diff options
Diffstat (limited to 'target-i386')
-rw-r--r-- | target-i386/cpu.c | 642 | ||||
-rw-r--r-- | target-i386/cpu.h | 55 | ||||
-rw-r--r-- | target-i386/fpu_helper.c | 108 | ||||
-rw-r--r-- | target-i386/helper.c | 19 | ||||
-rw-r--r-- | target-i386/kvm.c | 104 | ||||
-rw-r--r-- | target-i386/kvm_i386.h | 2 | ||||
-rw-r--r-- | target-i386/monitor.c | 3 | ||||
-rw-r--r-- | target-i386/seg_helper.c | 36 | ||||
-rw-r--r-- | target-i386/translate.c | 8 |
9 files changed, 560 insertions, 417 deletions
diff --git a/target-i386/cpu.c b/target-i386/cpu.c index 6a1afab595..1c57fce81b 100644 --- a/target-i386/cpu.c +++ b/target-i386/cpu.c @@ -57,6 +57,7 @@ #define CPUID_2_L1D_32KB_8WAY_64B 0x2c #define CPUID_2_L1I_32KB_8WAY_64B 0x30 #define CPUID_2_L2_2MB_8WAY_64B 0x7d +#define CPUID_2_L3_16MB_16WAY_64B 0x4d /* CPUID Leaf 4 constants: */ @@ -131,11 +132,18 @@ #define L2_LINES_PER_TAG 1 #define L2_SIZE_KB_AMD 512 -/* No L3 cache: */ +/* Level 3 unified cache: */ #define L3_SIZE_KB 0 /* disabled */ #define L3_ASSOCIATIVITY 0 /* disabled */ #define L3_LINES_PER_TAG 0 /* disabled */ #define L3_LINE_SIZE 0 /* disabled */ +#define L3_N_LINE_SIZE 64 +#define L3_N_ASSOCIATIVITY 16 +#define L3_N_SETS 16384 +#define L3_N_PARTITIONS 1 +#define L3_N_DESCRIPTOR CPUID_2_L3_16MB_16WAY_64B +#define L3_N_LINES_PER_TAG 1 +#define L3_N_SIZE_KB_AMD 16384 /* TLB definitions: */ @@ -173,181 +181,6 @@ static void x86_cpu_vendor_words2str(char *dst, uint32_t vendor1, dst[CPUID_VENDOR_SZ] = '\0'; } -/* feature flags taken from "Intel Processor Identification and the CPUID - * Instruction" and AMD's "CPUID Specification". In cases of disagreement - * between feature naming conventions, aliases may be added. - */ -static const char *feature_name[] = { - "fpu", "vme", "de", "pse", - "tsc", "msr", "pae", "mce", - "cx8", "apic", NULL, "sep", - "mtrr", "pge", "mca", "cmov", - "pat", "pse36", "pn" /* Intel psn */, "clflush" /* Intel clfsh */, - NULL, "ds" /* Intel dts */, "acpi", "mmx", - "fxsr", "sse", "sse2", "ss", - "ht" /* Intel htt */, "tm", "ia64", "pbe", -}; -static const char *ext_feature_name[] = { - "pni|sse3" /* Intel,AMD sse3 */, "pclmulqdq|pclmuldq", "dtes64", "monitor", - "ds_cpl", "vmx", "smx", "est", - "tm2", "ssse3", "cid", NULL, - "fma", "cx16", "xtpr", "pdcm", - NULL, "pcid", "dca", "sse4.1|sse4_1", - "sse4.2|sse4_2", "x2apic", "movbe", "popcnt", - "tsc-deadline", "aes", "xsave", "osxsave", - "avx", "f16c", "rdrand", "hypervisor", -}; -/* Feature names that are already defined on feature_name[] but are set on - * CPUID[8000_0001].EDX on AMD CPUs don't have their names on - * ext2_feature_name[]. They are copied automatically to cpuid_ext2_features - * if and only if CPU vendor is AMD. - */ -static const char *ext2_feature_name[] = { - NULL /* fpu */, NULL /* vme */, NULL /* de */, NULL /* pse */, - NULL /* tsc */, NULL /* msr */, NULL /* pae */, NULL /* mce */, - NULL /* cx8 */ /* AMD CMPXCHG8B */, NULL /* apic */, NULL, "syscall", - NULL /* mtrr */, NULL /* pge */, NULL /* mca */, NULL /* cmov */, - NULL /* pat */, NULL /* pse36 */, NULL, NULL /* Linux mp */, - "nx|xd", NULL, "mmxext", NULL /* mmx */, - NULL /* fxsr */, "fxsr_opt|ffxsr", "pdpe1gb" /* AMD Page1GB */, "rdtscp", - NULL, "lm|i64", "3dnowext", "3dnow", -}; -static const char *ext3_feature_name[] = { - "lahf_lm" /* AMD LahfSahf */, "cmp_legacy", "svm", "extapic" /* AMD ExtApicSpace */, - "cr8legacy" /* AMD AltMovCr8 */, "abm", "sse4a", "misalignsse", - "3dnowprefetch", "osvw", "ibs", "xop", - "skinit", "wdt", NULL, "lwp", - "fma4", "tce", NULL, "nodeid_msr", - NULL, "tbm", "topoext", "perfctr_core", - "perfctr_nb", NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *ext4_feature_name[] = { - NULL, NULL, "xstore", "xstore-en", - NULL, NULL, "xcrypt", "xcrypt-en", - "ace2", "ace2-en", "phe", "phe-en", - "pmm", "pmm-en", NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *kvm_feature_name[] = { - "kvmclock", "kvm_nopiodelay", "kvm_mmu", "kvmclock", - "kvm_asyncpf", "kvm_steal_time", "kvm_pv_eoi", "kvm_pv_unhalt", - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - "kvmclock-stable-bit", NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *hyperv_priv_feature_name[] = { - NULL /* hv_msr_vp_runtime_access */, NULL /* hv_msr_time_refcount_access */, - NULL /* hv_msr_synic_access */, NULL /* hv_msr_stimer_access */, - NULL /* hv_msr_apic_access */, NULL /* hv_msr_hypercall_access */, - NULL /* hv_vpindex_access */, NULL /* hv_msr_reset_access */, - NULL /* hv_msr_stats_access */, NULL /* hv_reftsc_access */, - NULL /* hv_msr_idle_access */, NULL /* hv_msr_frequency_access */, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *hyperv_ident_feature_name[] = { - NULL /* hv_create_partitions */, NULL /* hv_access_partition_id */, - NULL /* hv_access_memory_pool */, NULL /* hv_adjust_message_buffers */, - NULL /* hv_post_messages */, NULL /* hv_signal_events */, - NULL /* hv_create_port */, NULL /* hv_connect_port */, - NULL /* hv_access_stats */, NULL, NULL, NULL /* hv_debugging */, - NULL /* hv_cpu_power_management */, NULL /* hv_configure_profiler */, - NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *hyperv_misc_feature_name[] = { - NULL /* hv_mwait */, NULL /* hv_guest_debugging */, - NULL /* hv_perf_monitor */, NULL /* hv_cpu_dynamic_part */, - NULL /* hv_hypercall_params_xmm */, NULL /* hv_guest_idle_state */, - NULL, NULL, - NULL, NULL, NULL /* hv_guest_crash_msr */, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *svm_feature_name[] = { - "npt", "lbrv", "svm_lock", "nrip_save", - "tsc_scale", "vmcb_clean", "flushbyasid", "decodeassists", - NULL, NULL, "pause_filter", NULL, - "pfthreshold", NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *cpuid_7_0_ebx_feature_name[] = { - "fsgsbase", "tsc_adjust", NULL, "bmi1", "hle", "avx2", NULL, "smep", - "bmi2", "erms", "invpcid", "rtm", NULL, NULL, "mpx", NULL, - "avx512f", NULL, "rdseed", "adx", "smap", NULL, "pcommit", "clflushopt", - "clwb", NULL, "avx512pf", "avx512er", "avx512cd", NULL, NULL, NULL, -}; - -static const char *cpuid_7_0_ecx_feature_name[] = { - NULL, NULL, "umip", "pku", - "ospke", NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, "rdpid", NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *cpuid_apm_edx_feature_name[] = { - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - "invtsc", NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *cpuid_xsave_feature_name[] = { - "xsaveopt", "xsavec", "xgetbv1", "xsaves", - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - -static const char *cpuid_6_feature_name[] = { - NULL, NULL, "arat", NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, -}; - #define I486_FEATURES (CPUID_FP87 | CPUID_VME | CPUID_PSE) #define PENTIUM_FEATURES (I486_FEATURES | CPUID_DE | CPUID_TSC | \ CPUID_MSR | CPUID_MCE | CPUID_CX8 | CPUID_MMX | CPUID_APIC) @@ -413,96 +246,266 @@ static const char *cpuid_6_feature_name[] = { CPUID_XSAVE_XSAVEC, CPUID_XSAVE_XSAVES */ typedef struct FeatureWordInfo { - const char **feat_names; + /* feature flags names are taken from "Intel Processor Identification and + * the CPUID Instruction" and AMD's "CPUID Specification". + * In cases of disagreement between feature naming conventions, + * aliases may be added. + */ + const char *feat_names[32]; uint32_t cpuid_eax; /* Input EAX for CPUID */ bool cpuid_needs_ecx; /* CPUID instruction uses ECX as input */ uint32_t cpuid_ecx; /* Input ECX value for CPUID */ int cpuid_reg; /* output register (R_* constant) */ uint32_t tcg_features; /* Feature flags supported by TCG */ uint32_t unmigratable_flags; /* Feature flags known to be unmigratable */ + uint32_t migratable_flags; /* Feature flags known to be migratable */ } FeatureWordInfo; static FeatureWordInfo feature_word_info[FEATURE_WORDS] = { [FEAT_1_EDX] = { - .feat_names = feature_name, + .feat_names = { + "fpu", "vme", "de", "pse", + "tsc", "msr", "pae", "mce", + "cx8", "apic", NULL, "sep", + "mtrr", "pge", "mca", "cmov", + "pat", "pse36", "pn" /* Intel psn */, "clflush" /* Intel clfsh */, + NULL, "ds" /* Intel dts */, "acpi", "mmx", + "fxsr", "sse", "sse2", "ss", + "ht" /* Intel htt */, "tm", "ia64", "pbe", + }, .cpuid_eax = 1, .cpuid_reg = R_EDX, .tcg_features = TCG_FEATURES, }, [FEAT_1_ECX] = { - .feat_names = ext_feature_name, + .feat_names = { + "pni|sse3" /* Intel,AMD sse3 */, "pclmulqdq|pclmuldq", "dtes64", "monitor", + "ds_cpl", "vmx", "smx", "est", + "tm2", "ssse3", "cid", NULL, + "fma", "cx16", "xtpr", "pdcm", + NULL, "pcid", "dca", "sse4.1|sse4_1", + "sse4.2|sse4_2", "x2apic", "movbe", "popcnt", + "tsc-deadline", "aes", "xsave", "osxsave", + "avx", "f16c", "rdrand", "hypervisor", + }, .cpuid_eax = 1, .cpuid_reg = R_ECX, .tcg_features = TCG_EXT_FEATURES, }, + /* Feature names that are already defined on feature_name[] but + * are set on CPUID[8000_0001].EDX on AMD CPUs don't have their + * names on feat_names below. They are copied automatically + * to features[FEAT_8000_0001_EDX] if and only if CPU vendor is AMD. + */ [FEAT_8000_0001_EDX] = { - .feat_names = ext2_feature_name, + .feat_names = { + NULL /* fpu */, NULL /* vme */, NULL /* de */, NULL /* pse */, + NULL /* tsc */, NULL /* msr */, NULL /* pae */, NULL /* mce */, + NULL /* cx8 */, NULL /* apic */, NULL, "syscall", + NULL /* mtrr */, NULL /* pge */, NULL /* mca */, NULL /* cmov */, + NULL /* pat */, NULL /* pse36 */, NULL, NULL /* Linux mp */, + "nx|xd", NULL, "mmxext", NULL /* mmx */, + NULL /* fxsr */, "fxsr_opt|ffxsr", "pdpe1gb", "rdtscp", + NULL, "lm|i64", "3dnowext", "3dnow", + }, .cpuid_eax = 0x80000001, .cpuid_reg = R_EDX, .tcg_features = TCG_EXT2_FEATURES, }, [FEAT_8000_0001_ECX] = { - .feat_names = ext3_feature_name, + .feat_names = { + "lahf_lm", "cmp_legacy", "svm", "extapic", + "cr8legacy", "abm", "sse4a", "misalignsse", + "3dnowprefetch", "osvw", "ibs", "xop", + "skinit", "wdt", NULL, "lwp", + "fma4", "tce", NULL, "nodeid_msr", + NULL, "tbm", "topoext", "perfctr_core", + "perfctr_nb", NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 0x80000001, .cpuid_reg = R_ECX, .tcg_features = TCG_EXT3_FEATURES, }, [FEAT_C000_0001_EDX] = { - .feat_names = ext4_feature_name, + .feat_names = { + NULL, NULL, "xstore", "xstore-en", + NULL, NULL, "xcrypt", "xcrypt-en", + "ace2", "ace2-en", "phe", "phe-en", + "pmm", "pmm-en", NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 0xC0000001, .cpuid_reg = R_EDX, .tcg_features = TCG_EXT4_FEATURES, }, [FEAT_KVM] = { - .feat_names = kvm_feature_name, + .feat_names = { + "kvmclock", "kvm_nopiodelay", "kvm_mmu", "kvmclock", + "kvm_asyncpf", "kvm_steal_time", "kvm_pv_eoi", "kvm_pv_unhalt", + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + "kvmclock-stable-bit", NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = KVM_CPUID_FEATURES, .cpuid_reg = R_EAX, .tcg_features = TCG_KVM_FEATURES, }, [FEAT_HYPERV_EAX] = { - .feat_names = hyperv_priv_feature_name, + .feat_names = { + NULL /* hv_msr_vp_runtime_access */, NULL /* hv_msr_time_refcount_access */, + NULL /* hv_msr_synic_access */, NULL /* hv_msr_stimer_access */, + NULL /* hv_msr_apic_access */, NULL /* hv_msr_hypercall_access */, + NULL /* hv_vpindex_access */, NULL /* hv_msr_reset_access */, + NULL /* hv_msr_stats_access */, NULL /* hv_reftsc_access */, + NULL /* hv_msr_idle_access */, NULL /* hv_msr_frequency_access */, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 0x40000003, .cpuid_reg = R_EAX, }, [FEAT_HYPERV_EBX] = { - .feat_names = hyperv_ident_feature_name, + .feat_names = { + NULL /* hv_create_partitions */, NULL /* hv_access_partition_id */, + NULL /* hv_access_memory_pool */, NULL /* hv_adjust_message_buffers */, + NULL /* hv_post_messages */, NULL /* hv_signal_events */, + NULL /* hv_create_port */, NULL /* hv_connect_port */, + NULL /* hv_access_stats */, NULL, NULL, NULL /* hv_debugging */, + NULL /* hv_cpu_power_management */, NULL /* hv_configure_profiler */, + NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 0x40000003, .cpuid_reg = R_EBX, }, [FEAT_HYPERV_EDX] = { - .feat_names = hyperv_misc_feature_name, + .feat_names = { + NULL /* hv_mwait */, NULL /* hv_guest_debugging */, + NULL /* hv_perf_monitor */, NULL /* hv_cpu_dynamic_part */, + NULL /* hv_hypercall_params_xmm */, NULL /* hv_guest_idle_state */, + NULL, NULL, + NULL, NULL, NULL /* hv_guest_crash_msr */, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 0x40000003, .cpuid_reg = R_EDX, }, [FEAT_SVM] = { - .feat_names = svm_feature_name, + .feat_names = { + "npt", "lbrv", "svm_lock", "nrip_save", + "tsc_scale", "vmcb_clean", "flushbyasid", "decodeassists", + NULL, NULL, "pause_filter", NULL, + "pfthreshold", NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 0x8000000A, .cpuid_reg = R_EDX, .tcg_features = TCG_SVM_FEATURES, }, [FEAT_7_0_EBX] = { - .feat_names = cpuid_7_0_ebx_feature_name, + .feat_names = { + "fsgsbase", "tsc_adjust", NULL, "bmi1", + "hle", "avx2", NULL, "smep", + "bmi2", "erms", "invpcid", "rtm", + NULL, NULL, "mpx", NULL, + "avx512f", "avx512dq", "rdseed", "adx", + "smap", "avx512ifma", "pcommit", "clflushopt", + "clwb", NULL, "avx512pf", "avx512er", + "avx512cd", NULL, "avx512bw", "avx512vl", + }, .cpuid_eax = 7, .cpuid_needs_ecx = true, .cpuid_ecx = 0, .cpuid_reg = R_EBX, .tcg_features = TCG_7_0_EBX_FEATURES, }, [FEAT_7_0_ECX] = { - .feat_names = cpuid_7_0_ecx_feature_name, + .feat_names = { + NULL, "avx512vbmi", "umip", "pku", + "ospke", NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, "rdpid", NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 7, .cpuid_needs_ecx = true, .cpuid_ecx = 0, .cpuid_reg = R_ECX, .tcg_features = TCG_7_0_ECX_FEATURES, }, [FEAT_8000_0007_EDX] = { - .feat_names = cpuid_apm_edx_feature_name, + .feat_names = { + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + "invtsc", NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 0x80000007, .cpuid_reg = R_EDX, .tcg_features = TCG_APM_FEATURES, .unmigratable_flags = CPUID_APM_INVTSC, }, [FEAT_XSAVE] = { - .feat_names = cpuid_xsave_feature_name, + .feat_names = { + "xsaveopt", "xsavec", "xgetbv1", "xsaves", + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 0xd, .cpuid_needs_ecx = true, .cpuid_ecx = 1, .cpuid_reg = R_EAX, .tcg_features = TCG_XSAVE_FEATURES, }, [FEAT_6_EAX] = { - .feat_names = cpuid_6_feature_name, + .feat_names = { + NULL, NULL, "arat", NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, .cpuid_eax = 6, .cpuid_reg = R_EAX, .tcg_features = TCG_6_EAX_FEATURES, }, + [FEAT_XSAVE_COMP_LO] = { + .cpuid_eax = 0xD, + .cpuid_needs_ecx = true, .cpuid_ecx = 0, + .cpuid_reg = R_EAX, + .tcg_features = ~0U, + .migratable_flags = XSTATE_FP_MASK | XSTATE_SSE_MASK | + XSTATE_YMM_MASK | XSTATE_BNDREGS_MASK | XSTATE_BNDCSR_MASK | + XSTATE_OPMASK_MASK | XSTATE_ZMM_Hi256_MASK | XSTATE_Hi16_ZMM_MASK | + XSTATE_PKRU_MASK, + }, + [FEAT_XSAVE_COMP_HI] = { + .cpuid_eax = 0xD, + .cpuid_needs_ecx = true, .cpuid_ecx = 0, + .cpuid_reg = R_EDX, + .tcg_features = ~0U, + }, }; typedef struct X86RegisterInfo32 { @@ -526,7 +529,12 @@ static const X86RegisterInfo32 x86_reg_info_32[CPU_NB_REGS32] = { }; #undef REGISTER -const ExtSaveArea x86_ext_save_areas[] = { +typedef struct ExtSaveArea { + uint32_t feature, bits; + uint32_t offset, size; +} ExtSaveArea; + +static const ExtSaveArea x86_ext_save_areas[] = { [XSTATE_YMM_BIT] = { .feature = FEAT_1_ECX, .bits = CPUID_EXT_AVX, .offset = offsetof(X86XSaveArea, avx_state), @@ -557,6 +565,26 @@ const ExtSaveArea x86_ext_save_areas[] = { .size = sizeof(XSavePKRU) }, }; +static uint32_t xsave_area_size(uint64_t mask) +{ + int i; + uint64_t ret = sizeof(X86LegacyXSaveArea) + sizeof(X86XSaveHeader); + + for (i = 2; i < ARRAY_SIZE(x86_ext_save_areas); i++) { + const ExtSaveArea *esa = &x86_ext_save_areas[i]; + if ((mask >> i) & 1) { + ret = MAX(ret, esa->offset + esa->size); + } + } + return ret; +} + +static inline uint64_t x86_cpu_xsave_components(X86CPU *cpu) +{ + return ((uint64_t)cpu->env.features[FEAT_XSAVE_COMP_HI]) << 32 | + cpu->env.features[FEAT_XSAVE_COMP_LO]; +} + const char *get_register_name_32(unsigned int reg) { if (reg >= CPU_NB_REGS32) { @@ -577,15 +605,13 @@ static uint32_t x86_cpu_get_migratable_flags(FeatureWord w) for (i = 0; i < 32; i++) { uint32_t f = 1U << i; - /* If the feature name is unknown, it is not supported by QEMU yet */ - if (!wi->feat_names[i]) { - continue; - } - /* Skip features known to QEMU, but explicitly marked as unmigratable */ - if (wi->unmigratable_flags & f) { - continue; + + /* If the feature name is known, it is implicitly considered migratable, + * unless it is explicitly set in unmigratable_flags */ + if ((wi->migratable_flags & f) || + (wi->feat_names[i] && !(wi->unmigratable_flags & f))) { + r |= f; } - r |= f; } return r; } @@ -694,8 +720,7 @@ static void add_flagname_to_bitmaps(const char *flagname, FeatureWord w; for (w = 0; w < FEATURE_WORDS; w++) { FeatureWordInfo *wi = &feature_word_info[w]; - if (wi->feat_names && - lookup_feature(&words[w], flagname, NULL, wi->feat_names)) { + if (lookup_feature(&words[w], flagname, NULL, wi->feat_names)) { break; } } @@ -744,7 +769,6 @@ struct X86CPUDefinition { const char *name; uint32_t level; uint32_t xlevel; - uint32_t xlevel2; /* vendor is zero-terminated, 12 character ASCII string */ char vendor[CPUID_VENDOR_SZ + 1]; int family; @@ -1404,9 +1428,9 @@ static X86CPUDefinition builtin_x86_defs[] = { .name = "Opteron_G3", .level = 5, .vendor = CPUID_VENDOR_AMD, - .family = 15, - .model = 6, - .stepping = 1, + .family = 16, + .model = 2, + .stepping = 3, .features[FEAT_1_EDX] = CPUID_VME | CPUID_SSE2 | CPUID_SSE | CPUID_FXSR | CPUID_MMX | CPUID_CLFLUSH | CPUID_PSE36 | CPUID_PAT | CPUID_CMOV | CPUID_MCA | @@ -1627,9 +1651,12 @@ static void host_x86_cpu_initfn(Object *obj) /* If KVM is disabled, x86_cpu_realizefn() will report an error later */ if (kvm_enabled()) { - env->cpuid_level = kvm_arch_get_supported_cpuid(s, 0x0, 0, R_EAX); - env->cpuid_xlevel = kvm_arch_get_supported_cpuid(s, 0x80000000, 0, R_EAX); - env->cpuid_xlevel2 = kvm_arch_get_supported_cpuid(s, 0xC0000000, 0, R_EAX); + env->cpuid_min_level = + kvm_arch_get_supported_cpuid(s, 0x0, 0, R_EAX); + env->cpuid_min_xlevel = + kvm_arch_get_supported_cpuid(s, 0x80000000, 0, R_EAX); + env->cpuid_min_xlevel2 = + kvm_arch_get_supported_cpuid(s, 0xC0000000, 0, R_EAX); if (lmce_supported()) { object_property_set_bool(OBJECT(cpu), true, "lmce", &error_abort); @@ -2192,12 +2219,13 @@ static void x86_cpu_load_def(X86CPU *cpu, X86CPUDefinition *def, Error **errp) char host_vendor[CPUID_VENDOR_SZ + 1]; FeatureWord w; - object_property_set_int(OBJECT(cpu), def->level, "level", errp); + /* CPU models only set _minimum_ values for level/xlevel: */ + object_property_set_int(OBJECT(cpu), def->level, "min-level", errp); + object_property_set_int(OBJECT(cpu), def->xlevel, "min-xlevel", errp); + object_property_set_int(OBJECT(cpu), def->family, "family", errp); object_property_set_int(OBJECT(cpu), def->model, "model", errp); object_property_set_int(OBJECT(cpu), def->stepping, "stepping", errp); - object_property_set_int(OBJECT(cpu), def->xlevel, "xlevel", errp); - object_property_set_int(OBJECT(cpu), def->xlevel2, "xlevel2", errp); object_property_set_str(OBJECT(cpu), def->model_id, "model-id", errp); for (w = 0; w < FEATURE_WORDS; w++) { env->features[w] = def->features[w]; @@ -2275,6 +2303,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, { X86CPU *cpu = x86_env_get_cpu(env); CPUState *cs = CPU(cpu); + uint32_t pkg_offset; /* test if maximum index reached */ if (index & 0x80000000) { @@ -2328,7 +2357,11 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, } *eax = 1; /* Number of CPUID[EAX=2] calls required */ *ebx = 0; - *ecx = 0; + if (!cpu->enable_l3_cache) { + *ecx = 0; + } else { + *ecx = L3_N_DESCRIPTOR; + } *edx = (L1D_DESCRIPTOR << 16) | \ (L1I_DESCRIPTOR << 8) | \ (L2_DESCRIPTOR); @@ -2374,6 +2407,25 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, *ecx = L2_SETS - 1; *edx = CPUID_4_NO_INVD_SHARING; break; + case 3: /* L3 cache info */ + if (!cpu->enable_l3_cache) { + *eax = 0; + *ebx = 0; + *ecx = 0; + *edx = 0; + break; + } + *eax |= CPUID_4_TYPE_UNIFIED | \ + CPUID_4_LEVEL(3) | \ + CPUID_4_SELF_INIT_LEVEL; + pkg_offset = apicid_pkg_offset(cs->nr_cores, cs->nr_threads); + *eax |= ((1 << pkg_offset) - 1) << 14; + *ebx = (L3_N_LINE_SIZE - 1) | \ + ((L3_N_PARTITIONS - 1) << 12) | \ + ((L3_N_ASSOCIATIVITY - 1) << 22); + *ecx = L3_N_SETS - 1; + *edx = CPUID_4_INCLUSIVE | CPUID_4_COMPLEX_IDX; + break; default: /* end of info */ *eax = 0; *ebx = 0; @@ -2454,13 +2506,13 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, switch (count) { case 0: - *eax = apicid_core_offset(smp_cores, smp_threads); - *ebx = smp_threads; + *eax = apicid_core_offset(cs->nr_cores, cs->nr_threads); + *ebx = cs->nr_threads; *ecx |= CPUID_TOPOLOGY_LEVEL_SMT; break; case 1: - *eax = apicid_pkg_offset(smp_cores, smp_threads); - *ebx = smp_cores * smp_threads; + *eax = apicid_pkg_offset(cs->nr_cores, cs->nr_threads); + *ebx = cs->nr_cores * cs->nr_threads; *ecx |= CPUID_TOPOLOGY_LEVEL_CORE; break; default: @@ -2473,10 +2525,6 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, *ebx &= 0xffff; /* The count doesn't need to be reliable. */ break; case 0xD: { - KVMState *s = cs->kvm_state; - uint64_t ena_mask; - int i; - /* Processor Extended State */ *eax = 0; *ebx = 0; @@ -2485,36 +2533,17 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, if (!(env->features[FEAT_1_ECX] & CPUID_EXT_XSAVE)) { break; } - if (kvm_enabled()) { - ena_mask = kvm_arch_get_supported_cpuid(s, 0xd, 0, R_EDX); - ena_mask <<= 32; - ena_mask |= kvm_arch_get_supported_cpuid(s, 0xd, 0, R_EAX); - } else { - ena_mask = -1; - } if (count == 0) { - *ecx = 0x240; - for (i = 2; i < ARRAY_SIZE(x86_ext_save_areas); i++) { - const ExtSaveArea *esa = &x86_ext_save_areas[i]; - if ((env->features[esa->feature] & esa->bits) == esa->bits - && ((ena_mask >> i) & 1) != 0) { - if (i < 32) { - *eax |= 1u << i; - } else { - *edx |= 1u << (i - 32); - } - *ecx = MAX(*ecx, esa->offset + esa->size); - } - } - *eax |= ena_mask & (XSTATE_FP_MASK | XSTATE_SSE_MASK); + *ecx = xsave_area_size(x86_cpu_xsave_components(cpu)); + *eax = env->features[FEAT_XSAVE_COMP_LO]; + *edx = env->features[FEAT_XSAVE_COMP_HI]; *ebx = *ecx; } else if (count == 1) { *eax = env->features[FEAT_XSAVE]; } else if (count < ARRAY_SIZE(x86_ext_save_areas)) { - const ExtSaveArea *esa = &x86_ext_save_areas[count]; - if ((env->features[esa->feature] & esa->bits) == esa->bits - && ((ena_mask >> count) & 1) != 0) { + if ((x86_cpu_xsave_components(cpu) >> count) & 1) { + const ExtSaveArea *esa = &x86_ext_save_areas[count]; *eax = esa->size; *ebx = esa->offset; } @@ -2585,9 +2614,15 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, *ecx = (L2_SIZE_KB_AMD << 16) | \ (AMD_ENC_ASSOC(L2_ASSOCIATIVITY) << 12) | \ (L2_LINES_PER_TAG << 8) | (L2_LINE_SIZE); - *edx = ((L3_SIZE_KB/512) << 18) | \ - (AMD_ENC_ASSOC(L3_ASSOCIATIVITY) << 12) | \ - (L3_LINES_PER_TAG << 8) | (L3_LINE_SIZE); + if (!cpu->enable_l3_cache) { + *edx = ((L3_SIZE_KB / 512) << 18) | \ + (AMD_ENC_ASSOC(L3_ASSOCIATIVITY) << 12) | \ + (L3_LINES_PER_TAG << 8) | (L3_LINE_SIZE); + } else { + *edx = ((L3_N_SIZE_KB_AMD / 512) << 18) | \ + (AMD_ENC_ASSOC(L3_N_ASSOCIATIVITY) << 12) | \ + (L3_N_LINES_PER_TAG << 8) | (L3_N_LINE_SIZE); + } break; case 0x80000007: *eax = 0; @@ -2669,7 +2704,7 @@ static void x86_cpu_reset(CPUState *s) xcc->parent_reset(s); - memset(env, 0, offsetof(CPUX86State, cpuid_level)); + memset(env, 0, offsetof(CPUX86State, end_reset_fields)); tlb_flush(s, 1); @@ -2743,7 +2778,7 @@ static void x86_cpu_reset(CPUState *s) } for (i = 2; i < ARRAY_SIZE(x86_ext_save_areas); i++) { const ExtSaveArea *esa = &x86_ext_save_areas[i]; - if ((env->features[esa->feature] & esa->bits) == esa->bits) { + if (env->features[esa->feature] & esa->bits) { xcr0 |= 1ull << i; } } @@ -2906,6 +2941,61 @@ static uint32_t x86_host_phys_bits(void) return host_phys_bits; } +static void x86_cpu_adjust_level(X86CPU *cpu, uint32_t *min, uint32_t value) +{ + if (*min < value) { + *min = value; + } +} + +/* Increase cpuid_min_{level,xlevel,xlevel2} automatically, if appropriate */ +static void x86_cpu_adjust_feat_level(X86CPU *cpu, FeatureWord w) +{ + CPUX86State *env = &cpu->env; + FeatureWordInfo *fi = &feature_word_info[w]; + uint32_t eax = fi->cpuid_eax; + uint32_t region = eax & 0xF0000000; + + if (!env->features[w]) { + return; + } + + switch (region) { + case 0x00000000: + x86_cpu_adjust_level(cpu, &env->cpuid_min_level, eax); + break; + case 0x80000000: + x86_cpu_adjust_level(cpu, &env->cpuid_min_xlevel, eax); + break; + case 0xC0000000: + x86_cpu_adjust_level(cpu, &env->cpuid_min_xlevel2, eax); + break; + } +} + +/* Calculate XSAVE components based on the configured CPU feature flags */ +static void x86_cpu_enable_xsave_components(X86CPU *cpu) +{ + CPUX86State *env = &cpu->env; + int i; + uint64_t mask; + + if (!(env->features[FEAT_1_ECX] & CPUID_EXT_XSAVE)) { + return; + } + + mask = (XSTATE_FP_MASK | XSTATE_SSE_MASK); + for (i = 2; i < ARRAY_SIZE(x86_ext_save_areas); i++) { + const ExtSaveArea *esa = &x86_ext_save_areas[i]; + if (env->features[esa->feature] & esa->bits) { + mask |= (1ULL << i); + } + } + + env->features[FEAT_XSAVE_COMP_LO] = mask; + env->features[FEAT_XSAVE_COMP_HI] = mask >> 32; +} + #define IS_INTEL_CPU(env) ((env)->cpuid_vendor1 == CPUID_VENDOR_INTEL_1 && \ (env)->cpuid_vendor2 == CPUID_VENDOR_INTEL_2 && \ (env)->cpuid_vendor3 == CPUID_VENDOR_INTEL_3) @@ -2951,8 +3041,40 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp) cpu->env.features[w] &= ~minus_features[w]; } - if (env->features[FEAT_7_0_EBX] && env->cpuid_level < 7) { - env->cpuid_level = 7; + if (!kvm_enabled() || !cpu->expose_kvm) { + env->features[FEAT_KVM] = 0; + } + + x86_cpu_enable_xsave_components(cpu); + + /* CPUID[EAX=7,ECX=0].EBX always increased level automatically: */ + x86_cpu_adjust_feat_level(cpu, FEAT_7_0_EBX); + if (cpu->full_cpuid_auto_level) { + x86_cpu_adjust_feat_level(cpu, FEAT_1_EDX); + x86_cpu_adjust_feat_level(cpu, FEAT_1_ECX); + x86_cpu_adjust_feat_level(cpu, FEAT_6_EAX); + x86_cpu_adjust_feat_level(cpu, FEAT_7_0_ECX); + x86_cpu_adjust_feat_level(cpu, FEAT_8000_0001_EDX); + x86_cpu_adjust_feat_level(cpu, FEAT_8000_0001_ECX); + x86_cpu_adjust_feat_level(cpu, FEAT_8000_0007_EDX); + x86_cpu_adjust_feat_level(cpu, FEAT_C000_0001_EDX); + x86_cpu_adjust_feat_level(cpu, FEAT_SVM); + x86_cpu_adjust_feat_level(cpu, FEAT_XSAVE); + /* SVM requires CPUID[0x8000000A] */ + if (env->features[FEAT_8000_0001_ECX] & CPUID_EXT3_SVM) { + x86_cpu_adjust_level(cpu, &env->cpuid_min_xlevel, 0x8000000A); + } + } + + /* Set cpuid_*level* based on cpuid_min_*level, if not explicitly set */ + if (env->cpuid_level == UINT32_MAX) { + env->cpuid_level = env->cpuid_min_level; + } + if (env->cpuid_xlevel == UINT32_MAX) { + env->cpuid_xlevel = env->cpuid_min_xlevel; + } + if (env->cpuid_xlevel2 == UINT32_MAX) { + env->cpuid_xlevel2 = env->cpuid_min_xlevel2; } if (x86_cpu_filter_features(cpu) && cpu->enforce_cpuid) { @@ -3215,9 +3337,6 @@ static void x86_cpu_register_feature_bit_props(X86CPU *cpu, char **names; FeatureWordInfo *fi = &feature_word_info[w]; - if (!fi->feat_names) { - return; - } if (!fi->feat_names[bitnr]) { return; } @@ -3358,12 +3477,17 @@ static Property x86_cpu_properties[] = { DEFINE_PROP_UINT32("phys-bits", X86CPU, phys_bits, 0), DEFINE_PROP_BOOL("host-phys-bits", X86CPU, host_phys_bits, false), DEFINE_PROP_BOOL("fill-mtrr-mask", X86CPU, fill_mtrr_mask, true), - DEFINE_PROP_UINT32("level", X86CPU, env.cpuid_level, 0), - DEFINE_PROP_UINT32("xlevel", X86CPU, env.cpuid_xlevel, 0), - DEFINE_PROP_UINT32("xlevel2", X86CPU, env.cpuid_xlevel2, 0), + DEFINE_PROP_UINT32("level", X86CPU, env.cpuid_level, UINT32_MAX), + DEFINE_PROP_UINT32("xlevel", X86CPU, env.cpuid_xlevel, UINT32_MAX), + DEFINE_PROP_UINT32("xlevel2", X86CPU, env.cpuid_xlevel2, UINT32_MAX), + DEFINE_PROP_UINT32("min-level", X86CPU, env.cpuid_min_level, 0), + DEFINE_PROP_UINT32("min-xlevel", X86CPU, env.cpuid_min_xlevel, 0), + DEFINE_PROP_UINT32("min-xlevel2", X86CPU, env.cpuid_min_xlevel2, 0), + DEFINE_PROP_BOOL("full-cpuid-auto-level", X86CPU, full_cpuid_auto_level, true), DEFINE_PROP_STRING("hv-vendor-id", X86CPU, hyperv_vendor_id), DEFINE_PROP_BOOL("cpuid-0xb", X86CPU, enable_cpuid_0xb, true), DEFINE_PROP_BOOL("lmce", X86CPU, enable_lmce, false), + DEFINE_PROP_BOOL("l3-cache", X86CPU, enable_l3_cache, true), DEFINE_PROP_END_OF_LIST() }; diff --git a/target-i386/cpu.h b/target-i386/cpu.h index 65615c0f3b..e64569854f 100644 --- a/target-i386/cpu.h +++ b/target-i386/cpu.h @@ -453,6 +453,8 @@ typedef enum FeatureWord { FEAT_SVM, /* CPUID[8000_000A].EDX */ FEAT_XSAVE, /* CPUID[EAX=0xd,ECX=1].EAX */ FEAT_6_EAX, /* CPUID[6].EAX */ + FEAT_XSAVE_COMP_LO, /* CPUID[EAX=0xd,ECX=0].EAX */ + FEAT_XSAVE_COMP_HI, /* CPUID[EAX=0xd,ECX=0].EDX */ FEATURE_WORDS, } FeatureWord; @@ -606,16 +608,21 @@ typedef uint32_t FeatureWordArray[FEATURE_WORDS]; #define CPUID_7_0_EBX_RTM (1U << 11) #define CPUID_7_0_EBX_MPX (1U << 14) #define CPUID_7_0_EBX_AVX512F (1U << 16) /* AVX-512 Foundation */ +#define CPUID_7_0_EBX_AVX512DQ (1U << 17) /* AVX-512 Doubleword & Quadword Instrs */ #define CPUID_7_0_EBX_RDSEED (1U << 18) #define CPUID_7_0_EBX_ADX (1U << 19) #define CPUID_7_0_EBX_SMAP (1U << 20) +#define CPUID_7_0_EBX_AVX512IFMA (1U << 21) /* AVX-512 Integer Fused Multiply Add */ #define CPUID_7_0_EBX_PCOMMIT (1U << 22) /* Persistent Commit */ #define CPUID_7_0_EBX_CLFLUSHOPT (1U << 23) /* Flush a Cache Line Optimized */ #define CPUID_7_0_EBX_CLWB (1U << 24) /* Cache Line Write Back */ #define CPUID_7_0_EBX_AVX512PF (1U << 26) /* AVX-512 Prefetch */ #define CPUID_7_0_EBX_AVX512ER (1U << 27) /* AVX-512 Exponential and Reciprocal */ #define CPUID_7_0_EBX_AVX512CD (1U << 28) /* AVX-512 Conflict Detection */ +#define CPUID_7_0_EBX_AVX512BW (1U << 30) /* AVX-512 Byte and Word Instructions */ +#define CPUID_7_0_EBX_AVX512VL (1U << 31) /* AVX-512 Vector Length Extensions */ +#define CPUID_7_0_ECX_VBMI (1U << 1) /* AVX-512 Vector Byte Manipulation Instrs */ #define CPUID_7_0_ECX_UMIP (1U << 2) #define CPUID_7_0_ECX_PKU (1U << 3) #define CPUID_7_0_ECX_OSPKE (1U << 4) @@ -691,6 +698,13 @@ typedef uint32_t FeatureWordArray[FEATURE_WORDS]; /* Use a clearer name for this. */ #define CPU_INTERRUPT_INIT CPU_INTERRUPT_RESET +/* Instead of computing the condition codes after each x86 instruction, + * QEMU just stores one operand (called CC_SRC), the result + * (called CC_DST) and the type of operation (called CC_OP). When the + * condition codes are needed, the condition codes can be calculated + * using this information. Condition codes are not generated if they + * are only needed for conditional branches. + */ typedef enum { CC_OP_DYNAMIC, /* must use dynamic code to get cc_op */ CC_OP_EFLAGS, /* all cc are explicitly computed, CC_SRC = flags */ @@ -872,7 +886,8 @@ typedef union X86LegacyXSaveArea { typedef struct X86XSaveHeader { uint64_t xstate_bv; uint64_t xcomp_bv; - uint8_t reserved[48]; + uint64_t reserve0; + uint8_t reserved[40]; } X86XSaveHeader; /* Ext. save area 2: AVX State */ @@ -1030,6 +1045,9 @@ typedef struct CPUX86State { uint64_t tsc; uint64_t tsc_adjust; uint64_t tsc_deadline; + uint64_t tsc_aux; + + uint64_t xcr0; uint64_t mcg_status; uint64_t msr_ia32_misc_enable; @@ -1046,6 +1064,8 @@ typedef struct CPUX86State { uint64_t pat; uint32_t smbase; + uint32_t pkru; + /* End of state preserved by INIT (dummy marker). */ struct {} end_init_save; @@ -1097,11 +1117,15 @@ typedef struct CPUX86State { CPU_COMMON /* Fields from here on are preserved across CPU reset. */ + struct {} end_reset_fields; /* processor features (e.g. for CPUID insn) */ - uint32_t cpuid_level; - uint32_t cpuid_xlevel; - uint32_t cpuid_xlevel2; + /* Minimum level/xlevel/xlevel2, based on CPU model + features */ + uint32_t cpuid_min_level, cpuid_min_xlevel, cpuid_min_xlevel2; + /* Maximum level/xlevel/xlevel2 value for auto-assignment: */ + uint32_t cpuid_max_level, cpuid_max_xlevel, cpuid_max_xlevel2; + /* Actual level/xlevel/xlevel2 value: */ + uint32_t cpuid_level, cpuid_xlevel, cpuid_xlevel2; uint32_t cpuid_vendor1; uint32_t cpuid_vendor2; uint32_t cpuid_vendor3; @@ -1130,20 +1154,15 @@ typedef struct CPUX86State { uint64_t mcg_ctl; uint64_t mcg_ext_ctl; uint64_t mce_banks[MCE_BANKS_DEF*4]; - - uint64_t tsc_aux; + uint64_t xstate_bv; /* vmstate */ uint16_t fpus_vmstate; uint16_t fptag_vmstate; uint16_t fpregs_format_vmstate; - uint64_t xstate_bv; - uint64_t xcr0; uint64_t xss; - uint32_t pkru; - TPRAccess tpr_access_type; } CPUX86State; @@ -1202,9 +1221,18 @@ struct X86CPU { */ bool enable_lmce; + /* Compatibility bits for old machine types. + * If true present virtual l3 cache for VM, the vcpus in the same virtual + * socket share an virtual l3 cache. + */ + bool enable_l3_cache; + /* Compatibility bits for old machine types: */ bool enable_cpuid_0xb; + /* Enable auto level-increase for all CPUID leaves */ + bool full_cpuid_auto_level; + /* if true fill the top bits of the MTRR_PHYSMASKn variable range */ bool fill_mtrr_mask; @@ -1381,13 +1409,6 @@ int cpu_x86_signal_handler(int host_signum, void *pinfo, void *puc); /* cpu.c */ -typedef struct ExtSaveArea { - uint32_t feature, bits; - uint32_t offset, size; -} ExtSaveArea; - -extern const ExtSaveArea x86_ext_save_areas[]; - void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, uint32_t *eax, uint32_t *ebx, uint32_t *ecx, uint32_t *edx); diff --git a/target-i386/fpu_helper.c b/target-i386/fpu_helper.c index 929489bfc8..2049a8c01d 100644 --- a/target-i386/fpu_helper.c +++ b/target-i386/fpu_helper.c @@ -1110,6 +1110,8 @@ void cpu_x86_frstor(CPUX86State *env, target_ulong ptr, int data32) } #endif +#define XO(X) offsetof(X86XSaveArea, X) + static void do_xsave_fpu(CPUX86State *env, target_ulong ptr, uintptr_t ra) { int fpus, fptag, i; @@ -1120,17 +1122,18 @@ static void do_xsave_fpu(CPUX86State *env, target_ulong ptr, uintptr_t ra) for (i = 0; i < 8; i++) { fptag |= (env->fptags[i] << i); } - cpu_stw_data_ra(env, ptr, env->fpuc, ra); - cpu_stw_data_ra(env, ptr + 2, fpus, ra); - cpu_stw_data_ra(env, ptr + 4, fptag ^ 0xff, ra); + + cpu_stw_data_ra(env, ptr + XO(legacy.fcw), env->fpuc, ra); + cpu_stw_data_ra(env, ptr + XO(legacy.fsw), fpus, ra); + cpu_stw_data_ra(env, ptr + XO(legacy.ftw), fptag ^ 0xff, ra); /* In 32-bit mode this is eip, sel, dp, sel. In 64-bit mode this is rip, rdp. But in either case we don't write actual data, just zeros. */ - cpu_stq_data_ra(env, ptr + 0x08, 0, ra); /* eip+sel; rip */ - cpu_stq_data_ra(env, ptr + 0x10, 0, ra); /* edp+sel; rdp */ + cpu_stq_data_ra(env, ptr + XO(legacy.fpip), 0, ra); /* eip+sel; rip */ + cpu_stq_data_ra(env, ptr + XO(legacy.fpdp), 0, ra); /* edp+sel; rdp */ - addr = ptr + 0x20; + addr = ptr + XO(legacy.fpregs); for (i = 0; i < 8; i++) { floatx80 tmp = ST(i); helper_fstt(env, tmp, addr, ra); @@ -1140,8 +1143,8 @@ static void do_xsave_fpu(CPUX86State *env, target_ulong ptr, uintptr_t ra) static void do_xsave_mxcsr(CPUX86State *env, target_ulong ptr, uintptr_t ra) { - cpu_stl_data_ra(env, ptr + 0x18, env->mxcsr, ra); /* mxcsr */ - cpu_stl_data_ra(env, ptr + 0x1c, 0x0000ffff, ra); /* mxcsr_mask */ + cpu_stl_data_ra(env, ptr + XO(legacy.mxcsr), env->mxcsr, ra); + cpu_stl_data_ra(env, ptr + XO(legacy.mxcsr_mask), 0x0000ffff, ra); } static void do_xsave_sse(CPUX86State *env, target_ulong ptr, uintptr_t ra) @@ -1155,7 +1158,7 @@ static void do_xsave_sse(CPUX86State *env, target_ulong ptr, uintptr_t ra) nb_xmm_regs = 8; } - addr = ptr + 0xa0; + addr = ptr + XO(legacy.xmm_regs); for (i = 0; i < nb_xmm_regs; i++) { cpu_stq_data_ra(env, addr, env->xmm_regs[i].ZMM_Q(0), ra); cpu_stq_data_ra(env, addr + 8, env->xmm_regs[i].ZMM_Q(1), ra); @@ -1163,8 +1166,9 @@ static void do_xsave_sse(CPUX86State *env, target_ulong ptr, uintptr_t ra) } } -static void do_xsave_bndregs(CPUX86State *env, target_ulong addr, uintptr_t ra) +static void do_xsave_bndregs(CPUX86State *env, target_ulong ptr, uintptr_t ra) { + target_ulong addr = ptr + offsetof(XSaveBNDREG, bnd_regs); int i; for (i = 0; i < 4; i++, addr += 16) { @@ -1173,15 +1177,17 @@ static void do_xsave_bndregs(CPUX86State *env, target_ulong addr, uintptr_t ra) } } -static void do_xsave_bndcsr(CPUX86State *env, target_ulong addr, uintptr_t ra) +static void do_xsave_bndcsr(CPUX86State *env, target_ulong ptr, uintptr_t ra) { - cpu_stq_data_ra(env, addr, env->bndcs_regs.cfgu, ra); - cpu_stq_data_ra(env, addr + 8, env->bndcs_regs.sts, ra); + cpu_stq_data_ra(env, ptr + offsetof(XSaveBNDCSR, bndcsr.cfgu), + env->bndcs_regs.cfgu, ra); + cpu_stq_data_ra(env, ptr + offsetof(XSaveBNDCSR, bndcsr.sts), + env->bndcs_regs.sts, ra); } -static void do_xsave_pkru(CPUX86State *env, target_ulong addr, uintptr_t ra) +static void do_xsave_pkru(CPUX86State *env, target_ulong ptr, uintptr_t ra) { - cpu_stq_data_ra(env, addr, env->pkru, ra); + cpu_stq_data_ra(env, ptr, env->pkru, ra); } void helper_fxsave(CPUX86State *env, target_ulong ptr) @@ -1250,22 +1256,19 @@ static void do_xsave(CPUX86State *env, target_ulong ptr, uint64_t rfbm, do_xsave_sse(env, ptr, ra); } if (opt & XSTATE_BNDREGS_MASK) { - target_ulong off = x86_ext_save_areas[XSTATE_BNDREGS_BIT].offset; - do_xsave_bndregs(env, ptr + off, ra); + do_xsave_bndregs(env, ptr + XO(bndreg_state), ra); } if (opt & XSTATE_BNDCSR_MASK) { - target_ulong off = x86_ext_save_areas[XSTATE_BNDCSR_BIT].offset; - do_xsave_bndcsr(env, ptr + off, ra); + do_xsave_bndcsr(env, ptr + XO(bndcsr_state), ra); } if (opt & XSTATE_PKRU_MASK) { - target_ulong off = x86_ext_save_areas[XSTATE_PKRU_BIT].offset; - do_xsave_pkru(env, ptr + off, ra); + do_xsave_pkru(env, ptr + XO(pkru_state), ra); } /* Update the XSTATE_BV field. */ - old_bv = cpu_ldq_data_ra(env, ptr + 512, ra); + old_bv = cpu_ldq_data_ra(env, ptr + XO(header.xstate_bv), ra); new_bv = (old_bv & ~rfbm) | (inuse & rfbm); - cpu_stq_data_ra(env, ptr + 512, new_bv, ra); + cpu_stq_data_ra(env, ptr + XO(header.xstate_bv), new_bv, ra); } void helper_xsave(CPUX86State *env, target_ulong ptr, uint64_t rfbm) @@ -1281,12 +1284,13 @@ void helper_xsaveopt(CPUX86State *env, target_ulong ptr, uint64_t rfbm) static void do_xrstor_fpu(CPUX86State *env, target_ulong ptr, uintptr_t ra) { - int i, fpus, fptag; + int i, fpuc, fpus, fptag; target_ulong addr; - cpu_set_fpuc(env, cpu_lduw_data_ra(env, ptr, ra)); - fpus = cpu_lduw_data_ra(env, ptr + 2, ra); - fptag = cpu_lduw_data_ra(env, ptr + 4, ra); + fpuc = cpu_lduw_data_ra(env, ptr + XO(legacy.fcw), ra); + fpus = cpu_lduw_data_ra(env, ptr + XO(legacy.fsw), ra); + fptag = cpu_lduw_data_ra(env, ptr + XO(legacy.ftw), ra); + cpu_set_fpuc(env, fpuc); env->fpstt = (fpus >> 11) & 7; env->fpus = fpus & ~0x3800; fptag ^= 0xff; @@ -1294,7 +1298,7 @@ static void do_xrstor_fpu(CPUX86State *env, target_ulong ptr, uintptr_t ra) env->fptags[i] = ((fptag >> i) & 1); } - addr = ptr + 0x20; + addr = ptr + XO(legacy.fpregs); for (i = 0; i < 8; i++) { floatx80 tmp = helper_fldt(env, addr, ra); ST(i) = tmp; @@ -1304,7 +1308,7 @@ static void do_xrstor_fpu(CPUX86State *env, target_ulong ptr, uintptr_t ra) static void do_xrstor_mxcsr(CPUX86State *env, target_ulong ptr, uintptr_t ra) { - cpu_set_mxcsr(env, cpu_ldl_data_ra(env, ptr + 0x18, ra)); + cpu_set_mxcsr(env, cpu_ldl_data_ra(env, ptr + XO(legacy.mxcsr), ra)); } static void do_xrstor_sse(CPUX86State *env, target_ulong ptr, uintptr_t ra) @@ -1318,7 +1322,7 @@ static void do_xrstor_sse(CPUX86State *env, target_ulong ptr, uintptr_t ra) nb_xmm_regs = 8; } - addr = ptr + 0xa0; + addr = ptr + XO(legacy.xmm_regs); for (i = 0; i < nb_xmm_regs; i++) { env->xmm_regs[i].ZMM_Q(0) = cpu_ldq_data_ra(env, addr, ra); env->xmm_regs[i].ZMM_Q(1) = cpu_ldq_data_ra(env, addr + 8, ra); @@ -1326,8 +1330,9 @@ static void do_xrstor_sse(CPUX86State *env, target_ulong ptr, uintptr_t ra) } } -static void do_xrstor_bndregs(CPUX86State *env, target_ulong addr, uintptr_t ra) +static void do_xrstor_bndregs(CPUX86State *env, target_ulong ptr, uintptr_t ra) { + target_ulong addr = ptr + offsetof(XSaveBNDREG, bnd_regs); int i; for (i = 0; i < 4; i++, addr += 16) { @@ -1336,16 +1341,18 @@ static void do_xrstor_bndregs(CPUX86State *env, target_ulong addr, uintptr_t ra) } } -static void do_xrstor_bndcsr(CPUX86State *env, target_ulong addr, uintptr_t ra) +static void do_xrstor_bndcsr(CPUX86State *env, target_ulong ptr, uintptr_t ra) { /* FIXME: Extend highest implemented bit of linear address. */ - env->bndcs_regs.cfgu = cpu_ldq_data_ra(env, addr, ra); - env->bndcs_regs.sts = cpu_ldq_data_ra(env, addr + 8, ra); + env->bndcs_regs.cfgu + = cpu_ldq_data_ra(env, ptr + offsetof(XSaveBNDCSR, bndcsr.cfgu), ra); + env->bndcs_regs.sts + = cpu_ldq_data_ra(env, ptr + offsetof(XSaveBNDCSR, bndcsr.sts), ra); } -static void do_xrstor_pkru(CPUX86State *env, target_ulong addr, uintptr_t ra) +static void do_xrstor_pkru(CPUX86State *env, target_ulong ptr, uintptr_t ra) { - env->pkru = cpu_ldq_data_ra(env, addr, ra); + env->pkru = cpu_ldq_data_ra(env, ptr, ra); } void helper_fxrstor(CPUX86State *env, target_ulong ptr) @@ -1373,7 +1380,7 @@ void helper_fxrstor(CPUX86State *env, target_ulong ptr) void helper_xrstor(CPUX86State *env, target_ulong ptr, uint64_t rfbm) { uintptr_t ra = GETPC(); - uint64_t xstate_bv, xcomp_bv0, xcomp_bv1; + uint64_t xstate_bv, xcomp_bv, reserve0; rfbm &= env->xcr0; @@ -1387,7 +1394,7 @@ void helper_xrstor(CPUX86State *env, target_ulong ptr, uint64_t rfbm) raise_exception_ra(env, EXCP0D_GPF, ra); } - xstate_bv = cpu_ldq_data_ra(env, ptr + 512, ra); + xstate_bv = cpu_ldq_data_ra(env, ptr + XO(header.xstate_bv), ra); if ((int64_t)xstate_bv < 0) { /* FIXME: Compact form. */ @@ -1396,15 +1403,19 @@ void helper_xrstor(CPUX86State *env, target_ulong ptr, uint64_t rfbm) /* Standard form. */ - /* The XSTATE field must not set bits not present in XCR0. */ + /* The XSTATE_BV field must not set bits not present in XCR0. */ if (xstate_bv & ~env->xcr0) { raise_exception_ra(env, EXCP0D_GPF, ra); } - /* The XCOMP field must be zero. */ - xcomp_bv0 = cpu_ldq_data_ra(env, ptr + 520, ra); - xcomp_bv1 = cpu_ldq_data_ra(env, ptr + 528, ra); - if (xcomp_bv0 || xcomp_bv1) { + /* The XCOMP_BV field must be zero. Note that, as of the April 2016 + revision, the description of the XSAVE Header (Vol 1, Sec 13.4.2) + describes only XCOMP_BV, but the description of the standard form + of XRSTOR (Vol 1, Sec 13.8.1) checks bytes 23:8 for zero, which + includes the next 64-bit field. */ + xcomp_bv = cpu_ldq_data_ra(env, ptr + XO(header.xcomp_bv), ra); + reserve0 = cpu_ldq_data_ra(env, ptr + XO(header.reserve0), ra); + if (xcomp_bv || reserve0) { raise_exception_ra(env, EXCP0D_GPF, ra); } @@ -1430,8 +1441,7 @@ void helper_xrstor(CPUX86State *env, target_ulong ptr, uint64_t rfbm) } if (rfbm & XSTATE_BNDREGS_MASK) { if (xstate_bv & XSTATE_BNDREGS_MASK) { - target_ulong off = x86_ext_save_areas[XSTATE_BNDREGS_BIT].offset; - do_xrstor_bndregs(env, ptr + off, ra); + do_xrstor_bndregs(env, ptr + XO(bndreg_state), ra); env->hflags |= HF_MPX_IU_MASK; } else { memset(env->bnd_regs, 0, sizeof(env->bnd_regs)); @@ -1440,8 +1450,7 @@ void helper_xrstor(CPUX86State *env, target_ulong ptr, uint64_t rfbm) } if (rfbm & XSTATE_BNDCSR_MASK) { if (xstate_bv & XSTATE_BNDCSR_MASK) { - target_ulong off = x86_ext_save_areas[XSTATE_BNDCSR_BIT].offset; - do_xrstor_bndcsr(env, ptr + off, ra); + do_xrstor_bndcsr(env, ptr + XO(bndcsr_state), ra); } else { memset(&env->bndcs_regs, 0, sizeof(env->bndcs_regs)); } @@ -1450,8 +1459,7 @@ void helper_xrstor(CPUX86State *env, target_ulong ptr, uint64_t rfbm) if (rfbm & XSTATE_PKRU_MASK) { uint64_t old_pkru = env->pkru; if (xstate_bv & XSTATE_PKRU_MASK) { - target_ulong off = x86_ext_save_areas[XSTATE_PKRU_BIT].offset; - do_xrstor_pkru(env, ptr + off, ra); + do_xrstor_pkru(env, ptr + XO(pkru_state), ra); } else { env->pkru = 0; } @@ -1462,6 +1470,8 @@ void helper_xrstor(CPUX86State *env, target_ulong ptr, uint64_t rfbm) } } +#undef XO + uint64_t helper_xgetbv(CPUX86State *env, uint32_t ecx) { /* The OS must have enabled XSAVE. */ diff --git a/target-i386/helper.c b/target-i386/helper.c index 1c250b8245..9bc961bff3 100644 --- a/target-i386/helper.c +++ b/target-i386/helper.c @@ -1113,7 +1113,6 @@ out: typedef struct MCEInjectionParams { Monitor *mon; - X86CPU *cpu; int bank; uint64_t status; uint64_t mcg_status; @@ -1122,14 +1121,14 @@ typedef struct MCEInjectionParams { int flags; } MCEInjectionParams; -static void do_inject_x86_mce(void *data) +static void do_inject_x86_mce(CPUState *cs, void *data) { MCEInjectionParams *params = data; - CPUX86State *cenv = ¶ms->cpu->env; - CPUState *cpu = CPU(params->cpu); + X86CPU *cpu = X86_CPU(cs); + CPUX86State *cenv = &cpu->env; uint64_t *banks = cenv->mce_banks + 4 * params->bank; - cpu_synchronize_state(cpu); + cpu_synchronize_state(cs); /* * If there is an MCE exception being processed, ignore this SRAO MCE @@ -1149,7 +1148,7 @@ static void do_inject_x86_mce(void *data) if ((cenv->mcg_cap & MCG_CTL_P) && cenv->mcg_ctl != ~(uint64_t)0) { monitor_printf(params->mon, "CPU %d: Uncorrected error reporting disabled\n", - cpu->cpu_index); + cs->cpu_index); return; } @@ -1161,7 +1160,7 @@ static void do_inject_x86_mce(void *data) monitor_printf(params->mon, "CPU %d: Uncorrected error reporting disabled for" " bank %d\n", - cpu->cpu_index, params->bank); + cs->cpu_index, params->bank); return; } @@ -1170,7 +1169,7 @@ static void do_inject_x86_mce(void *data) monitor_printf(params->mon, "CPU %d: Previous MCE still in progress, raising" " triple fault\n", - cpu->cpu_index); + cs->cpu_index); qemu_log_mask(CPU_LOG_RESET, "Triple fault\n"); qemu_system_reset_request(); return; @@ -1182,7 +1181,7 @@ static void do_inject_x86_mce(void *data) banks[3] = params->misc; cenv->mcg_status = params->mcg_status; banks[1] = params->status; - cpu_interrupt(cpu, CPU_INTERRUPT_MCE); + cpu_interrupt(cs, CPU_INTERRUPT_MCE); } else if (!(banks[1] & MCI_STATUS_VAL) || !(banks[1] & MCI_STATUS_UC)) { if (banks[1] & MCI_STATUS_VAL) { @@ -1204,7 +1203,6 @@ void cpu_x86_inject_mce(Monitor *mon, X86CPU *cpu, int bank, CPUX86State *cenv = &cpu->env; MCEInjectionParams params = { .mon = mon, - .cpu = cpu, .bank = bank, .status = status, .mcg_status = mcg_status, @@ -1245,7 +1243,6 @@ void cpu_x86_inject_mce(Monitor *mon, X86CPU *cpu, int bank, if (other_cs == cs) { continue; } - params.cpu = X86_CPU(other_cs); run_on_cpu(other_cs, do_inject_x86_mce, ¶ms); } } diff --git a/target-i386/kvm.c b/target-i386/kvm.c index d1a25c5465..ee1f53e569 100644 --- a/target-i386/kvm.c +++ b/target-i386/kvm.c @@ -83,23 +83,17 @@ static bool has_msr_tsc_aux; static bool has_msr_tsc_adjust; static bool has_msr_tsc_deadline; static bool has_msr_feature_control; -static bool has_msr_async_pf_en; -static bool has_msr_pv_eoi_en; static bool has_msr_misc_enable; static bool has_msr_smbase; static bool has_msr_bndcfgs; -static bool has_msr_kvm_steal_time; static int lm_capable_kernel; static bool has_msr_hv_hypercall; -static bool has_msr_hv_vapic; -static bool has_msr_hv_tsc; static bool has_msr_hv_crash; static bool has_msr_hv_reset; static bool has_msr_hv_vpindex; static bool has_msr_hv_runtime; static bool has_msr_hv_synic; static bool has_msr_hv_stimer; -static bool has_msr_mtrr; static bool has_msr_xss; static bool has_msr_architectural_pmu; @@ -156,10 +150,8 @@ static int kvm_get_tsc(CPUState *cs) return 0; } -static inline void do_kvm_synchronize_tsc(void *arg) +static inline void do_kvm_synchronize_tsc(CPUState *cpu, void *arg) { - CPUState *cpu = arg; - kvm_get_tsc(cpu); } @@ -169,7 +161,7 @@ void kvm_synchronize_all_tsc(void) if (kvm_enabled()) { CPU_FOREACH(cpu) { - run_on_cpu(cpu, do_kvm_synchronize_tsc, cpu); + run_on_cpu(cpu, do_kvm_synchronize_tsc, NULL); } } } @@ -604,20 +596,22 @@ static int hyperv_handle_properties(CPUState *cs) X86CPU *cpu = X86_CPU(cs); CPUX86State *env = &cpu->env; + if (cpu->hyperv_time && + kvm_check_extension(cs->kvm_state, KVM_CAP_HYPERV_TIME) <= 0) { + cpu->hyperv_time = false; + } + if (cpu->hyperv_relaxed_timing) { env->features[FEAT_HYPERV_EAX] |= HV_X64_MSR_HYPERCALL_AVAILABLE; } if (cpu->hyperv_vapic) { env->features[FEAT_HYPERV_EAX] |= HV_X64_MSR_HYPERCALL_AVAILABLE; env->features[FEAT_HYPERV_EAX] |= HV_X64_MSR_APIC_ACCESS_AVAILABLE; - has_msr_hv_vapic = true; } - if (cpu->hyperv_time && - kvm_check_extension(cs->kvm_state, KVM_CAP_HYPERV_TIME) > 0) { + if (cpu->hyperv_time) { env->features[FEAT_HYPERV_EAX] |= HV_X64_MSR_HYPERCALL_AVAILABLE; env->features[FEAT_HYPERV_EAX] |= HV_X64_MSR_TIME_REF_COUNT_AVAILABLE; env->features[FEAT_HYPERV_EAX] |= 0x200; - has_msr_hv_tsc = true; } if (cpu->hyperv_crash && has_msr_hv_crash) { env->features[FEAT_HYPERV_EDX] |= HV_X64_GUEST_CRASH_MSR_AVAILABLE; @@ -729,7 +723,7 @@ int kvm_arch_init_vcpu(CPUState *cs) if (cpu->hyperv_relaxed_timing) { c->eax |= HV_X64_RELAXED_TIMING_RECOMMENDED; } - if (has_msr_hv_vapic) { + if (cpu->hyperv_vapic) { c->eax |= HV_X64_APIC_ACCESS_RECOMMENDED; } c->ebx = cpu->hyperv_spinlock_attempts; @@ -755,12 +749,6 @@ int kvm_arch_init_vcpu(CPUState *cs) c = &cpuid_data.entries[cpuid_i++]; c->function = KVM_CPUID_FEATURES | kvm_base; c->eax = env->features[FEAT_KVM]; - - has_msr_async_pf_en = c->eax & (1 << KVM_FEATURE_ASYNC_PF); - - has_msr_pv_eoi_en = c->eax & (1 << KVM_FEATURE_PV_EOI); - - has_msr_kvm_steal_time = c->eax & (1 << KVM_FEATURE_STEAL_TIME); } cpu_x86_cpuid(env, 0, 0, &limit, &unused, &unused, &unused); @@ -975,9 +963,6 @@ int kvm_arch_init_vcpu(CPUState *cs) } cpu->kvm_msr_buf = g_malloc0(MSR_BUF_SIZE); - if (env->features[FEAT_1_EDX] & CPUID_MTRR) { - has_msr_mtrr = true; - } if (!(env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_RDTSCP)) { has_msr_tsc_aux = false; } @@ -1532,6 +1517,22 @@ static void kvm_msr_entry_add(X86CPU *cpu, uint32_t index, uint64_t value) msrs->nmsrs++; } +static int kvm_put_one_msr(X86CPU *cpu, int index, uint64_t value) +{ + kvm_msr_buf_reset(cpu); + kvm_msr_entry_add(cpu, index, value); + + return kvm_vcpu_ioctl(CPU(cpu), KVM_SET_MSRS, cpu->kvm_msr_buf); +} + +void kvm_put_apicbase(X86CPU *cpu, uint64_t value) +{ + int ret; + + ret = kvm_put_one_msr(cpu, MSR_IA32_APICBASE, value); + assert(ret == 1); +} + static int kvm_put_tscdeadline_msr(X86CPU *cpu) { CPUX86State *env = &cpu->env; @@ -1541,10 +1542,7 @@ static int kvm_put_tscdeadline_msr(X86CPU *cpu) return 0; } - kvm_msr_buf_reset(cpu); - kvm_msr_entry_add(cpu, MSR_IA32_TSCDEADLINE, env->tsc_deadline); - - ret = kvm_vcpu_ioctl(CPU(cpu), KVM_SET_MSRS, cpu->kvm_msr_buf); + ret = kvm_put_one_msr(cpu, MSR_IA32_TSCDEADLINE, env->tsc_deadline); if (ret < 0) { return ret; } @@ -1567,11 +1565,8 @@ static int kvm_put_msr_feature_control(X86CPU *cpu) return 0; } - kvm_msr_buf_reset(cpu); - kvm_msr_entry_add(cpu, MSR_IA32_FEATURE_CONTROL, - cpu->env.msr_ia32_feature_control); - - ret = kvm_vcpu_ioctl(CPU(cpu), KVM_SET_MSRS, cpu->kvm_msr_buf); + ret = kvm_put_one_msr(cpu, MSR_IA32_FEATURE_CONTROL, + cpu->env.msr_ia32_feature_control); if (ret < 0) { return ret; } @@ -1633,13 +1628,13 @@ static int kvm_put_msrs(X86CPU *cpu, int level) kvm_msr_entry_add(cpu, MSR_IA32_TSC, env->tsc); kvm_msr_entry_add(cpu, MSR_KVM_SYSTEM_TIME, env->system_time_msr); kvm_msr_entry_add(cpu, MSR_KVM_WALL_CLOCK, env->wall_clock_msr); - if (has_msr_async_pf_en) { + if (env->features[FEAT_KVM] & (1 << KVM_FEATURE_ASYNC_PF)) { kvm_msr_entry_add(cpu, MSR_KVM_ASYNC_PF_EN, env->async_pf_en_msr); } - if (has_msr_pv_eoi_en) { + if (env->features[FEAT_KVM] & (1 << KVM_FEATURE_PV_EOI)) { kvm_msr_entry_add(cpu, MSR_KVM_PV_EOI_EN, env->pv_eoi_en_msr); } - if (has_msr_kvm_steal_time) { + if (env->features[FEAT_KVM] & (1 << KVM_FEATURE_STEAL_TIME)) { kvm_msr_entry_add(cpu, MSR_KVM_STEAL_TIME, env->steal_time_msr); } if (has_msr_architectural_pmu) { @@ -1675,11 +1670,11 @@ static int kvm_put_msrs(X86CPU *cpu, int level) kvm_msr_entry_add(cpu, HV_X64_MSR_HYPERCALL, env->msr_hv_hypercall); } - if (has_msr_hv_vapic) { + if (cpu->hyperv_vapic) { kvm_msr_entry_add(cpu, HV_X64_MSR_APIC_ASSIST_PAGE, env->msr_hv_vapic); } - if (has_msr_hv_tsc) { + if (cpu->hyperv_time) { kvm_msr_entry_add(cpu, HV_X64_MSR_REFERENCE_TSC, env->msr_hv_tsc); } if (has_msr_hv_crash) { @@ -1725,7 +1720,7 @@ static int kvm_put_msrs(X86CPU *cpu, int level) env->msr_hv_stimer_count[j]); } } - if (has_msr_mtrr) { + if (env->features[FEAT_1_EDX] & CPUID_MTRR) { uint64_t phys_mask = MAKE_64BIT_MASK(0, cpu->phys_bits); kvm_msr_entry_add(cpu, MSR_MTRRdefType, env->mtrr_deftype); @@ -2042,13 +2037,13 @@ static int kvm_get_msrs(X86CPU *cpu) #endif kvm_msr_entry_add(cpu, MSR_KVM_SYSTEM_TIME, 0); kvm_msr_entry_add(cpu, MSR_KVM_WALL_CLOCK, 0); - if (has_msr_async_pf_en) { + if (env->features[FEAT_KVM] & (1 << KVM_FEATURE_ASYNC_PF)) { kvm_msr_entry_add(cpu, MSR_KVM_ASYNC_PF_EN, 0); } - if (has_msr_pv_eoi_en) { + if (env->features[FEAT_KVM] & (1 << KVM_FEATURE_PV_EOI)) { kvm_msr_entry_add(cpu, MSR_KVM_PV_EOI_EN, 0); } - if (has_msr_kvm_steal_time) { + if (env->features[FEAT_KVM] & (1 << KVM_FEATURE_STEAL_TIME)) { kvm_msr_entry_add(cpu, MSR_KVM_STEAL_TIME, 0); } if (has_msr_architectural_pmu) { @@ -2080,10 +2075,10 @@ static int kvm_get_msrs(X86CPU *cpu) kvm_msr_entry_add(cpu, HV_X64_MSR_HYPERCALL, 0); kvm_msr_entry_add(cpu, HV_X64_MSR_GUEST_OS_ID, 0); } - if (has_msr_hv_vapic) { + if (cpu->hyperv_vapic) { kvm_msr_entry_add(cpu, HV_X64_MSR_APIC_ASSIST_PAGE, 0); } - if (has_msr_hv_tsc) { + if (cpu->hyperv_time) { kvm_msr_entry_add(cpu, HV_X64_MSR_REFERENCE_TSC, 0); } if (has_msr_hv_crash) { @@ -2115,7 +2110,7 @@ static int kvm_get_msrs(X86CPU *cpu) kvm_msr_entry_add(cpu, msr, 0); } } - if (has_msr_mtrr) { + if (env->features[FEAT_1_EDX] & CPUID_MTRR) { kvm_msr_entry_add(cpu, MSR_MTRRdefType, 0); kvm_msr_entry_add(cpu, MSR_MTRRfix64K_00000, 0); kvm_msr_entry_add(cpu, MSR_MTRRfix16K_80000, 0); @@ -2416,19 +2411,6 @@ static int kvm_get_apic(X86CPU *cpu) return 0; } -static int kvm_put_apic(X86CPU *cpu) -{ - DeviceState *apic = cpu->apic_state; - struct kvm_lapic_state kapic; - - if (apic && kvm_irqchip_in_kernel()) { - kvm_put_apic_state(apic, &kapic); - - return kvm_vcpu_ioctl(CPU(cpu), KVM_SET_LAPIC, &kapic); - } - return 0; -} - static int kvm_put_vcpu_events(X86CPU *cpu, int level) { CPUState *cs = CPU(cpu); @@ -2455,6 +2437,7 @@ static int kvm_put_vcpu_events(X86CPU *cpu, int level) events.nmi.pad = 0; events.sipi_vector = env->sipi_vector; + events.flags = 0; if (has_msr_smbase) { events.smi.smm = !!(env->hflags & HF_SMM_MASK); @@ -2474,7 +2457,6 @@ static int kvm_put_vcpu_events(X86CPU *cpu, int level) events.flags |= KVM_VCPUEVENT_VALID_SMM; } - events.flags = 0; if (level >= KVM_PUT_RESET_STATE) { events.flags |= KVM_VCPUEVENT_VALID_NMI_PENDING | KVM_VCPUEVENT_VALID_SIPI_VECTOR; @@ -2670,10 +2652,6 @@ int kvm_arch_put_registers(CPUState *cpu, int level) if (ret < 0) { return ret; } - ret = kvm_put_apic(x86_cpu); - if (ret < 0) { - return ret; - } } ret = kvm_put_tscdeadline_msr(x86_cpu); diff --git a/target-i386/kvm_i386.h b/target-i386/kvm_i386.h index 42b00af1b1..36407e0a5d 100644 --- a/target-i386/kvm_i386.h +++ b/target-i386/kvm_i386.h @@ -41,4 +41,6 @@ int kvm_device_msix_set_vector(KVMState *s, uint32_t dev_id, uint32_t vector, int kvm_device_msix_assign(KVMState *s, uint32_t dev_id); int kvm_device_msix_deassign(KVMState *s, uint32_t dev_id); +void kvm_put_apicbase(X86CPU *cpu, uint64_t value); + #endif diff --git a/target-i386/monitor.c b/target-i386/monitor.c index fccfe40ab7..9a3b4d746e 100644 --- a/target-i386/monitor.c +++ b/target-i386/monitor.c @@ -504,7 +504,8 @@ void hmp_info_local_apic(Monitor *mon, const QDict *qdict) void hmp_info_io_apic(Monitor *mon, const QDict *qdict) { - if (kvm_irqchip_in_kernel()) { + if (kvm_irqchip_in_kernel() && + !kvm_irqchip_is_split()) { kvm_ioapic_dump_state(mon, qdict); } else { ioapic_dump_state(mon, qdict); diff --git a/target-i386/seg_helper.c b/target-i386/seg_helper.c index 6cbdf17426..fb79f3180d 100644 --- a/target-i386/seg_helper.c +++ b/target-i386/seg_helper.c @@ -1137,25 +1137,27 @@ static void do_interrupt_real(CPUX86State *env, int intno, int is_int, static void do_interrupt_user(CPUX86State *env, int intno, int is_int, int error_code, target_ulong next_eip) { - SegmentCache *dt; - target_ulong ptr; - int dpl, cpl, shift; - uint32_t e2; + if (is_int) { + SegmentCache *dt; + target_ulong ptr; + int dpl, cpl, shift; + uint32_t e2; - dt = &env->idt; - if (env->hflags & HF_LMA_MASK) { - shift = 4; - } else { - shift = 3; - } - ptr = dt->base + (intno << shift); - e2 = cpu_ldl_kernel(env, ptr + 4); + dt = &env->idt; + if (env->hflags & HF_LMA_MASK) { + shift = 4; + } else { + shift = 3; + } + ptr = dt->base + (intno << shift); + e2 = cpu_ldl_kernel(env, ptr + 4); - dpl = (e2 >> DESC_DPL_SHIFT) & 3; - cpl = env->hflags & HF_CPL_MASK; - /* check privilege if software int */ - if (is_int && dpl < cpl) { - raise_exception_err(env, EXCP0D_GPF, (intno << shift) + 2); + dpl = (e2 >> DESC_DPL_SHIFT) & 3; + cpl = env->hflags & HF_CPL_MASK; + /* check privilege if software int */ + if (dpl < cpl) { + raise_exception_err(env, EXCP0D_GPF, (intno << shift) + 2); + } } /* Since we emulate only user space, we cannot do more than diff --git a/target-i386/translate.c b/target-i386/translate.c index fa2ac48173..9447557911 100644 --- a/target-i386/translate.c +++ b/target-i386/translate.c @@ -8012,13 +8012,21 @@ static target_ulong disas_insn(CPUX86State *env, DisasContext *s, || (prefixes & PREFIX_LOCK)) { goto illegal_op; } + tcg_gen_mb(TCG_MO_ST_ST | TCG_BAR_SC); break; case 0xe8 ... 0xef: /* lfence */ + if (!(s->cpuid_features & CPUID_SSE) + || (prefixes & PREFIX_LOCK)) { + goto illegal_op; + } + tcg_gen_mb(TCG_MO_LD_LD | TCG_BAR_SC); + break; case 0xf0 ... 0xf7: /* mfence */ if (!(s->cpuid_features & CPUID_SSE2) || (prefixes & PREFIX_LOCK)) { goto illegal_op; } + tcg_gen_mb(TCG_MO_ALL | TCG_BAR_SC); break; default: |