIT 1, 2 + * EPITs */ for (i = 0; i < FSL_IMX6UL_NUM_EPITS; i++) { snprintf(name, NAME_SIZE, "epit%d", i + 1); @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_init(Object *obj) } /* - * eCSPI + * eCSPIs */ for (i = 0; i < FSL_IMX6UL_NUM_ECSPIS; i++) { snprintf(name, NAME_SIZE, "spi%d", i + 1); @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_init(Object *obj) } /* - * I2C + * I2Cs */ for (i = 0; i < FSL_IMX6UL_NUM_I2CS; i++) { snprintf(name, NAME_SIZE, "i2c%d", i + 1); @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_init(Object *obj) } /* - * UART + * UARTs */ for (i = 0; i < FSL_IMX6UL_NUM_UARTS; i++) { snprintf(name, NAME_SIZE, "uart%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_init(Object *obj) } /* - * Ethernet + * Ethernets */ for (i = 0; i < FSL_IMX6UL_NUM_ETHS; i++) { snprintf(name, NAME_SIZE, "eth%d", i); object_initialize_child(obj, name, &s->eth[i], TYPE_IMX_ENET); } - /* USB */ + /* + * USB PHYs + */ for (i = 0; i < FSL_IMX6UL_NUM_USB_PHYS; i++) { snprintf(name, NAME_SIZE, "usbphy%d", i); object_initialize_child(obj, name, &s->usbphy[i], TYPE_IMX_USBPHY); } + + /* + * USBs + */ for (i = 0; i < FSL_IMX6UL_NUM_USBS; i++) { snprintf(name, NAME_SIZE, "usb%d", i); object_initialize_child(obj, name, &s->usb[i], TYPE_CHIPIDEA); } /* - * SDHCI + * SDHCIs */ for (i = 0; i < FSL_IMX6UL_NUM_USDHCS; i++) { snprintf(name, NAME_SIZE, "usdhc%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_init(Object *obj) } /* - * Watchdog + * Watchdogs */ for (i = 0; i < FSL_IMX6UL_NUM_WDTS; i++) { snprintf(name, NAME_SIZE, "wdt%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) * A7MPCORE DAP */ create_unimplemented_device("a7mpcore-dap", FSL_IMX6UL_A7MPCORE_DAP_ADDR, - 0x100000); + FSL_IMX6UL_A7MPCORE_DAP_SIZE); /* - * GPT 1, 2 + * GPTs */ for (i = 0; i < FSL_IMX6UL_NUM_GPTS; i++) { static const hwaddr FSL_IMX6UL_GPTn_ADDR[FSL_IMX6UL_NUM_GPTS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) } /* - * EPIT 1, 2 + * EPITs */ for (i = 0; i < FSL_IMX6UL_NUM_EPITS; i++) { static const hwaddr FSL_IMX6UL_EPITn_ADDR[FSL_IMX6UL_NUM_EPITS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) } /* - * GPIO + * GPIOs */ for (i = 0; i < FSL_IMX6UL_NUM_GPIOS; i++) { static const hwaddr FSL_IMX6UL_GPIOn_ADDR[FSL_IMX6UL_NUM_GPIOS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) } /* - * IOMUXC and IOMUXC_GPR + * IOMUXC */ - for (i = 0; i < 1; i++) { - static const hwaddr FSL_IMX6UL_IOMUXCn_ADDR[FSL_IMX6UL_NUM_IOMUXCS] = { - FSL_IMX6UL_IOMUXC_ADDR, - FSL_IMX6UL_IOMUXC_GPR_ADDR, - }; - - snprintf(name, NAME_SIZE, "iomuxc%d", i); - create_unimplemented_device(name, FSL_IMX6UL_IOMUXCn_ADDR[i], 0x4000); - } + create_unimplemented_device("iomuxc", FSL_IMX6UL_IOMUXC_ADDR, + FSL_IMX6UL_IOMUXC_SIZE); + create_unimplemented_device("iomuxc_gpr", FSL_IMX6UL_IOMUXC_GPR_ADDR, + FSL_IMX6UL_IOMUXC_GPR_SIZE); /* * CCM @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) sysbus_realize(SYS_BUS_DEVICE(&s->gpcv2), &error_abort); sysbus_mmio_map(SYS_BUS_DEVICE(&s->gpcv2), 0, FSL_IMX6UL_GPC_ADDR); - /* Initialize all ECSPI */ + /* + * ECSPIs + */ for (i = 0; i < FSL_IMX6UL_NUM_ECSPIS; i++) { static const hwaddr FSL_IMX6UL_SPIn_ADDR[FSL_IMX6UL_NUM_ECSPIS] = { FSL_IMX6UL_ECSPI1_ADDR, @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) } /* - * I2C + * I2Cs */ for (i = 0; i < FSL_IMX6UL_NUM_I2CS; i++) { static const hwaddr FSL_IMX6UL_I2Cn_ADDR[FSL_IMX6UL_NUM_I2CS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) } /* - * UART + * UARTs */ for (i = 0; i < FSL_IMX6UL_NUM_UARTS; i++) { static const hwaddr FSL_IMX6UL_UARTn_ADDR[FSL_IMX6UL_NUM_UARTS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) } /* - * Ethernet + * Ethernets * * We must use two loops since phy_connected affects the other interface * and we have to set all properties before calling sysbus_realize(). @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) FSL_IMX6UL_ENETn_TIMER_IRQ[i])); } - /* USB */ + /* + * USB PHYs + */ for (i = 0; i < FSL_IMX6UL_NUM_USB_PHYS; i++) { + static const hwaddr + FSL_IMX6UL_USB_PHYn_ADDR[FSL_IMX6UL_NUM_USB_PHYS] = { + FSL_IMX6UL_USBPHY1_ADDR, + FSL_IMX6UL_USBPHY2_ADDR, + }; + sysbus_realize(SYS_BUS_DEVICE(&s->usbphy[i]), &error_abort); sysbus_mmio_map(SYS_BUS_DEVICE(&s->usbphy[i]), 0, - FSL_IMX6UL_USBPHY1_ADDR + i * 0x1000); + FSL_IMX6UL_USB_PHYn_ADDR[i]); } + /* + * USBs + */ for (i = 0; i < FSL_IMX6UL_NUM_USBS; i++) { + static const hwaddr FSL_IMX6UL_USB02_USBn_ADDR[FSL_IMX6UL_NUM_USBS] = { + FSL_IMX6UL_USBO2_USB1_ADDR, + FSL_IMX6UL_USBO2_USB2_ADDR, + }; + static const int FSL_IMX6UL_USBn_IRQ[] = { FSL_IMX6UL_USB1_IRQ, FSL_IMX6UL_USB2_IRQ, }; + sysbus_realize(SYS_BUS_DEVICE(&s->usb[i]), &error_abort); sysbus_mmio_map(SYS_BUS_DEVICE(&s->usb[i]), 0, - FSL_IMX6UL_USBO2_USB_ADDR + i * 0x200); + FSL_IMX6UL_USB02_USBn_ADDR[i]); sysbus_connect_irq(SYS_BUS_DEVICE(&s->usb[i]), 0, qdev_get_gpio_in(DEVICE(&s->a7mpcore), FSL_IMX6UL_USBn_IRQ[i])); } /* - * USDHC + * USDHCs */ for (i = 0; i < FSL_IMX6UL_NUM_USDHCS; i++) { static const hwaddr FSL_IMX6UL_USDHCn_ADDR[FSL_IMX6UL_NUM_USDHCS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) sysbus_mmio_map(SYS_BUS_DEVICE(&s->snvs), 0, FSL_IMX6UL_SNVS_HP_ADDR); /* - * Watchdog + * Watchdogs */ for (i = 0; i < FSL_IMX6UL_NUM_WDTS; i++) { static const hwaddr FSL_IMX6UL_WDOGn_ADDR[FSL_IMX6UL_NUM_WDTS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) FSL_IMX6UL_WDOG2_ADDR, FSL_IMX6UL_WDOG3_ADDR, }; + static const int FSL_IMX6UL_WDOGn_IRQ[FSL_IMX6UL_NUM_WDTS] = { FSL_IMX6UL_WDOG1_IRQ, FSL_IMX6UL_WDOG2_IRQ, @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) /* * SDMA */ - create_unimplemented_device("sdma", FSL_IMX6UL_SDMA_ADDR, 0x4000); + create_unimplemented_device("sdma", FSL_IMX6UL_SDMA_ADDR, + FSL_IMX6UL_SDMA_SIZE); /* - * SAI (Audio SSI (Synchronous Serial Interface)) + * SAIs (Audio SSI (Synchronous Serial Interface)) */ - create_unimplemented_device("sai1", FSL_IMX6UL_SAI1_ADDR, 0x4000); - create_unimplemented_device("sai2", FSL_IMX6UL_SAI2_ADDR, 0x4000); - create_unimplemented_device("sai3", FSL_IMX6UL_SAI3_ADDR, 0x4000); + for (i = 0; i < FSL_IMX6UL_NUM_SAIS; i++) { + static const hwaddr FSL_IMX6UL_SAIn_ADDR[FSL_IMX6UL_NUM_SAIS] = { + FSL_IMX6UL_SAI1_ADDR, + FSL_IMX6UL_SAI2_ADDR, + FSL_IMX6UL_SAI3_ADDR, + }; + + snprintf(name, NAME_SIZE, "sai%d", i); + create_unimplemented_device(name, FSL_IMX6UL_SAIn_ADDR[i], + FSL_IMX6UL_SAIn_SIZE); + } /* - * PWM + * PWMs */ - create_unimplemented_device("pwm1", FSL_IMX6UL_PWM1_ADDR, 0x4000); - create_unimplemented_device("pwm2", FSL_IMX6UL_PWM2_ADDR, 0x4000); - create_unimplemented_device("pwm3", FSL_IMX6UL_PWM3_ADDR, 0x4000); - create_unimplemented_device("pwm4", FSL_IMX6UL_PWM4_ADDR, 0x4000); + for (i = 0; i < FSL_IMX6UL_NUM_PWMS; i++) { + static const hwaddr FSL_IMX6UL_PWMn_ADDR[FSL_IMX6UL_NUM_PWMS] = { + FSL_IMX6UL_PWM1_ADDR, + FSL_IMX6UL_PWM2_ADDR, + FSL_IMX6UL_PWM3_ADDR, + FSL_IMX6UL_PWM4_ADDR, + }; + + snprintf(name, NAME_SIZE, "pwm%d", i); + create_unimplemented_device(name, FSL_IMX6UL_PWMn_ADDR[i], + FSL_IMX6UL_PWMn_SIZE); + } /* * Audio ASRC (asynchronous sample rate converter) */ - create_unimplemented_device("asrc", FSL_IMX6UL_ASRC_ADDR, 0x4000); + create_unimplemented_device("asrc", FSL_IMX6UL_ASRC_ADDR, + FSL_IMX6UL_ASRC_SIZE); /* - * CAN + * CANs */ - create_unimplemented_device("can1", FSL_IMX6UL_CAN1_ADDR, 0x4000); - create_unimplemented_device("can2", FSL_IMX6UL_CAN2_ADDR, 0x4000); + for (i = 0; i < FSL_IMX6UL_NUM_CANS; i++) { + static const hwaddr FSL_IMX6UL_CANn_ADDR[FSL_IMX6UL_NUM_CANS] = { + FSL_IMX6UL_CAN1_ADDR, + FSL_IMX6UL_CAN2_ADDR, + }; + + snprintf(name, NAME_SIZE, "can%d", i); + create_unimplemented_device(name, FSL_IMX6UL_CANn_ADDR[i], + FSL_IMX6UL_CANn_SIZE); + } /* * APHB_DMA @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) }; snprintf(name, NAME_SIZE, "adc%d", i); - create_unimplemented_device(name, FSL_IMX6UL_ADCn_ADDR[i], 0x4000); + create_unimplemented_device(name, FSL_IMX6UL_ADCn_ADDR[i], + FSL_IMX6UL_ADCn_SIZE); } /* * LCD */ - create_unimplemented_device("lcdif", FSL_IMX6UL_LCDIF_ADDR, 0x4000); + create_unimplemented_device("lcdif", FSL_IMX6UL_LCDIF_ADDR, + FSL_IMX6UL_LCDIF_SIZE); /* * ROM memory -- 2.34.1 From: Jean-Christophe Dubois <jcd@tribudubois.net> * Add TZASC as unimplemented device. - Allow bare metal application to access this (unimplemented) device * Add CSU as unimplemented device. - Allow bare metal application to access this (unimplemented) device * Add 4 missing PWM devices Signed-off-by: Jean-Christophe Dubois <jcd@tribudubois.net> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-id: 59e4dc56e14eccfefd379275ec19048dff9c10b3.1692964892.git.jcd@tribudubois.net Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/fsl-imx6ul.h | 2 +- hw/arm/fsl-imx6ul.c | 16 ++++++++++++++++ 2 files changed, 17 insertions(+), 1 deletion(-) diff --git a/include/hw/arm/fsl-imx6ul.h b/include/hw/arm/fsl-imx6ul.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/fsl-imx6ul.h +++ b/include/hw/arm/fsl-imx6ul.h @@ -XXX,XX +XXX,XX @@ enum FslIMX6ULConfiguration { FSL_IMX6UL_NUM_USBS = 2, FSL_IMX6UL_NUM_SAIS = 3, FSL_IMX6UL_NUM_CANS = 2, - FSL_IMX6UL_NUM_PWMS = 4, + FSL_IMX6UL_NUM_PWMS = 8, }; struct FslIMX6ULState { diff --git a/hw/arm/fsl-imx6ul.c b/hw/arm/fsl-imx6ul.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/fsl-imx6ul.c +++ b/hw/arm/fsl-imx6ul.c @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) FSL_IMX6UL_PWM2_ADDR, FSL_IMX6UL_PWM3_ADDR, FSL_IMX6UL_PWM4_ADDR, + FSL_IMX6UL_PWM5_ADDR, + FSL_IMX6UL_PWM6_ADDR, + FSL_IMX6UL_PWM7_ADDR, + FSL_IMX6UL_PWM8_ADDR, }; snprintf(name, NAME_SIZE, "pwm%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx6ul_realize(DeviceState *dev, Error **errp) create_unimplemented_device("lcdif", FSL_IMX6UL_LCDIF_ADDR, FSL_IMX6UL_LCDIF_SIZE); + /* + * CSU + */ + create_unimplemented_device("csu", FSL_IMX6UL_CSU_ADDR, + FSL_IMX6UL_CSU_SIZE); + + /* + * TZASC + */ + create_unimplemented_device("tzasc", FSL_IMX6UL_TZASC_ADDR, + FSL_IMX6UL_TZASC_SIZE); + /* * ROM memory */ -- 2.34.1 From: Jean-Christophe Dubois <jcd@tribudubois.net> * Add Addr and size definition for all i.MX7 devices in i.MX7 header file. * Use those newly defined named constants whenever possible. * Standardize the way we init a familly of unimplemented devices - SAI - PWM - CAN * Add/rework few comments Signed-off-by: Jean-Christophe Dubois <jcd@tribudubois.net> Message-id: 59e195d33e4d486a8d131392acd46633c8c10ed7.1692964892.git.jcd@tribudubois.net Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/fsl-imx7.h | 330 ++++++++++++++++++++++++++++---------- hw/arm/fsl-imx7.c | 130 ++++++++++----- 2 files changed, 335 insertions(+), 125 deletions(-) diff --git a/include/hw/arm/fsl-imx7.h b/include/hw/arm/fsl-imx7.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/fsl-imx7.h +++ b/include/hw/arm/fsl-imx7.h @@ -XXX,XX +XXX,XX @@ #include "hw/misc/imx7_ccm.h" #include "hw/misc/imx7_snvs.h" #include "hw/misc/imx7_gpr.h" -#include "hw/misc/imx6_src.h" #include "hw/watchdog/wdt_imx2.h" #include "hw/gpio/imx_gpio.h" #include "hw/char/imx_serial.h" @@ -XXX,XX +XXX,XX @@ #include "hw/usb/chipidea.h" #include "cpu.h" #include "qom/object.h" +#include "qemu/units.h" #define TYPE_FSL_IMX7 "fsl-imx7" OBJECT_DECLARE_SIMPLE_TYPE(FslIMX7State, FSL_IMX7) @@ -XXX,XX +XXX,XX @@ enum FslIMX7Configuration { FSL_IMX7_NUM_ECSPIS = 4, FSL_IMX7_NUM_USBS = 3, FSL_IMX7_NUM_ADCS = 2, + FSL_IMX7_NUM_SAIS = 3, + FSL_IMX7_NUM_CANS = 2, + FSL_IMX7_NUM_PWMS = 4, }; struct FslIMX7State { @@ -XXX,XX +XXX,XX @@ struct FslIMX7State { enum FslIMX7MemoryMap { FSL_IMX7_MMDC_ADDR = 0x80000000, - FSL_IMX7_MMDC_SIZE = 2 * 1024 * 1024 * 1024UL, + FSL_IMX7_MMDC_SIZE = (2 * GiB), - FSL_IMX7_GPIO1_ADDR = 0x30200000, - FSL_IMX7_GPIO2_ADDR = 0x30210000, - FSL_IMX7_GPIO3_ADDR = 0x30220000, - FSL_IMX7_GPIO4_ADDR = 0x30230000, - FSL_IMX7_GPIO5_ADDR = 0x30240000, - FSL_IMX7_GPIO6_ADDR = 0x30250000, - FSL_IMX7_GPIO7_ADDR = 0x30260000, + FSL_IMX7_QSPI1_MEM_ADDR = 0x60000000, + FSL_IMX7_QSPI1_MEM_SIZE = (256 * MiB), - FSL_IMX7_IOMUXC_LPSR_GPR_ADDR = 0x30270000, + FSL_IMX7_PCIE1_MEM_ADDR = 0x40000000, + FSL_IMX7_PCIE1_MEM_SIZE = (256 * MiB), - FSL_IMX7_WDOG1_ADDR = 0x30280000, - FSL_IMX7_WDOG2_ADDR = 0x30290000, - FSL_IMX7_WDOG3_ADDR = 0x302A0000, - FSL_IMX7_WDOG4_ADDR = 0x302B0000, + FSL_IMX7_QSPI1_RX_BUF_ADDR = 0x34000000, + FSL_IMX7_QSPI1_RX_BUF_SIZE = (32 * MiB), - FSL_IMX7_IOMUXC_LPSR_ADDR = 0x302C0000, + /* PCIe Peripherals */ + FSL_IMX7_PCIE_REG_ADDR = 0x33800000, - FSL_IMX7_GPT1_ADDR = 0x302D0000, - FSL_IMX7_GPT2_ADDR = 0x302E0000, - FSL_IMX7_GPT3_ADDR = 0x302F0000, - FSL_IMX7_GPT4_ADDR = 0x30300000, + /* MMAP Peripherals */ + FSL_IMX7_DMA_APBH_ADDR = 0x33000000, + FSL_IMX7_DMA_APBH_SIZE = 0x8000, - FSL_IMX7_IOMUXC_ADDR = 0x30330000, - FSL_IMX7_IOMUXC_GPR_ADDR = 0x30340000, - FSL_IMX7_IOMUXCn_SIZE = 0x1000, + /* GPV configuration */ + FSL_IMX7_GPV6_ADDR = 0x32600000, + FSL_IMX7_GPV5_ADDR = 0x32500000, + FSL_IMX7_GPV4_ADDR = 0x32400000, + FSL_IMX7_GPV3_ADDR = 0x32300000, + FSL_IMX7_GPV2_ADDR = 0x32200000, + FSL_IMX7_GPV1_ADDR = 0x32100000, + FSL_IMX7_GPV0_ADDR = 0x32000000, + FSL_IMX7_GPVn_SIZE = (1 * MiB), - FSL_IMX7_OCOTP_ADDR = 0x30350000, - FSL_IMX7_OCOTP_SIZE = 0x10000, + /* Arm Peripherals */ + FSL_IMX7_A7MPCORE_ADDR = 0x31000000, - FSL_IMX7_ANALOG_ADDR = 0x30360000, - FSL_IMX7_SNVS_ADDR = 0x30370000, - FSL_IMX7_CCM_ADDR = 0x30380000, + /* AIPS-3 Begin */ - FSL_IMX7_SRC_ADDR = 0x30390000, - FSL_IMX7_SRC_SIZE = 0x1000, + FSL_IMX7_ENET2_ADDR = 0x30BF0000, + FSL_IMX7_ENET1_ADDR = 0x30BE0000, - FSL_IMX7_ADC1_ADDR = 0x30610000, - FSL_IMX7_ADC2_ADDR = 0x30620000, - FSL_IMX7_ADCn_SIZE = 0x1000, + FSL_IMX7_SDMA_ADDR = 0x30BD0000, + FSL_IMX7_SDMA_SIZE = (4 * KiB), - FSL_IMX7_PWM1_ADDR = 0x30660000, - FSL_IMX7_PWM2_ADDR = 0x30670000, - FSL_IMX7_PWM3_ADDR = 0x30680000, - FSL_IMX7_PWM4_ADDR = 0x30690000, - FSL_IMX7_PWMn_SIZE = 0x10000, + FSL_IMX7_EIM_ADDR = 0x30BC0000, + FSL_IMX7_EIM_SIZE = (4 * KiB), - FSL_IMX7_PCIE_PHY_ADDR = 0x306D0000, - FSL_IMX7_PCIE_PHY_SIZE = 0x10000, + FSL_IMX7_QSPI_ADDR = 0x30BB0000, + FSL_IMX7_QSPI_SIZE = 0x8000, - FSL_IMX7_GPC_ADDR = 0x303A0000, + FSL_IMX7_SIM2_ADDR = 0x30BA0000, + FSL_IMX7_SIM1_ADDR = 0x30B90000, + FSL_IMX7_SIMn_SIZE = (4 * KiB), + + FSL_IMX7_USDHC3_ADDR = 0x30B60000, + FSL_IMX7_USDHC2_ADDR = 0x30B50000, + FSL_IMX7_USDHC1_ADDR = 0x30B40000, + + FSL_IMX7_USB3_ADDR = 0x30B30000, + FSL_IMX7_USBMISC3_ADDR = 0x30B30200, + FSL_IMX7_USB2_ADDR = 0x30B20000, + FSL_IMX7_USBMISC2_ADDR = 0x30B20200, + FSL_IMX7_USB1_ADDR = 0x30B10000, + FSL_IMX7_USBMISC1_ADDR = 0x30B10200, + FSL_IMX7_USBMISCn_SIZE = 0x200, + + FSL_IMX7_USB_PL301_ADDR = 0x30AD0000, + FSL_IMX7_USB_PL301_SIZE = (64 * KiB), + + FSL_IMX7_SEMAPHORE_HS_ADDR = 0x30AC0000, + FSL_IMX7_SEMAPHORE_HS_SIZE = (64 * KiB), + + FSL_IMX7_MUB_ADDR = 0x30AB0000, + FSL_IMX7_MUA_ADDR = 0x30AA0000, + FSL_IMX7_MUn_SIZE = (KiB), + + FSL_IMX7_UART7_ADDR = 0x30A90000, + FSL_IMX7_UART6_ADDR = 0x30A80000, + FSL_IMX7_UART5_ADDR = 0x30A70000, + FSL_IMX7_UART4_ADDR = 0x30A60000, + + FSL_IMX7_I2C4_ADDR = 0x30A50000, + FSL_IMX7_I2C3_ADDR = 0x30A40000, + FSL_IMX7_I2C2_ADDR = 0x30A30000, + FSL_IMX7_I2C1_ADDR = 0x30A20000, + + FSL_IMX7_CAN2_ADDR = 0x30A10000, + FSL_IMX7_CAN1_ADDR = 0x30A00000, + FSL_IMX7_CANn_SIZE = (4 * KiB), + + FSL_IMX7_AIPS3_CONF_ADDR = 0x309F0000, + FSL_IMX7_AIPS3_CONF_SIZE = (64 * KiB), FSL_IMX7_CAAM_ADDR = 0x30900000, - FSL_IMX7_CAAM_SIZE = 0x40000, + FSL_IMX7_CAAM_SIZE = (256 * KiB), - FSL_IMX7_CAN1_ADDR = 0x30A00000, - FSL_IMX7_CAN2_ADDR = 0x30A10000, - FSL_IMX7_CANn_SIZE = 0x10000, + FSL_IMX7_SPBA_ADDR = 0x308F0000, + FSL_IMX7_SPBA_SIZE = (4 * KiB), - FSL_IMX7_I2C1_ADDR = 0x30A20000, - FSL_IMX7_I2C2_ADDR = 0x30A30000, - FSL_IMX7_I2C3_ADDR = 0x30A40000, - FSL_IMX7_I2C4_ADDR = 0x30A50000, + FSL_IMX7_SAI3_ADDR = 0x308C0000, + FSL_IMX7_SAI2_ADDR = 0x308B0000, + FSL_IMX7_SAI1_ADDR = 0x308A0000, + FSL_IMX7_SAIn_SIZE = (4 * KiB), - FSL_IMX7_ECSPI1_ADDR = 0x30820000, - FSL_IMX7_ECSPI2_ADDR = 0x30830000, - FSL_IMX7_ECSPI3_ADDR = 0x30840000, - FSL_IMX7_ECSPI4_ADDR = 0x30630000, - - FSL_IMX7_LCDIF_ADDR = 0x30730000, - FSL_IMX7_LCDIF_SIZE = 0x1000, - - FSL_IMX7_UART1_ADDR = 0x30860000, + FSL_IMX7_UART3_ADDR = 0x30880000, /* * Some versions of the reference manual claim that UART2 is @ * 0x30870000, but experiments with HW + DT files in upstream @@ -XXX,XX +XXX,XX @@ enum FslIMX7MemoryMap { * actually located @ 0x30890000 */ FSL_IMX7_UART2_ADDR = 0x30890000, - FSL_IMX7_UART3_ADDR = 0x30880000, - FSL_IMX7_UART4_ADDR = 0x30A60000, - FSL_IMX7_UART5_ADDR = 0x30A70000, - FSL_IMX7_UART6_ADDR = 0x30A80000, - FSL_IMX7_UART7_ADDR = 0x30A90000, + FSL_IMX7_UART1_ADDR = 0x30860000, - FSL_IMX7_SAI1_ADDR = 0x308A0000, - FSL_IMX7_SAI2_ADDR = 0x308B0000, - FSL_IMX7_SAI3_ADDR = 0x308C0000, - FSL_IMX7_SAIn_SIZE = 0x10000, + FSL_IMX7_ECSPI3_ADDR = 0x30840000, + FSL_IMX7_ECSPI2_ADDR = 0x30830000, + FSL_IMX7_ECSPI1_ADDR = 0x30820000, + FSL_IMX7_ECSPIn_SIZE = (4 * KiB), - FSL_IMX7_ENET1_ADDR = 0x30BE0000, - FSL_IMX7_ENET2_ADDR = 0x30BF0000, + /* AIPS-3 End */ - FSL_IMX7_USB1_ADDR = 0x30B10000, - FSL_IMX7_USBMISC1_ADDR = 0x30B10200, - FSL_IMX7_USB2_ADDR = 0x30B20000, - FSL_IMX7_USBMISC2_ADDR = 0x30B20200, - FSL_IMX7_USB3_ADDR = 0x30B30000, - FSL_IMX7_USBMISC3_ADDR = 0x30B30200, - FSL_IMX7_USBMISCn_SIZE = 0x200, + /* AIPS-2 Begin */ - FSL_IMX7_USDHC1_ADDR = 0x30B40000, - FSL_IMX7_USDHC2_ADDR = 0x30B50000, - FSL_IMX7_USDHC3_ADDR = 0x30B60000, + FSL_IMX7_AXI_DEBUG_MON_ADDR = 0x307E0000, + FSL_IMX7_AXI_DEBUG_MON_SIZE = (64 * KiB), - FSL_IMX7_SDMA_ADDR = 0x30BD0000, - FSL_IMX7_SDMA_SIZE = 0x1000, + FSL_IMX7_PERFMON2_ADDR = 0x307D0000, + FSL_IMX7_PERFMON1_ADDR = 0x307C0000, + FSL_IMX7_PERFMONn_SIZE = (64 * KiB), + + FSL_IMX7_DDRC_ADDR = 0x307A0000, + FSL_IMX7_DDRC_SIZE = (4 * KiB), + + FSL_IMX7_DDRC_PHY_ADDR = 0x30790000, + FSL_IMX7_DDRC_PHY_SIZE = (4 * KiB), + + FSL_IMX7_TZASC_ADDR = 0x30780000, + FSL_IMX7_TZASC_SIZE = (64 * KiB), + + FSL_IMX7_MIPI_DSI_ADDR = 0x30760000, + FSL_IMX7_MIPI_DSI_SIZE = (4 * KiB), + + FSL_IMX7_MIPI_CSI_ADDR = 0x30750000, + FSL_IMX7_MIPI_CSI_SIZE = 0x4000, + + FSL_IMX7_LCDIF_ADDR = 0x30730000, + FSL_IMX7_LCDIF_SIZE = 0x8000, + + FSL_IMX7_CSI_ADDR = 0x30710000, + FSL_IMX7_CSI_SIZE = (4 * KiB), + + FSL_IMX7_PXP_ADDR = 0x30700000, + FSL_IMX7_PXP_SIZE = 0x4000, + + FSL_IMX7_EPDC_ADDR = 0x306F0000, + FSL_IMX7_EPDC_SIZE = (4 * KiB), + + FSL_IMX7_PCIE_PHY_ADDR = 0x306D0000, + FSL_IMX7_PCIE_PHY_SIZE = (4 * KiB), + + FSL_IMX7_SYSCNT_CTRL_ADDR = 0x306C0000, + FSL_IMX7_SYSCNT_CMP_ADDR = 0x306B0000, + FSL_IMX7_SYSCNT_RD_ADDR = 0x306A0000, + + FSL_IMX7_PWM4_ADDR = 0x30690000, + FSL_IMX7_PWM3_ADDR = 0x30680000, + FSL_IMX7_PWM2_ADDR = 0x30670000, + FSL_IMX7_PWM1_ADDR = 0x30660000, + FSL_IMX7_PWMn_SIZE = (4 * KiB), + + FSL_IMX7_FlEXTIMER2_ADDR = 0x30650000, + FSL_IMX7_FlEXTIMER1_ADDR = 0x30640000, + FSL_IMX7_FLEXTIMERn_SIZE = (4 * KiB), + + FSL_IMX7_ECSPI4_ADDR = 0x30630000, + + FSL_IMX7_ADC2_ADDR = 0x30620000, + FSL_IMX7_ADC1_ADDR = 0x30610000, + FSL_IMX7_ADCn_SIZE = (4 * KiB), + + FSL_IMX7_AIPS2_CONF_ADDR = 0x305F0000, + FSL_IMX7_AIPS2_CONF_SIZE = (64 * KiB), + + /* AIPS-2 End */ + + /* AIPS-1 Begin */ + + FSL_IMX7_CSU_ADDR = 0x303E0000, + FSL_IMX7_CSU_SIZE = (64 * KiB), + + FSL_IMX7_RDC_ADDR = 0x303D0000, + FSL_IMX7_RDC_SIZE = (4 * KiB), + + FSL_IMX7_SEMAPHORE2_ADDR = 0x303C0000, + FSL_IMX7_SEMAPHORE1_ADDR = 0x303B0000, + FSL_IMX7_SEMAPHOREn_SIZE = (4 * KiB), + + FSL_IMX7_GPC_ADDR = 0x303A0000, + + FSL_IMX7_SRC_ADDR = 0x30390000, + FSL_IMX7_SRC_SIZE = (4 * KiB), + + FSL_IMX7_CCM_ADDR = 0x30380000, + + FSL_IMX7_SNVS_HP_ADDR = 0x30370000, + + FSL_IMX7_ANALOG_ADDR = 0x30360000, + + FSL_IMX7_OCOTP_ADDR = 0x30350000, + FSL_IMX7_OCOTP_SIZE = 0x10000, + + FSL_IMX7_IOMUXC_GPR_ADDR = 0x30340000, + FSL_IMX7_IOMUXC_GPR_SIZE = (4 * KiB), + + FSL_IMX7_IOMUXC_ADDR = 0x30330000, + FSL_IMX7_IOMUXC_SIZE = (4 * KiB), + + FSL_IMX7_KPP_ADDR = 0x30320000, + FSL_IMX7_KPP_SIZE = (4 * KiB), + + FSL_IMX7_ROMCP_ADDR = 0x30310000, + FSL_IMX7_ROMCP_SIZE = (4 * KiB), + + FSL_IMX7_GPT4_ADDR = 0x30300000, + FSL_IMX7_GPT3_ADDR = 0x302F0000, + FSL_IMX7_GPT2_ADDR = 0x302E0000, + FSL_IMX7_GPT1_ADDR = 0x302D0000, + + FSL_IMX7_IOMUXC_LPSR_ADDR = 0x302C0000, + FSL_IMX7_IOMUXC_LPSR_SIZE = (4 * KiB), + + FSL_IMX7_WDOG4_ADDR = 0x302B0000, + FSL_IMX7_WDOG3_ADDR = 0x302A0000, + FSL_IMX7_WDOG2_ADDR = 0x30290000, + FSL_IMX7_WDOG1_ADDR = 0x30280000, + + FSL_IMX7_IOMUXC_LPSR_GPR_ADDR = 0x30270000, + + FSL_IMX7_GPIO7_ADDR = 0x30260000, + FSL_IMX7_GPIO6_ADDR = 0x30250000, + FSL_IMX7_GPIO5_ADDR = 0x30240000, + FSL_IMX7_GPIO4_ADDR = 0x30230000, + FSL_IMX7_GPIO3_ADDR = 0x30220000, + FSL_IMX7_GPIO2_ADDR = 0x30210000, + FSL_IMX7_GPIO1_ADDR = 0x30200000, + + FSL_IMX7_AIPS1_CONF_ADDR = 0x301F0000, + FSL_IMX7_AIPS1_CONF_SIZE = (64 * KiB), - FSL_IMX7_A7MPCORE_ADDR = 0x31000000, FSL_IMX7_A7MPCORE_DAP_ADDR = 0x30000000, + FSL_IMX7_A7MPCORE_DAP_SIZE = (1 * MiB), - FSL_IMX7_PCIE_REG_ADDR = 0x33800000, - FSL_IMX7_PCIE_REG_SIZE = 16 * 1024, + /* AIPS-1 End */ - FSL_IMX7_GPR_ADDR = 0x30340000, + FSL_IMX7_EIM_CS0_ADDR = 0x28000000, + FSL_IMX7_EIM_CS0_SIZE = (128 * MiB), - FSL_IMX7_DMA_APBH_ADDR = 0x33000000, - FSL_IMX7_DMA_APBH_SIZE = 0x2000, + FSL_IMX7_OCRAM_PXP_ADDR = 0x00940000, + FSL_IMX7_OCRAM_PXP_SIZE = (32 * KiB), + + FSL_IMX7_OCRAM_EPDC_ADDR = 0x00920000, + FSL_IMX7_OCRAM_EPDC_SIZE = (128 * KiB), + + FSL_IMX7_OCRAM_MEM_ADDR = 0x00900000, + FSL_IMX7_OCRAM_MEM_SIZE = (128 * KiB), + + FSL_IMX7_TCMU_ADDR = 0x00800000, + FSL_IMX7_TCMU_SIZE = (32 * KiB), + + FSL_IMX7_TCML_ADDR = 0x007F8000, + FSL_IMX7_TCML_SIZE = (32 * KiB), + + FSL_IMX7_OCRAM_S_ADDR = 0x00180000, + FSL_IMX7_OCRAM_S_SIZE = (32 * KiB), + + FSL_IMX7_CAAM_MEM_ADDR = 0x00100000, + FSL_IMX7_CAAM_MEM_SIZE = (32 * KiB), + + FSL_IMX7_ROM_ADDR = 0x00000000, + FSL_IMX7_ROM_SIZE = (96 * KiB), }; enum FslIMX7IRQs { diff --git a/hw/arm/fsl-imx7.c b/hw/arm/fsl-imx7.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/fsl-imx7.c +++ b/hw/arm/fsl-imx7.c @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_init(Object *obj) char name[NAME_SIZE]; int i; + /* + * CPUs + */ for (i = 0; i < MIN(ms->smp.cpus, FSL_IMX7_NUM_CPUS); i++) { snprintf(name, NAME_SIZE, "cpu%d", i); object_initialize_child(obj, name, &s->cpu[i], @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_init(Object *obj) TYPE_A15MPCORE_PRIV); /* - * GPIOs 1 to 7 + * GPIOs */ for (i = 0; i < FSL_IMX7_NUM_GPIOS; i++) { snprintf(name, NAME_SIZE, "gpio%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_init(Object *obj) } /* - * GPT1, 2, 3, 4 + * GPTs */ for (i = 0; i < FSL_IMX7_NUM_GPTS; i++) { snprintf(name, NAME_SIZE, "gpt%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_init(Object *obj) */ object_initialize_child(obj, "gpcv2", &s->gpcv2, TYPE_IMX_GPCV2); + /* + * ECSPIs + */ for (i = 0; i < FSL_IMX7_NUM_ECSPIS; i++) { snprintf(name, NAME_SIZE, "spi%d", i + 1); object_initialize_child(obj, name, &s->spi[i], TYPE_IMX_SPI); } - + /* + * I2Cs + */ for (i = 0; i < FSL_IMX7_NUM_I2CS; i++) { snprintf(name, NAME_SIZE, "i2c%d", i + 1); object_initialize_child(obj, name, &s->i2c[i], TYPE_IMX_I2C); } /* - * UART + * UARTs */ for (i = 0; i < FSL_IMX7_NUM_UARTS; i++) { snprintf(name, NAME_SIZE, "uart%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_init(Object *obj) } /* - * Ethernet + * Ethernets */ for (i = 0; i < FSL_IMX7_NUM_ETHS; i++) { snprintf(name, NAME_SIZE, "eth%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_init(Object *obj) } /* - * SDHCI + * SDHCIs */ for (i = 0; i < FSL_IMX7_NUM_USDHCS; i++) { snprintf(name, NAME_SIZE, "usdhc%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_init(Object *obj) object_initialize_child(obj, "snvs", &s->snvs, TYPE_IMX7_SNVS); /* - * Watchdog + * Watchdogs */ for (i = 0; i < FSL_IMX7_NUM_WDTS; i++) { snprintf(name, NAME_SIZE, "wdt%d", i); @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_init(Object *obj) */ object_initialize_child(obj, "gpr", &s->gpr, TYPE_IMX7_GPR); + /* + * PCIE + */ object_initialize_child(obj, "pcie", &s->pcie, TYPE_DESIGNWARE_PCIE_HOST); + /* + * USBs + */ for (i = 0; i < FSL_IMX7_NUM_USBS; i++) { snprintf(name, NAME_SIZE, "usb%d", i); object_initialize_child(obj, name, &s->usb[i], TYPE_CHIPIDEA); @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) return; } + /* + * CPUs + */ for (i = 0; i < smp_cpus; i++) { o = OBJECT(&s->cpu[i]); @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) * A7MPCORE DAP */ create_unimplemented_device("a7mpcore-dap", FSL_IMX7_A7MPCORE_DAP_ADDR, - 0x100000); + FSL_IMX7_A7MPCORE_DAP_SIZE); /* - * GPT1, 2, 3, 4 + * GPTs */ for (i = 0; i < FSL_IMX7_NUM_GPTS; i++) { static const hwaddr FSL_IMX7_GPTn_ADDR[FSL_IMX7_NUM_GPTS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) FSL_IMX7_GPTn_IRQ[i])); } + /* + * GPIOs + */ for (i = 0; i < FSL_IMX7_NUM_GPIOS; i++) { static const hwaddr FSL_IMX7_GPIOn_ADDR[FSL_IMX7_NUM_GPIOS] = { FSL_IMX7_GPIO1_ADDR, @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) /* * IOMUXC and IOMUXC_LPSR */ - for (i = 0; i < FSL_IMX7_NUM_IOMUXCS; i++) { - static const hwaddr FSL_IMX7_IOMUXCn_ADDR[FSL_IMX7_NUM_IOMUXCS] = { - FSL_IMX7_IOMUXC_ADDR, - FSL_IMX7_IOMUXC_LPSR_ADDR, - }; - - snprintf(name, NAME_SIZE, "iomuxc%d", i); - create_unimplemented_device(name, FSL_IMX7_IOMUXCn_ADDR[i], - FSL_IMX7_IOMUXCn_SIZE); - } + create_unimplemented_device("iomuxc", FSL_IMX7_IOMUXC_ADDR, + FSL_IMX7_IOMUXC_SIZE); + create_unimplemented_device("iomuxc_lspr", FSL_IMX7_IOMUXC_LPSR_ADDR, + FSL_IMX7_IOMUXC_LPSR_SIZE); /* * CCM @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) sysbus_realize(SYS_BUS_DEVICE(&s->gpcv2), &error_abort); sysbus_mmio_map(SYS_BUS_DEVICE(&s->gpcv2), 0, FSL_IMX7_GPC_ADDR); - /* Initialize all ECSPI */ + /* + * ECSPIs + */ for (i = 0; i < FSL_IMX7_NUM_ECSPIS; i++) { static const hwaddr FSL_IMX7_SPIn_ADDR[FSL_IMX7_NUM_ECSPIS] = { FSL_IMX7_ECSPI1_ADDR, @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) FSL_IMX7_SPIn_IRQ[i])); } + /* + * I2Cs + */ for (i = 0; i < FSL_IMX7_NUM_I2CS; i++) { static const hwaddr FSL_IMX7_I2Cn_ADDR[FSL_IMX7_NUM_I2CS] = { FSL_IMX7_I2C1_ADDR, @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) } /* - * UART + * UARTs */ for (i = 0; i < FSL_IMX7_NUM_UARTS; i++) { static const hwaddr FSL_IMX7_UARTn_ADDR[FSL_IMX7_NUM_UARTS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) } /* - * Ethernet + * Ethernets * * We must use two loops since phy_connected affects the other interface * and we have to set all properties before calling sysbus_realize(). @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) } /* - * USDHC + * USDHCs */ for (i = 0; i < FSL_IMX7_NUM_USDHCS; i++) { static const hwaddr FSL_IMX7_USDHCn_ADDR[FSL_IMX7_NUM_USDHCS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) * SNVS */ sysbus_realize(SYS_BUS_DEVICE(&s->snvs), &error_abort); - sysbus_mmio_map(SYS_BUS_DEVICE(&s->snvs), 0, FSL_IMX7_SNVS_ADDR); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->snvs), 0, FSL_IMX7_SNVS_HP_ADDR); /* * SRC @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) create_unimplemented_device("src", FSL_IMX7_SRC_ADDR, FSL_IMX7_SRC_SIZE); /* - * Watchdog + * Watchdogs */ for (i = 0; i < FSL_IMX7_NUM_WDTS; i++) { static const hwaddr FSL_IMX7_WDOGn_ADDR[FSL_IMX7_NUM_WDTS] = { @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) create_unimplemented_device("caam", FSL_IMX7_CAAM_ADDR, FSL_IMX7_CAAM_SIZE); /* - * PWM + * PWMs */ - create_unimplemented_device("pwm1", FSL_IMX7_PWM1_ADDR, FSL_IMX7_PWMn_SIZE); - create_unimplemented_device("pwm2", FSL_IMX7_PWM2_ADDR, FSL_IMX7_PWMn_SIZE); - create_unimplemented_device("pwm3", FSL_IMX7_PWM3_ADDR, FSL_IMX7_PWMn_SIZE); - create_unimplemented_device("pwm4", FSL_IMX7_PWM4_ADDR, FSL_IMX7_PWMn_SIZE); + for (i = 0; i < FSL_IMX7_NUM_PWMS; i++) { + static const hwaddr FSL_IMX7_PWMn_ADDR[FSL_IMX7_NUM_PWMS] = { + FSL_IMX7_PWM1_ADDR, + FSL_IMX7_PWM2_ADDR, + FSL_IMX7_PWM3_ADDR, + FSL_IMX7_PWM4_ADDR, + }; + + snprintf(name, NAME_SIZE, "pwm%d", i); + create_unimplemented_device(name, FSL_IMX7_PWMn_ADDR[i], + FSL_IMX7_PWMn_SIZE); + } /* - * CAN + * CANs */ - create_unimplemented_device("can1", FSL_IMX7_CAN1_ADDR, FSL_IMX7_CANn_SIZE); - create_unimplemented_device("can2", FSL_IMX7_CAN2_ADDR, FSL_IMX7_CANn_SIZE); + for (i = 0; i < FSL_IMX7_NUM_CANS; i++) { + static const hwaddr FSL_IMX7_CANn_ADDR[FSL_IMX7_NUM_CANS] = { + FSL_IMX7_CAN1_ADDR, + FSL_IMX7_CAN2_ADDR, + }; + + snprintf(name, NAME_SIZE, "can%d", i); + create_unimplemented_device(name, FSL_IMX7_CANn_ADDR[i], + FSL_IMX7_CANn_SIZE); + } /* - * SAI (Audio SSI (Synchronous Serial Interface)) + * SAIs (Audio SSI (Synchronous Serial Interface)) */ - create_unimplemented_device("sai1", FSL_IMX7_SAI1_ADDR, FSL_IMX7_SAIn_SIZE); - create_unimplemented_device("sai2", FSL_IMX7_SAI2_ADDR, FSL_IMX7_SAIn_SIZE); - create_unimplemented_device("sai2", FSL_IMX7_SAI3_ADDR, FSL_IMX7_SAIn_SIZE); + for (i = 0; i < FSL_IMX7_NUM_SAIS; i++) { + static const hwaddr FSL_IMX7_SAIn_ADDR[FSL_IMX7_NUM_SAIS] = { + FSL_IMX7_SAI1_ADDR, + FSL_IMX7_SAI2_ADDR, + FSL_IMX7_SAI3_ADDR, + }; + + snprintf(name, NAME_SIZE, "sai%d", i); + create_unimplemented_device(name, FSL_IMX7_SAIn_ADDR[i], + FSL_IMX7_SAIn_SIZE); + } /* * OCOTP @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) create_unimplemented_device("ocotp", FSL_IMX7_OCOTP_ADDR, FSL_IMX7_OCOTP_SIZE); + /* + * GPR + */ sysbus_realize(SYS_BUS_DEVICE(&s->gpr), &error_abort); - sysbus_mmio_map(SYS_BUS_DEVICE(&s->gpr), 0, FSL_IMX7_GPR_ADDR); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->gpr), 0, FSL_IMX7_IOMUXC_GPR_ADDR); + /* + * PCIE + */ sysbus_realize(SYS_BUS_DEVICE(&s->pcie), &error_abort); sysbus_mmio_map(SYS_BUS_DEVICE(&s->pcie), 0, FSL_IMX7_PCIE_REG_ADDR); @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) irq = qdev_get_gpio_in(DEVICE(&s->a7mpcore), FSL_IMX7_PCI_INTD_IRQ); sysbus_connect_irq(SYS_BUS_DEVICE(&s->pcie), 3, irq); - + /* + * USBs + */ for (i = 0; i < FSL_IMX7_NUM_USBS; i++) { static const hwaddr FSL_IMX7_USBMISCn_ADDR[FSL_IMX7_NUM_USBS] = { FSL_IMX7_USBMISC1_ADDR, @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) */ create_unimplemented_device("pcie-phy", FSL_IMX7_PCIE_PHY_ADDR, FSL_IMX7_PCIE_PHY_SIZE); + } static Property fsl_imx7_properties[] = { -- 2.34.1 From: Jean-Christophe Dubois <jcd@tribudubois.net> * Add TZASC as unimplemented device. - Allow bare metal application to access this (unimplemented) device * Add CSU as unimplemented device. - Allow bare metal application to access this (unimplemented) device * Add various memory segments - OCRAM - OCRAM EPDC - OCRAM PXP - OCRAM S - ROM - CAAM Signed-off-by: Jean-Christophe Dubois <jcd@tribudubois.net> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-id: f887a3483996ba06d40bd62ffdfb0ecf68621987.1692964892.git.jcd@tribudubois.net Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/fsl-imx7.h | 7 +++++ hw/arm/fsl-imx7.c | 63 +++++++++++++++++++++++++++++++++++++++ 2 files changed, 70 insertions(+) diff --git a/include/hw/arm/fsl-imx7.h b/include/hw/arm/fsl-imx7.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/fsl-imx7.h +++ b/include/hw/arm/fsl-imx7.h @@ -XXX,XX +XXX,XX @@ struct FslIMX7State { IMX7GPRState gpr; ChipideaState usb[FSL_IMX7_NUM_USBS]; DesignwarePCIEHost pcie; + MemoryRegion rom; + MemoryRegion caam; + MemoryRegion ocram; + MemoryRegion ocram_epdc; + MemoryRegion ocram_pxp; + MemoryRegion ocram_s; + uint32_t phy_num[FSL_IMX7_NUM_ETHS]; bool phy_connected[FSL_IMX7_NUM_ETHS]; }; diff --git a/hw/arm/fsl-imx7.c b/hw/arm/fsl-imx7.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/fsl-imx7.c +++ b/hw/arm/fsl-imx7.c @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) create_unimplemented_device("pcie-phy", FSL_IMX7_PCIE_PHY_ADDR, FSL_IMX7_PCIE_PHY_SIZE); + /* + * CSU + */ + create_unimplemented_device("csu", FSL_IMX7_CSU_ADDR, + FSL_IMX7_CSU_SIZE); + + /* + * TZASC + */ + create_unimplemented_device("tzasc", FSL_IMX7_TZASC_ADDR, + FSL_IMX7_TZASC_SIZE); + + /* + * OCRAM memory + */ + memory_region_init_ram(&s->ocram, NULL, "imx7.ocram", + FSL_IMX7_OCRAM_MEM_SIZE, + &error_abort); + memory_region_add_subregion(get_system_memory(), FSL_IMX7_OCRAM_MEM_ADDR, + &s->ocram); + + /* + * OCRAM EPDC memory + */ + memory_region_init_ram(&s->ocram_epdc, NULL, "imx7.ocram_epdc", + FSL_IMX7_OCRAM_EPDC_SIZE, + &error_abort); + memory_region_add_subregion(get_system_memory(), FSL_IMX7_OCRAM_EPDC_ADDR, + &s->ocram_epdc); + + /* + * OCRAM PXP memory + */ + memory_region_init_ram(&s->ocram_pxp, NULL, "imx7.ocram_pxp", + FSL_IMX7_OCRAM_PXP_SIZE, + &error_abort); + memory_region_add_subregion(get_system_memory(), FSL_IMX7_OCRAM_PXP_ADDR, + &s->ocram_pxp); + + /* + * OCRAM_S memory + */ + memory_region_init_ram(&s->ocram_s, NULL, "imx7.ocram_s", + FSL_IMX7_OCRAM_S_SIZE, + &error_abort); + memory_region_add_subregion(get_system_memory(), FSL_IMX7_OCRAM_S_ADDR, + &s->ocram_s); + + /* + * ROM memory + */ + memory_region_init_rom(&s->rom, OBJECT(dev), "imx7.rom", + FSL_IMX7_ROM_SIZE, &error_abort); + memory_region_add_subregion(get_system_memory(), FSL_IMX7_ROM_ADDR, + &s->rom); + + /* + * CAAM memory + */ + memory_region_init_rom(&s->caam, OBJECT(dev), "imx7.caam", + FSL_IMX7_CAAM_MEM_SIZE, &error_abort); + memory_region_add_subregion(get_system_memory(), FSL_IMX7_CAAM_MEM_ADDR, + &s->caam); } static Property fsl_imx7_properties[] = { -- 2.34.1 From: Jean-Christophe Dubois <jcd@tribudubois.net> The SRC device is normally used to start the secondary CPU. When running Linux directly, QEMU is emulating a PSCI interface that UBOOT is installing at boot time and therefore the fact that the SRC device is unimplemented is hidden as Qemu respond directly to PSCI requets without using the SRC device. But if you try to run a more bare metal application (maybe uboot itself), then it is not possible to start the secondary CPU as the SRC is an unimplemented device. This patch adds the ability to start the secondary CPU through the SRC device so that you can use this feature in bare metal applications. Signed-off-by: Jean-Christophe Dubois <jcd@tribudubois.net> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: ce9a0162defd2acee5dc7f8a674743de0cded569.1692964892.git.jcd@tribudubois.net Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/fsl-imx7.h | 3 +- include/hw/misc/imx7_src.h | 66 +++++++++ hw/arm/fsl-imx7.c | 8 +- hw/misc/imx7_src.c | 276 +++++++++++++++++++++++++++++++++++++ hw/misc/meson.build | 1 + hw/misc/trace-events | 4 + 6 files changed, 356 insertions(+), 2 deletions(-) create mode 100644 include/hw/misc/imx7_src.h create mode 100644 hw/misc/imx7_src.c diff --git a/include/hw/arm/fsl-imx7.h b/include/hw/arm/fsl-imx7.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/fsl-imx7.h +++ b/include/hw/arm/fsl-imx7.h @@ -XXX,XX +XXX,XX @@ #include "hw/misc/imx7_ccm.h" #include "hw/misc/imx7_snvs.h" #include "hw/misc/imx7_gpr.h" +#include "hw/misc/imx7_src.h" #include "hw/watchdog/wdt_imx2.h" #include "hw/gpio/imx_gpio.h" #include "hw/char/imx_serial.h" @@ -XXX,XX +XXX,XX @@ struct FslIMX7State { IMX7CCMState ccm; IMX7AnalogState analog; IMX7SNVSState snvs; + IMX7SRCState src; IMXGPCv2State gpcv2; IMXSPIState spi[FSL_IMX7_NUM_ECSPIS]; IMXI2CState i2c[FSL_IMX7_NUM_I2CS]; @@ -XXX,XX +XXX,XX @@ enum FslIMX7MemoryMap { FSL_IMX7_GPC_ADDR = 0x303A0000, FSL_IMX7_SRC_ADDR = 0x30390000, - FSL_IMX7_SRC_SIZE = (4 * KiB), FSL_IMX7_CCM_ADDR = 0x30380000, diff --git a/include/hw/misc/imx7_src.h b/include/hw/misc/imx7_src.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/misc/imx7_src.h @@ -XXX,XX +XXX,XX @@ +/* + * IMX7 System Reset Controller + * + * Copyright (C) 2023 Jean-Christophe Dubois <jcd@tribudubois.net> + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef IMX7_SRC_H +#define IMX7_SRC_H + +#include "hw/sysbus.h" +#include "qemu/bitops.h" +#include "qom/object.h" + +#define SRC_SCR 0 +#define SRC_A7RCR0 1 +#define SRC_A7RCR1 2 +#define SRC_M4RCR 3 +#define SRC_ERCR 5 +#define SRC_HSICPHY_RCR 7 +#define SRC_USBOPHY1_RCR 8 +#define SRC_USBOPHY2_RCR 9 +#define SRC_MPIPHY_RCR 10 +#define SRC_PCIEPHY_RCR 11 +#define SRC_SBMR1 22 +#define SRC_SRSR 23 +#define SRC_SISR 26 +#define SRC_SIMR 27 +#define SRC_SBMR2 28 +#define SRC_GPR1 29 +#define SRC_GPR2 30 +#define SRC_GPR3 31 +#define SRC_GPR4 32 +#define SRC_GPR5 33 +#define SRC_GPR6 34 +#define SRC_GPR7 35 +#define SRC_GPR8 36 +#define SRC_GPR9 37 +#define SRC_GPR10 38 +#define SRC_MAX 39 + +/* SRC_A7SCR1 */ +#define R_CORE1_ENABLE_SHIFT 1 +#define R_CORE1_ENABLE_LENGTH 1 +/* SRC_A7SCR0 */ +#define R_CORE1_RST_SHIFT 5 +#define R_CORE1_RST_LENGTH 1 +#define R_CORE0_RST_SHIFT 4 +#define R_CORE0_RST_LENGTH 1 + +#define TYPE_IMX7_SRC "imx7.src" +OBJECT_DECLARE_SIMPLE_TYPE(IMX7SRCState, IMX7_SRC) + +struct IMX7SRCState { + /* <private> */ + SysBusDevice parent_obj; + + /* <public> */ + MemoryRegion iomem; + + uint32_t regs[SRC_MAX]; +}; + +#endif /* IMX7_SRC_H */ diff --git a/hw/arm/fsl-imx7.c b/hw/arm/fsl-imx7.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/fsl-imx7.c +++ b/hw/arm/fsl-imx7.c @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_init(Object *obj) */ object_initialize_child(obj, "gpcv2", &s->gpcv2, TYPE_IMX_GPCV2); + /* + * SRC + */ + object_initialize_child(obj, "src", &s->src, TYPE_IMX7_SRC); + /* * ECSPIs */ @@ -XXX,XX +XXX,XX @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp) /* * SRC */ - create_unimplemented_device("src", FSL_IMX7_SRC_ADDR, FSL_IMX7_SRC_SIZE); + sysbus_realize(SYS_BUS_DEVICE(&s->src), &error_abort); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->src), 0, FSL_IMX7_SRC_ADDR); /* * Watchdogs diff --git a/hw/misc/imx7_src.c b/hw/misc/imx7_src.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/misc/imx7_src.c @@ -XXX,XX +XXX,XX @@ +/* + * IMX7 System Reset Controller + * + * Copyright (c) 2023 Jean-Christophe Dubois <jcd@tribudubois.net> + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + * + */ + +#include "qemu/osdep.h" +#include "hw/misc/imx7_src.h" +#include "migration/vmstate.h" +#include "qemu/bitops.h" +#include "qemu/log.h" +#include "qemu/main-loop.h" +#include "qemu/module.h" +#include "target/arm/arm-powerctl.h" +#include "hw/core/cpu.h" +#include "hw/registerfields.h" + +#include "trace.h" + +static const char *imx7_src_reg_name(uint32_t reg) +{ + static char unknown[20]; + + switch (reg) { + case SRC_SCR: + return "SRC_SCR"; + case SRC_A7RCR0: + return "SRC_A7RCR0"; + case SRC_A7RCR1: + return "SRC_A7RCR1"; + case SRC_M4RCR: + return "SRC_M4RCR"; + case SRC_ERCR: + return "SRC_ERCR"; + case SRC_HSICPHY_RCR: + return "SRC_HSICPHY_RCR"; + case SRC_USBOPHY1_RCR: + return "SRC_USBOPHY1_RCR"; + case SRC_USBOPHY2_RCR: + return "SRC_USBOPHY2_RCR"; + case SRC_PCIEPHY_RCR: + return "SRC_PCIEPHY_RCR"; + case SRC_SBMR1: + return "SRC_SBMR1"; + case SRC_SRSR: + return "SRC_SRSR"; + case SRC_SISR: + return "SRC_SISR"; + case SRC_SIMR: + return "SRC_SIMR"; + case SRC_SBMR2: + return "SRC_SBMR2"; + case SRC_GPR1: + return "SRC_GPR1"; + case SRC_GPR2: + return "SRC_GPR2"; + case SRC_GPR3: + return "SRC_GPR3"; + case SRC_GPR4: + return "SRC_GPR4"; + case SRC_GPR5: + return "SRC_GPR5"; + case SRC_GPR6: + return "SRC_GPR6"; + case SRC_GPR7: + return "SRC_GPR7"; + case SRC_GPR8: + return "SRC_GPR8"; + case SRC_GPR9: + return "SRC_GPR9"; + case SRC_GPR10: + return "SRC_GPR10"; + default: + sprintf(unknown, "%u ?", reg); + return unknown; + } +} + +static const VMStateDescription vmstate_imx7_src = { + .name = TYPE_IMX7_SRC, + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32_ARRAY(regs, IMX7SRCState, SRC_MAX), + VMSTATE_END_OF_LIST() + }, +}; + +static void imx7_src_reset(DeviceState *dev) +{ + IMX7SRCState *s = IMX7_SRC(dev); + + memset(s->regs, 0, sizeof(s->regs)); + + /* Set reset values */ + s->regs[SRC_SCR] = 0xA0; + s->regs[SRC_SRSR] = 0x1; + s->regs[SRC_SIMR] = 0x1F; +} + +static uint64_t imx7_src_read(void *opaque, hwaddr offset, unsigned size) +{ + uint32_t value = 0; + IMX7SRCState *s = (IMX7SRCState *)opaque; + uint32_t index = offset >> 2; + + if (index < SRC_MAX) { + value = s->regs[index]; + } else { + qemu_log_mask(LOG_GUEST_ERROR, "[%s]%s: Bad register at offset 0x%" + HWADDR_PRIx "\n", TYPE_IMX7_SRC, __func__, offset); + } + + trace_imx7_src_read(imx7_src_reg_name(index), value); + + return value; +} + + +/* + * The reset is asynchronous so we need to defer clearing the reset + * bit until the work is completed. + */ + +struct SRCSCRResetInfo { + IMX7SRCState *s; + uint32_t reset_bit; +}; + +static void imx7_clear_reset_bit(CPUState *cpu, run_on_cpu_data data) +{ + struct SRCSCRResetInfo *ri = data.host_ptr; + IMX7SRCState *s = ri->s; + + assert(qemu_mutex_iothread_locked()); + + s->regs[SRC_A7RCR0] = deposit32(s->regs[SRC_A7RCR0], ri->reset_bit, 1, 0); + + trace_imx7_src_write(imx7_src_reg_name(SRC_A7RCR0), s->regs[SRC_A7RCR0]); + + g_free(ri); +} + +static void imx7_defer_clear_reset_bit(uint32_t cpuid, + IMX7SRCState *s, + uint32_t reset_shift) +{ + struct SRCSCRResetInfo *ri; + CPUState *cpu = arm_get_cpu_by_id(cpuid); + + if (!cpu) { + return; + } + + ri = g_new(struct SRCSCRResetInfo, 1); + ri->s = s; + ri->reset_bit = reset_shift; + + async_run_on_cpu(cpu, imx7_clear_reset_bit, RUN_ON_CPU_HOST_PTR(ri)); +} + + +static void imx7_src_write(void *opaque, hwaddr offset, uint64_t value, + unsigned size) +{ + IMX7SRCState *s = (IMX7SRCState *)opaque; + uint32_t index = offset >> 2; + long unsigned int change_mask; + uint32_t current_value = value; + + if (index >= SRC_MAX) { + qemu_log_mask(LOG_GUEST_ERROR, "[%s]%s: Bad register at offset 0x%" + HWADDR_PRIx "\n", TYPE_IMX7_SRC, __func__, offset); + return; + } + + trace_imx7_src_write(imx7_src_reg_name(SRC_A7RCR0), s->regs[SRC_A7RCR0]); + + change_mask = s->regs[index] ^ (uint32_t)current_value; + + switch (index) { + case SRC_A7RCR0: + if (FIELD_EX32(change_mask, CORE0, RST)) { + arm_reset_cpu(0); + imx7_defer_clear_reset_bit(0, s, R_CORE0_RST_SHIFT); + } + if (FIELD_EX32(change_mask, CORE1, RST)) { + arm_reset_cpu(1); + imx7_defer_clear_reset_bit(1, s, R_CORE1_RST_SHIFT); + } + s->regs[index] = current_value; + break; + case SRC_A7RCR1: + /* + * On real hardware when the system reset controller starts a + * secondary CPU it runs through some boot ROM code which reads + * the SRC_GPRX registers controlling the start address and branches + * to it. + * Here we are taking a short cut and branching directly to the + * requested address (we don't want to run the boot ROM code inside + * QEMU) + */ + if (FIELD_EX32(change_mask, CORE1, ENABLE)) { + if (FIELD_EX32(current_value, CORE1, ENABLE)) { + /* CORE 1 is brought up */ + arm_set_cpu_on(1, s->regs[SRC_GPR3], s->regs[SRC_GPR4], + 3, false); + } else { + /* CORE 1 is shut down */ + arm_set_cpu_off(1); + } + /* We clear the reset bits as the processor changed state */ + imx7_defer_clear_reset_bit(1, s, R_CORE1_RST_SHIFT); + clear_bit(R_CORE1_RST_SHIFT, &change_mask); + } + s->regs[index] = current_value; + break; + default: + s->regs[index] = current_value; + break; + } +} + +static const struct MemoryRegionOps imx7_src_ops = { + .read = imx7_src_read, + .write = imx7_src_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + /* + * Our device would not work correctly if the guest was doing + * unaligned access. This might not be a limitation on the real + * device but in practice there is no reason for a guest to access + * this device unaligned. + */ + .min_access_size = 4, + .max_access_size = 4, + .unaligned = false, + }, +}; + +static void imx7_src_realize(DeviceState *dev, Error **errp) +{ + IMX7SRCState *s = IMX7_SRC(dev); + + memory_region_init_io(&s->iomem, OBJECT(dev), &imx7_src_ops, s, + TYPE_IMX7_SRC, 0x1000); + sysbus_init_mmio(SYS_BUS_DEVICE(dev), &s->iomem); +} + +static void imx7_src_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->realize = imx7_src_realize; + dc->reset = imx7_src_reset; + dc->vmsd = &vmstate_imx7_src; + dc->desc = "i.MX6 System Reset Controller"; +} + +static const TypeInfo imx7_src_info = { + .name = TYPE_IMX7_SRC, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(IMX7SRCState), + .class_init = imx7_src_class_init, +}; + +static void imx7_src_register_types(void) +{ + type_register_static(&imx7_src_info); +} + +type_init(imx7_src_register_types) diff --git a/hw/misc/meson.build b/hw/misc/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/meson.build +++ b/hw/misc/meson.build @@ -XXX,XX +XXX,XX @@ system_ss.add(when: 'CONFIG_IMX', if_true: files( 'imx6_src.c', 'imx6ul_ccm.c', 'imx7_ccm.c', + 'imx7_src.c', 'imx7_gpr.c', 'imx7_snvs.c', 'imx_ccm.c', diff --git a/hw/misc/trace-events b/hw/misc/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/trace-events +++ b/hw/misc/trace-events @@ -XXX,XX +XXX,XX @@ ccm_clock_freq(uint32_t clock, uint32_t freq) "(Clock = %d) = %d" ccm_read_reg(const char *reg_name, uint32_t value) "reg[%s] <= 0x%" PRIx32 ccm_write_reg(const char *reg_name, uint32_t value) "reg[%s] => 0x%" PRIx32 +# imx7_src.c +imx7_src_read(const char *reg_name, uint32_t value) "reg[%s] => 0x%" PRIx32 +imx7_src_write(const char *reg_name, uint32_t value) "reg[%s] <= 0x%" PRIx32 + # iotkit-sysinfo.c iotkit_sysinfo_read(uint64_t offset, uint64_t data, unsigned size) "IoTKit SysInfo read: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %u" iotkit_sysinfo_write(uint64_t offset, uint64_t data, unsigned size) "IoTKit SysInfo write: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %u" -- 2.34.1 The architecture requires (R_TYTWB) that an attempt to return from EL3 when SCR_EL3.{NSE,NS} are {1,0} is an illegal exception return. (This enforces that the CPU can't ever be executing below EL3 with the NSE,NS bits indicating an invalid security state.) We were missing this check; add it. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230807150618.101357-1-peter.maydell@linaro.org --- target/arm/tcg/helper-a64.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/target/arm/tcg/helper-a64.c b/target/arm/tcg/helper-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/helper-a64.c +++ b/target/arm/tcg/helper-a64.c @@ -XXX,XX +XXX,XX @@ void HELPER(exception_return)(CPUARMState *env, uint64_t new_pc) spsr &= ~PSTATE_SS; } + /* + * FEAT_RME forbids return from EL3 with an invalid security state. + * We don't need an explicit check for FEAT_RME here because we enforce + * in scr_write() that you can't set the NSE bit without it. + */ + if (cur_el == 3 && (env->cp15.scr_el3 & (SCR_NS | SCR_NSE)) == SCR_NSE) { + goto illegal_return; + } + new_el = el_from_spsr(spsr); if (new_el == -1) { goto illegal_return; -- 2.34.1 In the m48t59 device we almost always use 64-bit arithmetic when dealing with time_t deltas. The one exception is in set_alarm(), which currently uses a plain 'int' to hold the difference between two time_t values. Switch to int64_t instead to avoid any possible overflow issues. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> --- hw/rtc/m48t59.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/rtc/m48t59.c b/hw/rtc/m48t59.c index XXXXXXX..XXXXXXX 100644 --- a/hw/rtc/m48t59.c +++ b/hw/rtc/m48t59.c @@ -XXX,XX +XXX,XX @@ static void alarm_cb (void *opaque) static void set_alarm(M48t59State *NVRAM) { - int diff; + int64_t diff; if (NVRAM->alrm_timer != NULL) { timer_del(NVRAM->alrm_timer); diff = qemu_timedate_diff(&NVRAM->alarm) - NVRAM->time_offset; -- 2.34.1 In the twl92230 device, use int64_t for the two state fields sec_offset and alm_sec, because we set these to values that are either time_t or differences between two time_t values. These fields aren't saved in vmstate anywhere, so we can safely widen them. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> --- hw/rtc/twl92230.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/hw/rtc/twl92230.c b/hw/rtc/twl92230.c index XXXXXXX..XXXXXXX 100644 --- a/hw/rtc/twl92230.c +++ b/hw/rtc/twl92230.c @@ -XXX,XX +XXX,XX @@ struct MenelausState { struct tm tm; struct tm new; struct tm alm; - int sec_offset; - int alm_sec; + int64_t sec_offset; + int64_t alm_sec; int next_comp; } rtc; uint16_t rtc_next_vmstate; -- 2.34.1 In the aspeed_rtc device we store a difference between two time_t values in an 'int'. This is not really correct when time_t could be 64 bits. Enlarge the field to 'int64_t'. This is a migration compatibility break for the aspeed boards. While we are changing the vmstate, remove the accidental duplicate of the offset field. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Cédric Le Goater <clg@kaod.org> --- include/hw/rtc/aspeed_rtc.h | 2 +- hw/rtc/aspeed_rtc.c | 5 ++--- 2 files changed, 3 insertions(+), 4 deletions(-) diff --git a/include/hw/rtc/aspeed_rtc.h b/include/hw/rtc/aspeed_rtc.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/rtc/aspeed_rtc.h +++ b/include/hw/rtc/aspeed_rtc.h @@ -XXX,XX +XXX,XX @@ struct AspeedRtcState { qemu_irq irq; uint32_t reg[0x18]; - int offset; + int64_t offset; }; diff --git a/hw/rtc/aspeed_rtc.c b/hw/rtc/aspeed_rtc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/rtc/aspeed_rtc.c +++ b/hw/rtc/aspeed_rtc.c @@ -XXX,XX +XXX,XX @@ static const MemoryRegionOps aspeed_rtc_ops = { static const VMStateDescription vmstate_aspeed_rtc = { .name = TYPE_ASPEED_RTC, - .version_id = 1, + .version_id = 2, .fields = (VMStateField[]) { VMSTATE_UINT32_ARRAY(reg, AspeedRtcState, 0x18), - VMSTATE_INT32(offset, AspeedRtcState), - VMSTATE_INT32(offset, AspeedRtcState), + VMSTATE_INT64(offset, AspeedRtcState), VMSTATE_END_OF_LIST() } }; -- 2.34.1 The functions qemu_get_timedate() and qemu_timedate_diff() take and return a time offset as an integer. Coverity points out that means that when an RTC device implementation holds an offset as a time_t, as the m48t59 does, the time_t will get truncated. (CID 1507157, 1517772). The functions work with time_t internally, so make them use that type in their APIs. Note that this won't help any Y2038 issues where either the device model itself is keeping the offset in a 32-bit integer, or where the hardware under emulation has Y2038 or other rollover problems. If we missed any cases of the former then hopefully Coverity will warn us about them since after this patch we'd be truncating a time_t in assignments from qemu_timedate_diff().) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> --- include/sysemu/rtc.h | 4 ++-- softmmu/rtc.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/include/sysemu/rtc.h b/include/sysemu/rtc.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/rtc.h +++ b/include/sysemu/rtc.h @@ -XXX,XX +XXX,XX @@ * The behaviour of the clock whose value this function returns will * depend on the -rtc command line option passed by the user. */ -void qemu_get_timedate(struct tm *tm, int offset); +void qemu_get_timedate(struct tm *tm, time_t offset); /** * qemu_timedate_diff: Return difference between a struct tm and the RTC @@ -XXX,XX +XXX,XX @@ void qemu_get_timedate(struct tm *tm, int offset); * a timestamp one hour further ahead than the current RTC time * then this function will return 3600. */ -int qemu_timedate_diff(struct tm *tm); +time_t qemu_timedate_diff(struct tm *tm); #endif diff --git a/softmmu/rtc.c b/softmmu/rtc.c index XXXXXXX..XXXXXXX 100644 --- a/softmmu/rtc.c +++ b/softmmu/rtc.c @@ -XXX,XX +XXX,XX @@ static time_t qemu_ref_timedate(QEMUClockType clock) return value; } -void qemu_get_timedate(struct tm *tm, int offset) +void qemu_get_timedate(struct tm *tm, time_t offset) { time_t ti = qemu_ref_timedate(rtc_clock); @@ -XXX,XX +XXX,XX @@ void qemu_get_timedate(struct tm *tm, int offset) } } -int qemu_timedate_diff(struct tm *tm) +time_t qemu_timedate_diff(struct tm *tm) { time_t seconds; -- 2.34.1 Where architecturally one ARM_FEATURE_X flag implies another ARM_FEATURE_Y, we allow the CPU init function to only set X, and then set Y for it. Currently we do this in two places -- we set a few flags in arm_cpu_post_init() because we need them to decide which properties to create on the CPU object, and then we do the rest in arm_cpu_realizefn(). However, this is fragile, because it's easy to add a new property and not notice that this means that an X-implies-Y check now has to move from realize to post-init. As a specific example, the pmsav7-dregion property is conditional on ARM_FEATURE_PMSA && ARM_FEATURE_V7, which means it won't appear on the Cortex-M33 and -M55, because they set ARM_FEATURE_V8 and rely on V8-implies-V7, which doesn't happen until the realizefn. Move all of these X-implies-Y checks into a new function, which we call at the top of arm_cpu_post_init(), so the feature bits are available at that point. This does now give us the reverse issue, that if there's a feature bit which is enabled or disabled by the setting of a property then then X-implies-Y features that are dependent on that property need to be in realize, not in this new function. But the only one of those is the "EL3 implies VBAR" which is already in the right place, so putting things this way round seems better to me. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230724174335.2150499-2-peter.maydell@linaro.org --- target/arm/cpu.c | 179 +++++++++++++++++++++++++---------------------- 1 file changed, 97 insertions(+), 82 deletions(-) diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ unsigned int gt_cntfrq_period_ns(ARMCPU *cpu) NANOSECONDS_PER_SECOND / cpu->gt_cntfrq_hz : 1; } +static void arm_cpu_propagate_feature_implications(ARMCPU *cpu) +{ + CPUARMState *env = &cpu->env; + bool no_aa32 = false; + + /* + * Some features automatically imply others: set the feature + * bits explicitly for these cases. + */ + + if (arm_feature(env, ARM_FEATURE_M)) { + set_feature(env, ARM_FEATURE_PMSA); + } + + if (arm_feature(env, ARM_FEATURE_V8)) { + if (arm_feature(env, ARM_FEATURE_M)) { + set_feature(env, ARM_FEATURE_V7); + } else { + set_feature(env, ARM_FEATURE_V7VE); + } + } + + /* + * There exist AArch64 cpus without AArch32 support. When KVM + * queries ID_ISAR0_EL1 on such a host, the value is UNKNOWN. + * Similarly, we cannot check ID_AA64PFR0 without AArch64 support. + * As a general principle, we also do not make ID register + * consistency checks anywhere unless using TCG, because only + * for TCG would a consistency-check failure be a QEMU bug. + */ + if (arm_feature(&cpu->env, ARM_FEATURE_AARCH64)) { + no_aa32 = !cpu_isar_feature(aa64_aa32, cpu); + } + + if (arm_feature(env, ARM_FEATURE_V7VE)) { + /* + * v7 Virtualization Extensions. In real hardware this implies + * EL2 and also the presence of the Security Extensions. + * For QEMU, for backwards-compatibility we implement some + * CPUs or CPU configs which have no actual EL2 or EL3 but do + * include the various other features that V7VE implies. + * Presence of EL2 itself is ARM_FEATURE_EL2, and of the + * Security Extensions is ARM_FEATURE_EL3. + */ + assert(!tcg_enabled() || no_aa32 || + cpu_isar_feature(aa32_arm_div, cpu)); + set_feature(env, ARM_FEATURE_LPAE); + set_feature(env, ARM_FEATURE_V7); + } + if (arm_feature(env, ARM_FEATURE_V7)) { + set_feature(env, ARM_FEATURE_VAPA); + set_feature(env, ARM_FEATURE_THUMB2); + set_feature(env, ARM_FEATURE_MPIDR); + if (!arm_feature(env, ARM_FEATURE_M)) { + set_feature(env, ARM_FEATURE_V6K); + } else { + set_feature(env, ARM_FEATURE_V6); + } + + /* + * Always define VBAR for V7 CPUs even if it doesn't exist in + * non-EL3 configs. This is needed by some legacy boards. + */ + set_feature(env, ARM_FEATURE_VBAR); + } + if (arm_feature(env, ARM_FEATURE_V6K)) { + set_feature(env, ARM_FEATURE_V6); + set_feature(env, ARM_FEATURE_MVFR); + } + if (arm_feature(env, ARM_FEATURE_V6)) { + set_feature(env, ARM_FEATURE_V5); + if (!arm_feature(env, ARM_FEATURE_M)) { + assert(!tcg_enabled() || no_aa32 || + cpu_isar_feature(aa32_jazelle, cpu)); + set_feature(env, ARM_FEATURE_AUXCR); + } + } + if (arm_feature(env, ARM_FEATURE_V5)) { + set_feature(env, ARM_FEATURE_V4T); + } + if (arm_feature(env, ARM_FEATURE_LPAE)) { + set_feature(env, ARM_FEATURE_V7MP); + } + if (arm_feature(env, ARM_FEATURE_CBAR_RO)) { + set_feature(env, ARM_FEATURE_CBAR); + } + if (arm_feature(env, ARM_FEATURE_THUMB2) && + !arm_feature(env, ARM_FEATURE_M)) { + set_feature(env, ARM_FEATURE_THUMB_DSP); + } +} + void arm_cpu_post_init(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); - /* M profile implies PMSA. We have to do this here rather than - * in realize with the other feature-implication checks because - * we look at the PMSA bit to see if we should add some properties. + /* + * Some features imply others. Figure this out now, because we + * are going to look at the feature bits in deciding which + * properties to add. */ - if (arm_feature(&cpu->env, ARM_FEATURE_M)) { - set_feature(&cpu->env, ARM_FEATURE_PMSA); - } + arm_cpu_propagate_feature_implications(cpu); if (arm_feature(&cpu->env, ARM_FEATURE_CBAR) || arm_feature(&cpu->env, ARM_FEATURE_CBAR_RO)) { @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) CPUARMState *env = &cpu->env; int pagebits; Error *local_err = NULL; - bool no_aa32 = false; /* Use pc-relative instructions in system-mode */ #ifndef CONFIG_USER_ONLY @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) cpu->isar.id_isar3 = u; } - /* Some features automatically imply others: */ - if (arm_feature(env, ARM_FEATURE_V8)) { - if (arm_feature(env, ARM_FEATURE_M)) { - set_feature(env, ARM_FEATURE_V7); - } else { - set_feature(env, ARM_FEATURE_V7VE); - } - } - - /* - * There exist AArch64 cpus without AArch32 support. When KVM - * queries ID_ISAR0_EL1 on such a host, the value is UNKNOWN. - * Similarly, we cannot check ID_AA64PFR0 without AArch64 support. - * As a general principle, we also do not make ID register - * consistency checks anywhere unless using TCG, because only - * for TCG would a consistency-check failure be a QEMU bug. - */ - if (arm_feature(&cpu->env, ARM_FEATURE_AARCH64)) { - no_aa32 = !cpu_isar_feature(aa64_aa32, cpu); - } - - if (arm_feature(env, ARM_FEATURE_V7VE)) { - /* v7 Virtualization Extensions. In real hardware this implies - * EL2 and also the presence of the Security Extensions. - * For QEMU, for backwards-compatibility we implement some - * CPUs or CPU configs which have no actual EL2 or EL3 but do - * include the various other features that V7VE implies. - * Presence of EL2 itself is ARM_FEATURE_EL2, and of the - * Security Extensions is ARM_FEATURE_EL3. - */ - assert(!tcg_enabled() || no_aa32 || - cpu_isar_feature(aa32_arm_div, cpu)); - set_feature(env, ARM_FEATURE_LPAE); - set_feature(env, ARM_FEATURE_V7); - } - if (arm_feature(env, ARM_FEATURE_V7)) { - set_feature(env, ARM_FEATURE_VAPA); - set_feature(env, ARM_FEATURE_THUMB2); - set_feature(env, ARM_FEATURE_MPIDR); - if (!arm_feature(env, ARM_FEATURE_M)) { - set_feature(env, ARM_FEATURE_V6K); - } else { - set_feature(env, ARM_FEATURE_V6); - } - - /* Always define VBAR for V7 CPUs even if it doesn't exist in - * non-EL3 configs. This is needed by some legacy boards. - */ - set_feature(env, ARM_FEATURE_VBAR); - } - if (arm_feature(env, ARM_FEATURE_V6K)) { - set_feature(env, ARM_FEATURE_V6); - set_feature(env, ARM_FEATURE_MVFR); - } - if (arm_feature(env, ARM_FEATURE_V6)) { - set_feature(env, ARM_FEATURE_V5); - if (!arm_feature(env, ARM_FEATURE_M)) { - assert(!tcg_enabled() || no_aa32 || - cpu_isar_feature(aa32_jazelle, cpu)); - set_feature(env, ARM_FEATURE_AUXCR); - } - } - if (arm_feature(env, ARM_FEATURE_V5)) { - set_feature(env, ARM_FEATURE_V4T); - } - if (arm_feature(env, ARM_FEATURE_LPAE)) { - set_feature(env, ARM_FEATURE_V7MP); - } - if (arm_feature(env, ARM_FEATURE_CBAR_RO)) { - set_feature(env, ARM_FEATURE_CBAR); - } - if (arm_feature(env, ARM_FEATURE_THUMB2) && - !arm_feature(env, ARM_FEATURE_M)) { - set_feature(env, ARM_FEATURE_THUMB_DSP); - } /* * We rely on no XScale CPU having VFP so we can use the same bits in the -- 2.34.1 M-profile CPUs generally allow configuration of the number of MPU regions that they have. We don't currently model this, so our implementations of some of the board models provide CPUs with the wrong number of regions. RTOSes like Zephyr that hardcode the expected number of regions may therefore not run on the model if they are set up to run on real hardware. Add properties mpu-ns-regions and mpu-s-regions to the ARMV7M object, matching the ability of hardware to configure the number of Secure and NonSecure regions separately. Our actual CPU implementation doesn't currently support that, and it happens that none of the MPS boards we model set the number of regions differently for Secure vs NonSecure, so we provide an interface to the boards and SoCs that won't need to change if we ever do add that functionality in future, but make it an error to configure the two properties to different values. (The property name on the CPU is the somewhat misnamed-for-M-profile "pmsav7-dregion", so we don't follow that naming convention for the properties here. The TRM doesn't say what the CPU configuration variable names are, so we pick something, and follow the lowercase convention we already have for properties here.) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-id: 20230724174335.2150499-3-peter.maydell@linaro.org --- include/hw/arm/armv7m.h | 8 ++++++++ hw/arm/armv7m.c | 21 +++++++++++++++++++++ 2 files changed, 29 insertions(+) diff --git a/include/hw/arm/armv7m.h b/include/hw/arm/armv7m.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/armv7m.h +++ b/include/hw/arm/armv7m.h @@ -XXX,XX +XXX,XX @@ OBJECT_DECLARE_SIMPLE_TYPE(ARMv7MState, ARMV7M) * + Property "vfp": enable VFP (forwarded to CPU object) * + Property "dsp": enable DSP (forwarded to CPU object) * + Property "enable-bitband": expose bitbanded IO + * + Property "mpu-ns-regions": number of Non-Secure MPU regions (forwarded + * to CPU object pmsav7-dregion property; default is whatever the default + * for the CPU is) + * + Property "mpu-s-regions": number of Secure MPU regions (default is + * whatever the default for the CPU is; must currently be set to the same + * value as mpu-ns-regions if the CPU implements the Security Extension) * + Clock input "refclk" is the external reference clock for the systick timers * + Clock input "cpuclk" is the main CPU clock */ @@ -XXX,XX +XXX,XX @@ struct ARMv7MState { Object *idau; uint32_t init_svtor; uint32_t init_nsvtor; + uint32_t mpu_ns_regions; + uint32_t mpu_s_regions; bool enable_bitband; bool start_powered_off; bool vfp; diff --git a/hw/arm/armv7m.c b/hw/arm/armv7m.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/armv7m.c +++ b/hw/arm/armv7m.c @@ -XXX,XX +XXX,XX @@ static void armv7m_realize(DeviceState *dev, Error **errp) } } + /* + * Real M-profile hardware can be configured with a different number of + * MPU regions for Secure vs NonSecure. QEMU's CPU implementation doesn't + * support that yet, so catch attempts to select that. + */ + if (arm_feature(&s->cpu->env, ARM_FEATURE_M_SECURITY) && + s->mpu_ns_regions != s->mpu_s_regions) { + error_setg(errp, + "mpu-ns-regions and mpu-s-regions properties must have the same value"); + return; + } + if (s->mpu_ns_regions != UINT_MAX && + object_property_find(OBJECT(s->cpu), "pmsav7-dregion")) { + if (!object_property_set_uint(OBJECT(s->cpu), "pmsav7-dregion", + s->mpu_ns_regions, errp)) { + return; + } + } + /* * Tell the CPU where the NVIC is; it will fail realize if it doesn't * have one. Similarly, tell the NVIC where its CPU is. @@ -XXX,XX +XXX,XX @@ static Property armv7m_properties[] = { false), DEFINE_PROP_BOOL("vfp", ARMv7MState, vfp, true), DEFINE_PROP_BOOL("dsp", ARMv7MState, dsp, true), + DEFINE_PROP_UINT32("mpu-ns-regions", ARMv7MState, mpu_ns_regions, UINT_MAX), + DEFINE_PROP_UINT32("mpu-s-regions", ARMv7MState, mpu_s_regions, UINT_MAX), DEFINE_PROP_END_OF_LIST(), }; -- 2.34.1 The IoTKit, SSE200 and SSE300 all default to 8 MPU regions. The MPS2/MPS3 FPGA images don't override these except in the case of AN547, which uses 16 MPU regions. Define properties on the ARMSSE object for the MPU regions (using the same names as the documented RTL configuration settings, and following the pattern we already have for this device of using all-caps names as the RTL does), and set them in the board code. We don't actually need to override the default except on AN547, but it's simpler code to have the board code set them always rather than tracking which board subtypes want to set them to a non-default value separately from what that value is. Tho overall effect is that for mps2-an505, mps2-an521 and mps3-an524 we now correctly use 8 MPU regions, while mps3-an547 stays at its current 16 regions. It's possible some guest code wrongly depended on the previous incorrectly modeled number of memory regions. (Such guest code should ideally check the number of regions via the MPU_TYPE register.) The old behaviour can be obtained with additional -global arguments to QEMU: For mps2-an521 and mps2-an524: -global sse-200.CPU0_MPU_NS=16 -global sse-200.CPU0_MPU_S=16 -global sse-200.CPU1_MPU_NS=16 -global sse-200.CPU1_MPU_S=16 For mps2-an505: -global sse-200.CPU0_MPU_NS=16 -global sse-200.CPU0_MPU_S=16 NB that the way the implementation allows this use of -global is slightly fragile: if the board code explicitly sets the properties on the sse-200 object, this overrides the -global command line option. So we rely on: - the boards that need fixing all happen to use the SSE defaults - we can write the board code to only set the property if it is different from the default, rather than having all boards explicitly set the property - the board that does need to use a non-default value happens to need to set it to the same value (16) we previously used This works, but there are some kinds of refactoring of the mps2-tz.c code that would break the support for -global here. Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1772 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-id: 20230724174335.2150499-4-peter.maydell@linaro.org --- include/hw/arm/armsse.h | 5 +++++ hw/arm/armsse.c | 16 ++++++++++++++++ hw/arm/mps2-tz.c | 29 +++++++++++++++++++++++++++++ 3 files changed, 50 insertions(+) diff --git a/include/hw/arm/armsse.h b/include/hw/arm/armsse.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/armsse.h +++ b/include/hw/arm/armsse.h @@ -XXX,XX +XXX,XX @@ * (matching the hardware) is that for CPU0 in an IoTKit and CPU1 in an * SSE-200 both are present; CPU0 in an SSE-200 has neither. * Since the IoTKit has only one CPU, it does not have the CPU1_* properties. + * + QOM properties "CPU0_MPU_NS", "CPU0_MPU_S", "CPU1_MPU_NS" and "CPU1_MPU_S" + * which set the number of MPU regions on the CPUs. If there is only one + * CPU the CPU1 properties are not present. * + Named GPIO inputs "EXP_IRQ" 0..n are the expansion interrupts for CPU 0, * which are wired to its NVIC lines 32 .. n+32 * + Named GPIO inputs "EXP_CPU1_IRQ" 0..n are the expansion interrupts for @@ -XXX,XX +XXX,XX @@ struct ARMSSE { uint32_t exp_numirq; uint32_t sram_addr_width; uint32_t init_svtor; + uint32_t cpu_mpu_ns[SSE_MAX_CPUS]; + uint32_t cpu_mpu_s[SSE_MAX_CPUS]; bool cpu_fpu[SSE_MAX_CPUS]; bool cpu_dsp[SSE_MAX_CPUS]; }; diff --git a/hw/arm/armsse.c b/hw/arm/armsse.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/armsse.c +++ b/hw/arm/armsse.c @@ -XXX,XX +XXX,XX @@ static Property iotkit_properties[] = { DEFINE_PROP_UINT32("init-svtor", ARMSSE, init_svtor, 0x10000000), DEFINE_PROP_BOOL("CPU0_FPU", ARMSSE, cpu_fpu[0], true), DEFINE_PROP_BOOL("CPU0_DSP", ARMSSE, cpu_dsp[0], true), + DEFINE_PROP_UINT32("CPU0_MPU_NS", ARMSSE, cpu_mpu_ns[0], 8), + DEFINE_PROP_UINT32("CPU0_MPU_S", ARMSSE, cpu_mpu_s[0], 8), DEFINE_PROP_END_OF_LIST() }; @@ -XXX,XX +XXX,XX @@ static Property sse200_properties[] = { DEFINE_PROP_BOOL("CPU0_DSP", ARMSSE, cpu_dsp[0], false), DEFINE_PROP_BOOL("CPU1_FPU", ARMSSE, cpu_fpu[1], true), DEFINE_PROP_BOOL("CPU1_DSP", ARMSSE, cpu_dsp[1], true), + DEFINE_PROP_UINT32("CPU0_MPU_NS", ARMSSE, cpu_mpu_ns[0], 8), + DEFINE_PROP_UINT32("CPU0_MPU_S", ARMSSE, cpu_mpu_s[0], 8), + DEFINE_PROP_UINT32("CPU1_MPU_NS", ARMSSE, cpu_mpu_ns[1], 8), + DEFINE_PROP_UINT32("CPU1_MPU_S", ARMSSE, cpu_mpu_s[1], 8), DEFINE_PROP_END_OF_LIST() }; @@ -XXX,XX +XXX,XX @@ static Property sse300_properties[] = { DEFINE_PROP_UINT32("init-svtor", ARMSSE, init_svtor, 0x10000000), DEFINE_PROP_BOOL("CPU0_FPU", ARMSSE, cpu_fpu[0], true), DEFINE_PROP_BOOL("CPU0_DSP", ARMSSE, cpu_dsp[0], true), + DEFINE_PROP_UINT32("CPU0_MPU_NS", ARMSSE, cpu_mpu_ns[0], 8), + DEFINE_PROP_UINT32("CPU0_MPU_S", ARMSSE, cpu_mpu_s[0], 8), DEFINE_PROP_END_OF_LIST() }; @@ -XXX,XX +XXX,XX @@ static void armsse_realize(DeviceState *dev, Error **errp) return; } } + if (!object_property_set_uint(cpuobj, "mpu-ns-regions", + s->cpu_mpu_ns[i], errp)) { + return; + } + if (!object_property_set_uint(cpuobj, "mpu-s-regions", + s->cpu_mpu_s[i], errp)) { + return; + } if (i > 0) { memory_region_add_subregion_overlap(&s->cpu_container[i], 0, diff --git a/hw/arm/mps2-tz.c b/hw/arm/mps2-tz.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/mps2-tz.c +++ b/hw/arm/mps2-tz.c @@ -XXX,XX +XXX,XX @@ struct MPS2TZMachineClass { int uart_overflow_irq; /* number of the combined UART overflow IRQ */ uint32_t init_svtor; /* init-svtor setting for SSE */ uint32_t sram_addr_width; /* SRAM_ADDR_WIDTH setting for SSE */ + uint32_t cpu0_mpu_ns; /* CPU0_MPU_NS setting for SSE */ + uint32_t cpu0_mpu_s; /* CPU0_MPU_S setting for SSE */ + uint32_t cpu1_mpu_ns; /* CPU1_MPU_NS setting for SSE */ + uint32_t cpu1_mpu_s; /* CPU1_MPU_S setting for SSE */ const RAMInfo *raminfo; const char *armsse_type; uint32_t boot_ram_size; /* size of ram at address 0; 0 == find in raminfo */ @@ -XXX,XX +XXX,XX @@ OBJECT_DECLARE_TYPE(MPS2TZMachineState, MPS2TZMachineClass, MPS2TZ_MACHINE) #define MPS3_DDR_SIZE (2 * GiB) #endif +/* For cpu{0,1}_mpu_{ns,s}, means "leave at SSE's default value" */ +#define MPU_REGION_DEFAULT UINT32_MAX + static const uint32_t an505_oscclk[] = { 40000000, 24580000, @@ -XXX,XX +XXX,XX @@ static void mps2tz_common_init(MachineState *machine) OBJECT(system_memory), &error_abort); qdev_prop_set_uint32(iotkitdev, "EXP_NUMIRQ", mmc->numirq); qdev_prop_set_uint32(iotkitdev, "init-svtor", mmc->init_svtor); + if (mmc->cpu0_mpu_ns != MPU_REGION_DEFAULT) { + qdev_prop_set_uint32(iotkitdev, "CPU0_MPU_NS", mmc->cpu0_mpu_ns); + } + if (mmc->cpu0_mpu_s != MPU_REGION_DEFAULT) { + qdev_prop_set_uint32(iotkitdev, "CPU0_MPU_S", mmc->cpu0_mpu_s); + } + if (object_property_find(OBJECT(iotkitdev), "CPU1_MPU_NS")) { + if (mmc->cpu1_mpu_ns != MPU_REGION_DEFAULT) { + qdev_prop_set_uint32(iotkitdev, "CPU1_MPU_NS", mmc->cpu1_mpu_ns); + } + if (mmc->cpu1_mpu_s != MPU_REGION_DEFAULT) { + qdev_prop_set_uint32(iotkitdev, "CPU1_MPU_S", mmc->cpu1_mpu_s); + } + } qdev_prop_set_uint32(iotkitdev, "SRAM_ADDR_WIDTH", mmc->sram_addr_width); qdev_connect_clock_in(iotkitdev, "MAINCLK", mms->sysclk); qdev_connect_clock_in(iotkitdev, "S32KCLK", mms->s32kclk); @@ -XXX,XX +XXX,XX @@ static void mps2tz_class_init(ObjectClass *oc, void *data) { MachineClass *mc = MACHINE_CLASS(oc); IDAUInterfaceClass *iic = IDAU_INTERFACE_CLASS(oc); + MPS2TZMachineClass *mmc = MPS2TZ_MACHINE_CLASS(oc); mc->init = mps2tz_common_init; mc->reset = mps2_machine_reset; iic->check = mps2_tz_idau_check; + + /* Most machines leave these at the SSE defaults */ + mmc->cpu0_mpu_ns = MPU_REGION_DEFAULT; + mmc->cpu0_mpu_s = MPU_REGION_DEFAULT; + mmc->cpu1_mpu_ns = MPU_REGION_DEFAULT; + mmc->cpu1_mpu_s = MPU_REGION_DEFAULT; } static void mps2tz_set_default_ram_info(MPS2TZMachineClass *mmc) @@ -XXX,XX +XXX,XX @@ static void mps3tz_an547_class_init(ObjectClass *oc, void *data) mmc->numirq = 96; mmc->uart_overflow_irq = 48; mmc->init_svtor = 0x00000000; + mmc->cpu0_mpu_s = mmc->cpu0_mpu_ns = 16; mmc->sram_addr_width = 21; mmc->raminfo = an547_raminfo; mmc->armsse_type = TYPE_SSE300; -- 2.34.1 Hi; here's this week's arm pullreq. Mostly this is my work on FEAT_MOPS and FEAT_HBC, but there are some other bits and pieces in there too, including a recent set of elf2dmp patches. thanks -- PMM The following changes since commit 55394dcbec8f0c29c30e792c102a0edd50a52bf4: Merge tag 'pull-loongarch-20230920' of https://gitlab.com/gaosong/qemu into staging (2023-09-20 13:56:18 -0400) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20230921 for you to fetch changes up to 231f6a7d66254a58bedbee458591b780e0a507b1: elf2dmp: rework PDB_STREAM_INDEXES::segments obtaining (2023-09-21 16:13:54 +0100) ---------------------------------------------------------------- target-arm queue: * target/m68k: Add URL to semihosting spec * docs/devel/loads-stores: Fix git grep regexes * hw/arm/boot: Set SCR_EL3.FGTEn when booting kernel * linux-user: Correct SME feature names reported in cpuinfo * linux-user: Add missing arm32 hwcaps * Don't skip MTE checks for LDRT/STRT at EL0 * Implement FEAT_HBC * Implement FEAT_MOPS * audio/jackaudio: Avoid dynamic stack allocation * sbsa-ref: add non-secure EL2 virtual timer * elf2dmp: improve Win2022, Win11 and large dumps ---------------------------------------------------------------- Fabian Vogt (1): hw/arm/boot: Set SCR_EL3.FGTEn when booting kernel Marcin Juszkiewicz (1): sbsa-ref: add non-secure EL2 virtual timer Peter Maydell (23): target/m68k: Add URL to semihosting spec docs/devel/loads-stores: Fix git grep regexes linux-user/elfload.c: Correct SME feature names reported in cpuinfo linux-user/elfload.c: Add missing arm and arm64 hwcap values linux-user/elfload.c: Report previously missing arm32 hwcaps target/arm: Update AArch64 ID register field definitions target/arm: Update user-mode ID reg mask values target/arm: Implement FEAT_HBC target/arm: Remove unused allocation_tag_mem() argument target/arm: Don't skip MTE checks for LDRT/STRT at EL0 target/arm: Implement FEAT_MOPS enable bits target/arm: Pass unpriv bool to get_a64_user_mem_index() target/arm: Define syndrome function for MOPS exceptions target/arm: New function allocation_tag_mem_probe() target/arm: Implement MTE tag-checking functions for FEAT_MOPS target/arm: Implement the SET* instructions target/arm: Define new TB flag for ATA0 target/arm: Implement the SETG* instructions target/arm: Implement MTE tag-checking functions for FEAT_MOPS copies target/arm: Implement the CPY* instructions target/arm: Enable FEAT_MOPS for CPU 'max' audio/jackaudio: Avoid dynamic stack allocation in qjack_client_init audio/jackaudio: Avoid dynamic stack allocation in qjack_process() Viktor Prutyanov (5): elf2dmp: replace PE export name check with PDB name check elf2dmp: introduce physical block alignment elf2dmp: introduce merging of physical memory runs elf2dmp: use Linux mmap with MAP_NORESERVE when possible elf2dmp: rework PDB_STREAM_INDEXES::segments obtaining docs/devel/loads-stores.rst | 40 +- docs/system/arm/emulation.rst | 2 + contrib/elf2dmp/addrspace.h | 1 + contrib/elf2dmp/pdb.h | 2 +- contrib/elf2dmp/qemu_elf.h | 2 + target/arm/cpu.h | 35 ++ target/arm/internals.h | 55 +++ target/arm/syndrome.h | 12 + target/arm/tcg/helper-a64.h | 14 + target/arm/tcg/translate.h | 4 +- target/arm/tcg/a64.decode | 38 +- audio/jackaudio.c | 21 +- contrib/elf2dmp/addrspace.c | 31 +- contrib/elf2dmp/main.c | 154 ++++---- contrib/elf2dmp/pdb.c | 15 +- contrib/elf2dmp/qemu_elf.c | 68 +++- hw/arm/boot.c | 4 + hw/arm/sbsa-ref.c | 2 + linux-user/elfload.c | 72 +++- target/arm/helper.c | 39 +- target/arm/tcg/cpu64.c | 5 + target/arm/tcg/helper-a64.c | 878 +++++++++++++++++++++++++++++++++++++++++ target/arm/tcg/hflags.c | 21 + target/arm/tcg/mte_helper.c | 281 +++++++++++-- target/arm/tcg/translate-a64.c | 164 +++++++- target/m68k/m68k-semi.c | 4 + tests/tcg/aarch64/sysregs.c | 4 +- 27 files changed, 1768 insertions(+), 200 deletions(-) The spec for m68k semihosting is documented in the libgloss sources. Add a comment with the URL for it, as we already have for nios2 semihosting. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-id: 20230801154451.3505492-1-peter.maydell@linaro.org --- target/m68k/m68k-semi.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/target/m68k/m68k-semi.c b/target/m68k/m68k-semi.c index XXXXXXX..XXXXXXX 100644 --- a/target/m68k/m68k-semi.c +++ b/target/m68k/m68k-semi.c @@ -XXX,XX +XXX,XX @@ * * You should have received a copy of the GNU General Public License * along with this program; if not, see <http://www.gnu.org/licenses/>. + * + * The semihosting protocol implemented here is described in the + * libgloss sources: + * https://sourceware.org/git/?p=newlib-cygwin.git;a=blob;f=libgloss/m68k/m68k-semi.txt;hb=HEAD */ #include "qemu/osdep.h" -- 2.34.1 The loads-and-stores documentation includes git grep regexes to find occurrences of the various functions. Some of these regexes have errors, typically failing to escape the '?', '(' and ')' when they should be metacharacters (since these are POSIX basic REs). We also weren't consistent about whether to have a ':' on the end of the line introducing the list of regexes in each section. Fix the errors. The following shell rune will complain about any REs in the file which don't have any matches in the codebase: for re in $(sed -ne 's/ - ``\(\\<.*\)``/\1/p' docs/devel/loads-stores.rst); do git grep -q "$re" || echo "no matches for re $re"; done Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-id: 20230904161703.3996734-1-peter.maydell@linaro.org --- docs/devel/loads-stores.rst | 40 ++++++++++++++++++------------------- 1 file changed, 20 insertions(+), 20 deletions(-) diff --git a/docs/devel/loads-stores.rst b/docs/devel/loads-stores.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/devel/loads-stores.rst +++ b/docs/devel/loads-stores.rst @@ -XXX,XX +XXX,XX @@ which stores ``val`` to ``ptr`` as an ``{endian}`` order value of size ``sz`` bytes. -Regexes for git grep +Regexes for git grep: - ``\<ld[us]\?[bwlq]\(_[hbl]e\)\?_p\>`` - ``\<st[bwlq]\(_[hbl]e\)\?_p\>`` - ``\<st24\(_[hbl]e\)\?_p\>`` - - ``\<ldn_\([hbl]e\)?_p\>`` - - ``\<stn_\([hbl]e\)?_p\>`` + - ``\<ldn_\([hbl]e\)\?_p\>`` + - ``\<stn_\([hbl]e\)\?_p\>`` ``cpu_{ld,st}*_mmu`` ~~~~~~~~~~~~~~~~~~~~ @@ -XXX,XX +XXX,XX @@ store: ``cpu_st{size}{end}_mmu(env, ptr, val, oi, retaddr)`` - ``_le`` : little endian Regexes for git grep: - - ``\<cpu_ld[bwlq](_[bl]e)\?_mmu\>`` - - ``\<cpu_st[bwlq](_[bl]e)\?_mmu\>`` + - ``\<cpu_ld[bwlq]\(_[bl]e\)\?_mmu\>`` + - ``\<cpu_st[bwlq]\(_[bl]e\)\?_mmu\>`` ``cpu_{ld,st}*_mmuidx_ra`` @@ -XXX,XX +XXX,XX @@ store: ``cpu_st{size}{end}_mmuidx_ra(env, ptr, val, mmuidx, retaddr)`` - ``_le`` : little endian Regexes for git grep: - - ``\<cpu_ld[us]\?[bwlq](_[bl]e)\?_mmuidx_ra\>`` - - ``\<cpu_st[bwlq](_[bl]e)\?_mmuidx_ra\>`` + - ``\<cpu_ld[us]\?[bwlq]\(_[bl]e\)\?_mmuidx_ra\>`` + - ``\<cpu_st[bwlq]\(_[bl]e\)\?_mmuidx_ra\>`` ``cpu_{ld,st}*_data_ra`` ~~~~~~~~~~~~~~~~~~~~~~~~ @@ -XXX,XX +XXX,XX @@ store: ``cpu_st{size}{end}_data_ra(env, ptr, val, ra)`` - ``_le`` : little endian Regexes for git grep: - - ``\<cpu_ld[us]\?[bwlq](_[bl]e)\?_data_ra\>`` - - ``\<cpu_st[bwlq](_[bl]e)\?_data_ra\>`` + - ``\<cpu_ld[us]\?[bwlq]\(_[bl]e\)\?_data_ra\>`` + - ``\<cpu_st[bwlq]\(_[bl]e\)\?_data_ra\>`` ``cpu_{ld,st}*_data`` ~~~~~~~~~~~~~~~~~~~~~ @@ -XXX,XX +XXX,XX @@ store: ``cpu_st{size}{end}_data(env, ptr, val)`` - ``_be`` : big endian - ``_le`` : little endian -Regexes for git grep - - ``\<cpu_ld[us]\?[bwlq](_[bl]e)\?_data\>`` - - ``\<cpu_st[bwlq](_[bl]e)\?_data\+\>`` +Regexes for git grep: + - ``\<cpu_ld[us]\?[bwlq]\(_[bl]e\)\?_data\>`` + - ``\<cpu_st[bwlq]\(_[bl]e\)\?_data\+\>`` ``cpu_ld*_code`` ~~~~~~~~~~~~~~~~ @@ -XXX,XX +XXX,XX @@ swap: ``translator_ld{sign}{size}_swap(env, ptr, swap)`` - ``l`` : 32 bits - ``q`` : 64 bits -Regexes for git grep +Regexes for git grep: - ``\<translator_ld[us]\?[bwlq]\(_swap\)\?\>`` ``helper_{ld,st}*_mmu`` @@ -XXX,XX +XXX,XX @@ store: ``helper_{size}_mmu(env, addr, val, opindex, retaddr)`` - ``l`` : 32 bits - ``q`` : 64 bits -Regexes for git grep +Regexes for git grep: - ``\<helper_ld[us]\?[bwlq]_mmu\>`` - ``\<helper_st[bwlq]_mmu\>`` @@ -XXX,XX +XXX,XX @@ succeeded using a MemTxResult return code. The ``_{endian}`` suffix is omitted for byte accesses. -Regexes for git grep +Regexes for git grep: - ``\<address_space_\(read\|write\|rw\)\>`` - ``\<address_space_ldu\?[bwql]\(_[lb]e\)\?\>`` - ``\<address_space_st[bwql]\(_[lb]e\)\?\>`` @@ -XXX,XX +XXX,XX @@ Note that portions of the write which attempt to write data to a device will be silently ignored -- only real RAM and ROM will be written to. -Regexes for git grep +Regexes for git grep: - ``address_space_write_rom`` ``{ld,st}*_phys`` @@ -XXX,XX +XXX,XX @@ device doing the access has no way to report such an error. The ``_{endian}_`` infix is omitted for byte accesses. -Regexes for git grep +Regexes for git grep: - ``\<ldu\?[bwlq]\(_[bl]e\)\?_phys\>`` - ``\<st[bwlq]\(_[bl]e\)\?_phys\>`` @@ -XXX,XX +XXX,XX @@ For new code they are better avoided: ``cpu_physical_memory_rw`` -Regexes for git grep +Regexes for git grep: - ``\<cpu_physical_memory_\(read\|write\|rw\)\>`` ``cpu_memory_rw_debug`` @@ -XXX,XX +XXX,XX @@ make sure our existing code is doing things correctly. ``dma_memory_rw`` -Regexes for git grep +Regexes for git grep: - ``\<dma_memory_\(read\|write\|rw\)\>`` - ``\<ldu\?[bwlq]\(_[bl]e\)\?_dma\>`` - ``\<st[bwlq]\(_[bl]e\)\?_dma\>`` @@ -XXX,XX +XXX,XX @@ correct address space for that device. The ``_{endian}_`` infix is omitted for byte accesses. -Regexes for git grep +Regexes for git grep: - ``\<pci_dma_\(read\|write\|rw\)\>`` - ``\<ldu\?[bwlq]\(_[bl]e\)\?_pci_dma\>`` - ``\<st[bwlq]\(_[bl]e\)\?_pci_dma\>`` -- 2.34.1 From: Fabian Vogt <fvogt@suse.de> Just like d7ef5e16a17c sets SCR_EL3.HXEn for FEAT_HCX, this commit handles SCR_EL3.FGTEn for FEAT_FGT: When we direct boot a kernel on a CPU which emulates EL3, we need to set up the EL3 system registers as the Linux kernel documentation specifies: https://www.kernel.org/doc/Documentation/arm64/booting.rst > For CPUs with the Fine Grained Traps (FEAT_FGT) extension present: > - If EL3 is present and the kernel is entered at EL2: > - SCR_EL3.FGTEn (bit 27) must be initialised to 0b1. Cc: qemu-stable@nongnu.org Signed-off-by: Fabian Vogt <fvogt@suse.de> Message-id: 4831384.GXAFRqVoOG@linux-e202.suse.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/boot.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/hw/arm/boot.c b/hw/arm/boot.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/boot.c +++ b/hw/arm/boot.c @@ -XXX,XX +XXX,XX @@ static void do_cpu_reset(void *opaque) if (cpu_isar_feature(aa64_hcx, cpu)) { env->cp15.scr_el3 |= SCR_HXEN; } + if (cpu_isar_feature(aa64_fgt, cpu)) { + env->cp15.scr_el3 |= SCR_FGTEN; + } + /* AArch64 kernels never boot in secure mode */ assert(!info->secure_boot); /* This hook is only supported for AArch32 currently: -- 2.34.1 Some of the names we use for CPU features in linux-user's dummy /proc/cpuinfo don't match the strings in the real kernel in arch/arm64/kernel/cpuinfo.c. Specifically, the SME related features have an underscore in the HWCAP_FOO define name, but (like the SVE ones) they do not have an underscore in the string in cpuinfo. Correct the errors. Fixes: a55b9e7226708 ("linux-user: Emulate /proc/cpuinfo on aarch64 and arm") Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> --- linux-user/elfload.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/linux-user/elfload.c b/linux-user/elfload.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/elfload.c +++ b/linux-user/elfload.c @@ -XXX,XX +XXX,XX @@ const char *elf_hwcap2_str(uint32_t bit) [__builtin_ctz(ARM_HWCAP2_A64_RPRES )] = "rpres", [__builtin_ctz(ARM_HWCAP2_A64_MTE3 )] = "mte3", [__builtin_ctz(ARM_HWCAP2_A64_SME )] = "sme", - [__builtin_ctz(ARM_HWCAP2_A64_SME_I16I64 )] = "sme_i16i64", - [__builtin_ctz(ARM_HWCAP2_A64_SME_F64F64 )] = "sme_f64f64", - [__builtin_ctz(ARM_HWCAP2_A64_SME_I8I32 )] = "sme_i8i32", - [__builtin_ctz(ARM_HWCAP2_A64_SME_F16F32 )] = "sme_f16f32", - [__builtin_ctz(ARM_HWCAP2_A64_SME_B16F32 )] = "sme_b16f32", - [__builtin_ctz(ARM_HWCAP2_A64_SME_F32F32 )] = "sme_f32f32", - [__builtin_ctz(ARM_HWCAP2_A64_SME_FA64 )] = "sme_fa64", + [__builtin_ctz(ARM_HWCAP2_A64_SME_I16I64 )] = "smei16i64", + [__builtin_ctz(ARM_HWCAP2_A64_SME_F64F64 )] = "smef64f64", + [__builtin_ctz(ARM_HWCAP2_A64_SME_I8I32 )] = "smei8i32", + [__builtin_ctz(ARM_HWCAP2_A64_SME_F16F32 )] = "smef16f32", + [__builtin_ctz(ARM_HWCAP2_A64_SME_B16F32 )] = "smeb16f32", + [__builtin_ctz(ARM_HWCAP2_A64_SME_F32F32 )] = "smef32f32", + [__builtin_ctz(ARM_HWCAP2_A64_SME_FA64 )] = "smefa64", }; return bit < ARRAY_SIZE(hwcap_str) ? hwcap_str[bit] : NULL; -- 2.34.1 Our lists of Arm 32 and 64 bit hwcap values have lagged behind the Linux kernel. Update them to include all the bits defined as of upstream Linux git commit a48fa7efaf1161c1 (in the middle of the kernel 6.6 dev cycle). For 64-bit, we don't yet implement any of the features reported via these hwcap bits. For 32-bit we do in fact already implement them all; we'll add the code to set them in a subsequent commit. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> --- linux-user/elfload.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/linux-user/elfload.c b/linux-user/elfload.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/elfload.c +++ b/linux-user/elfload.c @@ -XXX,XX +XXX,XX @@ enum ARM_HWCAP_ARM_VFPD32 = 1 << 19, ARM_HWCAP_ARM_LPAE = 1 << 20, ARM_HWCAP_ARM_EVTSTRM = 1 << 21, + ARM_HWCAP_ARM_FPHP = 1 << 22, + ARM_HWCAP_ARM_ASIMDHP = 1 << 23, + ARM_HWCAP_ARM_ASIMDDP = 1 << 24, + ARM_HWCAP_ARM_ASIMDFHM = 1 << 25, + ARM_HWCAP_ARM_ASIMDBF16 = 1 << 26, + ARM_HWCAP_ARM_I8MM = 1 << 27, }; enum { @@ -XXX,XX +XXX,XX @@ enum { ARM_HWCAP2_ARM_SHA1 = 1 << 2, ARM_HWCAP2_ARM_SHA2 = 1 << 3, ARM_HWCAP2_ARM_CRC32 = 1 << 4, + ARM_HWCAP2_ARM_SB = 1 << 5, + ARM_HWCAP2_ARM_SSBS = 1 << 6, }; /* The commpage only exists for 32 bit kernels */ @@ -XXX,XX +XXX,XX @@ const char *elf_hwcap_str(uint32_t bit) [__builtin_ctz(ARM_HWCAP_ARM_VFPD32 )] = "vfpd32", [__builtin_ctz(ARM_HWCAP_ARM_LPAE )] = "lpae", [__builtin_ctz(ARM_HWCAP_ARM_EVTSTRM )] = "evtstrm", + [__builtin_ctz(ARM_HWCAP_ARM_FPHP )] = "fphp", + [__builtin_ctz(ARM_HWCAP_ARM_ASIMDHP )] = "asimdhp", + [__builtin_ctz(ARM_HWCAP_ARM_ASIMDDP )] = "asimddp", + [__builtin_ctz(ARM_HWCAP_ARM_ASIMDFHM )] = "asimdfhm", + [__builtin_ctz(ARM_HWCAP_ARM_ASIMDBF16)] = "asimdbf16", + [__builtin_ctz(ARM_HWCAP_ARM_I8MM )] = "i8mm", }; return bit < ARRAY_SIZE(hwcap_str) ? hwcap_str[bit] : NULL; @@ -XXX,XX +XXX,XX @@ const char *elf_hwcap2_str(uint32_t bit) [__builtin_ctz(ARM_HWCAP2_ARM_SHA1 )] = "sha1", [__builtin_ctz(ARM_HWCAP2_ARM_SHA2 )] = "sha2", [__builtin_ctz(ARM_HWCAP2_ARM_CRC32)] = "crc32", + [__builtin_ctz(ARM_HWCAP2_ARM_SB )] = "sb", + [__builtin_ctz(ARM_HWCAP2_ARM_SSBS )] = "ssbs", }; return bit < ARRAY_SIZE(hwcap_str) ? hwcap_str[bit] : NULL; @@ -XXX,XX +XXX,XX @@ enum { ARM_HWCAP2_A64_SME_B16F32 = 1 << 28, ARM_HWCAP2_A64_SME_F32F32 = 1 << 29, ARM_HWCAP2_A64_SME_FA64 = 1 << 30, + ARM_HWCAP2_A64_WFXT = 1ULL << 31, + ARM_HWCAP2_A64_EBF16 = 1ULL << 32, + ARM_HWCAP2_A64_SVE_EBF16 = 1ULL << 33, + ARM_HWCAP2_A64_CSSC = 1ULL << 34, + ARM_HWCAP2_A64_RPRFM = 1ULL << 35, + ARM_HWCAP2_A64_SVE2P1 = 1ULL << 36, + ARM_HWCAP2_A64_SME2 = 1ULL << 37, + ARM_HWCAP2_A64_SME2P1 = 1ULL << 38, + ARM_HWCAP2_A64_SME_I16I32 = 1ULL << 39, + ARM_HWCAP2_A64_SME_BI32I32 = 1ULL << 40, + ARM_HWCAP2_A64_SME_B16B16 = 1ULL << 41, + ARM_HWCAP2_A64_SME_F16F16 = 1ULL << 42, + ARM_HWCAP2_A64_MOPS = 1ULL << 43, + ARM_HWCAP2_A64_HBC = 1ULL << 44, }; #define ELF_HWCAP get_elf_hwcap() @@ -XXX,XX +XXX,XX @@ const char *elf_hwcap2_str(uint32_t bit) [__builtin_ctz(ARM_HWCAP2_A64_SME_B16F32 )] = "smeb16f32", [__builtin_ctz(ARM_HWCAP2_A64_SME_F32F32 )] = "smef32f32", [__builtin_ctz(ARM_HWCAP2_A64_SME_FA64 )] = "smefa64", + [__builtin_ctz(ARM_HWCAP2_A64_WFXT )] = "wfxt", + [__builtin_ctzll(ARM_HWCAP2_A64_EBF16 )] = "ebf16", + [__builtin_ctzll(ARM_HWCAP2_A64_SVE_EBF16 )] = "sveebf16", + [__builtin_ctzll(ARM_HWCAP2_A64_CSSC )] = "cssc", + [__builtin_ctzll(ARM_HWCAP2_A64_RPRFM )] = "rprfm", + [__builtin_ctzll(ARM_HWCAP2_A64_SVE2P1 )] = "sve2p1", + [__builtin_ctzll(ARM_HWCAP2_A64_SME2 )] = "sme2", + [__builtin_ctzll(ARM_HWCAP2_A64_SME2P1 )] = "sme2p1", + [__builtin_ctzll(ARM_HWCAP2_A64_SME_I16I32 )] = "smei16i32", + [__builtin_ctzll(ARM_HWCAP2_A64_SME_BI32I32)] = "smebi32i32", + [__builtin_ctzll(ARM_HWCAP2_A64_SME_B16B16 )] = "smeb16b16", + [__builtin_ctzll(ARM_HWCAP2_A64_SME_F16F16 )] = "smef16f16", + [__builtin_ctzll(ARM_HWCAP2_A64_MOPS )] = "mops", + [__builtin_ctzll(ARM_HWCAP2_A64_HBC )] = "hbc", }; return bit < ARRAY_SIZE(hwcap_str) ? hwcap_str[bit] : NULL; -- 2.34.1 Add the code to report the arm32 hwcaps we were previously missing: ss, ssbs, fphp, asimdhp, asimddp, asimdfhm, asimdbf16, i8mm Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> --- linux-user/elfload.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/linux-user/elfload.c b/linux-user/elfload.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/elfload.c +++ b/linux-user/elfload.c @@ -XXX,XX +XXX,XX @@ uint32_t get_elf_hwcap(void) } } GET_FEATURE_ID(aa32_simdfmac, ARM_HWCAP_ARM_VFPv4); + /* + * MVFR1.FPHP and .SIMDHP must be in sync, and QEMU uses the same + * isar_feature function for both. The kernel reports them as two hwcaps. + */ + GET_FEATURE_ID(aa32_fp16_arith, ARM_HWCAP_ARM_FPHP); + GET_FEATURE_ID(aa32_fp16_arith, ARM_HWCAP_ARM_ASIMDHP); + GET_FEATURE_ID(aa32_dp, ARM_HWCAP_ARM_ASIMDDP); + GET_FEATURE_ID(aa32_fhm, ARM_HWCAP_ARM_ASIMDFHM); + GET_FEATURE_ID(aa32_bf16, ARM_HWCAP_ARM_ASIMDBF16); + GET_FEATURE_ID(aa32_i8mm, ARM_HWCAP_ARM_I8MM); return hwcaps; } @@ -XXX,XX +XXX,XX @@ uint32_t get_elf_hwcap2(void) GET_FEATURE_ID(aa32_sha1, ARM_HWCAP2_ARM_SHA1); GET_FEATURE_ID(aa32_sha2, ARM_HWCAP2_ARM_SHA2); GET_FEATURE_ID(aa32_crc32, ARM_HWCAP2_ARM_CRC32); + GET_FEATURE_ID(aa32_sb, ARM_HWCAP2_ARM_SB); + GET_FEATURE_ID(aa32_ssbs, ARM_HWCAP2_ARM_SSBS); return hwcaps; } -- 2.34.1 Update our AArch64 ID register field definitions from the 2023-06 system register XML release: https://developer.arm.com/documentation/ddi0601/2023-06/ Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> --- target/arm/cpu.h | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/target/arm/cpu.h b/target/arm/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -XXX,XX +XXX,XX @@ FIELD(ID_AA64ISAR0, SHA1, 8, 4) FIELD(ID_AA64ISAR0, SHA2, 12, 4) FIELD(ID_AA64ISAR0, CRC32, 16, 4) FIELD(ID_AA64ISAR0, ATOMIC, 20, 4) +FIELD(ID_AA64ISAR0, TME, 24, 4) FIELD(ID_AA64ISAR0, RDM, 28, 4) FIELD(ID_AA64ISAR0, SHA3, 32, 4) FIELD(ID_AA64ISAR0, SM3, 36, 4) @@ -XXX,XX +XXX,XX @@ FIELD(ID_AA64ISAR2, APA3, 12, 4) FIELD(ID_AA64ISAR2, MOPS, 16, 4) FIELD(ID_AA64ISAR2, BC, 20, 4) FIELD(ID_AA64ISAR2, PAC_FRAC, 24, 4) +FIELD(ID_AA64ISAR2, CLRBHB, 28, 4) +FIELD(ID_AA64ISAR2, SYSREG_128, 32, 4) +FIELD(ID_AA64ISAR2, SYSINSTR_128, 36, 4) +FIELD(ID_AA64ISAR2, PRFMSLC, 40, 4) +FIELD(ID_AA64ISAR2, RPRFM, 48, 4) +FIELD(ID_AA64ISAR2, CSSC, 52, 4) +FIELD(ID_AA64ISAR2, ATS1A, 60, 4) FIELD(ID_AA64PFR0, EL0, 0, 4) FIELD(ID_AA64PFR0, EL1, 4, 4) @@ -XXX,XX +XXX,XX @@ FIELD(ID_AA64PFR1, SME, 24, 4) FIELD(ID_AA64PFR1, RNDR_TRAP, 28, 4) FIELD(ID_AA64PFR1, CSV2_FRAC, 32, 4) FIELD(ID_AA64PFR1, NMI, 36, 4) +FIELD(ID_AA64PFR1, MTE_FRAC, 40, 4) +FIELD(ID_AA64PFR1, GCS, 44, 4) +FIELD(ID_AA64PFR1, THE, 48, 4) +FIELD(ID_AA64PFR1, MTEX, 52, 4) +FIELD(ID_AA64PFR1, DF2, 56, 4) +FIELD(ID_AA64PFR1, PFAR, 60, 4) FIELD(ID_AA64MMFR0, PARANGE, 0, 4) FIELD(ID_AA64MMFR0, ASIDBITS, 4, 4) @@ -XXX,XX +XXX,XX @@ FIELD(ID_AA64MMFR1, AFP, 44, 4) FIELD(ID_AA64MMFR1, NTLBPA, 48, 4) FIELD(ID_AA64MMFR1, TIDCP1, 52, 4) FIELD(ID_AA64MMFR1, CMOW, 56, 4) +FIELD(ID_AA64MMFR1, ECBHB, 60, 4) FIELD(ID_AA64MMFR2, CNP, 0, 4) FIELD(ID_AA64MMFR2, UAO, 4, 4) @@ -XXX,XX +XXX,XX @@ FIELD(ID_AA64DFR0, DEBUGVER, 0, 4) FIELD(ID_AA64DFR0, TRACEVER, 4, 4) FIELD(ID_AA64DFR0, PMUVER, 8, 4) FIELD(ID_AA64DFR0, BRPS, 12, 4) +FIELD(ID_AA64DFR0, PMSS, 16, 4) FIELD(ID_AA64DFR0, WRPS, 20, 4) +FIELD(ID_AA64DFR0, SEBEP, 24, 4) FIELD(ID_AA64DFR0, CTX_CMPS, 28, 4) FIELD(ID_AA64DFR0, PMSVER, 32, 4) FIELD(ID_AA64DFR0, DOUBLELOCK, 36, 4) @@ -XXX,XX +XXX,XX @@ FIELD(ID_AA64DFR0, TRACEFILT, 40, 4) FIELD(ID_AA64DFR0, TRACEBUFFER, 44, 4) FIELD(ID_AA64DFR0, MTPMU, 48, 4) FIELD(ID_AA64DFR0, BRBE, 52, 4) +FIELD(ID_AA64DFR0, EXTTRCBUFF, 56, 4) FIELD(ID_AA64DFR0, HPMN0, 60, 4) FIELD(ID_AA64ZFR0, SVEVER, 0, 4) FIELD(ID_AA64ZFR0, AES, 4, 4) FIELD(ID_AA64ZFR0, BITPERM, 16, 4) FIELD(ID_AA64ZFR0, BFLOAT16, 20, 4) +FIELD(ID_AA64ZFR0, B16B16, 24, 4) FIELD(ID_AA64ZFR0, SHA3, 32, 4) FIELD(ID_AA64ZFR0, SM4, 40, 4) FIELD(ID_AA64ZFR0, I8MM, 44, 4) @@ -XXX,XX +XXX,XX @@ FIELD(ID_AA64ZFR0, F32MM, 52, 4) FIELD(ID_AA64ZFR0, F64MM, 56, 4) FIELD(ID_AA64SMFR0, F32F32, 32, 1) +FIELD(ID_AA64SMFR0, BI32I32, 33, 1) FIELD(ID_AA64SMFR0, B16F32, 34, 1) FIELD(ID_AA64SMFR0, F16F32, 35, 1) FIELD(ID_AA64SMFR0, I8I32, 36, 4) +FIELD(ID_AA64SMFR0, F16F16, 42, 1) +FIELD(ID_AA64SMFR0, B16B16, 43, 1) +FIELD(ID_AA64SMFR0, I16I32, 44, 4) FIELD(ID_AA64SMFR0, F64F64, 48, 1) FIELD(ID_AA64SMFR0, I16I64, 52, 4) FIELD(ID_AA64SMFR0, SMEVER, 56, 4) -- 2.34.1 For user-only mode we reveal a subset of the AArch64 ID registers to the guest, to emulate the kernel's trap-and-emulate-ID-regs handling. Update the feature bit masks to match upstream kernel commit a48fa7efaf1161c1c. None of these features are yet implemented by QEMU, so this doesn't yet have a behavioural change, but implementation of FEAT_MOPS and FEAT_HBC is imminent. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> --- target/arm/helper.c | 11 ++++++++++- tests/tcg/aarch64/sysregs.c | 4 ++-- 2 files changed, 12 insertions(+), 3 deletions(-) diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) R_ID_AA64ZFR0_F64MM_MASK }, { .name = "ID_AA64SMFR0_EL1", .exported_bits = R_ID_AA64SMFR0_F32F32_MASK | + R_ID_AA64SMFR0_BI32I32_MASK | R_ID_AA64SMFR0_B16F32_MASK | R_ID_AA64SMFR0_F16F32_MASK | R_ID_AA64SMFR0_I8I32_MASK | + R_ID_AA64SMFR0_F16F16_MASK | + R_ID_AA64SMFR0_B16B16_MASK | + R_ID_AA64SMFR0_I16I32_MASK | R_ID_AA64SMFR0_F64F64_MASK | R_ID_AA64SMFR0_I16I64_MASK | + R_ID_AA64SMFR0_SMEVER_MASK | R_ID_AA64SMFR0_FA64_MASK }, { .name = "ID_AA64MMFR0_EL1", .exported_bits = R_ID_AA64MMFR0_ECV_MASK, @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) .exported_bits = R_ID_AA64ISAR2_WFXT_MASK | R_ID_AA64ISAR2_RPRES_MASK | R_ID_AA64ISAR2_GPA3_MASK | - R_ID_AA64ISAR2_APA3_MASK }, + R_ID_AA64ISAR2_APA3_MASK | + R_ID_AA64ISAR2_MOPS_MASK | + R_ID_AA64ISAR2_BC_MASK | + R_ID_AA64ISAR2_RPRFM_MASK | + R_ID_AA64ISAR2_CSSC_MASK }, { .name = "ID_AA64ISAR*_EL1_RESERVED", .is_glob = true }, }; diff --git a/tests/tcg/aarch64/sysregs.c b/tests/tcg/aarch64/sysregs.c index XXXXXXX..XXXXXXX 100644 --- a/tests/tcg/aarch64/sysregs.c +++ b/tests/tcg/aarch64/sysregs.c @@ -XXX,XX +XXX,XX @@ int main(void) */ get_cpu_reg_check_mask(id_aa64isar0_el1, _m(f0ff,ffff,f0ff,fff0)); get_cpu_reg_check_mask(id_aa64isar1_el1, _m(00ff,f0ff,ffff,ffff)); - get_cpu_reg_check_mask(SYS_ID_AA64ISAR2_EL1, _m(0000,0000,0000,ffff)); + get_cpu_reg_check_mask(SYS_ID_AA64ISAR2_EL1, _m(00ff,0000,00ff,ffff)); /* TGran4 & TGran64 as pegged to -1 */ get_cpu_reg_check_mask(id_aa64mmfr0_el1, _m(f000,0000,ff00,0000)); get_cpu_reg_check_mask(id_aa64mmfr1_el1, _m(0000,f000,0000,0000)); @@ -XXX,XX +XXX,XX @@ int main(void) get_cpu_reg_check_mask(id_aa64dfr0_el1, _m(0000,0000,0000,0006)); get_cpu_reg_check_zero(id_aa64dfr1_el1); get_cpu_reg_check_mask(SYS_ID_AA64ZFR0_EL1, _m(0ff0,ff0f,00ff,00ff)); - get_cpu_reg_check_mask(SYS_ID_AA64SMFR0_EL1, _m(80f1,00fd,0000,0000)); + get_cpu_reg_check_mask(SYS_ID_AA64SMFR0_EL1, _m(8ff1,fcff,0000,0000)); get_cpu_reg_check_zero(id_aa64afr0_el1); get_cpu_reg_check_zero(id_aa64afr1_el1); -- 2.34.1 FEAT_HBC (Hinted conditional branches) provides a new instruction BC.cond, which behaves exactly like the existing B.cond except that it provides a hint to the branch predictor about the likely behaviour of the branch. Since QEMU does not implement branch prediction, we can treat this identically to B.cond. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu.h | 5 +++++ target/arm/tcg/a64.decode | 3 ++- linux-user/elfload.c | 1 + target/arm/tcg/cpu64.c | 4 ++++ target/arm/tcg/translate-a64.c | 4 ++++ 6 files changed, 17 insertions(+), 1 deletion(-) diff --git a/docs/system/arm/emulation.rst b/docs/system/arm/emulation.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/emulation.rst +++ b/docs/system/arm/emulation.rst @@ -XXX,XX +XXX,XX @@ the following architecture extensions: - FEAT_FlagM2 (Enhancements to flag manipulation instructions) - FEAT_GTG (Guest translation granule size) - FEAT_HAFDBS (Hardware management of the access flag and dirty bit state) +- FEAT_HBC (Hinted conditional branches) - FEAT_HCX (Support for the HCRX_EL2 register) - FEAT_HPDS (Hierarchical permission disables) - FEAT_HPDS2 (Translation table page-based hardware attributes) diff --git a/target/arm/cpu.h b/target/arm/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_i8mm(const ARMISARegisters *id) return FIELD_EX64(id->id_aa64isar1, ID_AA64ISAR1, I8MM) != 0; } +static inline bool isar_feature_aa64_hbc(const ARMISARegisters *id) +{ + return FIELD_EX64(id->id_aa64isar2, ID_AA64ISAR2, BC) != 0; +} + static inline bool isar_feature_aa64_tgran4_lpa2(const ARMISARegisters *id) { return FIELD_SEX64(id->id_aa64mmfr0, ID_AA64MMFR0, TGRAN4) >= 1; diff --git a/target/arm/tcg/a64.decode b/target/arm/tcg/a64.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/a64.decode +++ b/target/arm/tcg/a64.decode @@ -XXX,XX +XXX,XX @@ CBZ sf:1 011010 nz:1 ................... rt:5 &cbz imm=%imm19 TBZ . 011011 nz:1 ..... .............. rt:5 &tbz imm=%imm14 bitpos=%imm31_19 -B_cond 0101010 0 ................... 0 cond:4 imm=%imm19 +# B.cond and BC.cond +B_cond 0101010 0 ................... c:1 cond:4 imm=%imm19 BR 1101011 0000 11111 000000 rn:5 00000 &r BLR 1101011 0001 11111 000000 rn:5 00000 &r diff --git a/linux-user/elfload.c b/linux-user/elfload.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/elfload.c +++ b/linux-user/elfload.c @@ -XXX,XX +XXX,XX @@ uint32_t get_elf_hwcap2(void) GET_FEATURE_ID(aa64_sme_f64f64, ARM_HWCAP2_A64_SME_F64F64); GET_FEATURE_ID(aa64_sme_i16i64, ARM_HWCAP2_A64_SME_I16I64); GET_FEATURE_ID(aa64_sme_fa64, ARM_HWCAP2_A64_SME_FA64); + GET_FEATURE_ID(aa64_hbc, ARM_HWCAP2_A64_HBC); return hwcaps; } diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu64.c +++ b/target/arm/tcg/cpu64.c @@ -XXX,XX +XXX,XX @@ void aarch64_max_tcg_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64ISAR1, I8MM, 1); /* FEAT_I8MM */ cpu->isar.id_aa64isar1 = t; + t = cpu->isar.id_aa64isar2; + t = FIELD_DP64(t, ID_AA64ISAR2, BC, 1); /* FEAT_HBC */ + cpu->isar.id_aa64isar2 = t; + t = cpu->isar.id_aa64pfr0; t = FIELD_DP64(t, ID_AA64PFR0, FP, 1); /* FEAT_FP16 */ t = FIELD_DP64(t, ID_AA64PFR0, ADVSIMD, 1); /* FEAT_FP16 */ diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ static bool trans_TBZ(DisasContext *s, arg_tbz *a) static bool trans_B_cond(DisasContext *s, arg_B_cond *a) { + /* BC.cond is only present with FEAT_HBC */ + if (a->c && !dc_isar_feature(aa64_hbc, s)) { + return false; + } reset_btype(s); if (a->cond < 0x0e) { /* genuinely conditional branches */ -- 2.34.1 The allocation_tag_mem() function takes an argument tag_size, but it never uses it. Remove the argument. In mte_probe_int() in particular this also lets us delete the code computing the value we were passing in. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> --- target/arm/tcg/mte_helper.c | 42 +++++++++++++------------------------ 1 file changed, 14 insertions(+), 28 deletions(-) diff --git a/target/arm/tcg/mte_helper.c b/target/arm/tcg/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/mte_helper.c +++ b/target/arm/tcg/mte_helper.c @@ -XXX,XX +XXX,XX @@ static int choose_nonexcluded_tag(int tag, int offset, uint16_t exclude) * @ptr_access: the access to use for the virtual address * @ptr_size: the number of bytes in the normal memory access * @tag_access: the access to use for the tag memory - * @tag_size: the number of bytes in the tag memory access * @ra: the return address for exception handling * * Our tag memory is formatted as a sequence of little-endian nibbles. @@ -XXX,XX +XXX,XX @@ static int choose_nonexcluded_tag(int tag, int offset, uint16_t exclude) * a pointer to the corresponding tag byte. Exit with exception if the * virtual address is not accessible for @ptr_access. * - * The @ptr_size and @tag_size values may not have an obvious relation - * due to the alignment of @ptr, and the number of tag checks required. - * * If there is no tag storage corresponding to @ptr, return NULL. */ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, uint64_t ptr, MMUAccessType ptr_access, int ptr_size, MMUAccessType tag_access, - int tag_size, uintptr_t ra) + uintptr_t ra) { #ifdef CONFIG_USER_ONLY uint64_t clean_ptr = useronly_clean_ptr(ptr); @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(ldg)(CPUARMState *env, uint64_t ptr, uint64_t xt) /* Trap if accessing an invalid page. */ mem = allocation_tag_mem(env, mmu_idx, ptr, MMU_DATA_LOAD, 1, - MMU_DATA_LOAD, 1, GETPC()); + MMU_DATA_LOAD, GETPC()); /* Load if page supports tags. */ if (mem) { @@ -XXX,XX +XXX,XX @@ static inline void do_stg(CPUARMState *env, uint64_t ptr, uint64_t xt, /* Trap if accessing an invalid page. */ mem = allocation_tag_mem(env, mmu_idx, ptr, MMU_DATA_STORE, TAG_GRANULE, - MMU_DATA_STORE, 1, ra); + MMU_DATA_STORE, ra); /* Store if page supports tags. */ if (mem) { @@ -XXX,XX +XXX,XX @@ static inline void do_st2g(CPUARMState *env, uint64_t ptr, uint64_t xt, if (ptr & TAG_GRANULE) { /* Two stores unaligned mod TAG_GRANULE*2 -- modify two bytes. */ mem1 = allocation_tag_mem(env, mmu_idx, ptr, MMU_DATA_STORE, - TAG_GRANULE, MMU_DATA_STORE, 1, ra); + TAG_GRANULE, MMU_DATA_STORE, ra); mem2 = allocation_tag_mem(env, mmu_idx, ptr + TAG_GRANULE, MMU_DATA_STORE, TAG_GRANULE, - MMU_DATA_STORE, 1, ra); + MMU_DATA_STORE, ra); /* Store if page(s) support tags. */ if (mem1) { @@ -XXX,XX +XXX,XX @@ static inline void do_st2g(CPUARMState *env, uint64_t ptr, uint64_t xt, } else { /* Two stores aligned mod TAG_GRANULE*2 -- modify one byte. */ mem1 = allocation_tag_mem(env, mmu_idx, ptr, MMU_DATA_STORE, - 2 * TAG_GRANULE, MMU_DATA_STORE, 1, ra); + 2 * TAG_GRANULE, MMU_DATA_STORE, ra); if (mem1) { tag |= tag << 4; qatomic_set(mem1, tag); @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(ldgm)(CPUARMState *env, uint64_t ptr) /* Trap if accessing an invalid page. */ tag_mem = allocation_tag_mem(env, mmu_idx, ptr, MMU_DATA_LOAD, - gm_bs_bytes, MMU_DATA_LOAD, - gm_bs_bytes / (2 * TAG_GRANULE), ra); + gm_bs_bytes, MMU_DATA_LOAD, ra); /* The tag is squashed to zero if the page does not support tags. */ if (!tag_mem) { @@ -XXX,XX +XXX,XX @@ void HELPER(stgm)(CPUARMState *env, uint64_t ptr, uint64_t val) /* Trap if accessing an invalid page. */ tag_mem = allocation_tag_mem(env, mmu_idx, ptr, MMU_DATA_STORE, - gm_bs_bytes, MMU_DATA_LOAD, - gm_bs_bytes / (2 * TAG_GRANULE), ra); + gm_bs_bytes, MMU_DATA_LOAD, ra); /* * Tag store only happens if the page support tags, @@ -XXX,XX +XXX,XX @@ void HELPER(stzgm_tags)(CPUARMState *env, uint64_t ptr, uint64_t val) ptr &= -dcz_bytes; mem = allocation_tag_mem(env, mmu_idx, ptr, MMU_DATA_STORE, dcz_bytes, - MMU_DATA_STORE, tag_bytes, ra); + MMU_DATA_STORE, ra); if (mem) { int tag_pair = (val & 0xf) * 0x11; memset(mem, tag_pair, tag_bytes); @@ -XXX,XX +XXX,XX @@ static int mte_probe_int(CPUARMState *env, uint32_t desc, uint64_t ptr, int mmu_idx, ptr_tag, bit55; uint64_t ptr_last, prev_page, next_page; uint64_t tag_first, tag_last; - uint64_t tag_byte_first, tag_byte_last; - uint32_t sizem1, tag_count, tag_size, n, c; + uint32_t sizem1, tag_count, n, c; uint8_t *mem1, *mem2; MMUAccessType type; @@ -XXX,XX +XXX,XX @@ static int mte_probe_int(CPUARMState *env, uint32_t desc, uint64_t ptr, tag_last = QEMU_ALIGN_DOWN(ptr_last, TAG_GRANULE); tag_count = ((tag_last - tag_first) / TAG_GRANULE) + 1; - /* Round the bounds to twice the tag granule, and compute the bytes. */ - tag_byte_first = QEMU_ALIGN_DOWN(ptr, 2 * TAG_GRANULE); - tag_byte_last = QEMU_ALIGN_DOWN(ptr_last, 2 * TAG_GRANULE); - /* Locate the page boundaries. */ prev_page = ptr & TARGET_PAGE_MASK; next_page = prev_page + TARGET_PAGE_SIZE; if (likely(tag_last - prev_page < TARGET_PAGE_SIZE)) { /* Memory access stays on one page. */ - tag_size = ((tag_byte_last - tag_byte_first) / (2 * TAG_GRANULE)) + 1; mem1 = allocation_tag_mem(env, mmu_idx, ptr, type, sizem1 + 1, - MMU_DATA_LOAD, tag_size, ra); + MMU_DATA_LOAD, ra); if (!mem1) { return 1; } @@ -XXX,XX +XXX,XX @@ static int mte_probe_int(CPUARMState *env, uint32_t desc, uint64_t ptr, n = checkN(mem1, ptr & TAG_GRANULE, ptr_tag, tag_count); } else { /* Memory access crosses to next page. */ - tag_size = (next_page - tag_byte_first) / (2 * TAG_GRANULE); mem1 = allocation_tag_mem(env, mmu_idx, ptr, type, next_page - ptr, - MMU_DATA_LOAD, tag_size, ra); + MMU_DATA_LOAD, ra); - tag_size = ((tag_byte_last - next_page) / (2 * TAG_GRANULE)) + 1; mem2 = allocation_tag_mem(env, mmu_idx, next_page, type, ptr_last - next_page + 1, - MMU_DATA_LOAD, tag_size, ra); + MMU_DATA_LOAD, ra); /* * Perform all of the comparisons. @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(mte_check_zva)(CPUARMState *env, uint32_t desc, uint64_t ptr) mmu_idx = FIELD_EX32(desc, MTEDESC, MIDX); (void) probe_write(env, ptr, 1, mmu_idx, ra); mem = allocation_tag_mem(env, mmu_idx, align_ptr, MMU_DATA_STORE, - dcz_bytes, MMU_DATA_LOAD, tag_bytes, ra); + dcz_bytes, MMU_DATA_LOAD, ra); if (!mem) { goto done; } -- 2.34.1 The LDRT/STRT "unprivileged load/store" instructions behave like normal ones if executed at EL0. We handle this correctly for the load/store semantics, but get the MTE checking wrong. We always look at s->mte_active[is_unpriv] to see whether we should be doing MTE checks, but in hflags.c when we set the TB flags that will be used to fill the mte_active[] array we only set the MTE0_ACTIVE bit if UNPRIV is true (i.e. we are not at EL0). This means that a LDRT at EL0 will see s->mte_active[1] as 0, and will not do MTE checks even when MTE is enabled. To avoid the translate-time code having to do an explicit check on s->unpriv to see if it is OK to index into the mte_active[] array, duplicate MTE_ACTIVE into MTE0_ACTIVE when UNPRIV is false. (This isn't a very serious bug because generally nobody executes LDRT/STRT at EL0, because they have no use there.) Cc: qemu-stable@nongnu.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-2-peter.maydell@linaro.org --- target/arm/tcg/hflags.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/target/arm/tcg/hflags.c b/target/arm/tcg/hflags.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/hflags.c +++ b/target/arm/tcg/hflags.c @@ -XXX,XX +XXX,XX @@ static CPUARMTBFlags rebuild_hflags_a64(CPUARMState *env, int el, int fp_el, && !(env->pstate & PSTATE_TCO) && (sctlr & (el == 0 ? SCTLR_TCF0 : SCTLR_TCF))) { DP_TBFLAG_A64(flags, MTE_ACTIVE, 1); + if (!EX_TBFLAG_A64(flags, UNPRIV)) { + /* + * In non-unpriv contexts (eg EL0), unpriv load/stores + * act like normal ones; duplicate the MTE info to + * avoid translate-a64.c having to check UNPRIV to see + * whether it is OK to index into MTE_ACTIVE[]. + */ + DP_TBFLAG_A64(flags, MTE0_ACTIVE, 1); + } } } /* And again for unprivileged accesses, if required. */ -- 2.34.1 FEAT_MOPS defines a handful of new enable bits: * HCRX_EL2.MSCEn, SCTLR_EL1.MSCEn, SCTLR_EL2.MSCen: define whether the new insns should UNDEF or not * HCRX_EL2.MCE2: defines whether memops exceptions from EL1 should be taken to EL1 or EL2 Since we don't sanitise what bits can be written for the SCTLR registers, we only need to handle the new bits in HCRX_EL2, and define SCTLR_MSCEN for the new SCTLR bit value. The precedence of "HCRX bits acts as 0 if SCR_EL3.HXEn is 0" versus "bit acts as 1 if EL2 disabled" is not clear from the register definition text, but it is clear in the CheckMOPSEnabled() pseudocode(), so we follow that. We'll have to check whether other bits we need to implement in future follow the same logic or not. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-3-peter.maydell@linaro.org --- target/arm/cpu.h | 6 ++++++ target/arm/helper.c | 28 +++++++++++++++++++++------- 2 files changed, 27 insertions(+), 7 deletions(-) diff --git a/target/arm/cpu.h b/target/arm/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -XXX,XX +XXX,XX @@ void pmu_init(ARMCPU *cpu); #define SCTLR_EnIB (1U << 30) /* v8.3, AArch64 only */ #define SCTLR_EnIA (1U << 31) /* v8.3, AArch64 only */ #define SCTLR_DSSBS_32 (1U << 31) /* v8.5, AArch32 only */ +#define SCTLR_MSCEN (1ULL << 33) /* FEAT_MOPS */ #define SCTLR_BT0 (1ULL << 35) /* v8.5-BTI */ #define SCTLR_BT1 (1ULL << 36) /* v8.5-BTI */ #define SCTLR_ITFSB (1ULL << 37) /* v8.5-MemTag */ @@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_doublelock(const ARMISARegisters *id) return FIELD_SEX64(id->id_aa64dfr0, ID_AA64DFR0, DOUBLELOCK) >= 0; } +static inline bool isar_feature_aa64_mops(const ARMISARegisters *id) +{ + return FIELD_EX64(id->id_aa64isar2, ID_AA64ISAR2, MOPS); +} + /* * Feature tests for "does this exist in either 32-bit or 64-bit?" */ diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ static void hcrx_write(CPUARMState *env, const ARMCPRegInfo *ri, { uint64_t valid_mask = 0; - /* No features adding bits to HCRX are implemented. */ + /* FEAT_MOPS adds MSCEn and MCE2 */ + if (cpu_isar_feature(aa64_mops, env_archcpu(env))) { + valid_mask |= HCRX_MSCEN | HCRX_MCE2; + } /* Clear RES0 bits. */ env->cp15.hcrx_el2 = value & valid_mask; @@ -XXX,XX +XXX,XX @@ uint64_t arm_hcrx_el2_eff(CPUARMState *env) { /* * The bits in this register behave as 0 for all purposes other than - * direct reads of the register if: - * - EL2 is not enabled in the current security state, - * - SCR_EL3.HXEn is 0. + * direct reads of the register if SCR_EL3.HXEn is 0. + * If EL2 is not enabled in the current security state, then the + * bit may behave as if 0, or as if 1, depending on the bit. + * For the moment, we treat the EL2-disabled case as taking + * priority over the HXEn-disabled case. This is true for the only + * bit for a feature which we implement where the answer is different + * for the two cases (MSCEn for FEAT_MOPS). + * This may need to be revisited for future bits. */ - if (!arm_is_el2_enabled(env) - || (arm_feature(env, ARM_FEATURE_EL3) - && !(env->cp15.scr_el3 & SCR_HXEN))) { + if (!arm_is_el2_enabled(env)) { + uint64_t hcrx = 0; + if (cpu_isar_feature(aa64_mops, env_archcpu(env))) { + /* MSCEn behaves as 1 if EL2 is not enabled */ + hcrx |= HCRX_MSCEN; + } + return hcrx; + } + if (arm_feature(env, ARM_FEATURE_EL3) && !(env->cp15.scr_el3 & SCR_HXEN)) { return 0; } return env->cp15.hcrx_el2; -- 2.34.1 In every place that we call the get_a64_user_mem_index() function we do it like this: memidx = a->unpriv ? get_a64_user_mem_index(s) : get_mem_index(s); Refactor so the caller passes in the bool that says whether they want the 'unpriv' or 'normal' mem_index rather than having to do the ?: themselves. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20230912140434.1333369-4-peter.maydell@linaro.org --- target/arm/tcg/translate-a64.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ void a64_translate_init(void) } /* - * Return the core mmu_idx to use for A64 "unprivileged load/store" insns + * Return the core mmu_idx to use for A64 load/store insns which + * have a "unprivileged load/store" variant. Those insns access + * EL0 if executed from an EL which has control over EL0 (usually + * EL1) but behave like normal loads and stores if executed from + * elsewhere (eg EL3). + * + * @unpriv : true for the unprivileged encoding; false for the + * normal encoding (in which case we will return the same + * thing as get_mem_index(). */ -static int get_a64_user_mem_index(DisasContext *s) +static int get_a64_user_mem_index(DisasContext *s, bool unpriv) { /* * If AccType_UNPRIV is not used, the insn uses AccType_NORMAL, @@ -XXX,XX +XXX,XX @@ static int get_a64_user_mem_index(DisasContext *s) */ ARMMMUIdx useridx = s->mmu_idx; - if (s->unpriv) { + if (unpriv && s->unpriv) { /* * We have pre-computed the condition for AccType_UNPRIV. * Therefore we should never get here with a mmu_idx for @@ -XXX,XX +XXX,XX @@ static void op_addr_ldst_imm_pre(DisasContext *s, arg_ldst_imm *a, if (!a->p) { tcg_gen_addi_i64(*dirty_addr, *dirty_addr, offset); } - memidx = a->unpriv ? get_a64_user_mem_index(s) : get_mem_index(s); + memidx = get_a64_user_mem_index(s, a->unpriv); *clean_addr = gen_mte_check1_mmuidx(s, *dirty_addr, is_store, a->w || a->rn != 31, mop, a->unpriv, memidx); @@ -XXX,XX +XXX,XX @@ static bool trans_STR_i(DisasContext *s, arg_ldst_imm *a) { bool iss_sf, iss_valid = !a->w; TCGv_i64 clean_addr, dirty_addr, tcg_rt; - int memidx = a->unpriv ? get_a64_user_mem_index(s) : get_mem_index(s); + int memidx = get_a64_user_mem_index(s, a->unpriv); MemOp mop = finalize_memop(s, a->sz + a->sign * MO_SIGN); op_addr_ldst_imm_pre(s, a, &clean_addr, &dirty_addr, a->imm, true, mop); @@ -XXX,XX +XXX,XX @@ static bool trans_LDR_i(DisasContext *s, arg_ldst_imm *a) { bool iss_sf, iss_valid = !a->w; TCGv_i64 clean_addr, dirty_addr, tcg_rt; - int memidx = a->unpriv ? get_a64_user_mem_index(s) : get_mem_index(s); + int memidx = get_a64_user_mem_index(s, a->unpriv); MemOp mop = finalize_memop(s, a->sz + a->sign * MO_SIGN); op_addr_ldst_imm_pre(s, a, &clean_addr, &dirty_addr, a->imm, false, mop); -- 2.34.1 The FEAT_MOPS memory operations can raise a Memory Copy or Memory Set exception if a copy or set instruction is executed when the CPU register state is not correct for that instruction. Define the usual syn_* function that constructs the syndrome register value for these exceptions. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-5-peter.maydell@linaro.org --- target/arm/syndrome.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/target/arm/syndrome.h b/target/arm/syndrome.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/syndrome.h +++ b/target/arm/syndrome.h @@ -XXX,XX +XXX,XX @@ enum arm_exception_class { EC_DATAABORT = 0x24, EC_DATAABORT_SAME_EL = 0x25, EC_SPALIGNMENT = 0x26, + EC_MOP = 0x27, EC_AA32_FPTRAP = 0x28, EC_AA64_FPTRAP = 0x2c, EC_SERROR = 0x2f, @@ -XXX,XX +XXX,XX @@ static inline uint32_t syn_serror(uint32_t extra) return (EC_SERROR << ARM_EL_EC_SHIFT) | ARM_EL_IL | extra; } +static inline uint32_t syn_mop(bool is_set, bool is_setg, int options, + bool epilogue, bool wrong_option, bool option_a, + int destreg, int srcreg, int sizereg) +{ + return (EC_MOP << ARM_EL_EC_SHIFT) | ARM_EL_IL | + (is_set << 24) | (is_setg << 23) | (options << 19) | + (epilogue << 18) | (wrong_option << 17) | (option_a << 16) | + (destreg << 10) | (srcreg << 5) | sizereg; +} + + #endif /* TARGET_ARM_SYNDROME_H */ -- 2.34.1 For the FEAT_MOPS operations, the existing allocation_tag_mem() function almost does what we want, but it will take a watchpoint exception even for an ra == 0 probe request, and it requires that the caller guarantee that the memory is accessible. For FEAT_MOPS we want a function that will not take any kind of exception, and will return NULL for the not-accessible case. Rename allocation_tag_mem() to allocation_tag_mem_probe() and add an extra 'probe' argument that lets us distinguish these cases; allocation_tag_mem() is now a wrapper that always passes 'false'. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-6-peter.maydell@linaro.org --- target/arm/tcg/mte_helper.c | 48 ++++++++++++++++++++++++++++--------- 1 file changed, 37 insertions(+), 11 deletions(-) diff --git a/target/arm/tcg/mte_helper.c b/target/arm/tcg/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/mte_helper.c +++ b/target/arm/tcg/mte_helper.c @@ -XXX,XX +XXX,XX @@ static int choose_nonexcluded_tag(int tag, int offset, uint16_t exclude) } /** - * allocation_tag_mem: + * allocation_tag_mem_probe: * @env: the cpu environment * @ptr_mmu_idx: the addressing regime to use for the virtual address * @ptr: the virtual address for which to look up tag memory * @ptr_access: the access to use for the virtual address * @ptr_size: the number of bytes in the normal memory access * @tag_access: the access to use for the tag memory + * @probe: true to merely probe, never taking an exception * @ra: the return address for exception handling * * Our tag memory is formatted as a sequence of little-endian nibbles. @@ -XXX,XX +XXX,XX @@ static int choose_nonexcluded_tag(int tag, int offset, uint16_t exclude) * for the higher addr. * * Here, resolve the physical address from the virtual address, and return - * a pointer to the corresponding tag byte. Exit with exception if the - * virtual address is not accessible for @ptr_access. + * a pointer to the corresponding tag byte. * * If there is no tag storage corresponding to @ptr, return NULL. + * + * If the page is inaccessible for @ptr_access, or has a watchpoint, there are + * three options: + * (1) probe = true, ra = 0 : pure probe -- we return NULL if the page is not + * accessible, and do not take watchpoint traps. The calling code must + * handle those cases in the right priority compared to MTE traps. + * (2) probe = false, ra = 0 : probe, no fault expected -- the caller guarantees + * that the page is going to be accessible. We will take watchpoint traps. + * (3) probe = false, ra != 0 : non-probe -- we will take both memory access + * traps and watchpoint traps. + * (probe = true, ra != 0 is invalid and will assert.) */ -static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, - uint64_t ptr, MMUAccessType ptr_access, - int ptr_size, MMUAccessType tag_access, - uintptr_t ra) +static uint8_t *allocation_tag_mem_probe(CPUARMState *env, int ptr_mmu_idx, + uint64_t ptr, MMUAccessType ptr_access, + int ptr_size, MMUAccessType tag_access, + bool probe, uintptr_t ra) { #ifdef CONFIG_USER_ONLY uint64_t clean_ptr = useronly_clean_ptr(ptr); @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, uint8_t *tags; uintptr_t index; + assert(!(probe && ra)); + if (!(flags & (ptr_access == MMU_DATA_STORE ? PAGE_WRITE_ORG : PAGE_READ))) { cpu_loop_exit_sigsegv(env_cpu(env), ptr, ptr_access, !(flags & PAGE_VALID), ra); @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, * exception for inaccessible pages, and resolves the virtual address * into the softmmu tlb. * - * When RA == 0, this is for mte_probe. The page is expected to be - * valid. Indicate to probe_access_flags no-fault, then assert that - * we received a valid page. + * When RA == 0, this is either a pure probe or a no-fault-expected probe. + * Indicate to probe_access_flags no-fault, then either return NULL + * for the pure probe, or assert that we received a valid page for the + * no-fault-expected probe. */ flags = probe_access_full(env, ptr, 0, ptr_access, ptr_mmu_idx, ra == 0, &host, &full, ra); + if (probe && (flags & TLB_INVALID_MASK)) { + return NULL; + } assert(!(flags & TLB_INVALID_MASK)); /* If the virtual page MemAttr != Tagged, access unchecked. */ @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, } /* Any debug exception has priority over a tag check exception. */ - if (unlikely(flags & TLB_WATCHPOINT)) { + if (!probe && unlikely(flags & TLB_WATCHPOINT)) { int wp = ptr_access == MMU_DATA_LOAD ? BP_MEM_READ : BP_MEM_WRITE; assert(ra != 0); cpu_check_watchpoint(env_cpu(env), ptr, ptr_size, attrs, wp, ra); @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, #endif } +static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, + uint64_t ptr, MMUAccessType ptr_access, + int ptr_size, MMUAccessType tag_access, + uintptr_t ra) +{ + return allocation_tag_mem_probe(env, ptr_mmu_idx, ptr, ptr_access, + ptr_size, tag_access, false, ra); +} + uint64_t HELPER(irg)(CPUARMState *env, uint64_t rn, uint64_t rm) { uint16_t exclude = extract32(rm | env->cp15.gcr_el1, 0, 16); -- 2.34.1 The FEAT_MOPS instructions need a couple of helper routines that check for MTE tag failures: * mte_mops_probe() checks whether there is going to be a tag error in the next up-to-a-page worth of data * mte_check_fail() is an existing function to record the fact of a tag failure, which we need to make global so we can call it from helper-a64.c Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-7-peter.maydell@linaro.org --- target/arm/internals.h | 28 +++++++++++++++++++ target/arm/tcg/mte_helper.c | 54 +++++++++++++++++++++++++++++++++++-- 2 files changed, 80 insertions(+), 2 deletions(-) diff --git a/target/arm/internals.h b/target/arm/internals.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/internals.h +++ b/target/arm/internals.h @@ -XXX,XX +XXX,XX @@ FIELD(MTEDESC, SIZEM1, 12, SIMD_DATA_BITS - 12) /* size - 1 */ bool mte_probe(CPUARMState *env, uint32_t desc, uint64_t ptr); uint64_t mte_check(CPUARMState *env, uint32_t desc, uint64_t ptr, uintptr_t ra); +/** + * mte_mops_probe: Check where the next MTE failure is for a FEAT_MOPS operation + * @env: CPU env + * @ptr: start address of memory region (dirty pointer) + * @size: length of region (guaranteed not to cross a page boundary) + * @desc: MTEDESC descriptor word (0 means no MTE checks) + * Returns: the size of the region that can be copied without hitting + * an MTE tag failure + * + * Note that we assume that the caller has already checked the TBI + * and TCMA bits with mte_checks_needed() and an MTE check is definitely + * required. + */ +uint64_t mte_mops_probe(CPUARMState *env, uint64_t ptr, uint64_t size, + uint32_t desc); + +/** + * mte_check_fail: Record an MTE tag check failure + * @env: CPU env + * @desc: MTEDESC descriptor word + * @dirty_ptr: Failing dirty address + * @ra: TCG retaddr + * + * This may never return (if the MTE tag checks are configured to fault). + */ +void mte_check_fail(CPUARMState *env, uint32_t desc, + uint64_t dirty_ptr, uintptr_t ra); + static inline int allocation_tag_from_addr(uint64_t ptr) { return extract64(ptr, 56, 4); diff --git a/target/arm/tcg/mte_helper.c b/target/arm/tcg/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/mte_helper.c +++ b/target/arm/tcg/mte_helper.c @@ -XXX,XX +XXX,XX @@ static void mte_async_check_fail(CPUARMState *env, uint64_t dirty_ptr, } /* Record a tag check failure. */ -static void mte_check_fail(CPUARMState *env, uint32_t desc, - uint64_t dirty_ptr, uintptr_t ra) +void mte_check_fail(CPUARMState *env, uint32_t desc, + uint64_t dirty_ptr, uintptr_t ra) { int mmu_idx = FIELD_EX32(desc, MTEDESC, MIDX); ARMMMUIdx arm_mmu_idx = core_to_aa64_mmu_idx(mmu_idx); @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(mte_check_zva)(CPUARMState *env, uint32_t desc, uint64_t ptr) done: return useronly_clean_ptr(ptr); } + +uint64_t mte_mops_probe(CPUARMState *env, uint64_t ptr, uint64_t size, + uint32_t desc) +{ + int mmu_idx, tag_count; + uint64_t ptr_tag, tag_first, tag_last; + void *mem; + bool w = FIELD_EX32(desc, MTEDESC, WRITE); + uint32_t n; + + mmu_idx = FIELD_EX32(desc, MTEDESC, MIDX); + /* True probe; this will never fault */ + mem = allocation_tag_mem_probe(env, mmu_idx, ptr, + w ? MMU_DATA_STORE : MMU_DATA_LOAD, + size, MMU_DATA_LOAD, true, 0); + if (!mem) { + return size; + } + + /* + * TODO: checkN() is not designed for checks of the size we expect + * for FEAT_MOPS operations, so we should implement this differently. + * Maybe we should do something like + * if (region start and size are aligned nicely) { + * do direct loads of 64 tag bits at a time; + * } else { + * call checkN() + * } + */ + /* Round the bounds to the tag granule, and compute the number of tags. */ + ptr_tag = allocation_tag_from_addr(ptr); + tag_first = QEMU_ALIGN_DOWN(ptr, TAG_GRANULE); + tag_last = QEMU_ALIGN_DOWN(ptr + size - 1, TAG_GRANULE); + tag_count = ((tag_last - tag_first) / TAG_GRANULE) + 1; + n = checkN(mem, ptr & TAG_GRANULE, ptr_tag, tag_count); + if (likely(n == tag_count)) { + return size; + } + + /* + * Failure; for the first granule, it's at @ptr. Otherwise + * it's at the first byte of the nth granule. Calculate how + * many bytes we can access without hitting that failure. + */ + if (n == 0) { + return 0; + } else { + return n * TAG_GRANULE - (ptr - tag_first); + } +} -- 2.34.1 Implement the SET* instructions which collectively implement a "memset" operation. These come in a set of three, eg SETP (prologue), SETM (main), SETE (epilogue), and each of those has different flavours to indicate whether memory accesses should be unpriv or non-temporal. This commit does not include the "memset with tag setting" SETG* instructions. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-8-peter.maydell@linaro.org --- target/arm/tcg/helper-a64.h | 4 + target/arm/tcg/a64.decode | 16 ++ target/arm/tcg/helper-a64.c | 344 +++++++++++++++++++++++++++++++++ target/arm/tcg/translate-a64.c | 49 +++++ 4 files changed, 413 insertions(+) diff --git a/target/arm/tcg/helper-a64.h b/target/arm/tcg/helper-a64.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/helper-a64.h +++ b/target/arm/tcg/helper-a64.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_3(stzgm_tags, TCG_CALL_NO_WG, void, env, i64, i64) DEF_HELPER_FLAGS_4(unaligned_access, TCG_CALL_NO_WG, noreturn, env, i64, i32, i32) + +DEF_HELPER_3(setp, void, env, i32, i32) +DEF_HELPER_3(setm, void, env, i32, i32) +DEF_HELPER_3(sete, void, env, i32, i32) diff --git a/target/arm/tcg/a64.decode b/target/arm/tcg/a64.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/a64.decode +++ b/target/arm/tcg/a64.decode @@ -XXX,XX +XXX,XX @@ LDGM 11011001 11 1 ......... 00 ..... ..... @ldst_tag_mult p=0 w=0 STZ2G 11011001 11 1 ......... 01 ..... ..... @ldst_tag p=1 w=1 STZ2G 11011001 11 1 ......... 10 ..... ..... @ldst_tag p=0 w=0 STZ2G 11011001 11 1 ......... 11 ..... ..... @ldst_tag p=0 w=1 + +# Memory operations (memset, memcpy, memmove) +# Each of these comes in a set of three, eg SETP (prologue), SETM (main), +# SETE (epilogue), and each of those has different flavours to +# indicate whether memory accesses should be unpriv or non-temporal. +# We don't distinguish temporal and non-temporal accesses, but we +# do need to report it in syndrome register values. + +# Memset +&set rs rn rd unpriv nontemp +# op2 bit 1 is nontemporal bit +@set .. ......... rs:5 .. nontemp:1 unpriv:1 .. rn:5 rd:5 &set + +SETP 00 011001110 ..... 00 . . 01 ..... ..... @set +SETM 00 011001110 ..... 01 . . 01 ..... ..... @set +SETE 00 011001110 ..... 10 . . 01 ..... ..... @set diff --git a/target/arm/tcg/helper-a64.c b/target/arm/tcg/helper-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/helper-a64.c +++ b/target/arm/tcg/helper-a64.c @@ -XXX,XX +XXX,XX @@ void HELPER(unaligned_access)(CPUARMState *env, uint64_t addr, arm_cpu_do_unaligned_access(env_cpu(env), addr, access_type, mmu_idx, GETPC()); } + +/* Memory operations (memset, memmove, memcpy) */ + +/* + * Return true if the CPY* and SET* insns can execute; compare + * pseudocode CheckMOPSEnabled(), though we refactor it a little. + */ +static bool mops_enabled(CPUARMState *env) +{ + int el = arm_current_el(env); + + if (el < 2 && + (arm_hcr_el2_eff(env) & (HCR_E2H | HCR_TGE)) != (HCR_E2H | HCR_TGE) && + !(arm_hcrx_el2_eff(env) & HCRX_MSCEN)) { + return false; + } + + if (el == 0) { + if (!el_is_in_host(env, 0)) { + return env->cp15.sctlr_el[1] & SCTLR_MSCEN; + } else { + return env->cp15.sctlr_el[2] & SCTLR_MSCEN; + } + } + return true; +} + +static void check_mops_enabled(CPUARMState *env, uintptr_t ra) +{ + if (!mops_enabled(env)) { + raise_exception_ra(env, EXCP_UDEF, syn_uncategorized(), + exception_target_el(env), ra); + } +} + +/* + * Return the target exception level for an exception due + * to mismatched arguments in a FEAT_MOPS copy or set. + * Compare pseudocode MismatchedCpySetTargetEL() + */ +static int mops_mismatch_exception_target_el(CPUARMState *env) +{ + int el = arm_current_el(env); + + if (el > 1) { + return el; + } + if (el == 0 && (arm_hcr_el2_eff(env) & HCR_TGE)) { + return 2; + } + if (el == 1 && (arm_hcrx_el2_eff(env) & HCRX_MCE2)) { + return 2; + } + return 1; +} + +/* + * Check whether an M or E instruction was executed with a CF value + * indicating the wrong option for this implementation. + * Assumes we are always Option A. + */ +static void check_mops_wrong_option(CPUARMState *env, uint32_t syndrome, + uintptr_t ra) +{ + if (env->CF != 0) { + syndrome |= 1 << 17; /* Set the wrong-option bit */ + raise_exception_ra(env, EXCP_UDEF, syndrome, + mops_mismatch_exception_target_el(env), ra); + } +} + +/* + * Return the maximum number of bytes we can transfer starting at addr + * without crossing a page boundary. + */ +static uint64_t page_limit(uint64_t addr) +{ + return TARGET_PAGE_ALIGN(addr + 1) - addr; +} + +/* + * Perform part of a memory set on an area of guest memory starting at + * toaddr (a dirty address) and extending for setsize bytes. + * + * Returns the number of bytes actually set, which might be less than + * setsize; the caller should loop until the whole set has been done. + * The caller should ensure that the guest registers are correct + * for the possibility that the first byte of the set encounters + * an exception or watchpoint. We guarantee not to take any faults + * for bytes other than the first. + */ +static uint64_t set_step(CPUARMState *env, uint64_t toaddr, + uint64_t setsize, uint32_t data, int memidx, + uint32_t *mtedesc, uintptr_t ra) +{ + void *mem; + + setsize = MIN(setsize, page_limit(toaddr)); + if (*mtedesc) { + uint64_t mtesize = mte_mops_probe(env, toaddr, setsize, *mtedesc); + if (mtesize == 0) { + /* Trap, or not. All CPU state is up to date */ + mte_check_fail(env, *mtedesc, toaddr, ra); + /* Continue, with no further MTE checks required */ + *mtedesc = 0; + } else { + /* Advance to the end, or to the tag mismatch */ + setsize = MIN(setsize, mtesize); + } + } + + toaddr = useronly_clean_ptr(toaddr); + /* + * Trapless lookup: returns NULL for invalid page, I/O, + * watchpoints, clean pages, etc. + */ + mem = tlb_vaddr_to_host(env, toaddr, MMU_DATA_STORE, memidx); + +#ifndef CONFIG_USER_ONLY + if (unlikely(!mem)) { + /* + * Slow-path: just do one byte write. This will handle the + * watchpoint, invalid page, etc handling correctly. + * For clean code pages, the next iteration will see + * the page dirty and will use the fast path. + */ + cpu_stb_mmuidx_ra(env, toaddr, data, memidx, ra); + return 1; + } +#endif + /* Easy case: just memset the host memory */ + memset(mem, data, setsize); + return setsize; +} + +typedef uint64_t StepFn(CPUARMState *env, uint64_t toaddr, + uint64_t setsize, uint32_t data, + int memidx, uint32_t *mtedesc, uintptr_t ra); + +/* Extract register numbers from a MOPS exception syndrome value */ +static int mops_destreg(uint32_t syndrome) +{ + return extract32(syndrome, 10, 5); +} + +static int mops_srcreg(uint32_t syndrome) +{ + return extract32(syndrome, 5, 5); +} + +static int mops_sizereg(uint32_t syndrome) +{ + return extract32(syndrome, 0, 5); +} + +/* + * Return true if TCMA and TBI bits mean we need to do MTE checks. + * We only need to do this once per MOPS insn, not for every page. + */ +static bool mte_checks_needed(uint64_t ptr, uint32_t desc) +{ + int bit55 = extract64(ptr, 55, 1); + + /* + * Note that tbi_check() returns true for "access checked" but + * tcma_check() returns true for "access unchecked". + */ + if (!tbi_check(desc, bit55)) { + return false; + } + return !tcma_check(desc, bit55, allocation_tag_from_addr(ptr)); +} + +/* + * For the Memory Set operation, our implementation chooses + * always to use "option A", where we update Xd to the final + * address in the SETP insn, and set Xn to be -(bytes remaining). + * On SETM and SETE insns we only need update Xn. + * + * @env: CPU + * @syndrome: syndrome value for mismatch exceptions + * (also contains the register numbers we need to use) + * @mtedesc: MTE descriptor word + * @stepfn: function which does a single part of the set operation + * @is_setg: true if this is the tag-setting SETG variant + */ +static void do_setp(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc, + StepFn *stepfn, bool is_setg, uintptr_t ra) +{ + /* Prologue: we choose to do up to the next page boundary */ + int rd = mops_destreg(syndrome); + int rs = mops_srcreg(syndrome); + int rn = mops_sizereg(syndrome); + uint8_t data = env->xregs[rs]; + uint32_t memidx = FIELD_EX32(mtedesc, MTEDESC, MIDX); + uint64_t toaddr = env->xregs[rd]; + uint64_t setsize = env->xregs[rn]; + uint64_t stagesetsize, step; + + check_mops_enabled(env, ra); + + if (setsize > INT64_MAX) { + setsize = INT64_MAX; + } + + if (!mte_checks_needed(toaddr, mtedesc)) { + mtedesc = 0; + } + + stagesetsize = MIN(setsize, page_limit(toaddr)); + while (stagesetsize) { + env->xregs[rd] = toaddr; + env->xregs[rn] = setsize; + step = stepfn(env, toaddr, stagesetsize, data, memidx, &mtedesc, ra); + toaddr += step; + setsize -= step; + stagesetsize -= step; + } + /* Insn completed, so update registers to the Option A format */ + env->xregs[rd] = toaddr + setsize; + env->xregs[rn] = -setsize; + + /* Set NZCV = 0000 to indicate we are an Option A implementation */ + env->NF = 0; + env->ZF = 1; /* our env->ZF encoding is inverted */ + env->CF = 0; + env->VF = 0; + return; +} + +void HELPER(setp)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) +{ + do_setp(env, syndrome, mtedesc, set_step, false, GETPC()); +} + +static void do_setm(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc, + StepFn *stepfn, bool is_setg, uintptr_t ra) +{ + /* Main: we choose to do all the full-page chunks */ + CPUState *cs = env_cpu(env); + int rd = mops_destreg(syndrome); + int rs = mops_srcreg(syndrome); + int rn = mops_sizereg(syndrome); + uint8_t data = env->xregs[rs]; + uint64_t toaddr = env->xregs[rd] + env->xregs[rn]; + uint64_t setsize = -env->xregs[rn]; + uint32_t memidx = FIELD_EX32(mtedesc, MTEDESC, MIDX); + uint64_t step, stagesetsize; + + check_mops_enabled(env, ra); + + /* + * We're allowed to NOP out "no data to copy" before the consistency + * checks; we choose to do so. + */ + if (env->xregs[rn] == 0) { + return; + } + + check_mops_wrong_option(env, syndrome, ra); + + /* + * Our implementation will work fine even if we have an unaligned + * destination address, and because we update Xn every time around + * the loop below and the return value from stepfn() may be less + * than requested, we might find toaddr is unaligned. So we don't + * have an IMPDEF check for alignment here. + */ + + if (!mte_checks_needed(toaddr, mtedesc)) { + mtedesc = 0; + } + + /* Do the actual memset: we leave the last partial page to SETE */ + stagesetsize = setsize & TARGET_PAGE_MASK; + while (stagesetsize > 0) { + step = stepfn(env, toaddr, setsize, data, memidx, &mtedesc, ra); + toaddr += step; + setsize -= step; + stagesetsize -= step; + env->xregs[rn] = -setsize; + if (stagesetsize > 0 && unlikely(cpu_loop_exit_requested(cs))) { + cpu_loop_exit_restore(cs, ra); + } + } +} + +void HELPER(setm)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) +{ + do_setm(env, syndrome, mtedesc, set_step, false, GETPC()); +} + +static void do_sete(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc, + StepFn *stepfn, bool is_setg, uintptr_t ra) +{ + /* Epilogue: do the last partial page */ + int rd = mops_destreg(syndrome); + int rs = mops_srcreg(syndrome); + int rn = mops_sizereg(syndrome); + uint8_t data = env->xregs[rs]; + uint64_t toaddr = env->xregs[rd] + env->xregs[rn]; + uint64_t setsize = -env->xregs[rn]; + uint32_t memidx = FIELD_EX32(mtedesc, MTEDESC, MIDX); + uint64_t step; + + check_mops_enabled(env, ra); + + /* + * We're allowed to NOP out "no data to copy" before the consistency + * checks; we choose to do so. + */ + if (setsize == 0) { + return; + } + + check_mops_wrong_option(env, syndrome, ra); + + /* + * Our implementation has no address alignment requirements, but + * we do want to enforce the "less than a page" size requirement, + * so we don't need to have the "check for interrupts" here. + */ + if (setsize >= TARGET_PAGE_SIZE) { + raise_exception_ra(env, EXCP_UDEF, syndrome, + mops_mismatch_exception_target_el(env), ra); + } + + if (!mte_checks_needed(toaddr, mtedesc)) { + mtedesc = 0; + } + + /* Do the actual memset */ + while (setsize > 0) { + step = stepfn(env, toaddr, setsize, data, memidx, &mtedesc, ra); + toaddr += step; + setsize -= step; + env->xregs[rn] = -setsize; + } +} + +void HELPER(sete)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) +{ + do_sete(env, syndrome, mtedesc, set_step, false, GETPC()); +} diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ TRANS_FEAT(STZG, aa64_mte_insn_reg, do_STG, a, true, false) TRANS_FEAT(ST2G, aa64_mte_insn_reg, do_STG, a, false, true) TRANS_FEAT(STZ2G, aa64_mte_insn_reg, do_STG, a, true, true) +typedef void SetFn(TCGv_env, TCGv_i32, TCGv_i32); + +static bool do_SET(DisasContext *s, arg_set *a, bool is_epilogue, SetFn fn) +{ + int memidx; + uint32_t syndrome, desc = 0; + + /* + * UNPREDICTABLE cases: we choose to UNDEF, which allows + * us to pull this check before the CheckMOPSEnabled() test + * (which we do in the helper function) + */ + if (a->rs == a->rn || a->rs == a->rd || a->rn == a->rd || + a->rd == 31 || a->rn == 31) { + return false; + } + + memidx = get_a64_user_mem_index(s, a->unpriv); + + /* + * We pass option_a == true, matching our implementation; + * we pass wrong_option == false: helper function may set that bit. + */ + syndrome = syn_mop(true, false, (a->nontemp << 1) | a->unpriv, + is_epilogue, false, true, a->rd, a->rs, a->rn); + + if (s->mte_active[a->unpriv]) { + /* We may need to do MTE tag checking, so assemble the descriptor */ + desc = FIELD_DP32(desc, MTEDESC, TBI, s->tbid); + desc = FIELD_DP32(desc, MTEDESC, TCMA, s->tcma); + desc = FIELD_DP32(desc, MTEDESC, WRITE, true); + /* SIZEM1 and ALIGN we leave 0 (byte write) */ + } + /* The helper function always needs the memidx even with MTE disabled */ + desc = FIELD_DP32(desc, MTEDESC, MIDX, memidx); + + /* + * The helper needs the register numbers, but since they're in + * the syndrome anyway, we let it extract them from there rather + * than passing in an extra three integer arguments. + */ + fn(cpu_env, tcg_constant_i32(syndrome), tcg_constant_i32(desc)); + return true; +} + +TRANS_FEAT(SETP, aa64_mops, do_SET, a, false, gen_helper_setp) +TRANS_FEAT(SETM, aa64_mops, do_SET, a, false, gen_helper_setm) +TRANS_FEAT(SETE, aa64_mops, do_SET, a, true, gen_helper_sete) + typedef void ArithTwoOp(TCGv_i64, TCGv_i64, TCGv_i64); static bool gen_rri(DisasContext *s, arg_rri_sf *a, -- 2.34.1 Currently the only tag-setting instructions always do so in the context of the current EL, and so we only need one ATA bit in the TB flags. The FEAT_MOPS SETG instructions include ones which set tags for a non-privileged access, so we now also need the equivalent "are tags enabled?" information for EL0. Add the new TB flag, and convert the existing 'bool ata' field in DisasContext to a 'bool ata[2]' that can be indexed by the is_unpriv bit in an instruction, similarly to mte[2]. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-9-peter.maydell@linaro.org --- target/arm/cpu.h | 1 + target/arm/tcg/translate.h | 4 ++-- target/arm/tcg/hflags.c | 12 ++++++++++++ target/arm/tcg/translate-a64.c | 23 ++++++++++++----------- 4 files changed, 27 insertions(+), 13 deletions(-) diff --git a/target/arm/cpu.h b/target/arm/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -XXX,XX +XXX,XX @@ FIELD(TBFLAG_A64, SVL, 24, 4) FIELD(TBFLAG_A64, SME_TRAP_NONSTREAMING, 28, 1) FIELD(TBFLAG_A64, FGT_ERET, 29, 1) FIELD(TBFLAG_A64, NAA, 30, 1) +FIELD(TBFLAG_A64, ATA0, 31, 1) /* * Helpers for using the above. diff --git a/target/arm/tcg/translate.h b/target/arm/tcg/translate.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate.h +++ b/target/arm/tcg/translate.h @@ -XXX,XX +XXX,XX @@ typedef struct DisasContext { bool unpriv; /* True if v8.3-PAuth is active. */ bool pauth_active; - /* True if v8.5-MTE access to tags is enabled. */ - bool ata; + /* True if v8.5-MTE access to tags is enabled; index with is_unpriv. */ + bool ata[2]; /* True if v8.5-MTE tag checks affect the PE; index with is_unpriv. */ bool mte_active[2]; /* True with v8.5-BTI and SCTLR_ELx.BT* set. */ diff --git a/target/arm/tcg/hflags.c b/target/arm/tcg/hflags.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/hflags.c +++ b/target/arm/tcg/hflags.c @@ -XXX,XX +XXX,XX @@ static CPUARMTBFlags rebuild_hflags_a64(CPUARMState *env, int el, int fp_el, && allocation_tag_access_enabled(env, 0, sctlr)) { DP_TBFLAG_A64(flags, MTE0_ACTIVE, 1); } + /* + * For unpriv tag-setting accesses we alse need ATA0. Again, in + * contexts where unpriv and normal insns are the same we + * duplicate the ATA bit to save effort for translate-a64.c. + */ + if (EX_TBFLAG_A64(flags, UNPRIV)) { + if (allocation_tag_access_enabled(env, 0, sctlr)) { + DP_TBFLAG_A64(flags, ATA0, 1); + } + } else { + DP_TBFLAG_A64(flags, ATA0, EX_TBFLAG_A64(flags, ATA)); + } /* Cache TCMA as well as TBI. */ DP_TBFLAG_A64(flags, TCMA, aa64_va_parameter_tcma(tcr, mmu_idx)); } diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void handle_sys(DisasContext *s, bool isread, clean_addr = clean_data_tbi(s, tcg_rt); gen_probe_access(s, clean_addr, MMU_DATA_STORE, MO_8); - if (s->ata) { + if (s->ata[0]) { /* Extract the tag from the register to match STZGM. */ tag = tcg_temp_new_i64(); tcg_gen_shri_i64(tag, tcg_rt, 56); @@ -XXX,XX +XXX,XX @@ static void handle_sys(DisasContext *s, bool isread, clean_addr = clean_data_tbi(s, tcg_rt); gen_helper_dc_zva(cpu_env, clean_addr); - if (s->ata) { + if (s->ata[0]) { /* Extract the tag from the register to match STZGM. */ tag = tcg_temp_new_i64(); tcg_gen_shri_i64(tag, tcg_rt, 56); @@ -XXX,XX +XXX,XX @@ static bool trans_STGP(DisasContext *s, arg_ldstpair *a) tcg_gen_qemu_st_i128(tmp, clean_addr, get_mem_index(s), mop); /* Perform the tag store, if tag access enabled. */ - if (s->ata) { + if (s->ata[0]) { if (tb_cflags(s->base.tb) & CF_PARALLEL) { gen_helper_stg_parallel(cpu_env, dirty_addr, dirty_addr); } else { @@ -XXX,XX +XXX,XX @@ static bool trans_STZGM(DisasContext *s, arg_ldst_tag *a) tcg_gen_addi_i64(addr, addr, a->imm); tcg_rt = cpu_reg(s, a->rt); - if (s->ata) { + if (s->ata[0]) { gen_helper_stzgm_tags(cpu_env, addr, tcg_rt); } /* @@ -XXX,XX +XXX,XX @@ static bool trans_STGM(DisasContext *s, arg_ldst_tag *a) tcg_gen_addi_i64(addr, addr, a->imm); tcg_rt = cpu_reg(s, a->rt); - if (s->ata) { + if (s->ata[0]) { gen_helper_stgm(cpu_env, addr, tcg_rt); } else { MMUAccessType acc = MMU_DATA_STORE; @@ -XXX,XX +XXX,XX @@ static bool trans_LDGM(DisasContext *s, arg_ldst_tag *a) tcg_gen_addi_i64(addr, addr, a->imm); tcg_rt = cpu_reg(s, a->rt); - if (s->ata) { + if (s->ata[0]) { gen_helper_ldgm(tcg_rt, cpu_env, addr); } else { MMUAccessType acc = MMU_DATA_LOAD; @@ -XXX,XX +XXX,XX @@ static bool trans_LDG(DisasContext *s, arg_ldst_tag *a) tcg_gen_andi_i64(addr, addr, -TAG_GRANULE); tcg_rt = cpu_reg(s, a->rt); - if (s->ata) { + if (s->ata[0]) { gen_helper_ldg(tcg_rt, cpu_env, addr, tcg_rt); } else { /* @@ -XXX,XX +XXX,XX @@ static bool do_STG(DisasContext *s, arg_ldst_tag *a, bool is_zero, bool is_pair) tcg_gen_addi_i64(addr, addr, a->imm); } tcg_rt = cpu_reg_sp(s, a->rt); - if (!s->ata) { + if (!s->ata[0]) { /* * For STG and ST2G, we need to check alignment and probe memory. * TODO: For STZG and STZ2G, we could rely on the stores below, @@ -XXX,XX +XXX,XX @@ static bool gen_add_sub_imm_with_tags(DisasContext *s, arg_rri_tag *a, tcg_rn = cpu_reg_sp(s, a->rn); tcg_rd = cpu_reg_sp(s, a->rd); - if (s->ata) { + if (s->ata[0]) { gen_helper_addsubg(tcg_rd, cpu_env, tcg_rn, tcg_constant_i32(imm), tcg_constant_i32(a->uimm4)); @@ -XXX,XX +XXX,XX @@ static void disas_data_proc_2src(DisasContext *s, uint32_t insn) if (sf == 0 || !dc_isar_feature(aa64_mte_insn_reg, s)) { goto do_unallocated; } - if (s->ata) { + if (s->ata[0]) { gen_helper_irg(cpu_reg_sp(s, rd), cpu_env, cpu_reg_sp(s, rn), cpu_reg(s, rm)); } else { @@ -XXX,XX +XXX,XX @@ static void aarch64_tr_init_disas_context(DisasContextBase *dcbase, dc->bt = EX_TBFLAG_A64(tb_flags, BT); dc->btype = EX_TBFLAG_A64(tb_flags, BTYPE); dc->unpriv = EX_TBFLAG_A64(tb_flags, UNPRIV); - dc->ata = EX_TBFLAG_A64(tb_flags, ATA); + dc->ata[0] = EX_TBFLAG_A64(tb_flags, ATA); + dc->ata[1] = EX_TBFLAG_A64(tb_flags, ATA0); dc->mte_active[0] = EX_TBFLAG_A64(tb_flags, MTE_ACTIVE); dc->mte_active[1] = EX_TBFLAG_A64(tb_flags, MTE0_ACTIVE); dc->pstate_sm = EX_TBFLAG_A64(tb_flags, PSTATE_SM); -- 2.34.1 The FEAT_MOPS SETG* instructions are very similar to the SET* instructions, but as well as setting memory contents they also set the MTE tags. They are architecturally required to operate on tag-granule aligned regions only. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-10-peter.maydell@linaro.org --- target/arm/internals.h | 10 ++++ target/arm/tcg/helper-a64.h | 3 ++ target/arm/tcg/a64.decode | 5 ++ target/arm/tcg/helper-a64.c | 86 ++++++++++++++++++++++++++++++++-- target/arm/tcg/mte_helper.c | 40 ++++++++++++++++ target/arm/tcg/translate-a64.c | 20 +++++--- 6 files changed, 155 insertions(+), 9 deletions(-) diff --git a/target/arm/internals.h b/target/arm/internals.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/internals.h +++ b/target/arm/internals.h @@ -XXX,XX +XXX,XX @@ uint64_t mte_mops_probe(CPUARMState *env, uint64_t ptr, uint64_t size, void mte_check_fail(CPUARMState *env, uint32_t desc, uint64_t dirty_ptr, uintptr_t ra); +/** + * mte_mops_set_tags: Set MTE tags for a portion of a FEAT_MOPS operation + * @env: CPU env + * @dirty_ptr: Start address of memory region (dirty pointer) + * @size: length of region (guaranteed not to cross page boundary) + * @desc: MTEDESC descriptor word + */ +void mte_mops_set_tags(CPUARMState *env, uint64_t dirty_ptr, uint64_t size, + uint32_t desc); + static inline int allocation_tag_from_addr(uint64_t ptr) { return extract64(ptr, 56, 4); diff --git a/target/arm/tcg/helper-a64.h b/target/arm/tcg/helper-a64.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/helper-a64.h +++ b/target/arm/tcg/helper-a64.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_4(unaligned_access, TCG_CALL_NO_WG, DEF_HELPER_3(setp, void, env, i32, i32) DEF_HELPER_3(setm, void, env, i32, i32) DEF_HELPER_3(sete, void, env, i32, i32) +DEF_HELPER_3(setgp, void, env, i32, i32) +DEF_HELPER_3(setgm, void, env, i32, i32) +DEF_HELPER_3(setge, void, env, i32, i32) diff --git a/target/arm/tcg/a64.decode b/target/arm/tcg/a64.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/a64.decode +++ b/target/arm/tcg/a64.decode @@ -XXX,XX +XXX,XX @@ STZ2G 11011001 11 1 ......... 11 ..... ..... @ldst_tag p=0 w=1 SETP 00 011001110 ..... 00 . . 01 ..... ..... @set SETM 00 011001110 ..... 01 . . 01 ..... ..... @set SETE 00 011001110 ..... 10 . . 01 ..... ..... @set + +# Like SET, but also setting MTE tags +SETGP 00 011101110 ..... 00 . . 01 ..... ..... @set +SETGM 00 011101110 ..... 01 . . 01 ..... ..... @set +SETGE 00 011101110 ..... 10 . . 01 ..... ..... @set diff --git a/target/arm/tcg/helper-a64.c b/target/arm/tcg/helper-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/helper-a64.c +++ b/target/arm/tcg/helper-a64.c @@ -XXX,XX +XXX,XX @@ static uint64_t set_step(CPUARMState *env, uint64_t toaddr, return setsize; } +/* + * Similar, but setting tags. The architecture requires us to do this + * in 16-byte chunks. SETP accesses are not tag checked; they set + * the tags. + */ +static uint64_t set_step_tags(CPUARMState *env, uint64_t toaddr, + uint64_t setsize, uint32_t data, int memidx, + uint32_t *mtedesc, uintptr_t ra) +{ + void *mem; + uint64_t cleanaddr; + + setsize = MIN(setsize, page_limit(toaddr)); + + cleanaddr = useronly_clean_ptr(toaddr); + /* + * Trapless lookup: returns NULL for invalid page, I/O, + * watchpoints, clean pages, etc. + */ + mem = tlb_vaddr_to_host(env, cleanaddr, MMU_DATA_STORE, memidx); + +#ifndef CONFIG_USER_ONLY + if (unlikely(!mem)) { + /* + * Slow-path: just do one write. This will handle the + * watchpoint, invalid page, etc handling correctly. + * The architecture requires that we do 16 bytes at a time, + * and we know both ptr and size are 16 byte aligned. + * For clean code pages, the next iteration will see + * the page dirty and will use the fast path. + */ + uint64_t repldata = data * 0x0101010101010101ULL; + MemOpIdx oi16 = make_memop_idx(MO_TE | MO_128, memidx); + cpu_st16_mmu(env, toaddr, int128_make128(repldata, repldata), oi16, ra); + mte_mops_set_tags(env, toaddr, 16, *mtedesc); + return 16; + } +#endif + /* Easy case: just memset the host memory */ + memset(mem, data, setsize); + mte_mops_set_tags(env, toaddr, setsize, *mtedesc); + return setsize; +} + typedef uint64_t StepFn(CPUARMState *env, uint64_t toaddr, uint64_t setsize, uint32_t data, int memidx, uint32_t *mtedesc, uintptr_t ra); @@ -XXX,XX +XXX,XX @@ static bool mte_checks_needed(uint64_t ptr, uint32_t desc) return !tcma_check(desc, bit55, allocation_tag_from_addr(ptr)); } +/* Take an exception if the SETG addr/size are not granule aligned */ +static void check_setg_alignment(CPUARMState *env, uint64_t ptr, uint64_t size, + uint32_t memidx, uintptr_t ra) +{ + if ((size != 0 && !QEMU_IS_ALIGNED(ptr, TAG_GRANULE)) || + !QEMU_IS_ALIGNED(size, TAG_GRANULE)) { + arm_cpu_do_unaligned_access(env_cpu(env), ptr, MMU_DATA_STORE, + memidx, ra); + + } +} + /* * For the Memory Set operation, our implementation chooses * always to use "option A", where we update Xd to the final @@ -XXX,XX +XXX,XX @@ static void do_setp(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc, if (setsize > INT64_MAX) { setsize = INT64_MAX; + if (is_setg) { + setsize &= ~0xf; + } } - if (!mte_checks_needed(toaddr, mtedesc)) { + if (unlikely(is_setg)) { + check_setg_alignment(env, toaddr, setsize, memidx, ra); + } else if (!mte_checks_needed(toaddr, mtedesc)) { mtedesc = 0; } @@ -XXX,XX +XXX,XX @@ void HELPER(setp)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) do_setp(env, syndrome, mtedesc, set_step, false, GETPC()); } +void HELPER(setgp)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) +{ + do_setp(env, syndrome, mtedesc, set_step_tags, true, GETPC()); +} + static void do_setm(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc, StepFn *stepfn, bool is_setg, uintptr_t ra) { @@ -XXX,XX +XXX,XX @@ static void do_setm(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc, * have an IMPDEF check for alignment here. */ - if (!mte_checks_needed(toaddr, mtedesc)) { + if (unlikely(is_setg)) { + check_setg_alignment(env, toaddr, setsize, memidx, ra); + } else if (!mte_checks_needed(toaddr, mtedesc)) { mtedesc = 0; } @@ -XXX,XX +XXX,XX @@ void HELPER(setm)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) do_setm(env, syndrome, mtedesc, set_step, false, GETPC()); } +void HELPER(setgm)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) +{ + do_setm(env, syndrome, mtedesc, set_step_tags, true, GETPC()); +} + static void do_sete(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc, StepFn *stepfn, bool is_setg, uintptr_t ra) { @@ -XXX,XX +XXX,XX @@ static void do_sete(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc, mops_mismatch_exception_target_el(env), ra); } - if (!mte_checks_needed(toaddr, mtedesc)) { + if (unlikely(is_setg)) { + check_setg_alignment(env, toaddr, setsize, memidx, ra); + } else if (!mte_checks_needed(toaddr, mtedesc)) { mtedesc = 0; } @@ -XXX,XX +XXX,XX @@ void HELPER(sete)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) { do_sete(env, syndrome, mtedesc, set_step, false, GETPC()); } + +void HELPER(setge)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) +{ + do_sete(env, syndrome, mtedesc, set_step_tags, true, GETPC()); +} diff --git a/target/arm/tcg/mte_helper.c b/target/arm/tcg/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/mte_helper.c +++ b/target/arm/tcg/mte_helper.c @@ -XXX,XX +XXX,XX @@ uint64_t mte_mops_probe(CPUARMState *env, uint64_t ptr, uint64_t size, return n * TAG_GRANULE - (ptr - tag_first); } } + +void mte_mops_set_tags(CPUARMState *env, uint64_t ptr, uint64_t size, + uint32_t desc) +{ + int mmu_idx, tag_count; + uint64_t ptr_tag; + void *mem; + + if (!desc) { + /* Tags not actually enabled */ + return; + } + + mmu_idx = FIELD_EX32(desc, MTEDESC, MIDX); + /* True probe: this will never fault */ + mem = allocation_tag_mem_probe(env, mmu_idx, ptr, MMU_DATA_STORE, size, + MMU_DATA_STORE, true, 0); + if (!mem) { + return; + } + + /* + * We know that ptr and size are both TAG_GRANULE aligned; store + * the tag from the pointer value into the tag memory. + */ + ptr_tag = allocation_tag_from_addr(ptr); + tag_count = size / TAG_GRANULE; + if (ptr & TAG_GRANULE) { + /* Not 2*TAG_GRANULE-aligned: store tag to first nibble */ + store_tag1_parallel(TAG_GRANULE, mem, ptr_tag); + mem++; + tag_count--; + } + memset(mem, ptr_tag | (ptr_tag << 4), tag_count / 2); + if (tag_count & 1) { + /* Final trailing unaligned nibble */ + mem += tag_count / 2; + store_tag1_parallel(0, mem, ptr_tag); + } +} diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ TRANS_FEAT(STZ2G, aa64_mte_insn_reg, do_STG, a, true, true) typedef void SetFn(TCGv_env, TCGv_i32, TCGv_i32); -static bool do_SET(DisasContext *s, arg_set *a, bool is_epilogue, SetFn fn) +static bool do_SET(DisasContext *s, arg_set *a, bool is_epilogue, + bool is_setg, SetFn fn) { int memidx; uint32_t syndrome, desc = 0; + if (is_setg && !dc_isar_feature(aa64_mte, s)) { + return false; + } + /* * UNPREDICTABLE cases: we choose to UNDEF, which allows * us to pull this check before the CheckMOPSEnabled() test @@ -XXX,XX +XXX,XX @@ static bool do_SET(DisasContext *s, arg_set *a, bool is_epilogue, SetFn fn) * We pass option_a == true, matching our implementation; * we pass wrong_option == false: helper function may set that bit. */ - syndrome = syn_mop(true, false, (a->nontemp << 1) | a->unpriv, + syndrome = syn_mop(true, is_setg, (a->nontemp << 1) | a->unpriv, is_epilogue, false, true, a->rd, a->rs, a->rn); - if (s->mte_active[a->unpriv]) { + if (is_setg ? s->ata[a->unpriv] : s->mte_active[a->unpriv]) { /* We may need to do MTE tag checking, so assemble the descriptor */ desc = FIELD_DP32(desc, MTEDESC, TBI, s->tbid); desc = FIELD_DP32(desc, MTEDESC, TCMA, s->tcma); @@ -XXX,XX +XXX,XX @@ static bool do_SET(DisasContext *s, arg_set *a, bool is_epilogue, SetFn fn) return true; } -TRANS_FEAT(SETP, aa64_mops, do_SET, a, false, gen_helper_setp) -TRANS_FEAT(SETM, aa64_mops, do_SET, a, false, gen_helper_setm) -TRANS_FEAT(SETE, aa64_mops, do_SET, a, true, gen_helper_sete) +TRANS_FEAT(SETP, aa64_mops, do_SET, a, false, false, gen_helper_setp) +TRANS_FEAT(SETM, aa64_mops, do_SET, a, false, false, gen_helper_setm) +TRANS_FEAT(SETE, aa64_mops, do_SET, a, true, false, gen_helper_sete) +TRANS_FEAT(SETGP, aa64_mops, do_SET, a, false, true, gen_helper_setgp) +TRANS_FEAT(SETGM, aa64_mops, do_SET, a, false, true, gen_helper_setgm) +TRANS_FEAT(SETGE, aa64_mops, do_SET, a, true, true, gen_helper_setge) typedef void ArithTwoOp(TCGv_i64, TCGv_i64, TCGv_i64); -- 2.34.1 The FEAT_MOPS memory copy operations need an extra helper routine for checking for MTE tag checking failures beyond the ones we already added for memory set operations: * mte_mops_probe_rev() does the same job as mte_mops_probe(), but it checks tags starting at the provided address and working backwards, rather than forwards Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-11-peter.maydell@linaro.org --- target/arm/internals.h | 17 +++++++ target/arm/tcg/mte_helper.c | 99 +++++++++++++++++++++++++++++++++++++ 2 files changed, 116 insertions(+) diff --git a/target/arm/internals.h b/target/arm/internals.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/internals.h +++ b/target/arm/internals.h @@ -XXX,XX +XXX,XX @@ uint64_t mte_check(CPUARMState *env, uint32_t desc, uint64_t ptr, uintptr_t ra); uint64_t mte_mops_probe(CPUARMState *env, uint64_t ptr, uint64_t size, uint32_t desc); +/** + * mte_mops_probe_rev: Check where the next MTE failure is for a FEAT_MOPS + * operation going in the reverse direction + * @env: CPU env + * @ptr: *end* address of memory region (dirty pointer) + * @size: length of region (guaranteed not to cross a page boundary) + * @desc: MTEDESC descriptor word (0 means no MTE checks) + * Returns: the size of the region that can be copied without hitting + * an MTE tag failure + * + * Note that we assume that the caller has already checked the TBI + * and TCMA bits with mte_checks_needed() and an MTE check is definitely + * required. + */ +uint64_t mte_mops_probe_rev(CPUARMState *env, uint64_t ptr, uint64_t size, + uint32_t desc); + /** * mte_check_fail: Record an MTE tag check failure * @env: CPU env diff --git a/target/arm/tcg/mte_helper.c b/target/arm/tcg/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/mte_helper.c +++ b/target/arm/tcg/mte_helper.c @@ -XXX,XX +XXX,XX @@ static int checkN(uint8_t *mem, int odd, int cmp, int count) return n; } +/** + * checkNrev: + * @tag: tag memory to test + * @odd: true to begin testing at tags at odd nibble + * @cmp: the tag to compare against + * @count: number of tags to test + * + * Return the number of successful tests. + * Thus a return value < @count indicates a failure. + * + * This is like checkN, but it runs backwards, checking the + * tags starting with @tag and then the tags preceding it. + * This is needed by the backwards-memory-copying operations. + */ +static int checkNrev(uint8_t *mem, int odd, int cmp, int count) +{ + int n = 0, diff; + + /* Replicate the test tag and compare. */ + cmp *= 0x11; + diff = *mem-- ^ cmp; + + if (!odd) { + goto start_even; + } + + while (1) { + /* Test odd tag. */ + if (unlikely((diff) & 0xf0)) { + break; + } + if (++n == count) { + break; + } + + start_even: + /* Test even tag. */ + if (unlikely((diff) & 0x0f)) { + break; + } + if (++n == count) { + break; + } + + diff = *mem-- ^ cmp; + } + return n; +} + /** * mte_probe_int() - helper for mte_probe and mte_check * @env: CPU environment @@ -XXX,XX +XXX,XX @@ uint64_t mte_mops_probe(CPUARMState *env, uint64_t ptr, uint64_t size, } } +uint64_t mte_mops_probe_rev(CPUARMState *env, uint64_t ptr, uint64_t size, + uint32_t desc) +{ + int mmu_idx, tag_count; + uint64_t ptr_tag, tag_first, tag_last; + void *mem; + bool w = FIELD_EX32(desc, MTEDESC, WRITE); + uint32_t n; + + mmu_idx = FIELD_EX32(desc, MTEDESC, MIDX); + /* True probe; this will never fault */ + mem = allocation_tag_mem_probe(env, mmu_idx, ptr, + w ? MMU_DATA_STORE : MMU_DATA_LOAD, + size, MMU_DATA_LOAD, true, 0); + if (!mem) { + return size; + } + + /* + * TODO: checkNrev() is not designed for checks of the size we expect + * for FEAT_MOPS operations, so we should implement this differently. + * Maybe we should do something like + * if (region start and size are aligned nicely) { + * do direct loads of 64 tag bits at a time; + * } else { + * call checkN() + * } + */ + /* Round the bounds to the tag granule, and compute the number of tags. */ + ptr_tag = allocation_tag_from_addr(ptr); + tag_first = QEMU_ALIGN_DOWN(ptr - (size - 1), TAG_GRANULE); + tag_last = QEMU_ALIGN_DOWN(ptr, TAG_GRANULE); + tag_count = ((tag_last - tag_first) / TAG_GRANULE) + 1; + n = checkNrev(mem, ptr & TAG_GRANULE, ptr_tag, tag_count); + if (likely(n == tag_count)) { + return size; + } + + /* + * Failure; for the first granule, it's at @ptr. Otherwise + * it's at the last byte of the nth granule. Calculate how + * many bytes we can access without hitting that failure. + */ + if (n == 0) { + return 0; + } else { + return (n - 1) * TAG_GRANULE + ((ptr + 1) - tag_last); + } +} + void mte_mops_set_tags(CPUARMState *env, uint64_t ptr, uint64_t size, uint32_t desc) { -- 2.34.1 The FEAT_MOPS CPY* instructions implement memory copies. These come in both "always forwards" (memcpy-style) and "overlap OK" (memmove-style) flavours. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-12-peter.maydell@linaro.org --- target/arm/tcg/helper-a64.h | 7 + target/arm/tcg/a64.decode | 14 + target/arm/tcg/helper-a64.c | 454 +++++++++++++++++++++++++++++++++ target/arm/tcg/translate-a64.c | 60 +++++ 4 files changed, 535 insertions(+) diff --git a/target/arm/tcg/helper-a64.h b/target/arm/tcg/helper-a64.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/helper-a64.h +++ b/target/arm/tcg/helper-a64.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_3(sete, void, env, i32, i32) DEF_HELPER_3(setgp, void, env, i32, i32) DEF_HELPER_3(setgm, void, env, i32, i32) DEF_HELPER_3(setge, void, env, i32, i32) + +DEF_HELPER_4(cpyp, void, env, i32, i32, i32) +DEF_HELPER_4(cpym, void, env, i32, i32, i32) +DEF_HELPER_4(cpye, void, env, i32, i32, i32) +DEF_HELPER_4(cpyfp, void, env, i32, i32, i32) +DEF_HELPER_4(cpyfm, void, env, i32, i32, i32) +DEF_HELPER_4(cpyfe, void, env, i32, i32, i32) diff --git a/target/arm/tcg/a64.decode b/target/arm/tcg/a64.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/a64.decode +++ b/target/arm/tcg/a64.decode @@ -XXX,XX +XXX,XX @@ SETE 00 011001110 ..... 10 . . 01 ..... ..... @set SETGP 00 011101110 ..... 00 . . 01 ..... ..... @set SETGM 00 011101110 ..... 01 . . 01 ..... ..... @set SETGE 00 011101110 ..... 10 . . 01 ..... ..... @set + +# Memmove/Memcopy: the CPY insns allow overlapping src/dest and +# copy in the correct direction; the CPYF insns always copy forwards. +# +# options has the nontemporal and unpriv bits for src and dest +&cpy rs rn rd options +@cpy .. ... . ..... rs:5 options:4 .. rn:5 rd:5 &cpy + +CPYFP 00 011 0 01000 ..... .... 01 ..... ..... @cpy +CPYFM 00 011 0 01010 ..... .... 01 ..... ..... @cpy +CPYFE 00 011 0 01100 ..... .... 01 ..... ..... @cpy +CPYP 00 011 1 01000 ..... .... 01 ..... ..... @cpy +CPYM 00 011 1 01010 ..... .... 01 ..... ..... @cpy +CPYE 00 011 1 01100 ..... .... 01 ..... ..... @cpy diff --git a/target/arm/tcg/helper-a64.c b/target/arm/tcg/helper-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/helper-a64.c +++ b/target/arm/tcg/helper-a64.c @@ -XXX,XX +XXX,XX @@ static uint64_t page_limit(uint64_t addr) return TARGET_PAGE_ALIGN(addr + 1) - addr; } +/* + * Return the number of bytes we can copy starting from addr and working + * backwards without crossing a page boundary. + */ +static uint64_t page_limit_rev(uint64_t addr) +{ + return (addr & ~TARGET_PAGE_MASK) + 1; +} + /* * Perform part of a memory set on an area of guest memory starting at * toaddr (a dirty address) and extending for setsize bytes. @@ -XXX,XX +XXX,XX @@ void HELPER(setge)(CPUARMState *env, uint32_t syndrome, uint32_t mtedesc) { do_sete(env, syndrome, mtedesc, set_step_tags, true, GETPC()); } + +/* + * Perform part of a memory copy from the guest memory at fromaddr + * and extending for copysize bytes, to the guest memory at + * toaddr. Both addreses are dirty. + * + * Returns the number of bytes actually set, which might be less than + * copysize; the caller should loop until the whole copy has been done. + * The caller should ensure that the guest registers are correct + * for the possibility that the first byte of the copy encounters + * an exception or watchpoint. We guarantee not to take any faults + * for bytes other than the first. + */ +static uint64_t copy_step(CPUARMState *env, uint64_t toaddr, uint64_t fromaddr, + uint64_t copysize, int wmemidx, int rmemidx, + uint32_t *wdesc, uint32_t *rdesc, uintptr_t ra) +{ + void *rmem; + void *wmem; + + /* Don't cross a page boundary on either source or destination */ + copysize = MIN(copysize, page_limit(toaddr)); + copysize = MIN(copysize, page_limit(fromaddr)); + /* + * Handle MTE tag checks: either handle the tag mismatch for byte 0, + * or else copy up to but not including the byte with the mismatch. + */ + if (*rdesc) { + uint64_t mtesize = mte_mops_probe(env, fromaddr, copysize, *rdesc); + if (mtesize == 0) { + mte_check_fail(env, *rdesc, fromaddr, ra); + *rdesc = 0; + } else { + copysize = MIN(copysize, mtesize); + } + } + if (*wdesc) { + uint64_t mtesize = mte_mops_probe(env, toaddr, copysize, *wdesc); + if (mtesize == 0) { + mte_check_fail(env, *wdesc, toaddr, ra); + *wdesc = 0; + } else { + copysize = MIN(copysize, mtesize); + } + } + + toaddr = useronly_clean_ptr(toaddr); + fromaddr = useronly_clean_ptr(fromaddr); + /* Trapless lookup of whether we can get a host memory pointer */ + wmem = tlb_vaddr_to_host(env, toaddr, MMU_DATA_STORE, wmemidx); + rmem = tlb_vaddr_to_host(env, fromaddr, MMU_DATA_LOAD, rmemidx); + +#ifndef CONFIG_USER_ONLY + /* + * If we don't have host memory for both source and dest then just + * do a single byte copy. This will handle watchpoints, invalid pages, + * etc correctly. For clean code pages, the next iteration will see + * the page dirty and will use the fast path. + */ + if (unlikely(!rmem || !wmem)) { + uint8_t byte; + if (rmem) { + byte = *(uint8_t *)rmem; + } else { + byte = cpu_ldub_mmuidx_ra(env, fromaddr, rmemidx, ra); + } + if (wmem) { + *(uint8_t *)wmem = byte; + } else { + cpu_stb_mmuidx_ra(env, toaddr, byte, wmemidx, ra); + } + return 1; + } +#endif + /* Easy case: just memmove the host memory */ + memmove(wmem, rmem, copysize); + return copysize; +} + +/* + * Do part of a backwards memory copy. Here toaddr and fromaddr point + * to the *last* byte to be copied. + */ +static uint64_t copy_step_rev(CPUARMState *env, uint64_t toaddr, + uint64_t fromaddr, + uint64_t copysize, int wmemidx, int rmemidx, + uint32_t *wdesc, uint32_t *rdesc, uintptr_t ra) +{ + void *rmem; + void *wmem; + + /* Don't cross a page boundary on either source or destination */ + copysize = MIN(copysize, page_limit_rev(toaddr)); + copysize = MIN(copysize, page_limit_rev(fromaddr)); + + /* + * Handle MTE tag checks: either handle the tag mismatch for byte 0, + * or else copy up to but not including the byte with the mismatch. + */ + if (*rdesc) { + uint64_t mtesize = mte_mops_probe_rev(env, fromaddr, copysize, *rdesc); + if (mtesize == 0) { + mte_check_fail(env, *rdesc, fromaddr, ra); + *rdesc = 0; + } else { + copysize = MIN(copysize, mtesize); + } + } + if (*wdesc) { + uint64_t mtesize = mte_mops_probe_rev(env, toaddr, copysize, *wdesc); + if (mtesize == 0) { + mte_check_fail(env, *wdesc, toaddr, ra); + *wdesc = 0; + } else { + copysize = MIN(copysize, mtesize); + } + } + + toaddr = useronly_clean_ptr(toaddr); + fromaddr = useronly_clean_ptr(fromaddr); + /* Trapless lookup of whether we can get a host memory pointer */ + wmem = tlb_vaddr_to_host(env, toaddr, MMU_DATA_STORE, wmemidx); + rmem = tlb_vaddr_to_host(env, fromaddr, MMU_DATA_LOAD, rmemidx); + +#ifndef CONFIG_USER_ONLY + /* + * If we don't have host memory for both source and dest then just + * do a single byte copy. This will handle watchpoints, invalid pages, + * etc correctly. For clean code pages, the next iteration will see + * the page dirty and will use the fast path. + */ + if (unlikely(!rmem || !wmem)) { + uint8_t byte; + if (rmem) { + byte = *(uint8_t *)rmem; + } else { + byte = cpu_ldub_mmuidx_ra(env, fromaddr, rmemidx, ra); + } + if (wmem) { + *(uint8_t *)wmem = byte; + } else { + cpu_stb_mmuidx_ra(env, toaddr, byte, wmemidx, ra); + } + return 1; + } +#endif + /* + * Easy case: just memmove the host memory. Note that wmem and + * rmem here point to the *last* byte to copy. + */ + memmove(wmem - (copysize - 1), rmem - (copysize - 1), copysize); + return copysize; +} + +/* + * for the Memory Copy operation, our implementation chooses always + * to use "option A", where we update Xd and Xs to the final addresses + * in the CPYP insn, and then in CPYM and CPYE only need to update Xn. + * + * @env: CPU + * @syndrome: syndrome value for mismatch exceptions + * (also contains the register numbers we need to use) + * @wdesc: MTE descriptor for the writes (destination) + * @rdesc: MTE descriptor for the reads (source) + * @move: true if this is CPY (memmove), false for CPYF (memcpy forwards) + */ +static void do_cpyp(CPUARMState *env, uint32_t syndrome, uint32_t wdesc, + uint32_t rdesc, uint32_t move, uintptr_t ra) +{ + int rd = mops_destreg(syndrome); + int rs = mops_srcreg(syndrome); + int rn = mops_sizereg(syndrome); + uint32_t rmemidx = FIELD_EX32(rdesc, MTEDESC, MIDX); + uint32_t wmemidx = FIELD_EX32(wdesc, MTEDESC, MIDX); + bool forwards = true; + uint64_t toaddr = env->xregs[rd]; + uint64_t fromaddr = env->xregs[rs]; + uint64_t copysize = env->xregs[rn]; + uint64_t stagecopysize, step; + + check_mops_enabled(env, ra); + + + if (move) { + /* + * Copy backwards if necessary. The direction for a non-overlapping + * copy is IMPDEF; we choose forwards. + */ + if (copysize > 0x007FFFFFFFFFFFFFULL) { + copysize = 0x007FFFFFFFFFFFFFULL; + } + uint64_t fs = extract64(fromaddr, 0, 56); + uint64_t ts = extract64(toaddr, 0, 56); + uint64_t fe = extract64(fromaddr + copysize, 0, 56); + + if (fs < ts && fe > ts) { + forwards = false; + } + } else { + if (copysize > INT64_MAX) { + copysize = INT64_MAX; + } + } + + if (!mte_checks_needed(fromaddr, rdesc)) { + rdesc = 0; + } + if (!mte_checks_needed(toaddr, wdesc)) { + wdesc = 0; + } + + if (forwards) { + stagecopysize = MIN(copysize, page_limit(toaddr)); + stagecopysize = MIN(stagecopysize, page_limit(fromaddr)); + while (stagecopysize) { + env->xregs[rd] = toaddr; + env->xregs[rs] = fromaddr; + env->xregs[rn] = copysize; + step = copy_step(env, toaddr, fromaddr, stagecopysize, + wmemidx, rmemidx, &wdesc, &rdesc, ra); + toaddr += step; + fromaddr += step; + copysize -= step; + stagecopysize -= step; + } + /* Insn completed, so update registers to the Option A format */ + env->xregs[rd] = toaddr + copysize; + env->xregs[rs] = fromaddr + copysize; + env->xregs[rn] = -copysize; + } else { + /* + * In a reverse copy the to and from addrs in Xs and Xd are the start + * of the range, but it's more convenient for us to work with pointers + * to the last byte being copied. + */ + toaddr += copysize - 1; + fromaddr += copysize - 1; + stagecopysize = MIN(copysize, page_limit_rev(toaddr)); + stagecopysize = MIN(stagecopysize, page_limit_rev(fromaddr)); + while (stagecopysize) { + env->xregs[rn] = copysize; + step = copy_step_rev(env, toaddr, fromaddr, stagecopysize, + wmemidx, rmemidx, &wdesc, &rdesc, ra); + copysize -= step; + stagecopysize -= step; + toaddr -= step; + fromaddr -= step; + } + /* + * Insn completed, so update registers to the Option A format. + * For a reverse copy this is no different to the CPYP input format. + */ + env->xregs[rn] = copysize; + } + + /* Set NZCV = 0000 to indicate we are an Option A implementation */ + env->NF = 0; + env->ZF = 1; /* our env->ZF encoding is inverted */ + env->CF = 0; + env->VF = 0; + return; +} + +void HELPER(cpyp)(CPUARMState *env, uint32_t syndrome, uint32_t wdesc, + uint32_t rdesc) +{ + do_cpyp(env, syndrome, wdesc, rdesc, true, GETPC()); +} + +void HELPER(cpyfp)(CPUARMState *env, uint32_t syndrome, uint32_t wdesc, + uint32_t rdesc) +{ + do_cpyp(env, syndrome, wdesc, rdesc, false, GETPC()); +} + +static void do_cpym(CPUARMState *env, uint32_t syndrome, uint32_t wdesc, + uint32_t rdesc, uint32_t move, uintptr_t ra) +{ + /* Main: we choose to copy until less than a page remaining */ + CPUState *cs = env_cpu(env); + int rd = mops_destreg(syndrome); + int rs = mops_srcreg(syndrome); + int rn = mops_sizereg(syndrome); + uint32_t rmemidx = FIELD_EX32(rdesc, MTEDESC, MIDX); + uint32_t wmemidx = FIELD_EX32(wdesc, MTEDESC, MIDX); + bool forwards = true; + uint64_t toaddr, fromaddr, copysize, step; + + check_mops_enabled(env, ra); + + /* We choose to NOP out "no data to copy" before consistency checks */ + if (env->xregs[rn] == 0) { + return; + } + + check_mops_wrong_option(env, syndrome, ra); + + if (move) { + forwards = (int64_t)env->xregs[rn] < 0; + } + + if (forwards) { + toaddr = env->xregs[rd] + env->xregs[rn]; + fromaddr = env->xregs[rs] + env->xregs[rn]; + copysize = -env->xregs[rn]; + } else { + copysize = env->xregs[rn]; + /* This toaddr and fromaddr point to the *last* byte to copy */ + toaddr = env->xregs[rd] + copysize - 1; + fromaddr = env->xregs[rs] + copysize - 1; + } + + if (!mte_checks_needed(fromaddr, rdesc)) { + rdesc = 0; + } + if (!mte_checks_needed(toaddr, wdesc)) { + wdesc = 0; + } + + /* Our implementation has no particular parameter requirements for CPYM */ + + /* Do the actual memmove */ + if (forwards) { + while (copysize >= TARGET_PAGE_SIZE) { + step = copy_step(env, toaddr, fromaddr, copysize, + wmemidx, rmemidx, &wdesc, &rdesc, ra); + toaddr += step; + fromaddr += step; + copysize -= step; + env->xregs[rn] = -copysize; + if (copysize >= TARGET_PAGE_SIZE && + unlikely(cpu_loop_exit_requested(cs))) { + cpu_loop_exit_restore(cs, ra); + } + } + } else { + while (copysize >= TARGET_PAGE_SIZE) { + step = copy_step_rev(env, toaddr, fromaddr, copysize, + wmemidx, rmemidx, &wdesc, &rdesc, ra); + toaddr -= step; + fromaddr -= step; + copysize -= step; + env->xregs[rn] = copysize; + if (copysize >= TARGET_PAGE_SIZE && + unlikely(cpu_loop_exit_requested(cs))) { + cpu_loop_exit_restore(cs, ra); + } + } + } +} + +void HELPER(cpym)(CPUARMState *env, uint32_t syndrome, uint32_t wdesc, + uint32_t rdesc) +{ + do_cpym(env, syndrome, wdesc, rdesc, true, GETPC()); +} + +void HELPER(cpyfm)(CPUARMState *env, uint32_t syndrome, uint32_t wdesc, + uint32_t rdesc) +{ + do_cpym(env, syndrome, wdesc, rdesc, false, GETPC()); +} + +static void do_cpye(CPUARMState *env, uint32_t syndrome, uint32_t wdesc, + uint32_t rdesc, uint32_t move, uintptr_t ra) +{ + /* Epilogue: do the last partial page */ + int rd = mops_destreg(syndrome); + int rs = mops_srcreg(syndrome); + int rn = mops_sizereg(syndrome); + uint32_t rmemidx = FIELD_EX32(rdesc, MTEDESC, MIDX); + uint32_t wmemidx = FIELD_EX32(wdesc, MTEDESC, MIDX); + bool forwards = true; + uint64_t toaddr, fromaddr, copysize, step; + + check_mops_enabled(env, ra); + + /* We choose to NOP out "no data to copy" before consistency checks */ + if (env->xregs[rn] == 0) { + return; + } + + check_mops_wrong_option(env, syndrome, ra); + + if (move) { + forwards = (int64_t)env->xregs[rn] < 0; + } + + if (forwards) { + toaddr = env->xregs[rd] + env->xregs[rn]; + fromaddr = env->xregs[rs] + env->xregs[rn]; + copysize = -env->xregs[rn]; + } else { + copysize = env->xregs[rn]; + /* This toaddr and fromaddr point to the *last* byte to copy */ + toaddr = env->xregs[rd] + copysize - 1; + fromaddr = env->xregs[rs] + copysize - 1; + } + + if (!mte_checks_needed(fromaddr, rdesc)) { + rdesc = 0; + } + if (!mte_checks_needed(toaddr, wdesc)) { + wdesc = 0; + } + + /* Check the size; we don't want to have do a check-for-interrupts */ + if (copysize >= TARGET_PAGE_SIZE) { + raise_exception_ra(env, EXCP_UDEF, syndrome, + mops_mismatch_exception_target_el(env), ra); + } + + /* Do the actual memmove */ + if (forwards) { + while (copysize > 0) { + step = copy_step(env, toaddr, fromaddr, copysize, + wmemidx, rmemidx, &wdesc, &rdesc, ra); + toaddr += step; + fromaddr += step; + copysize -= step; + env->xregs[rn] = -copysize; + } + } else { + while (copysize > 0) { + step = copy_step_rev(env, toaddr, fromaddr, copysize, + wmemidx, rmemidx, &wdesc, &rdesc, ra); + toaddr -= step; + fromaddr -= step; + copysize -= step; + env->xregs[rn] = copysize; + } + } +} + +void HELPER(cpye)(CPUARMState *env, uint32_t syndrome, uint32_t wdesc, + uint32_t rdesc) +{ + do_cpye(env, syndrome, wdesc, rdesc, true, GETPC()); +} + +void HELPER(cpyfe)(CPUARMState *env, uint32_t syndrome, uint32_t wdesc, + uint32_t rdesc) +{ + do_cpye(env, syndrome, wdesc, rdesc, false, GETPC()); +} diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ TRANS_FEAT(SETGP, aa64_mops, do_SET, a, false, true, gen_helper_setgp) TRANS_FEAT(SETGM, aa64_mops, do_SET, a, false, true, gen_helper_setgm) TRANS_FEAT(SETGE, aa64_mops, do_SET, a, true, true, gen_helper_setge) +typedef void CpyFn(TCGv_env, TCGv_i32, TCGv_i32, TCGv_i32); + +static bool do_CPY(DisasContext *s, arg_cpy *a, bool is_epilogue, CpyFn fn) +{ + int rmemidx, wmemidx; + uint32_t syndrome, rdesc = 0, wdesc = 0; + bool wunpriv = extract32(a->options, 0, 1); + bool runpriv = extract32(a->options, 1, 1); + + /* + * UNPREDICTABLE cases: we choose to UNDEF, which allows + * us to pull this check before the CheckMOPSEnabled() test + * (which we do in the helper function) + */ + if (a->rs == a->rn || a->rs == a->rd || a->rn == a->rd || + a->rd == 31 || a->rs == 31 || a->rn == 31) { + return false; + } + + rmemidx = get_a64_user_mem_index(s, runpriv); + wmemidx = get_a64_user_mem_index(s, wunpriv); + + /* + * We pass option_a == true, matching our implementation; + * we pass wrong_option == false: helper function may set that bit. + */ + syndrome = syn_mop(false, false, a->options, is_epilogue, + false, true, a->rd, a->rs, a->rn); + + /* If we need to do MTE tag checking, assemble the descriptors */ + if (s->mte_active[runpriv]) { + rdesc = FIELD_DP32(rdesc, MTEDESC, TBI, s->tbid); + rdesc = FIELD_DP32(rdesc, MTEDESC, TCMA, s->tcma); + } + if (s->mte_active[wunpriv]) { + wdesc = FIELD_DP32(wdesc, MTEDESC, TBI, s->tbid); + wdesc = FIELD_DP32(wdesc, MTEDESC, TCMA, s->tcma); + wdesc = FIELD_DP32(wdesc, MTEDESC, WRITE, true); + } + /* The helper function needs these parts of the descriptor regardless */ + rdesc = FIELD_DP32(rdesc, MTEDESC, MIDX, rmemidx); + wdesc = FIELD_DP32(wdesc, MTEDESC, MIDX, wmemidx); + + /* + * The helper needs the register numbers, but since they're in + * the syndrome anyway, we let it extract them from there rather + * than passing in an extra three integer arguments. + */ + fn(cpu_env, tcg_constant_i32(syndrome), tcg_constant_i32(wdesc), + tcg_constant_i32(rdesc)); + return true; +} + +TRANS_FEAT(CPYP, aa64_mops, do_CPY, a, false, gen_helper_cpyp) +TRANS_FEAT(CPYM, aa64_mops, do_CPY, a, false, gen_helper_cpym) +TRANS_FEAT(CPYE, aa64_mops, do_CPY, a, true, gen_helper_cpye) +TRANS_FEAT(CPYFP, aa64_mops, do_CPY, a, false, gen_helper_cpyfp) +TRANS_FEAT(CPYFM, aa64_mops, do_CPY, a, false, gen_helper_cpyfm) +TRANS_FEAT(CPYFE, aa64_mops, do_CPY, a, true, gen_helper_cpyfe) + typedef void ArithTwoOp(TCGv_i64, TCGv_i64, TCGv_i64); static bool gen_rri(DisasContext *s, arg_rri_sf *a, -- 2.34.1 Enable FEAT_MOPS on the AArch64 'max' CPU, and add it to the list of features we implement. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20230912140434.1333369-13-peter.maydell@linaro.org --- docs/system/arm/emulation.rst | 1 + linux-user/elfload.c | 1 + target/arm/tcg/cpu64.c | 1 + 3 files changed, 3 insertions(+) diff --git a/docs/system/arm/emulation.rst b/docs/system/arm/emulation.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/emulation.rst +++ b/docs/system/arm/emulation.rst @@ -XXX,XX +XXX,XX @@ the following architecture extensions: - FEAT_LSE (Large System Extensions) - FEAT_LSE2 (Large System Extensions v2) - FEAT_LVA (Large Virtual Address space) +- FEAT_MOPS (Standardization of memory operations) - FEAT_MTE (Memory Tagging Extension) - FEAT_MTE2 (Memory Tagging Extension) - FEAT_MTE3 (MTE Asymmetric Fault Handling) diff --git a/linux-user/elfload.c b/linux-user/elfload.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/elfload.c +++ b/linux-user/elfload.c @@ -XXX,XX +XXX,XX @@ uint32_t get_elf_hwcap2(void) GET_FEATURE_ID(aa64_sme_i16i64, ARM_HWCAP2_A64_SME_I16I64); GET_FEATURE_ID(aa64_sme_fa64, ARM_HWCAP2_A64_SME_FA64); GET_FEATURE_ID(aa64_hbc, ARM_HWCAP2_A64_HBC); + GET_FEATURE_ID(aa64_mops, ARM_HWCAP2_A64_MOPS); return hwcaps; } diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu64.c +++ b/target/arm/tcg/cpu64.c @@ -XXX,XX +XXX,XX @@ void aarch64_max_tcg_initfn(Object *obj) cpu->isar.id_aa64isar1 = t; t = cpu->isar.id_aa64isar2; + t = FIELD_DP64(t, ID_AA64ISAR2, MOPS, 1); /* FEAT_MOPS */ t = FIELD_DP64(t, ID_AA64ISAR2, BC, 1); /* FEAT_HBC */ cpu->isar.id_aa64isar2 = t; -- 2.34.1 Avoid a dynamic stack allocation in qjack_client_init(), by using a g_autofree heap allocation instead. (We stick with allocate + snprintf() because the JACK API requires the name to be no more than its maximum size, so g_strdup_printf() would require an extra truncation step.) The codebase has very few VLAs, and if we can get rid of them all we can make the compiler error on new additions. This is a defensive measure against security bugs where an on-stack dynamic allocation isn't correctly size-checked (e.g. CVE-2021-3527). Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Reviewed-by: Francisco Iglesias <frasse.iglesias@gmail.com> Reviewed-by: Christian Schoenebeck <qemu_oss@crudebyte.com> Message-id: 20230818155846.1651287-2-peter.maydell@linaro.org --- audio/jackaudio.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/audio/jackaudio.c b/audio/jackaudio.c index XXXXXXX..XXXXXXX 100644 --- a/audio/jackaudio.c +++ b/audio/jackaudio.c @@ -XXX,XX +XXX,XX @@ static void qjack_client_connect_ports(QJackClient *c) static int qjack_client_init(QJackClient *c) { jack_status_t status; - char client_name[jack_client_name_size()]; + int client_name_len = jack_client_name_size(); /* includes NUL */ + g_autofree char *client_name = g_new(char, client_name_len); jack_options_t options = JackNullOption; if (c->state == QJACK_STATE_RUNNING) { @@ -XXX,XX +XXX,XX @@ static int qjack_client_init(QJackClient *c) c->connect_ports = true; - snprintf(client_name, sizeof(client_name), "%s-%s", + snprintf(client_name, client_name_len, "%s-%s", c->out ? "out" : "in", c->opt->client_name ? c->opt->client_name : audio_application_name()); -- 2.34.1 Avoid a dynamic stack allocation in qjack_process(). Since this function is a JACK process callback, we are not permitted to malloc() here, so we allocate a working buffer in qjack_client_init() instead. The codebase has very few VLAs, and if we can get rid of them all we can make the compiler error on new additions. This is a defensive measure against security bugs where an on-stack dynamic allocation isn't correctly size-checked (e.g. CVE-2021-3527). Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Reviewed-by: Francisco Iglesias <frasse.iglesias@gmail.com> Reviewed-by: Christian Schoenebeck <qemu_oss@crudebyte.com> Message-id: 20230818155846.1651287-3-peter.maydell@linaro.org --- audio/jackaudio.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/audio/jackaudio.c b/audio/jackaudio.c index XXXXXXX..XXXXXXX 100644 --- a/audio/jackaudio.c +++ b/audio/jackaudio.c @@ -XXX,XX +XXX,XX @@ typedef struct QJackClient { int buffersize; jack_port_t **port; QJackBuffer fifo; + + /* Used as workspace by qjack_process() */ + float **process_buffers; } QJackClient; @@ -XXX,XX +XXX,XX @@ static int qjack_process(jack_nframes_t nframes, void *arg) } /* get the buffers for the ports */ - float *buffers[c->nchannels]; for (int i = 0; i < c->nchannels; ++i) { - buffers[i] = jack_port_get_buffer(c->port[i], nframes); + c->process_buffers[i] = jack_port_get_buffer(c->port[i], nframes); } if (c->out) { if (likely(c->enabled)) { - qjack_buffer_read_l(&c->fifo, buffers, nframes); + qjack_buffer_read_l(&c->fifo, c->process_buffers, nframes); } else { for (int i = 0; i < c->nchannels; ++i) { - memset(buffers[i], 0, nframes * sizeof(float)); + memset(c->process_buffers[i], 0, nframes * sizeof(float)); } } } else { if (likely(c->enabled)) { - qjack_buffer_write_l(&c->fifo, buffers, nframes); + qjack_buffer_write_l(&c->fifo, c->process_buffers, nframes); } } @@ -XXX,XX +XXX,XX @@ static int qjack_client_init(QJackClient *c) jack_get_client_name(c->client)); } + /* Allocate working buffer for process callback */ + c->process_buffers = g_new(float *, c->nchannels); + jack_set_process_callback(c->client, qjack_process , c); jack_set_port_registration_callback(c->client, qjack_port_registration, c); jack_set_xrun_callback(c->client, qjack_xrun, c); @@ -XXX,XX +XXX,XX @@ static void qjack_client_fini_locked(QJackClient *c) qjack_buffer_free(&c->fifo); g_free(c->port); + g_free(c->process_buffers); c->state = QJACK_STATE_DISCONNECTED; /* fallthrough */ -- 2.34.1 From: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> Armv8.1+ cpus have Virtual Host Extension (VHE) which added non-secure EL2 virtual timer. This change adds it to fullfil Arm BSA (Base System Architecture) requirements. Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> Message-id: 20230913140610.214893-2-marcin.juszkiewicz@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/sbsa-ref.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/hw/arm/sbsa-ref.c b/hw/arm/sbsa-ref.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/sbsa-ref.c +++ b/hw/arm/sbsa-ref.c @@ -XXX,XX +XXX,XX @@ #define ARCH_TIMER_S_EL1_IRQ 13 #define ARCH_TIMER_NS_EL1_IRQ 14 #define ARCH_TIMER_NS_EL2_IRQ 10 +#define ARCH_TIMER_NS_EL2_VIRT_IRQ 12 enum { SBSA_FLASH, @@ -XXX,XX +XXX,XX @@ static void create_gic(SBSAMachineState *sms, MemoryRegion *mem) [GTIMER_VIRT] = ARCH_TIMER_VIRT_IRQ, [GTIMER_HYP] = ARCH_TIMER_NS_EL2_IRQ, [GTIMER_SEC] = ARCH_TIMER_S_EL1_IRQ, + [GTIMER_HYPVIRT] = ARCH_TIMER_NS_EL2_VIRT_IRQ, }; for (irq = 0; irq < ARRAY_SIZE(timer_irq); irq++) { -- 2.34.1 From: Viktor Prutyanov <viktor@daynix.com> PE export name check introduced in d399d6b179 isn't reliable enough, because a page with the export directory may be not present for some reason. On the other hand, elf2dmp retrieves the PDB name in any case. It can be also used to check that a PE image is the kernel image. So, check PDB name when searching for Windows kernel image. Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2165917 Signed-off-by: Viktor Prutyanov <viktor@daynix.com> Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com> Message-id: 20230915170153.10959-2-viktor@daynix.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- contrib/elf2dmp/main.c | 93 +++++++++++++++--------------------------- 1 file changed, 33 insertions(+), 60 deletions(-) diff --git a/contrib/elf2dmp/main.c b/contrib/elf2dmp/main.c index XXXXXXX..XXXXXXX 100644 --- a/contrib/elf2dmp/main.c +++ b/contrib/elf2dmp/main.c @@ -XXX,XX +XXX,XX @@ static int write_dump(struct pa_space *ps, return fclose(dmp_file); } -static bool pe_check_export_name(uint64_t base, void *start_addr, - struct va_space *vs) -{ - IMAGE_EXPORT_DIRECTORY export_dir; - const char *pe_name; - - if (pe_get_data_dir_entry(base, start_addr, IMAGE_FILE_EXPORT_DIRECTORY, - &export_dir, sizeof(export_dir), vs)) { - return false; - } - - pe_name = va_space_resolve(vs, base + export_dir.Name); - if (!pe_name) { - return false; - } - - return !strcmp(pe_name, PE_NAME); -} - -static int pe_get_pdb_symstore_hash(uint64_t base, void *start_addr, - char *hash, struct va_space *vs) +static bool pe_check_pdb_name(uint64_t base, void *start_addr, + struct va_space *vs, OMFSignatureRSDS *rsds) { const char sign_rsds[4] = "RSDS"; IMAGE_DEBUG_DIRECTORY debug_dir; - OMFSignatureRSDS rsds; - char *pdb_name; - size_t pdb_name_sz; - size_t i; + char pdb_name[sizeof(PDB_NAME)]; if (pe_get_data_dir_entry(base, start_addr, IMAGE_FILE_DEBUG_DIRECTORY, &debug_dir, sizeof(debug_dir), vs)) { eprintf("Failed to get Debug Directory\n"); - return 1; + return false; } if (debug_dir.Type != IMAGE_DEBUG_TYPE_CODEVIEW) { - return 1; + eprintf("Debug Directory type is not CodeView\n"); + return false; } if (va_space_rw(vs, base + debug_dir.AddressOfRawData, - &rsds, sizeof(rsds), 0)) { - return 1; + rsds, sizeof(*rsds), 0)) { + eprintf("Failed to resolve OMFSignatureRSDS\n"); + return false; } - printf("CodeView signature is \'%.4s\'\n", rsds.Signature); - - if (memcmp(&rsds.Signature, sign_rsds, sizeof(sign_rsds))) { - return 1; + if (memcmp(&rsds->Signature, sign_rsds, sizeof(sign_rsds))) { + eprintf("CodeView signature is \'%.4s\', \'%s\' expected\n", + rsds->Signature, sign_rsds); + return false; } - pdb_name_sz = debug_dir.SizeOfData - sizeof(rsds); - pdb_name = malloc(pdb_name_sz); - if (!pdb_name) { - return 1; + if (debug_dir.SizeOfData - sizeof(*rsds) != sizeof(PDB_NAME)) { + eprintf("PDB name size doesn't match\n"); + return false; } if (va_space_rw(vs, base + debug_dir.AddressOfRawData + - offsetof(OMFSignatureRSDS, name), pdb_name, pdb_name_sz, 0)) { - free(pdb_name); - return 1; + offsetof(OMFSignatureRSDS, name), pdb_name, sizeof(PDB_NAME), + 0)) { + eprintf("Failed to resolve PDB name\n"); + return false; } printf("PDB name is \'%s\', \'%s\' expected\n", pdb_name, PDB_NAME); - if (strcmp(pdb_name, PDB_NAME)) { - eprintf("Unexpected PDB name, it seems the kernel isn't found\n"); - free(pdb_name); - return 1; - } + return !strcmp(pdb_name, PDB_NAME); +} - free(pdb_name); - - sprintf(hash, "%.08x%.04x%.04x%.02x%.02x", rsds.guid.a, rsds.guid.b, - rsds.guid.c, rsds.guid.d[0], rsds.guid.d[1]); +static void pe_get_pdb_symstore_hash(OMFSignatureRSDS *rsds, char *hash) +{ + sprintf(hash, "%.08x%.04x%.04x%.02x%.02x", rsds->guid.a, rsds->guid.b, + rsds->guid.c, rsds->guid.d[0], rsds->guid.d[1]); hash += 20; - for (i = 0; i < 6; i++, hash += 2) { - sprintf(hash, "%.02x", rsds.guid.e[i]); + for (unsigned int i = 0; i < 6; i++, hash += 2) { + sprintf(hash, "%.02x", rsds->guid.e[i]); } - sprintf(hash, "%.01x", rsds.age); - - return 0; + sprintf(hash, "%.01x", rsds->age); } int main(int argc, char *argv[]) @@ -XXX,XX +XXX,XX @@ int main(int argc, char *argv[]) KDDEBUGGER_DATA64 *kdbg; uint64_t KdVersionBlock; bool kernel_found = false; + OMFSignatureRSDS rsds; if (argc != 3) { eprintf("usage:\n\t%s elf_file dmp_file\n", argv[0]); @@ -XXX,XX +XXX,XX @@ int main(int argc, char *argv[]) } if (*(uint16_t *)nt_start_addr == 0x5a4d) { /* MZ */ - if (pe_check_export_name(KernBase, nt_start_addr, &vs)) { + printf("Checking candidate KernBase = 0x%016"PRIx64"\n", KernBase); + if (pe_check_pdb_name(KernBase, nt_start_addr, &vs, &rsds)) { kernel_found = true; break; } @@ -XXX,XX +XXX,XX @@ int main(int argc, char *argv[]) printf("KernBase = 0x%016"PRIx64", signature is \'%.2s\'\n", KernBase, (char *)nt_start_addr); - if (pe_get_pdb_symstore_hash(KernBase, nt_start_addr, pdb_hash, &vs)) { - eprintf("Failed to get PDB symbol store hash\n"); - err = 1; - goto out_ps; - } + pe_get_pdb_symstore_hash(&rsds, pdb_hash); sprintf(pdb_url, "%s%s/%s/%s", SYM_URL_BASE, PDB_NAME, pdb_hash, PDB_NAME); printf("PDB URL is %s\n", pdb_url); -- 2.34.1 From: Viktor Prutyanov <viktor@daynix.com> Physical memory ranges may not be aligned to page size in QEMU ELF, but DMP can only contain page-aligned runs. So, align them. Signed-off-by: Viktor Prutyanov <viktor@daynix.com> Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com> Message-id: 20230915170153.10959-3-viktor@daynix.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- contrib/elf2dmp/addrspace.h | 1 + contrib/elf2dmp/addrspace.c | 31 +++++++++++++++++++++++++++++-- contrib/elf2dmp/main.c | 5 +++-- 3 files changed, 33 insertions(+), 4 deletions(-) diff --git a/contrib/elf2dmp/addrspace.h b/contrib/elf2dmp/addrspace.h index XXXXXXX..XXXXXXX 100644 --- a/contrib/elf2dmp/addrspace.h +++ b/contrib/elf2dmp/addrspace.h @@ -XXX,XX +XXX,XX @@ #define ELF2DMP_PAGE_BITS 12 #define ELF2DMP_PAGE_SIZE (1ULL << ELF2DMP_PAGE_BITS) +#define ELF2DMP_PAGE_MASK (ELF2DMP_PAGE_SIZE - 1) #define ELF2DMP_PFN_MASK (~(ELF2DMP_PAGE_SIZE - 1)) #define INVALID_PA UINT64_MAX diff --git a/contrib/elf2dmp/addrspace.c b/contrib/elf2dmp/addrspace.c index XXXXXXX..XXXXXXX 100644 --- a/contrib/elf2dmp/addrspace.c +++ b/contrib/elf2dmp/addrspace.c @@ -XXX,XX +XXX,XX @@ static struct pa_block *pa_space_find_block(struct pa_space *ps, uint64_t pa) for (i = 0; i < ps->block_nr; i++) { if (ps->block[i].paddr <= pa && - pa <= ps->block[i].paddr + ps->block[i].size) { + pa < ps->block[i].paddr + ps->block[i].size) { return ps->block + i; } } @@ -XXX,XX +XXX,XX @@ static uint8_t *pa_space_resolve(struct pa_space *ps, uint64_t pa) return block->addr + (pa - block->paddr); } +static void pa_block_align(struct pa_block *b) +{ + uint64_t low_align = ((b->paddr - 1) | ELF2DMP_PAGE_MASK) + 1 - b->paddr; + uint64_t high_align = (b->paddr + b->size) & ELF2DMP_PAGE_MASK; + + if (low_align == 0 && high_align == 0) { + return; + } + + if (low_align + high_align < b->size) { + printf("Block 0x%"PRIx64"+:0x%"PRIx64" will be aligned to " + "0x%"PRIx64"+:0x%"PRIx64"\n", b->paddr, b->size, + b->paddr + low_align, b->size - low_align - high_align); + b->size -= low_align + high_align; + } else { + printf("Block 0x%"PRIx64"+:0x%"PRIx64" is too small to align\n", + b->paddr, b->size); + b->size = 0; + } + + b->addr += low_align; + b->paddr += low_align; +} + int pa_space_create(struct pa_space *ps, QEMU_Elf *qemu_elf) { Elf64_Half phdr_nr = elf_getphdrnum(qemu_elf->map); @@ -XXX,XX +XXX,XX @@ int pa_space_create(struct pa_space *ps, QEMU_Elf *qemu_elf) .paddr = phdr[i].p_paddr, .size = phdr[i].p_filesz, }; - block_i++; + pa_block_align(&ps->block[block_i]); + block_i = ps->block[block_i].size ? (block_i + 1) : block_i; } } + ps->block_nr = block_i; + return 0; } diff --git a/contrib/elf2dmp/main.c b/contrib/elf2dmp/main.c index XXXXXXX..XXXXXXX 100644 --- a/contrib/elf2dmp/main.c +++ b/contrib/elf2dmp/main.c @@ -XXX,XX +XXX,XX @@ static int write_dump(struct pa_space *ps, for (i = 0; i < ps->block_nr; i++) { struct pa_block *b = &ps->block[i]; - printf("Writing block #%zu/%zu to file...\n", i, ps->block_nr); + printf("Writing block #%zu/%zu of %"PRIu64" bytes to file...\n", i, + ps->block_nr, b->size); if (fwrite(b->addr, b->size, 1, dmp_file) != 1) { - eprintf("Failed to write dump header\n"); + eprintf("Failed to write block\n"); fclose(dmp_file); return 1; } -- 2.34.1 From: Viktor Prutyanov <viktor@daynix.com> DMP supports 42 physical memory runs at most. So, merge adjacent physical memory ranges from QEMU ELF when possible to minimize total number of runs. Signed-off-by: Viktor Prutyanov <viktor@daynix.com> Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com> Message-id: 20230915170153.10959-4-viktor@daynix.com [PMM: fixed format string for printing size_t values] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- contrib/elf2dmp/main.c | 56 ++++++++++++++++++++++++++++++++++++------ 1 file changed, 48 insertions(+), 8 deletions(-) diff --git a/contrib/elf2dmp/main.c b/contrib/elf2dmp/main.c index XXXXXXX..XXXXXXX 100644 --- a/contrib/elf2dmp/main.c +++ b/contrib/elf2dmp/main.c @@ -XXX,XX +XXX,XX @@ #define PE_NAME "ntoskrnl.exe" #define INITIAL_MXCSR 0x1f80 +#define MAX_NUMBER_OF_RUNS 42 typedef struct idt_desc { uint16_t offset1; /* offset bits 0..15 */ @@ -XXX,XX +XXX,XX @@ static int fix_dtb(struct va_space *vs, QEMU_Elf *qe) return 1; } +static void try_merge_runs(struct pa_space *ps, + WinDumpPhyMemDesc64 *PhysicalMemoryBlock) +{ + unsigned int merge_cnt = 0, run_idx = 0; + + PhysicalMemoryBlock->NumberOfRuns = 0; + + for (size_t idx = 0; idx < ps->block_nr; idx++) { + struct pa_block *blk = ps->block + idx; + struct pa_block *next = blk + 1; + + PhysicalMemoryBlock->NumberOfPages += blk->size / ELF2DMP_PAGE_SIZE; + + if (idx + 1 != ps->block_nr && blk->paddr + blk->size == next->paddr) { + printf("Block #%zu 0x%"PRIx64"+:0x%"PRIx64" and %u previous will be" + " merged\n", idx, blk->paddr, blk->size, merge_cnt); + merge_cnt++; + } else { + struct pa_block *first_merged = blk - merge_cnt; + + printf("Block #%zu 0x%"PRIx64"+:0x%"PRIx64" and %u previous will be" + " merged to 0x%"PRIx64"+:0x%"PRIx64" (run #%u)\n", + idx, blk->paddr, blk->size, merge_cnt, first_merged->paddr, + blk->paddr + blk->size - first_merged->paddr, run_idx); + PhysicalMemoryBlock->Run[run_idx] = (WinDumpPhyMemRun64) { + .BasePage = first_merged->paddr / ELF2DMP_PAGE_SIZE, + .PageCount = (blk->paddr + blk->size - first_merged->paddr) / + ELF2DMP_PAGE_SIZE, + }; + PhysicalMemoryBlock->NumberOfRuns++; + run_idx++; + merge_cnt = 0; + } + } +} + static int fill_header(WinDumpHeader64 *hdr, struct pa_space *ps, struct va_space *vs, uint64_t KdDebuggerDataBlock, KDDEBUGGER_DATA64 *kdbg, uint64_t KdVersionBlock, int nr_cpus) @@ -XXX,XX +XXX,XX @@ static int fill_header(WinDumpHeader64 *hdr, struct pa_space *ps, KUSD_OFFSET_PRODUCT_TYPE); DBGKD_GET_VERSION64 kvb; WinDumpHeader64 h; - size_t i; QEMU_BUILD_BUG_ON(KUSD_OFFSET_SUITE_MASK >= ELF2DMP_PAGE_SIZE); QEMU_BUILD_BUG_ON(KUSD_OFFSET_PRODUCT_TYPE >= ELF2DMP_PAGE_SIZE); @@ -XXX,XX +XXX,XX @@ static int fill_header(WinDumpHeader64 *hdr, struct pa_space *ps, .RequiredDumpSpace = sizeof(h), }; - for (i = 0; i < ps->block_nr; i++) { - h.PhysicalMemoryBlock.NumberOfPages += - ps->block[i].size / ELF2DMP_PAGE_SIZE; - h.PhysicalMemoryBlock.Run[i] = (WinDumpPhyMemRun64) { - .BasePage = ps->block[i].paddr / ELF2DMP_PAGE_SIZE, - .PageCount = ps->block[i].size / ELF2DMP_PAGE_SIZE, - }; + if (h.PhysicalMemoryBlock.NumberOfRuns <= MAX_NUMBER_OF_RUNS) { + for (size_t idx = 0; idx < ps->block_nr; idx++) { + h.PhysicalMemoryBlock.NumberOfPages += + ps->block[idx].size / ELF2DMP_PAGE_SIZE; + h.PhysicalMemoryBlock.Run[idx] = (WinDumpPhyMemRun64) { + .BasePage = ps->block[idx].paddr / ELF2DMP_PAGE_SIZE, + .PageCount = ps->block[idx].size / ELF2DMP_PAGE_SIZE, + }; + } + } else { + try_merge_runs(ps, &h.PhysicalMemoryBlock); } h.RequiredDumpSpace += -- 2.34.1 From: Viktor Prutyanov <viktor@daynix.com> Glib's g_mapped_file_new maps file with PROT_READ|PROT_WRITE and MAP_PRIVATE. This leads to premature physical memory allocation of dump file size on Linux hosts and may fail. On Linux, mapping the file with MAP_NORESERVE limits the allocation by available memory. Signed-off-by: Viktor Prutyanov <viktor@daynix.com> Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com> Message-id: 20230915170153.10959-5-viktor@daynix.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- contrib/elf2dmp/qemu_elf.h | 2 ++ contrib/elf2dmp/qemu_elf.c | 68 +++++++++++++++++++++++++++++++------- 2 files changed, 58 insertions(+), 12 deletions(-) diff --git a/contrib/elf2dmp/qemu_elf.h b/contrib/elf2dmp/qemu_elf.h index XXXXXXX..XXXXXXX 100644 --- a/contrib/elf2dmp/qemu_elf.h +++ b/contrib/elf2dmp/qemu_elf.h @@ -XXX,XX +XXX,XX @@ typedef struct QEMUCPUState { int is_system(QEMUCPUState *s); typedef struct QEMU_Elf { +#ifndef CONFIG_LINUX GMappedFile *gmf; +#endif size_t size; void *map; QEMUCPUState **state; diff --git a/contrib/elf2dmp/qemu_elf.c b/contrib/elf2dmp/qemu_elf.c index XXXXXXX..XXXXXXX 100644 --- a/contrib/elf2dmp/qemu_elf.c +++ b/contrib/elf2dmp/qemu_elf.c @@ -XXX,XX +XXX,XX @@ static bool check_ehdr(QEMU_Elf *qe) return true; } -int QEMU_Elf_init(QEMU_Elf *qe, const char *filename) +static int QEMU_Elf_map(QEMU_Elf *qe, const char *filename) { +#ifdef CONFIG_LINUX + struct stat st; + int fd; + + printf("Using Linux mmap\n"); + + fd = open(filename, O_RDONLY, 0); + if (fd == -1) { + eprintf("Failed to open ELF dump file \'%s\'\n", filename); + return 1; + } + + if (fstat(fd, &st)) { + eprintf("Failed to get size of ELF dump file\n"); + close(fd); + return 1; + } + qe->size = st.st_size; + + qe->map = mmap(NULL, qe->size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_NORESERVE, fd, 0); + if (qe->map == MAP_FAILED) { + eprintf("Failed to map ELF file\n"); + close(fd); + return 1; + } + + close(fd); +#else GError *gerr = NULL; - int err = 0; + + printf("Using GLib mmap\n"); qe->gmf = g_mapped_file_new(filename, TRUE, &gerr); if (gerr) { @@ -XXX,XX +XXX,XX @@ int QEMU_Elf_init(QEMU_Elf *qe, const char *filename) qe->map = g_mapped_file_get_contents(qe->gmf); qe->size = g_mapped_file_get_length(qe->gmf); +#endif + + return 0; +} + +static void QEMU_Elf_unmap(QEMU_Elf *qe) +{ +#ifdef CONFIG_LINUX + munmap(qe->map, qe->size); +#else + g_mapped_file_unref(qe->gmf); +#endif +} + +int QEMU_Elf_init(QEMU_Elf *qe, const char *filename) +{ + if (QEMU_Elf_map(qe, filename)) { + return 1; + } if (!check_ehdr(qe)) { eprintf("Input file has the wrong format\n"); - err = 1; - goto out_unmap; + QEMU_Elf_unmap(qe); + return 1; } if (init_states(qe)) { eprintf("Failed to extract QEMU CPU states\n"); - err = 1; - goto out_unmap; + QEMU_Elf_unmap(qe); + return 1; } return 0; - -out_unmap: - g_mapped_file_unref(qe->gmf); - - return err; } void QEMU_Elf_exit(QEMU_Elf *qe) { exit_states(qe); - g_mapped_file_unref(qe->gmf); + QEMU_Elf_unmap(qe); } -- 2.34.1 From: Viktor Prutyanov <viktor@daynix.com> PDB for Windows 11 kernel has slightly different structure compared to previous versions. Since elf2dmp don't use the other fields, copy only 'segments' field from PDB_STREAM_INDEXES. Signed-off-by: Viktor Prutyanov <viktor@daynix.com> Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com> Message-id: 20230915170153.10959-6-viktor@daynix.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- contrib/elf2dmp/pdb.h | 2 +- contrib/elf2dmp/pdb.c | 15 ++++----------- 2 files changed, 5 insertions(+), 12 deletions(-) diff --git a/contrib/elf2dmp/pdb.h b/contrib/elf2dmp/pdb.h index XXXXXXX..XXXXXXX 100644 --- a/contrib/elf2dmp/pdb.h +++ b/contrib/elf2dmp/pdb.h @@ -XXX,XX +XXX,XX @@ struct pdb_reader { } ds; uint32_t file_used[1024]; PDB_SYMBOLS *symbols; - PDB_STREAM_INDEXES sidx; + uint16_t segments; uint8_t *modimage; char *segs; size_t segs_size; diff --git a/contrib/elf2dmp/pdb.c b/contrib/elf2dmp/pdb.c index XXXXXXX..XXXXXXX 100644 --- a/contrib/elf2dmp/pdb.c +++ b/contrib/elf2dmp/pdb.c @@ -XXX,XX +XXX,XX @@ static void *pdb_ds_read_file(struct pdb_reader* r, uint32_t file_number) static int pdb_init_segments(struct pdb_reader *r) { char *segs; - unsigned stream_idx = r->sidx.segments; + unsigned stream_idx = r->segments; segs = pdb_ds_read_file(r, stream_idx); if (!segs) { @@ -XXX,XX +XXX,XX @@ static int pdb_init_symbols(struct pdb_reader *r) { int err = 0; PDB_SYMBOLS *symbols; - PDB_STREAM_INDEXES *sidx = &r->sidx; - - memset(sidx, -1, sizeof(*sidx)); symbols = pdb_ds_read_file(r, 3); if (!symbols) { @@ -XXX,XX +XXX,XX @@ static int pdb_init_symbols(struct pdb_reader *r) r->symbols = symbols; - if (symbols->stream_index_size != sizeof(PDB_STREAM_INDEXES)) { - err = 1; - goto out_symbols; - } - - memcpy(sidx, (const char *)symbols + sizeof(PDB_SYMBOLS) + + r->segments = *(uint16_t *)((const char *)symbols + sizeof(PDB_SYMBOLS) + symbols->module_size + symbols->offset_size + symbols->hash_size + symbols->srcmodule_size + - symbols->pdbimport_size + symbols->unknown2_size, sizeof(*sidx)); + symbols->pdbimport_size + symbols->unknown2_size + + offsetof(PDB_STREAM_INDEXES, segments)); /* Read global symbol table */ r->modimage = pdb_ds_read_file(r, symbols->gsym_file); -- 2.34.1