The following changes since commit a97978bcc2d1f650c7d411428806e5b03082b8c7: Merge remote-tracking branch 'remotes/dg-gitlab/tags/ppc-for-6.1-20210603' into staging (2021-06-03 10:00:35 +0100) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20210603 for you to fetch changes up to 1c861885894d840235954060050d240259f5340b: tests/unit/test-vmstate: Assert that dup() and mkstemp() succeed (2021-06-03 16:43:27 +0100) ---------------------------------------------------------------- target-arm queue: * Some not-yet-enabled preliminaries for M-profile MVE support * Consistently use "Cortex-Axx", not "Cortex Axx" in docs, comments * docs: Fix installation of man pages with Sphinx 4.x * Mark LDS{MIN,MAX} as signed operations * Fix missing syndrome value for DAIF and PAC check exceptions * Implement BFloat16 extensions * Refactoring of hvf accelerator code in preparation for aarch64 support * Fix some coverity nits in test code ---------------------------------------------------------------- Alexander Graf (12): hvf: Move assert_hvf_ok() into common directory hvf: Move vcpu thread functions into common directory hvf: Move cpu functions into common directory hvf: Move hvf internal definitions into common header hvf: Make hvf_set_phys_mem() static hvf: Remove use of hv_uvaddr_t and hv_gpaddr_t hvf: Split out common code on vcpu init and destroy hvf: Use cpu_synchronize_state() hvf: Make synchronize functions static hvf: Remove hvf-accel-ops.h hvf: Introduce hvf vcpu struct hvf: Simplify post reset/init/loadvm hooks Damien Goutte-Gattat (1): docs: Fix installation of man pages with Sphinx 4.x Jamie Iles (4): target/arm: fix missing exception class target/arm: fold do_raise_exception into raise_exception target/arm: use raise_exception_ra for MTE check failure target/arm: use raise_exception_ra for stack limit exception Peter Maydell (15): target/arm: Add isar feature check functions for MVE target/arm: Update feature checks for insns which are "MVE or FP" target/arm: Move fpsp/fpdp isar check into callers of do_vfp_2op_sp/dp target/arm: Add MVE check to VMOV_reg_sp and VMOV_reg_dp target/arm: Fix return values in fp_sysreg_checks() target/arm: Implement M-profile VPR register target/arm: Make FPSCR.LTPSIZE writable for MVE target/arm: Allow board models to specify initial NS VTOR arm: Consistently use "Cortex-Axx", not "Cortex Axx" tests/qtest/bios-tables-test: Check for dup2() failure tests/qtest/e1000e-test: Check qemu_recv() succeeded tests/qtest/hd-geo-test: Fix checks on mkstemp() return value tests/qtest/pflash-cfi02-test: Avoid potential integer overflow tests/qtest/tpm-tests: Remove unnecessary NULL checks tests/unit/test-vmstate: Assert that dup() and mkstemp() succeed Richard Henderson (13): target/arm: Mark LDS{MIN,MAX} as signed operations target/arm: Add isar_feature_{aa32, aa64, aa64_sve}_bf16 target/arm: Unify unallocated path in disas_fp_1src target/arm: Implement scalar float32 to bfloat16 conversion target/arm: Implement vector float32 to bfloat16 conversion softfpu: Add float_round_to_odd_inf target/arm: Implement bfloat16 dot product (vector) target/arm: Implement bfloat16 dot product (indexed) target/arm: Implement bfloat16 matrix multiply accumulate target/arm: Implement bfloat widening fma (vector) target/arm: Implement bfloat widening fma (indexed) linux-user/aarch64: Enable hwcap bits for bfloat16 target/arm: Enable BFloat16 extensions docs/conf.py | 1 + docs/system/arm/aspeed.rst | 4 +- docs/system/arm/nuvoton.rst | 6 +- docs/system/arm/sabrelite.rst | 2 +- include/fpu/softfloat-types.h | 4 +- include/hw/arm/allwinner-h3.h | 2 +- include/hw/arm/armv7m.h | 2 + include/hw/core/cpu.h | 3 +- include/sysemu/hvf_int.h | 58 +++++ target/arm/cpu.h | 48 +++- target/arm/helper-sve.h | 4 + target/arm/helper.h | 15 ++ target/i386/hvf/hvf-accel-ops.h | 23 -- target/i386/hvf/hvf-i386.h | 33 +-- target/i386/hvf/vmx.h | 24 +- target/i386/hvf/x86hvf.h | 2 - target/arm/neon-dp.decode | 1 + target/arm/neon-shared.decode | 11 + target/arm/sve.decode | 19 +- target/arm/vfp.decode | 2 + accel/hvf/hvf-accel-ops.c | 471 ++++++++++++++++++++++++++++++++++++++++ accel/hvf/hvf-all.c | 47 ++++ hw/arm/armv7m.c | 7 + hw/arm/aspeed.c | 6 +- hw/arm/mcimx6ul-evk.c | 2 +- hw/arm/mcimx7d-sabre.c | 2 +- hw/arm/npcm7xx_boards.c | 4 +- hw/arm/sabrelite.c | 2 +- hw/misc/npcm7xx_clk.c | 2 +- linux-user/elfload.c | 2 + target/arm/cpu.c | 13 ++ target/arm/cpu64.c | 3 + target/arm/cpu_tcg.c | 1 + target/arm/m_helper.c | 5 +- target/arm/machine.c | 20 ++ target/arm/mte_helper.c | 12 +- target/arm/op_helper.c | 32 ++- target/arm/sve_helper.c | 2 + target/arm/translate-a64.c | 155 +++++++++++-- target/arm/translate-neon.c | 91 ++++++++ target/arm/translate-sve.c | 112 ++++++++++ target/arm/translate-vfp.c | 164 ++++++++++---- target/arm/vec_helper.c | 140 +++++++++++- target/arm/vfp_helper.c | 21 +- target/i386/hvf/hvf-accel-ops.c | 146 ------------- target/i386/hvf/hvf.c | 464 +++++---------------------------------- target/i386/hvf/x86.c | 28 +-- target/i386/hvf/x86_descr.c | 26 +-- target/i386/hvf/x86_emu.c | 62 +++--- target/i386/hvf/x86_mmu.c | 4 +- target/i386/hvf/x86_task.c | 12 +- target/i386/hvf/x86hvf.c | 222 +++++++++---------- tests/qtest/bios-tables-test.c | 8 +- tests/qtest/e1000e-test.c | 3 +- tests/qtest/hd-geo-test.c | 4 +- tests/qtest/pflash-cfi02-test.c | 2 +- tests/qtest/tpm-tests.c | 12 +- tests/unit/test-vmstate.c | 5 +- fpu/softfloat-parts.c.inc | 6 +- MAINTAINERS | 8 + accel/hvf/meson.build | 7 + accel/meson.build | 1 + target/i386/hvf/meson.build | 1 - 63 files changed, 1666 insertions(+), 935 deletions(-) create mode 100644 include/sysemu/hvf_int.h delete mode 100644 target/i386/hvf/hvf-accel-ops.h create mode 100644 accel/hvf/hvf-accel-ops.c create mode 100644 accel/hvf/hvf-all.c delete mode 100644 target/i386/hvf/hvf-accel-ops.c create mode 100644 accel/hvf/meson.build Add the isar feature check functions we will need for v8.1M MVE: * a check for MVE present: this corresponds to the pseudocode's CheckDecodeFaults(ExtType_Mve) * a check for the optional floating-point part of MVE: this corresponds to CheckDecodeFaults(ExtType_MveFp) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210520152840.24453-2-peter.maydell@linaro.org --- target/arm/cpu.h | 22 ++++++++++++++++++++++ 1 file changed, 22 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 @@ static inline bool isar_feature_aa32_fp16_arith(const ARMISARegisters *id) } } +static inline bool isar_feature_aa32_mve(const ARMISARegisters *id) +{ + /* + * Return true if MVE is supported (either integer or floating point). + * We must check for M-profile as the MVFR1 field means something + * else for A-profile. + */ + return isar_feature_aa32_mprofile(id) && + FIELD_EX32(id->mvfr1, MVFR1, MVE) > 0; +} + +static inline bool isar_feature_aa32_mve_fp(const ARMISARegisters *id) +{ + /* + * Return true if MVE is supported (either integer or floating point). + * We must check for M-profile as the MVFR1 field means something + * else for A-profile. + */ + return isar_feature_aa32_mprofile(id) && + FIELD_EX32(id->mvfr1, MVFR1, MVE) >= 2; +} + static inline bool isar_feature_aa32_vfp_simd(const ARMISARegisters *id) { /* -- 2.20.1 Some v8M instructions are present if either the floating point extension or MVE is implemented. Update our implementation of them to check for MVE as well as for FP. This is all the insns which use CheckDecodeFaults(ExtType_MveOrFp) or CheckDecodeFaults(ExtType_MveOrDpFp) in their pseudocode, which are essentially the loads and stores, moves and sysreg accesses, except for VMOV_reg_sp and VMOV_reg_dp, which we handle in subsequent patches because they need a refactor to provide a place to put the new MVE check. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210520152840.24453-3-peter.maydell@linaro.org --- target/arm/translate-vfp.c | 48 +++++++++++++++++++++++--------------- 1 file changed, 29 insertions(+), 19 deletions(-) diff --git a/target/arm/translate-vfp.c b/target/arm/translate-vfp.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-vfp.c +++ b/target/arm/translate-vfp.c @@ -XXX,XX +XXX,XX @@ static bool trans_VMOV_to_gp(DisasContext *s, arg_VMOV_to_gp *a) /* VMOV scalar to general purpose register */ TCGv_i32 tmp; - /* SIZE == MO_32 is a VFP instruction; otherwise NEON. */ - if (a->size == MO_32 - ? !dc_isar_feature(aa32_fpsp_v2, s) - : !arm_dc_feature(s, ARM_FEATURE_NEON)) { - return false; + /* + * SIZE == MO_32 is a VFP instruction; otherwise NEON. MVE has + * all sizes, whether the CPU has fp or not. + */ + if (!dc_isar_feature(aa32_mve, s)) { + if (a->size == MO_32 + ? !dc_isar_feature(aa32_fpsp_v2, s) + : !arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } } /* UNDEF accesses to D16-D31 if they don't exist */ @@ -XXX,XX +XXX,XX @@ static bool trans_VMOV_from_gp(DisasContext *s, arg_VMOV_from_gp *a) /* VMOV general purpose register to scalar */ TCGv_i32 tmp; - /* SIZE == MO_32 is a VFP instruction; otherwise NEON. */ - if (a->size == MO_32 - ? !dc_isar_feature(aa32_fpsp_v2, s) - : !arm_dc_feature(s, ARM_FEATURE_NEON)) { - return false; + /* + * SIZE == MO_32 is a VFP instruction; otherwise NEON. MVE has + * all sizes, whether the CPU has fp or not. + */ + if (!dc_isar_feature(aa32_mve, s)) { + if (a->size == MO_32 + ? !dc_isar_feature(aa32_fpsp_v2, s) + : !arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } } /* UNDEF accesses to D16-D31 if they don't exist */ @@ -XXX,XX +XXX,XX @@ typedef enum FPSysRegCheckResult { static FPSysRegCheckResult fp_sysreg_checks(DisasContext *s, int regno) { - if (!dc_isar_feature(aa32_fpsp_v2, s)) { + if (!dc_isar_feature(aa32_fpsp_v2, s) && !dc_isar_feature(aa32_mve, s)) { return FPSysRegCheckFailed; } @@ -XXX,XX +XXX,XX @@ static bool trans_VMOV_single(DisasContext *s, arg_VMOV_single *a) { TCGv_i32 tmp; - if (!dc_isar_feature(aa32_fpsp_v2, s)) { + if (!dc_isar_feature(aa32_fpsp_v2, s) && !dc_isar_feature(aa32_mve, s)) { return false; } @@ -XXX,XX +XXX,XX @@ static bool trans_VMOV_64_sp(DisasContext *s, arg_VMOV_64_sp *a) { TCGv_i32 tmp; - if (!dc_isar_feature(aa32_fpsp_v2, s)) { + if (!dc_isar_feature(aa32_fpsp_v2, s) && !dc_isar_feature(aa32_mve, s)) { return false; } @@ -XXX,XX +XXX,XX @@ static bool trans_VMOV_64_dp(DisasContext *s, arg_VMOV_64_dp *a) * floating point register. Note that this does not require support * for double precision arithmetic. */ - if (!dc_isar_feature(aa32_fpsp_v2, s)) { + if (!dc_isar_feature(aa32_fpsp_v2, s) && !dc_isar_feature(aa32_mve, s)) { return false; } @@ -XXX,XX +XXX,XX @@ static bool trans_VLDR_VSTR_hp(DisasContext *s, arg_VLDR_VSTR_sp *a) uint32_t offset; TCGv_i32 addr, tmp; - if (!dc_isar_feature(aa32_fp16_arith, s)) { + if (!dc_isar_feature(aa32_fpsp_v2, s) && !dc_isar_feature(aa32_mve, s)) { return false; } @@ -XXX,XX +XXX,XX @@ static bool trans_VLDR_VSTR_sp(DisasContext *s, arg_VLDR_VSTR_sp *a) uint32_t offset; TCGv_i32 addr, tmp; - if (!dc_isar_feature(aa32_fpsp_v2, s)) { + if (!dc_isar_feature(aa32_fpsp_v2, s) && !dc_isar_feature(aa32_mve, s)) { return false; } @@ -XXX,XX +XXX,XX @@ static bool trans_VLDR_VSTR_dp(DisasContext *s, arg_VLDR_VSTR_dp *a) TCGv_i64 tmp; /* Note that this does not require support for double arithmetic. */ - if (!dc_isar_feature(aa32_fpsp_v2, s)) { + if (!dc_isar_feature(aa32_fpsp_v2, s) && !dc_isar_feature(aa32_mve, s)) { return false; } @@ -XXX,XX +XXX,XX @@ static bool trans_VLDM_VSTM_sp(DisasContext *s, arg_VLDM_VSTM_sp *a) TCGv_i32 addr, tmp; int i, n; - if (!dc_isar_feature(aa32_fpsp_v2, s)) { + if (!dc_isar_feature(aa32_fpsp_v2, s) && !dc_isar_feature(aa32_mve, s)) { return false; } @@ -XXX,XX +XXX,XX @@ static bool trans_VLDM_VSTM_dp(DisasContext *s, arg_VLDM_VSTM_dp *a) int i, n; /* Note that this does not require support for double arithmetic. */ - if (!dc_isar_feature(aa32_fpsp_v2, s)) { + if (!dc_isar_feature(aa32_fpsp_v2, s) && !dc_isar_feature(aa32_mve, s)) { return false; } -- 2.20.1 The do_vfp_2op_sp() and do_vfp_2op_dp() functions currently check whether floating point is supported via the aa32_fpdp_v2 and aa32_fpsp_v2 isar checks. For v8.1M MVE support, the VMOV_reg trans functions (but not any of the others) need to update this to also allow the insn if MVE is implemented. Move the check out of the do_ function and into its callsites (which are all implemented via the DO_VFP_2OP macro), so we have a place to change the check for the VMOV insns. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210520152840.24453-4-peter.maydell@linaro.org --- target/arm/translate-vfp.c | 37 +++++++++++++++++++------------------ 1 file changed, 19 insertions(+), 18 deletions(-) diff --git a/target/arm/translate-vfp.c b/target/arm/translate-vfp.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-vfp.c +++ b/target/arm/translate-vfp.c @@ -XXX,XX +XXX,XX @@ static bool do_vfp_2op_sp(DisasContext *s, VFPGen2OpSPFn *fn, int vd, int vm) int veclen = s->vec_len; TCGv_i32 f0, fd; - if (!dc_isar_feature(aa32_fpsp_v2, s)) { - return false; - } + /* Note that the caller must check the aa32_fpsp_v2 feature. */ if (!dc_isar_feature(aa32_fpshvec, s) && (veclen != 0 || s->vec_stride != 0)) { @@ -XXX,XX +XXX,XX @@ static bool do_vfp_2op_hp(DisasContext *s, VFPGen2OpSPFn *fn, int vd, int vm) */ TCGv_i32 f0; + /* Note that the caller must check the aa32_fp16_arith feature */ + if (!dc_isar_feature(aa32_fp16_arith, s)) { return false; } @@ -XXX,XX +XXX,XX @@ static bool do_vfp_2op_dp(DisasContext *s, VFPGen2OpDPFn *fn, int vd, int vm) int veclen = s->vec_len; TCGv_i64 f0, fd; - if (!dc_isar_feature(aa32_fpdp_v2, s)) { - return false; - } + /* Note that the caller must check the aa32_fpdp_v2 feature. */ /* UNDEF accesses to D16-D31 if they don't exist */ if (!dc_isar_feature(aa32_simd_r32, s) && ((vd | vm) & 0x10)) { @@ -XXX,XX +XXX,XX @@ static bool trans_VMOV_imm_dp(DisasContext *s, arg_VMOV_imm_dp *a) return true; } -#define DO_VFP_2OP(INSN, PREC, FN) \ +#define DO_VFP_2OP(INSN, PREC, FN, CHECK) \ static bool trans_##INSN##_##PREC(DisasContext *s, \ arg_##INSN##_##PREC *a) \ { \ + if (!dc_isar_feature(CHECK, s)) { \ + return false; \ + } \ return do_vfp_2op_##PREC(s, FN, a->vd, a->vm); \ } -DO_VFP_2OP(VMOV_reg, sp, tcg_gen_mov_i32) -DO_VFP_2OP(VMOV_reg, dp, tcg_gen_mov_i64) +DO_VFP_2OP(VMOV_reg, sp, tcg_gen_mov_i32, aa32_fpsp_v2) +DO_VFP_2OP(VMOV_reg, dp, tcg_gen_mov_i64, aa32_fpdp_v2) -DO_VFP_2OP(VABS, hp, gen_helper_vfp_absh) -DO_VFP_2OP(VABS, sp, gen_helper_vfp_abss) -DO_VFP_2OP(VABS, dp, gen_helper_vfp_absd) +DO_VFP_2OP(VABS, hp, gen_helper_vfp_absh, aa32_fp16_arith) +DO_VFP_2OP(VABS, sp, gen_helper_vfp_abss, aa32_fpsp_v2) +DO_VFP_2OP(VABS, dp, gen_helper_vfp_absd, aa32_fpdp_v2) -DO_VFP_2OP(VNEG, hp, gen_helper_vfp_negh) -DO_VFP_2OP(VNEG, sp, gen_helper_vfp_negs) -DO_VFP_2OP(VNEG, dp, gen_helper_vfp_negd) +DO_VFP_2OP(VNEG, hp, gen_helper_vfp_negh, aa32_fp16_arith) +DO_VFP_2OP(VNEG, sp, gen_helper_vfp_negs, aa32_fpsp_v2) +DO_VFP_2OP(VNEG, dp, gen_helper_vfp_negd, aa32_fpdp_v2) static void gen_VSQRT_hp(TCGv_i32 vd, TCGv_i32 vm) { @@ -XXX,XX +XXX,XX @@ static void gen_VSQRT_dp(TCGv_i64 vd, TCGv_i64 vm) gen_helper_vfp_sqrtd(vd, vm, cpu_env); } -DO_VFP_2OP(VSQRT, hp, gen_VSQRT_hp) -DO_VFP_2OP(VSQRT, sp, gen_VSQRT_sp) -DO_VFP_2OP(VSQRT, dp, gen_VSQRT_dp) +DO_VFP_2OP(VSQRT, hp, gen_VSQRT_hp, aa32_fp16_arith) +DO_VFP_2OP(VSQRT, sp, gen_VSQRT_sp, aa32_fpsp_v2) +DO_VFP_2OP(VSQRT, dp, gen_VSQRT_dp, aa32_fpdp_v2) static bool trans_VCMP_hp(DisasContext *s, arg_VCMP_sp *a) { -- 2.20.1 Split out the handling of VMOV_reg_sp and VMOV_reg_dp so that we can permit the insns if either FP or MVE are present. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210520152840.24453-5-peter.maydell@linaro.org --- target/arm/translate-vfp.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/target/arm/translate-vfp.c b/target/arm/translate-vfp.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-vfp.c +++ b/target/arm/translate-vfp.c @@ -XXX,XX +XXX,XX @@ static bool trans_VMOV_imm_dp(DisasContext *s, arg_VMOV_imm_dp *a) return do_vfp_2op_##PREC(s, FN, a->vd, a->vm); \ } -DO_VFP_2OP(VMOV_reg, sp, tcg_gen_mov_i32, aa32_fpsp_v2) -DO_VFP_2OP(VMOV_reg, dp, tcg_gen_mov_i64, aa32_fpdp_v2) +#define DO_VFP_VMOV(INSN, PREC, FN) \ + static bool trans_##INSN##_##PREC(DisasContext *s, \ + arg_##INSN##_##PREC *a) \ + { \ + if (!dc_isar_feature(aa32_fp##PREC##_v2, s) && \ + !dc_isar_feature(aa32_mve, s)) { \ + return false; \ + } \ + return do_vfp_2op_##PREC(s, FN, a->vd, a->vm); \ + } + +DO_VFP_VMOV(VMOV_reg, sp, tcg_gen_mov_i32) +DO_VFP_VMOV(VMOV_reg, dp, tcg_gen_mov_i64) DO_VFP_2OP(VABS, hp, gen_helper_vfp_absh, aa32_fp16_arith) DO_VFP_2OP(VABS, sp, gen_helper_vfp_abss, aa32_fpsp_v2) -- 2.20.1 The fp_sysreg_checks() function is supposed to be returning an FPSysRegCheckResult, which is an enum with three possible values. However, three places in the function "return false" (a hangover from a previous iteration of the design where the function just returned a bool). Make these return FPSysRegCheckFailed instead (for no functional change, since both false and FPSysRegCheckFailed are zero). Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210520152840.24453-6-peter.maydell@linaro.org --- target/arm/translate-vfp.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/target/arm/translate-vfp.c b/target/arm/translate-vfp.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-vfp.c +++ b/target/arm/translate-vfp.c @@ -XXX,XX +XXX,XX @@ static FPSysRegCheckResult fp_sysreg_checks(DisasContext *s, int regno) break; case ARM_VFP_FPSCR_NZCVQC: if (!arm_dc_feature(s, ARM_FEATURE_V8_1M)) { - return false; + return FPSysRegCheckFailed; } break; case ARM_VFP_FPCXT_S: case ARM_VFP_FPCXT_NS: if (!arm_dc_feature(s, ARM_FEATURE_V8_1M)) { - return false; + return FPSysRegCheckFailed; } if (!s->v8m_secure) { - return false; + return FPSysRegCheckFailed; } break; default: -- 2.20.1 If MVE is implemented for an M-profile CPU then it has a VPR register, which tracks predication information. Implement the read and write handling of this register, and the migration of its state. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210520152840.24453-7-peter.maydell@linaro.org --- target/arm/cpu.h | 6 ++++++ target/arm/machine.c | 19 +++++++++++++++++++ target/arm/translate-vfp.c | 38 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 63 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 @@ typedef struct CPUARMState { uint32_t cpacr[M_REG_NUM_BANKS]; uint32_t nsacr; int ltpsize; + uint32_t vpr; } v7m; /* Information associated with an exception about to be taken: @@ -XXX,XX +XXX,XX @@ FIELD(V7M_FPCCR, ASPEN, 31, 1) R_V7M_FPCCR_UFRDY_MASK | \ R_V7M_FPCCR_ASPEN_MASK) +/* v7M VPR bits */ +FIELD(V7M_VPR, P0, 0, 16) +FIELD(V7M_VPR, MASK01, 16, 4) +FIELD(V7M_VPR, MASK23, 20, 4) + /* * System register ID fields. */ diff --git a/target/arm/machine.c b/target/arm/machine.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/machine.c +++ b/target/arm/machine.c @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_m_fp = { } }; +static bool mve_needed(void *opaque) +{ + ARMCPU *cpu = opaque; + + return cpu_isar_feature(aa32_mve, cpu); +} + +static const VMStateDescription vmstate_m_mve = { + .name = "cpu/m/mve", + .version_id = 1, + .minimum_version_id = 1, + .needed = mve_needed, + .fields = (VMStateField[]) { + VMSTATE_UINT32(env.v7m.vpr, ARMCPU), + VMSTATE_END_OF_LIST() + }, +}; + static const VMStateDescription vmstate_m = { .name = "cpu/m", .version_id = 4, @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_m = { &vmstate_m_other_sp, &vmstate_m_v8m, &vmstate_m_fp, + &vmstate_m_mve, NULL } }; diff --git a/target/arm/translate-vfp.c b/target/arm/translate-vfp.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-vfp.c +++ b/target/arm/translate-vfp.c @@ -XXX,XX +XXX,XX @@ static FPSysRegCheckResult fp_sysreg_checks(DisasContext *s, int regno) return FPSysRegCheckFailed; } break; + case ARM_VFP_VPR: + case ARM_VFP_P0: + if (!dc_isar_feature(aa32_mve, s)) { + return FPSysRegCheckFailed; + } + break; default: return FPSysRegCheckFailed; } @@ -XXX,XX +XXX,XX @@ static bool gen_M_fp_sysreg_write(DisasContext *s, int regno, tcg_temp_free_i32(sfpa); break; } + case ARM_VFP_VPR: + /* Behaves as NOP if not privileged */ + if (IS_USER(s)) { + break; + } + tmp = loadfn(s, opaque); + store_cpu_field(tmp, v7m.vpr); + break; + case ARM_VFP_P0: + { + TCGv_i32 vpr; + tmp = loadfn(s, opaque); + vpr = load_cpu_field(v7m.vpr); + tcg_gen_deposit_i32(vpr, vpr, tmp, + R_V7M_VPR_P0_SHIFT, R_V7M_VPR_P0_LENGTH); + store_cpu_field(vpr, v7m.vpr); + tcg_temp_free_i32(tmp); + break; + } default: g_assert_not_reached(); } @@ -XXX,XX +XXX,XX @@ static bool gen_M_fp_sysreg_read(DisasContext *s, int regno, tcg_temp_free_i32(fpscr); break; } + case ARM_VFP_VPR: + /* Behaves as NOP if not privileged */ + if (IS_USER(s)) { + break; + } + tmp = load_cpu_field(v7m.vpr); + storefn(s, opaque, tmp); + break; + case ARM_VFP_P0: + tmp = load_cpu_field(v7m.vpr); + tcg_gen_extract_i32(tmp, tmp, R_V7M_VPR_P0_SHIFT, R_V7M_VPR_P0_LENGTH); + storefn(s, opaque, tmp); + break; default: g_assert_not_reached(); } -- 2.20.1 The M-profile FPSCR has an LTPSIZE field, but if MVE is not implemented it is read-only and always reads as 4; this is how QEMU currently handles it. Make the field writable when MVE is implemented. We can safely add the field to the MVE migration struct because currently no CPUs enable MVE and so the migration struct is never used. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210520152840.24453-8-peter.maydell@linaro.org --- target/arm/cpu.h | 3 ++- target/arm/machine.c | 1 + target/arm/vfp_helper.c | 9 ++++++--- 3 files changed, 9 insertions(+), 4 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 @@ typedef struct CPUARMState { uint32_t fpdscr[M_REG_NUM_BANKS]; uint32_t cpacr[M_REG_NUM_BANKS]; uint32_t nsacr; - int ltpsize; + uint32_t ltpsize; uint32_t vpr; } v7m; @@ -XXX,XX +XXX,XX @@ void vfp_set_fpscr(CPUARMState *env, uint32_t val); #define FPCR_LTPSIZE_SHIFT 16 /* LTPSIZE, M-profile only */ #define FPCR_LTPSIZE_MASK (7 << FPCR_LTPSIZE_SHIFT) +#define FPCR_LTPSIZE_LENGTH 3 #define FPCR_NZCV_MASK (FPCR_N | FPCR_Z | FPCR_C | FPCR_V) #define FPCR_NZCVQC_MASK (FPCR_NZCV_MASK | FPCR_QC) diff --git a/target/arm/machine.c b/target/arm/machine.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/machine.c +++ b/target/arm/machine.c @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_m_mve = { .needed = mve_needed, .fields = (VMStateField[]) { VMSTATE_UINT32(env.v7m.vpr, ARMCPU), + VMSTATE_UINT32(env.v7m.ltpsize, ARMCPU), VMSTATE_END_OF_LIST() }, }; diff --git a/target/arm/vfp_helper.c b/target/arm/vfp_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vfp_helper.c +++ b/target/arm/vfp_helper.c @@ -XXX,XX +XXX,XX @@ uint32_t vfp_get_fpscr(CPUARMState *env) void HELPER(vfp_set_fpscr)(CPUARMState *env, uint32_t val) { + ARMCPU *cpu = env_archcpu(env); + /* When ARMv8.2-FP16 is not supported, FZ16 is RES0. */ - if (!cpu_isar_feature(any_fp16, env_archcpu(env))) { + if (!cpu_isar_feature(any_fp16, cpu)) { val &= ~FPCR_FZ16; } @@ -XXX,XX +XXX,XX @@ void HELPER(vfp_set_fpscr)(CPUARMState *env, uint32_t val) * because in v7A no-short-vector-support cores still had to * allow Stride/Len to be written with the only effect that * some insns are required to UNDEF if the guest sets them. - * - * TODO: if M-profile MVE implemented, set LTPSIZE. */ env->vfp.vec_len = extract32(val, 16, 3); env->vfp.vec_stride = extract32(val, 20, 2); + } else if (cpu_isar_feature(aa32_mve, cpu)) { + env->v7m.ltpsize = extract32(val, FPCR_LTPSIZE_SHIFT, + FPCR_LTPSIZE_LENGTH); } if (arm_feature(env, ARM_FEATURE_NEON)) { -- 2.20.1 Currently we allow board models to specify the initial value of the Secure VTOR register, using an init-svtor property on the TYPE_ARMV7M object which is plumbed through to the CPU. Allow board models to also specify the initial value of the Non-secure VTOR via a similar init-nsvtor property. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210520152840.24453-10-peter.maydell@linaro.org --- include/hw/arm/armv7m.h | 2 ++ target/arm/cpu.h | 2 ++ hw/arm/armv7m.c | 7 +++++++ target/arm/cpu.c | 10 ++++++++++ 4 files changed, 21 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) * devices will be automatically layered on top of this view.) * + Property "idau": IDAU interface (forwarded to CPU object) * + Property "init-svtor": secure VTOR reset value (forwarded to CPU object) + * + Property "init-nsvtor": non-secure VTOR reset value (forwarded to CPU object) * + Property "vfp": enable VFP (forwarded to CPU object) * + Property "dsp": enable DSP (forwarded to CPU object) * + Property "enable-bitband": expose bitbanded IO @@ -XXX,XX +XXX,XX @@ struct ARMv7MState { MemoryRegion *board_memory; Object *idau; uint32_t init_svtor; + uint32_t init_nsvtor; bool enable_bitband; bool start_powered_off; bool vfp; 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 ARMCPU { /* For v8M, initial value of the Secure VTOR */ uint32_t init_svtor; + /* For v8M, initial value of the Non-secure VTOR */ + uint32_t init_nsvtor; /* [QEMU_]KVM_ARM_TARGET_* constant for this CPU, or * QEMU_KVM_ARM_TARGET_NONE if the kernel doesn't support this CPU type. 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) return; } } + if (object_property_find(OBJECT(s->cpu), "init-nsvtor")) { + if (!object_property_set_uint(OBJECT(s->cpu), "init-nsvtor", + s->init_nsvtor, errp)) { + return; + } + } if (object_property_find(OBJECT(s->cpu), "start-powered-off")) { if (!object_property_set_bool(OBJECT(s->cpu), "start-powered-off", s->start_powered_off, errp)) { @@ -XXX,XX +XXX,XX @@ static Property armv7m_properties[] = { MemoryRegion *), DEFINE_PROP_LINK("idau", ARMv7MState, idau, TYPE_IDAU_INTERFACE, Object *), DEFINE_PROP_UINT32("init-svtor", ARMv7MState, init_svtor, 0), + DEFINE_PROP_UINT32("init-nsvtor", ARMv7MState, init_nsvtor, 0), DEFINE_PROP_BOOL("enable-bitband", ARMv7MState, enable_bitband, false), DEFINE_PROP_BOOL("start-powered-off", ARMv7MState, start_powered_off, false), 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_reset(DeviceState *dev) env->regs[14] = 0xffffffff; env->v7m.vecbase[M_REG_S] = cpu->init_svtor & 0xffffff80; + env->v7m.vecbase[M_REG_NS] = cpu->init_nsvtor & 0xffffff80; /* Load the initial SP and PC from offset 0 and 4 in the vector table */ vecbase = env->v7m.vecbase[env->v7m.secure]; @@ -XXX,XX +XXX,XX @@ void arm_cpu_post_init(Object *obj) &cpu->init_svtor, OBJ_PROP_FLAG_READWRITE); } + if (arm_feature(&cpu->env, ARM_FEATURE_M)) { + /* + * Initial value of the NS VTOR (for cores without the Security + * extension, this is the only VTOR) + */ + object_property_add_uint32_ptr(obj, "init-nsvtor", + &cpu->init_nsvtor, + OBJ_PROP_FLAG_READWRITE); + } qdev_property_add_static(DEVICE(obj), &arm_cpu_cfgend_property); -- 2.20.1 The official punctuation for Arm CPU names uses a hyphen, like "Cortex-A9". We mostly follow this, but in a few places usage without the hyphen has crept in. Fix those so we consistently use the same way of writing the CPU name. This commit was created with: git grep -z -l 'Cortex ' | xargs -0 sed -i 's/Cortex /Cortex-/' Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20210527095152.10968-1-peter.maydell@linaro.org --- docs/system/arm/aspeed.rst | 4 ++-- docs/system/arm/nuvoton.rst | 6 +++--- docs/system/arm/sabrelite.rst | 2 +- include/hw/arm/allwinner-h3.h | 2 +- hw/arm/aspeed.c | 6 +++--- hw/arm/mcimx6ul-evk.c | 2 +- hw/arm/mcimx7d-sabre.c | 2 +- hw/arm/npcm7xx_boards.c | 4 ++-- hw/arm/sabrelite.c | 2 +- hw/misc/npcm7xx_clk.c | 2 +- 10 files changed, 16 insertions(+), 16 deletions(-) diff --git a/docs/system/arm/aspeed.rst b/docs/system/arm/aspeed.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/aspeed.rst +++ b/docs/system/arm/aspeed.rst @@ -XXX,XX +XXX,XX @@ The QEMU Aspeed machines model BMCs of various OpenPOWER systems and Aspeed evaluation boards. They are based on different releases of the Aspeed SoC : the AST2400 integrating an ARM926EJ-S CPU (400MHz), the AST2500 with an ARM1176JZS CPU (800MHz) and more recently the AST2600 -with dual cores ARM Cortex A7 CPUs (1.2GHz). +with dual cores ARM Cortex-A7 CPUs (1.2GHz). The SoC comes with RAM, Gigabit ethernet, USB, SD/MMC, USB, SPI, I2C, etc. @@ -XXX,XX +XXX,XX @@ AST2500 SoC based machines : AST2600 SoC based machines : -- ``ast2600-evb`` Aspeed AST2600 Evaluation board (Cortex A7) +- ``ast2600-evb`` Aspeed AST2600 Evaluation board (Cortex-A7) - ``tacoma-bmc`` OpenPOWER Witherspoon POWER9 AST2600 BMC Supported devices diff --git a/docs/system/arm/nuvoton.rst b/docs/system/arm/nuvoton.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/nuvoton.rst +++ b/docs/system/arm/nuvoton.rst @@ -XXX,XX +XXX,XX @@ Nuvoton iBMC boards (``npcm750-evb``, ``quanta-gsj``) The `Nuvoton iBMC`_ chips (NPCM7xx) are a family of ARM-based SoCs that are designed to be used as Baseboard Management Controllers (BMCs) in various -servers. They all feature one or two ARM Cortex A9 CPU cores, as well as an +servers. They all feature one or two ARM Cortex-A9 CPU cores, as well as an assortment of peripherals targeted for either Enterprise or Data Center / Hyperscale applications. The former is a superset of the latter, so NPCM750 has all the peripherals of NPCM730 and more. .. _Nuvoton iBMC: https://www.nuvoton.com/products/cloud-computing/ibmc/ -The NPCM750 SoC has two Cortex A9 cores and is targeted for the Enterprise +The NPCM750 SoC has two Cortex-A9 cores and is targeted for the Enterprise segment. The following machines are based on this chip : - ``npcm750-evb`` Nuvoton NPCM750 Evaluation board -The NPCM730 SoC has two Cortex A9 cores and is targeted for Data Center and +The NPCM730 SoC has two Cortex-A9 cores and is targeted for Data Center and Hyperscale applications. The following machines are based on this chip : - ``quanta-gsj`` Quanta GSJ server BMC diff --git a/docs/system/arm/sabrelite.rst b/docs/system/arm/sabrelite.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/sabrelite.rst +++ b/docs/system/arm/sabrelite.rst @@ -XXX,XX +XXX,XX @@ Supported devices The SABRE Lite machine supports the following devices: - * Up to 4 Cortex A9 cores + * Up to 4 Cortex-A9 cores * Generic Interrupt Controller * 1 Clock Controller Module * 1 System Reset Controller diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ */ /* - * The Allwinner H3 is a System on Chip containing four ARM Cortex A7 + * The Allwinner H3 is a System on Chip containing four ARM Cortex-A7 * processor cores. Features and specifications include DDR2/DDR3 memory, * SD/MMC storage cards, 10/100/1000Mbit Ethernet, USB 2.0, HDMI and * various I/O modules. diff --git a/hw/arm/aspeed.c b/hw/arm/aspeed.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed.c +++ b/hw/arm/aspeed.c @@ -XXX,XX +XXX,XX @@ static void aspeed_machine_ast2600_evb_class_init(ObjectClass *oc, void *data) MachineClass *mc = MACHINE_CLASS(oc); AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); - mc->desc = "Aspeed AST2600 EVB (Cortex A7)"; + mc->desc = "Aspeed AST2600 EVB (Cortex-A7)"; amc->soc_name = "ast2600-a1"; amc->hw_strap1 = AST2600_EVB_HW_STRAP1; amc->hw_strap2 = AST2600_EVB_HW_STRAP2; @@ -XXX,XX +XXX,XX @@ static void aspeed_machine_tacoma_class_init(ObjectClass *oc, void *data) MachineClass *mc = MACHINE_CLASS(oc); AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); - mc->desc = "OpenPOWER Tacoma BMC (Cortex A7)"; + mc->desc = "OpenPOWER Tacoma BMC (Cortex-A7)"; amc->soc_name = "ast2600-a1"; amc->hw_strap1 = TACOMA_BMC_HW_STRAP1; amc->hw_strap2 = TACOMA_BMC_HW_STRAP2; @@ -XXX,XX +XXX,XX @@ static void aspeed_machine_rainier_class_init(ObjectClass *oc, void *data) MachineClass *mc = MACHINE_CLASS(oc); AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); - mc->desc = "IBM Rainier BMC (Cortex A7)"; + mc->desc = "IBM Rainier BMC (Cortex-A7)"; amc->soc_name = "ast2600-a1"; amc->hw_strap1 = RAINIER_BMC_HW_STRAP1; amc->hw_strap2 = RAINIER_BMC_HW_STRAP2; diff --git a/hw/arm/mcimx6ul-evk.c b/hw/arm/mcimx6ul-evk.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/mcimx6ul-evk.c +++ b/hw/arm/mcimx6ul-evk.c @@ -XXX,XX +XXX,XX @@ static void mcimx6ul_evk_init(MachineState *machine) static void mcimx6ul_evk_machine_init(MachineClass *mc) { - mc->desc = "Freescale i.MX6UL Evaluation Kit (Cortex A7)"; + mc->desc = "Freescale i.MX6UL Evaluation Kit (Cortex-A7)"; mc->init = mcimx6ul_evk_init; mc->max_cpus = FSL_IMX6UL_NUM_CPUS; mc->default_ram_id = "mcimx6ul-evk.ram"; diff --git a/hw/arm/mcimx7d-sabre.c b/hw/arm/mcimx7d-sabre.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/mcimx7d-sabre.c +++ b/hw/arm/mcimx7d-sabre.c @@ -XXX,XX +XXX,XX @@ static void mcimx7d_sabre_init(MachineState *machine) static void mcimx7d_sabre_machine_init(MachineClass *mc) { - mc->desc = "Freescale i.MX7 DUAL SABRE (Cortex A7)"; + mc->desc = "Freescale i.MX7 DUAL SABRE (Cortex-A7)"; mc->init = mcimx7d_sabre_init; mc->max_cpus = FSL_IMX7_NUM_CPUS; mc->default_ram_id = "mcimx7d-sabre.ram"; diff --git a/hw/arm/npcm7xx_boards.c b/hw/arm/npcm7xx_boards.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/npcm7xx_boards.c +++ b/hw/arm/npcm7xx_boards.c @@ -XXX,XX +XXX,XX @@ static void npcm750_evb_machine_class_init(ObjectClass *oc, void *data) npcm7xx_set_soc_type(nmc, TYPE_NPCM750); - mc->desc = "Nuvoton NPCM750 Evaluation Board (Cortex A9)"; + mc->desc = "Nuvoton NPCM750 Evaluation Board (Cortex-A9)"; mc->init = npcm750_evb_init; mc->default_ram_size = 512 * MiB; }; @@ -XXX,XX +XXX,XX @@ static void gsj_machine_class_init(ObjectClass *oc, void *data) npcm7xx_set_soc_type(nmc, TYPE_NPCM730); - mc->desc = "Quanta GSJ (Cortex A9)"; + mc->desc = "Quanta GSJ (Cortex-A9)"; mc->init = quanta_gsj_init; mc->default_ram_size = 512 * MiB; }; diff --git a/hw/arm/sabrelite.c b/hw/arm/sabrelite.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/sabrelite.c +++ b/hw/arm/sabrelite.c @@ -XXX,XX +XXX,XX @@ static void sabrelite_init(MachineState *machine) static void sabrelite_machine_init(MachineClass *mc) { - mc->desc = "Freescale i.MX6 Quad SABRE Lite Board (Cortex A9)"; + mc->desc = "Freescale i.MX6 Quad SABRE Lite Board (Cortex-A9)"; mc->init = sabrelite_init; mc->max_cpus = FSL_IMX6_NUM_CPUS; mc->ignore_memory_transaction_failures = true; diff --git a/hw/misc/npcm7xx_clk.c b/hw/misc/npcm7xx_clk.c index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/npcm7xx_clk.c +++ b/hw/misc/npcm7xx_clk.c @@ -XXX,XX +XXX,XX @@ #define NPCM7XX_CLOCK_REF_HZ (25000000) /* Register Field Definitions */ -#define NPCM7XX_CLK_WDRCR_CA9C BIT(0) /* Cortex A9 Cores */ +#define NPCM7XX_CLK_WDRCR_CA9C BIT(0) /* Cortex-A9 Cores */ #define PLLCON_LOKI BIT(31) #define PLLCON_LOKS BIT(30) -- 2.20.1 From: Damien Goutte-Gattat <dgouttegattat@incenp.org> The 4.x branch of Sphinx introduces a breaking change, as generated man pages are now written to subdirectories corresponding to the manual section they belong to. This results in `make install` erroring out when attempting to install the man pages, because they are not where it expects to find them. This patch restores the behavior of Sphinx 3.x regarding man pages. Resolves: https://gitlab.com/qemu-project/qemu/-/issues/256 Signed-off-by: Damien Goutte-Gattat <dgouttegattat@incenp.org> Message-id: 20210503161422.15028-1-dgouttegattat@incenp.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/conf.py | 1 + 1 file changed, 1 insertion(+) diff --git a/docs/conf.py b/docs/conf.py index XXXXXXX..XXXXXXX 100644 --- a/docs/conf.py +++ b/docs/conf.py @@ -XXX,XX +XXX,XX @@ ['Stefan Hajnoczi <stefanha@redhat.com>', 'Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>'], 1), ] +man_make_section_directory = False # -- Options for Texinfo output ------------------------------------------- -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> The operands to tcg_gen_atomic_fetch_s{min,max}_i64 must be signed, so that the inputs are properly extended. Zero extend the result afterward, as needed. Resolves: https://gitlab.com/qemu-project/qemu/-/issues/364 Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20210602020720.47679-1-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate-a64.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_ldst_atomic(DisasContext *s, uint32_t insn, int o3_opc = extract32(insn, 12, 4); bool r = extract32(insn, 22, 1); bool a = extract32(insn, 23, 1); - TCGv_i64 tcg_rs, clean_addr; + TCGv_i64 tcg_rs, tcg_rt, clean_addr; AtomicThreeOpFn *fn = NULL; + MemOp mop = s->be_data | size | MO_ALIGN; if (is_vector || !dc_isar_feature(aa64_atomics, s)) { unallocated_encoding(s); @@ -XXX,XX +XXX,XX @@ static void disas_ldst_atomic(DisasContext *s, uint32_t insn, break; case 004: /* LDSMAX */ fn = tcg_gen_atomic_fetch_smax_i64; + mop |= MO_SIGN; break; case 005: /* LDSMIN */ fn = tcg_gen_atomic_fetch_smin_i64; + mop |= MO_SIGN; break; case 006: /* LDUMAX */ fn = tcg_gen_atomic_fetch_umax_i64; @@ -XXX,XX +XXX,XX @@ static void disas_ldst_atomic(DisasContext *s, uint32_t insn, } tcg_rs = read_cpu_reg(s, rs, true); + tcg_rt = cpu_reg(s, rt); if (o3_opc == 1) { /* LDCLR */ tcg_gen_not_i64(tcg_rs, tcg_rs); @@ -XXX,XX +XXX,XX @@ static void disas_ldst_atomic(DisasContext *s, uint32_t insn, /* The tcg atomic primitives are all full barriers. Therefore we * can ignore the Acquire and Release bits of this instruction. */ - fn(cpu_reg(s, rt), clean_addr, tcg_rs, get_mem_index(s), - s->be_data | size | MO_ALIGN); + fn(tcg_rt, clean_addr, tcg_rs, get_mem_index(s), mop); + + if ((mop & MO_SIGN) && size != MO_64) { + tcg_gen_ext32u_i64(tcg_rt, tcg_rt); + } } /* -- 2.20.1 From: Jamie Iles <jamie@nuviainc.com> The DAIF and PAC checks used raise_exception_ra to raise an exception and unwind CPU state but raise_exception_ra is currently designed for handling data aborts as the syndrome is partially precomputed and encoded in the TB and then merged in merge_syn_data_abort when handling the data abort. Using raise_exception_ra for DAIF and PAC checks results in an empty syndrome being retrieved from data[2] in restore_state_to_opc and setting ESR to 0. This manifested as: kvm [571]: Unknown exception class: esr: 0x000000 – Unknown/Uncategorized when launching a KVM guest when the host qemu used a CPU supporting EL2+pointer authentication and enabling pointer authentication in the guest. Rework raise_exception_ra such that the state is restored before raising the exception so that the exception is not clobbered by restore_state_to_opc. Fixes: 0d43e1a2d29a ("target/arm: Add PAuth helpers") Cc: Richard Henderson <richard.henderson@linaro.org> Cc: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Jamie Iles <jamie@nuviainc.com> [PMM: added comment] Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/op_helper.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/target/arm/op_helper.c b/target/arm/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/op_helper.c +++ b/target/arm/op_helper.c @@ -XXX,XX +XXX,XX @@ void raise_exception(CPUARMState *env, uint32_t excp, void raise_exception_ra(CPUARMState *env, uint32_t excp, uint32_t syndrome, uint32_t target_el, uintptr_t ra) { - CPUState *cs = do_raise_exception(env, excp, syndrome, target_el); - cpu_loop_exit_restore(cs, ra); + CPUState *cs = env_cpu(env); + + /* + * restore_state_to_opc() will set env->exception.syndrome, so + * we must restore CPU state here before setting the syndrome + * the caller passed us, and cannot use cpu_loop_exit_restore(). + */ + cpu_restore_state(cs, ra, true); + raise_exception(env, excp, syndrome, target_el); } uint64_t HELPER(neon_tbl)(CPUARMState *env, uint32_t desc, -- 2.20.1 From: Jamie Iles <jamie@nuviainc.com> Now that there are no other users of do_raise_exception, fold it into raise_exception. Cc: Richard Henderson <richard.henderson@linaro.org> Cc: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Jamie Iles <jamie@nuviainc.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/op_helper.c | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) diff --git a/target/arm/op_helper.c b/target/arm/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/op_helper.c +++ b/target/arm/op_helper.c @@ -XXX,XX +XXX,XX @@ #define SIGNBIT (uint32_t)0x80000000 #define SIGNBIT64 ((uint64_t)1 << 63) -static CPUState *do_raise_exception(CPUARMState *env, uint32_t excp, - uint32_t syndrome, uint32_t target_el) +void raise_exception(CPUARMState *env, uint32_t excp, + uint32_t syndrome, uint32_t target_el) { CPUState *cs = env_cpu(env); @@ -XXX,XX +XXX,XX @@ static CPUState *do_raise_exception(CPUARMState *env, uint32_t excp, cs->exception_index = excp; env->exception.syndrome = syndrome; env->exception.target_el = target_el; - - return cs; -} - -void raise_exception(CPUARMState *env, uint32_t excp, - uint32_t syndrome, uint32_t target_el) -{ - CPUState *cs = do_raise_exception(env, excp, syndrome, target_el); cpu_loop_exit(cs); } -- 2.20.1 From: Jamie Iles <jamie@nuviainc.com> Now that raise_exception_ra restores the state before raising the exception we can use restore_exception_ra to perform the state restore + exception raising without clobbering the syndrome. Cc: Richard Henderson <richard.henderson@linaro.org> Cc: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Jamie Iles <jamie@nuviainc.com> [PMM: Keep the one line of the comment that is still relevant] Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/mte_helper.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) diff --git a/target/arm/mte_helper.c b/target/arm/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/mte_helper.c +++ b/target/arm/mte_helper.c @@ -XXX,XX +XXX,XX @@ static void mte_check_fail(CPUARMState *env, uint32_t desc, switch (tcf) { case 1: - /* - * Tag check fail causes a synchronous exception. - * - * In restore_state_to_opc, we set the exception syndrome - * for the load or store operation. Unwind first so we - * may overwrite that with the syndrome for the tag check. - */ - cpu_restore_state(env_cpu(env), ra, true); + /* Tag check fail causes a synchronous exception. */ env->exception.vaddress = dirty_ptr; is_write = FIELD_EX32(desc, MTEDESC, WRITE); syn = syn_data_abort_no_iss(arm_current_el(env) != 0, 0, 0, 0, 0, is_write, 0x11); - raise_exception(env, EXCP_DATA_ABORT, syn, exception_target_el(env)); + raise_exception_ra(env, EXCP_DATA_ABORT, syn, + exception_target_el(env), ra); /* noreturn, but fall through to the assert anyway */ case 0: -- 2.20.1 From: Jamie Iles <jamie@nuviainc.com> The sequence cpu_restore_state() + raise_exception() is equivalent to raise_exception_ra(), so use that instead. (In this case we never cared about the syndrome value, because M-profile doesn't use the syndrome; the old code was just written unnecessarily awkwardly.) Cc: Richard Henderson <richard.henderson@linaro.org> Cc: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Jamie Iles <jamie@nuviainc.com> [PMM: Retain edited version of comment; rewrite commit message] Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/m_helper.c | 5 +---- target/arm/op_helper.c | 9 +++------ 2 files changed, 4 insertions(+), 10 deletions(-) diff --git a/target/arm/m_helper.c b/target/arm/m_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/m_helper.c +++ b/target/arm/m_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(v7m_msr)(CPUARMState *env, uint32_t maskreg, uint32_t val) limit = is_psp ? env->v7m.psplim[false] : env->v7m.msplim[false]; if (val < limit) { - CPUState *cs = env_cpu(env); - - cpu_restore_state(cs, GETPC(), true); - raise_exception(env, EXCP_STKOF, 0, 1); + raise_exception_ra(env, EXCP_STKOF, 0, 1, GETPC()); } if (is_psp) { diff --git a/target/arm/op_helper.c b/target/arm/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/op_helper.c +++ b/target/arm/op_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(v8m_stackcheck)(CPUARMState *env, uint32_t newvalue) * raising an exception if the limit is breached. */ if (newvalue < v7m_sp_limit(env)) { - CPUState *cs = env_cpu(env); - /* * Stack limit exceptions are a rare case, so rather than syncing - * PC/condbits before the call, we use cpu_restore_state() to - * get them right before raising the exception. + * PC/condbits before the call, we use raise_exception_ra() so + * that cpu_restore_state() will sort them out. */ - cpu_restore_state(cs, GETPC(), true); - raise_exception(env, EXCP_STKOF, 0, 1); + raise_exception_ra(env, EXCP_STKOF, 0, 1, GETPC()); } } -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Note that the SVE BFLOAT16 support does not require SVE2, it is an independent extension. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-2-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 15 +++++++++++++++ 1 file changed, 15 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 @@ static inline bool isar_feature_aa32_predinv(const ARMISARegisters *id) return FIELD_EX32(id->id_isar6, ID_ISAR6, SPECRES) != 0; } +static inline bool isar_feature_aa32_bf16(const ARMISARegisters *id) +{ + return FIELD_EX32(id->id_isar6, ID_ISAR6, BF16) != 0; +} + static inline bool isar_feature_aa32_i8mm(const ARMISARegisters *id) { return FIELD_EX32(id->id_isar6, ID_ISAR6, I8MM) != 0; @@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_dcpodp(const ARMISARegisters *id) return FIELD_EX64(id->id_aa64isar1, ID_AA64ISAR1, DPB) >= 2; } +static inline bool isar_feature_aa64_bf16(const ARMISARegisters *id) +{ + return FIELD_EX64(id->id_aa64isar1, ID_AA64ISAR1, BF16) != 0; +} + static inline bool isar_feature_aa64_fp_simd(const ARMISARegisters *id) { /* We always set the AdvSIMD and FP fields identically. */ @@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_sve2_bitperm(const ARMISARegisters *id) return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, BITPERM) != 0; } +static inline bool isar_feature_aa64_sve_bf16(const ARMISARegisters *id) +{ + return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, BFLOAT16) != 0; +} + static inline bool isar_feature_aa64_sve2_sha3(const ARMISARegisters *id) { return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, SHA3) != 0; -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-3-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate-a64.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_fp_1src(DisasContext *s, uint32_t insn) int rd = extract32(insn, 0, 5); if (mos) { - unallocated_encoding(s); - return; + goto do_unallocated; } switch (opcode) { @@ -XXX,XX +XXX,XX @@ static void disas_fp_1src(DisasContext *s, uint32_t insn) /* FCVT between half, single and double precision */ int dtype = extract32(opcode, 0, 2); if (type == 2 || dtype == type) { - unallocated_encoding(s); - return; + goto do_unallocated; } if (!fp_access_check(s)) { return; @@ -XXX,XX +XXX,XX @@ static void disas_fp_1src(DisasContext *s, uint32_t insn) case 0x10 ... 0x13: /* FRINT{32,64}{X,Z} */ if (type > 1 || !dc_isar_feature(aa64_frint, s)) { - unallocated_encoding(s); - return; + goto do_unallocated; } /* fall through */ case 0x0 ... 0x3: @@ -XXX,XX +XXX,XX @@ static void disas_fp_1src(DisasContext *s, uint32_t insn) break; case 3: if (!dc_isar_feature(aa64_fp16, s)) { - unallocated_encoding(s); - return; + goto do_unallocated; } if (!fp_access_check(s)) { @@ -XXX,XX +XXX,XX @@ static void disas_fp_1src(DisasContext *s, uint32_t insn) handle_fp_1src_half(s, opcode, rd, rn); break; default: - unallocated_encoding(s); + goto do_unallocated; } break; default: + do_unallocated: unallocated_encoding(s); break; } -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> This is the 64-bit BFCVT and the 32-bit VCVT{B,T}.BF16.F32. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-4-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 1 + target/arm/vfp.decode | 2 ++ target/arm/translate-a64.c | 19 +++++++++++++++++++ target/arm/translate-vfp.c | 24 ++++++++++++++++++++++++ target/arm/vfp_helper.c | 5 +++++ 5 files changed, 51 insertions(+) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_3(vfp_cmped, void, f64, f64, env) DEF_HELPER_2(vfp_fcvtds, f64, f32, env) DEF_HELPER_2(vfp_fcvtsd, f32, f64, env) +DEF_HELPER_FLAGS_2(bfcvt, TCG_CALL_NO_RWG, i32, f32, ptr) DEF_HELPER_2(vfp_uitoh, f16, i32, ptr) DEF_HELPER_2(vfp_uitos, f32, i32, ptr) diff --git a/target/arm/vfp.decode b/target/arm/vfp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vfp.decode +++ b/target/arm/vfp.decode @@ -XXX,XX +XXX,XX @@ VCVT_f64_f16 ---- 1110 1.11 0010 .... 1011 t:1 1.0 .... \ # VCVTB and VCVTT to f16: Vd format is always vd_sp; # Vm format depends on size bit +VCVT_b16_f32 ---- 1110 1.11 0011 .... 1001 t:1 1.0 .... \ + vd=%vd_sp vm=%vm_sp VCVT_f16_f32 ---- 1110 1.11 0011 .... 1010 t:1 1.0 .... \ vd=%vd_sp vm=%vm_sp VCVT_f16_f64 ---- 1110 1.11 0011 .... 1011 t:1 1.0 .... \ diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void handle_fp_1src_single(DisasContext *s, int opcode, int rd, int rn) case 0x3: /* FSQRT */ gen_helper_vfp_sqrts(tcg_res, tcg_op, cpu_env); goto done; + case 0x6: /* BFCVT */ + gen_fpst = gen_helper_bfcvt; + break; case 0x8: /* FRINTN */ case 0x9: /* FRINTP */ case 0xa: /* FRINTM */ @@ -XXX,XX +XXX,XX @@ static void disas_fp_1src(DisasContext *s, uint32_t insn) } break; + case 0x6: + switch (type) { + case 1: /* BFCVT */ + if (!dc_isar_feature(aa64_bf16, s)) { + goto do_unallocated; + } + if (!fp_access_check(s)) { + return; + } + handle_fp_1src_single(s, opcode, rd, rn); + break; + default: + goto do_unallocated; + } + break; + default: do_unallocated: unallocated_encoding(s); diff --git a/target/arm/translate-vfp.c b/target/arm/translate-vfp.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-vfp.c +++ b/target/arm/translate-vfp.c @@ -XXX,XX +XXX,XX @@ static bool trans_VCVT_f64_f16(DisasContext *s, arg_VCVT_f64_f16 *a) return true; } +static bool trans_VCVT_b16_f32(DisasContext *s, arg_VCVT_b16_f32 *a) +{ + TCGv_ptr fpst; + TCGv_i32 tmp; + + if (!dc_isar_feature(aa32_bf16, s)) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + fpst = fpstatus_ptr(FPST_FPCR); + tmp = tcg_temp_new_i32(); + + vfp_load_reg32(tmp, a->vm); + gen_helper_bfcvt(tmp, tmp, fpst); + tcg_gen_st16_i32(tmp, cpu_env, vfp_f16_offset(a->vd, a->t)); + tcg_temp_free_ptr(fpst); + tcg_temp_free_i32(tmp); + return true; +} + static bool trans_VCVT_f16_f32(DisasContext *s, arg_VCVT_f16_f32 *a) { TCGv_ptr fpst; diff --git a/target/arm/vfp_helper.c b/target/arm/vfp_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vfp_helper.c +++ b/target/arm/vfp_helper.c @@ -XXX,XX +XXX,XX @@ float32 VFP_HELPER(fcvts, d)(float64 x, CPUARMState *env) return float64_to_float32(x, &env->vfp.fp_status); } +uint32_t HELPER(bfcvt)(float32 x, void *status) +{ + return float32_to_bfloat16(x, status); +} + /* * VFP3 fixed point conversion. The AArch32 versions of fix-to-float * must always round-to-nearest; the AArch64 ones honour the FPSCR -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> This is BFCVT{N,T} for both AArch64 AdvSIMD and SVE, and VCVT.BF16.F32 for AArch32 NEON. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-5-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper-sve.h | 4 ++++ target/arm/helper.h | 1 + target/arm/neon-dp.decode | 1 + target/arm/sve.decode | 2 ++ target/arm/sve_helper.c | 2 ++ target/arm/translate-a64.c | 17 ++++++++++++++ target/arm/translate-neon.c | 45 +++++++++++++++++++++++++++++++++++++ target/arm/translate-sve.c | 16 +++++++++++++ target/arm/vfp_helper.c | 7 ++++++ 9 files changed, 95 insertions(+) diff --git a/target/arm/helper-sve.h b/target/arm/helper-sve.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper-sve.h +++ b/target/arm/helper-sve.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_5(sve_fcvt_hd, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_5(sve_fcvt_sd, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_5(sve_bfcvt, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_5(sve_fcvtzs_hh, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_5(sve2_fcvtnt_sh, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_5(sve2_fcvtnt_ds, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_5(sve_bfcvtnt, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_5(sve2_fcvtlt_hs, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_3(vfp_cmped, void, f64, f64, env) DEF_HELPER_2(vfp_fcvtds, f64, f32, env) DEF_HELPER_2(vfp_fcvtsd, f32, f64, env) DEF_HELPER_FLAGS_2(bfcvt, TCG_CALL_NO_RWG, i32, f32, ptr) +DEF_HELPER_FLAGS_2(bfcvt_pair, TCG_CALL_NO_RWG, i32, i64, ptr) DEF_HELPER_2(vfp_uitoh, f16, i32, ptr) DEF_HELPER_2(vfp_uitos, f32, i32, ptr) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ Vimm_1r 1111 001 . 1 . 000 ... .... cmode:4 0 . op:1 1 .... @1reg_imm VRINTZ 1111 001 11 . 11 .. 10 .... 0 1011 . . 0 .... @2misc VCVT_F16_F32 1111 001 11 . 11 .. 10 .... 0 1100 0 . 0 .... @2misc_q0 + VCVT_B16_F32 1111 001 11 . 11 .. 10 .... 0 1100 1 . 0 .... @2misc_q0 VRINTM 1111 001 11 . 11 .. 10 .... 0 1101 . . 0 .... @2misc diff --git a/target/arm/sve.decode b/target/arm/sve.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve.decode +++ b/target/arm/sve.decode @@ -XXX,XX +XXX,XX @@ FNMLS_zpzzz 01100101 .. 1 ..... 111 ... ..... ..... @rdn_pg_rm_ra # SVE floating-point convert precision FCVT_sh 01100101 10 0010 00 101 ... ..... ..... @rd_pg_rn_e0 FCVT_hs 01100101 10 0010 01 101 ... ..... ..... @rd_pg_rn_e0 +BFCVT 01100101 10 0010 10 101 ... ..... ..... @rd_pg_rn_e0 FCVT_dh 01100101 11 0010 00 101 ... ..... ..... @rd_pg_rn_e0 FCVT_hd 01100101 11 0010 01 101 ... ..... ..... @rd_pg_rn_e0 FCVT_ds 01100101 11 0010 10 101 ... ..... ..... @rd_pg_rn_e0 @@ -XXX,XX +XXX,XX @@ RAX1 01000101 00 1 ..... 11110 1 ..... ..... @rd_rn_rm_e0 FCVTXNT_ds 01100100 00 0010 10 101 ... ..... ..... @rd_pg_rn_e0 FCVTX_ds 01100101 00 0010 10 101 ... ..... ..... @rd_pg_rn_e0 FCVTNT_sh 01100100 10 0010 00 101 ... ..... ..... @rd_pg_rn_e0 +BFCVTNT 01100100 10 0010 10 101 ... ..... ..... @rd_pg_rn_e0 FCVTLT_hs 01100100 10 0010 01 101 ... ..... ..... @rd_pg_rn_e0 FCVTNT_ds 01100100 11 0010 10 101 ... ..... ..... @rd_pg_rn_e0 FCVTLT_sd 01100100 11 0010 11 101 ... ..... ..... @rd_pg_rn_e0 diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ static inline uint64_t vfp_float64_to_uint64_rtz(float64 f, float_status *s) DO_ZPZ_FP(sve_fcvt_sh, uint32_t, H1_4, sve_f32_to_f16) DO_ZPZ_FP(sve_fcvt_hs, uint32_t, H1_4, sve_f16_to_f32) +DO_ZPZ_FP(sve_bfcvt, uint32_t, H1_4, float32_to_bfloat16) DO_ZPZ_FP(sve_fcvt_dh, uint64_t, , sve_f64_to_f16) DO_ZPZ_FP(sve_fcvt_hd, uint64_t, , sve_f16_to_f64) DO_ZPZ_FP(sve_fcvt_ds, uint64_t, , float64_to_float32) @@ -XXX,XX +XXX,XX @@ void HELPER(NAME)(void *vd, void *vn, void *vg, void *status, uint32_t desc) \ } while (i != 0); \ } +DO_FCVTNT(sve_bfcvtnt, uint32_t, uint16_t, H1_4, H1_2, float32_to_bfloat16) DO_FCVTNT(sve2_fcvtnt_sh, uint32_t, uint16_t, H1_4, H1_2, sve_f32_to_f16) DO_FCVTNT(sve2_fcvtnt_ds, uint64_t, uint32_t, , H1_4, float64_to_float32) diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void handle_2misc_narrow(DisasContext *s, bool scalar, tcg_temp_free_i32(ahp); } break; + case 0x36: /* BFCVTN, BFCVTN2 */ + { + TCGv_ptr fpst = fpstatus_ptr(FPST_FPCR); + gen_helper_bfcvt_pair(tcg_res[pass], tcg_op, fpst); + tcg_temp_free_ptr(fpst); + } + break; case 0x56: /* FCVTXN, FCVTXN2 */ /* 64 bit to 32 bit float conversion * with von Neumann rounding (round to odd) @@ -XXX,XX +XXX,XX @@ static void disas_simd_two_reg_misc(DisasContext *s, uint32_t insn) } handle_2misc_narrow(s, false, opcode, 0, is_q, size - 1, rn, rd); return; + case 0x36: /* BFCVTN, BFCVTN2 */ + if (!dc_isar_feature(aa64_bf16, s) || size != 2) { + unallocated_encoding(s); + return; + } + if (!fp_access_check(s)) { + return; + } + handle_2misc_narrow(s, false, opcode, 0, is_q, size - 1, rn, rd); + return; case 0x17: /* FCVTL, FCVTL2 */ if (!fp_access_check(s)) { return; diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.c +++ b/target/arm/translate-neon.c @@ -XXX,XX +XXX,XX @@ static bool trans_VSHLL(DisasContext *s, arg_2misc *a) return true; } +static bool trans_VCVT_B16_F32(DisasContext *s, arg_2misc *a) +{ + TCGv_ptr fpst; + TCGv_i64 tmp; + TCGv_i32 dst0, dst1; + + if (!dc_isar_feature(aa32_bf16, s)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vm) & 0x10)) { + return false; + } + + if ((a->vm & 1) || (a->size != 1)) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + fpst = fpstatus_ptr(FPST_STD); + tmp = tcg_temp_new_i64(); + dst0 = tcg_temp_new_i32(); + dst1 = tcg_temp_new_i32(); + + read_neon_element64(tmp, a->vm, 0, MO_64); + gen_helper_bfcvt_pair(dst0, tmp, fpst); + + read_neon_element64(tmp, a->vm, 1, MO_64); + gen_helper_bfcvt_pair(dst1, tmp, fpst); + + write_neon_element32(dst0, a->vd, 0, MO_32); + write_neon_element32(dst1, a->vd, 1, MO_32); + + tcg_temp_free_i64(tmp); + tcg_temp_free_i32(dst0); + tcg_temp_free_i32(dst1); + tcg_temp_free_ptr(fpst); + return true; +} + static bool trans_VCVT_F16_F32(DisasContext *s, arg_2misc *a) { TCGv_ptr fpst; diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static bool trans_FCVT_hs(DisasContext *s, arg_rpr_esz *a) return do_zpz_ptr(s, a->rd, a->rn, a->pg, false, gen_helper_sve_fcvt_hs); } +static bool trans_BFCVT(DisasContext *s, arg_rpr_esz *a) +{ + if (!dc_isar_feature(aa64_sve_bf16, s)) { + return false; + } + return do_zpz_ptr(s, a->rd, a->rn, a->pg, false, gen_helper_sve_bfcvt); +} + static bool trans_FCVT_dh(DisasContext *s, arg_rpr_esz *a) { return do_zpz_ptr(s, a->rd, a->rn, a->pg, false, gen_helper_sve_fcvt_dh); @@ -XXX,XX +XXX,XX @@ static bool trans_FCVTNT_sh(DisasContext *s, arg_rpr_esz *a) return do_zpz_ptr(s, a->rd, a->rn, a->pg, false, gen_helper_sve2_fcvtnt_sh); } +static bool trans_BFCVTNT(DisasContext *s, arg_rpr_esz *a) +{ + if (!dc_isar_feature(aa64_sve_bf16, s)) { + return false; + } + return do_zpz_ptr(s, a->rd, a->rn, a->pg, false, gen_helper_sve_bfcvtnt); +} + static bool trans_FCVTNT_ds(DisasContext *s, arg_rpr_esz *a) { if (!dc_isar_feature(aa64_sve2, s)) { diff --git a/target/arm/vfp_helper.c b/target/arm/vfp_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vfp_helper.c +++ b/target/arm/vfp_helper.c @@ -XXX,XX +XXX,XX @@ uint32_t HELPER(bfcvt)(float32 x, void *status) return float32_to_bfloat16(x, status); } +uint32_t HELPER(bfcvt_pair)(uint64_t pair, void *status) +{ + bfloat16 lo = float32_to_bfloat16(extract64(pair, 0, 32), status); + bfloat16 hi = float32_to_bfloat16(extract64(pair, 32, 32), status); + return deposit32(lo, 16, 16, hi); +} + /* * VFP3 fixed point conversion. The AArch32 versions of fix-to-float * must always round-to-nearest; the AArch64 ones honour the FPSCR -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> For Arm BFDOT and BFMMLA, we need a version of round-to-odd that overflows to infinity, instead of the max normal number. Cc: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-6-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/fpu/softfloat-types.h | 4 +++- fpu/softfloat-parts.c.inc | 6 ++++-- 2 files changed, 7 insertions(+), 3 deletions(-) diff --git a/include/fpu/softfloat-types.h b/include/fpu/softfloat-types.h index XXXXXXX..XXXXXXX 100644 --- a/include/fpu/softfloat-types.h +++ b/include/fpu/softfloat-types.h @@ -XXX,XX +XXX,XX @@ typedef enum __attribute__((__packed__)) { float_round_up = 2, float_round_to_zero = 3, float_round_ties_away = 4, - /* Not an IEEE rounding mode: round to the closest odd mantissa value */ + /* Not an IEEE rounding mode: round to closest odd, overflow to max */ float_round_to_odd = 5, + /* Not an IEEE rounding mode: round to closest odd, overflow to inf */ + float_round_to_odd_inf = 6, } FloatRoundMode; /* diff --git a/fpu/softfloat-parts.c.inc b/fpu/softfloat-parts.c.inc index XXXXXXX..XXXXXXX 100644 --- a/fpu/softfloat-parts.c.inc +++ b/fpu/softfloat-parts.c.inc @@ -XXX,XX +XXX,XX @@ static void partsN(uncanon)(FloatPartsN *p, float_status *s, g_assert_not_reached(); } + overflow_norm = false; switch (s->float_rounding_mode) { case float_round_nearest_even: - overflow_norm = false; inc = ((p->frac_lo & roundeven_mask) != frac_lsbm1 ? frac_lsbm1 : 0); break; case float_round_ties_away: - overflow_norm = false; inc = frac_lsbm1; break; case float_round_to_zero: @@ -XXX,XX +XXX,XX @@ static void partsN(uncanon)(FloatPartsN *p, float_status *s, break; case float_round_to_odd: overflow_norm = true; + /* fall through */ + case float_round_to_odd_inf: inc = p->frac_lo & frac_lsb ? 0 : round_mask; break; default: @@ -XXX,XX +XXX,XX @@ static void partsN(uncanon)(FloatPartsN *p, float_status *s, ? frac_lsbm1 : 0); break; case float_round_to_odd: + case float_round_to_odd_inf: inc = p->frac_lo & frac_lsb ? 0 : round_mask; break; default: -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> This is BFDOT for both AArch64 AdvSIMD and SVE, and VDOT.BF16 for AArch32 NEON. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-7-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 3 +++ target/arm/neon-shared.decode | 2 ++ target/arm/sve.decode | 3 +++ target/arm/translate-a64.c | 20 ++++++++++++++++++ target/arm/translate-neon.c | 9 ++++++++ target/arm/translate-sve.c | 12 +++++++++++ target/arm/vec_helper.c | 40 +++++++++++++++++++++++++++++++++++ 7 files changed, 89 insertions(+) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_5(gvec_ummla_b, TCG_CALL_NO_RWG, DEF_HELPER_FLAGS_5(gvec_usmmla_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_5(gvec_bfdot, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, ptr, i32) + #ifdef TARGET_AARCH64 #include "helper-a64.h" #include "helper-sve.h" diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VUDOT 1111 110 00 . 10 .... .... 1101 . q:1 . 1 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp VUSDOT 1111 110 01 . 10 .... .... 1101 . q:1 . 0 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp +VDOT_b16 1111 110 00 . 00 .... .... 1101 . q:1 . 0 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp # VFM[AS]L VFML 1111 110 0 s:1 . 10 .... .... 1000 . 0 . 1 .... \ diff --git a/target/arm/sve.decode b/target/arm/sve.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve.decode +++ b/target/arm/sve.decode @@ -XXX,XX +XXX,XX @@ FMLALT_zzzw 01100100 10 1 ..... 10 0 00 1 ..... ..... @rda_rn_rm_e0 FMLSLB_zzzw 01100100 10 1 ..... 10 1 00 0 ..... ..... @rda_rn_rm_e0 FMLSLT_zzzw 01100100 10 1 ..... 10 1 00 1 ..... ..... @rda_rn_rm_e0 +### SVE2 floating-point bfloat16 dot-product +BFDOT_zzzz 01100100 01 1 ..... 10 0 00 0 ..... ..... @rda_rn_rm_e0 + ### SVE2 floating-point multiply-add long (indexed) FMLALB_zzxw 01100100 10 1 ..... 0100.0 ..... ..... @rrxr_3a esz=2 FMLALT_zzxw 01100100 10 1 ..... 0100.1 ..... ..... @rrxr_3a esz=2 diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_simd_three_reg_same_extra(DisasContext *s, uint32_t insn) } feature = dc_isar_feature(aa64_fcma, s); break; + case 0x1f: /* BFDOT */ + switch (size) { + case 1: + feature = dc_isar_feature(aa64_bf16, s); + break; + default: + unallocated_encoding(s); + return; + } + break; default: unallocated_encoding(s); return; @@ -XXX,XX +XXX,XX @@ static void disas_simd_three_reg_same_extra(DisasContext *s, uint32_t insn) } return; + case 0xf: /* BFDOT */ + switch (size) { + case 1: + gen_gvec_op4_ool(s, is_q, rd, rn, rm, rd, 0, gen_helper_gvec_bfdot); + break; + default: + g_assert_not_reached(); + } + return; + default: g_assert_not_reached(); } diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.c +++ b/target/arm/translate-neon.c @@ -XXX,XX +XXX,XX @@ static bool trans_VUSDOT(DisasContext *s, arg_VUSDOT *a) gen_helper_gvec_usdot_b); } +static bool trans_VDOT_b16(DisasContext *s, arg_VDOT_b16 *a) +{ + if (!dc_isar_feature(aa32_bf16, s)) { + return false; + } + return do_neon_ddda(s, a->q * 7, a->vd, a->vn, a->vm, 0, + gen_helper_gvec_bfdot); +} + static bool trans_VFML(DisasContext *s, arg_VFML *a) { int opr_sz; diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static bool trans_UMMLA(DisasContext *s, arg_rrrr_esz *a) { return do_i8mm_zzzz_ool(s, a, gen_helper_gvec_ummla_b, 0); } + +static bool trans_BFDOT_zzzz(DisasContext *s, arg_rrrr_esz *a) +{ + if (!dc_isar_feature(aa64_sve_bf16, s)) { + return false; + } + if (sve_access_check(s)) { + gen_gvec_ool_zzzz(s, gen_helper_gvec_bfdot, + a->rd, a->rn, a->rm, a->ra, 0); + } + return true; +} diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ static void do_mmla_b(void *vd, void *vn, void *vm, void *va, uint32_t desc, DO_MMLA_B(gvec_smmla_b, do_smmla_b) DO_MMLA_B(gvec_ummla_b, do_ummla_b) DO_MMLA_B(gvec_usmmla_b, do_usmmla_b) + +/* + * BFloat16 Dot Product + */ + +static float32 bfdotadd(float32 sum, uint32_t e1, uint32_t e2) +{ + /* FPCR is ignored for BFDOT and BFMMLA. */ + float_status bf_status = { + .tininess_before_rounding = float_tininess_before_rounding, + .float_rounding_mode = float_round_to_odd_inf, + .flush_to_zero = true, + .flush_inputs_to_zero = true, + .default_nan_mode = true, + }; + float32 t1, t2; + + /* + * Extract each BFloat16 from the element pair, and shift + * them such that they become float32. + */ + t1 = float32_mul(e1 << 16, e2 << 16, &bf_status); + t2 = float32_mul(e1 & 0xffff0000u, e2 & 0xffff0000u, &bf_status); + t1 = float32_add(t1, t2, &bf_status); + t1 = float32_add(sum, t1, &bf_status); + + return t1; +} + +void HELPER(gvec_bfdot)(void *vd, void *vn, void *vm, void *va, uint32_t desc) +{ + intptr_t i, opr_sz = simd_oprsz(desc); + float32 *d = vd, *a = va; + uint32_t *n = vn, *m = vm; + + for (i = 0; i < opr_sz / 4; ++i) { + d[i] = bfdotadd(a[i], n[i], m[i]); + } + clear_tail(d, opr_sz, simd_maxsz(desc)); +} -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> This is BFDOT for both AArch64 AdvSIMD and SVE, and VDOT.BF16 for AArch32 NEON. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-8-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 2 ++ target/arm/neon-shared.decode | 2 ++ target/arm/sve.decode | 3 +++ target/arm/translate-a64.c | 41 +++++++++++++++++++++++++++-------- target/arm/translate-neon.c | 9 ++++++++ target/arm/translate-sve.c | 12 ++++++++++ target/arm/vec_helper.c | 20 +++++++++++++++++ 7 files changed, 80 insertions(+), 9 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_5(gvec_usmmla_b, TCG_CALL_NO_RWG, DEF_HELPER_FLAGS_5(gvec_bfdot, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_5(gvec_bfdot_idx, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, ptr, i32) #ifdef TARGET_AARCH64 #include "helper-a64.h" diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VUSDOT_scalar 1111 1110 1 . 00 .... .... 1101 . q:1 index:1 0 vm:4 \ vn=%vn_dp vd=%vd_dp VSUDOT_scalar 1111 1110 1 . 00 .... .... 1101 . q:1 index:1 1 vm:4 \ vn=%vn_dp vd=%vd_dp +VDOT_b16_scal 1111 1110 0 . 00 .... .... 1101 . q:1 index:1 0 vm:4 \ + vn=%vn_dp vd=%vd_dp %vfml_scalar_q0_rm 0:3 5:1 %vfml_scalar_q1_index 5:1 3:1 diff --git a/target/arm/sve.decode b/target/arm/sve.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve.decode +++ b/target/arm/sve.decode @@ -XXX,XX +XXX,XX @@ FMLALB_zzxw 01100100 10 1 ..... 0100.0 ..... ..... @rrxr_3a esz=2 FMLALT_zzxw 01100100 10 1 ..... 0100.1 ..... ..... @rrxr_3a esz=2 FMLSLB_zzxw 01100100 10 1 ..... 0110.0 ..... ..... @rrxr_3a esz=2 FMLSLT_zzxw 01100100 10 1 ..... 0110.1 ..... ..... @rrxr_3a esz=2 + +### SVE2 floating-point bfloat16 dot-product (indexed) +BFDOT_zzxz 01100100 01 1 ..... 010000 ..... ..... @rrxr_2 esz=2 diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_simd_indexed(DisasContext *s, uint32_t insn) return; } break; - case 0x0f: /* SUDOT, USDOT */ - if (is_scalar || (size & 1) || !dc_isar_feature(aa64_i8mm, s)) { + case 0x0f: + switch (size) { + case 0: /* SUDOT */ + case 2: /* USDOT */ + if (is_scalar || !dc_isar_feature(aa64_i8mm, s)) { + unallocated_encoding(s); + return; + } + break; + case 1: /* BFDOT */ + if (is_scalar || !dc_isar_feature(aa64_bf16, s)) { + unallocated_encoding(s); + return; + } + break; + default: unallocated_encoding(s); return; } @@ -XXX,XX +XXX,XX @@ static void disas_simd_indexed(DisasContext *s, uint32_t insn) u ? gen_helper_gvec_udot_idx_b : gen_helper_gvec_sdot_idx_b); return; - case 0x0f: /* SUDOT, USDOT */ - gen_gvec_op4_ool(s, is_q, rd, rn, rm, rd, index, - extract32(insn, 23, 1) - ? gen_helper_gvec_usdot_idx_b - : gen_helper_gvec_sudot_idx_b); - return; - + case 0x0f: + switch (extract32(insn, 22, 2)) { + case 0: /* SUDOT */ + gen_gvec_op4_ool(s, is_q, rd, rn, rm, rd, index, + gen_helper_gvec_sudot_idx_b); + return; + case 1: /* BFDOT */ + gen_gvec_op4_ool(s, is_q, rd, rn, rm, rd, index, + gen_helper_gvec_bfdot_idx); + return; + case 2: /* USDOT */ + gen_gvec_op4_ool(s, is_q, rd, rn, rm, rd, index, + gen_helper_gvec_usdot_idx_b); + return; + } + g_assert_not_reached(); case 0x11: /* FCMLA #0 */ case 0x13: /* FCMLA #90 */ case 0x15: /* FCMLA #180 */ diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.c +++ b/target/arm/translate-neon.c @@ -XXX,XX +XXX,XX @@ static bool trans_VSUDOT_scalar(DisasContext *s, arg_VSUDOT_scalar *a) gen_helper_gvec_sudot_idx_b); } +static bool trans_VDOT_b16_scal(DisasContext *s, arg_VDOT_b16_scal *a) +{ + if (!dc_isar_feature(aa32_bf16, s)) { + return false; + } + return do_neon_ddda(s, a->q * 6, a->vd, a->vn, a->vm, a->index, + gen_helper_gvec_bfdot_idx); +} + static bool trans_VFML_scalar(DisasContext *s, arg_VFML_scalar *a) { int opr_sz; diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static bool trans_BFDOT_zzzz(DisasContext *s, arg_rrrr_esz *a) } return true; } + +static bool trans_BFDOT_zzxz(DisasContext *s, arg_rrxr_esz *a) +{ + if (!dc_isar_feature(aa64_sve_bf16, s)) { + return false; + } + if (sve_access_check(s)) { + gen_gvec_ool_zzzz(s, gen_helper_gvec_bfdot_idx, + a->rd, a->rn, a->rm, a->ra, a->index); + } + return true; +} diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(gvec_bfdot)(void *vd, void *vn, void *vm, void *va, uint32_t desc) } clear_tail(d, opr_sz, simd_maxsz(desc)); } + +void HELPER(gvec_bfdot_idx)(void *vd, void *vn, void *vm, + void *va, uint32_t desc) +{ + intptr_t i, j, opr_sz = simd_oprsz(desc); + intptr_t index = simd_data(desc); + intptr_t elements = opr_sz / 4; + intptr_t eltspersegment = MIN(16 / 4, elements); + float32 *d = vd, *a = va; + uint32_t *n = vn, *m = vm; + + for (i = 0; i < elements; i += eltspersegment) { + uint32_t m_idx = m[i + H4(index)]; + + for (j = i; j < i + eltspersegment; j++) { + d[j] = bfdotadd(a[j], n[j], m_idx); + } + } + clear_tail(d, opr_sz, simd_maxsz(desc)); +} -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> This is BFMMLA for both AArch64 AdvSIMD and SVE, and VMMLA.BF16 for AArch32 NEON. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-9-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 3 +++ target/arm/neon-shared.decode | 2 ++ target/arm/sve.decode | 6 +++-- target/arm/translate-a64.c | 10 +++++++++ target/arm/translate-neon.c | 9 ++++++++ target/arm/translate-sve.c | 12 ++++++++++ target/arm/vec_helper.c | 42 ++++++++++++++++++++++++++++++++++- 7 files changed, 81 insertions(+), 3 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_5(gvec_bfdot, TCG_CALL_NO_RWG, DEF_HELPER_FLAGS_5(gvec_bfdot_idx, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_5(gvec_bfmmla, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, ptr, i32) + #ifdef TARGET_AARCH64 #include "helper-a64.h" #include "helper-sve.h" diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VUMMLA 1111 1100 0.10 .... .... 1100 .1.1 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp VUSMMLA 1111 1100 1.10 .... .... 1100 .1.0 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp +VMMLA_b16 1111 1100 0.00 .... .... 1100 .1.0 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp VCMLA_scalar 1111 1110 0 . rot:2 .... .... 1000 . q:1 index:1 0 vm:4 \ vn=%vn_dp vd=%vd_dp size=1 diff --git a/target/arm/sve.decode b/target/arm/sve.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve.decode +++ b/target/arm/sve.decode @@ -XXX,XX +XXX,XX @@ SQRDCMLAH_zzzz 01000100 esz:2 0 rm:5 0011 rot:2 rn:5 rd:5 ra=%reg_movprfx USDOT_zzzz 01000100 .. 0 ..... 011 110 ..... ..... @rda_rn_rm ### SVE2 floating point matrix multiply accumulate - -FMMLA 01100100 .. 1 ..... 111001 ..... ..... @rda_rn_rm +{ + BFMMLA 01100100 01 1 ..... 111 001 ..... ..... @rda_rn_rm_e0 + FMMLA 01100100 .. 1 ..... 111 001 ..... ..... @rda_rn_rm +} ### SVE2 Memory Gather Load Group diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_simd_three_reg_same_extra(DisasContext *s, uint32_t insn) } feature = dc_isar_feature(aa64_fcma, s); break; + case 0x1d: /* BFMMLA */ + if (size != MO_16 || !is_q) { + unallocated_encoding(s); + return; + } + feature = dc_isar_feature(aa64_bf16, s); + break; case 0x1f: /* BFDOT */ switch (size) { case 1: @@ -XXX,XX +XXX,XX @@ static void disas_simd_three_reg_same_extra(DisasContext *s, uint32_t insn) } return; + case 0xd: /* BFMMLA */ + gen_gvec_op4_ool(s, is_q, rd, rn, rm, rd, 0, gen_helper_gvec_bfmmla); + return; case 0xf: /* BFDOT */ switch (size) { case 1: diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.c +++ b/target/arm/translate-neon.c @@ -XXX,XX +XXX,XX @@ static bool trans_VUSMMLA(DisasContext *s, arg_VUSMMLA *a) return do_neon_ddda(s, 7, a->vd, a->vn, a->vm, 0, gen_helper_gvec_usmmla_b); } + +static bool trans_VMMLA_b16(DisasContext *s, arg_VMMLA_b16 *a) +{ + if (!dc_isar_feature(aa32_bf16, s)) { + return false; + } + return do_neon_ddda(s, 7, a->vd, a->vn, a->vm, 0, + gen_helper_gvec_bfmmla); +} diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static bool trans_BFDOT_zzxz(DisasContext *s, arg_rrxr_esz *a) } return true; } + +static bool trans_BFMMLA(DisasContext *s, arg_rrrr_esz *a) +{ + if (!dc_isar_feature(aa64_sve_bf16, s)) { + return false; + } + if (sve_access_check(s)) { + gen_gvec_ool_zzzz(s, gen_helper_gvec_bfmmla, + a->rd, a->rn, a->rm, a->ra, 0); + } + return true; +} diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ static void do_mmla_b(void *vd, void *vn, void *vm, void *va, uint32_t desc, * Process the entire segment at once, writing back the * results only after we've consumed all of the inputs. * - * Key to indicies by column: + * Key to indices by column: * i j i j */ sum0 = a[H4(0 + 0)]; @@ -XXX,XX +XXX,XX @@ void HELPER(gvec_bfdot_idx)(void *vd, void *vn, void *vm, } clear_tail(d, opr_sz, simd_maxsz(desc)); } + +void HELPER(gvec_bfmmla)(void *vd, void *vn, void *vm, void *va, uint32_t desc) +{ + intptr_t s, opr_sz = simd_oprsz(desc); + float32 *d = vd, *a = va; + uint32_t *n = vn, *m = vm; + + for (s = 0; s < opr_sz / 4; s += 4) { + float32 sum00, sum01, sum10, sum11; + + /* + * Process the entire segment at once, writing back the + * results only after we've consumed all of the inputs. + * + * Key to indicies by column: + * i j i k j k + */ + sum00 = a[s + H4(0 + 0)]; + sum00 = bfdotadd(sum00, n[s + H4(0 + 0)], m[s + H4(0 + 0)]); + sum00 = bfdotadd(sum00, n[s + H4(0 + 1)], m[s + H4(0 + 1)]); + + sum01 = a[s + H4(0 + 1)]; + sum01 = bfdotadd(sum01, n[s + H4(0 + 0)], m[s + H4(2 + 0)]); + sum01 = bfdotadd(sum01, n[s + H4(0 + 1)], m[s + H4(2 + 1)]); + + sum10 = a[s + H4(2 + 0)]; + sum10 = bfdotadd(sum10, n[s + H4(2 + 0)], m[s + H4(0 + 0)]); + sum10 = bfdotadd(sum10, n[s + H4(2 + 1)], m[s + H4(0 + 1)]); + + sum11 = a[s + H4(2 + 1)]; + sum11 = bfdotadd(sum11, n[s + H4(2 + 0)], m[s + H4(2 + 0)]); + sum11 = bfdotadd(sum11, n[s + H4(2 + 1)], m[s + H4(2 + 1)]); + + d[s + H4(0 + 0)] = sum00; + d[s + H4(0 + 1)] = sum01; + d[s + H4(2 + 0)] = sum10; + d[s + H4(2 + 1)] = sum11; + } + clear_tail(d, opr_sz, simd_maxsz(desc)); +} -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> This is BFMLAL{B,T} for both AArch64 AdvSIMD and SVE, and VFMA{B,T}.BF16 for AArch32 NEON. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-10-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 3 +++ target/arm/neon-shared.decode | 3 +++ target/arm/sve.decode | 3 +++ target/arm/translate-a64.c | 13 +++++++++---- target/arm/translate-neon.c | 9 +++++++++ target/arm/translate-sve.c | 30 ++++++++++++++++++++++++++++++ target/arm/vec_helper.c | 16 ++++++++++++++++ 7 files changed, 73 insertions(+), 4 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_5(gvec_bfdot_idx, TCG_CALL_NO_RWG, DEF_HELPER_FLAGS_5(gvec_bfmmla, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_6(gvec_bfmlal, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, ptr, ptr, i32) + #ifdef TARGET_AARCH64 #include "helper-a64.h" #include "helper-sve.h" diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VUSMMLA 1111 1100 1.10 .... .... 1100 .1.0 .... \ VMMLA_b16 1111 1100 0.00 .... .... 1100 .1.0 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp +VFMA_b16 1111 110 0 0.11 .... .... 1000 . q:1 . 1 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp + VCMLA_scalar 1111 1110 0 . rot:2 .... .... 1000 . q:1 index:1 0 vm:4 \ vn=%vn_dp vd=%vd_dp size=1 VCMLA_scalar 1111 1110 1 . rot:2 .... .... 1000 . q:1 . 0 .... \ diff --git a/target/arm/sve.decode b/target/arm/sve.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve.decode +++ b/target/arm/sve.decode @@ -XXX,XX +XXX,XX @@ FMLALT_zzzw 01100100 10 1 ..... 10 0 00 1 ..... ..... @rda_rn_rm_e0 FMLSLB_zzzw 01100100 10 1 ..... 10 1 00 0 ..... ..... @rda_rn_rm_e0 FMLSLT_zzzw 01100100 10 1 ..... 10 1 00 1 ..... ..... @rda_rn_rm_e0 +BFMLALB_zzzw 01100100 11 1 ..... 10 0 00 0 ..... ..... @rda_rn_rm_e0 +BFMLALT_zzzw 01100100 11 1 ..... 10 0 00 1 ..... ..... @rda_rn_rm_e0 + ### SVE2 floating-point bfloat16 dot-product BFDOT_zzzz 01100100 01 1 ..... 10 0 00 0 ..... ..... @rda_rn_rm_e0 diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_simd_three_reg_same_extra(DisasContext *s, uint32_t insn) } feature = dc_isar_feature(aa64_bf16, s); break; - case 0x1f: /* BFDOT */ + case 0x1f: switch (size) { - case 1: + case 1: /* BFDOT */ + case 3: /* BFMLAL{B,T} */ feature = dc_isar_feature(aa64_bf16, s); break; default: @@ -XXX,XX +XXX,XX @@ static void disas_simd_three_reg_same_extra(DisasContext *s, uint32_t insn) case 0xd: /* BFMMLA */ gen_gvec_op4_ool(s, is_q, rd, rn, rm, rd, 0, gen_helper_gvec_bfmmla); return; - case 0xf: /* BFDOT */ + case 0xf: switch (size) { - case 1: + case 1: /* BFDOT */ gen_gvec_op4_ool(s, is_q, rd, rn, rm, rd, 0, gen_helper_gvec_bfdot); break; + case 3: /* BFMLAL{B,T} */ + gen_gvec_op4_fpst(s, 1, rd, rn, rm, rd, false, is_q, + gen_helper_gvec_bfmlal); + break; default: g_assert_not_reached(); } diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.c +++ b/target/arm/translate-neon.c @@ -XXX,XX +XXX,XX @@ static bool trans_VMMLA_b16(DisasContext *s, arg_VMMLA_b16 *a) return do_neon_ddda(s, 7, a->vd, a->vn, a->vm, 0, gen_helper_gvec_bfmmla); } + +static bool trans_VFMA_b16(DisasContext *s, arg_VFMA_b16 *a) +{ + if (!dc_isar_feature(aa32_bf16, s)) { + return false; + } + return do_neon_ddda_fpst(s, 7, a->vd, a->vn, a->vm, a->q, FPST_STD, + gen_helper_gvec_bfmlal); +} diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static bool trans_BFMMLA(DisasContext *s, arg_rrrr_esz *a) } return true; } + +static bool do_BFMLAL_zzzw(DisasContext *s, arg_rrrr_esz *a, bool sel) +{ + if (!dc_isar_feature(aa64_sve_bf16, s)) { + return false; + } + if (sve_access_check(s)) { + TCGv_ptr status = fpstatus_ptr(FPST_FPCR); + unsigned vsz = vec_full_reg_size(s); + + tcg_gen_gvec_4_ptr(vec_full_reg_offset(s, a->rd), + vec_full_reg_offset(s, a->rn), + vec_full_reg_offset(s, a->rm), + vec_full_reg_offset(s, a->ra), + status, vsz, vsz, sel, + gen_helper_gvec_bfmlal); + tcg_temp_free_ptr(status); + } + return true; +} + +static bool trans_BFMLALB_zzzw(DisasContext *s, arg_rrrr_esz *a) +{ + return do_BFMLAL_zzzw(s, a, false); +} + +static bool trans_BFMLALT_zzzw(DisasContext *s, arg_rrrr_esz *a) +{ + return do_BFMLAL_zzzw(s, a, true); +} diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(gvec_bfmmla)(void *vd, void *vn, void *vm, void *va, uint32_t desc) } clear_tail(d, opr_sz, simd_maxsz(desc)); } + +void HELPER(gvec_bfmlal)(void *vd, void *vn, void *vm, void *va, + void *stat, uint32_t desc) +{ + intptr_t i, opr_sz = simd_oprsz(desc); + intptr_t sel = simd_data(desc); + float32 *d = vd, *a = va; + bfloat16 *n = vn, *m = vm; + + for (i = 0; i < opr_sz / 4; ++i) { + float32 nn = n[H2(i * 2 + sel)] << 16; + float32 mm = m[H2(i * 2 + sel)] << 16; + d[H4(i)] = float32_muladd(nn, mm, a[H4(i)], 0, stat); + } + clear_tail(d, opr_sz, simd_maxsz(desc)); +} -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> This is BFMLAL{B,T} for both AArch64 AdvSIMD and SVE, and VFMA{B,T}.BF16 for AArch32 NEON. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-11-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 2 ++ target/arm/neon-shared.decode | 2 ++ target/arm/sve.decode | 2 ++ target/arm/translate-a64.c | 15 ++++++++++++++- target/arm/translate-neon.c | 10 ++++++++++ target/arm/translate-sve.c | 30 ++++++++++++++++++++++++++++++ target/arm/vec_helper.c | 22 ++++++++++++++++++++++ 7 files changed, 82 insertions(+), 1 deletion(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_5(gvec_bfmmla, TCG_CALL_NO_RWG, DEF_HELPER_FLAGS_6(gvec_bfmlal, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_6(gvec_bfmlal_idx, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, ptr, ptr, i32) #ifdef TARGET_AARCH64 #include "helper-a64.h" diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VFML_scalar 1111 1110 0 . 0 s:1 .... .... 1000 . 0 . 1 index:1 ... \ rm=%vfml_scalar_q0_rm vn=%vn_sp vd=%vd_dp q=0 VFML_scalar 1111 1110 0 . 0 s:1 .... .... 1000 . 1 . 1 . rm:3 \ index=%vfml_scalar_q1_index vn=%vn_dp vd=%vd_dp q=1 +VFMA_b16_scal 1111 1110 0.11 .... .... 1000 . q:1 . 1 . vm:3 \ + index=%vfml_scalar_q1_index vn=%vn_dp vd=%vd_dp diff --git a/target/arm/sve.decode b/target/arm/sve.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve.decode +++ b/target/arm/sve.decode @@ -XXX,XX +XXX,XX @@ FMLALB_zzxw 01100100 10 1 ..... 0100.0 ..... ..... @rrxr_3a esz=2 FMLALT_zzxw 01100100 10 1 ..... 0100.1 ..... ..... @rrxr_3a esz=2 FMLSLB_zzxw 01100100 10 1 ..... 0110.0 ..... ..... @rrxr_3a esz=2 FMLSLT_zzxw 01100100 10 1 ..... 0110.1 ..... ..... @rrxr_3a esz=2 +BFMLALB_zzxw 01100100 11 1 ..... 0100.0 ..... ..... @rrxr_3a esz=2 +BFMLALT_zzxw 01100100 11 1 ..... 0100.1 ..... ..... @rrxr_3a esz=2 ### SVE2 floating-point bfloat16 dot-product (indexed) BFDOT_zzxz 01100100 01 1 ..... 010000 ..... ..... @rrxr_2 esz=2 diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_simd_indexed(DisasContext *s, uint32_t insn) unallocated_encoding(s); return; } + size = MO_32; break; case 1: /* BFDOT */ if (is_scalar || !dc_isar_feature(aa64_bf16, s)) { unallocated_encoding(s); return; } + size = MO_32; + break; + case 3: /* BFMLAL{B,T} */ + if (is_scalar || !dc_isar_feature(aa64_bf16, s)) { + unallocated_encoding(s); + return; + } + /* can't set is_fp without other incorrect size checks */ + size = MO_16; break; default: unallocated_encoding(s); return; } - size = MO_32; break; case 0x11: /* FCMLA #0 */ case 0x13: /* FCMLA #90 */ @@ -XXX,XX +XXX,XX @@ static void disas_simd_indexed(DisasContext *s, uint32_t insn) gen_gvec_op4_ool(s, is_q, rd, rn, rm, rd, index, gen_helper_gvec_usdot_idx_b); return; + case 3: /* BFMLAL{B,T} */ + gen_gvec_op4_fpst(s, 1, rd, rn, rm, rd, 0, (index << 1) | is_q, + gen_helper_gvec_bfmlal_idx); + return; } g_assert_not_reached(); case 0x11: /* FCMLA #0 */ diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.c +++ b/target/arm/translate-neon.c @@ -XXX,XX +XXX,XX @@ static bool trans_VFMA_b16(DisasContext *s, arg_VFMA_b16 *a) return do_neon_ddda_fpst(s, 7, a->vd, a->vn, a->vm, a->q, FPST_STD, gen_helper_gvec_bfmlal); } + +static bool trans_VFMA_b16_scal(DisasContext *s, arg_VFMA_b16_scal *a) +{ + if (!dc_isar_feature(aa32_bf16, s)) { + return false; + } + return do_neon_ddda_fpst(s, 6, a->vd, a->vn, a->vm, + (a->index << 1) | a->q, FPST_STD, + gen_helper_gvec_bfmlal_idx); +} diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static bool trans_BFMLALT_zzzw(DisasContext *s, arg_rrrr_esz *a) { return do_BFMLAL_zzzw(s, a, true); } + +static bool do_BFMLAL_zzxw(DisasContext *s, arg_rrxr_esz *a, bool sel) +{ + if (!dc_isar_feature(aa64_sve_bf16, s)) { + return false; + } + if (sve_access_check(s)) { + TCGv_ptr status = fpstatus_ptr(FPST_FPCR); + unsigned vsz = vec_full_reg_size(s); + + tcg_gen_gvec_4_ptr(vec_full_reg_offset(s, a->rd), + vec_full_reg_offset(s, a->rn), + vec_full_reg_offset(s, a->rm), + vec_full_reg_offset(s, a->ra), + status, vsz, vsz, (a->index << 1) | sel, + gen_helper_gvec_bfmlal_idx); + tcg_temp_free_ptr(status); + } + return true; +} + +static bool trans_BFMLALB_zzxw(DisasContext *s, arg_rrxr_esz *a) +{ + return do_BFMLAL_zzxw(s, a, false); +} + +static bool trans_BFMLALT_zzxw(DisasContext *s, arg_rrxr_esz *a) +{ + return do_BFMLAL_zzxw(s, a, true); +} diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(gvec_bfmlal)(void *vd, void *vn, void *vm, void *va, } clear_tail(d, opr_sz, simd_maxsz(desc)); } + +void HELPER(gvec_bfmlal_idx)(void *vd, void *vn, void *vm, + void *va, void *stat, uint32_t desc) +{ + intptr_t i, j, opr_sz = simd_oprsz(desc); + intptr_t sel = extract32(desc, SIMD_DATA_SHIFT, 1); + intptr_t index = extract32(desc, SIMD_DATA_SHIFT + 1, 3); + intptr_t elements = opr_sz / 4; + intptr_t eltspersegment = MIN(16 / 4, elements); + float32 *d = vd, *a = va; + bfloat16 *n = vn, *m = vm; + + for (i = 0; i < elements; i += eltspersegment) { + float32 m_idx = m[H2(2 * i + index)] << 16; + + for (j = i; j < i + eltspersegment; j++) { + float32 n_j = n[H2(2 * j + sel)] << 16; + d[H4(j)] = float32_muladd(n_j, m_idx, a[H4(j)], 0, stat); + } + } + clear_tail(d, opr_sz, simd_maxsz(desc)); +} -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-12-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- linux-user/elfload.c | 2 ++ 1 file changed, 2 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 @@ static uint32_t get_elf_hwcap2(void) GET_FEATURE_ID(aa64_sve_i8mm, ARM_HWCAP2_A64_SVEI8MM); GET_FEATURE_ID(aa64_sve_f32mm, ARM_HWCAP2_A64_SVEF32MM); GET_FEATURE_ID(aa64_sve_f64mm, ARM_HWCAP2_A64_SVEF64MM); + GET_FEATURE_ID(aa64_sve_bf16, ARM_HWCAP2_A64_SVEBF16); GET_FEATURE_ID(aa64_i8mm, ARM_HWCAP2_A64_I8MM); + GET_FEATURE_ID(aa64_bf16, ARM_HWCAP2_A64_BF16); GET_FEATURE_ID(aa64_rndr, ARM_HWCAP2_A64_RNG); GET_FEATURE_ID(aa64_bti, ARM_HWCAP2_A64_BTI); GET_FEATURE_ID(aa64_mte, ARM_HWCAP2_A64_MTE); -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Disable BF16 again for !have_neon and !have_vfp during realize. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210525225817.400336-13-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.c | 3 +++ target/arm/cpu64.c | 3 +++ target/arm/cpu_tcg.c | 1 + 3 files changed, 7 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) u = cpu->isar.id_isar6; u = FIELD_DP32(u, ID_ISAR6, JSCVT, 0); + u = FIELD_DP32(u, ID_ISAR6, BF16, 0); cpu->isar.id_isar6 = u; u = cpu->isar.mvfr0; @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) t = cpu->isar.id_aa64isar1; t = FIELD_DP64(t, ID_AA64ISAR1, FCMA, 0); + t = FIELD_DP64(t, ID_AA64ISAR1, BF16, 0); t = FIELD_DP64(t, ID_AA64ISAR1, I8MM, 0); cpu->isar.id_aa64isar1 = t; @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) u = cpu->isar.id_isar6; u = FIELD_DP32(u, ID_ISAR6, DP, 0); u = FIELD_DP32(u, ID_ISAR6, FHM, 0); + u = FIELD_DP32(u, ID_ISAR6, BF16, 0); u = FIELD_DP32(u, ID_ISAR6, I8MM, 0); cpu->isar.id_isar6 = u; diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64ISAR1, FCMA, 1); t = FIELD_DP64(t, ID_AA64ISAR1, SB, 1); t = FIELD_DP64(t, ID_AA64ISAR1, SPECRES, 1); + t = FIELD_DP64(t, ID_AA64ISAR1, BF16, 1); t = FIELD_DP64(t, ID_AA64ISAR1, FRINTTS, 1); t = FIELD_DP64(t, ID_AA64ISAR1, LRCPC, 2); /* ARMv8.4-RCPC */ t = FIELD_DP64(t, ID_AA64ISAR1, I8MM, 1); @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64ZFR0, SVEVER, 1); t = FIELD_DP64(t, ID_AA64ZFR0, AES, 2); /* PMULL */ t = FIELD_DP64(t, ID_AA64ZFR0, BITPERM, 1); + t = FIELD_DP64(t, ID_AA64ZFR0, BFLOAT16, 1); t = FIELD_DP64(t, ID_AA64ZFR0, SHA3, 1); t = FIELD_DP64(t, ID_AA64ZFR0, SM4, 1); t = FIELD_DP64(t, ID_AA64ZFR0, I8MM, 1); @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) u = FIELD_DP32(u, ID_ISAR6, FHM, 1); u = FIELD_DP32(u, ID_ISAR6, SB, 1); u = FIELD_DP32(u, ID_ISAR6, SPECRES, 1); + u = FIELD_DP32(u, ID_ISAR6, BF16, 1); u = FIELD_DP32(u, ID_ISAR6, I8MM, 1); cpu->isar.id_isar6 = u; diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ static void arm_max_initfn(Object *obj) t = FIELD_DP32(t, ID_ISAR6, FHM, 1); t = FIELD_DP32(t, ID_ISAR6, SB, 1); t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); + t = FIELD_DP32(t, ID_ISAR6, BF16, 1); t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); cpu->isar.id_isar6 = t; -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> Until now, Hypervisor.framework has only been available on x86_64 systems. With Apple Silicon shipping now, it extends its reach to aarch64. To prepare for support for multiple architectures, let's start moving common code out into its own accel directory. This patch moves assert_hvf_ok() and introduces generic build infrastructure. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-2-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/sysemu/hvf_int.h | 18 +++++++++++++++ accel/hvf/hvf-all.c | 47 ++++++++++++++++++++++++++++++++++++++++ target/i386/hvf/hvf.c | 33 +--------------------------- MAINTAINERS | 8 +++++++ accel/hvf/meson.build | 6 +++++ accel/meson.build | 1 + 6 files changed, 81 insertions(+), 32 deletions(-) create mode 100644 include/sysemu/hvf_int.h create mode 100644 accel/hvf/hvf-all.c create mode 100644 accel/hvf/meson.build diff --git a/include/sysemu/hvf_int.h b/include/sysemu/hvf_int.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/sysemu/hvf_int.h @@ -XXX,XX +XXX,XX @@ +/* + * QEMU Hypervisor.framework (HVF) support + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + * + */ + +/* header to be included in HVF-specific code */ + +#ifndef HVF_INT_H +#define HVF_INT_H + +#include <Hypervisor/hv.h> + +void assert_hvf_ok(hv_return_t ret); + +#endif diff --git a/accel/hvf/hvf-all.c b/accel/hvf/hvf-all.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/accel/hvf/hvf-all.c @@ -XXX,XX +XXX,XX @@ +/* + * QEMU Hypervisor.framework support + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + * Contributions after 2012-01-13 are licensed under the terms of the + * GNU GPL, version 2 or (at your option) any later version. + */ + +#include "qemu/osdep.h" +#include "qemu-common.h" +#include "qemu/error-report.h" +#include "sysemu/hvf.h" +#include "sysemu/hvf_int.h" + +void assert_hvf_ok(hv_return_t ret) +{ + if (ret == HV_SUCCESS) { + return; + } + + switch (ret) { + case HV_ERROR: + error_report("Error: HV_ERROR"); + break; + case HV_BUSY: + error_report("Error: HV_BUSY"); + break; + case HV_BAD_ARGUMENT: + error_report("Error: HV_BAD_ARGUMENT"); + break; + case HV_NO_RESOURCES: + error_report("Error: HV_NO_RESOURCES"); + break; + case HV_NO_DEVICE: + error_report("Error: HV_NO_DEVICE"); + break; + case HV_UNSUPPORTED: + error_report("Error: HV_UNSUPPORTED"); + break; + default: + error_report("Unknown Error"); + } + + abort(); +} diff --git a/target/i386/hvf/hvf.c b/target/i386/hvf/hvf.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/hvf.c +++ b/target/i386/hvf/hvf.c @@ -XXX,XX +XXX,XX @@ #include "qemu/error-report.h" #include "sysemu/hvf.h" +#include "sysemu/hvf_int.h" #include "sysemu/runstate.h" #include "hvf-i386.h" #include "vmcs.h" @@ -XXX,XX +XXX,XX @@ HVFState *hvf_state; -static void assert_hvf_ok(hv_return_t ret) -{ - if (ret == HV_SUCCESS) { - return; - } - - switch (ret) { - case HV_ERROR: - error_report("Error: HV_ERROR"); - break; - case HV_BUSY: - error_report("Error: HV_BUSY"); - break; - case HV_BAD_ARGUMENT: - error_report("Error: HV_BAD_ARGUMENT"); - break; - case HV_NO_RESOURCES: - error_report("Error: HV_NO_RESOURCES"); - break; - case HV_NO_DEVICE: - error_report("Error: HV_NO_DEVICE"); - break; - case HV_UNSUPPORTED: - error_report("Error: HV_UNSUPPORTED"); - break; - default: - error_report("Unknown Error"); - } - - abort(); -} - /* Memory slots */ hvf_slot *hvf_find_overlap_slot(uint64_t start, uint64_t size) { diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ M: Roman Bolshakov <r.bolshakov@yadro.com> W: https://wiki.qemu.org/Features/HVF S: Maintained F: target/i386/hvf/ + +HVF +M: Cameron Esfahani <dirty@apple.com> +M: Roman Bolshakov <r.bolshakov@yadro.com> +W: https://wiki.qemu.org/Features/HVF +S: Maintained +F: accel/hvf/ F: include/sysemu/hvf.h +F: include/sysemu/hvf_int.h WHPX CPUs M: Sunil Muthuswamy <sunilmut@microsoft.com> diff --git a/accel/hvf/meson.build b/accel/hvf/meson.build new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/accel/hvf/meson.build @@ -XXX,XX +XXX,XX @@ +hvf_ss = ss.source_set() +hvf_ss.add(files( + 'hvf-all.c', +)) + +specific_ss.add_all(when: 'CONFIG_HVF', if_true: hvf_ss) diff --git a/accel/meson.build b/accel/meson.build index XXXXXXX..XXXXXXX 100644 --- a/accel/meson.build +++ b/accel/meson.build @@ -XXX,XX +XXX,XX @@ specific_ss.add(files('accel-common.c')) softmmu_ss.add(files('accel-softmmu.c')) user_ss.add(files('accel-user.c')) +subdir('hvf') subdir('qtest') subdir('kvm') subdir('tcg') -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> Until now, Hypervisor.framework has only been available on x86_64 systems. With Apple Silicon shipping now, it extends its reach to aarch64. To prepare for support for multiple architectures, let's start moving common code out into its own accel directory. This patch moves the vCPU thread loop over. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-3-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- {target/i386 => accel}/hvf/hvf-accel-ops.h | 0 {target/i386 => accel}/hvf/hvf-accel-ops.c | 0 target/i386/hvf/x86hvf.c | 2 +- accel/hvf/meson.build | 1 + target/i386/hvf/meson.build | 1 - 5 files changed, 2 insertions(+), 2 deletions(-) rename {target/i386 => accel}/hvf/hvf-accel-ops.h (100%) rename {target/i386 => accel}/hvf/hvf-accel-ops.c (100%) diff --git a/target/i386/hvf/hvf-accel-ops.h b/accel/hvf/hvf-accel-ops.h similarity index 100% rename from target/i386/hvf/hvf-accel-ops.h rename to accel/hvf/hvf-accel-ops.h diff --git a/target/i386/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c similarity index 100% rename from target/i386/hvf/hvf-accel-ops.c rename to accel/hvf/hvf-accel-ops.c diff --git a/target/i386/hvf/x86hvf.c b/target/i386/hvf/x86hvf.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86hvf.c +++ b/target/i386/hvf/x86hvf.c @@ -XXX,XX +XXX,XX @@ #include <Hypervisor/hv.h> #include <Hypervisor/hv_vmx.h> -#include "hvf-accel-ops.h" +#include "accel/hvf/hvf-accel-ops.h" void hvf_set_segment(struct CPUState *cpu, struct vmx_segment *vmx_seg, SegmentCache *qseg, bool is_tr) diff --git a/accel/hvf/meson.build b/accel/hvf/meson.build index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/meson.build +++ b/accel/hvf/meson.build @@ -XXX,XX +XXX,XX @@ hvf_ss = ss.source_set() hvf_ss.add(files( 'hvf-all.c', + 'hvf-accel-ops.c', )) specific_ss.add_all(when: 'CONFIG_HVF', if_true: hvf_ss) diff --git a/target/i386/hvf/meson.build b/target/i386/hvf/meson.build index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/meson.build +++ b/target/i386/hvf/meson.build @@ -XXX,XX +XXX,XX @@ i386_softmmu_ss.add(when: [hvf, 'CONFIG_HVF'], if_true: files( 'hvf.c', - 'hvf-accel-ops.c', 'x86.c', 'x86_cpuid.c', 'x86_decode.c', -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> Until now, Hypervisor.framework has only been available on x86_64 systems. With Apple Silicon shipping now, it extends its reach to aarch64. To prepare for support for multiple architectures, let's start moving common code out into its own accel directory. This patch moves CPU and memory operations over. While at it, make sure the code is consumable on non-i386 systems. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-4-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/sysemu/hvf_int.h | 4 + target/i386/hvf/hvf-i386.h | 2 - target/i386/hvf/x86hvf.h | 2 - accel/hvf/hvf-accel-ops.c | 308 ++++++++++++++++++++++++++++++++++++- target/i386/hvf/hvf.c | 302 ------------------------------------ 5 files changed, 311 insertions(+), 307 deletions(-) diff --git a/include/sysemu/hvf_int.h b/include/sysemu/hvf_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/hvf_int.h +++ b/include/sysemu/hvf_int.h @@ -XXX,XX +XXX,XX @@ #include <Hypervisor/hv.h> +void hvf_set_phys_mem(MemoryRegionSection *, bool); void assert_hvf_ok(hv_return_t ret); +hvf_slot *hvf_find_overlap_slot(uint64_t, uint64_t); +int hvf_put_registers(CPUState *); +int hvf_get_registers(CPUState *); #endif diff --git a/target/i386/hvf/hvf-i386.h b/target/i386/hvf/hvf-i386.h index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/hvf-i386.h +++ b/target/i386/hvf/hvf-i386.h @@ -XXX,XX +XXX,XX @@ struct HVFState { }; extern HVFState *hvf_state; -void hvf_set_phys_mem(MemoryRegionSection *, bool); void hvf_handle_io(CPUArchState *, uint16_t, void *, int, int, int); -hvf_slot *hvf_find_overlap_slot(uint64_t, uint64_t); #ifdef NEED_CPU_H /* Functions exported to host specific mode */ diff --git a/target/i386/hvf/x86hvf.h b/target/i386/hvf/x86hvf.h index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86hvf.h +++ b/target/i386/hvf/x86hvf.h @@ -XXX,XX +XXX,XX @@ #include "x86_descr.h" int hvf_process_events(CPUState *); -int hvf_put_registers(CPUState *); -int hvf_get_registers(CPUState *); bool hvf_inject_interrupts(CPUState *); void hvf_set_segment(struct CPUState *cpu, struct vmx_segment *vmx_seg, SegmentCache *qseg, bool is_tr); diff --git a/accel/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.c +++ b/accel/hvf/hvf-accel-ops.c @@ -XXX,XX +XXX,XX @@ #include "qemu/osdep.h" #include "qemu/error-report.h" #include "qemu/main-loop.h" +#include "exec/address-spaces.h" +#include "exec/exec-all.h" +#include "sysemu/cpus.h" #include "sysemu/hvf.h" +#include "sysemu/hvf_int.h" #include "sysemu/runstate.h" -#include "target/i386/cpu.h" #include "qemu/guest-random.h" #include "hvf-accel-ops.h" +HVFState *hvf_state; + +/* Memory slots */ + +hvf_slot *hvf_find_overlap_slot(uint64_t start, uint64_t size) +{ + hvf_slot *slot; + int x; + for (x = 0; x < hvf_state->num_slots; ++x) { + slot = &hvf_state->slots[x]; + if (slot->size && start < (slot->start + slot->size) && + (start + size) > slot->start) { + return slot; + } + } + return NULL; +} + +struct mac_slot { + int present; + uint64_t size; + uint64_t gpa_start; + uint64_t gva; +}; + +struct mac_slot mac_slots[32]; + +static int do_hvf_set_memory(hvf_slot *slot, hv_memory_flags_t flags) +{ + struct mac_slot *macslot; + hv_return_t ret; + + macslot = &mac_slots[slot->slot_id]; + + if (macslot->present) { + if (macslot->size != slot->size) { + macslot->present = 0; + ret = hv_vm_unmap(macslot->gpa_start, macslot->size); + assert_hvf_ok(ret); + } + } + + if (!slot->size) { + return 0; + } + + macslot->present = 1; + macslot->gpa_start = slot->start; + macslot->size = slot->size; + ret = hv_vm_map((hv_uvaddr_t)slot->mem, slot->start, slot->size, flags); + assert_hvf_ok(ret); + return 0; +} + +void hvf_set_phys_mem(MemoryRegionSection *section, bool add) +{ + hvf_slot *mem; + MemoryRegion *area = section->mr; + bool writeable = !area->readonly && !area->rom_device; + hv_memory_flags_t flags; + + if (!memory_region_is_ram(area)) { + if (writeable) { + return; + } else if (!memory_region_is_romd(area)) { + /* + * If the memory device is not in romd_mode, then we actually want + * to remove the hvf memory slot so all accesses will trap. + */ + add = false; + } + } + + mem = hvf_find_overlap_slot( + section->offset_within_address_space, + int128_get64(section->size)); + + if (mem && add) { + if (mem->size == int128_get64(section->size) && + mem->start == section->offset_within_address_space && + mem->mem == (memory_region_get_ram_ptr(area) + + section->offset_within_region)) { + return; /* Same region was attempted to register, go away. */ + } + } + + /* Region needs to be reset. set the size to 0 and remap it. */ + if (mem) { + mem->size = 0; + if (do_hvf_set_memory(mem, 0)) { + error_report("Failed to reset overlapping slot"); + abort(); + } + } + + if (!add) { + return; + } + + if (area->readonly || + (!memory_region_is_ram(area) && memory_region_is_romd(area))) { + flags = HV_MEMORY_READ | HV_MEMORY_EXEC; + } else { + flags = HV_MEMORY_READ | HV_MEMORY_WRITE | HV_MEMORY_EXEC; + } + + /* Now make a new slot. */ + int x; + + for (x = 0; x < hvf_state->num_slots; ++x) { + mem = &hvf_state->slots[x]; + if (!mem->size) { + break; + } + } + + if (x == hvf_state->num_slots) { + error_report("No free slots"); + abort(); + } + + mem->size = int128_get64(section->size); + mem->mem = memory_region_get_ram_ptr(area) + section->offset_within_region; + mem->start = section->offset_within_address_space; + mem->region = area; + + if (do_hvf_set_memory(mem, flags)) { + error_report("Error registering new memory slot"); + abort(); + } +} + +static void do_hvf_cpu_synchronize_state(CPUState *cpu, run_on_cpu_data arg) +{ + if (!cpu->vcpu_dirty) { + hvf_get_registers(cpu); + cpu->vcpu_dirty = true; + } +} + +void hvf_cpu_synchronize_state(CPUState *cpu) +{ + if (!cpu->vcpu_dirty) { + run_on_cpu(cpu, do_hvf_cpu_synchronize_state, RUN_ON_CPU_NULL); + } +} + +static void do_hvf_cpu_synchronize_post_reset(CPUState *cpu, + run_on_cpu_data arg) +{ + hvf_put_registers(cpu); + cpu->vcpu_dirty = false; +} + +void hvf_cpu_synchronize_post_reset(CPUState *cpu) +{ + run_on_cpu(cpu, do_hvf_cpu_synchronize_post_reset, RUN_ON_CPU_NULL); +} + +static void do_hvf_cpu_synchronize_post_init(CPUState *cpu, + run_on_cpu_data arg) +{ + hvf_put_registers(cpu); + cpu->vcpu_dirty = false; +} + +void hvf_cpu_synchronize_post_init(CPUState *cpu) +{ + run_on_cpu(cpu, do_hvf_cpu_synchronize_post_init, RUN_ON_CPU_NULL); +} + +static void do_hvf_cpu_synchronize_pre_loadvm(CPUState *cpu, + run_on_cpu_data arg) +{ + cpu->vcpu_dirty = true; +} + +void hvf_cpu_synchronize_pre_loadvm(CPUState *cpu) +{ + run_on_cpu(cpu, do_hvf_cpu_synchronize_pre_loadvm, RUN_ON_CPU_NULL); +} + +static void hvf_set_dirty_tracking(MemoryRegionSection *section, bool on) +{ + hvf_slot *slot; + + slot = hvf_find_overlap_slot( + section->offset_within_address_space, + int128_get64(section->size)); + + /* protect region against writes; begin tracking it */ + if (on) { + slot->flags |= HVF_SLOT_LOG; + hv_vm_protect((hv_gpaddr_t)slot->start, (size_t)slot->size, + HV_MEMORY_READ); + /* stop tracking region*/ + } else { + slot->flags &= ~HVF_SLOT_LOG; + hv_vm_protect((hv_gpaddr_t)slot->start, (size_t)slot->size, + HV_MEMORY_READ | HV_MEMORY_WRITE); + } +} + +static void hvf_log_start(MemoryListener *listener, + MemoryRegionSection *section, int old, int new) +{ + if (old != 0) { + return; + } + + hvf_set_dirty_tracking(section, 1); +} + +static void hvf_log_stop(MemoryListener *listener, + MemoryRegionSection *section, int old, int new) +{ + if (new != 0) { + return; + } + + hvf_set_dirty_tracking(section, 0); +} + +static void hvf_log_sync(MemoryListener *listener, + MemoryRegionSection *section) +{ + /* + * sync of dirty pages is handled elsewhere; just make sure we keep + * tracking the region. + */ + hvf_set_dirty_tracking(section, 1); +} + +static void hvf_region_add(MemoryListener *listener, + MemoryRegionSection *section) +{ + hvf_set_phys_mem(section, true); +} + +static void hvf_region_del(MemoryListener *listener, + MemoryRegionSection *section) +{ + hvf_set_phys_mem(section, false); +} + +static MemoryListener hvf_memory_listener = { + .priority = 10, + .region_add = hvf_region_add, + .region_del = hvf_region_del, + .log_start = hvf_log_start, + .log_stop = hvf_log_stop, + .log_sync = hvf_log_sync, +}; + +static void dummy_signal(int sig) +{ +} + +bool hvf_allowed; + +static int hvf_accel_init(MachineState *ms) +{ + int x; + hv_return_t ret; + HVFState *s; + + ret = hv_vm_create(HV_VM_DEFAULT); + assert_hvf_ok(ret); + + s = g_new0(HVFState, 1); + + s->num_slots = 32; + for (x = 0; x < s->num_slots; ++x) { + s->slots[x].size = 0; + s->slots[x].slot_id = x; + } + + hvf_state = s; + memory_listener_register(&hvf_memory_listener, &address_space_memory); + return 0; +} + +static void hvf_accel_class_init(ObjectClass *oc, void *data) +{ + AccelClass *ac = ACCEL_CLASS(oc); + ac->name = "HVF"; + ac->init_machine = hvf_accel_init; + ac->allowed = &hvf_allowed; +} + +static const TypeInfo hvf_accel_type = { + .name = TYPE_HVF_ACCEL, + .parent = TYPE_ACCEL, + .class_init = hvf_accel_class_init, +}; + +static void hvf_type_init(void) +{ + type_register_static(&hvf_accel_type); +} + +type_init(hvf_type_init); + /* * The HVF-specific vCPU thread function. This one should only run when the host * CPU supports the VMX "unrestricted guest" feature. diff --git a/target/i386/hvf/hvf.c b/target/i386/hvf/hvf.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/hvf.c +++ b/target/i386/hvf/hvf.c @@ -XXX,XX +XXX,XX @@ #include "hvf-accel-ops.h" -HVFState *hvf_state; - -/* Memory slots */ -hvf_slot *hvf_find_overlap_slot(uint64_t start, uint64_t size) -{ - hvf_slot *slot; - int x; - for (x = 0; x < hvf_state->num_slots; ++x) { - slot = &hvf_state->slots[x]; - if (slot->size && start < (slot->start + slot->size) && - (start + size) > slot->start) { - return slot; - } - } - return NULL; -} - -struct mac_slot { - int present; - uint64_t size; - uint64_t gpa_start; - uint64_t gva; -}; - -struct mac_slot mac_slots[32]; - -static int do_hvf_set_memory(hvf_slot *slot, hv_memory_flags_t flags) -{ - struct mac_slot *macslot; - hv_return_t ret; - - macslot = &mac_slots[slot->slot_id]; - - if (macslot->present) { - if (macslot->size != slot->size) { - macslot->present = 0; - ret = hv_vm_unmap(macslot->gpa_start, macslot->size); - assert_hvf_ok(ret); - } - } - - if (!slot->size) { - return 0; - } - - macslot->present = 1; - macslot->gpa_start = slot->start; - macslot->size = slot->size; - ret = hv_vm_map((hv_uvaddr_t)slot->mem, slot->start, slot->size, flags); - assert_hvf_ok(ret); - return 0; -} - -void hvf_set_phys_mem(MemoryRegionSection *section, bool add) -{ - hvf_slot *mem; - MemoryRegion *area = section->mr; - bool writeable = !area->readonly && !area->rom_device; - hv_memory_flags_t flags; - - if (!memory_region_is_ram(area)) { - if (writeable) { - return; - } else if (!memory_region_is_romd(area)) { - /* - * If the memory device is not in romd_mode, then we actually want - * to remove the hvf memory slot so all accesses will trap. - */ - add = false; - } - } - - mem = hvf_find_overlap_slot( - section->offset_within_address_space, - int128_get64(section->size)); - - if (mem && add) { - if (mem->size == int128_get64(section->size) && - mem->start == section->offset_within_address_space && - mem->mem == (memory_region_get_ram_ptr(area) + - section->offset_within_region)) { - return; /* Same region was attempted to register, go away. */ - } - } - - /* Region needs to be reset. set the size to 0 and remap it. */ - if (mem) { - mem->size = 0; - if (do_hvf_set_memory(mem, 0)) { - error_report("Failed to reset overlapping slot"); - abort(); - } - } - - if (!add) { - return; - } - - if (area->readonly || - (!memory_region_is_ram(area) && memory_region_is_romd(area))) { - flags = HV_MEMORY_READ | HV_MEMORY_EXEC; - } else { - flags = HV_MEMORY_READ | HV_MEMORY_WRITE | HV_MEMORY_EXEC; - } - - /* Now make a new slot. */ - int x; - - for (x = 0; x < hvf_state->num_slots; ++x) { - mem = &hvf_state->slots[x]; - if (!mem->size) { - break; - } - } - - if (x == hvf_state->num_slots) { - error_report("No free slots"); - abort(); - } - - mem->size = int128_get64(section->size); - mem->mem = memory_region_get_ram_ptr(area) + section->offset_within_region; - mem->start = section->offset_within_address_space; - mem->region = area; - - if (do_hvf_set_memory(mem, flags)) { - error_report("Error registering new memory slot"); - abort(); - } -} - void vmx_update_tpr(CPUState *cpu) { /* TODO: need integrate APIC handling */ @@ -XXX,XX +XXX,XX @@ void hvf_handle_io(CPUArchState *env, uint16_t port, void *buffer, } } -static void do_hvf_cpu_synchronize_state(CPUState *cpu, run_on_cpu_data arg) -{ - if (!cpu->vcpu_dirty) { - hvf_get_registers(cpu); - cpu->vcpu_dirty = true; - } -} - -void hvf_cpu_synchronize_state(CPUState *cpu) -{ - if (!cpu->vcpu_dirty) { - run_on_cpu(cpu, do_hvf_cpu_synchronize_state, RUN_ON_CPU_NULL); - } -} - -static void do_hvf_cpu_synchronize_post_reset(CPUState *cpu, - run_on_cpu_data arg) -{ - hvf_put_registers(cpu); - cpu->vcpu_dirty = false; -} - -void hvf_cpu_synchronize_post_reset(CPUState *cpu) -{ - run_on_cpu(cpu, do_hvf_cpu_synchronize_post_reset, RUN_ON_CPU_NULL); -} - -static void do_hvf_cpu_synchronize_post_init(CPUState *cpu, - run_on_cpu_data arg) -{ - hvf_put_registers(cpu); - cpu->vcpu_dirty = false; -} - -void hvf_cpu_synchronize_post_init(CPUState *cpu) -{ - run_on_cpu(cpu, do_hvf_cpu_synchronize_post_init, RUN_ON_CPU_NULL); -} - -static void do_hvf_cpu_synchronize_pre_loadvm(CPUState *cpu, - run_on_cpu_data arg) -{ - cpu->vcpu_dirty = true; -} - -void hvf_cpu_synchronize_pre_loadvm(CPUState *cpu) -{ - run_on_cpu(cpu, do_hvf_cpu_synchronize_pre_loadvm, RUN_ON_CPU_NULL); -} - static bool ept_emulation_fault(hvf_slot *slot, uint64_t gpa, uint64_t ept_qual) { int read, write; @@ -XXX,XX +XXX,XX @@ static bool ept_emulation_fault(hvf_slot *slot, uint64_t gpa, uint64_t ept_qual) return false; } -static void hvf_set_dirty_tracking(MemoryRegionSection *section, bool on) -{ - hvf_slot *slot; - - slot = hvf_find_overlap_slot( - section->offset_within_address_space, - int128_get64(section->size)); - - /* protect region against writes; begin tracking it */ - if (on) { - slot->flags |= HVF_SLOT_LOG; - hv_vm_protect((hv_gpaddr_t)slot->start, (size_t)slot->size, - HV_MEMORY_READ); - /* stop tracking region*/ - } else { - slot->flags &= ~HVF_SLOT_LOG; - hv_vm_protect((hv_gpaddr_t)slot->start, (size_t)slot->size, - HV_MEMORY_READ | HV_MEMORY_WRITE); - } -} - -static void hvf_log_start(MemoryListener *listener, - MemoryRegionSection *section, int old, int new) -{ - if (old != 0) { - return; - } - - hvf_set_dirty_tracking(section, 1); -} - -static void hvf_log_stop(MemoryListener *listener, - MemoryRegionSection *section, int old, int new) -{ - if (new != 0) { - return; - } - - hvf_set_dirty_tracking(section, 0); -} - -static void hvf_log_sync(MemoryListener *listener, - MemoryRegionSection *section) -{ - /* - * sync of dirty pages is handled elsewhere; just make sure we keep - * tracking the region. - */ - hvf_set_dirty_tracking(section, 1); -} - -static void hvf_region_add(MemoryListener *listener, - MemoryRegionSection *section) -{ - hvf_set_phys_mem(section, true); -} - -static void hvf_region_del(MemoryListener *listener, - MemoryRegionSection *section) -{ - hvf_set_phys_mem(section, false); -} - -static MemoryListener hvf_memory_listener = { - .priority = 10, - .region_add = hvf_region_add, - .region_del = hvf_region_del, - .log_start = hvf_log_start, - .log_stop = hvf_log_stop, - .log_sync = hvf_log_sync, -}; - void hvf_vcpu_destroy(CPUState *cpu) { X86CPU *x86_cpu = X86_CPU(cpu); @@ -XXX,XX +XXX,XX @@ void hvf_vcpu_destroy(CPUState *cpu) assert_hvf_ok(ret); } -static void dummy_signal(int sig) -{ -} - static void init_tsc_freq(CPUX86State *env) { size_t length; @@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu) return ret; } - -bool hvf_allowed; - -static int hvf_accel_init(MachineState *ms) -{ - int x; - hv_return_t ret; - HVFState *s; - - ret = hv_vm_create(HV_VM_DEFAULT); - assert_hvf_ok(ret); - - s = g_new0(HVFState, 1); - - s->num_slots = 32; - for (x = 0; x < s->num_slots; ++x) { - s->slots[x].size = 0; - s->slots[x].slot_id = x; - } - - hvf_state = s; - memory_listener_register(&hvf_memory_listener, &address_space_memory); - return 0; -} - -static void hvf_accel_class_init(ObjectClass *oc, void *data) -{ - AccelClass *ac = ACCEL_CLASS(oc); - ac->name = "HVF"; - ac->init_machine = hvf_accel_init; - ac->allowed = &hvf_allowed; -} - -static const TypeInfo hvf_accel_type = { - .name = TYPE_HVF_ACCEL, - .parent = TYPE_ACCEL, - .class_init = hvf_accel_class_init, -}; - -static void hvf_type_init(void) -{ - type_register_static(&hvf_accel_type); -} - -type_init(hvf_type_init); -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> Until now, Hypervisor.framework has only been available on x86_64 systems. With Apple Silicon shipping now, it extends its reach to aarch64. To prepare for support for multiple architectures, let's start moving common code out into its own accel directory. This patch moves a few internal struct and constant defines over. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-5-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/sysemu/hvf_int.h | 30 ++++++++++++++++++++++++++++++ target/i386/hvf/hvf-i386.h | 31 +------------------------------ 2 files changed, 31 insertions(+), 30 deletions(-) diff --git a/include/sysemu/hvf_int.h b/include/sysemu/hvf_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/hvf_int.h +++ b/include/sysemu/hvf_int.h @@ -XXX,XX +XXX,XX @@ #include <Hypervisor/hv.h> +/* hvf_slot flags */ +#define HVF_SLOT_LOG (1 << 0) + +typedef struct hvf_slot { + uint64_t start; + uint64_t size; + uint8_t *mem; + int slot_id; + uint32_t flags; + MemoryRegion *region; +} hvf_slot; + +typedef struct hvf_vcpu_caps { + uint64_t vmx_cap_pinbased; + uint64_t vmx_cap_procbased; + uint64_t vmx_cap_procbased2; + uint64_t vmx_cap_entry; + uint64_t vmx_cap_exit; + uint64_t vmx_cap_preemption_timer; +} hvf_vcpu_caps; + +struct HVFState { + AccelState parent; + hvf_slot slots[32]; + int num_slots; + + hvf_vcpu_caps *hvf_caps; +}; +extern HVFState *hvf_state; + void hvf_set_phys_mem(MemoryRegionSection *, bool); void assert_hvf_ok(hv_return_t ret); hvf_slot *hvf_find_overlap_slot(uint64_t, uint64_t); diff --git a/target/i386/hvf/hvf-i386.h b/target/i386/hvf/hvf-i386.h index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/hvf-i386.h +++ b/target/i386/hvf/hvf-i386.h @@ -XXX,XX +XXX,XX @@ #include "qemu/accel.h" #include "sysemu/hvf.h" +#include "sysemu/hvf_int.h" #include "cpu.h" #include "x86.h" -/* hvf_slot flags */ -#define HVF_SLOT_LOG (1 << 0) - -typedef struct hvf_slot { - uint64_t start; - uint64_t size; - uint8_t *mem; - int slot_id; - uint32_t flags; - MemoryRegion *region; -} hvf_slot; - -typedef struct hvf_vcpu_caps { - uint64_t vmx_cap_pinbased; - uint64_t vmx_cap_procbased; - uint64_t vmx_cap_procbased2; - uint64_t vmx_cap_entry; - uint64_t vmx_cap_exit; - uint64_t vmx_cap_preemption_timer; -} hvf_vcpu_caps; - -struct HVFState { - AccelState parent; - hvf_slot slots[32]; - int num_slots; - - hvf_vcpu_caps *hvf_caps; -}; -extern HVFState *hvf_state; - void hvf_handle_io(CPUArchState *, uint16_t, void *, int, int, int); #ifdef NEED_CPU_H -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> The hvf_set_phys_mem() function is only called within the same file. Make it static. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-6-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/sysemu/hvf_int.h | 1 - accel/hvf/hvf-accel-ops.c | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/include/sysemu/hvf_int.h b/include/sysemu/hvf_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/hvf_int.h +++ b/include/sysemu/hvf_int.h @@ -XXX,XX +XXX,XX @@ struct HVFState { }; extern HVFState *hvf_state; -void hvf_set_phys_mem(MemoryRegionSection *, bool); void assert_hvf_ok(hv_return_t ret); hvf_slot *hvf_find_overlap_slot(uint64_t, uint64_t); int hvf_put_registers(CPUState *); diff --git a/accel/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.c +++ b/accel/hvf/hvf-accel-ops.c @@ -XXX,XX +XXX,XX @@ static int do_hvf_set_memory(hvf_slot *slot, hv_memory_flags_t flags) return 0; } -void hvf_set_phys_mem(MemoryRegionSection *section, bool add) +static void hvf_set_phys_mem(MemoryRegionSection *section, bool add) { hvf_slot *mem; MemoryRegion *area = section->mr; -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> The ARM version of Hypervisor.framework no longer defines these two types, so let's just revert to standard ones. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-7-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- accel/hvf/hvf-accel-ops.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/accel/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.c +++ b/accel/hvf/hvf-accel-ops.c @@ -XXX,XX +XXX,XX @@ static int do_hvf_set_memory(hvf_slot *slot, hv_memory_flags_t flags) macslot->present = 1; macslot->gpa_start = slot->start; macslot->size = slot->size; - ret = hv_vm_map((hv_uvaddr_t)slot->mem, slot->start, slot->size, flags); + ret = hv_vm_map(slot->mem, slot->start, slot->size, flags); assert_hvf_ok(ret); return 0; } @@ -XXX,XX +XXX,XX @@ static void hvf_set_dirty_tracking(MemoryRegionSection *section, bool on) /* protect region against writes; begin tracking it */ if (on) { slot->flags |= HVF_SLOT_LOG; - hv_vm_protect((hv_gpaddr_t)slot->start, (size_t)slot->size, + hv_vm_protect((uintptr_t)slot->start, (size_t)slot->size, HV_MEMORY_READ); /* stop tracking region*/ } else { slot->flags &= ~HVF_SLOT_LOG; - hv_vm_protect((hv_gpaddr_t)slot->start, (size_t)slot->size, + hv_vm_protect((uintptr_t)slot->start, (size_t)slot->size, HV_MEMORY_READ | HV_MEMORY_WRITE); } } -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> Until now, Hypervisor.framework has only been available on x86_64 systems. With Apple Silicon shipping now, it extends its reach to aarch64. To prepare for support for multiple architectures, let's start moving common code out into its own accel directory. This patch splits the vcpu init and destroy functions into a generic and an architecture specific portion. This also allows us to move the generic functions into the generic hvf code, removing exported functions. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-8-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- accel/hvf/hvf-accel-ops.h | 2 -- include/sysemu/hvf_int.h | 2 ++ accel/hvf/hvf-accel-ops.c | 30 ++++++++++++++++++++++++++++++ target/i386/hvf/hvf.c | 23 ++--------------------- 4 files changed, 34 insertions(+), 23 deletions(-) diff --git a/accel/hvf/hvf-accel-ops.h b/accel/hvf/hvf-accel-ops.h index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.h +++ b/accel/hvf/hvf-accel-ops.h @@ -XXX,XX +XXX,XX @@ #include "sysemu/cpus.h" -int hvf_init_vcpu(CPUState *); int hvf_vcpu_exec(CPUState *); void hvf_cpu_synchronize_state(CPUState *); void hvf_cpu_synchronize_post_reset(CPUState *); void hvf_cpu_synchronize_post_init(CPUState *); void hvf_cpu_synchronize_pre_loadvm(CPUState *); -void hvf_vcpu_destroy(CPUState *); #endif /* HVF_CPUS_H */ diff --git a/include/sysemu/hvf_int.h b/include/sysemu/hvf_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/hvf_int.h +++ b/include/sysemu/hvf_int.h @@ -XXX,XX +XXX,XX @@ struct HVFState { extern HVFState *hvf_state; void assert_hvf_ok(hv_return_t ret); +int hvf_arch_init_vcpu(CPUState *cpu); +void hvf_arch_vcpu_destroy(CPUState *cpu); hvf_slot *hvf_find_overlap_slot(uint64_t, uint64_t); int hvf_put_registers(CPUState *); int hvf_get_registers(CPUState *); diff --git a/accel/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.c +++ b/accel/hvf/hvf-accel-ops.c @@ -XXX,XX +XXX,XX @@ static void hvf_type_init(void) type_init(hvf_type_init); +static void hvf_vcpu_destroy(CPUState *cpu) +{ + hv_return_t ret = hv_vcpu_destroy(cpu->hvf_fd); + assert_hvf_ok(ret); + + hvf_arch_vcpu_destroy(cpu); +} + +static int hvf_init_vcpu(CPUState *cpu) +{ + int r; + + /* init cpu signals */ + sigset_t set; + struct sigaction sigact; + + memset(&sigact, 0, sizeof(sigact)); + sigact.sa_handler = dummy_signal; + sigaction(SIG_IPI, &sigact, NULL); + + pthread_sigmask(SIG_BLOCK, NULL, &set); + sigdelset(&set, SIG_IPI); + + r = hv_vcpu_create((hv_vcpuid_t *)&cpu->hvf_fd, HV_VCPU_DEFAULT); + cpu->vcpu_dirty = 1; + assert_hvf_ok(r); + + return hvf_arch_init_vcpu(cpu); +} + /* * The HVF-specific vCPU thread function. This one should only run when the host * CPU supports the VMX "unrestricted guest" feature. diff --git a/target/i386/hvf/hvf.c b/target/i386/hvf/hvf.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/hvf.c +++ b/target/i386/hvf/hvf.c @@ -XXX,XX +XXX,XX @@ static bool ept_emulation_fault(hvf_slot *slot, uint64_t gpa, uint64_t ept_qual) return false; } -void hvf_vcpu_destroy(CPUState *cpu) +void hvf_arch_vcpu_destroy(CPUState *cpu) { X86CPU *x86_cpu = X86_CPU(cpu); CPUX86State *env = &x86_cpu->env; - hv_return_t ret = hv_vcpu_destroy((hv_vcpuid_t)cpu->hvf_fd); g_free(env->hvf_mmio_buf); - assert_hvf_ok(ret); } static void init_tsc_freq(CPUX86State *env) @@ -XXX,XX +XXX,XX @@ static inline bool apic_bus_freq_is_known(CPUX86State *env) return env->apic_bus_freq != 0; } -int hvf_init_vcpu(CPUState *cpu) +int hvf_arch_init_vcpu(CPUState *cpu) { - X86CPU *x86cpu = X86_CPU(cpu); CPUX86State *env = &x86cpu->env; - int r; - - /* init cpu signals */ - sigset_t set; - struct sigaction sigact; - - memset(&sigact, 0, sizeof(sigact)); - sigact.sa_handler = dummy_signal; - sigaction(SIG_IPI, &sigact, NULL); - - pthread_sigmask(SIG_BLOCK, NULL, &set); - sigdelset(&set, SIG_IPI); init_emu(); init_decoder(); @@ -XXX,XX +XXX,XX @@ int hvf_init_vcpu(CPUState *cpu) } } - r = hv_vcpu_create((hv_vcpuid_t *)&cpu->hvf_fd, HV_VCPU_DEFAULT); - cpu->vcpu_dirty = 1; - assert_hvf_ok(r); - if (hv_vmx_read_capability(HV_VMX_CAP_PINBASED, &hvf_state->hvf_caps->vmx_cap_pinbased)) { abort(); -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> There is no reason to call the hvf specific hvf_cpu_synchronize_state() when we can just use the generic cpu_synchronize_state() instead. This allows us to have less dependency on internal function definitions and allows us to make hvf_cpu_synchronize_state() static. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-9-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- accel/hvf/hvf-accel-ops.h | 1 - accel/hvf/hvf-accel-ops.c | 2 +- target/i386/hvf/x86hvf.c | 9 ++++----- 3 files changed, 5 insertions(+), 7 deletions(-) diff --git a/accel/hvf/hvf-accel-ops.h b/accel/hvf/hvf-accel-ops.h index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.h +++ b/accel/hvf/hvf-accel-ops.h @@ -XXX,XX +XXX,XX @@ #include "sysemu/cpus.h" int hvf_vcpu_exec(CPUState *); -void hvf_cpu_synchronize_state(CPUState *); void hvf_cpu_synchronize_post_reset(CPUState *); void hvf_cpu_synchronize_post_init(CPUState *); void hvf_cpu_synchronize_pre_loadvm(CPUState *); diff --git a/accel/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.c +++ b/accel/hvf/hvf-accel-ops.c @@ -XXX,XX +XXX,XX @@ static void do_hvf_cpu_synchronize_state(CPUState *cpu, run_on_cpu_data arg) } } -void hvf_cpu_synchronize_state(CPUState *cpu) +static void hvf_cpu_synchronize_state(CPUState *cpu) { if (!cpu->vcpu_dirty) { run_on_cpu(cpu, do_hvf_cpu_synchronize_state, RUN_ON_CPU_NULL); diff --git a/target/i386/hvf/x86hvf.c b/target/i386/hvf/x86hvf.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86hvf.c +++ b/target/i386/hvf/x86hvf.c @@ -XXX,XX +XXX,XX @@ #include "cpu.h" #include "x86_descr.h" #include "x86_decode.h" +#include "sysemu/hw_accel.h" #include "hw/i386/apic_internal.h" #include <Hypervisor/hv.h> #include <Hypervisor/hv_vmx.h> -#include "accel/hvf/hvf-accel-ops.h" - void hvf_set_segment(struct CPUState *cpu, struct vmx_segment *vmx_seg, SegmentCache *qseg, bool is_tr) { @@ -XXX,XX +XXX,XX @@ int hvf_process_events(CPUState *cpu_state) env->eflags = rreg(cpu_state->hvf_fd, HV_X86_RFLAGS); if (cpu_state->interrupt_request & CPU_INTERRUPT_INIT) { - hvf_cpu_synchronize_state(cpu_state); + cpu_synchronize_state(cpu_state); do_cpu_init(cpu); } @@ -XXX,XX +XXX,XX @@ int hvf_process_events(CPUState *cpu_state) cpu_state->halted = 0; } if (cpu_state->interrupt_request & CPU_INTERRUPT_SIPI) { - hvf_cpu_synchronize_state(cpu_state); + cpu_synchronize_state(cpu_state); do_cpu_sipi(cpu); } if (cpu_state->interrupt_request & CPU_INTERRUPT_TPR) { cpu_state->interrupt_request &= ~CPU_INTERRUPT_TPR; - hvf_cpu_synchronize_state(cpu_state); + cpu_synchronize_state(cpu_state); apic_handle_tpr_access_report(cpu->apic_state, env->eip, env->tpr_access_type); } -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> The hvf accel synchronize functions are only used as input for local callback functions, so we can make them static. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-10-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- accel/hvf/hvf-accel-ops.h | 3 --- accel/hvf/hvf-accel-ops.c | 6 +++--- 2 files changed, 3 insertions(+), 6 deletions(-) diff --git a/accel/hvf/hvf-accel-ops.h b/accel/hvf/hvf-accel-ops.h index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.h +++ b/accel/hvf/hvf-accel-ops.h @@ -XXX,XX +XXX,XX @@ #include "sysemu/cpus.h" int hvf_vcpu_exec(CPUState *); -void hvf_cpu_synchronize_post_reset(CPUState *); -void hvf_cpu_synchronize_post_init(CPUState *); -void hvf_cpu_synchronize_pre_loadvm(CPUState *); #endif /* HVF_CPUS_H */ diff --git a/accel/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.c +++ b/accel/hvf/hvf-accel-ops.c @@ -XXX,XX +XXX,XX @@ static void do_hvf_cpu_synchronize_post_reset(CPUState *cpu, cpu->vcpu_dirty = false; } -void hvf_cpu_synchronize_post_reset(CPUState *cpu) +static void hvf_cpu_synchronize_post_reset(CPUState *cpu) { run_on_cpu(cpu, do_hvf_cpu_synchronize_post_reset, RUN_ON_CPU_NULL); } @@ -XXX,XX +XXX,XX @@ static void do_hvf_cpu_synchronize_post_init(CPUState *cpu, cpu->vcpu_dirty = false; } -void hvf_cpu_synchronize_post_init(CPUState *cpu) +static void hvf_cpu_synchronize_post_init(CPUState *cpu) { run_on_cpu(cpu, do_hvf_cpu_synchronize_post_init, RUN_ON_CPU_NULL); } @@ -XXX,XX +XXX,XX @@ static void do_hvf_cpu_synchronize_pre_loadvm(CPUState *cpu, cpu->vcpu_dirty = true; } -void hvf_cpu_synchronize_pre_loadvm(CPUState *cpu) +static void hvf_cpu_synchronize_pre_loadvm(CPUState *cpu) { run_on_cpu(cpu, do_hvf_cpu_synchronize_pre_loadvm, RUN_ON_CPU_NULL); } -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> We can move the definition of hvf_vcpu_exec() into our internal hvf header, obsoleting the need for hvf-accel-ops.h. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-11-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- accel/hvf/hvf-accel-ops.h | 17 ----------------- include/sysemu/hvf_int.h | 1 + accel/hvf/hvf-accel-ops.c | 2 -- target/i386/hvf/hvf.c | 2 -- 4 files changed, 1 insertion(+), 21 deletions(-) delete mode 100644 accel/hvf/hvf-accel-ops.h diff --git a/accel/hvf/hvf-accel-ops.h b/accel/hvf/hvf-accel-ops.h deleted file mode 100644 index XXXXXXX..XXXXXXX --- a/accel/hvf/hvf-accel-ops.h +++ /dev/null @@ -XXX,XX +XXX,XX @@ -/* - * Accelerator CPUS Interface - * - * Copyright 2020 SUSE LLC - * - * 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 HVF_CPUS_H -#define HVF_CPUS_H - -#include "sysemu/cpus.h" - -int hvf_vcpu_exec(CPUState *); - -#endif /* HVF_CPUS_H */ diff --git a/include/sysemu/hvf_int.h b/include/sysemu/hvf_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/hvf_int.h +++ b/include/sysemu/hvf_int.h @@ -XXX,XX +XXX,XX @@ extern HVFState *hvf_state; void assert_hvf_ok(hv_return_t ret); int hvf_arch_init_vcpu(CPUState *cpu); void hvf_arch_vcpu_destroy(CPUState *cpu); +int hvf_vcpu_exec(CPUState *); hvf_slot *hvf_find_overlap_slot(uint64_t, uint64_t); int hvf_put_registers(CPUState *); int hvf_get_registers(CPUState *); diff --git a/accel/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.c +++ b/accel/hvf/hvf-accel-ops.c @@ -XXX,XX +XXX,XX @@ #include "sysemu/runstate.h" #include "qemu/guest-random.h" -#include "hvf-accel-ops.h" - HVFState *hvf_state; /* Memory slots */ diff --git a/target/i386/hvf/hvf.c b/target/i386/hvf/hvf.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/hvf.c +++ b/target/i386/hvf/hvf.c @@ -XXX,XX +XXX,XX @@ #include "qemu/accel.h" #include "target/i386/cpu.h" -#include "hvf-accel-ops.h" - void vmx_update_tpr(CPUState *cpu) { /* TODO: need integrate APIC handling */ -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> We will need more than a single field for hvf going forward. To keep the global vcpu struct uncluttered, let's allocate a special hvf vcpu struct, similar to how hax does it. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Roman Bolshakov <r.bolshakov@yadro.com> Tested-by: Roman Bolshakov <r.bolshakov@yadro.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-12-agraf@csgraf.de Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/core/cpu.h | 3 +- include/sysemu/hvf_int.h | 4 + target/i386/hvf/vmx.h | 24 +++-- accel/hvf/hvf-accel-ops.c | 8 +- target/i386/hvf/hvf.c | 104 +++++++++--------- target/i386/hvf/x86.c | 28 ++--- target/i386/hvf/x86_descr.c | 26 ++--- target/i386/hvf/x86_emu.c | 62 +++++------ target/i386/hvf/x86_mmu.c | 4 +- target/i386/hvf/x86_task.c | 12 +-- target/i386/hvf/x86hvf.c | 210 ++++++++++++++++++------------------ 11 files changed, 248 insertions(+), 237 deletions(-) diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/core/cpu.h +++ b/include/hw/core/cpu.h @@ -XXX,XX +XXX,XX @@ struct KVMState; struct kvm_run; struct hax_vcpu_state; +struct hvf_vcpu_state; #define TB_JMP_CACHE_BITS 12 #define TB_JMP_CACHE_SIZE (1 << TB_JMP_CACHE_BITS) @@ -XXX,XX +XXX,XX @@ struct CPUState { struct hax_vcpu_state *hax_vcpu; - int hvf_fd; + struct hvf_vcpu_state *hvf; /* track IOMMUs whose translations we've cached in the TCG TLB */ GArray *iommu_notifiers; diff --git a/include/sysemu/hvf_int.h b/include/sysemu/hvf_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/hvf_int.h +++ b/include/sysemu/hvf_int.h @@ -XXX,XX +XXX,XX @@ struct HVFState { }; extern HVFState *hvf_state; +struct hvf_vcpu_state { + int fd; +}; + void assert_hvf_ok(hv_return_t ret); int hvf_arch_init_vcpu(CPUState *cpu); void hvf_arch_vcpu_destroy(CPUState *cpu); diff --git a/target/i386/hvf/vmx.h b/target/i386/hvf/vmx.h index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/vmx.h +++ b/target/i386/hvf/vmx.h @@ -XXX,XX +XXX,XX @@ #include "vmcs.h" #include "cpu.h" #include "x86.h" +#include "sysemu/hvf.h" +#include "sysemu/hvf_int.h" #include "exec/address-spaces.h" @@ -XXX,XX +XXX,XX @@ static inline void macvm_set_rip(CPUState *cpu, uint64_t rip) uint64_t val; /* BUG, should take considering overlap.. */ - wreg(cpu->hvf_fd, HV_X86_RIP, rip); + wreg(cpu->hvf->fd, HV_X86_RIP, rip); env->eip = rip; /* after moving forward in rip, we need to clean INTERRUPTABILITY */ - val = rvmcs(cpu->hvf_fd, VMCS_GUEST_INTERRUPTIBILITY); + val = rvmcs(cpu->hvf->fd, VMCS_GUEST_INTERRUPTIBILITY); if (val & (VMCS_INTERRUPTIBILITY_STI_BLOCKING | VMCS_INTERRUPTIBILITY_MOVSS_BLOCKING)) { env->hflags &= ~HF_INHIBIT_IRQ_MASK; - wvmcs(cpu->hvf_fd, VMCS_GUEST_INTERRUPTIBILITY, + wvmcs(cpu->hvf->fd, VMCS_GUEST_INTERRUPTIBILITY, val & ~(VMCS_INTERRUPTIBILITY_STI_BLOCKING | VMCS_INTERRUPTIBILITY_MOVSS_BLOCKING)); } @@ -XXX,XX +XXX,XX @@ static inline void vmx_clear_nmi_blocking(CPUState *cpu) CPUX86State *env = &x86_cpu->env; env->hflags2 &= ~HF2_NMI_MASK; - uint32_t gi = (uint32_t) rvmcs(cpu->hvf_fd, VMCS_GUEST_INTERRUPTIBILITY); + uint32_t gi = (uint32_t) rvmcs(cpu->hvf->fd, VMCS_GUEST_INTERRUPTIBILITY); gi &= ~VMCS_INTERRUPTIBILITY_NMI_BLOCKING; - wvmcs(cpu->hvf_fd, VMCS_GUEST_INTERRUPTIBILITY, gi); + wvmcs(cpu->hvf->fd, VMCS_GUEST_INTERRUPTIBILITY, gi); } static inline void vmx_set_nmi_blocking(CPUState *cpu) @@ -XXX,XX +XXX,XX @@ static inline void vmx_set_nmi_blocking(CPUState *cpu) CPUX86State *env = &x86_cpu->env; env->hflags2 |= HF2_NMI_MASK; - uint32_t gi = (uint32_t)rvmcs(cpu->hvf_fd, VMCS_GUEST_INTERRUPTIBILITY); + uint32_t gi = (uint32_t)rvmcs(cpu->hvf->fd, VMCS_GUEST_INTERRUPTIBILITY); gi |= VMCS_INTERRUPTIBILITY_NMI_BLOCKING; - wvmcs(cpu->hvf_fd, VMCS_GUEST_INTERRUPTIBILITY, gi); + wvmcs(cpu->hvf->fd, VMCS_GUEST_INTERRUPTIBILITY, gi); } static inline void vmx_set_nmi_window_exiting(CPUState *cpu) { uint64_t val; - val = rvmcs(cpu->hvf_fd, VMCS_PRI_PROC_BASED_CTLS); - wvmcs(cpu->hvf_fd, VMCS_PRI_PROC_BASED_CTLS, val | + val = rvmcs(cpu->hvf->fd, VMCS_PRI_PROC_BASED_CTLS); + wvmcs(cpu->hvf->fd, VMCS_PRI_PROC_BASED_CTLS, val | VMCS_PRI_PROC_BASED_CTLS_NMI_WINDOW_EXITING); } @@ -XXX,XX +XXX,XX @@ static inline void vmx_clear_nmi_window_exiting(CPUState *cpu) { uint64_t val; - val = rvmcs(cpu->hvf_fd, VMCS_PRI_PROC_BASED_CTLS); - wvmcs(cpu->hvf_fd, VMCS_PRI_PROC_BASED_CTLS, val & + val = rvmcs(cpu->hvf->fd, VMCS_PRI_PROC_BASED_CTLS); + wvmcs(cpu->hvf->fd, VMCS_PRI_PROC_BASED_CTLS, val & ~VMCS_PRI_PROC_BASED_CTLS_NMI_WINDOW_EXITING); } diff --git a/accel/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.c +++ b/accel/hvf/hvf-accel-ops.c @@ -XXX,XX +XXX,XX @@ type_init(hvf_type_init); static void hvf_vcpu_destroy(CPUState *cpu) { - hv_return_t ret = hv_vcpu_destroy(cpu->hvf_fd); + hv_return_t ret = hv_vcpu_destroy(cpu->hvf->fd); assert_hvf_ok(ret); hvf_arch_vcpu_destroy(cpu); + g_free(cpu->hvf); + cpu->hvf = NULL; } static int hvf_init_vcpu(CPUState *cpu) { int r; + cpu->hvf = g_malloc0(sizeof(*cpu->hvf)); + /* init cpu signals */ sigset_t set; struct sigaction sigact; @@ -XXX,XX +XXX,XX @@ static int hvf_init_vcpu(CPUState *cpu) pthread_sigmask(SIG_BLOCK, NULL, &set); sigdelset(&set, SIG_IPI); - r = hv_vcpu_create((hv_vcpuid_t *)&cpu->hvf_fd, HV_VCPU_DEFAULT); + r = hv_vcpu_create((hv_vcpuid_t *)&cpu->hvf->fd, HV_VCPU_DEFAULT); cpu->vcpu_dirty = 1; assert_hvf_ok(r); diff --git a/target/i386/hvf/hvf.c b/target/i386/hvf/hvf.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/hvf.c +++ b/target/i386/hvf/hvf.c @@ -XXX,XX +XXX,XX @@ void vmx_update_tpr(CPUState *cpu) int tpr = cpu_get_apic_tpr(x86_cpu->apic_state) << 4; int irr = apic_get_highest_priority_irr(x86_cpu->apic_state); - wreg(cpu->hvf_fd, HV_X86_TPR, tpr); + wreg(cpu->hvf->fd, HV_X86_TPR, tpr); if (irr == -1) { - wvmcs(cpu->hvf_fd, VMCS_TPR_THRESHOLD, 0); + wvmcs(cpu->hvf->fd, VMCS_TPR_THRESHOLD, 0); } else { - wvmcs(cpu->hvf_fd, VMCS_TPR_THRESHOLD, (irr > tpr) ? tpr >> 4 : + wvmcs(cpu->hvf->fd, VMCS_TPR_THRESHOLD, (irr > tpr) ? tpr >> 4 : irr >> 4); } } @@ -XXX,XX +XXX,XX @@ void vmx_update_tpr(CPUState *cpu) static void update_apic_tpr(CPUState *cpu) { X86CPU *x86_cpu = X86_CPU(cpu); - int tpr = rreg(cpu->hvf_fd, HV_X86_TPR) >> 4; + int tpr = rreg(cpu->hvf->fd, HV_X86_TPR) >> 4; cpu_set_apic_tpr(x86_cpu->apic_state, tpr); } @@ -XXX,XX +XXX,XX @@ int hvf_arch_init_vcpu(CPUState *cpu) } /* set VMCS control fields */ - wvmcs(cpu->hvf_fd, VMCS_PIN_BASED_CTLS, + wvmcs(cpu->hvf->fd, VMCS_PIN_BASED_CTLS, cap2ctrl(hvf_state->hvf_caps->vmx_cap_pinbased, VMCS_PIN_BASED_CTLS_EXTINT | VMCS_PIN_BASED_CTLS_NMI | VMCS_PIN_BASED_CTLS_VNMI)); - wvmcs(cpu->hvf_fd, VMCS_PRI_PROC_BASED_CTLS, + wvmcs(cpu->hvf->fd, VMCS_PRI_PROC_BASED_CTLS, cap2ctrl(hvf_state->hvf_caps->vmx_cap_procbased, VMCS_PRI_PROC_BASED_CTLS_HLT | VMCS_PRI_PROC_BASED_CTLS_MWAIT | VMCS_PRI_PROC_BASED_CTLS_TSC_OFFSET | VMCS_PRI_PROC_BASED_CTLS_TPR_SHADOW) | VMCS_PRI_PROC_BASED_CTLS_SEC_CONTROL); - wvmcs(cpu->hvf_fd, VMCS_SEC_PROC_BASED_CTLS, + wvmcs(cpu->hvf->fd, VMCS_SEC_PROC_BASED_CTLS, cap2ctrl(hvf_state->hvf_caps->vmx_cap_procbased2, VMCS_PRI_PROC_BASED2_CTLS_APIC_ACCESSES)); - wvmcs(cpu->hvf_fd, VMCS_ENTRY_CTLS, cap2ctrl(hvf_state->hvf_caps->vmx_cap_entry, + wvmcs(cpu->hvf->fd, VMCS_ENTRY_CTLS, cap2ctrl(hvf_state->hvf_caps->vmx_cap_entry, 0)); - wvmcs(cpu->hvf_fd, VMCS_EXCEPTION_BITMAP, 0); /* Double fault */ + wvmcs(cpu->hvf->fd, VMCS_EXCEPTION_BITMAP, 0); /* Double fault */ - wvmcs(cpu->hvf_fd, VMCS_TPR_THRESHOLD, 0); + wvmcs(cpu->hvf->fd, VMCS_TPR_THRESHOLD, 0); x86cpu = X86_CPU(cpu); x86cpu->env.xsave_buf = qemu_memalign(4096, 4096); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_STAR, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_LSTAR, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_CSTAR, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_FMASK, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_FSBASE, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_GSBASE, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_KERNELGSBASE, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_TSC_AUX, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_IA32_TSC, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_IA32_SYSENTER_CS, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_IA32_SYSENTER_EIP, 1); - hv_vcpu_enable_native_msr(cpu->hvf_fd, MSR_IA32_SYSENTER_ESP, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_STAR, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_LSTAR, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_CSTAR, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_FMASK, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_FSBASE, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_GSBASE, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_KERNELGSBASE, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_TSC_AUX, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_IA32_TSC, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_IA32_SYSENTER_CS, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_IA32_SYSENTER_EIP, 1); + hv_vcpu_enable_native_msr(cpu->hvf->fd, MSR_IA32_SYSENTER_ESP, 1); return 0; } @@ -XXX,XX +XXX,XX @@ static void hvf_store_events(CPUState *cpu, uint32_t ins_len, uint64_t idtvec_in } if (idtvec_info & VMCS_IDT_VEC_ERRCODE_VALID) { env->has_error_code = true; - env->error_code = rvmcs(cpu->hvf_fd, VMCS_IDT_VECTORING_ERROR); + env->error_code = rvmcs(cpu->hvf->fd, VMCS_IDT_VECTORING_ERROR); } } - if ((rvmcs(cpu->hvf_fd, VMCS_GUEST_INTERRUPTIBILITY) & + if ((rvmcs(cpu->hvf->fd, VMCS_GUEST_INTERRUPTIBILITY) & VMCS_INTERRUPTIBILITY_NMI_BLOCKING)) { env->hflags2 |= HF2_NMI_MASK; } else { env->hflags2 &= ~HF2_NMI_MASK; } - if (rvmcs(cpu->hvf_fd, VMCS_GUEST_INTERRUPTIBILITY) & + if (rvmcs(cpu->hvf->fd, VMCS_GUEST_INTERRUPTIBILITY) & (VMCS_INTERRUPTIBILITY_STI_BLOCKING | VMCS_INTERRUPTIBILITY_MOVSS_BLOCKING)) { env->hflags |= HF_INHIBIT_IRQ_MASK; @@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu) return EXCP_HLT; } - hv_return_t r = hv_vcpu_run(cpu->hvf_fd); + hv_return_t r = hv_vcpu_run(cpu->hvf->fd); assert_hvf_ok(r); /* handle VMEXIT */ - uint64_t exit_reason = rvmcs(cpu->hvf_fd, VMCS_EXIT_REASON); - uint64_t exit_qual = rvmcs(cpu->hvf_fd, VMCS_EXIT_QUALIFICATION); - uint32_t ins_len = (uint32_t)rvmcs(cpu->hvf_fd, + uint64_t exit_reason = rvmcs(cpu->hvf->fd, VMCS_EXIT_REASON); + uint64_t exit_qual = rvmcs(cpu->hvf->fd, VMCS_EXIT_QUALIFICATION); + uint32_t ins_len = (uint32_t)rvmcs(cpu->hvf->fd, VMCS_EXIT_INSTRUCTION_LENGTH); - uint64_t idtvec_info = rvmcs(cpu->hvf_fd, VMCS_IDT_VECTORING_INFO); + uint64_t idtvec_info = rvmcs(cpu->hvf->fd, VMCS_IDT_VECTORING_INFO); hvf_store_events(cpu, ins_len, idtvec_info); - rip = rreg(cpu->hvf_fd, HV_X86_RIP); - env->eflags = rreg(cpu->hvf_fd, HV_X86_RFLAGS); + rip = rreg(cpu->hvf->fd, HV_X86_RIP); + env->eflags = rreg(cpu->hvf->fd, HV_X86_RFLAGS); qemu_mutex_lock_iothread(); @@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu) case EXIT_REASON_EPT_FAULT: { hvf_slot *slot; - uint64_t gpa = rvmcs(cpu->hvf_fd, VMCS_GUEST_PHYSICAL_ADDRESS); + uint64_t gpa = rvmcs(cpu->hvf->fd, VMCS_GUEST_PHYSICAL_ADDRESS); if (((idtvec_info & VMCS_IDT_VEC_VALID) == 0) && ((exit_qual & EXIT_QUAL_NMIUDTI) != 0)) { @@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu) store_regs(cpu); break; } else if (!string && !in) { - RAX(env) = rreg(cpu->hvf_fd, HV_X86_RAX); + RAX(env) = rreg(cpu->hvf->fd, HV_X86_RAX); hvf_handle_io(env, port, &RAX(env), 1, size, 1); macvm_set_rip(cpu, rip + ins_len); break; @@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu) break; } case EXIT_REASON_CPUID: { - uint32_t rax = (uint32_t)rreg(cpu->hvf_fd, HV_X86_RAX); - uint32_t rbx = (uint32_t)rreg(cpu->hvf_fd, HV_X86_RBX); - uint32_t rcx = (uint32_t)rreg(cpu->hvf_fd, HV_X86_RCX); - uint32_t rdx = (uint32_t)rreg(cpu->hvf_fd, HV_X86_RDX); + uint32_t rax = (uint32_t)rreg(cpu->hvf->fd, HV_X86_RAX); + uint32_t rbx = (uint32_t)rreg(cpu->hvf->fd, HV_X86_RBX); + uint32_t rcx = (uint32_t)rreg(cpu->hvf->fd, HV_X86_RCX); + uint32_t rdx = (uint32_t)rreg(cpu->hvf->fd, HV_X86_RDX); if (rax == 1) { /* CPUID1.ecx.OSXSAVE needs to know CR4 */ - env->cr[4] = rvmcs(cpu->hvf_fd, VMCS_GUEST_CR4); + env->cr[4] = rvmcs(cpu->hvf->fd, VMCS_GUEST_CR4); } hvf_cpu_x86_cpuid(env, rax, rcx, &rax, &rbx, &rcx, &rdx); - wreg(cpu->hvf_fd, HV_X86_RAX, rax); - wreg(cpu->hvf_fd, HV_X86_RBX, rbx); - wreg(cpu->hvf_fd, HV_X86_RCX, rcx); - wreg(cpu->hvf_fd, HV_X86_RDX, rdx); + wreg(cpu->hvf->fd, HV_X86_RAX, rax); + wreg(cpu->hvf->fd, HV_X86_RBX, rbx); + wreg(cpu->hvf->fd, HV_X86_RCX, rcx); + wreg(cpu->hvf->fd, HV_X86_RDX, rdx); macvm_set_rip(cpu, rip + ins_len); break; @@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu) case EXIT_REASON_XSETBV: { X86CPU *x86_cpu = X86_CPU(cpu); CPUX86State *env = &x86_cpu->env; - uint32_t eax = (uint32_t)rreg(cpu->hvf_fd, HV_X86_RAX); - uint32_t ecx = (uint32_t)rreg(cpu->hvf_fd, HV_X86_RCX); - uint32_t edx = (uint32_t)rreg(cpu->hvf_fd, HV_X86_RDX); + uint32_t eax = (uint32_t)rreg(cpu->hvf->fd, HV_X86_RAX); + uint32_t ecx = (uint32_t)rreg(cpu->hvf->fd, HV_X86_RCX); + uint32_t edx = (uint32_t)rreg(cpu->hvf->fd, HV_X86_RDX); if (ecx) { macvm_set_rip(cpu, rip + ins_len); break; } env->xcr0 = ((uint64_t)edx << 32) | eax; - wreg(cpu->hvf_fd, HV_X86_XCR0, env->xcr0 | 1); + wreg(cpu->hvf->fd, HV_X86_XCR0, env->xcr0 | 1); macvm_set_rip(cpu, rip + ins_len); break; } @@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu) switch (cr) { case 0x0: { - macvm_set_cr0(cpu->hvf_fd, RRX(env, reg)); + macvm_set_cr0(cpu->hvf->fd, RRX(env, reg)); break; } case 4: { - macvm_set_cr4(cpu->hvf_fd, RRX(env, reg)); + macvm_set_cr4(cpu->hvf->fd, RRX(env, reg)); break; } case 8: { @@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu) break; } case EXIT_REASON_TASK_SWITCH: { - uint64_t vinfo = rvmcs(cpu->hvf_fd, VMCS_IDT_VECTORING_INFO); + uint64_t vinfo = rvmcs(cpu->hvf->fd, VMCS_IDT_VECTORING_INFO); x68_segment_selector sel = {.sel = exit_qual & 0xffff}; vmx_handle_task_switch(cpu, sel, (exit_qual >> 30) & 0x3, vinfo & VMCS_INTR_VALID, vinfo & VECTORING_INFO_VECTOR_MASK, vinfo @@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu) break; } case EXIT_REASON_RDPMC: - wreg(cpu->hvf_fd, HV_X86_RAX, 0); - wreg(cpu->hvf_fd, HV_X86_RDX, 0); + wreg(cpu->hvf->fd, HV_X86_RAX, 0); + wreg(cpu->hvf->fd, HV_X86_RDX, 0); macvm_set_rip(cpu, rip + ins_len); break; case VMX_REASON_VMCALL: diff --git a/target/i386/hvf/x86.c b/target/i386/hvf/x86.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86.c +++ b/target/i386/hvf/x86.c @@ -XXX,XX +XXX,XX @@ bool x86_read_segment_descriptor(struct CPUState *cpu, } if (GDT_SEL == sel.ti) { - base = rvmcs(cpu->hvf_fd, VMCS_GUEST_GDTR_BASE); - limit = rvmcs(cpu->hvf_fd, VMCS_GUEST_GDTR_LIMIT); + base = rvmcs(cpu->hvf->fd, VMCS_GUEST_GDTR_BASE); + limit = rvmcs(cpu->hvf->fd, VMCS_GUEST_GDTR_LIMIT); } else { - base = rvmcs(cpu->hvf_fd, VMCS_GUEST_LDTR_BASE); - limit = rvmcs(cpu->hvf_fd, VMCS_GUEST_LDTR_LIMIT); + base = rvmcs(cpu->hvf->fd, VMCS_GUEST_LDTR_BASE); + limit = rvmcs(cpu->hvf->fd, VMCS_GUEST_LDTR_LIMIT); } if (sel.index * 8 >= limit) { @@ -XXX,XX +XXX,XX @@ bool x86_write_segment_descriptor(struct CPUState *cpu, uint32_t limit; if (GDT_SEL == sel.ti) { - base = rvmcs(cpu->hvf_fd, VMCS_GUEST_GDTR_BASE); - limit = rvmcs(cpu->hvf_fd, VMCS_GUEST_GDTR_LIMIT); + base = rvmcs(cpu->hvf->fd, VMCS_GUEST_GDTR_BASE); + limit = rvmcs(cpu->hvf->fd, VMCS_GUEST_GDTR_LIMIT); } else { - base = rvmcs(cpu->hvf_fd, VMCS_GUEST_LDTR_BASE); - limit = rvmcs(cpu->hvf_fd, VMCS_GUEST_LDTR_LIMIT); + base = rvmcs(cpu->hvf->fd, VMCS_GUEST_LDTR_BASE); + limit = rvmcs(cpu->hvf->fd, VMCS_GUEST_LDTR_LIMIT); } if (sel.index * 8 >= limit) { @@ -XXX,XX +XXX,XX @@ bool x86_write_segment_descriptor(struct CPUState *cpu, bool x86_read_call_gate(struct CPUState *cpu, struct x86_call_gate *idt_desc, int gate) { - target_ulong base = rvmcs(cpu->hvf_fd, VMCS_GUEST_IDTR_BASE); - uint32_t limit = rvmcs(cpu->hvf_fd, VMCS_GUEST_IDTR_LIMIT); + target_ulong base = rvmcs(cpu->hvf->fd, VMCS_GUEST_IDTR_BASE); + uint32_t limit = rvmcs(cpu->hvf->fd, VMCS_GUEST_IDTR_LIMIT); memset(idt_desc, 0, sizeof(*idt_desc)); if (gate * 8 >= limit) { @@ -XXX,XX +XXX,XX @@ bool x86_read_call_gate(struct CPUState *cpu, struct x86_call_gate *idt_desc, bool x86_is_protected(struct CPUState *cpu) { - uint64_t cr0 = rvmcs(cpu->hvf_fd, VMCS_GUEST_CR0); + uint64_t cr0 = rvmcs(cpu->hvf->fd, VMCS_GUEST_CR0); return cr0 & CR0_PE; } @@ -XXX,XX +XXX,XX @@ bool x86_is_v8086(struct CPUState *cpu) bool x86_is_long_mode(struct CPUState *cpu) { - return rvmcs(cpu->hvf_fd, VMCS_GUEST_IA32_EFER) & MSR_EFER_LMA; + return rvmcs(cpu->hvf->fd, VMCS_GUEST_IA32_EFER) & MSR_EFER_LMA; } bool x86_is_long64_mode(struct CPUState *cpu) @@ -XXX,XX +XXX,XX @@ bool x86_is_long64_mode(struct CPUState *cpu) bool x86_is_paging_mode(struct CPUState *cpu) { - uint64_t cr0 = rvmcs(cpu->hvf_fd, VMCS_GUEST_CR0); + uint64_t cr0 = rvmcs(cpu->hvf->fd, VMCS_GUEST_CR0); return cr0 & CR0_PG; } bool x86_is_pae_enabled(struct CPUState *cpu) { - uint64_t cr4 = rvmcs(cpu->hvf_fd, VMCS_GUEST_CR4); + uint64_t cr4 = rvmcs(cpu->hvf->fd, VMCS_GUEST_CR4); return cr4 & CR4_PAE; } diff --git a/target/i386/hvf/x86_descr.c b/target/i386/hvf/x86_descr.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86_descr.c +++ b/target/i386/hvf/x86_descr.c @@ -XXX,XX +XXX,XX @@ static const struct vmx_segment_field { uint32_t vmx_read_segment_limit(CPUState *cpu, X86Seg seg) { - return (uint32_t)rvmcs(cpu->hvf_fd, vmx_segment_fields[seg].limit); + return (uint32_t)rvmcs(cpu->hvf->fd, vmx_segment_fields[seg].limit); } uint32_t vmx_read_segment_ar(CPUState *cpu, X86Seg seg) { - return (uint32_t)rvmcs(cpu->hvf_fd, vmx_segment_fields[seg].ar_bytes); + return (uint32_t)rvmcs(cpu->hvf->fd, vmx_segment_fields[seg].ar_bytes); } uint64_t vmx_read_segment_base(CPUState *cpu, X86Seg seg) { - return rvmcs(cpu->hvf_fd, vmx_segment_fields[seg].base); + return rvmcs(cpu->hvf->fd, vmx_segment_fields[seg].base); } x68_segment_selector vmx_read_segment_selector(CPUState *cpu, X86Seg seg) { x68_segment_selector sel; - sel.sel = rvmcs(cpu->hvf_fd, vmx_segment_fields[seg].selector); + sel.sel = rvmcs(cpu->hvf->fd, vmx_segment_fields[seg].selector); return sel; } void vmx_write_segment_selector(struct CPUState *cpu, x68_segment_selector selector, X86Seg seg) { - wvmcs(cpu->hvf_fd, vmx_segment_fields[seg].selector, selector.sel); + wvmcs(cpu->hvf->fd, vmx_segment_fields[seg].selector, selector.sel); } void vmx_read_segment_descriptor(struct CPUState *cpu, struct vmx_segment *desc, X86Seg seg) { - desc->sel = rvmcs(cpu->hvf_fd, vmx_segment_fields[seg].selector); - desc->base = rvmcs(cpu->hvf_fd, vmx_segment_fields[seg].base); - desc->limit = rvmcs(cpu->hvf_fd, vmx_segment_fields[seg].limit); - desc->ar = rvmcs(cpu->hvf_fd, vmx_segment_fields[seg].ar_bytes); + desc->sel = rvmcs(cpu->hvf->fd, vmx_segment_fields[seg].selector); + desc->base = rvmcs(cpu->hvf->fd, vmx_segment_fields[seg].base); + desc->limit = rvmcs(cpu->hvf->fd, vmx_segment_fields[seg].limit); + desc->ar = rvmcs(cpu->hvf->fd, vmx_segment_fields[seg].ar_bytes); } void vmx_write_segment_descriptor(CPUState *cpu, struct vmx_segment *desc, X86Seg seg) { const struct vmx_segment_field *sf = &vmx_segment_fields[seg]; - wvmcs(cpu->hvf_fd, sf->base, desc->base); - wvmcs(cpu->hvf_fd, sf->limit, desc->limit); - wvmcs(cpu->hvf_fd, sf->selector, desc->sel); - wvmcs(cpu->hvf_fd, sf->ar_bytes, desc->ar); + wvmcs(cpu->hvf->fd, sf->base, desc->base); + wvmcs(cpu->hvf->fd, sf->limit, desc->limit); + wvmcs(cpu->hvf->fd, sf->selector, desc->sel); + wvmcs(cpu->hvf->fd, sf->ar_bytes, desc->ar); } void x86_segment_descriptor_to_vmx(struct CPUState *cpu, x68_segment_selector selector, struct x86_segment_descriptor *desc, struct vmx_segment *vmx_desc) diff --git a/target/i386/hvf/x86_emu.c b/target/i386/hvf/x86_emu.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86_emu.c +++ b/target/i386/hvf/x86_emu.c @@ -XXX,XX +XXX,XX @@ void simulate_rdmsr(struct CPUState *cpu) switch (msr) { case MSR_IA32_TSC: - val = rdtscp() + rvmcs(cpu->hvf_fd, VMCS_TSC_OFFSET); + val = rdtscp() + rvmcs(cpu->hvf->fd, VMCS_TSC_OFFSET); break; case MSR_IA32_APICBASE: val = cpu_get_apic_base(X86_CPU(cpu)->apic_state); @@ -XXX,XX +XXX,XX @@ void simulate_rdmsr(struct CPUState *cpu) val = x86_cpu->ucode_rev; break; case MSR_EFER: - val = rvmcs(cpu->hvf_fd, VMCS_GUEST_IA32_EFER); + val = rvmcs(cpu->hvf->fd, VMCS_GUEST_IA32_EFER); break; case MSR_FSBASE: - val = rvmcs(cpu->hvf_fd, VMCS_GUEST_FS_BASE); + val = rvmcs(cpu->hvf->fd, VMCS_GUEST_FS_BASE); break; case MSR_GSBASE: - val = rvmcs(cpu->hvf_fd, VMCS_GUEST_GS_BASE); + val = rvmcs(cpu->hvf->fd, VMCS_GUEST_GS_BASE); break; case MSR_KERNELGSBASE: - val = rvmcs(cpu->hvf_fd, VMCS_HOST_FS_BASE); + val = rvmcs(cpu->hvf->fd, VMCS_HOST_FS_BASE); break; case MSR_STAR: abort(); @@ -XXX,XX +XXX,XX @@ void simulate_wrmsr(struct CPUState *cpu) cpu_set_apic_base(X86_CPU(cpu)->apic_state, data); break; case MSR_FSBASE: - wvmcs(cpu->hvf_fd, VMCS_GUEST_FS_BASE, data); + wvmcs(cpu->hvf->fd, VMCS_GUEST_FS_BASE, data); break; case MSR_GSBASE: - wvmcs(cpu->hvf_fd, VMCS_GUEST_GS_BASE, data); + wvmcs(cpu->hvf->fd, VMCS_GUEST_GS_BASE, data); break; case MSR_KERNELGSBASE: - wvmcs(cpu->hvf_fd, VMCS_HOST_FS_BASE, data); + wvmcs(cpu->hvf->fd, VMCS_HOST_FS_BASE, data); break; case MSR_STAR: abort(); @@ -XXX,XX +XXX,XX @@ void simulate_wrmsr(struct CPUState *cpu) break; case MSR_EFER: /*printf("new efer %llx\n", EFER(cpu));*/ - wvmcs(cpu->hvf_fd, VMCS_GUEST_IA32_EFER, data); + wvmcs(cpu->hvf->fd, VMCS_GUEST_IA32_EFER, data); if (data & MSR_EFER_NXE) { - hv_vcpu_invalidate_tlb(cpu->hvf_fd); + hv_vcpu_invalidate_tlb(cpu->hvf->fd); } break; case MSR_MTRRphysBase(0): @@ -XXX,XX +XXX,XX @@ void load_regs(struct CPUState *cpu) CPUX86State *env = &x86_cpu->env; int i = 0; - RRX(env, R_EAX) = rreg(cpu->hvf_fd, HV_X86_RAX); - RRX(env, R_EBX) = rreg(cpu->hvf_fd, HV_X86_RBX); - RRX(env, R_ECX) = rreg(cpu->hvf_fd, HV_X86_RCX); - RRX(env, R_EDX) = rreg(cpu->hvf_fd, HV_X86_RDX); - RRX(env, R_ESI) = rreg(cpu->hvf_fd, HV_X86_RSI); - RRX(env, R_EDI) = rreg(cpu->hvf_fd, HV_X86_RDI); - RRX(env, R_ESP) = rreg(cpu->hvf_fd, HV_X86_RSP); - RRX(env, R_EBP) = rreg(cpu->hvf_fd, HV_X86_RBP); + RRX(env, R_EAX) = rreg(cpu->hvf->fd, HV_X86_RAX); + RRX(env, R_EBX) = rreg(cpu->hvf->fd, HV_X86_RBX); + RRX(env, R_ECX) = rreg(cpu->hvf->fd, HV_X86_RCX); + RRX(env, R_EDX) = rreg(cpu->hvf->fd, HV_X86_RDX); + RRX(env, R_ESI) = rreg(cpu->hvf->fd, HV_X86_RSI); + RRX(env, R_EDI) = rreg(cpu->hvf->fd, HV_X86_RDI); + RRX(env, R_ESP) = rreg(cpu->hvf->fd, HV_X86_RSP); + RRX(env, R_EBP) = rreg(cpu->hvf->fd, HV_X86_RBP); for (i = 8; i < 16; i++) { - RRX(env, i) = rreg(cpu->hvf_fd, HV_X86_RAX + i); + RRX(env, i) = rreg(cpu->hvf->fd, HV_X86_RAX + i); } - env->eflags = rreg(cpu->hvf_fd, HV_X86_RFLAGS); + env->eflags = rreg(cpu->hvf->fd, HV_X86_RFLAGS); rflags_to_lflags(env); - env->eip = rreg(cpu->hvf_fd, HV_X86_RIP); + env->eip = rreg(cpu->hvf->fd, HV_X86_RIP); } void store_regs(struct CPUState *cpu) @@ -XXX,XX +XXX,XX @@ void store_regs(struct CPUState *cpu) CPUX86State *env = &x86_cpu->env; int i = 0; - wreg(cpu->hvf_fd, HV_X86_RAX, RAX(env)); - wreg(cpu->hvf_fd, HV_X86_RBX, RBX(env)); - wreg(cpu->hvf_fd, HV_X86_RCX, RCX(env)); - wreg(cpu->hvf_fd, HV_X86_RDX, RDX(env)); - wreg(cpu->hvf_fd, HV_X86_RSI, RSI(env)); - wreg(cpu->hvf_fd, HV_X86_RDI, RDI(env)); - wreg(cpu->hvf_fd, HV_X86_RBP, RBP(env)); - wreg(cpu->hvf_fd, HV_X86_RSP, RSP(env)); + wreg(cpu->hvf->fd, HV_X86_RAX, RAX(env)); + wreg(cpu->hvf->fd, HV_X86_RBX, RBX(env)); + wreg(cpu->hvf->fd, HV_X86_RCX, RCX(env)); + wreg(cpu->hvf->fd, HV_X86_RDX, RDX(env)); + wreg(cpu->hvf->fd, HV_X86_RSI, RSI(env)); + wreg(cpu->hvf->fd, HV_X86_RDI, RDI(env)); + wreg(cpu->hvf->fd, HV_X86_RBP, RBP(env)); + wreg(cpu->hvf->fd, HV_X86_RSP, RSP(env)); for (i = 8; i < 16; i++) { - wreg(cpu->hvf_fd, HV_X86_RAX + i, RRX(env, i)); + wreg(cpu->hvf->fd, HV_X86_RAX + i, RRX(env, i)); } lflags_to_rflags(env); - wreg(cpu->hvf_fd, HV_X86_RFLAGS, env->eflags); + wreg(cpu->hvf->fd, HV_X86_RFLAGS, env->eflags); macvm_set_rip(cpu, env->eip); } diff --git a/target/i386/hvf/x86_mmu.c b/target/i386/hvf/x86_mmu.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86_mmu.c +++ b/target/i386/hvf/x86_mmu.c @@ -XXX,XX +XXX,XX @@ static bool test_pt_entry(struct CPUState *cpu, struct gpt_translation *pt, pt->err_code |= MMU_PAGE_PT; } - uint32_t cr0 = rvmcs(cpu->hvf_fd, VMCS_GUEST_CR0); + uint32_t cr0 = rvmcs(cpu->hvf->fd, VMCS_GUEST_CR0); /* check protection */ if (cr0 & CR0_WP) { if (pt->write_access && !pte_write_access(pte)) { @@ -XXX,XX +XXX,XX @@ static bool walk_gpt(struct CPUState *cpu, target_ulong addr, int err_code, { int top_level, level; bool is_large = false; - target_ulong cr3 = rvmcs(cpu->hvf_fd, VMCS_GUEST_CR3); + target_ulong cr3 = rvmcs(cpu->hvf->fd, VMCS_GUEST_CR3); uint64_t page_mask = pae ? PAE_PTE_PAGE_MASK : LEGACY_PTE_PAGE_MASK; memset(pt, 0, sizeof(*pt)); diff --git a/target/i386/hvf/x86_task.c b/target/i386/hvf/x86_task.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86_task.c +++ b/target/i386/hvf/x86_task.c @@ -XXX,XX +XXX,XX @@ static void load_state_from_tss32(CPUState *cpu, struct x86_tss_segment32 *tss) X86CPU *x86_cpu = X86_CPU(cpu); CPUX86State *env = &x86_cpu->env; - wvmcs(cpu->hvf_fd, VMCS_GUEST_CR3, tss->cr3); + wvmcs(cpu->hvf->fd, VMCS_GUEST_CR3, tss->cr3); env->eip = tss->eip; env->eflags = tss->eflags | 2; @@ -XXX,XX +XXX,XX @@ static int task_switch_32(CPUState *cpu, x68_segment_selector tss_sel, x68_segme void vmx_handle_task_switch(CPUState *cpu, x68_segment_selector tss_sel, int reason, bool gate_valid, uint8_t gate, uint64_t gate_type) { - uint64_t rip = rreg(cpu->hvf_fd, HV_X86_RIP); + uint64_t rip = rreg(cpu->hvf->fd, HV_X86_RIP); if (!gate_valid || (gate_type != VMCS_INTR_T_HWEXCEPTION && gate_type != VMCS_INTR_T_HWINTR && gate_type != VMCS_INTR_T_NMI)) { - int ins_len = rvmcs(cpu->hvf_fd, VMCS_EXIT_INSTRUCTION_LENGTH); + int ins_len = rvmcs(cpu->hvf->fd, VMCS_EXIT_INSTRUCTION_LENGTH); macvm_set_rip(cpu, rip + ins_len); return; } @@ -XXX,XX +XXX,XX @@ void vmx_handle_task_switch(CPUState *cpu, x68_segment_selector tss_sel, int rea //ret = task_switch_16(cpu, tss_sel, old_tss_sel, old_tss_base, &next_tss_desc); VM_PANIC("task_switch_16"); - macvm_set_cr0(cpu->hvf_fd, rvmcs(cpu->hvf_fd, VMCS_GUEST_CR0) | CR0_TS); + macvm_set_cr0(cpu->hvf->fd, rvmcs(cpu->hvf->fd, VMCS_GUEST_CR0) | CR0_TS); x86_segment_descriptor_to_vmx(cpu, tss_sel, &next_tss_desc, &vmx_seg); vmx_write_segment_descriptor(cpu, &vmx_seg, R_TR); store_regs(cpu); - hv_vcpu_invalidate_tlb(cpu->hvf_fd); - hv_vcpu_flush(cpu->hvf_fd); + hv_vcpu_invalidate_tlb(cpu->hvf->fd); + hv_vcpu_flush(cpu->hvf->fd); } diff --git a/target/i386/hvf/x86hvf.c b/target/i386/hvf/x86hvf.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86hvf.c +++ b/target/i386/hvf/x86hvf.c @@ -XXX,XX +XXX,XX @@ void hvf_put_xsave(CPUState *cpu_state) x86_cpu_xsave_all_areas(X86_CPU(cpu_state), xsave); - if (hv_vcpu_write_fpstate(cpu_state->hvf_fd, (void*)xsave, 4096)) { + if (hv_vcpu_write_fpstate(cpu_state->hvf->fd, (void*)xsave, 4096)) { abort(); } } @@ -XXX,XX +XXX,XX @@ void hvf_put_segments(CPUState *cpu_state) CPUX86State *env = &X86_CPU(cpu_state)->env; struct vmx_segment seg; - wvmcs(cpu_state->hvf_fd, VMCS_GUEST_IDTR_LIMIT, env->idt.limit); - wvmcs(cpu_state->hvf_fd, VMCS_GUEST_IDTR_BASE, env->idt.base); + wvmcs(cpu_state->hvf->fd, VMCS_GUEST_IDTR_LIMIT, env->idt.limit); + wvmcs(cpu_state->hvf->fd, VMCS_GUEST_IDTR_BASE, env->idt.base); - wvmcs(cpu_state->hvf_fd, VMCS_GUEST_GDTR_LIMIT, env->gdt.limit); - wvmcs(cpu_state->hvf_fd, VMCS_GUEST_GDTR_BASE, env->gdt.base); + wvmcs(cpu_state->hvf->fd, VMCS_GUEST_GDTR_LIMIT, env->gdt.limit); + wvmcs(cpu_state->hvf->fd, VMCS_GUEST_GDTR_BASE, env->gdt.base); - /* wvmcs(cpu_state->hvf_fd, VMCS_GUEST_CR2, env->cr[2]); */ - wvmcs(cpu_state->hvf_fd, VMCS_GUEST_CR3, env->cr[3]); + /* wvmcs(cpu_state->hvf->fd, VMCS_GUEST_CR2, env->cr[2]); */ + wvmcs(cpu_state->hvf->fd, VMCS_GUEST_CR3, env->cr[3]); vmx_update_tpr(cpu_state); - wvmcs(cpu_state->hvf_fd, VMCS_GUEST_IA32_EFER, env->efer); + wvmcs(cpu_state->hvf->fd, VMCS_GUEST_IA32_EFER, env->efer); - macvm_set_cr4(cpu_state->hvf_fd, env->cr[4]); - macvm_set_cr0(cpu_state->hvf_fd, env->cr[0]); + macvm_set_cr4(cpu_state->hvf->fd, env->cr[4]); + macvm_set_cr0(cpu_state->hvf->fd, env->cr[0]); hvf_set_segment(cpu_state, &seg, &env->segs[R_CS], false); vmx_write_segment_descriptor(cpu_state, &seg, R_CS); @@ -XXX,XX +XXX,XX @@ void hvf_put_segments(CPUState *cpu_state) hvf_set_segment(cpu_state, &seg, &env->ldt, false); vmx_write_segment_descriptor(cpu_state, &seg, R_LDTR); - hv_vcpu_flush(cpu_state->hvf_fd); + hv_vcpu_flush(cpu_state->hvf->fd); } void hvf_put_msrs(CPUState *cpu_state) { CPUX86State *env = &X86_CPU(cpu_state)->env; - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_IA32_SYSENTER_CS, + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_IA32_SYSENTER_CS, env->sysenter_cs); - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_IA32_SYSENTER_ESP, + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_IA32_SYSENTER_ESP, env->sysenter_esp); - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_IA32_SYSENTER_EIP, + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_IA32_SYSENTER_EIP, env->sysenter_eip); - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_STAR, env->star); + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_STAR, env->star); #ifdef TARGET_X86_64 - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_CSTAR, env->cstar); - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_KERNELGSBASE, env->kernelgsbase); - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_FMASK, env->fmask); - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_LSTAR, env->lstar); + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_CSTAR, env->cstar); + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_KERNELGSBASE, env->kernelgsbase); + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_FMASK, env->fmask); + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_LSTAR, env->lstar); #endif - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_GSBASE, env->segs[R_GS].base); - hv_vcpu_write_msr(cpu_state->hvf_fd, MSR_FSBASE, env->segs[R_FS].base); + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_GSBASE, env->segs[R_GS].base); + hv_vcpu_write_msr(cpu_state->hvf->fd, MSR_FSBASE, env->segs[R_FS].base); } @@ -XXX,XX +XXX,XX @@ void hvf_get_xsave(CPUState *cpu_state) xsave = X86_CPU(cpu_state)->env.xsave_buf; - if (hv_vcpu_read_fpstate(cpu_state->hvf_fd, (void*)xsave, 4096)) { + if (hv_vcpu_read_fpstate(cpu_state->hvf->fd, (void*)xsave, 4096)) { abort(); } @@ -XXX,XX +XXX,XX @@ void hvf_get_segments(CPUState *cpu_state) vmx_read_segment_descriptor(cpu_state, &seg, R_LDTR); hvf_get_segment(&env->ldt, &seg); - env->idt.limit = rvmcs(cpu_state->hvf_fd, VMCS_GUEST_IDTR_LIMIT); - env->idt.base = rvmcs(cpu_state->hvf_fd, VMCS_GUEST_IDTR_BASE); - env->gdt.limit = rvmcs(cpu_state->hvf_fd, VMCS_GUEST_GDTR_LIMIT); - env->gdt.base = rvmcs(cpu_state->hvf_fd, VMCS_GUEST_GDTR_BASE); + env->idt.limit = rvmcs(cpu_state->hvf->fd, VMCS_GUEST_IDTR_LIMIT); + env->idt.base = rvmcs(cpu_state->hvf->fd, VMCS_GUEST_IDTR_BASE); + env->gdt.limit = rvmcs(cpu_state->hvf->fd, VMCS_GUEST_GDTR_LIMIT); + env->gdt.base = rvmcs(cpu_state->hvf->fd, VMCS_GUEST_GDTR_BASE); - env->cr[0] = rvmcs(cpu_state->hvf_fd, VMCS_GUEST_CR0); + env->cr[0] = rvmcs(cpu_state->hvf->fd, VMCS_GUEST_CR0); env->cr[2] = 0; - env->cr[3] = rvmcs(cpu_state->hvf_fd, VMCS_GUEST_CR3); - env->cr[4] = rvmcs(cpu_state->hvf_fd, VMCS_GUEST_CR4); + env->cr[3] = rvmcs(cpu_state->hvf->fd, VMCS_GUEST_CR3); + env->cr[4] = rvmcs(cpu_state->hvf->fd, VMCS_GUEST_CR4); - env->efer = rvmcs(cpu_state->hvf_fd, VMCS_GUEST_IA32_EFER); + env->efer = rvmcs(cpu_state->hvf->fd, VMCS_GUEST_IA32_EFER); } void hvf_get_msrs(CPUState *cpu_state) @@ -XXX,XX +XXX,XX @@ void hvf_get_msrs(CPUState *cpu_state) CPUX86State *env = &X86_CPU(cpu_state)->env; uint64_t tmp; - hv_vcpu_read_msr(cpu_state->hvf_fd, MSR_IA32_SYSENTER_CS, &tmp); + hv_vcpu_read_msr(cpu_state->hvf->fd, MSR_IA32_SYSENTER_CS, &tmp); env->sysenter_cs = tmp; - hv_vcpu_read_msr(cpu_state->hvf_fd, MSR_IA32_SYSENTER_ESP, &tmp); + hv_vcpu_read_msr(cpu_state->hvf->fd, MSR_IA32_SYSENTER_ESP, &tmp); env->sysenter_esp = tmp; - hv_vcpu_read_msr(cpu_state->hvf_fd, MSR_IA32_SYSENTER_EIP, &tmp); + hv_vcpu_read_msr(cpu_state->hvf->fd, MSR_IA32_SYSENTER_EIP, &tmp); env->sysenter_eip = tmp; - hv_vcpu_read_msr(cpu_state->hvf_fd, MSR_STAR, &env->star); + hv_vcpu_read_msr(cpu_state->hvf->fd, MSR_STAR, &env->star); #ifdef TARGET_X86_64 - hv_vcpu_read_msr(cpu_state->hvf_fd, MSR_CSTAR, &env->cstar); - hv_vcpu_read_msr(cpu_state->hvf_fd, MSR_KERNELGSBASE, &env->kernelgsbase); - hv_vcpu_read_msr(cpu_state->hvf_fd, MSR_FMASK, &env->fmask); - hv_vcpu_read_msr(cpu_state->hvf_fd, MSR_LSTAR, &env->lstar); + hv_vcpu_read_msr(cpu_state->hvf->fd, MSR_CSTAR, &env->cstar); + hv_vcpu_read_msr(cpu_state->hvf->fd, MSR_KERNELGSBASE, &env->kernelgsbase); + hv_vcpu_read_msr(cpu_state->hvf->fd, MSR_FMASK, &env->fmask); + hv_vcpu_read_msr(cpu_state->hvf->fd, MSR_LSTAR, &env->lstar); #endif - hv_vcpu_read_msr(cpu_state->hvf_fd, MSR_IA32_APICBASE, &tmp); + hv_vcpu_read_msr(cpu_state->hvf->fd, MSR_IA32_APICBASE, &tmp); - env->tsc = rdtscp() + rvmcs(cpu_state->hvf_fd, VMCS_TSC_OFFSET); + env->tsc = rdtscp() + rvmcs(cpu_state->hvf->fd, VMCS_TSC_OFFSET); } int hvf_put_registers(CPUState *cpu_state) @@ -XXX,XX +XXX,XX @@ int hvf_put_registers(CPUState *cpu_state) X86CPU *x86cpu = X86_CPU(cpu_state); CPUX86State *env = &x86cpu->env; - wreg(cpu_state->hvf_fd, HV_X86_RAX, env->regs[R_EAX]); - wreg(cpu_state->hvf_fd, HV_X86_RBX, env->regs[R_EBX]); - wreg(cpu_state->hvf_fd, HV_X86_RCX, env->regs[R_ECX]); - wreg(cpu_state->hvf_fd, HV_X86_RDX, env->regs[R_EDX]); - wreg(cpu_state->hvf_fd, HV_X86_RBP, env->regs[R_EBP]); - wreg(cpu_state->hvf_fd, HV_X86_RSP, env->regs[R_ESP]); - wreg(cpu_state->hvf_fd, HV_X86_RSI, env->regs[R_ESI]); - wreg(cpu_state->hvf_fd, HV_X86_RDI, env->regs[R_EDI]); - wreg(cpu_state->hvf_fd, HV_X86_R8, env->regs[8]); - wreg(cpu_state->hvf_fd, HV_X86_R9, env->regs[9]); - wreg(cpu_state->hvf_fd, HV_X86_R10, env->regs[10]); - wreg(cpu_state->hvf_fd, HV_X86_R11, env->regs[11]); - wreg(cpu_state->hvf_fd, HV_X86_R12, env->regs[12]); - wreg(cpu_state->hvf_fd, HV_X86_R13, env->regs[13]); - wreg(cpu_state->hvf_fd, HV_X86_R14, env->regs[14]); - wreg(cpu_state->hvf_fd, HV_X86_R15, env->regs[15]); - wreg(cpu_state->hvf_fd, HV_X86_RFLAGS, env->eflags); - wreg(cpu_state->hvf_fd, HV_X86_RIP, env->eip); + wreg(cpu_state->hvf->fd, HV_X86_RAX, env->regs[R_EAX]); + wreg(cpu_state->hvf->fd, HV_X86_RBX, env->regs[R_EBX]); + wreg(cpu_state->hvf->fd, HV_X86_RCX, env->regs[R_ECX]); + wreg(cpu_state->hvf->fd, HV_X86_RDX, env->regs[R_EDX]); + wreg(cpu_state->hvf->fd, HV_X86_RBP, env->regs[R_EBP]); + wreg(cpu_state->hvf->fd, HV_X86_RSP, env->regs[R_ESP]); + wreg(cpu_state->hvf->fd, HV_X86_RSI, env->regs[R_ESI]); + wreg(cpu_state->hvf->fd, HV_X86_RDI, env->regs[R_EDI]); + wreg(cpu_state->hvf->fd, HV_X86_R8, env->regs[8]); + wreg(cpu_state->hvf->fd, HV_X86_R9, env->regs[9]); + wreg(cpu_state->hvf->fd, HV_X86_R10, env->regs[10]); + wreg(cpu_state->hvf->fd, HV_X86_R11, env->regs[11]); + wreg(cpu_state->hvf->fd, HV_X86_R12, env->regs[12]); + wreg(cpu_state->hvf->fd, HV_X86_R13, env->regs[13]); + wreg(cpu_state->hvf->fd, HV_X86_R14, env->regs[14]); + wreg(cpu_state->hvf->fd, HV_X86_R15, env->regs[15]); + wreg(cpu_state->hvf->fd, HV_X86_RFLAGS, env->eflags); + wreg(cpu_state->hvf->fd, HV_X86_RIP, env->eip); - wreg(cpu_state->hvf_fd, HV_X86_XCR0, env->xcr0); + wreg(cpu_state->hvf->fd, HV_X86_XCR0, env->xcr0); hvf_put_xsave(cpu_state); @@ -XXX,XX +XXX,XX @@ int hvf_put_registers(CPUState *cpu_state) hvf_put_msrs(cpu_state); - wreg(cpu_state->hvf_fd, HV_X86_DR0, env->dr[0]); - wreg(cpu_state->hvf_fd, HV_X86_DR1, env->dr[1]); - wreg(cpu_state->hvf_fd, HV_X86_DR2, env->dr[2]); - wreg(cpu_state->hvf_fd, HV_X86_DR3, env->dr[3]); - wreg(cpu_state->hvf_fd, HV_X86_DR4, env->dr[4]); - wreg(cpu_state->hvf_fd, HV_X86_DR5, env->dr[5]); - wreg(cpu_state->hvf_fd, HV_X86_DR6, env->dr[6]); - wreg(cpu_state->hvf_fd, HV_X86_DR7, env->dr[7]); + wreg(cpu_state->hvf->fd, HV_X86_DR0, env->dr[0]); + wreg(cpu_state->hvf->fd, HV_X86_DR1, env->dr[1]); + wreg(cpu_state->hvf->fd, HV_X86_DR2, env->dr[2]); + wreg(cpu_state->hvf->fd, HV_X86_DR3, env->dr[3]); + wreg(cpu_state->hvf->fd, HV_X86_DR4, env->dr[4]); + wreg(cpu_state->hvf->fd, HV_X86_DR5, env->dr[5]); + wreg(cpu_state->hvf->fd, HV_X86_DR6, env->dr[6]); + wreg(cpu_state->hvf->fd, HV_X86_DR7, env->dr[7]); return 0; } @@ -XXX,XX +XXX,XX @@ int hvf_get_registers(CPUState *cpu_state) X86CPU *x86cpu = X86_CPU(cpu_state); CPUX86State *env = &x86cpu->env; - env->regs[R_EAX] = rreg(cpu_state->hvf_fd, HV_X86_RAX); - env->regs[R_EBX] = rreg(cpu_state->hvf_fd, HV_X86_RBX); - env->regs[R_ECX] = rreg(cpu_state->hvf_fd, HV_X86_RCX); - env->regs[R_EDX] = rreg(cpu_state->hvf_fd, HV_X86_RDX); - env->regs[R_EBP] = rreg(cpu_state->hvf_fd, HV_X86_RBP); - env->regs[R_ESP] = rreg(cpu_state->hvf_fd, HV_X86_RSP); - env->regs[R_ESI] = rreg(cpu_state->hvf_fd, HV_X86_RSI); - env->regs[R_EDI] = rreg(cpu_state->hvf_fd, HV_X86_RDI); - env->regs[8] = rreg(cpu_state->hvf_fd, HV_X86_R8); - env->regs[9] = rreg(cpu_state->hvf_fd, HV_X86_R9); - env->regs[10] = rreg(cpu_state->hvf_fd, HV_X86_R10); - env->regs[11] = rreg(cpu_state->hvf_fd, HV_X86_R11); - env->regs[12] = rreg(cpu_state->hvf_fd, HV_X86_R12); - env->regs[13] = rreg(cpu_state->hvf_fd, HV_X86_R13); - env->regs[14] = rreg(cpu_state->hvf_fd, HV_X86_R14); - env->regs[15] = rreg(cpu_state->hvf_fd, HV_X86_R15); + env->regs[R_EAX] = rreg(cpu_state->hvf->fd, HV_X86_RAX); + env->regs[R_EBX] = rreg(cpu_state->hvf->fd, HV_X86_RBX); + env->regs[R_ECX] = rreg(cpu_state->hvf->fd, HV_X86_RCX); + env->regs[R_EDX] = rreg(cpu_state->hvf->fd, HV_X86_RDX); + env->regs[R_EBP] = rreg(cpu_state->hvf->fd, HV_X86_RBP); + env->regs[R_ESP] = rreg(cpu_state->hvf->fd, HV_X86_RSP); + env->regs[R_ESI] = rreg(cpu_state->hvf->fd, HV_X86_RSI); + env->regs[R_EDI] = rreg(cpu_state->hvf->fd, HV_X86_RDI); + env->regs[8] = rreg(cpu_state->hvf->fd, HV_X86_R8); + env->regs[9] = rreg(cpu_state->hvf->fd, HV_X86_R9); + env->regs[10] = rreg(cpu_state->hvf->fd, HV_X86_R10); + env->regs[11] = rreg(cpu_state->hvf->fd, HV_X86_R11); + env->regs[12] = rreg(cpu_state->hvf->fd, HV_X86_R12); + env->regs[13] = rreg(cpu_state->hvf->fd, HV_X86_R13); + env->regs[14] = rreg(cpu_state->hvf->fd, HV_X86_R14); + env->regs[15] = rreg(cpu_state->hvf->fd, HV_X86_R15); - env->eflags = rreg(cpu_state->hvf_fd, HV_X86_RFLAGS); - env->eip = rreg(cpu_state->hvf_fd, HV_X86_RIP); + env->eflags = rreg(cpu_state->hvf->fd, HV_X86_RFLAGS); + env->eip = rreg(cpu_state->hvf->fd, HV_X86_RIP); hvf_get_xsave(cpu_state); - env->xcr0 = rreg(cpu_state->hvf_fd, HV_X86_XCR0); + env->xcr0 = rreg(cpu_state->hvf->fd, HV_X86_XCR0); hvf_get_segments(cpu_state); hvf_get_msrs(cpu_state); - env->dr[0] = rreg(cpu_state->hvf_fd, HV_X86_DR0); - env->dr[1] = rreg(cpu_state->hvf_fd, HV_X86_DR1); - env->dr[2] = rreg(cpu_state->hvf_fd, HV_X86_DR2); - env->dr[3] = rreg(cpu_state->hvf_fd, HV_X86_DR3); - env->dr[4] = rreg(cpu_state->hvf_fd, HV_X86_DR4); - env->dr[5] = rreg(cpu_state->hvf_fd, HV_X86_DR5); - env->dr[6] = rreg(cpu_state->hvf_fd, HV_X86_DR6); - env->dr[7] = rreg(cpu_state->hvf_fd, HV_X86_DR7); + env->dr[0] = rreg(cpu_state->hvf->fd, HV_X86_DR0); + env->dr[1] = rreg(cpu_state->hvf->fd, HV_X86_DR1); + env->dr[2] = rreg(cpu_state->hvf->fd, HV_X86_DR2); + env->dr[3] = rreg(cpu_state->hvf->fd, HV_X86_DR3); + env->dr[4] = rreg(cpu_state->hvf->fd, HV_X86_DR4); + env->dr[5] = rreg(cpu_state->hvf->fd, HV_X86_DR5); + env->dr[6] = rreg(cpu_state->hvf->fd, HV_X86_DR6); + env->dr[7] = rreg(cpu_state->hvf->fd, HV_X86_DR7); x86_update_hflags(env); return 0; @@ -XXX,XX +XXX,XX @@ int hvf_get_registers(CPUState *cpu_state) static void vmx_set_int_window_exiting(CPUState *cpu) { uint64_t val; - val = rvmcs(cpu->hvf_fd, VMCS_PRI_PROC_BASED_CTLS); - wvmcs(cpu->hvf_fd, VMCS_PRI_PROC_BASED_CTLS, val | + val = rvmcs(cpu->hvf->fd, VMCS_PRI_PROC_BASED_CTLS); + wvmcs(cpu->hvf->fd, VMCS_PRI_PROC_BASED_CTLS, val | VMCS_PRI_PROC_BASED_CTLS_INT_WINDOW_EXITING); } void vmx_clear_int_window_exiting(CPUState *cpu) { uint64_t val; - val = rvmcs(cpu->hvf_fd, VMCS_PRI_PROC_BASED_CTLS); - wvmcs(cpu->hvf_fd, VMCS_PRI_PROC_BASED_CTLS, val & + val = rvmcs(cpu->hvf->fd, VMCS_PRI_PROC_BASED_CTLS); + wvmcs(cpu->hvf->fd, VMCS_PRI_PROC_BASED_CTLS, val & ~VMCS_PRI_PROC_BASED_CTLS_INT_WINDOW_EXITING); } @@ -XXX,XX +XXX,XX @@ bool hvf_inject_interrupts(CPUState *cpu_state) uint64_t info = 0; if (have_event) { info = vector | intr_type | VMCS_INTR_VALID; - uint64_t reason = rvmcs(cpu_state->hvf_fd, VMCS_EXIT_REASON); + uint64_t reason = rvmcs(cpu_state->hvf->fd, VMCS_EXIT_REASON); if (env->nmi_injected && reason != EXIT_REASON_TASK_SWITCH) { vmx_clear_nmi_blocking(cpu_state); } @@ -XXX,XX +XXX,XX @@ bool hvf_inject_interrupts(CPUState *cpu_state) info &= ~(1 << 12); /* clear undefined bit */ if (intr_type == VMCS_INTR_T_SWINTR || intr_type == VMCS_INTR_T_SWEXCEPTION) { - wvmcs(cpu_state->hvf_fd, VMCS_ENTRY_INST_LENGTH, env->ins_len); + wvmcs(cpu_state->hvf->fd, VMCS_ENTRY_INST_LENGTH, env->ins_len); } if (env->has_error_code) { - wvmcs(cpu_state->hvf_fd, VMCS_ENTRY_EXCEPTION_ERROR, + wvmcs(cpu_state->hvf->fd, VMCS_ENTRY_EXCEPTION_ERROR, env->error_code); /* Indicate that VMCS_ENTRY_EXCEPTION_ERROR is valid */ info |= VMCS_INTR_DEL_ERRCODE; } /*printf("reinject %lx err %d\n", info, err);*/ - wvmcs(cpu_state->hvf_fd, VMCS_ENTRY_INTR_INFO, info); + wvmcs(cpu_state->hvf->fd, VMCS_ENTRY_INTR_INFO, info); }; } @@ -XXX,XX +XXX,XX @@ bool hvf_inject_interrupts(CPUState *cpu_state) if (!(env->hflags2 & HF2_NMI_MASK) && !(info & VMCS_INTR_VALID)) { cpu_state->interrupt_request &= ~CPU_INTERRUPT_NMI; info = VMCS_INTR_VALID | VMCS_INTR_T_NMI | EXCP02_NMI; - wvmcs(cpu_state->hvf_fd, VMCS_ENTRY_INTR_INFO, info); + wvmcs(cpu_state->hvf->fd, VMCS_ENTRY_INTR_INFO, info); } else { vmx_set_nmi_window_exiting(cpu_state); } @@ -XXX,XX +XXX,XX @@ bool hvf_inject_interrupts(CPUState *cpu_state) int line = cpu_get_pic_interrupt(&x86cpu->env); cpu_state->interrupt_request &= ~CPU_INTERRUPT_HARD; if (line >= 0) { - wvmcs(cpu_state->hvf_fd, VMCS_ENTRY_INTR_INFO, line | + wvmcs(cpu_state->hvf->fd, VMCS_ENTRY_INTR_INFO, line | VMCS_INTR_VALID | VMCS_INTR_T_HWINTR); } } @@ -XXX,XX +XXX,XX @@ int hvf_process_events(CPUState *cpu_state) X86CPU *cpu = X86_CPU(cpu_state); CPUX86State *env = &cpu->env; - env->eflags = rreg(cpu_state->hvf_fd, HV_X86_RFLAGS); + env->eflags = rreg(cpu_state->hvf->fd, HV_X86_RFLAGS); if (cpu_state->interrupt_request & CPU_INTERRUPT_INIT) { cpu_synchronize_state(cpu_state); -- 2.20.1 From: Alexander Graf <agraf@csgraf.de> The hooks we have that call us after reset, init and loadvm really all just want to say "The reference of all register state is in the QEMU vcpu struct, please push it". We already have a working pushing mechanism though called cpu->vcpu_dirty, so we can just reuse that for all of the above, syncing state properly the next time we actually execute a vCPU. This fixes PSCI resets on ARM, as they modify CPU state even after the post init call has completed, but before we execute the vCPU again. To also make the scheme work for x86, we have to make sure we don't move stale eflags into our env when the vcpu state is dirty. Signed-off-by: Alexander Graf <agraf@csgraf.de> Reviewed-by: Roman Bolshakov <r.bolshakov@yadro.com> Tested-by: Roman Bolshakov <r.bolshakov@yadro.com> Reviewed-by: Sergio Lopez <slp@redhat.com> Message-id: 20210519202253.76782-13-agraf@csgraf.de Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- accel/hvf/hvf-accel-ops.c | 27 +++++++-------------------- target/i386/hvf/x86hvf.c | 5 ++++- 2 files changed, 11 insertions(+), 21 deletions(-) diff --git a/accel/hvf/hvf-accel-ops.c b/accel/hvf/hvf-accel-ops.c index XXXXXXX..XXXXXXX 100644 --- a/accel/hvf/hvf-accel-ops.c +++ b/accel/hvf/hvf-accel-ops.c @@ -XXX,XX +XXX,XX @@ static void hvf_cpu_synchronize_state(CPUState *cpu) } } -static void do_hvf_cpu_synchronize_post_reset(CPUState *cpu, - run_on_cpu_data arg) +static void do_hvf_cpu_synchronize_set_dirty(CPUState *cpu, + run_on_cpu_data arg) { - hvf_put_registers(cpu); - cpu->vcpu_dirty = false; + /* QEMU state is the reference, push it to HVF now and on next entry */ + cpu->vcpu_dirty = true; } static void hvf_cpu_synchronize_post_reset(CPUState *cpu) { - run_on_cpu(cpu, do_hvf_cpu_synchronize_post_reset, RUN_ON_CPU_NULL); -} - -static void do_hvf_cpu_synchronize_post_init(CPUState *cpu, - run_on_cpu_data arg) -{ - hvf_put_registers(cpu); - cpu->vcpu_dirty = false; + run_on_cpu(cpu, do_hvf_cpu_synchronize_set_dirty, RUN_ON_CPU_NULL); } static void hvf_cpu_synchronize_post_init(CPUState *cpu) { - run_on_cpu(cpu, do_hvf_cpu_synchronize_post_init, RUN_ON_CPU_NULL); -} - -static void do_hvf_cpu_synchronize_pre_loadvm(CPUState *cpu, - run_on_cpu_data arg) -{ - cpu->vcpu_dirty = true; + run_on_cpu(cpu, do_hvf_cpu_synchronize_set_dirty, RUN_ON_CPU_NULL); } static void hvf_cpu_synchronize_pre_loadvm(CPUState *cpu) { - run_on_cpu(cpu, do_hvf_cpu_synchronize_pre_loadvm, RUN_ON_CPU_NULL); + run_on_cpu(cpu, do_hvf_cpu_synchronize_set_dirty, RUN_ON_CPU_NULL); } static void hvf_set_dirty_tracking(MemoryRegionSection *section, bool on) diff --git a/target/i386/hvf/x86hvf.c b/target/i386/hvf/x86hvf.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/hvf/x86hvf.c +++ b/target/i386/hvf/x86hvf.c @@ -XXX,XX +XXX,XX @@ int hvf_process_events(CPUState *cpu_state) X86CPU *cpu = X86_CPU(cpu_state); CPUX86State *env = &cpu->env; - env->eflags = rreg(cpu_state->hvf->fd, HV_X86_RFLAGS); + if (!cpu_state->vcpu_dirty) { + /* light weight sync for CPU_INTERRUPT_HARD and IF_MASK */ + env->eflags = rreg(cpu_state->hvf->fd, HV_X86_RFLAGS); + } if (cpu_state->interrupt_request & CPU_INTERRUPT_INIT) { cpu_synchronize_state(cpu_state); -- 2.20.1 Coverity notes that we don't check for dup2() failing. Add some assertions so that if it does ever happen we get some indication. (This is similar to how we handle other "don't expect this syscall to fail" checks in this test code.) Fixes: Coverity CID 1432346 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Stefan Berger <stefanb@linux.ibm.com> Message-id: 20210525134458.6675-2-peter.maydell@linaro.org --- tests/qtest/bios-tables-test.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/tests/qtest/bios-tables-test.c b/tests/qtest/bios-tables-test.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/bios-tables-test.c +++ b/tests/qtest/bios-tables-test.c @@ -XXX,XX +XXX,XX @@ static void test_acpi_asl(test_data *data) exp_sdt->asl_file, sdt->asl_file); int out = dup(STDOUT_FILENO); int ret G_GNUC_UNUSED; + int dupret; - dup2(STDERR_FILENO, STDOUT_FILENO); + g_assert(out >= 0); + dupret = dup2(STDERR_FILENO, STDOUT_FILENO); + g_assert(dupret >= 0); ret = system(diff) ; - dup2(out, STDOUT_FILENO); + dupret = dup2(out, STDOUT_FILENO); + g_assert(dupret >= 0); close(out); g_free(diff); } -- 2.20.1 The e1000e_send_verify() test calls qemu_recv() but doesn't check that the call succeeded, which annoys Coverity. Add an explicit test check for the length of the data. (This is a test check, not a "we assume this syscall always succeeds", so we use g_assert_cmpint() rather than g_assert().) Fixes: Coverity CID 1432324 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Stefan Berger <stefanb@linux.ibm.com> Message-id: 20210525134458.6675-3-peter.maydell@linaro.org --- tests/qtest/e1000e-test.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/qtest/e1000e-test.c b/tests/qtest/e1000e-test.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/e1000e-test.c +++ b/tests/qtest/e1000e-test.c @@ -XXX,XX +XXX,XX @@ static void e1000e_send_verify(QE1000E *d, int *test_sockets, QGuestAllocator *a /* Check data sent to the backend */ ret = qemu_recv(test_sockets[0], &recv_len, sizeof(recv_len), 0); g_assert_cmpint(ret, == , sizeof(recv_len)); - qemu_recv(test_sockets[0], buffer, 64, 0); + ret = qemu_recv(test_sockets[0], buffer, 64, 0); + g_assert_cmpint(ret, >=, 5); g_assert_cmpstr(buffer, == , "TEST"); /* Free test data buffer */ -- 2.20.1 Coverity notices that the checks against mkstemp() failing in create_qcow2_with_mbr() are wrong: mkstemp returns -1 on failure but the check is just "g_assert(fd)". Fix to use "g_assert(fd >= 0)", matching the correct check in create_test_img(). Fixes: Coverity CID 1432274 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Stefan Berger <stefanb@linux.ibm.com> Message-id: 20210525134458.6675-4-peter.maydell@linaro.org --- tests/qtest/hd-geo-test.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/qtest/hd-geo-test.c b/tests/qtest/hd-geo-test.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/hd-geo-test.c +++ b/tests/qtest/hd-geo-test.c @@ -XXX,XX +XXX,XX @@ static char *create_qcow2_with_mbr(MBRpartitions mbr, uint64_t sectors) } fd = mkstemp(raw_path); - g_assert(fd); + g_assert(fd >= 0); close(fd); fd = open(raw_path, O_WRONLY); @@ -XXX,XX +XXX,XX @@ static char *create_qcow2_with_mbr(MBRpartitions mbr, uint64_t sectors) close(fd); fd = mkstemp(qcow2_path); - g_assert(fd); + g_assert(fd >= 0); close(fd); qemu_img_path = getenv("QTEST_QEMU_IMG"); -- 2.20.1 Coverity points out that we calculate a 64-bit value using 32-bit arithmetic; add the cast to force the multiply to be done as 64-bits. (The overflow will never happen with the current test data.) Fixes: Coverity CID 1432320 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Stefan Berger <stefanb@linux.ibm.com> Message-id: 20210525134458.6675-5-peter.maydell@linaro.org --- tests/qtest/pflash-cfi02-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/qtest/pflash-cfi02-test.c b/tests/qtest/pflash-cfi02-test.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/pflash-cfi02-test.c +++ b/tests/qtest/pflash-cfi02-test.c @@ -XXX,XX +XXX,XX @@ static void test_geometry(const void *opaque) for (int region = 0; region < nb_erase_regions; ++region) { for (uint32_t i = 0; i < c->nb_blocs[region]; ++i) { - uint64_t byte_addr = i * c->sector_len[region]; + uint64_t byte_addr = (uint64_t)i * c->sector_len[region]; g_assert_cmphex(flash_read(c, byte_addr), ==, bank_mask(c)); } } -- 2.20.1 Coverity points out that in tpm_test_swtpm_migration_test() we assume that src_tpm_addr and dst_tpm_addr are non-NULL (we pass them to tpm_util_migration_start_qemu() which will unconditionally dereference them) but then later explicitly check them for NULL. Remove the pointless checks. Fixes: Coverity CID 1432367, 1432359 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Stefan Berger <stefanb@linux.ibm.com> Message-id: 20210525134458.6675-6-peter.maydell@linaro.org --- tests/qtest/tpm-tests.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/tests/qtest/tpm-tests.c b/tests/qtest/tpm-tests.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/tpm-tests.c +++ b/tests/qtest/tpm-tests.c @@ -XXX,XX +XXX,XX @@ void tpm_test_swtpm_migration_test(const char *src_tpm_path, qtest_quit(src_qemu); tpm_util_swtpm_kill(dst_tpm_pid); - if (dst_tpm_addr) { - g_unlink(dst_tpm_addr->u.q_unix.path); - qapi_free_SocketAddress(dst_tpm_addr); - } + g_unlink(dst_tpm_addr->u.q_unix.path); + qapi_free_SocketAddress(dst_tpm_addr); tpm_util_swtpm_kill(src_tpm_pid); - if (src_tpm_addr) { - g_unlink(src_tpm_addr->u.q_unix.path); - qapi_free_SocketAddress(src_tpm_addr); - } + g_unlink(src_tpm_addr->u.q_unix.path); + qapi_free_SocketAddress(src_tpm_addr); } -- 2.20.1 Coverity complains that we don't check for failures from dup() and mkstemp(); add asserts that these syscalls succeeded. Fixes: Coverity CID 1432516, 1432574 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Stefan Berger <stefanb@linux.ibm.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20210525134458.6675-7-peter.maydell@linaro.org --- tests/unit/test-vmstate.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/tests/unit/test-vmstate.c b/tests/unit/test-vmstate.c index XXXXXXX..XXXXXXX 100644 --- a/tests/unit/test-vmstate.c +++ b/tests/unit/test-vmstate.c @@ -XXX,XX +XXX,XX @@ static int temp_fd; /* Duplicate temp_fd and seek to the beginning of the file */ static QEMUFile *open_test_file(bool write) { - int fd = dup(temp_fd); + int fd; QIOChannel *ioc; QEMUFile *f; + fd = dup(temp_fd); + g_assert(fd >= 0); lseek(fd, 0, SEEK_SET); if (write) { g_assert_cmpint(ftruncate(fd, 0), ==, 0); @@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv) g_autofree char *temp_file = g_strdup_printf("%s/vmst.test.XXXXXX", g_get_tmp_dir()); temp_fd = mkstemp(temp_file); + g_assert(temp_fd >= 0); module_call_init(MODULE_INIT_QOM); -- 2.20.1 target-arm queue: the big stuff here is the final part of rth's patches for Cortex-A76 and Neoverse-N1 support; also present are Gavin's NUMA series and a few other things. thanks -- PMM The following changes since commit 554623226f800acf48a2ed568900c1c968ec9a8b: Merge tag 'qemu-sparc-20220508' of https://github.com/mcayland/qemu into staging (2022-05-08 17:03:26 -0500) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20220509 for you to fetch changes up to ae9141d4a3265553503bf07d3574b40f84615a34: hw/acpi/aml-build: Use existing CPU topology to build PPTT table (2022-05-09 11:47:55 +0100) ---------------------------------------------------------------- target-arm queue: * MAINTAINERS/.mailmap: update email for Leif Lindholm * hw/arm: add version information to sbsa-ref machine DT * Enable new features for -cpu max: FEAT_Debugv8p2, FEAT_Debugv8p4, FEAT_RAS (minimal version only), FEAT_IESB, FEAT_CSV2, FEAT_CSV2_2, FEAT_CSV3, FEAT_DGH * Emulate Cortex-A76 * Emulate Neoverse-N1 * Fix the virt board default NUMA topology ---------------------------------------------------------------- Gavin Shan (6): qapi/machine.json: Add cluster-id qtest/numa-test: Specify CPU topology in aarch64_numa_cpu() hw/arm/virt: Consider SMP configuration in CPU topology qtest/numa-test: Correct CPU and NUMA association in aarch64_numa_cpu() hw/arm/virt: Fix CPU's default NUMA node ID hw/acpi/aml-build: Use existing CPU topology to build PPTT table Leif Lindholm (2): MAINTAINERS/.mailmap: update email for Leif Lindholm hw/arm: add versioning to sbsa-ref machine DT Richard Henderson (24): target/arm: Handle cpreg registration for missing EL target/arm: Drop EL3 no EL2 fallbacks target/arm: Merge zcr reginfo target/arm: Adjust definition of CONTEXTIDR_EL2 target/arm: Move cortex impdef sysregs to cpu_tcg.c target/arm: Update qemu-system-arm -cpu max to cortex-a57 target/arm: Set ID_DFR0.PerfMon for qemu-system-arm -cpu max target/arm: Split out aa32_max_features target/arm: Annotate arm_max_initfn with FEAT identifiers target/arm: Use field names for manipulating EL2 and EL3 modes target/arm: Enable FEAT_Debugv8p2 for -cpu max target/arm: Enable FEAT_Debugv8p4 for -cpu max target/arm: Add minimal RAS registers target/arm: Enable SCR and HCR bits for RAS target/arm: Implement virtual SError exceptions target/arm: Implement ESB instruction target/arm: Enable FEAT_RAS for -cpu max target/arm: Enable FEAT_IESB for -cpu max target/arm: Enable FEAT_CSV2 for -cpu max target/arm: Enable FEAT_CSV2_2 for -cpu max target/arm: Enable FEAT_CSV3 for -cpu max target/arm: Enable FEAT_DGH for -cpu max target/arm: Define cortex-a76 target/arm: Define neoverse-n1 docs/system/arm/emulation.rst | 10 + docs/system/arm/virt.rst | 2 + qapi/machine.json | 6 +- target/arm/cpregs.h | 11 + target/arm/cpu.h | 23 ++ target/arm/helper.h | 1 + target/arm/internals.h | 16 ++ target/arm/syndrome.h | 5 + target/arm/a32.decode | 16 +- target/arm/t32.decode | 18 +- hw/acpi/aml-build.c | 111 ++++---- hw/arm/sbsa-ref.c | 16 ++ hw/arm/virt.c | 21 +- hw/core/machine-hmp-cmds.c | 4 + hw/core/machine.c | 16 ++ target/arm/cpu.c | 66 ++++- target/arm/cpu64.c | 353 ++++++++++++++----------- target/arm/cpu_tcg.c | 227 +++++++++++----- target/arm/helper.c | 600 +++++++++++++++++++++++++----------------- target/arm/op_helper.c | 43 +++ target/arm/translate-a64.c | 18 ++ target/arm/translate.c | 23 ++ tests/qtest/numa-test.c | 19 +- .mailmap | 3 +- MAINTAINERS | 2 +- 25 files changed, 1068 insertions(+), 562 deletions(-) From: Leif Lindholm <quic_llindhol@quicinc.com> NUVIA was acquired by Qualcomm in March 2021, but kept functioning on separate infrastructure for a transitional period. We've now switched over to contributing as Qualcomm Innovation Center (quicinc), so update my email address to reflect this. Signed-off-by: Leif Lindholm <quic_llindhol@quicinc.com> Message-id: 20220505113740.75565-1-quic_llindhol@quicinc.com Cc: Leif Lindholm <leif@nuviainc.com> Cc: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> [Fixed commit message typo] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- .mailmap | 3 ++- MAINTAINERS | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/.mailmap b/.mailmap index XXXXXXX..XXXXXXX 100644 --- a/.mailmap +++ b/.mailmap @@ -XXX,XX +XXX,XX @@ Greg Kurz <groug@kaod.org> <gkurz@linux.vnet.ibm.com> Huacai Chen <chenhuacai@kernel.org> <chenhc@lemote.com> Huacai Chen <chenhuacai@kernel.org> <chenhuacai@loongson.cn> James Hogan <jhogan@kernel.org> <james.hogan@imgtec.com> -Leif Lindholm <leif@nuviainc.com> <leif.lindholm@linaro.org> +Leif Lindholm <quic_llindhol@quicinc.com> <leif.lindholm@linaro.org> +Leif Lindholm <quic_llindhol@quicinc.com> <leif@nuviainc.com> Radoslaw Biernacki <rad@semihalf.com> <radoslaw.biernacki@linaro.org> Paul Burton <paulburton@kernel.org> <paul.burton@mips.com> Paul Burton <paulburton@kernel.org> <paul.burton@imgtec.com> diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ F: include/hw/ssi/imx_spi.h SBSA-REF M: Radoslaw Biernacki <rad@semihalf.com> M: Peter Maydell <peter.maydell@linaro.org> -R: Leif Lindholm <leif@nuviainc.com> +R: Leif Lindholm <quic_llindhol@quicinc.com> L: qemu-arm@nongnu.org S: Maintained F: hw/arm/sbsa-ref.c -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> More gracefully handle cpregs when EL2 and/or EL3 are missing. If the reg is entirely inaccessible, do not register it at all. If the reg is for EL2, and EL3 is present but EL2 is not, either discard, squash to res0, const, or keep unchanged. Per rule RJFFP, mark the 4 aarch32 hypervisor access registers with ARM_CP_EL3_NO_EL2_KEEP, and mark all of the EL2 address translation and tlb invalidation "regs" ARM_CP_EL3_NO_EL2_UNDEF. Mark the 2 virtualization processor id regs ARM_CP_EL3_NO_EL2_C_NZ. This will simplify cpreg registration for conditional arm features. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-2-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpregs.h | 11 +++ target/arm/helper.c | 178 ++++++++++++++++++++++++++++++-------------- 2 files changed, 133 insertions(+), 56 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 @@ enum { ARM_CP_SVE = 1 << 14, /* Flag: Do not expose in gdb sysreg xml. */ ARM_CP_NO_GDB = 1 << 15, + /* + * Flags: If EL3 but not EL2... + * - UNDEF: discard the cpreg, + * - KEEP: retain the cpreg as is, + * - C_NZ: set const on the cpreg, but retain resetvalue, + * - else: set const on the cpreg, zero resetvalue, aka RES0. + * See rule RJFFP in section D1.1.3 of DDI0487H.a. + */ + ARM_CP_EL3_NO_EL2_UNDEF = 1 << 16, + ARM_CP_EL3_NO_EL2_KEEP = 1 << 17, + ARM_CP_EL3_NO_EL2_C_NZ = 1 << 18, }; /* 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 v8_cp_reginfo[] = { .access = PL1_RW, .readfn = spsel_read, .writefn = spsel_write }, { .name = "FPEXC32_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 3, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_ALIAS | ARM_CP_FPU, + .access = PL2_RW, + .type = ARM_CP_ALIAS | ARM_CP_FPU | ARM_CP_EL3_NO_EL2_KEEP, .fieldoffset = offsetof(CPUARMState, vfp.xregs[ARM_VFP_FPEXC]) }, { .name = "DACR32_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 3, .crm = 0, .opc2 = 0, - .access = PL2_RW, .resetvalue = 0, + .access = PL2_RW, .resetvalue = 0, .type = ARM_CP_EL3_NO_EL2_KEEP, .writefn = dacr_write, .raw_writefn = raw_write, .fieldoffset = offsetof(CPUARMState, cp15.dacr32_el2) }, { .name = "IFSR32_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 0, .opc2 = 1, - .access = PL2_RW, .resetvalue = 0, + .access = PL2_RW, .resetvalue = 0, .type = ARM_CP_EL3_NO_EL2_KEEP, .fieldoffset = offsetof(CPUARMState, cp15.ifsr32_el2) }, { .name = "SPSR_IRQ", .state = ARM_CP_STATE_AA64, .type = ARM_CP_ALIAS, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_cp_reginfo[] = { .writefn = tlbimva_hyp_is_write }, { .name = "TLBI_ALLE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 7, .opc2 = 0, - .type = ARM_CP_NO_RAW, .access = PL2_W, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_alle2_write }, { .name = "TLBI_VAE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 7, .opc2 = 1, - .type = ARM_CP_NO_RAW, .access = PL2_W, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2_write }, { .name = "TLBI_VALE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 7, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2_write }, { .name = "TLBI_ALLE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 3, .opc2 = 0, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_alle2is_write }, { .name = "TLBI_VAE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 3, .opc2 = 1, - .type = ARM_CP_NO_RAW, .access = PL2_W, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2is_write }, { .name = "TLBI_VALE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 3, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2is_write }, #ifndef CONFIG_USER_ONLY /* Unlike the other EL2-related AT operations, these must @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_cp_reginfo[] = { { .name = "AT_S1E2R", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 7, .crm = 8, .opc2 = 0, .access = PL2_W, .accessfn = at_s1e2_access, - .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC, .writefn = ats_write64 }, + .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC | ARM_CP_EL3_NO_EL2_UNDEF, + .writefn = ats_write64 }, { .name = "AT_S1E2W", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 7, .crm = 8, .opc2 = 1, .access = PL2_W, .accessfn = at_s1e2_access, - .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC, .writefn = ats_write64 }, + .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC | ARM_CP_EL3_NO_EL2_UNDEF, + .writefn = ats_write64 }, /* The AArch32 ATS1H* operations are CONSTRAINED UNPREDICTABLE * if EL2 is not implemented; we choose to UNDEF. Behaviour at EL3 * with SCR.NS == 0 outside Monitor mode is UNPREDICTABLE; we choose @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo debug_cp_reginfo[] = { { .name = "DBGVCR32_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 2, .opc1 = 4, .crn = 0, .crm = 7, .opc2 = 0, .access = PL2_RW, .accessfn = access_tda, - .type = ARM_CP_NOP }, + .type = ARM_CP_NOP | ARM_CP_EL3_NO_EL2_KEEP }, /* Dummy MDCCINT_EL1, since we don't implement the Debug Communications * Channel but Linux may try to access this register. The 32-bit * alias is DBGDCCINT. @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbirange_reginfo[] = { .access = PL2_W, .type = ARM_CP_NOP }, { .name = "TLBI_RVAE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 2, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2is_write }, { .name = "TLBI_RVALE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 2, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2is_write }, { .name = "TLBI_RIPAS2E1", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 2, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbirange_reginfo[] = { .access = PL2_W, .type = ARM_CP_NOP }, { .name = "TLBI_RVAE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 5, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2is_write }, { .name = "TLBI_RVALE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 5, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2is_write }, { .name = "TLBI_RVAE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 6, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2_write }, { .name = "TLBI_RVALE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 6, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2_write }, { .name = "TLBI_RVAE3IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 6, .crn = 8, .crm = 2, .opc2 = 1, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbios_reginfo[] = { .writefn = tlbi_aa64_vae1is_write }, { .name = "TLBI_ALLE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 0, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_alle2is_write }, { .name = "TLBI_VAE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2is_write }, { .name = "TLBI_ALLE1OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 4, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbios_reginfo[] = { .writefn = tlbi_aa64_alle1is_write }, { .name = "TLBI_VALE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2is_write }, { .name = "TLBI_VMALLS12E1OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 6, @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) { .name = "VPIDR", .state = ARM_CP_STATE_AA32, .cp = 15, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 0, .access = PL2_RW, .accessfn = access_el3_aa32ns, - .resetvalue = cpu->midr, .type = ARM_CP_ALIAS, + .resetvalue = cpu->midr, + .type = ARM_CP_ALIAS | ARM_CP_EL3_NO_EL2_C_NZ, .fieldoffset = offsetoflow32(CPUARMState, cp15.vpidr_el2) }, { .name = "VPIDR_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 0, .access = PL2_RW, .resetvalue = cpu->midr, + .type = ARM_CP_EL3_NO_EL2_C_NZ, .fieldoffset = offsetof(CPUARMState, cp15.vpidr_el2) }, { .name = "VMPIDR", .state = ARM_CP_STATE_AA32, .cp = 15, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 5, .access = PL2_RW, .accessfn = access_el3_aa32ns, - .resetvalue = vmpidr_def, .type = ARM_CP_ALIAS, + .resetvalue = vmpidr_def, + .type = ARM_CP_ALIAS | ARM_CP_EL3_NO_EL2_C_NZ, .fieldoffset = offsetoflow32(CPUARMState, cp15.vmpidr_el2) }, { .name = "VMPIDR_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 5, - .access = PL2_RW, - .resetvalue = vmpidr_def, + .access = PL2_RW, .resetvalue = vmpidr_def, + .type = ARM_CP_EL3_NO_EL2_C_NZ, .fieldoffset = offsetof(CPUARMState, cp15.vmpidr_el2) }, }; define_arm_cp_regs(cpu, vpidr_regs); @@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r, int crm, int opc1, int opc2, const char *name) { + CPUARMState *env = &cpu->env; uint32_t key; ARMCPRegInfo *r2; bool is64 = r->type & ARM_CP_64BIT; bool ns = secstate & ARM_CP_SECSTATE_NS; int cp = r->cp; - bool isbanked; size_t name_len; + bool make_const; switch (state) { case ARM_CP_STATE_AA32: @@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r, } } + /* + * Eliminate registers that are not present because the EL is missing. + * Doing this here makes it easier to put all registers for a given + * feature into the same ARMCPRegInfo array and define them all at once. + */ + make_const = false; + if (arm_feature(env, ARM_FEATURE_EL3)) { + /* + * An EL2 register without EL2 but with EL3 is (usually) RES0. + * See rule RJFFP in section D1.1.3 of DDI0487H.a. + */ + int min_el = ctz32(r->access) / 2; + if (min_el == 2 && !arm_feature(env, ARM_FEATURE_EL2)) { + if (r->type & ARM_CP_EL3_NO_EL2_UNDEF) { + return; + } + make_const = !(r->type & ARM_CP_EL3_NO_EL2_KEEP); + } + } else { + CPAccessRights max_el = (arm_feature(env, ARM_FEATURE_EL2) + ? PL2_RW : PL1_RW); + if ((r->access & max_el) == 0) { + return; + } + } + /* Combine cpreg and name into one allocation. */ name_len = strlen(name) + 1; r2 = g_malloc(sizeof(*r2) + name_len); @@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r, r2->opaque = opaque; } - isbanked = r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1]; - if (isbanked) { + if (make_const) { + /* This should not have been a very special register to begin. */ + int old_special = r2->type & ARM_CP_SPECIAL_MASK; + assert(old_special == 0 || old_special == ARM_CP_NOP); /* - * Register is banked (using both entries in array). - * Overwriting fieldoffset as the array is only used to define - * banked registers but later only fieldoffset is used. + * Set the special function to CONST, retaining the other flags. + * This is important for e.g. ARM_CP_SVE so that we still + * take the SVE trap if CPTR_EL3.EZ == 0. */ - r2->fieldoffset = r->bank_fieldoffsets[ns]; - } + r2->type = (r2->type & ~ARM_CP_SPECIAL_MASK) | ARM_CP_CONST; + /* + * Usually, these registers become RES0, but there are a few + * special cases like VPIDR_EL2 which have a constant non-zero + * value with writes ignored. + */ + if (!(r->type & ARM_CP_EL3_NO_EL2_C_NZ)) { + r2->resetvalue = 0; + } + /* + * ARM_CP_CONST has precedence, so removing the callbacks and + * offsets are not strictly necessary, but it is potentially + * less confusing to debug later. + */ + r2->readfn = NULL; + r2->writefn = NULL; + r2->raw_readfn = NULL; + r2->raw_writefn = NULL; + r2->resetfn = NULL; + r2->fieldoffset = 0; + r2->bank_fieldoffsets[0] = 0; + r2->bank_fieldoffsets[1] = 0; + } else { + bool isbanked = r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1]; - if (state == ARM_CP_STATE_AA32) { if (isbanked) { /* - * If the register is banked then we don't need to migrate or - * reset the 32-bit instance in certain cases: - * - * 1) If the register has both 32-bit and 64-bit instances then we - * can count on the 64-bit instance taking care of the - * non-secure bank. - * 2) If ARMv8 is enabled then we can count on a 64-bit version - * taking care of the secure bank. This requires that separate - * 32 and 64-bit definitions are provided. + * Register is banked (using both entries in array). + * Overwriting fieldoffset as the array is only used to define + * banked registers but later only fieldoffset is used. */ - if ((r->state == ARM_CP_STATE_BOTH && ns) || - (arm_feature(&cpu->env, ARM_FEATURE_V8) && !ns)) { + r2->fieldoffset = r->bank_fieldoffsets[ns]; + } + if (state == ARM_CP_STATE_AA32) { + if (isbanked) { + /* + * If the register is banked then we don't need to migrate or + * reset the 32-bit instance in certain cases: + * + * 1) If the register has both 32-bit and 64-bit instances + * then we can count on the 64-bit instance taking care + * of the non-secure bank. + * 2) If ARMv8 is enabled then we can count on a 64-bit + * version taking care of the secure bank. This requires + * that separate 32 and 64-bit definitions are provided. + */ + if ((r->state == ARM_CP_STATE_BOTH && ns) || + (arm_feature(env, ARM_FEATURE_V8) && !ns)) { + r2->type |= ARM_CP_ALIAS; + } + } else if ((secstate != r->secure) && !ns) { + /* + * The register is not banked so we only want to allow + * migration of the non-secure instance. + */ r2->type |= ARM_CP_ALIAS; } - } else if ((secstate != r->secure) && !ns) { - /* - * The register is not banked so we only want to allow migration - * of the non-secure instance. - */ - r2->type |= ARM_CP_ALIAS; - } - if (HOST_BIG_ENDIAN && - r->state == ARM_CP_STATE_BOTH && r2->fieldoffset) { - r2->fieldoffset += sizeof(uint32_t); + if (HOST_BIG_ENDIAN && + r->state == ARM_CP_STATE_BOTH && r2->fieldoffset) { + r2->fieldoffset += sizeof(uint32_t); + } } } @@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r, * multiple times. Special registers (ie NOP/WFI) are * never migratable and not even raw-accessible. */ - if (r->type & ARM_CP_SPECIAL_MASK) { + if (r2->type & ARM_CP_SPECIAL_MASK) { r2->type |= ARM_CP_NO_RAW; } if (((r->crm == CP_ANY) && crm != 0) || -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Drop el3_no_el2_cp_reginfo, el3_no_el2_v8_cp_reginfo, and the local vpidr_regs definition, and rely on the squashing to ARM_CP_CONST while registering for v8. This is a behavior change for v7 cpus with Security Extensions and without Virtualization Extensions, in that the virtualization cpregs are now correctly not present. This would be a migration compatibility break, except that we have an existing bug in which migration of 32-bit cpus with Security Extensions enabled does not work. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-3-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 158 ++++---------------------------------------- 1 file changed, 13 insertions(+), 145 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 @@ static const ARMCPRegInfo v8_cp_reginfo[] = { .fieldoffset = offsetoflow32(CPUARMState, cp15.mdcr_el3) }, }; -/* Used to describe the behaviour of EL2 regs when EL2 does not exist. */ -static const ARMCPRegInfo el3_no_el2_cp_reginfo[] = { - { .name = "VBAR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 12, .crm = 0, .opc2 = 0, - .access = PL2_RW, - .readfn = arm_cp_read_zero, .writefn = arm_cp_write_ignore }, - { .name = "HCR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 0, - .access = PL2_RW, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HACR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 7, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "ESR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 2, .opc2 = 0, - .access = PL2_RW, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPTR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 2, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "MAIR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 10, .crm = 2, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "HMAIR1", .state = ARM_CP_STATE_AA32, - .cp = 15, .opc1 = 4, .crn = 10, .crm = 2, .opc2 = 1, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "AMAIR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 10, .crm = 3, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "HAMAIR1", .state = ARM_CP_STATE_AA32, - .cp = 15, .opc1 = 4, .crn = 10, .crm = 3, .opc2 = 1, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "AFSR0_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 1, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "AFSR1_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 1, .opc2 = 1, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "TCR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 0, .opc2 = 2, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "VTCR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 1, .opc2 = 2, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "VTTBR", .state = ARM_CP_STATE_AA32, - .cp = 15, .opc1 = 6, .crm = 2, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, - { .name = "VTTBR_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 1, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "SCTLR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 0, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "TPIDR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 13, .crm = 0, .opc2 = 2, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "TTBR0_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 0, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HTTBR", .cp = 15, .opc1 = 4, .crm = 2, - .access = PL2_RW, .type = ARM_CP_64BIT | ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "CNTHCTL_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 1, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CNTVOFF_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 0, .opc2 = 3, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CNTVOFF", .cp = 15, .opc1 = 4, .crm = 14, - .access = PL2_RW, .type = ARM_CP_64BIT | ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "CNTHP_CVAL_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 2, .opc2 = 2, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CNTHP_CVAL", .cp = 15, .opc1 = 6, .crm = 14, - .access = PL2_RW, .type = ARM_CP_64BIT | ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "CNTHP_TVAL_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 2, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CNTHP_CTL_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 2, .opc2 = 1, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "MDCR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 1, - .access = PL2_RW, .accessfn = access_tda, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HPFAR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 6, .crm = 0, .opc2 = 4, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HSTR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 3, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "FAR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 6, .crm = 0, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HIFAR", .state = ARM_CP_STATE_AA32, - .type = ARM_CP_CONST, - .cp = 15, .opc1 = 4, .crn = 6, .crm = 0, .opc2 = 2, - .access = PL2_RW, .resetvalue = 0 }, -}; - -/* Ditto, but for registers which exist in ARMv8 but not v7 */ -static const ARMCPRegInfo el3_no_el2_v8_cp_reginfo[] = { - { .name = "HCR2", .state = ARM_CP_STATE_AA32, - .cp = 15, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 4, - .access = PL2_RW, - .type = ARM_CP_CONST, .resetvalue = 0 }, -}; - static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask) { ARMCPU *cpu = env_archcpu(env); @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) define_arm_cp_regs(cpu, v8_idregs); define_arm_cp_regs(cpu, v8_cp_reginfo); } - if (arm_feature(env, ARM_FEATURE_EL2)) { + + /* + * Register the base EL2 cpregs. + * Pre v8, these registers are implemented only as part of the + * Virtualization Extensions (EL2 present). Beginning with v8, + * if EL2 is missing but EL3 is enabled, mostly these become + * RES0 from EL3, with some specific exceptions. + */ + if (arm_feature(env, ARM_FEATURE_EL2) + || (arm_feature(env, ARM_FEATURE_EL3) + && arm_feature(env, ARM_FEATURE_V8))) { uint64_t vmpidr_def = mpidr_read_val(env); ARMCPRegInfo vpidr_regs[] = { { .name = "VPIDR", .state = ARM_CP_STATE_AA32, @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) }; define_one_arm_cp_reg(cpu, &rvbar); } - } else { - /* If EL2 is missing but higher ELs are enabled, we need to - * register the no_el2 reginfos. - */ - if (arm_feature(env, ARM_FEATURE_EL3)) { - /* When EL3 exists but not EL2, VPIDR and VMPIDR take the value - * of MIDR_EL1 and MPIDR_EL1. - */ - ARMCPRegInfo vpidr_regs[] = { - { .name = "VPIDR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 0, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_CONST, .resetvalue = cpu->midr, - .fieldoffset = offsetof(CPUARMState, cp15.vpidr_el2) }, - { .name = "VMPIDR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 5, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_NO_RAW, - .writefn = arm_cp_write_ignore, .readfn = mpidr_read }, - }; - define_arm_cp_regs(cpu, vpidr_regs); - define_arm_cp_regs(cpu, el3_no_el2_cp_reginfo); - if (arm_feature(env, ARM_FEATURE_V8)) { - define_arm_cp_regs(cpu, el3_no_el2_v8_cp_reginfo); - } - } } + + /* Register the base EL3 cpregs. */ if (arm_feature(env, ARM_FEATURE_EL3)) { define_arm_cp_regs(cpu, el3_cp_reginfo); ARMCPRegInfo el3_regs[] = { -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Drop zcr_no_el2_reginfo and merge the 3 registers into one array, now that ZCR_EL2 can be squashed to RES0 and ZCR_EL3 dropped while registering. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-4-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 55 ++++++++++++++------------------------------- 1 file changed, 17 insertions(+), 38 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 @@ static void zcr_write(CPUARMState *env, const ARMCPRegInfo *ri, } } -static const ARMCPRegInfo zcr_el1_reginfo = { - .name = "ZCR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 0, .crn = 1, .crm = 2, .opc2 = 0, - .access = PL1_RW, .type = ARM_CP_SVE, - .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[1]), - .writefn = zcr_write, .raw_writefn = raw_write -}; - -static const ARMCPRegInfo zcr_el2_reginfo = { - .name = "ZCR_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 2, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_SVE, - .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[2]), - .writefn = zcr_write, .raw_writefn = raw_write -}; - -static const ARMCPRegInfo zcr_no_el2_reginfo = { - .name = "ZCR_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 2, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_SVE, - .readfn = arm_cp_read_zero, .writefn = arm_cp_write_ignore -}; - -static const ARMCPRegInfo zcr_el3_reginfo = { - .name = "ZCR_EL3", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 6, .crn = 1, .crm = 2, .opc2 = 0, - .access = PL3_RW, .type = ARM_CP_SVE, - .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[3]), - .writefn = zcr_write, .raw_writefn = raw_write +static const ARMCPRegInfo zcr_reginfo[] = { + { .name = "ZCR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 0, .crn = 1, .crm = 2, .opc2 = 0, + .access = PL1_RW, .type = ARM_CP_SVE, + .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[1]), + .writefn = zcr_write, .raw_writefn = raw_write }, + { .name = "ZCR_EL2", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 2, .opc2 = 0, + .access = PL2_RW, .type = ARM_CP_SVE, + .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[2]), + .writefn = zcr_write, .raw_writefn = raw_write }, + { .name = "ZCR_EL3", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 6, .crn = 1, .crm = 2, .opc2 = 0, + .access = PL3_RW, .type = ARM_CP_SVE, + .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[3]), + .writefn = zcr_write, .raw_writefn = raw_write }, }; void hw_watchpoint_update(ARMCPU *cpu, int n) @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) } if (cpu_isar_feature(aa64_sve, cpu)) { - define_one_arm_cp_reg(cpu, &zcr_el1_reginfo); - if (arm_feature(env, ARM_FEATURE_EL2)) { - define_one_arm_cp_reg(cpu, &zcr_el2_reginfo); - } else { - define_one_arm_cp_reg(cpu, &zcr_no_el2_reginfo); - } - if (arm_feature(env, ARM_FEATURE_EL3)) { - define_one_arm_cp_reg(cpu, &zcr_el3_reginfo); - } + define_arm_cp_regs(cpu, zcr_reginfo); } #ifdef TARGET_AARCH64 -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This register is present for either VHE or Debugv8p2. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-5-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 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 @@ static const ARMCPRegInfo jazelle_regs[] = { .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, }; +static const ARMCPRegInfo contextidr_el2 = { + .name = "CONTEXTIDR_EL2", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 4, .crn = 13, .crm = 0, .opc2 = 1, + .access = PL2_RW, + .fieldoffset = offsetof(CPUARMState, cp15.contextidr_el[2]) +}; + static const ARMCPRegInfo vhe_reginfo[] = { - { .name = "CONTEXTIDR_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 13, .crm = 0, .opc2 = 1, - .access = PL2_RW, - .fieldoffset = offsetof(CPUARMState, cp15.contextidr_el[2]) }, { .name = "TTBR1_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 0, .opc2 = 1, .access = PL2_RW, .writefn = vmsa_tcr_ttbr_el2_write, @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) define_one_arm_cp_reg(cpu, &ssbs_reginfo); } + if (cpu_isar_feature(aa64_vh, cpu) || + cpu_isar_feature(aa64_debugv8p2, cpu)) { + define_one_arm_cp_reg(cpu, &contextidr_el2); + } if (arm_feature(env, ARM_FEATURE_EL2) && cpu_isar_feature(aa64_vh, cpu)) { define_arm_cp_regs(cpu, vhe_reginfo); } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Previously we were defining some of these in user-only mode, but none of them are accessible from user-only, therefore define them only in system mode. This will shortly be used from cpu_tcg.c also. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-6-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/internals.h | 6 ++++ target/arm/cpu64.c | 64 +++--------------------------------------- target/arm/cpu_tcg.c | 59 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 69 insertions(+), 60 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 @@ int aarch64_fpu_gdb_get_reg(CPUARMState *env, GByteArray *buf, int reg); int aarch64_fpu_gdb_set_reg(CPUARMState *env, uint8_t *buf, int reg); #endif +#ifdef CONFIG_USER_ONLY +static inline void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu) { } +#else +void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu); +#endif + #endif diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ #include "hvf_arm.h" #include "qapi/visitor.h" #include "hw/qdev-properties.h" -#include "cpregs.h" +#include "internals.h" -#ifndef CONFIG_USER_ONLY -static uint64_t a57_a53_l2ctlr_read(CPUARMState *env, const ARMCPRegInfo *ri) -{ - ARMCPU *cpu = env_archcpu(env); - - /* Number of cores is in [25:24]; otherwise we RAZ */ - return (cpu->core_count - 1) << 24; -} -#endif - -static const ARMCPRegInfo cortex_a72_a57_a53_cp_reginfo[] = { -#ifndef CONFIG_USER_ONLY - { .name = "L2CTLR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 11, .crm = 0, .opc2 = 2, - .access = PL1_RW, .readfn = a57_a53_l2ctlr_read, - .writefn = arm_cp_write_ignore }, - { .name = "L2CTLR", - .cp = 15, .opc1 = 1, .crn = 9, .crm = 0, .opc2 = 2, - .access = PL1_RW, .readfn = a57_a53_l2ctlr_read, - .writefn = arm_cp_write_ignore }, -#endif - { .name = "L2ECTLR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 11, .crm = 0, .opc2 = 3, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "L2ECTLR", - .cp = 15, .opc1 = 1, .crn = 9, .crm = 0, .opc2 = 3, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "L2ACTLR", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 0, .opc2 = 0, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPUACTLR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 0, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPUACTLR", - .cp = 15, .opc1 = 0, .crm = 15, - .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, - { .name = "CPUECTLR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 1, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPUECTLR", - .cp = 15, .opc1 = 1, .crm = 15, - .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, - { .name = "CPUMERRSR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 2, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPUMERRSR", - .cp = 15, .opc1 = 2, .crm = 15, - .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, - { .name = "L2MERRSR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 3, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "L2MERRSR", - .cp = 15, .opc1 = 3, .crm = 15, - .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, -}; - static void aarch64_a57_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); @@ -XXX,XX +XXX,XX @@ static void aarch64_a57_initfn(Object *obj) cpu->gic_num_lrs = 4; cpu->gic_vpribits = 5; cpu->gic_vprebits = 5; - define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo); + define_cortex_a72_a57_a53_cp_reginfo(cpu); } static void aarch64_a53_initfn(Object *obj) @@ -XXX,XX +XXX,XX @@ static void aarch64_a53_initfn(Object *obj) cpu->gic_num_lrs = 4; cpu->gic_vpribits = 5; cpu->gic_vprebits = 5; - define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo); + define_cortex_a72_a57_a53_cp_reginfo(cpu); } static void aarch64_a72_initfn(Object *obj) @@ -XXX,XX +XXX,XX @@ static void aarch64_a72_initfn(Object *obj) cpu->gic_num_lrs = 4; cpu->gic_vpribits = 5; cpu->gic_vprebits = 5; - define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo); + define_cortex_a72_a57_a53_cp_reginfo(cpu); } void arm_cpu_sve_finalize(ARMCPU *cpu, Error **errp) diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ #endif #include "cpregs.h" +#ifndef CONFIG_USER_ONLY +static uint64_t l2ctlr_read(CPUARMState *env, const ARMCPRegInfo *ri) +{ + ARMCPU *cpu = env_archcpu(env); + + /* Number of cores is in [25:24]; otherwise we RAZ */ + return (cpu->core_count - 1) << 24; +} + +static const ARMCPRegInfo cortex_a72_a57_a53_cp_reginfo[] = { + { .name = "L2CTLR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 11, .crm = 0, .opc2 = 2, + .access = PL1_RW, .readfn = l2ctlr_read, + .writefn = arm_cp_write_ignore }, + { .name = "L2CTLR", + .cp = 15, .opc1 = 1, .crn = 9, .crm = 0, .opc2 = 2, + .access = PL1_RW, .readfn = l2ctlr_read, + .writefn = arm_cp_write_ignore }, + { .name = "L2ECTLR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 11, .crm = 0, .opc2 = 3, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "L2ECTLR", + .cp = 15, .opc1 = 1, .crn = 9, .crm = 0, .opc2 = 3, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "L2ACTLR", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 0, .opc2 = 0, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "CPUACTLR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 0, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "CPUACTLR", + .cp = 15, .opc1 = 0, .crm = 15, + .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, + { .name = "CPUECTLR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 1, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "CPUECTLR", + .cp = 15, .opc1 = 1, .crm = 15, + .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, + { .name = "CPUMERRSR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 2, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "CPUMERRSR", + .cp = 15, .opc1 = 2, .crm = 15, + .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, + { .name = "L2MERRSR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 3, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "L2MERRSR", + .cp = 15, .opc1 = 3, .crm = 15, + .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, +}; + +void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu) +{ + define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo); +} +#endif /* !CONFIG_USER_ONLY */ + /* CPU models. These are not needed for the AArch64 linux-user build. */ #if !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64) -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Instead of starting with cortex-a15 and adding v8 features to a v7 cpu, begin with a v8 cpu stripped of its aarch64 features. This fixes the long-standing to-do where we only enabled v8 features for user-only. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-7-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu_tcg.c | 151 ++++++++++++++++++++++++++----------------- 1 file changed, 92 insertions(+), 59 deletions(-) diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ static void arm_v7m_class_init(ObjectClass *oc, void *data) static void arm_max_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); + uint32_t t; - cortex_a15_initfn(obj); + /* aarch64_a57_initfn, advertising none of the aarch64 features */ + cpu->dtb_compatible = "arm,cortex-a57"; + set_feature(&cpu->env, ARM_FEATURE_V8); + set_feature(&cpu->env, ARM_FEATURE_NEON); + set_feature(&cpu->env, ARM_FEATURE_GENERIC_TIMER); + set_feature(&cpu->env, ARM_FEATURE_CBAR_RO); + set_feature(&cpu->env, ARM_FEATURE_EL2); + set_feature(&cpu->env, ARM_FEATURE_EL3); + set_feature(&cpu->env, ARM_FEATURE_PMU); + cpu->midr = 0x411fd070; + cpu->revidr = 0x00000000; + cpu->reset_fpsid = 0x41034070; + cpu->isar.mvfr0 = 0x10110222; + cpu->isar.mvfr1 = 0x12111111; + cpu->isar.mvfr2 = 0x00000043; + cpu->ctr = 0x8444c004; + cpu->reset_sctlr = 0x00c50838; + cpu->isar.id_pfr0 = 0x00000131; + cpu->isar.id_pfr1 = 0x00011011; + cpu->isar.id_dfr0 = 0x03010066; + cpu->id_afr0 = 0x00000000; + cpu->isar.id_mmfr0 = 0x10101105; + cpu->isar.id_mmfr1 = 0x40000000; + cpu->isar.id_mmfr2 = 0x01260000; + cpu->isar.id_mmfr3 = 0x02102211; + cpu->isar.id_isar0 = 0x02101110; + cpu->isar.id_isar1 = 0x13112111; + cpu->isar.id_isar2 = 0x21232042; + cpu->isar.id_isar3 = 0x01112131; + cpu->isar.id_isar4 = 0x00011142; + cpu->isar.id_isar5 = 0x00011121; + cpu->isar.id_isar6 = 0; + cpu->isar.dbgdidr = 0x3516d000; + cpu->clidr = 0x0a200023; + cpu->ccsidr[0] = 0x701fe00a; /* 32KB L1 dcache */ + cpu->ccsidr[1] = 0x201fe012; /* 48KB L1 icache */ + cpu->ccsidr[2] = 0x70ffe07a; /* 2048KB L2 cache */ + define_cortex_a72_a57_a53_cp_reginfo(cpu); - /* old-style VFP short-vector support */ - cpu->isar.mvfr0 = FIELD_DP32(cpu->isar.mvfr0, MVFR0, FPSHVEC, 1); + /* Add additional features supported by QEMU */ + t = cpu->isar.id_isar5; + t = FIELD_DP32(t, ID_ISAR5, AES, 2); + t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); + t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); + t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); + t = FIELD_DP32(t, ID_ISAR5, RDM, 1); + t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); + cpu->isar.id_isar5 = t; + + t = cpu->isar.id_isar6; + t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); + t = FIELD_DP32(t, ID_ISAR6, DP, 1); + t = FIELD_DP32(t, ID_ISAR6, FHM, 1); + t = FIELD_DP32(t, ID_ISAR6, SB, 1); + t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); + t = FIELD_DP32(t, ID_ISAR6, BF16, 1); + t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); + cpu->isar.id_isar6 = t; + + t = cpu->isar.mvfr1; + t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ + t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ + cpu->isar.mvfr1 = t; + + t = cpu->isar.mvfr2; + t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ + t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ + cpu->isar.mvfr2 = t; + + t = cpu->isar.id_mmfr3; + t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ + cpu->isar.id_mmfr3 = t; + + t = cpu->isar.id_mmfr4; + t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ + t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ + t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ + t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ + cpu->isar.id_mmfr4 = t; + + t = cpu->isar.id_pfr0; + t = FIELD_DP32(t, ID_PFR0, DIT, 1); + cpu->isar.id_pfr0 = t; + + t = cpu->isar.id_pfr2; + t = FIELD_DP32(t, ID_PFR2, SSBS, 1); + cpu->isar.id_pfr2 = t; #ifdef CONFIG_USER_ONLY /* - * We don't set these in system emulation mode for the moment, - * since we don't correctly set (all of) the ID registers to - * advertise them. + * Break with true ARMv8 and add back old-style VFP short-vector support. + * Only do this for user-mode, where -cpu max is the default, so that + * older v6 and v7 programs are more likely to work without adjustment. */ - set_feature(&cpu->env, ARM_FEATURE_V8); - { - uint32_t t; - - t = cpu->isar.id_isar5; - t = FIELD_DP32(t, ID_ISAR5, AES, 2); - t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); - t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); - t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); - t = FIELD_DP32(t, ID_ISAR5, RDM, 1); - t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); - cpu->isar.id_isar5 = t; - - t = cpu->isar.id_isar6; - t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); - t = FIELD_DP32(t, ID_ISAR6, DP, 1); - t = FIELD_DP32(t, ID_ISAR6, FHM, 1); - t = FIELD_DP32(t, ID_ISAR6, SB, 1); - t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); - t = FIELD_DP32(t, ID_ISAR6, BF16, 1); - t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); - cpu->isar.id_isar6 = t; - - t = cpu->isar.mvfr1; - t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ - t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ - cpu->isar.mvfr1 = t; - - t = cpu->isar.mvfr2; - t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ - t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ - cpu->isar.mvfr2 = t; - - t = cpu->isar.id_mmfr3; - t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ - cpu->isar.id_mmfr3 = t; - - t = cpu->isar.id_mmfr4; - t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ - t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ - t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ - t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ - cpu->isar.id_mmfr4 = t; - - t = cpu->isar.id_pfr0; - t = FIELD_DP32(t, ID_PFR0, DIT, 1); - cpu->isar.id_pfr0 = t; - - t = cpu->isar.id_pfr2; - t = FIELD_DP32(t, ID_PFR2, SSBS, 1); - cpu->isar.id_pfr2 = t; - } -#endif /* CONFIG_USER_ONLY */ + cpu->isar.mvfr0 = FIELD_DP32(cpu->isar.mvfr0, MVFR0, FPSHVEC, 1); +#endif } #endif /* !TARGET_AARCH64 */ -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> We set this for qemu-system-aarch64, but failed to do so for the strictly 32-bit emulation. Fixes: 3bec78447a9 ("target/arm: Provide ARMv8.4-PMU in '-cpu max'") Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-8-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu_tcg.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ static void arm_max_initfn(Object *obj) t = FIELD_DP32(t, ID_PFR2, SSBS, 1); cpu->isar.id_pfr2 = t; + t = cpu->isar.id_dfr0; + t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ + cpu->isar.id_dfr0 = t; + #ifdef CONFIG_USER_ONLY /* * Break with true ARMv8 and add back old-style VFP short-vector support. -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Share the code to set AArch32 max features so that we no longer have code drift between qemu{-system,}-{arm,aarch64}. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-9-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/internals.h | 2 + target/arm/cpu64.c | 50 +----------------- target/arm/cpu_tcg.c | 114 ++++++++++++++++++++++------------------- 3 files changed, 65 insertions(+), 101 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 @@ static inline void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu) { } void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu); #endif +void aa32_max_features(ARMCPU *cpu); + #endif diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); uint64_t t; - uint32_t u; if (kvm_enabled() || hvf_enabled()) { /* With KVM or HVF, '-cpu max' is identical to '-cpu host' */ @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64ZFR0, F64MM, 1); cpu->isar.id_aa64zfr0 = t; - /* Replicate the same data to the 32-bit id registers. */ - u = cpu->isar.id_isar5; - u = FIELD_DP32(u, ID_ISAR5, AES, 2); /* AES + PMULL */ - u = FIELD_DP32(u, ID_ISAR5, SHA1, 1); - u = FIELD_DP32(u, ID_ISAR5, SHA2, 1); - u = FIELD_DP32(u, ID_ISAR5, CRC32, 1); - u = FIELD_DP32(u, ID_ISAR5, RDM, 1); - u = FIELD_DP32(u, ID_ISAR5, VCMA, 1); - cpu->isar.id_isar5 = u; - - u = cpu->isar.id_isar6; - u = FIELD_DP32(u, ID_ISAR6, JSCVT, 1); - u = FIELD_DP32(u, ID_ISAR6, DP, 1); - u = FIELD_DP32(u, ID_ISAR6, FHM, 1); - u = FIELD_DP32(u, ID_ISAR6, SB, 1); - u = FIELD_DP32(u, ID_ISAR6, SPECRES, 1); - u = FIELD_DP32(u, ID_ISAR6, BF16, 1); - u = FIELD_DP32(u, ID_ISAR6, I8MM, 1); - cpu->isar.id_isar6 = u; - - u = cpu->isar.id_pfr0; - u = FIELD_DP32(u, ID_PFR0, DIT, 1); - cpu->isar.id_pfr0 = u; - - u = cpu->isar.id_pfr2; - u = FIELD_DP32(u, ID_PFR2, SSBS, 1); - cpu->isar.id_pfr2 = u; - - u = cpu->isar.id_mmfr3; - u = FIELD_DP32(u, ID_MMFR3, PAN, 2); /* ATS1E1 */ - cpu->isar.id_mmfr3 = u; - - u = cpu->isar.id_mmfr4; - u = FIELD_DP32(u, ID_MMFR4, HPDS, 1); /* AA32HPD */ - u = FIELD_DP32(u, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ - u = FIELD_DP32(u, ID_MMFR4, CNP, 1); /* TTCNP */ - u = FIELD_DP32(u, ID_MMFR4, XNX, 1); /* TTS2UXN */ - cpu->isar.id_mmfr4 = u; - t = cpu->isar.id_aa64dfr0; t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* v8.4-PMU */ cpu->isar.id_aa64dfr0 = t; - u = cpu->isar.id_dfr0; - u = FIELD_DP32(u, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ - cpu->isar.id_dfr0 = u; - - u = cpu->isar.mvfr1; - u = FIELD_DP32(u, MVFR1, FPHP, 3); /* v8.2-FP16 */ - u = FIELD_DP32(u, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ - cpu->isar.mvfr1 = u; + /* Replicate the same data to the 32-bit id registers. */ + aa32_max_features(cpu); #ifdef CONFIG_USER_ONLY /* diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ #endif #include "cpregs.h" + +/* Share AArch32 -cpu max features with AArch64. */ +void aa32_max_features(ARMCPU *cpu) +{ + uint32_t t; + + /* Add additional features supported by QEMU */ + t = cpu->isar.id_isar5; + t = FIELD_DP32(t, ID_ISAR5, AES, 2); + t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); + t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); + t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); + t = FIELD_DP32(t, ID_ISAR5, RDM, 1); + t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); + cpu->isar.id_isar5 = t; + + t = cpu->isar.id_isar6; + t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); + t = FIELD_DP32(t, ID_ISAR6, DP, 1); + t = FIELD_DP32(t, ID_ISAR6, FHM, 1); + t = FIELD_DP32(t, ID_ISAR6, SB, 1); + t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); + t = FIELD_DP32(t, ID_ISAR6, BF16, 1); + t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); + cpu->isar.id_isar6 = t; + + t = cpu->isar.mvfr1; + t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ + t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ + cpu->isar.mvfr1 = t; + + t = cpu->isar.mvfr2; + t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ + t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ + cpu->isar.mvfr2 = t; + + t = cpu->isar.id_mmfr3; + t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ + cpu->isar.id_mmfr3 = t; + + t = cpu->isar.id_mmfr4; + t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ + t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ + t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ + t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ + cpu->isar.id_mmfr4 = t; + + t = cpu->isar.id_pfr0; + t = FIELD_DP32(t, ID_PFR0, DIT, 1); + cpu->isar.id_pfr0 = t; + + t = cpu->isar.id_pfr2; + t = FIELD_DP32(t, ID_PFR2, SSBS, 1); + cpu->isar.id_pfr2 = t; + + t = cpu->isar.id_dfr0; + t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ + cpu->isar.id_dfr0 = t; +} + #ifndef CONFIG_USER_ONLY static uint64_t l2ctlr_read(CPUARMState *env, const ARMCPRegInfo *ri) { @@ -XXX,XX +XXX,XX @@ static void arm_v7m_class_init(ObjectClass *oc, void *data) static void arm_max_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); - uint32_t t; /* aarch64_a57_initfn, advertising none of the aarch64 features */ cpu->dtb_compatible = "arm,cortex-a57"; @@ -XXX,XX +XXX,XX @@ static void arm_max_initfn(Object *obj) cpu->ccsidr[2] = 0x70ffe07a; /* 2048KB L2 cache */ define_cortex_a72_a57_a53_cp_reginfo(cpu); - /* Add additional features supported by QEMU */ - t = cpu->isar.id_isar5; - t = FIELD_DP32(t, ID_ISAR5, AES, 2); - t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); - t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); - t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); - t = FIELD_DP32(t, ID_ISAR5, RDM, 1); - t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); - cpu->isar.id_isar5 = t; - - t = cpu->isar.id_isar6; - t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); - t = FIELD_DP32(t, ID_ISAR6, DP, 1); - t = FIELD_DP32(t, ID_ISAR6, FHM, 1); - t = FIELD_DP32(t, ID_ISAR6, SB, 1); - t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); - t = FIELD_DP32(t, ID_ISAR6, BF16, 1); - t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); - cpu->isar.id_isar6 = t; - - t = cpu->isar.mvfr1; - t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ - t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ - cpu->isar.mvfr1 = t; - - t = cpu->isar.mvfr2; - t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ - t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ - cpu->isar.mvfr2 = t; - - t = cpu->isar.id_mmfr3; - t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ - cpu->isar.id_mmfr3 = t; - - t = cpu->isar.id_mmfr4; - t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ - t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ - t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ - t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ - cpu->isar.id_mmfr4 = t; - - t = cpu->isar.id_pfr0; - t = FIELD_DP32(t, ID_PFR0, DIT, 1); - cpu->isar.id_pfr0 = t; - - t = cpu->isar.id_pfr2; - t = FIELD_DP32(t, ID_PFR2, SSBS, 1); - cpu->isar.id_pfr2 = t; - - t = cpu->isar.id_dfr0; - t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ - cpu->isar.id_dfr0 = t; + aa32_max_features(cpu); #ifdef CONFIG_USER_ONLY /* -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Update the legacy feature names to the current names. Provide feature names for id changes that were not marked. Sort the field updates into increasing bitfield order. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-10-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu64.c | 100 +++++++++++++++++++++---------------------- target/arm/cpu_tcg.c | 48 ++++++++++----------- 2 files changed, 74 insertions(+), 74 deletions(-) diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) cpu->midr = t; t = cpu->isar.id_aa64isar0; - t = FIELD_DP64(t, ID_AA64ISAR0, AES, 2); /* AES + PMULL */ - t = FIELD_DP64(t, ID_AA64ISAR0, SHA1, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, SHA2, 2); /* SHA512 */ + 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, ATOMIC, 2); - t = FIELD_DP64(t, ID_AA64ISAR0, RDM, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, SHA3, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, SM3, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, SM4, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, DP, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, FHM, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, TS, 2); /* v8.5-CondM */ - t = FIELD_DP64(t, ID_AA64ISAR0, TLB, 2); /* FEAT_TLBIRANGE */ - t = FIELD_DP64(t, ID_AA64ISAR0, RNDR, 1); + 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 */ + t = FIELD_DP64(t, ID_AA64ISAR0, SM3, 1); /* FEAT_SM3 */ + t = FIELD_DP64(t, ID_AA64ISAR0, SM4, 1); /* FEAT_SM4 */ + t = FIELD_DP64(t, ID_AA64ISAR0, DP, 1); /* FEAT_DotProd */ + t = FIELD_DP64(t, ID_AA64ISAR0, FHM, 1); /* FEAT_FHM */ + t = FIELD_DP64(t, ID_AA64ISAR0, TS, 2); /* FEAT_FlagM2 */ + t = FIELD_DP64(t, ID_AA64ISAR0, TLB, 2); /* FEAT_TLBIRANGE */ + t = FIELD_DP64(t, ID_AA64ISAR0, RNDR, 1); /* FEAT_RNG */ cpu->isar.id_aa64isar0 = t; t = cpu->isar.id_aa64isar1; - t = FIELD_DP64(t, ID_AA64ISAR1, DPB, 2); - t = FIELD_DP64(t, ID_AA64ISAR1, JSCVT, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, FCMA, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, SB, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, SPECRES, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, BF16, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, FRINTTS, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, LRCPC, 2); /* ARMv8.4-RCPC */ - t = FIELD_DP64(t, ID_AA64ISAR1, I8MM, 1); + t = FIELD_DP64(t, ID_AA64ISAR1, DPB, 2); /* FEAT_DPB2 */ + t = FIELD_DP64(t, ID_AA64ISAR1, JSCVT, 1); /* FEAT_JSCVT */ + t = FIELD_DP64(t, ID_AA64ISAR1, FCMA, 1); /* FEAT_FCMA */ + t = FIELD_DP64(t, ID_AA64ISAR1, LRCPC, 2); /* FEAT_LRCPC2 */ + t = FIELD_DP64(t, ID_AA64ISAR1, FRINTTS, 1); /* FEAT_FRINTTS */ + t = FIELD_DP64(t, ID_AA64ISAR1, SB, 1); /* FEAT_SB */ + t = FIELD_DP64(t, ID_AA64ISAR1, SPECRES, 1); /* FEAT_SPECRES */ + t = FIELD_DP64(t, ID_AA64ISAR1, BF16, 1); /* FEAT_BF16 */ + t = FIELD_DP64(t, ID_AA64ISAR1, I8MM, 1); /* FEAT_I8MM */ cpu->isar.id_aa64isar1 = 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 */ t = FIELD_DP64(t, ID_AA64PFR0, SVE, 1); - t = FIELD_DP64(t, ID_AA64PFR0, FP, 1); - t = FIELD_DP64(t, ID_AA64PFR0, ADVSIMD, 1); - t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); - t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); + t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ + t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ cpu->isar.id_aa64pfr0 = t; t = cpu->isar.id_aa64pfr1; - t = FIELD_DP64(t, ID_AA64PFR1, BT, 1); - t = FIELD_DP64(t, ID_AA64PFR1, SSBS, 2); + t = FIELD_DP64(t, ID_AA64PFR1, BT, 1); /* FEAT_BTI */ + t = FIELD_DP64(t, ID_AA64PFR1, SSBS, 2); /* FEAT_SSBS2 */ /* * Begin with full support for MTE. This will be downgraded to MTE=0 * during realize if the board provides no tag memory, much like * we do for EL2 with the virtualization=on property. */ - t = FIELD_DP64(t, ID_AA64PFR1, MTE, 3); + t = FIELD_DP64(t, ID_AA64PFR1, MTE, 3); /* FEAT_MTE3 */ cpu->isar.id_aa64pfr1 = t; t = cpu->isar.id_aa64mmfr0; @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) cpu->isar.id_aa64mmfr0 = t; t = cpu->isar.id_aa64mmfr1; - t = FIELD_DP64(t, ID_AA64MMFR1, HPDS, 1); /* HPD */ - t = FIELD_DP64(t, ID_AA64MMFR1, LO, 1); - t = FIELD_DP64(t, ID_AA64MMFR1, VH, 1); - t = FIELD_DP64(t, ID_AA64MMFR1, PAN, 2); /* ATS1E1 */ - t = FIELD_DP64(t, ID_AA64MMFR1, VMIDBITS, 2); /* VMID16 */ - t = FIELD_DP64(t, ID_AA64MMFR1, XNX, 1); /* TTS2UXN */ + 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, LO, 1); /* FEAT_LOR */ + t = FIELD_DP64(t, ID_AA64MMFR1, PAN, 2); /* FEAT_PAN2 */ + t = FIELD_DP64(t, ID_AA64MMFR1, XNX, 1); /* FEAT_XNX */ cpu->isar.id_aa64mmfr1 = t; t = cpu->isar.id_aa64mmfr2; - t = FIELD_DP64(t, ID_AA64MMFR2, UAO, 1); - t = FIELD_DP64(t, ID_AA64MMFR2, CNP, 1); /* TTCNP */ - t = FIELD_DP64(t, ID_AA64MMFR2, ST, 1); /* TTST */ - t = FIELD_DP64(t, ID_AA64MMFR2, VARANGE, 1); /* FEAT_LVA */ - t = FIELD_DP64(t, ID_AA64MMFR2, TTL, 1); /* FEAT_TTL */ - t = FIELD_DP64(t, ID_AA64MMFR2, BBM, 2); /* FEAT_BBM at level 2 */ + t = FIELD_DP64(t, ID_AA64MMFR2, CNP, 1); /* FEAT_TTCNP */ + t = FIELD_DP64(t, ID_AA64MMFR2, UAO, 1); /* FEAT_UAO */ + t = FIELD_DP64(t, ID_AA64MMFR2, VARANGE, 1); /* FEAT_LVA */ + t = FIELD_DP64(t, ID_AA64MMFR2, ST, 1); /* FEAT_TTST */ + t = FIELD_DP64(t, ID_AA64MMFR2, TTL, 1); /* FEAT_TTL */ + t = FIELD_DP64(t, ID_AA64MMFR2, BBM, 2); /* FEAT_BBM at level 2 */ cpu->isar.id_aa64mmfr2 = t; t = cpu->isar.id_aa64zfr0; t = FIELD_DP64(t, ID_AA64ZFR0, SVEVER, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, AES, 2); /* PMULL */ - t = FIELD_DP64(t, ID_AA64ZFR0, BITPERM, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, BFLOAT16, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, SHA3, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, SM4, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, I8MM, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, F32MM, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, F64MM, 1); + t = FIELD_DP64(t, ID_AA64ZFR0, AES, 2); /* FEAT_SVE_PMULL128 */ + t = FIELD_DP64(t, ID_AA64ZFR0, BITPERM, 1); /* FEAT_SVE_BitPerm */ + t = FIELD_DP64(t, ID_AA64ZFR0, BFLOAT16, 1); /* FEAT_BF16 */ + t = FIELD_DP64(t, ID_AA64ZFR0, SHA3, 1); /* FEAT_SVE_SHA3 */ + t = FIELD_DP64(t, ID_AA64ZFR0, SM4, 1); /* FEAT_SVE_SM4 */ + t = FIELD_DP64(t, ID_AA64ZFR0, I8MM, 1); /* FEAT_I8MM */ + t = FIELD_DP64(t, ID_AA64ZFR0, F32MM, 1); /* FEAT_F32MM */ + t = FIELD_DP64(t, ID_AA64ZFR0, F64MM, 1); /* FEAT_F64MM */ cpu->isar.id_aa64zfr0 = t; t = cpu->isar.id_aa64dfr0; - t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* v8.4-PMU */ + t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_aa64dfr0 = t; /* Replicate the same data to the 32-bit id registers. */ diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ void aa32_max_features(ARMCPU *cpu) /* Add additional features supported by QEMU */ t = cpu->isar.id_isar5; - t = FIELD_DP32(t, ID_ISAR5, AES, 2); - t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); - t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); + t = FIELD_DP32(t, ID_ISAR5, AES, 2); /* FEAT_PMULL */ + t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); /* FEAT_SHA1 */ + t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); /* FEAT_SHA256 */ t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); - t = FIELD_DP32(t, ID_ISAR5, RDM, 1); - t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); + t = FIELD_DP32(t, ID_ISAR5, RDM, 1); /* FEAT_RDM */ + t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); /* FEAT_FCMA */ cpu->isar.id_isar5 = t; t = cpu->isar.id_isar6; - t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); - t = FIELD_DP32(t, ID_ISAR6, DP, 1); - t = FIELD_DP32(t, ID_ISAR6, FHM, 1); - t = FIELD_DP32(t, ID_ISAR6, SB, 1); - t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); - t = FIELD_DP32(t, ID_ISAR6, BF16, 1); - t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); + t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); /* FEAT_JSCVT */ + t = FIELD_DP32(t, ID_ISAR6, DP, 1); /* Feat_DotProd */ + t = FIELD_DP32(t, ID_ISAR6, FHM, 1); /* FEAT_FHM */ + t = FIELD_DP32(t, ID_ISAR6, SB, 1); /* FEAT_SB */ + t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); /* FEAT_SPECRES */ + t = FIELD_DP32(t, ID_ISAR6, BF16, 1); /* FEAT_AA32BF16 */ + t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); /* FEAT_AA32I8MM */ cpu->isar.id_isar6 = t; t = cpu->isar.mvfr1; - t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ - t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ + t = FIELD_DP32(t, MVFR1, FPHP, 3); /* FEAT_FP16 */ + t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* FEAT_FP16 */ cpu->isar.mvfr1 = t; t = cpu->isar.mvfr2; - t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ - t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ + t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ + t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ cpu->isar.mvfr2 = t; t = cpu->isar.id_mmfr3; - t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ + t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* FEAT_PAN2 */ cpu->isar.id_mmfr3 = t; t = cpu->isar.id_mmfr4; - t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ - t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ - t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ - t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ + t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* FEAT_AA32HPD */ + 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*/ cpu->isar.id_mmfr4 = t; t = cpu->isar.id_pfr0; - t = FIELD_DP32(t, ID_PFR0, DIT, 1); + t = FIELD_DP32(t, ID_PFR0, DIT, 1); /* FEAT_DIT */ cpu->isar.id_pfr0 = t; t = cpu->isar.id_pfr2; - t = FIELD_DP32(t, ID_PFR2, SSBS, 1); + t = FIELD_DP32(t, ID_PFR2, SSBS, 1); /* FEAT_SSBS */ cpu->isar.id_pfr2 = t; t = cpu->isar.id_dfr0; - t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ + t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_dfr0 = t; } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Use FIELD_DP{32,64} to manipulate id_pfr1 and id_aa64pfr0 during arm_cpu_realizefn. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-11-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.c | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 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) */ unset_feature(env, ARM_FEATURE_EL3); - /* Disable the security extension feature bits in the processor feature - * registers as well. These are id_pfr1[7:4] and id_aa64pfr0[15:12]. + /* + * Disable the security extension feature bits in the processor + * feature registers as well. */ - cpu->isar.id_pfr1 &= ~0xf0; - cpu->isar.id_aa64pfr0 &= ~0xf000; + cpu->isar.id_pfr1 = FIELD_DP32(cpu->isar.id_pfr1, ID_PFR1, SECURITY, 0); + cpu->isar.id_aa64pfr0 = FIELD_DP64(cpu->isar.id_aa64pfr0, + ID_AA64PFR0, EL3, 0); } if (!cpu->has_el2) { @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) } if (!arm_feature(env, ARM_FEATURE_EL2)) { - /* Disable the hypervisor feature bits in the processor feature - * registers if we don't have EL2. These are id_pfr1[15:12] and - * id_aa64pfr0_el1[11:8]. + /* + * Disable the hypervisor feature bits in the processor feature + * registers if we don't have EL2. */ - cpu->isar.id_aa64pfr0 &= ~0xf00; - cpu->isar.id_pfr1 &= ~0xf000; + cpu->isar.id_aa64pfr0 = FIELD_DP64(cpu->isar.id_aa64pfr0, + ID_AA64PFR0, EL2, 0); + cpu->isar.id_pfr1 = FIELD_DP32(cpu->isar.id_pfr1, + ID_PFR1, VIRTUALIZATION, 0); } #ifndef CONFIG_USER_ONLY -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> The only portion of FEAT_Debugv8p2 that is relevant to QEMU is CONTEXTIDR_EL2, which is also conditionally implemented with FEAT_VHE. The rest of the debug extension concerns the External debug interface, which is outside the scope of QEMU. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-12-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu.c | 1 + target/arm/cpu64.c | 1 + target/arm/cpu_tcg.c | 2 ++ 4 files changed, 5 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_BTI (Branch Target Identification) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) +- FEAT_Debugv8p2 (Debug changes for v8.2) - FEAT_DotProd (Advanced SIMD dot product instructions) - FEAT_FCMA (Floating-point complex number instructions) - FEAT_FHM (Floating-point half-precision multiplication instructions) 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) * feature registers as well. */ cpu->isar.id_pfr1 = FIELD_DP32(cpu->isar.id_pfr1, ID_PFR1, SECURITY, 0); + cpu->isar.id_dfr0 = FIELD_DP32(cpu->isar.id_dfr0, ID_DFR0, COPSDBG, 0); cpu->isar.id_aa64pfr0 = FIELD_DP64(cpu->isar.id_aa64pfr0, ID_AA64PFR0, EL3, 0); } diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) cpu->isar.id_aa64zfr0 = t; t = cpu->isar.id_aa64dfr0; + t = FIELD_DP64(t, ID_AA64DFR0, DEBUGVER, 8); /* FEAT_Debugv8p2 */ t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_aa64dfr0 = t; diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ void aa32_max_features(ARMCPU *cpu) cpu->isar.id_pfr2 = t; t = cpu->isar.id_dfr0; + t = FIELD_DP32(t, ID_DFR0, COPDBG, 8); /* FEAT_Debugv8p2 */ + t = FIELD_DP32(t, ID_DFR0, COPSDBG, 8); /* FEAT_Debugv8p2 */ t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_dfr0 = t; } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This extension concerns changes to the External Debug interface, with Secure and Non-secure access to the debug registers, and all of it is outside the scope of QEMU. Indicating support for this is mandatory with FEAT_SEL2, which we do implement. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-13-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 2 +- target/arm/cpu_tcg.c | 4 ++-- 3 files changed, 4 insertions(+), 3 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_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) +- FEAT_Debugv8p4 (Debug changes for v8.4) - FEAT_DotProd (Advanced SIMD dot product instructions) - FEAT_FCMA (Floating-point complex number instructions) - FEAT_FHM (Floating-point half-precision multiplication instructions) diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) cpu->isar.id_aa64zfr0 = t; t = cpu->isar.id_aa64dfr0; - t = FIELD_DP64(t, ID_AA64DFR0, DEBUGVER, 8); /* FEAT_Debugv8p2 */ + t = FIELD_DP64(t, ID_AA64DFR0, DEBUGVER, 9); /* FEAT_Debugv8p4 */ t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_aa64dfr0 = t; diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ void aa32_max_features(ARMCPU *cpu) cpu->isar.id_pfr2 = t; t = cpu->isar.id_dfr0; - t = FIELD_DP32(t, ID_DFR0, COPDBG, 8); /* FEAT_Debugv8p2 */ - t = FIELD_DP32(t, ID_DFR0, COPSDBG, 8); /* FEAT_Debugv8p2 */ + t = FIELD_DP32(t, ID_DFR0, COPDBG, 9); /* FEAT_Debugv8p4 */ + t = FIELD_DP32(t, ID_DFR0, COPSDBG, 9); /* FEAT_Debugv8p4 */ t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_dfr0 = t; } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Add only the system registers required to implement zero error records. This means that all values for ERRSELR are out of range, which means that it and all of the indexed error record registers need not be implemented. Add the EL2 registers required for injecting virtual SError. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-14-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 5 +++ target/arm/helper.c | 84 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 89 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 @@ typedef struct CPUArchState { uint64_t tfsr_el[4]; /* tfsre0_el1 is index 0. */ uint64_t gcr_el1; uint64_t rgsr_el1; + + /* Minimal RAS registers */ + uint64_t disr_el1; + uint64_t vdisr_el2; + uint64_t vsesr_el2; } cp15; struct { 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 debug_lpae_cp_reginfo[] = { .access = PL0_R, .type = ARM_CP_CONST|ARM_CP_64BIT, .resetvalue = 0 }, }; +/* + * Check for traps to RAS registers, which are controlled + * by HCR_EL2.TERR and SCR_EL3.TERR. + */ +static CPAccessResult access_terr(CPUARMState *env, const ARMCPRegInfo *ri, + bool isread) +{ + int el = arm_current_el(env); + + if (el < 2 && (arm_hcr_el2_eff(env) & HCR_TERR)) { + return CP_ACCESS_TRAP_EL2; + } + if (el < 3 && (env->cp15.scr_el3 & SCR_TERR)) { + return CP_ACCESS_TRAP_EL3; + } + return CP_ACCESS_OK; +} + +static uint64_t disr_read(CPUARMState *env, const ARMCPRegInfo *ri) +{ + int el = arm_current_el(env); + + if (el < 2 && (arm_hcr_el2_eff(env) & HCR_AMO)) { + return env->cp15.vdisr_el2; + } + if (el < 3 && (env->cp15.scr_el3 & SCR_EA)) { + return 0; /* RAZ/WI */ + } + return env->cp15.disr_el1; +} + +static void disr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t val) +{ + int el = arm_current_el(env); + + if (el < 2 && (arm_hcr_el2_eff(env) & HCR_AMO)) { + env->cp15.vdisr_el2 = val; + return; + } + if (el < 3 && (env->cp15.scr_el3 & SCR_EA)) { + return; /* RAZ/WI */ + } + env->cp15.disr_el1 = val; +} + +/* + * Minimal RAS implementation with no Error Records. + * Which means that all of the Error Record registers: + * ERXADDR_EL1 + * ERXCTLR_EL1 + * ERXFR_EL1 + * ERXMISC0_EL1 + * ERXMISC1_EL1 + * ERXMISC2_EL1 + * ERXMISC3_EL1 + * ERXPFGCDN_EL1 (RASv1p1) + * ERXPFGCTL_EL1 (RASv1p1) + * ERXPFGF_EL1 (RASv1p1) + * ERXSTATUS_EL1 + * and + * ERRSELR_EL1 + * may generate UNDEFINED, which is the effect we get by not + * listing them at all. + */ +static const ARMCPRegInfo minimal_ras_reginfo[] = { + { .name = "DISR_EL1", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 0, .crn = 12, .crm = 1, .opc2 = 1, + .access = PL1_RW, .fieldoffset = offsetof(CPUARMState, cp15.disr_el1), + .readfn = disr_read, .writefn = disr_write, .raw_writefn = raw_write }, + { .name = "ERRIDR_EL1", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 0, .crn = 5, .crm = 3, .opc2 = 0, + .access = PL1_R, .accessfn = access_terr, + .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "VDISR_EL2", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 4, .crn = 12, .crm = 1, .opc2 = 1, + .access = PL2_RW, .fieldoffset = offsetof(CPUARMState, cp15.vdisr_el2) }, + { .name = "VSESR_EL2", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 2, .opc2 = 3, + .access = PL2_RW, .fieldoffset = offsetof(CPUARMState, cp15.vsesr_el2) }, +}; + /* Return the exception level to which exceptions should be taken * via SVEAccessTrap. If an exception should be routed through * AArch64.AdvSIMDFPAccessTrap, return 0; fp_exception_el should @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) if (cpu_isar_feature(aa64_ssbs, cpu)) { define_one_arm_cp_reg(cpu, &ssbs_reginfo); } + if (cpu_isar_feature(any_ras, cpu)) { + define_arm_cp_regs(cpu, minimal_ras_reginfo); + } if (cpu_isar_feature(aa64_vh, cpu) || cpu_isar_feature(aa64_debugv8p2, cpu)) { -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Enable writes to the TERR and TEA bits when RAS is enabled. These bits are otherwise RES0. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-15-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 9 +++++++++ 1 file changed, 9 insertions(+) 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 scr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) } valid_mask &= ~SCR_NET; + if (cpu_isar_feature(aa64_ras, cpu)) { + valid_mask |= SCR_TERR; + } if (cpu_isar_feature(aa64_lor, cpu)) { valid_mask |= SCR_TLOR; } @@ -XXX,XX +XXX,XX @@ static void scr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) } } else { valid_mask &= ~(SCR_RW | SCR_ST); + if (cpu_isar_feature(aa32_ras, cpu)) { + valid_mask |= SCR_TERR; + } } if (!arm_feature(env, ARM_FEATURE_EL2)) { @@ -XXX,XX +XXX,XX @@ static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask) if (cpu_isar_feature(aa64_vh, cpu)) { valid_mask |= HCR_E2H; } + if (cpu_isar_feature(aa64_ras, cpu)) { + valid_mask |= HCR_TERR | HCR_TEA; + } if (cpu_isar_feature(aa64_lor, cpu)) { valid_mask |= HCR_TLOR; } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Virtual SError exceptions are raised by setting HCR_EL2.VSE, and are routed to EL1 just like other virtual exceptions. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-16-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 2 ++ target/arm/internals.h | 8 ++++++++ target/arm/syndrome.h | 5 +++++ target/arm/cpu.c | 38 +++++++++++++++++++++++++++++++++++++- target/arm/helper.c | 40 +++++++++++++++++++++++++++++++++++++++- 5 files changed, 91 insertions(+), 2 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 @@ #define EXCP_LSERR 21 /* v8M LSERR SecureFault */ #define EXCP_UNALIGNED 22 /* v7M UNALIGNED UsageFault */ #define EXCP_DIVBYZERO 23 /* v7M DIVBYZERO UsageFault */ +#define EXCP_VSERR 24 /* NB: add new EXCP_ defines to the array in arm_log_exception() too */ #define ARMV7M_EXCP_RESET 1 @@ -XXX,XX +XXX,XX @@ enum { #define CPU_INTERRUPT_FIQ CPU_INTERRUPT_TGT_EXT_1 #define CPU_INTERRUPT_VIRQ CPU_INTERRUPT_TGT_EXT_2 #define CPU_INTERRUPT_VFIQ CPU_INTERRUPT_TGT_EXT_3 +#define CPU_INTERRUPT_VSERR CPU_INTERRUPT_TGT_INT_0 /* The usual mapping for an AArch64 system register to its AArch32 * counterpart is for the 32 bit world to have access to the lower 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_cpu_update_virq(ARMCPU *cpu); */ void arm_cpu_update_vfiq(ARMCPU *cpu); +/** + * arm_cpu_update_vserr: Update CPU_INTERRUPT_VSERR bit + * + * Update the CPU_INTERRUPT_VSERR bit in cs->interrupt_request, + * following a change to the HCR_EL2.VSE bit. + */ +void arm_cpu_update_vserr(ARMCPU *cpu); + /** * arm_mmu_idx_el: * @env: The cpu environment 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 @@ static inline uint32_t syn_pcalignment(void) return (EC_PCALIGNMENT << ARM_EL_EC_SHIFT) | ARM_EL_IL; } +static inline uint32_t syn_serror(uint32_t extra) +{ + return (EC_SERROR << ARM_EL_EC_SHIFT) | ARM_EL_IL | extra; +} + #endif /* TARGET_ARM_SYNDROME_H */ 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 bool arm_cpu_has_work(CPUState *cs) return (cpu->power_state != PSCI_OFF) && cs->interrupt_request & (CPU_INTERRUPT_FIQ | CPU_INTERRUPT_HARD - | CPU_INTERRUPT_VFIQ | CPU_INTERRUPT_VIRQ + | CPU_INTERRUPT_VFIQ | CPU_INTERRUPT_VIRQ | CPU_INTERRUPT_VSERR | CPU_INTERRUPT_EXITTB); } @@ -XXX,XX +XXX,XX @@ static inline bool arm_excp_unmasked(CPUState *cs, unsigned int excp_idx, return false; } return !(env->daif & PSTATE_I); + case EXCP_VSERR: + if (!(hcr_el2 & HCR_AMO) || (hcr_el2 & HCR_TGE)) { + /* VIRQs are only taken when hypervized. */ + return false; + } + return !(env->daif & PSTATE_A); default: g_assert_not_reached(); } @@ -XXX,XX +XXX,XX @@ static bool arm_cpu_exec_interrupt(CPUState *cs, int interrupt_request) goto found; } } + if (interrupt_request & CPU_INTERRUPT_VSERR) { + excp_idx = EXCP_VSERR; + target_el = 1; + if (arm_excp_unmasked(cs, excp_idx, target_el, + cur_el, secure, hcr_el2)) { + /* Taking a virtual abort clears HCR_EL2.VSE */ + env->cp15.hcr_el2 &= ~HCR_VSE; + cpu_reset_interrupt(cs, CPU_INTERRUPT_VSERR); + goto found; + } + } return false; found: @@ -XXX,XX +XXX,XX @@ void arm_cpu_update_vfiq(ARMCPU *cpu) } } +void arm_cpu_update_vserr(ARMCPU *cpu) +{ + /* + * Update the interrupt level for VSERR, which is the HCR_EL2.VSE bit. + */ + CPUARMState *env = &cpu->env; + CPUState *cs = CPU(cpu); + + bool new_state = env->cp15.hcr_el2 & HCR_VSE; + + if (new_state != ((cs->interrupt_request & CPU_INTERRUPT_VSERR) != 0)) { + if (new_state) { + cpu_interrupt(cs, CPU_INTERRUPT_VSERR); + } else { + cpu_reset_interrupt(cs, CPU_INTERRUPT_VSERR); + } + } +} + #ifndef CONFIG_USER_ONLY static void arm_cpu_set_irq(void *opaque, int irq, int level) { 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 uint64_t isr_read(CPUARMState *env, const ARMCPRegInfo *ri) } } - /* External aborts are not possible in QEMU so A bit is always clear */ + if (hcr_el2 & HCR_AMO) { + if (cs->interrupt_request & CPU_INTERRUPT_VSERR) { + ret |= CPSR_A; + } + } + return ret; } @@ -XXX,XX +XXX,XX @@ static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask) g_assert(qemu_mutex_iothread_locked()); arm_cpu_update_virq(cpu); arm_cpu_update_vfiq(cpu); + arm_cpu_update_vserr(cpu); } static void hcr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) @@ -XXX,XX +XXX,XX @@ void arm_log_exception(CPUState *cs) [EXCP_LSERR] = "v8M LSERR UsageFault", [EXCP_UNALIGNED] = "v7M UNALIGNED UsageFault", [EXCP_DIVBYZERO] = "v7M DIVBYZERO UsageFault", + [EXCP_VSERR] = "Virtual SERR", }; if (idx >= 0 && idx < ARRAY_SIZE(excnames)) { @@ -XXX,XX +XXX,XX @@ static void arm_cpu_do_interrupt_aarch32(CPUState *cs) mask = CPSR_A | CPSR_I | CPSR_F; offset = 4; break; + case EXCP_VSERR: + { + /* + * Note that this is reported as a data abort, but the DFAR + * has an UNKNOWN value. Construct the SError syndrome from + * AET and ExT fields. + */ + ARMMMUFaultInfo fi = { .type = ARMFault_AsyncExternal, }; + + if (extended_addresses_enabled(env)) { + env->exception.fsr = arm_fi_to_lfsc(&fi); + } else { + env->exception.fsr = arm_fi_to_sfsc(&fi); + } + env->exception.fsr |= env->cp15.vsesr_el2 & 0xd000; + A32_BANKED_CURRENT_REG_SET(env, dfsr, env->exception.fsr); + qemu_log_mask(CPU_LOG_INT, "...with IFSR 0x%x\n", + env->exception.fsr); + + new_mode = ARM_CPU_MODE_ABT; + addr = 0x10; + mask = CPSR_A | CPSR_I; + offset = 8; + } + break; case EXCP_SMC: new_mode = ARM_CPU_MODE_MON; addr = 0x08; @@ -XXX,XX +XXX,XX @@ static void arm_cpu_do_interrupt_aarch64(CPUState *cs) case EXCP_VFIQ: addr += 0x100; break; + case EXCP_VSERR: + addr += 0x180; + /* Construct the SError syndrome from IDS and ISS fields. */ + env->exception.syndrome = syn_serror(env->cp15.vsesr_el2 & 0x1ffffff); + env->cp15.esr_el[new_el] = env->exception.syndrome; + break; default: cpu_abort(cs, "Unhandled exception 0x%x\n", cs->exception_index); } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Check for and defer any pending virtual SError. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-17-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 1 + target/arm/a32.decode | 16 ++++++++------ target/arm/t32.decode | 18 ++++++++-------- target/arm/op_helper.c | 43 ++++++++++++++++++++++++++++++++++++++ target/arm/translate-a64.c | 17 +++++++++++++++ target/arm/translate.c | 23 ++++++++++++++++++++ 6 files changed, 103 insertions(+), 15 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_1(wfe, void, env) DEF_HELPER_1(yield, void, env) DEF_HELPER_1(pre_hvc, void, env) DEF_HELPER_2(pre_smc, void, env, i32) +DEF_HELPER_1(vesb, void, env) DEF_HELPER_3(cpsr_write, void, env, i32, i32) DEF_HELPER_2(cpsr_write_eret, void, env, i32) diff --git a/target/arm/a32.decode b/target/arm/a32.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/a32.decode +++ b/target/arm/a32.decode @@ -XXX,XX +XXX,XX @@ SMULTT .... 0001 0110 .... 0000 .... 1110 .... @rd0mn { { - YIELD ---- 0011 0010 0000 1111 ---- 0000 0001 - WFE ---- 0011 0010 0000 1111 ---- 0000 0010 - WFI ---- 0011 0010 0000 1111 ---- 0000 0011 + [ + YIELD ---- 0011 0010 0000 1111 ---- 0000 0001 + WFE ---- 0011 0010 0000 1111 ---- 0000 0010 + WFI ---- 0011 0010 0000 1111 ---- 0000 0011 - # TODO: Implement SEV, SEVL; may help SMP performance. - # SEV ---- 0011 0010 0000 1111 ---- 0000 0100 - # SEVL ---- 0011 0010 0000 1111 ---- 0000 0101 + # TODO: Implement SEV, SEVL; may help SMP performance. + # SEV ---- 0011 0010 0000 1111 ---- 0000 0100 + # SEVL ---- 0011 0010 0000 1111 ---- 0000 0101 + + ESB ---- 0011 0010 0000 1111 ---- 0001 0000 + ] # The canonical nop ends in 00000000, but the whole of the # rest of the space executes as nop if otherwise unsupported. diff --git a/target/arm/t32.decode b/target/arm/t32.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/t32.decode +++ b/target/arm/t32.decode @@ -XXX,XX +XXX,XX @@ CLZ 1111 1010 1011 ---- 1111 .... 1000 .... @rdm [ # Hints, and CPS { - YIELD 1111 0011 1010 1111 1000 0000 0000 0001 - WFE 1111 0011 1010 1111 1000 0000 0000 0010 - WFI 1111 0011 1010 1111 1000 0000 0000 0011 + [ + YIELD 1111 0011 1010 1111 1000 0000 0000 0001 + WFE 1111 0011 1010 1111 1000 0000 0000 0010 + WFI 1111 0011 1010 1111 1000 0000 0000 0011 - # TODO: Implement SEV, SEVL; may help SMP performance. - # SEV 1111 0011 1010 1111 1000 0000 0000 0100 - # SEVL 1111 0011 1010 1111 1000 0000 0000 0101 + # TODO: Implement SEV, SEVL; may help SMP performance. + # SEV 1111 0011 1010 1111 1000 0000 0000 0100 + # SEVL 1111 0011 1010 1111 1000 0000 0000 0101 - # For M-profile minimal-RAS ESB can be a NOP, which is the - # default behaviour since it is in the hint space. - # ESB 1111 0011 1010 1111 1000 0000 0001 0000 + ESB 1111 0011 1010 1111 1000 0000 0001 0000 + ] # The canonical nop ends in 0000 0000, but the whole rest # of the space is "reserved hint, behaves as nop". diff --git a/target/arm/op_helper.c b/target/arm/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/op_helper.c +++ b/target/arm/op_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(probe_access)(CPUARMState *env, target_ulong ptr, access_type, mmu_idx, ra); } } + +/* + * This function corresponds to AArch64.vESBOperation(). + * Note that the AArch32 version is not functionally different. + */ +void HELPER(vesb)(CPUARMState *env) +{ + /* + * The EL2Enabled() check is done inside arm_hcr_el2_eff, + * and will return HCR_EL2.VSE == 0, so nothing happens. + */ + uint64_t hcr = arm_hcr_el2_eff(env); + bool enabled = !(hcr & HCR_TGE) && (hcr & HCR_AMO); + bool pending = enabled && (hcr & HCR_VSE); + bool masked = (env->daif & PSTATE_A); + + /* If VSE pending and masked, defer the exception. */ + if (pending && masked) { + uint32_t syndrome; + + if (arm_el_is_aa64(env, 1)) { + /* Copy across IDS and ISS from VSESR. */ + syndrome = env->cp15.vsesr_el2 & 0x1ffffff; + } else { + ARMMMUFaultInfo fi = { .type = ARMFault_AsyncExternal }; + + if (extended_addresses_enabled(env)) { + syndrome = arm_fi_to_lfsc(&fi); + } else { + syndrome = arm_fi_to_sfsc(&fi); + } + /* Copy across AET and ExT from VSESR. */ + syndrome |= env->cp15.vsesr_el2 & 0xd000; + } + + /* Set VDISR_EL2.A along with the syndrome. */ + env->cp15.vdisr_el2 = syndrome | (1u << 31); + + /* Clear pending virtual SError */ + env->cp15.hcr_el2 &= ~HCR_VSE; + cpu_reset_interrupt(env_cpu(env), CPU_INTERRUPT_VSERR); + } +} diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void handle_hint(DisasContext *s, uint32_t insn, gen_helper_autib(cpu_X[17], cpu_env, cpu_X[17], cpu_X[16]); } break; + case 0b10000: /* ESB */ + /* Without RAS, we must implement this as NOP. */ + if (dc_isar_feature(aa64_ras, s)) { + /* + * QEMU does not have a source of physical SErrors, + * so we are only concerned with virtual SErrors. + * The pseudocode in the ARM for this case is + * if PSTATE.EL IN {EL0, EL1} && EL2Enabled() then + * AArch64.vESBOperation(); + * Most of the condition can be evaluated at translation time. + * Test for EL2 present, and defer test for SEL2 to runtime. + */ + if (s->current_el <= 1 && arm_dc_feature(s, ARM_FEATURE_EL2)) { + gen_helper_vesb(cpu_env); + } + } + break; case 0b11000: /* PACIAZ */ if (s->pauth_active) { gen_helper_pacia(cpu_X[30], cpu_env, cpu_X[30], diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static bool trans_WFI(DisasContext *s, arg_WFI *a) return true; } +static bool trans_ESB(DisasContext *s, arg_ESB *a) +{ + /* + * For M-profile, minimal-RAS ESB can be a NOP. + * Without RAS, we must implement this as NOP. + */ + if (!arm_dc_feature(s, ARM_FEATURE_M) && dc_isar_feature(aa32_ras, s)) { + /* + * QEMU does not have a source of physical SErrors, + * so we are only concerned with virtual SErrors. + * The pseudocode in the ARM for this case is + * if PSTATE.EL IN {EL0, EL1} && EL2Enabled() then + * AArch32.vESBOperation(); + * Most of the condition can be evaluated at translation time. + * Test for EL2 present, and defer test for SEL2 to runtime. + */ + if (s->current_el <= 1 && arm_dc_feature(s, ARM_FEATURE_EL2)) { + gen_helper_vesb(cpu_env); + } + } + return true; +} + static bool trans_NOP(DisasContext *s, arg_NOP *a) { return true; -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-18-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 1 + target/arm/cpu_tcg.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_PMULL (PMULL, PMULL2 instructions) - FEAT_PMUv3p1 (PMU Extensions v3.1) - FEAT_PMUv3p4 (PMU Extensions v3.4) +- FEAT_RAS (Reliability, availability, and serviceability) - FEAT_RDM (Advanced SIMD rounding double multiply accumulate instructions) - FEAT_RNG (Random number generator) - FEAT_SB (Speculation Barrier) diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) 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 */ + t = FIELD_DP64(t, ID_AA64PFR0, RAS, 1); /* FEAT_RAS */ t = FIELD_DP64(t, ID_AA64PFR0, SVE, 1); t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ void aa32_max_features(ARMCPU *cpu) t = cpu->isar.id_pfr0; t = FIELD_DP32(t, ID_PFR0, DIT, 1); /* FEAT_DIT */ + t = FIELD_DP32(t, ID_PFR0, RAS, 1); /* FEAT_RAS */ cpu->isar.id_pfr0 = t; t = cpu->isar.id_pfr2; -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This feature is AArch64 only, and applies to physical SErrors, which QEMU does not implement, thus the feature is a nop. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-19-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 1 + 2 files changed, 2 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_FlagM2 (Enhancements to flag manipulation instructions) - FEAT_HPDS (Hierarchical permission disables) - FEAT_I8MM (AArch64 Int8 matrix multiplication instructions) +- FEAT_IESB (Implicit error synchronization event) - FEAT_JSCVT (JavaScript conversion instructions) - FEAT_LOR (Limited ordering regions) - FEAT_LPA (Large Physical Address space) diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = cpu->isar.id_aa64mmfr2; t = FIELD_DP64(t, ID_AA64MMFR2, CNP, 1); /* FEAT_TTCNP */ t = FIELD_DP64(t, ID_AA64MMFR2, UAO, 1); /* FEAT_UAO */ + t = FIELD_DP64(t, ID_AA64MMFR2, IESB, 1); /* FEAT_IESB */ t = FIELD_DP64(t, ID_AA64MMFR2, VARANGE, 1); /* FEAT_LVA */ t = FIELD_DP64(t, ID_AA64MMFR2, ST, 1); /* FEAT_TTST */ t = FIELD_DP64(t, ID_AA64MMFR2, TTL, 1); /* FEAT_TTL */ -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This extension concerns branch speculation, which TCG does not implement. Thus we can trivially enable this feature. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-20-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 1 + target/arm/cpu_tcg.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_BBM at level 2 (Translation table break-before-make levels) - FEAT_BF16 (AArch64 BFloat16 instructions) - FEAT_BTI (Branch Target Identification) +- FEAT_CSV2 (Cache speculation variant 2) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64PFR0, SVE, 1); t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ + t = FIELD_DP64(t, ID_AA64PFR0, CSV2, 1); /* FEAT_CSV2 */ cpu->isar.id_aa64pfr0 = t; t = cpu->isar.id_aa64pfr1; diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ void aa32_max_features(ARMCPU *cpu) cpu->isar.id_mmfr4 = t; t = cpu->isar.id_pfr0; + t = FIELD_DP32(t, ID_PFR0, CSV2, 2); /* FEAT_CVS2 */ t = FIELD_DP32(t, ID_PFR0, DIT, 1); /* FEAT_DIT */ t = FIELD_DP32(t, ID_PFR0, RAS, 1); /* FEAT_RAS */ cpu->isar.id_pfr0 = t; -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> There is no branch prediction in TCG, therefore there is no need to actually include the context number into the predictor. Therefore all we need to do is add the state for SCXTNUM_ELx. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-21-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 3 ++ target/arm/cpu.h | 16 +++++++++ target/arm/cpu.c | 5 +++ target/arm/cpu64.c | 3 +- target/arm/helper.c | 61 ++++++++++++++++++++++++++++++++++- 5 files changed, 86 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_BF16 (AArch64 BFloat16 instructions) - FEAT_BTI (Branch Target Identification) - 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) +- FEAT_CSV2_2 (Cache speculation variant 2, version 2) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) 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 @@ typedef struct CPUArchState { ARMPACKey apdb; ARMPACKey apga; } keys; + + uint64_t scxtnum_el[4]; #endif #if defined(CONFIG_USER_ONLY) @@ -XXX,XX +XXX,XX @@ void pmu_init(ARMCPU *cpu); #define SCTLR_WXN (1U << 19) #define SCTLR_ST (1U << 20) /* up to ??, RAZ in v6 */ #define SCTLR_UWXN (1U << 20) /* v7 onward, AArch32 only */ +#define SCTLR_TSCXT (1U << 20) /* FEAT_CSV2_1p2, AArch64 only */ #define SCTLR_FI (1U << 21) /* up to v7, v8 RES0 */ #define SCTLR_IESB (1U << 21) /* v8.2-IESB, AArch64 only */ #define SCTLR_U (1U << 22) /* up to v6, RAO in v7 */ @@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_dit(const ARMISARegisters *id) return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, DIT) != 0; } +static inline bool isar_feature_aa64_scxtnum(const ARMISARegisters *id) +{ + int key = FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, CSV2); + if (key >= 2) { + return true; /* FEAT_CSV2_2 */ + } + if (key == 1) { + key = FIELD_EX64(id->id_aa64pfr1, ID_AA64PFR1, CSV2_FRAC); + return key >= 2; /* FEAT_CSV2_1p2 */ + } + return false; +} + static inline bool isar_feature_aa64_ssbs(const ARMISARegisters *id) { return FIELD_EX64(id->id_aa64pfr1, ID_AA64PFR1, SSBS) != 0; 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_reset(DeviceState *dev) */ env->cp15.gcr_el1 = 0x1ffff; } + /* + * Disable access to SCXTNUM_EL0 from CSV2_1p2. + * This is not yet exposed from the Linux kernel in any way. + */ + env->cp15.sctlr_el[1] |= SCTLR_TSCXT; #else /* Reset into the highest available EL */ if (arm_feature(env, ARM_FEATURE_EL3)) { diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64PFR0, SVE, 1); t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ - t = FIELD_DP64(t, ID_AA64PFR0, CSV2, 1); /* FEAT_CSV2 */ + t = FIELD_DP64(t, ID_AA64PFR0, CSV2, 2); /* FEAT_CSV2_2 */ cpu->isar.id_aa64pfr0 = t; t = cpu->isar.id_aa64pfr1; @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) * we do for EL2 with the virtualization=on property. */ t = FIELD_DP64(t, ID_AA64PFR1, MTE, 3); /* FEAT_MTE3 */ + t = FIELD_DP64(t, ID_AA64PFR1, CSV2_FRAC, 0); /* FEAT_CSV2_2 */ cpu->isar.id_aa64pfr1 = t; t = cpu->isar.id_aa64mmfr0; 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 scr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) if (cpu_isar_feature(aa64_mte, cpu)) { valid_mask |= SCR_ATA; } + if (cpu_isar_feature(aa64_scxtnum, cpu)) { + valid_mask |= SCR_ENSCXT; + } } else { valid_mask &= ~(SCR_RW | SCR_ST); if (cpu_isar_feature(aa32_ras, cpu)) { @@ -XXX,XX +XXX,XX @@ static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask) if (cpu_isar_feature(aa64_mte, cpu)) { valid_mask |= HCR_ATA | HCR_DCT | HCR_TID5; } + if (cpu_isar_feature(aa64_scxtnum, cpu)) { + valid_mask |= HCR_ENSCXT; + } } /* Clear RES0 bits. */ @@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu) { K(3, 0, 5, 6, 0), K(3, 4, 5, 6, 0), K(3, 5, 5, 6, 0), "TFSR_EL1", "TFSR_EL2", "TFSR_EL12", isar_feature_aa64_mte }, + { K(3, 0, 13, 0, 7), K(3, 4, 13, 0, 7), K(3, 5, 13, 0, 7), + "SCXTNUM_EL1", "SCXTNUM_EL2", "SCXTNUM_EL12", + isar_feature_aa64_scxtnum }, + /* TODO: ARMv8.2-SPE -- PMSCR_EL2 */ /* TODO: ARMv8.4-Trace -- TRFCR_EL2 */ }; @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo mte_el0_cacheop_reginfo[] = { }, }; -#endif +static CPAccessResult access_scxtnum(CPUARMState *env, const ARMCPRegInfo *ri, + bool isread) +{ + uint64_t hcr = arm_hcr_el2_eff(env); + int el = arm_current_el(env); + + if (el == 0 && !((hcr & HCR_E2H) && (hcr & HCR_TGE))) { + if (env->cp15.sctlr_el[1] & SCTLR_TSCXT) { + if (hcr & HCR_TGE) { + return CP_ACCESS_TRAP_EL2; + } + return CP_ACCESS_TRAP; + } + } else if (el < 2 && (env->cp15.sctlr_el[2] & SCTLR_TSCXT)) { + return CP_ACCESS_TRAP_EL2; + } + if (el < 2 && arm_is_el2_enabled(env) && !(hcr & HCR_ENSCXT)) { + return CP_ACCESS_TRAP_EL2; + } + if (el < 3 + && arm_feature(env, ARM_FEATURE_EL3) + && !(env->cp15.scr_el3 & SCR_ENSCXT)) { + return CP_ACCESS_TRAP_EL3; + } + return CP_ACCESS_OK; +} + +static const ARMCPRegInfo scxtnum_reginfo[] = { + { .name = "SCXTNUM_EL0", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 3, .crn = 13, .crm = 0, .opc2 = 7, + .access = PL0_RW, .accessfn = access_scxtnum, + .fieldoffset = offsetof(CPUARMState, scxtnum_el[0]) }, + { .name = "SCXTNUM_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 0, .crn = 13, .crm = 0, .opc2 = 7, + .access = PL1_RW, .accessfn = access_scxtnum, + .fieldoffset = offsetof(CPUARMState, scxtnum_el[1]) }, + { .name = "SCXTNUM_EL2", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 4, .crn = 13, .crm = 0, .opc2 = 7, + .access = PL2_RW, .accessfn = access_scxtnum, + .fieldoffset = offsetof(CPUARMState, scxtnum_el[2]) }, + { .name = "SCXTNUM_EL3", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 6, .crn = 13, .crm = 0, .opc2 = 7, + .access = PL3_RW, + .fieldoffset = offsetof(CPUARMState, scxtnum_el[3]) }, +}; +#endif /* TARGET_AARCH64 */ static CPAccessResult access_predinv(CPUARMState *env, const ARMCPRegInfo *ri, bool isread) @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) define_arm_cp_regs(cpu, mte_tco_ro_reginfo); define_arm_cp_regs(cpu, mte_el0_cacheop_reginfo); } + + if (cpu_isar_feature(aa64_scxtnum, cpu)) { + define_arm_cp_regs(cpu, scxtnum_reginfo); + } #endif if (cpu_isar_feature(any_predinv, cpu)) { -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This extension concerns cache speculation, which TCG does not implement. Thus we can trivially enable this feature. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-22-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 1 + target/arm/cpu_tcg.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_CSV2_1p1 (Cache speculation variant 2, version 1.1) - FEAT_CSV2_1p2 (Cache speculation variant 2, version 1.2) - FEAT_CSV2_2 (Cache speculation variant 2, version 2) +- FEAT_CSV3 (Cache speculation variant 3) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ t = FIELD_DP64(t, ID_AA64PFR0, CSV2, 2); /* FEAT_CSV2_2 */ + t = FIELD_DP64(t, ID_AA64PFR0, CSV3, 1); /* FEAT_CSV3 */ cpu->isar.id_aa64pfr0 = t; t = cpu->isar.id_aa64pfr1; diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu_tcg.c +++ b/target/arm/cpu_tcg.c @@ -XXX,XX +XXX,XX @@ void aa32_max_features(ARMCPU *cpu) cpu->isar.id_pfr0 = t; t = cpu->isar.id_pfr2; + t = FIELD_DP32(t, ID_PFR2, CSV3, 1); /* FEAT_CSV3 */ t = FIELD_DP32(t, ID_PFR2, SSBS, 1); /* FEAT_SSBS */ cpu->isar.id_pfr2 = t; -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This extension concerns not merging memory access, which TCG does not implement. Thus we can trivially enable this feature. Add a comment to handle_hint for the DGH instruction, but no code. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-23-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 1 + target/arm/translate-a64.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_CSV2_1p2 (Cache speculation variant 2, version 1.2) - FEAT_CSV2_2 (Cache speculation variant 2, version 2) - FEAT_CSV3 (Cache speculation variant 3) +- FEAT_DGH (Data gathering hint) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64ISAR1, SB, 1); /* FEAT_SB */ t = FIELD_DP64(t, ID_AA64ISAR1, SPECRES, 1); /* FEAT_SPECRES */ t = FIELD_DP64(t, ID_AA64ISAR1, BF16, 1); /* FEAT_BF16 */ + t = FIELD_DP64(t, ID_AA64ISAR1, DGH, 1); /* FEAT_DGH */ t = FIELD_DP64(t, ID_AA64ISAR1, I8MM, 1); /* FEAT_I8MM */ cpu->isar.id_aa64isar1 = t; diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void handle_hint(DisasContext *s, uint32_t insn, break; case 0b00100: /* SEV */ case 0b00101: /* SEVL */ + case 0b00110: /* DGH */ /* we treat all as NOP at least for now */ break; case 0b00111: /* XPACLRI */ -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Enable the a76 for virt and sbsa board use. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-24-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/virt.rst | 1 + hw/arm/sbsa-ref.c | 1 + hw/arm/virt.c | 1 + target/arm/cpu64.c | 66 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 69 insertions(+) diff --git a/docs/system/arm/virt.rst b/docs/system/arm/virt.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/virt.rst +++ b/docs/system/arm/virt.rst @@ -XXX,XX +XXX,XX @@ Supported guest CPU types: - ``cortex-a53`` (64-bit) - ``cortex-a57`` (64-bit) - ``cortex-a72`` (64-bit) +- ``cortex-a76`` (64-bit) - ``a64fx`` (64-bit) - ``host`` (with KVM only) - ``max`` (same as ``host`` for KVM; best possible emulation with TCG) 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 @@ static const int sbsa_ref_irqmap[] = { static const char * const valid_cpus[] = { ARM_CPU_TYPE_NAME("cortex-a57"), ARM_CPU_TYPE_NAME("cortex-a72"), + ARM_CPU_TYPE_NAME("cortex-a76"), ARM_CPU_TYPE_NAME("max"), }; diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ static const char *valid_cpus[] = { ARM_CPU_TYPE_NAME("cortex-a53"), ARM_CPU_TYPE_NAME("cortex-a57"), ARM_CPU_TYPE_NAME("cortex-a72"), + ARM_CPU_TYPE_NAME("cortex-a76"), ARM_CPU_TYPE_NAME("a64fx"), ARM_CPU_TYPE_NAME("host"), ARM_CPU_TYPE_NAME("max"), diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_a72_initfn(Object *obj) define_cortex_a72_a57_a53_cp_reginfo(cpu); } +static void aarch64_a76_initfn(Object *obj) +{ + ARMCPU *cpu = ARM_CPU(obj); + + cpu->dtb_compatible = "arm,cortex-a76"; + set_feature(&cpu->env, ARM_FEATURE_V8); + set_feature(&cpu->env, ARM_FEATURE_NEON); + set_feature(&cpu->env, ARM_FEATURE_GENERIC_TIMER); + set_feature(&cpu->env, ARM_FEATURE_AARCH64); + set_feature(&cpu->env, ARM_FEATURE_CBAR_RO); + set_feature(&cpu->env, ARM_FEATURE_EL2); + set_feature(&cpu->env, ARM_FEATURE_EL3); + set_feature(&cpu->env, ARM_FEATURE_PMU); + + /* Ordered by B2.4 AArch64 registers by functional group */ + cpu->clidr = 0x82000023; + cpu->ctr = 0x8444C004; + cpu->dcz_blocksize = 4; + cpu->isar.id_aa64dfr0 = 0x0000000010305408ull; + cpu->isar.id_aa64isar0 = 0x0000100010211120ull; + cpu->isar.id_aa64isar1 = 0x0000000000100001ull; + cpu->isar.id_aa64mmfr0 = 0x0000000000101122ull; + cpu->isar.id_aa64mmfr1 = 0x0000000010212122ull; + cpu->isar.id_aa64mmfr2 = 0x0000000000001011ull; + cpu->isar.id_aa64pfr0 = 0x1100000010111112ull; /* GIC filled in later */ + cpu->isar.id_aa64pfr1 = 0x0000000000000010ull; + cpu->id_afr0 = 0x00000000; + cpu->isar.id_dfr0 = 0x04010088; + cpu->isar.id_isar0 = 0x02101110; + cpu->isar.id_isar1 = 0x13112111; + cpu->isar.id_isar2 = 0x21232042; + cpu->isar.id_isar3 = 0x01112131; + cpu->isar.id_isar4 = 0x00010142; + cpu->isar.id_isar5 = 0x01011121; + cpu->isar.id_isar6 = 0x00000010; + cpu->isar.id_mmfr0 = 0x10201105; + cpu->isar.id_mmfr1 = 0x40000000; + cpu->isar.id_mmfr2 = 0x01260000; + cpu->isar.id_mmfr3 = 0x02122211; + cpu->isar.id_mmfr4 = 0x00021110; + cpu->isar.id_pfr0 = 0x10010131; + cpu->isar.id_pfr1 = 0x00010000; /* GIC filled in later */ + cpu->isar.id_pfr2 = 0x00000011; + cpu->midr = 0x414fd0b1; /* r4p1 */ + cpu->revidr = 0; + + /* From B2.18 CCSIDR_EL1 */ + cpu->ccsidr[0] = 0x701fe01a; /* 64KB L1 dcache */ + cpu->ccsidr[1] = 0x201fe01a; /* 64KB L1 icache */ + cpu->ccsidr[2] = 0x707fe03a; /* 512KB L2 cache */ + + /* From B2.93 SCTLR_EL3 */ + cpu->reset_sctlr = 0x30c50838; + + /* From B4.23 ICH_VTR_EL2 */ + cpu->gic_num_lrs = 4; + cpu->gic_vpribits = 5; + cpu->gic_vprebits = 5; + + /* From B5.1 AdvSIMD AArch64 register summary */ + cpu->isar.mvfr0 = 0x10110222; + cpu->isar.mvfr1 = 0x13211111; + cpu->isar.mvfr2 = 0x00000043; +} + void arm_cpu_sve_finalize(ARMCPU *cpu, Error **errp) { /* @@ -XXX,XX +XXX,XX @@ static const ARMCPUInfo aarch64_cpus[] = { { .name = "cortex-a57", .initfn = aarch64_a57_initfn }, { .name = "cortex-a53", .initfn = aarch64_a53_initfn }, { .name = "cortex-a72", .initfn = aarch64_a72_initfn }, + { .name = "cortex-a76", .initfn = aarch64_a76_initfn }, { .name = "a64fx", .initfn = aarch64_a64fx_initfn }, { .name = "max", .initfn = aarch64_max_initfn }, #if defined(CONFIG_KVM) || defined(CONFIG_HVF) -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Enable the n1 for virt and sbsa board use. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-25-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/virt.rst | 1 + hw/arm/sbsa-ref.c | 1 + hw/arm/virt.c | 1 + target/arm/cpu64.c | 66 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 69 insertions(+) diff --git a/docs/system/arm/virt.rst b/docs/system/arm/virt.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/virt.rst +++ b/docs/system/arm/virt.rst @@ -XXX,XX +XXX,XX @@ Supported guest CPU types: - ``cortex-a76`` (64-bit) - ``a64fx`` (64-bit) - ``host`` (with KVM only) +- ``neoverse-n1`` (64-bit) - ``max`` (same as ``host`` for KVM; best possible emulation with TCG) Note that the default is ``cortex-a15``, so for an AArch64 guest you must 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 @@ static const char * const valid_cpus[] = { ARM_CPU_TYPE_NAME("cortex-a57"), ARM_CPU_TYPE_NAME("cortex-a72"), ARM_CPU_TYPE_NAME("cortex-a76"), + ARM_CPU_TYPE_NAME("neoverse-n1"), ARM_CPU_TYPE_NAME("max"), }; diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ static const char *valid_cpus[] = { ARM_CPU_TYPE_NAME("cortex-a72"), ARM_CPU_TYPE_NAME("cortex-a76"), ARM_CPU_TYPE_NAME("a64fx"), + ARM_CPU_TYPE_NAME("neoverse-n1"), ARM_CPU_TYPE_NAME("host"), ARM_CPU_TYPE_NAME("max"), }; diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_a76_initfn(Object *obj) cpu->isar.mvfr2 = 0x00000043; } +static void aarch64_neoverse_n1_initfn(Object *obj) +{ + ARMCPU *cpu = ARM_CPU(obj); + + cpu->dtb_compatible = "arm,neoverse-n1"; + set_feature(&cpu->env, ARM_FEATURE_V8); + set_feature(&cpu->env, ARM_FEATURE_NEON); + set_feature(&cpu->env, ARM_FEATURE_GENERIC_TIMER); + set_feature(&cpu->env, ARM_FEATURE_AARCH64); + set_feature(&cpu->env, ARM_FEATURE_CBAR_RO); + set_feature(&cpu->env, ARM_FEATURE_EL2); + set_feature(&cpu->env, ARM_FEATURE_EL3); + set_feature(&cpu->env, ARM_FEATURE_PMU); + + /* Ordered by B2.4 AArch64 registers by functional group */ + cpu->clidr = 0x82000023; + cpu->ctr = 0x8444c004; + cpu->dcz_blocksize = 4; + cpu->isar.id_aa64dfr0 = 0x0000000110305408ull; + cpu->isar.id_aa64isar0 = 0x0000100010211120ull; + cpu->isar.id_aa64isar1 = 0x0000000000100001ull; + cpu->isar.id_aa64mmfr0 = 0x0000000000101125ull; + cpu->isar.id_aa64mmfr1 = 0x0000000010212122ull; + cpu->isar.id_aa64mmfr2 = 0x0000000000001011ull; + cpu->isar.id_aa64pfr0 = 0x1100000010111112ull; /* GIC filled in later */ + cpu->isar.id_aa64pfr1 = 0x0000000000000020ull; + cpu->id_afr0 = 0x00000000; + cpu->isar.id_dfr0 = 0x04010088; + cpu->isar.id_isar0 = 0x02101110; + cpu->isar.id_isar1 = 0x13112111; + cpu->isar.id_isar2 = 0x21232042; + cpu->isar.id_isar3 = 0x01112131; + cpu->isar.id_isar4 = 0x00010142; + cpu->isar.id_isar5 = 0x01011121; + cpu->isar.id_isar6 = 0x00000010; + cpu->isar.id_mmfr0 = 0x10201105; + cpu->isar.id_mmfr1 = 0x40000000; + cpu->isar.id_mmfr2 = 0x01260000; + cpu->isar.id_mmfr3 = 0x02122211; + cpu->isar.id_mmfr4 = 0x00021110; + cpu->isar.id_pfr0 = 0x10010131; + cpu->isar.id_pfr1 = 0x00010000; /* GIC filled in later */ + cpu->isar.id_pfr2 = 0x00000011; + cpu->midr = 0x414fd0c1; /* r4p1 */ + cpu->revidr = 0; + + /* From B2.23 CCSIDR_EL1 */ + cpu->ccsidr[0] = 0x701fe01a; /* 64KB L1 dcache */ + cpu->ccsidr[1] = 0x201fe01a; /* 64KB L1 icache */ + cpu->ccsidr[2] = 0x70ffe03a; /* 1MB L2 cache */ + + /* From B2.98 SCTLR_EL3 */ + cpu->reset_sctlr = 0x30c50838; + + /* From B4.23 ICH_VTR_EL2 */ + cpu->gic_num_lrs = 4; + cpu->gic_vpribits = 5; + cpu->gic_vprebits = 5; + + /* From B5.1 AdvSIMD AArch64 register summary */ + cpu->isar.mvfr0 = 0x10110222; + cpu->isar.mvfr1 = 0x13211111; + cpu->isar.mvfr2 = 0x00000043; +} + void arm_cpu_sve_finalize(ARMCPU *cpu, Error **errp) { /* @@ -XXX,XX +XXX,XX @@ static const ARMCPUInfo aarch64_cpus[] = { { .name = "cortex-a72", .initfn = aarch64_a72_initfn }, { .name = "cortex-a76", .initfn = aarch64_a76_initfn }, { .name = "a64fx", .initfn = aarch64_a64fx_initfn }, + { .name = "neoverse-n1", .initfn = aarch64_neoverse_n1_initfn }, { .name = "max", .initfn = aarch64_max_initfn }, #if defined(CONFIG_KVM) || defined(CONFIG_HVF) { .name = "host", .initfn = aarch64_host_initfn }, -- 2.25.1 From: Leif Lindholm <quic_llindhol@quicinc.com> The sbsa-ref machine is continuously evolving. Some of the changes we want to make in the near future, to align with real components (e.g. the GIC-700), will break compatibility for existing firmware. Introduce two new properties to the DT generated on machine generation: - machine-version-major To be incremented when a platform change makes the machine incompatible with existing firmware. - machine-version-minor To be incremented when functionality is added to the machine without causing incompatibility with existing firmware. to be reset to 0 when machine-version-major is incremented. This versioning scheme is *neither*: - A QEMU versioned machine type; a given version of QEMU will emulate a given version of the platform. - A reflection of level of SBSA (now SystemReady SR) support provided. The version will increment on guest-visible functional changes only, akin to a revision ID register found on a physical platform. These properties are both introduced with the value 0. (Hence, a machine where the DT is lacking these nodes is equivalent to version 0.0.) Signed-off-by: Leif Lindholm <quic_llindhol@quicinc.com> Message-id: 20220505113947.75714-1-quic_llindhol@quicinc.com Cc: Peter Maydell <peter.maydell@linaro.org> Cc: Radoslaw Biernacki <rad@semihalf.com> Cc: Cédric Le Goater <clg@kaod.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/sbsa-ref.c | 14 ++++++++++++++ 1 file changed, 14 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 @@ static void create_fdt(SBSAMachineState *sms) qemu_fdt_setprop_cell(fdt, "/", "#address-cells", 0x2); qemu_fdt_setprop_cell(fdt, "/", "#size-cells", 0x2); + /* + * This versioning scheme is for informing platform fw only. It is neither: + * - A QEMU versioned machine type; a given version of QEMU will emulate + * a given version of the platform. + * - A reflection of level of SBSA (now SystemReady SR) support provided. + * + * machine-version-major: updated when changes breaking fw compatibility + * are introduced. + * machine-version-minor: updated when features are added that don't break + * fw compatibility. + */ + qemu_fdt_setprop_cell(fdt, "/", "machine-version-major", 0); + qemu_fdt_setprop_cell(fdt, "/", "machine-version-minor", 0); + if (ms->numa_state->have_numa_distance) { int size = nb_numa_nodes * nb_numa_nodes * 3 * sizeof(uint32_t); uint32_t *matrix = g_malloc0(size); -- 2.25.1 From: Gavin Shan <gshan@redhat.com> This adds cluster-id in CPU instance properties, which will be used by arm/virt machine. Besides, the cluster-id is also verified or dumped in various spots: * hw/core/machine.c::machine_set_cpu_numa_node() to associate CPU with its NUMA node. * hw/core/machine.c::machine_numa_finish_cpu_init() to record CPU slots with no NUMA mapping set. * hw/core/machine-hmp-cmds.c::hmp_hotpluggable_cpus() to dump cluster-id. Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20220503140304.855514-2-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- qapi/machine.json | 6 ++++-- hw/core/machine-hmp-cmds.c | 4 ++++ hw/core/machine.c | 16 ++++++++++++++++ 3 files changed, 24 insertions(+), 2 deletions(-) diff --git a/qapi/machine.json b/qapi/machine.json index XXXXXXX..XXXXXXX 100644 --- a/qapi/machine.json +++ b/qapi/machine.json @@ -XXX,XX +XXX,XX @@ # @node-id: NUMA node ID the CPU belongs to # @socket-id: socket number within node/board the CPU belongs to # @die-id: die number within socket the CPU belongs to (since 4.1) -# @core-id: core number within die the CPU belongs to +# @cluster-id: cluster number within die the CPU belongs to (since 7.1) +# @core-id: core number within cluster the CPU belongs to # @thread-id: thread number within core the CPU belongs to # -# Note: currently there are 5 properties that could be present +# Note: currently there are 6 properties that could be present # but management should be prepared to pass through other # properties with device_add command to allow for future # interface extension. This also requires the filed names to be kept in @@ -XXX,XX +XXX,XX @@ 'data': { '*node-id': 'int', '*socket-id': 'int', '*die-id': 'int', + '*cluster-id': 'int', '*core-id': 'int', '*thread-id': 'int' } diff --git a/hw/core/machine-hmp-cmds.c b/hw/core/machine-hmp-cmds.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/machine-hmp-cmds.c +++ b/hw/core/machine-hmp-cmds.c @@ -XXX,XX +XXX,XX @@ void hmp_hotpluggable_cpus(Monitor *mon, const QDict *qdict) if (c->has_die_id) { monitor_printf(mon, " die-id: \"%" PRIu64 "\"\n", c->die_id); } + if (c->has_cluster_id) { + monitor_printf(mon, " cluster-id: \"%" PRIu64 "\"\n", + c->cluster_id); + } if (c->has_core_id) { monitor_printf(mon, " core-id: \"%" PRIu64 "\"\n", c->core_id); } diff --git a/hw/core/machine.c b/hw/core/machine.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/machine.c +++ b/hw/core/machine.c @@ -XXX,XX +XXX,XX @@ void machine_set_cpu_numa_node(MachineState *machine, return; } + if (props->has_cluster_id && !slot->props.has_cluster_id) { + error_setg(errp, "cluster-id is not supported"); + return; + } + if (props->has_socket_id && !slot->props.has_socket_id) { error_setg(errp, "socket-id is not supported"); return; @@ -XXX,XX +XXX,XX @@ void machine_set_cpu_numa_node(MachineState *machine, continue; } + if (props->has_cluster_id && + props->cluster_id != slot->props.cluster_id) { + continue; + } + if (props->has_die_id && props->die_id != slot->props.die_id) { continue; } @@ -XXX,XX +XXX,XX @@ static char *cpu_slot_to_string(const CPUArchId *cpu) } g_string_append_printf(s, "die-id: %"PRId64, cpu->props.die_id); } + if (cpu->props.has_cluster_id) { + if (s->len) { + g_string_append_printf(s, ", "); + } + g_string_append_printf(s, "cluster-id: %"PRId64, cpu->props.cluster_id); + } if (cpu->props.has_core_id) { if (s->len) { g_string_append_printf(s, ", "); -- 2.25.1 From: Gavin Shan <gshan@redhat.com> The CPU topology isn't enabled on arm/virt machine yet, but we're going to do it in next patch. After the CPU topology is enabled by next patch, "thread-id=1" becomes invalid because the CPU core is preferred on arm/virt machine. It means these two CPUs have 0/1 as their core IDs, but their thread IDs are all 0. It will trigger test failure as the following message indicates: [14/21 qemu:qtest+qtest-aarch64 / qtest-aarch64/numa-test ERROR 1.48s killed by signal 6 SIGABRT >>> G_TEST_DBUS_DAEMON=/home/gavin/sandbox/qemu.main/tests/dbus-vmstate-daemon.sh \ QTEST_QEMU_STORAGE_DAEMON_BINARY=./storage-daemon/qemu-storage-daemon \ QTEST_QEMU_BINARY=./qemu-system-aarch64 \ QTEST_QEMU_IMG=./qemu-img MALLOC_PERTURB_=83 \ /home/gavin/sandbox/qemu.main/build/tests/qtest/numa-test --tap -k ―――――――――――――――――――――――――――――――――――――――――――――― stderr: qemu-system-aarch64: -numa cpu,node-id=0,thread-id=1: no match found This fixes the issue by providing comprehensive SMP configurations in aarch64_numa_cpu(). The SMP configurations aren't used before the CPU topology is enabled in next patch. Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Message-id: 20220503140304.855514-3-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/qtest/numa-test.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/qtest/numa-test.c b/tests/qtest/numa-test.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/numa-test.c +++ b/tests/qtest/numa-test.c @@ -XXX,XX +XXX,XX @@ static void aarch64_numa_cpu(const void *data) QTestState *qts; g_autofree char *cli = NULL; - cli = make_cli(data, "-machine smp.cpus=2 " + cli = make_cli(data, "-machine " + "smp.cpus=2,smp.sockets=1,smp.clusters=1,smp.cores=1,smp.threads=2 " "-numa node,nodeid=0,memdev=ram -numa node,nodeid=1 " "-numa cpu,node-id=1,thread-id=0 " "-numa cpu,node-id=0,thread-id=1"); -- 2.25.1 From: Gavin Shan <gshan@redhat.com> Currently, the SMP configuration isn't considered when the CPU topology is populated. In this case, it's impossible to provide the default CPU-to-NUMA mapping or association based on the socket ID of the given CPU. This takes account of SMP configuration when the CPU topology is populated. The die ID for the given CPU isn't assigned since it's not supported on arm/virt machine. Besides, the used SMP configuration in qtest/numa-test/aarch64_numa_cpu() is corrcted to avoid testing failure Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20220503140304.855514-4-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ static const CPUArchIdList *virt_possible_cpu_arch_ids(MachineState *ms) int n; unsigned int max_cpus = ms->smp.max_cpus; VirtMachineState *vms = VIRT_MACHINE(ms); + MachineClass *mc = MACHINE_GET_CLASS(vms); if (ms->possible_cpus) { assert(ms->possible_cpus->len == max_cpus); @@ -XXX,XX +XXX,XX @@ static const CPUArchIdList *virt_possible_cpu_arch_ids(MachineState *ms) ms->possible_cpus->cpus[n].type = ms->cpu_type; ms->possible_cpus->cpus[n].arch_id = virt_cpu_mp_affinity(vms, n); + + assert(!mc->smp_props.dies_supported); + ms->possible_cpus->cpus[n].props.has_socket_id = true; + ms->possible_cpus->cpus[n].props.socket_id = + n / (ms->smp.clusters * ms->smp.cores * ms->smp.threads); + ms->possible_cpus->cpus[n].props.has_cluster_id = true; + ms->possible_cpus->cpus[n].props.cluster_id = + (n / (ms->smp.cores * ms->smp.threads)) % ms->smp.clusters; + ms->possible_cpus->cpus[n].props.has_core_id = true; + ms->possible_cpus->cpus[n].props.core_id = + (n / ms->smp.threads) % ms->smp.cores; ms->possible_cpus->cpus[n].props.has_thread_id = true; - ms->possible_cpus->cpus[n].props.thread_id = n; + ms->possible_cpus->cpus[n].props.thread_id = + n % ms->smp.threads; } return ms->possible_cpus; } -- 2.25.1 From: Gavin Shan <gshan@redhat.com> In aarch64_numa_cpu(), the CPU and NUMA association is something like below. Two threads in the same core/cluster/socket are associated with two individual NUMA nodes, which is unreal as Igor Mammedov mentioned. We don't expect the association to break NUMA-to-socket boundary, which matches with the real world. NUMA-node socket cluster core thread ------------------------------------------ 0 0 0 0 0 1 0 0 0 1 This corrects the topology for CPUs and their association with NUMA nodes. After this patch is applied, the CPU and NUMA association becomes something like below, which looks real. Besides, socket/cluster/core/thread IDs are all checked when the NUMA node IDs are verified. It helps to check if the CPU topology is properly populated or not. NUMA-node socket cluster core thread ------------------------------------------ 0 1 0 0 0 1 0 0 0 0 Suggested-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Gavin Shan <gshan@redhat.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20220503140304.855514-5-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/qtest/numa-test.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/tests/qtest/numa-test.c b/tests/qtest/numa-test.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/numa-test.c +++ b/tests/qtest/numa-test.c @@ -XXX,XX +XXX,XX @@ static void aarch64_numa_cpu(const void *data) g_autofree char *cli = NULL; cli = make_cli(data, "-machine " - "smp.cpus=2,smp.sockets=1,smp.clusters=1,smp.cores=1,smp.threads=2 " + "smp.cpus=2,smp.sockets=2,smp.clusters=1,smp.cores=1,smp.threads=1 " "-numa node,nodeid=0,memdev=ram -numa node,nodeid=1 " - "-numa cpu,node-id=1,thread-id=0 " - "-numa cpu,node-id=0,thread-id=1"); + "-numa cpu,node-id=0,socket-id=1,cluster-id=0,core-id=0,thread-id=0 " + "-numa cpu,node-id=1,socket-id=0,cluster-id=0,core-id=0,thread-id=0"); qts = qtest_init(cli); cpus = get_cpus(qts, &resp); g_assert(cpus); while ((e = qlist_pop(cpus))) { QDict *cpu, *props; - int64_t thread, node; + int64_t socket, cluster, core, thread, node; cpu = qobject_to(QDict, e); g_assert(qdict_haskey(cpu, "props")); @@ -XXX,XX +XXX,XX @@ static void aarch64_numa_cpu(const void *data) g_assert(qdict_haskey(props, "node-id")); node = qdict_get_int(props, "node-id"); + g_assert(qdict_haskey(props, "socket-id")); + socket = qdict_get_int(props, "socket-id"); + g_assert(qdict_haskey(props, "cluster-id")); + cluster = qdict_get_int(props, "cluster-id"); + g_assert(qdict_haskey(props, "core-id")); + core = qdict_get_int(props, "core-id"); g_assert(qdict_haskey(props, "thread-id")); thread = qdict_get_int(props, "thread-id"); - if (thread == 0) { + if (socket == 0 && cluster == 0 && core == 0 && thread == 0) { g_assert_cmpint(node, ==, 1); - } else if (thread == 1) { + } else if (socket == 1 && cluster == 0 && core == 0 && thread == 0) { g_assert_cmpint(node, ==, 0); } else { g_assert(false); -- 2.25.1 From: Gavin Shan <gshan@redhat.com> When CPU-to-NUMA association isn't explicitly provided by users, the default one is given by mc->get_default_cpu_node_id(). However, the CPU topology isn't fully considered in the default association and this causes CPU topology broken warnings on booting Linux guest. For example, the following warning messages are observed when the Linux guest is booted with the following command lines. /home/gavin/sandbox/qemu.main/build/qemu-system-aarch64 \ -accel kvm -machine virt,gic-version=host \ -cpu host \ -smp 6,sockets=2,cores=3,threads=1 \ -m 1024M,slots=16,maxmem=64G \ -object memory-backend-ram,id=mem0,size=128M \ -object memory-backend-ram,id=mem1,size=128M \ -object memory-backend-ram,id=mem2,size=128M \ -object memory-backend-ram,id=mem3,size=128M \ -object memory-backend-ram,id=mem4,size=128M \ -object memory-backend-ram,id=mem4,size=384M \ -numa node,nodeid=0,memdev=mem0 \ -numa node,nodeid=1,memdev=mem1 \ -numa node,nodeid=2,memdev=mem2 \ -numa node,nodeid=3,memdev=mem3 \ -numa node,nodeid=4,memdev=mem4 \ -numa node,nodeid=5,memdev=mem5 : alternatives: patching kernel code BUG: arch topology borken the CLS domain not a subset of the MC domain <the above error log repeats> BUG: arch topology borken the DIE domain not a subset of the NODE domain With current implementation of mc->get_default_cpu_node_id(), CPU#0 to CPU#5 are associated with NODE#0 to NODE#5 separately. That's incorrect because CPU#0/1/2 should be associated with same NUMA node because they're seated in same socket. This fixes the issue by considering the socket ID when the default CPU-to-NUMA association is provided in virt_possible_cpu_arch_ids(). With this applied, no more CPU topology broken warnings are seen from the Linux guest. The 6 CPUs are associated with NODE#0/1, but there are no CPUs associated with NODE#2/3/4/5. Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Message-id: 20220503140304.855514-6-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ virt_cpu_index_to_props(MachineState *ms, unsigned cpu_index) static int64_t virt_get_default_cpu_node_id(const MachineState *ms, int idx) { - return idx % ms->numa_state->num_nodes; + int64_t socket_id = ms->possible_cpus->cpus[idx].props.socket_id; + + return socket_id % ms->numa_state->num_nodes; } static const CPUArchIdList *virt_possible_cpu_arch_ids(MachineState *ms) -- 2.25.1 From: Gavin Shan <gshan@redhat.com> When the PPTT table is built, the CPU topology is re-calculated, but it's unecessary because the CPU topology has been populated in virt_possible_cpu_arch_ids() on arm/virt machine. This reworks build_pptt() to avoid by reusing the existing IDs in ms->possible_cpus. Currently, the only user of build_pptt() is arm/virt machine. Signed-off-by: Gavin Shan <gshan@redhat.com> Tested-by: Yanan Wang <wangyanan55@huawei.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Message-id: 20220503140304.855514-7-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/acpi/aml-build.c | 111 +++++++++++++++++++------------------------- 1 file changed, 48 insertions(+), 63 deletions(-) diff --git a/hw/acpi/aml-build.c b/hw/acpi/aml-build.c index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/aml-build.c +++ b/hw/acpi/aml-build.c @@ -XXX,XX +XXX,XX @@ void build_pptt(GArray *table_data, BIOSLinker *linker, MachineState *ms, const char *oem_id, const char *oem_table_id) { MachineClass *mc = MACHINE_GET_CLASS(ms); - GQueue *list = g_queue_new(); - guint pptt_start = table_data->len; - guint parent_offset; - guint length, i; - int uid = 0; - int socket; + CPUArchIdList *cpus = ms->possible_cpus; + int64_t socket_id = -1, cluster_id = -1, core_id = -1; + uint32_t socket_offset = 0, cluster_offset = 0, core_offset = 0; + uint32_t pptt_start = table_data->len; + int n; AcpiTable table = { .sig = "PPTT", .rev = 2, .oem_id = oem_id, .oem_table_id = oem_table_id }; acpi_table_begin(&table, table_data); - for (socket = 0; socket < ms->smp.sockets; socket++) { - g_queue_push_tail(list, - GUINT_TO_POINTER(table_data->len - pptt_start)); - build_processor_hierarchy_node( - table_data, - /* - * Physical package - represents the boundary - * of a physical package - */ - (1 << 0), - 0, socket, NULL, 0); - } - - if (mc->smp_props.clusters_supported) { - length = g_queue_get_length(list); - for (i = 0; i < length; i++) { - int cluster; - - parent_offset = GPOINTER_TO_UINT(g_queue_pop_head(list)); - for (cluster = 0; cluster < ms->smp.clusters; cluster++) { - g_queue_push_tail(list, - GUINT_TO_POINTER(table_data->len - pptt_start)); - build_processor_hierarchy_node( - table_data, - (0 << 0), /* not a physical package */ - parent_offset, cluster, NULL, 0); - } + /* + * This works with the assumption that cpus[n].props.*_id has been + * sorted from top to down levels in mc->possible_cpu_arch_ids(). + * Otherwise, the unexpected and duplicated containers will be + * created. + */ + for (n = 0; n < cpus->len; n++) { + if (cpus->cpus[n].props.socket_id != socket_id) { + assert(cpus->cpus[n].props.socket_id > socket_id); + socket_id = cpus->cpus[n].props.socket_id; + cluster_id = -1; + core_id = -1; + socket_offset = table_data->len - pptt_start; + build_processor_hierarchy_node(table_data, + (1 << 0), /* Physical package */ + 0, socket_id, NULL, 0); } - } - length = g_queue_get_length(list); - for (i = 0; i < length; i++) { - int core; - - parent_offset = GPOINTER_TO_UINT(g_queue_pop_head(list)); - for (core = 0; core < ms->smp.cores; core++) { - if (ms->smp.threads > 1) { - g_queue_push_tail(list, - GUINT_TO_POINTER(table_data->len - pptt_start)); - build_processor_hierarchy_node( - table_data, - (0 << 0), /* not a physical package */ - parent_offset, core, NULL, 0); - } else { - build_processor_hierarchy_node( - table_data, - (1 << 1) | /* ACPI Processor ID valid */ - (1 << 3), /* Node is a Leaf */ - parent_offset, uid++, NULL, 0); + if (mc->smp_props.clusters_supported) { + if (cpus->cpus[n].props.cluster_id != cluster_id) { + assert(cpus->cpus[n].props.cluster_id > cluster_id); + cluster_id = cpus->cpus[n].props.cluster_id; + core_id = -1; + cluster_offset = table_data->len - pptt_start; + build_processor_hierarchy_node(table_data, + (0 << 0), /* Not a physical package */ + socket_offset, cluster_id, NULL, 0); } + } else { + cluster_offset = socket_offset; } - } - length = g_queue_get_length(list); - for (i = 0; i < length; i++) { - int thread; + if (ms->smp.threads == 1) { + build_processor_hierarchy_node(table_data, + (1 << 1) | /* ACPI Processor ID valid */ + (1 << 3), /* Node is a Leaf */ + cluster_offset, n, NULL, 0); + } else { + if (cpus->cpus[n].props.core_id != core_id) { + assert(cpus->cpus[n].props.core_id > core_id); + core_id = cpus->cpus[n].props.core_id; + core_offset = table_data->len - pptt_start; + build_processor_hierarchy_node(table_data, + (0 << 0), /* Not a physical package */ + cluster_offset, core_id, NULL, 0); + } - parent_offset = GPOINTER_TO_UINT(g_queue_pop_head(list)); - for (thread = 0; thread < ms->smp.threads; thread++) { - build_processor_hierarchy_node( - table_data, + build_processor_hierarchy_node(table_data, (1 << 1) | /* ACPI Processor ID valid */ (1 << 2) | /* Processor is a Thread */ (1 << 3), /* Node is a Leaf */ - parent_offset, uid++, NULL, 0); + core_offset, n, NULL, 0); } } - g_queue_free(list); acpi_table_end(linker, &table); } -- 2.25.1