--- /dev/null
+#ifndef _ASM_X86_MSR_INDEX_H
+#define _ASM_X86_MSR_INDEX_H
+
+/* CPU model specific register (MSR) numbers */
+
+/* x86-64 specific MSRs */
+#define MSR_EFER 0xc0000080 /* extended feature register */
+#define MSR_STAR 0xc0000081 /* legacy mode SYSCALL target */
+#define MSR_LSTAR 0xc0000082 /* long mode SYSCALL target */
+#define MSR_CSTAR 0xc0000083 /* compat mode SYSCALL target */
+#define MSR_SYSCALL_MASK 0xc0000084 /* EFLAGS mask for syscall */
+#define MSR_FS_BASE 0xc0000100 /* 64bit FS base */
+#define MSR_GS_BASE 0xc0000101 /* 64bit GS base */
+#define MSR_KERNEL_GS_BASE 0xc0000102 /* SwapGS GS shadow */
+#define MSR_TSC_AUX 0xc0000103 /* Auxiliary TSC */
+
+/* EFER bits: */
+#define _EFER_SCE 0 /* SYSCALL/SYSRET */
+#define _EFER_LME 8 /* Long mode enable */
+#define _EFER_LMA 10 /* Long mode active (read-only) */
+#define _EFER_NX 11 /* No execute enable */
+#define _EFER_SVME 12 /* Enable virtualization */
+#define _EFER_LMSLE 13 /* Long Mode Segment Limit Enable */
+#define _EFER_FFXSR 14 /* Enable Fast FXSAVE/FXRSTOR */
+
+#define EFER_SCE (1<<_EFER_SCE)
+#define EFER_LME (1<<_EFER_LME)
+#define EFER_LMA (1<<_EFER_LMA)
+#define EFER_NX (1<<_EFER_NX)
+#define EFER_SVME (1<<_EFER_SVME)
+#define EFER_LMSLE (1<<_EFER_LMSLE)
+#define EFER_FFXSR (1<<_EFER_FFXSR)
+
+/* Intel MSRs. Some also available on other CPUs */
+#define MSR_IA32_PERFCTR0 0x000000c1
+#define MSR_IA32_PERFCTR1 0x000000c2
+#define MSR_FSB_FREQ 0x000000cd
+#define MSR_NHM_PLATFORM_INFO 0x000000ce
+
+#define MSR_NHM_SNB_PKG_CST_CFG_CTL 0x000000e2
+#define NHM_C3_AUTO_DEMOTE (1UL << 25)
+#define NHM_C1_AUTO_DEMOTE (1UL << 26)
+#define ATM_LNC_C6_AUTO_DEMOTE (1UL << 25)
+#define SNB_C1_AUTO_UNDEMOTE (1UL << 27)
+#define SNB_C3_AUTO_UNDEMOTE (1UL << 28)
+
+#define MSR_PLATFORM_INFO 0x000000ce
+#define MSR_MTRRcap 0x000000fe
+#define MSR_IA32_BBL_CR_CTL 0x00000119
+#define MSR_IA32_BBL_CR_CTL3 0x0000011e
+
+#define MSR_IA32_SYSENTER_CS 0x00000174
+#define MSR_IA32_SYSENTER_ESP 0x00000175
+#define MSR_IA32_SYSENTER_EIP 0x00000176
+
+#define MSR_IA32_MCG_CAP 0x00000179
+#define MSR_IA32_MCG_STATUS 0x0000017a
+#define MSR_IA32_MCG_CTL 0x0000017b
+
+#define MSR_OFFCORE_RSP_0 0x000001a6
+#define MSR_OFFCORE_RSP_1 0x000001a7
+#define MSR_NHM_TURBO_RATIO_LIMIT 0x000001ad
+#define MSR_IVT_TURBO_RATIO_LIMIT 0x000001ae
+
+#define MSR_LBR_SELECT 0x000001c8
+#define MSR_LBR_TOS 0x000001c9
+#define MSR_LBR_NHM_FROM 0x00000680
+#define MSR_LBR_NHM_TO 0x000006c0
+#define MSR_LBR_CORE_FROM 0x00000040
+#define MSR_LBR_CORE_TO 0x00000060
+
+#define MSR_IA32_PEBS_ENABLE 0x000003f1
+#define MSR_IA32_DS_AREA 0x00000600
+#define MSR_IA32_PERF_CAPABILITIES 0x00000345
+#define MSR_PEBS_LD_LAT_THRESHOLD 0x000003f6
+
+#define MSR_MTRRfix64K_00000 0x00000250
+#define MSR_MTRRfix16K_80000 0x00000258
+#define MSR_MTRRfix16K_A0000 0x00000259
+#define MSR_MTRRfix4K_C0000 0x00000268
+#define MSR_MTRRfix4K_C8000 0x00000269
+#define MSR_MTRRfix4K_D0000 0x0000026a
+#define MSR_MTRRfix4K_D8000 0x0000026b
+#define MSR_MTRRfix4K_E0000 0x0000026c
+#define MSR_MTRRfix4K_E8000 0x0000026d
+#define MSR_MTRRfix4K_F0000 0x0000026e
+#define MSR_MTRRfix4K_F8000 0x0000026f
+#define MSR_MTRRdefType 0x000002ff
+
+#define MSR_IA32_CR_PAT 0x00000277
+
+#define MSR_IA32_DEBUGCTLMSR 0x000001d9
+#define MSR_IA32_LASTBRANCHFROMIP 0x000001db
+#define MSR_IA32_LASTBRANCHTOIP 0x000001dc
+#define MSR_IA32_LASTINTFROMIP 0x000001dd
+#define MSR_IA32_LASTINTTOIP 0x000001de
+
+/* DEBUGCTLMSR bits (others vary by model): */
+#define DEBUGCTLMSR_LBR (1UL << 0) /* last branch recording */
+#define DEBUGCTLMSR_BTF (1UL << 1) /* single-step on branches */
+#define DEBUGCTLMSR_TR (1UL << 6)
+#define DEBUGCTLMSR_BTS (1UL << 7)
+#define DEBUGCTLMSR_BTINT (1UL << 8)
+#define DEBUGCTLMSR_BTS_OFF_OS (1UL << 9)
+#define DEBUGCTLMSR_BTS_OFF_USR (1UL << 10)
+#define DEBUGCTLMSR_FREEZE_LBRS_ON_PMI (1UL << 11)
+
+#define MSR_IA32_POWER_CTL 0x000001fc
+
+#define MSR_IA32_MC0_CTL 0x00000400
+#define MSR_IA32_MC0_STATUS 0x00000401
+#define MSR_IA32_MC0_ADDR 0x00000402
+#define MSR_IA32_MC0_MISC 0x00000403
+
+/* C-state Residency Counters */
+#define MSR_PKG_C3_RESIDENCY 0x000003f8
+#define MSR_PKG_C6_RESIDENCY 0x000003f9
+#define MSR_PKG_C7_RESIDENCY 0x000003fa
+#define MSR_CORE_C3_RESIDENCY 0x000003fc
+#define MSR_CORE_C6_RESIDENCY 0x000003fd
+#define MSR_CORE_C7_RESIDENCY 0x000003fe
+#define MSR_PKG_C2_RESIDENCY 0x0000060d
+#define MSR_PKG_C8_RESIDENCY 0x00000630
+#define MSR_PKG_C9_RESIDENCY 0x00000631
+#define MSR_PKG_C10_RESIDENCY 0x00000632
+
+/* Run Time Average Power Limiting (RAPL) Interface */
+
+#define MSR_RAPL_POWER_UNIT 0x00000606
+
+#define MSR_PKG_POWER_LIMIT 0x00000610
+#define MSR_PKG_ENERGY_STATUS 0x00000611
+#define MSR_PKG_PERF_STATUS 0x00000613
+#define MSR_PKG_POWER_INFO 0x00000614
+
+#define MSR_DRAM_POWER_LIMIT 0x00000618
+#define MSR_DRAM_ENERGY_STATUS 0x00000619
+#define MSR_DRAM_PERF_STATUS 0x0000061b
+#define MSR_DRAM_POWER_INFO 0x0000061c
+
+#define MSR_PP0_POWER_LIMIT 0x00000638
+#define MSR_PP0_ENERGY_STATUS 0x00000639
+#define MSR_PP0_POLICY 0x0000063a
+#define MSR_PP0_PERF_STATUS 0x0000063b
+
+#define MSR_PP1_POWER_LIMIT 0x00000640
+#define MSR_PP1_ENERGY_STATUS 0x00000641
+#define MSR_PP1_POLICY 0x00000642
+
+#define MSR_CORE_C1_RES 0x00000660
+
+#define MSR_AMD64_MC0_MASK 0xc0010044
+
+#define MSR_IA32_MCx_CTL(x) (MSR_IA32_MC0_CTL + 4*(x))
+#define MSR_IA32_MCx_STATUS(x) (MSR_IA32_MC0_STATUS + 4*(x))
+#define MSR_IA32_MCx_ADDR(x) (MSR_IA32_MC0_ADDR + 4*(x))
+#define MSR_IA32_MCx_MISC(x) (MSR_IA32_MC0_MISC + 4*(x))
+
+#define MSR_AMD64_MCx_MASK(x) (MSR_AMD64_MC0_MASK + (x))
+
+/* These are consecutive and not in the normal 4er MCE bank block */
+#define MSR_IA32_MC0_CTL2 0x00000280
+#define MSR_IA32_MCx_CTL2(x) (MSR_IA32_MC0_CTL2 + (x))
+
+#define MSR_P6_PERFCTR0 0x000000c1
+#define MSR_P6_PERFCTR1 0x000000c2
+#define MSR_P6_EVNTSEL0 0x00000186
+#define MSR_P6_EVNTSEL1 0x00000187
+
+#define MSR_KNC_PERFCTR0 0x00000020
+#define MSR_KNC_PERFCTR1 0x00000021
+#define MSR_KNC_EVNTSEL0 0x00000028
+#define MSR_KNC_EVNTSEL1 0x00000029
+
+/* Alternative perfctr range with full access. */
+#define MSR_IA32_PMC0 0x000004c1
+
+/* AMD64 MSRs. Not complete. See the architecture manual for a more
+ complete list. */
+
+#define MSR_AMD64_PATCH_LEVEL 0x0000008b
+#define MSR_AMD64_TSC_RATIO 0xc0000104
+#define MSR_AMD64_NB_CFG 0xc001001f
+#define MSR_AMD64_PATCH_LOADER 0xc0010020
+#define MSR_AMD64_OSVW_ID_LENGTH 0xc0010140
+#define MSR_AMD64_OSVW_STATUS 0xc0010141
+#define MSR_AMD64_LS_CFG 0xc0011020
+#define MSR_AMD64_DC_CFG 0xc0011022
+#define MSR_AMD64_BU_CFG2 0xc001102a
+#define MSR_AMD64_IBSFETCHCTL 0xc0011030
+#define MSR_AMD64_IBSFETCHLINAD 0xc0011031
+#define MSR_AMD64_IBSFETCHPHYSAD 0xc0011032
+#define MSR_AMD64_IBSFETCH_REG_COUNT 3
+#define MSR_AMD64_IBSFETCH_REG_MASK ((1UL<<MSR_AMD64_IBSFETCH_REG_COUNT)-1)
+#define MSR_AMD64_IBSOPCTL 0xc0011033
+#define MSR_AMD64_IBSOPRIP 0xc0011034
+#define MSR_AMD64_IBSOPDATA 0xc0011035
+#define MSR_AMD64_IBSOPDATA2 0xc0011036
+#define MSR_AMD64_IBSOPDATA3 0xc0011037
+#define MSR_AMD64_IBSDCLINAD 0xc0011038
+#define MSR_AMD64_IBSDCPHYSAD 0xc0011039
+#define MSR_AMD64_IBSOP_REG_COUNT 7
+#define MSR_AMD64_IBSOP_REG_MASK ((1UL<<MSR_AMD64_IBSOP_REG_COUNT)-1)
+#define MSR_AMD64_IBSCTL 0xc001103a
+#define MSR_AMD64_IBSBRTARGET 0xc001103b
+#define MSR_AMD64_IBS_REG_COUNT_MAX 8 /* includes MSR_AMD64_IBSBRTARGET */
+
+/* Fam 16h MSRs */
+#define MSR_F16H_L2I_PERF_CTL 0xc0010230
+#define MSR_F16H_L2I_PERF_CTR 0xc0010231
+
+/* Fam 15h MSRs */
+#define MSR_F15H_PERF_CTL 0xc0010200
+#define MSR_F15H_PERF_CTR 0xc0010201
+#define MSR_F15H_NB_PERF_CTL 0xc0010240
+#define MSR_F15H_NB_PERF_CTR 0xc0010241
+
+/* Fam 10h MSRs */
+#define MSR_FAM10H_MMIO_CONF_BASE 0xc0010058
+#define FAM10H_MMIO_CONF_ENABLE (1<<0)
+#define FAM10H_MMIO_CONF_BUSRANGE_MASK 0xf
+#define FAM10H_MMIO_CONF_BUSRANGE_SHIFT 2
+#define FAM10H_MMIO_CONF_BASE_MASK 0xfffffffULL
+#define FAM10H_MMIO_CONF_BASE_SHIFT 20
+#define MSR_FAM10H_NODE_ID 0xc001100c
+
+/* K8 MSRs */
+#define MSR_K8_TOP_MEM1 0xc001001a
+#define MSR_K8_TOP_MEM2 0xc001001d
+#define MSR_K8_SYSCFG 0xc0010010
+#define MSR_K8_INT_PENDING_MSG 0xc0010055
+/* C1E active bits in int pending message */
+#define K8_INTP_C1E_ACTIVE_MASK 0x18000000
+#define MSR_K8_TSEG_ADDR 0xc0010112
+#define K8_MTRRFIXRANGE_DRAM_ENABLE 0x00040000 /* MtrrFixDramEn bit */
+#define K8_MTRRFIXRANGE_DRAM_MODIFY 0x00080000 /* MtrrFixDramModEn bit */
+#define K8_MTRR_RDMEM_WRMEM_MASK 0x18181818 /* Mask: RdMem|WrMem */
+
+/* K7 MSRs */
+#define MSR_K7_EVNTSEL0 0xc0010000
+#define MSR_K7_PERFCTR0 0xc0010004
+#define MSR_K7_EVNTSEL1 0xc0010001
+#define MSR_K7_PERFCTR1 0xc0010005
+#define MSR_K7_EVNTSEL2 0xc0010002
+#define MSR_K7_PERFCTR2 0xc0010006
+#define MSR_K7_EVNTSEL3 0xc0010003
+#define MSR_K7_PERFCTR3 0xc0010007
+#define MSR_K7_CLK_CTL 0xc001001b
+#define MSR_K7_HWCR 0xc0010015
+#define MSR_K7_FID_VID_CTL 0xc0010041
+#define MSR_K7_FID_VID_STATUS 0xc0010042
+
+/* K6 MSRs */
+#define MSR_K6_WHCR 0xc0000082
+#define MSR_K6_UWCCR 0xc0000085
+#define MSR_K6_EPMR 0xc0000086
+#define MSR_K6_PSOR 0xc0000087
+#define MSR_K6_PFIR 0xc0000088
+
+/* Centaur-Hauls/IDT defined MSRs. */
+#define MSR_IDT_FCR1 0x00000107
+#define MSR_IDT_FCR2 0x00000108
+#define MSR_IDT_FCR3 0x00000109
+#define MSR_IDT_FCR4 0x0000010a
+
+#define MSR_IDT_MCR0 0x00000110
+#define MSR_IDT_MCR1 0x00000111
+#define MSR_IDT_MCR2 0x00000112
+#define MSR_IDT_MCR3 0x00000113
+#define MSR_IDT_MCR4 0x00000114
+#define MSR_IDT_MCR5 0x00000115
+#define MSR_IDT_MCR6 0x00000116
+#define MSR_IDT_MCR7 0x00000117
+#define MSR_IDT_MCR_CTRL 0x00000120
+
+/* VIA Cyrix defined MSRs*/
+#define MSR_VIA_FCR 0x00001107
+#define MSR_VIA_LONGHAUL 0x0000110a
+#define MSR_VIA_RNG 0x0000110b
+#define MSR_VIA_BCR2 0x00001147
+
+/* Transmeta defined MSRs */
+#define MSR_TMTA_LONGRUN_CTRL 0x80868010
+#define MSR_TMTA_LONGRUN_FLAGS 0x80868011
+#define MSR_TMTA_LRTI_READOUT 0x80868018
+#define MSR_TMTA_LRTI_VOLT_MHZ 0x8086801a
+
+/* Intel defined MSRs. */
+#define MSR_IA32_P5_MC_ADDR 0x00000000
+#define MSR_IA32_P5_MC_TYPE 0x00000001
+#define MSR_IA32_TSC 0x00000010
+#define MSR_IA32_PLATFORM_ID 0x00000017
+#define MSR_IA32_EBL_CR_POWERON 0x0000002a
+#define MSR_EBC_FREQUENCY_ID 0x0000002c
+#define MSR_SMI_COUNT 0x00000034
+#define MSR_IA32_FEATURE_CONTROL 0x0000003a
+#define MSR_IA32_TSC_ADJUST 0x0000003b
+
+#define FEATURE_CONTROL_LOCKED (1<<0)
+#define FEATURE_CONTROL_VMXON_ENABLED_INSIDE_SMX (1<<1)
+#define FEATURE_CONTROL_VMXON_ENABLED_OUTSIDE_SMX (1<<2)
+
+#define MSR_IA32_APICBASE 0x0000001b
+#define MSR_IA32_APICBASE_BSP (1<<8)
+#define MSR_IA32_APICBASE_ENABLE (1<<11)
+#define MSR_IA32_APICBASE_BASE (0xfffff<<12)
+
+#define MSR_IA32_TSCDEADLINE 0x000006e0
+
+#define MSR_IA32_UCODE_WRITE 0x00000079
+#define MSR_IA32_UCODE_REV 0x0000008b
+
+#define MSR_IA32_PERF_STATUS 0x00000198
+#define MSR_IA32_PERF_CTL 0x00000199
+#define MSR_AMD_PSTATE_DEF_BASE 0xc0010064
+#define MSR_AMD_PERF_STATUS 0xc0010063
+#define MSR_AMD_PERF_CTL 0xc0010062
+
+#define MSR_IA32_MPERF 0x000000e7
+#define MSR_IA32_APERF 0x000000e8
+
+#define MSR_IA32_THERM_CONTROL 0x0000019a
+#define MSR_IA32_THERM_INTERRUPT 0x0000019b
+
+#define THERM_INT_HIGH_ENABLE (1 << 0)
+#define THERM_INT_LOW_ENABLE (1 << 1)
+#define THERM_INT_PLN_ENABLE (1 << 24)
+
+#define MSR_IA32_THERM_STATUS 0x0000019c
+
+#define THERM_STATUS_PROCHOT (1 << 0)
+#define THERM_STATUS_POWER_LIMIT (1 << 10)
+
+#define MSR_THERM2_CTL 0x0000019d
+
+#define MSR_THERM2_CTL_TM_SELECT (1ULL << 16)
+
+#define MSR_IA32_MISC_ENABLE 0x000001a0
+
+#define MSR_IA32_TEMPERATURE_TARGET 0x000001a2
+
+#define MSR_IA32_ENERGY_PERF_BIAS 0x000001b0
+#define ENERGY_PERF_BIAS_PERFORMANCE 0
+#define ENERGY_PERF_BIAS_NORMAL 6
+#define ENERGY_PERF_BIAS_POWERSAVE 15
+
+#define MSR_IA32_PACKAGE_THERM_STATUS 0x000001b1
+
+#define PACKAGE_THERM_STATUS_PROCHOT (1 << 0)
+#define PACKAGE_THERM_STATUS_POWER_LIMIT (1 << 10)
+
+#define MSR_IA32_PACKAGE_THERM_INTERRUPT 0x000001b2
+
+#define PACKAGE_THERM_INT_HIGH_ENABLE (1 << 0)
+#define PACKAGE_THERM_INT_LOW_ENABLE (1 << 1)
+#define PACKAGE_THERM_INT_PLN_ENABLE (1 << 24)
+
+/* Thermal Thresholds Support */
+#define THERM_INT_THRESHOLD0_ENABLE (1 << 15)
+#define THERM_SHIFT_THRESHOLD0 8
+#define THERM_MASK_THRESHOLD0 (0x7f << THERM_SHIFT_THRESHOLD0)
+#define THERM_INT_THRESHOLD1_ENABLE (1 << 23)
+#define THERM_SHIFT_THRESHOLD1 16
+#define THERM_MASK_THRESHOLD1 (0x7f << THERM_SHIFT_THRESHOLD1)
+#define THERM_STATUS_THRESHOLD0 (1 << 6)
+#define THERM_LOG_THRESHOLD0 (1 << 7)
+#define THERM_STATUS_THRESHOLD1 (1 << 8)
+#define THERM_LOG_THRESHOLD1 (1 << 9)
+
+/* MISC_ENABLE bits: architectural */
+#define MSR_IA32_MISC_ENABLE_FAST_STRING (1ULL << 0)
+#define MSR_IA32_MISC_ENABLE_TCC (1ULL << 1)
+#define MSR_IA32_MISC_ENABLE_EMON (1ULL << 7)
+#define MSR_IA32_MISC_ENABLE_BTS_UNAVAIL (1ULL << 11)
+#define MSR_IA32_MISC_ENABLE_PEBS_UNAVAIL (1ULL << 12)
+#define MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP (1ULL << 16)
+#define MSR_IA32_MISC_ENABLE_MWAIT (1ULL << 18)
+#define MSR_IA32_MISC_ENABLE_LIMIT_CPUID (1ULL << 22)
+#define MSR_IA32_MISC_ENABLE_XTPR_DISABLE (1ULL << 23)
+#define MSR_IA32_MISC_ENABLE_XD_DISABLE (1ULL << 34)
+
+/* MISC_ENABLE bits: model-specific, meaning may vary from core to core */
+#define MSR_IA32_MISC_ENABLE_X87_COMPAT (1ULL << 2)
+#define MSR_IA32_MISC_ENABLE_TM1 (1ULL << 3)
+#define MSR_IA32_MISC_ENABLE_SPLIT_LOCK_DISABLE (1ULL << 4)
+#define MSR_IA32_MISC_ENABLE_L3CACHE_DISABLE (1ULL << 6)
+#define MSR_IA32_MISC_ENABLE_SUPPRESS_LOCK (1ULL << 8)
+#define MSR_IA32_MISC_ENABLE_PREFETCH_DISABLE (1ULL << 9)
+#define MSR_IA32_MISC_ENABLE_FERR (1ULL << 10)
+#define MSR_IA32_MISC_ENABLE_FERR_MULTIPLEX (1ULL << 10)
+#define MSR_IA32_MISC_ENABLE_TM2 (1ULL << 13)
+#define MSR_IA32_MISC_ENABLE_ADJ_PREF_DISABLE (1ULL << 19)
+#define MSR_IA32_MISC_ENABLE_SPEEDSTEP_LOCK (1ULL << 20)
+#define MSR_IA32_MISC_ENABLE_L1D_CONTEXT (1ULL << 24)
+#define MSR_IA32_MISC_ENABLE_DCU_PREF_DISABLE (1ULL << 37)
+#define MSR_IA32_MISC_ENABLE_TURBO_DISABLE (1ULL << 38)
+#define MSR_IA32_MISC_ENABLE_IP_PREF_DISABLE (1ULL << 39)
+
+#define MSR_IA32_TSC_DEADLINE 0x000006E0
+
+/* P4/Xeon+ specific */
+#define MSR_IA32_MCG_EAX 0x00000180
+#define MSR_IA32_MCG_EBX 0x00000181
+#define MSR_IA32_MCG_ECX 0x00000182
+#define MSR_IA32_MCG_EDX 0x00000183
+#define MSR_IA32_MCG_ESI 0x00000184
+#define MSR_IA32_MCG_EDI 0x00000185
+#define MSR_IA32_MCG_EBP 0x00000186
+#define MSR_IA32_MCG_ESP 0x00000187
+#define MSR_IA32_MCG_EFLAGS 0x00000188
+#define MSR_IA32_MCG_EIP 0x00000189
+#define MSR_IA32_MCG_RESERVED 0x0000018a
+
+/* Pentium IV performance counter MSRs */
+#define MSR_P4_BPU_PERFCTR0 0x00000300
+#define MSR_P4_BPU_PERFCTR1 0x00000301
+#define MSR_P4_BPU_PERFCTR2 0x00000302
+#define MSR_P4_BPU_PERFCTR3 0x00000303
+#define MSR_P4_MS_PERFCTR0 0x00000304
+#define MSR_P4_MS_PERFCTR1 0x00000305
+#define MSR_P4_MS_PERFCTR2 0x00000306
+#define MSR_P4_MS_PERFCTR3 0x00000307
+#define MSR_P4_FLAME_PERFCTR0 0x00000308
+#define MSR_P4_FLAME_PERFCTR1 0x00000309
+#define MSR_P4_FLAME_PERFCTR2 0x0000030a
+#define MSR_P4_FLAME_PERFCTR3 0x0000030b
+#define MSR_P4_IQ_PERFCTR0 0x0000030c
+#define MSR_P4_IQ_PERFCTR1 0x0000030d
+#define MSR_P4_IQ_PERFCTR2 0x0000030e
+#define MSR_P4_IQ_PERFCTR3 0x0000030f
+#define MSR_P4_IQ_PERFCTR4 0x00000310
+#define MSR_P4_IQ_PERFCTR5 0x00000311
+#define MSR_P4_BPU_CCCR0 0x00000360
+#define MSR_P4_BPU_CCCR1 0x00000361
+#define MSR_P4_BPU_CCCR2 0x00000362
+#define MSR_P4_BPU_CCCR3 0x00000363
+#define MSR_P4_MS_CCCR0 0x00000364
+#define MSR_P4_MS_CCCR1 0x00000365
+#define MSR_P4_MS_CCCR2 0x00000366
+#define MSR_P4_MS_CCCR3 0x00000367
+#define MSR_P4_FLAME_CCCR0 0x00000368
+#define MSR_P4_FLAME_CCCR1 0x00000369
+#define MSR_P4_FLAME_CCCR2 0x0000036a
+#define MSR_P4_FLAME_CCCR3 0x0000036b
+#define MSR_P4_IQ_CCCR0 0x0000036c
+#define MSR_P4_IQ_CCCR1 0x0000036d
+#define MSR_P4_IQ_CCCR2 0x0000036e
+#define MSR_P4_IQ_CCCR3 0x0000036f
+#define MSR_P4_IQ_CCCR4 0x00000370
+#define MSR_P4_IQ_CCCR5 0x00000371
+#define MSR_P4_ALF_ESCR0 0x000003ca
+#define MSR_P4_ALF_ESCR1 0x000003cb
+#define MSR_P4_BPU_ESCR0 0x000003b2
+#define MSR_P4_BPU_ESCR1 0x000003b3
+#define MSR_P4_BSU_ESCR0 0x000003a0
+#define MSR_P4_BSU_ESCR1 0x000003a1
+#define MSR_P4_CRU_ESCR0 0x000003b8
+#define MSR_P4_CRU_ESCR1 0x000003b9
+#define MSR_P4_CRU_ESCR2 0x000003cc
+#define MSR_P4_CRU_ESCR3 0x000003cd
+#define MSR_P4_CRU_ESCR4 0x000003e0
+#define MSR_P4_CRU_ESCR5 0x000003e1
+#define MSR_P4_DAC_ESCR0 0x000003a8
+#define MSR_P4_DAC_ESCR1 0x000003a9
+#define MSR_P4_FIRM_ESCR0 0x000003a4
+#define MSR_P4_FIRM_ESCR1 0x000003a5
+#define MSR_P4_FLAME_ESCR0 0x000003a6
+#define MSR_P4_FLAME_ESCR1 0x000003a7
+#define MSR_P4_FSB_ESCR0 0x000003a2
+#define MSR_P4_FSB_ESCR1 0x000003a3
+#define MSR_P4_IQ_ESCR0 0x000003ba
+#define MSR_P4_IQ_ESCR1 0x000003bb
+#define MSR_P4_IS_ESCR0 0x000003b4
+#define MSR_P4_IS_ESCR1 0x000003b5
+#define MSR_P4_ITLB_ESCR0 0x000003b6
+#define MSR_P4_ITLB_ESCR1 0x000003b7
+#define MSR_P4_IX_ESCR0 0x000003c8
+#define MSR_P4_IX_ESCR1 0x000003c9
+#define MSR_P4_MOB_ESCR0 0x000003aa
+#define MSR_P4_MOB_ESCR1 0x000003ab
+#define MSR_P4_MS_ESCR0 0x000003c0
+#define MSR_P4_MS_ESCR1 0x000003c1
+#define MSR_P4_PMH_ESCR0 0x000003ac
+#define MSR_P4_PMH_ESCR1 0x000003ad
+#define MSR_P4_RAT_ESCR0 0x000003bc
+#define MSR_P4_RAT_ESCR1 0x000003bd
+#define MSR_P4_SAAT_ESCR0 0x000003ae
+#define MSR_P4_SAAT_ESCR1 0x000003af
+#define MSR_P4_SSU_ESCR0 0x000003be
+#define MSR_P4_SSU_ESCR1 0x000003bf /* guess: not in manual */
+
+#define MSR_P4_TBPU_ESCR0 0x000003c2
+#define MSR_P4_TBPU_ESCR1 0x000003c3
+#define MSR_P4_TC_ESCR0 0x000003c4
+#define MSR_P4_TC_ESCR1 0x000003c5
+#define MSR_P4_U2L_ESCR0 0x000003b0
+#define MSR_P4_U2L_ESCR1 0x000003b1
+
+#define MSR_P4_PEBS_MATRIX_VERT 0x000003f2
+
+/* Intel Core-based CPU performance counters */
+#define MSR_CORE_PERF_FIXED_CTR0 0x00000309
+#define MSR_CORE_PERF_FIXED_CTR1 0x0000030a
+#define MSR_CORE_PERF_FIXED_CTR2 0x0000030b
+#define MSR_CORE_PERF_FIXED_CTR_CTRL 0x0000038d
+#define MSR_CORE_PERF_GLOBAL_STATUS 0x0000038e
+#define MSR_CORE_PERF_GLOBAL_CTRL 0x0000038f
+#define MSR_CORE_PERF_GLOBAL_OVF_CTRL 0x00000390
+
+/* Geode defined MSRs */
+#define MSR_GEODE_BUSCONT_CONF0 0x00001900
+
+/* Intel VT MSRs */
+#define MSR_IA32_VMX_BASIC 0x00000480
+#define MSR_IA32_VMX_PINBASED_CTLS 0x00000481
+#define MSR_IA32_VMX_PROCBASED_CTLS 0x00000482
+#define MSR_IA32_VMX_EXIT_CTLS 0x00000483
+#define MSR_IA32_VMX_ENTRY_CTLS 0x00000484
+#define MSR_IA32_VMX_MISC 0x00000485
+#define MSR_IA32_VMX_CR0_FIXED0 0x00000486
+#define MSR_IA32_VMX_CR0_FIXED1 0x00000487
+#define MSR_IA32_VMX_CR4_FIXED0 0x00000488
+#define MSR_IA32_VMX_CR4_FIXED1 0x00000489
+#define MSR_IA32_VMX_VMCS_ENUM 0x0000048a
+#define MSR_IA32_VMX_PROCBASED_CTLS2 0x0000048b
+#define MSR_IA32_VMX_EPT_VPID_CAP 0x0000048c
+#define MSR_IA32_VMX_TRUE_PINBASED_CTLS 0x0000048d
+#define MSR_IA32_VMX_TRUE_PROCBASED_CTLS 0x0000048e
+#define MSR_IA32_VMX_TRUE_EXIT_CTLS 0x0000048f
+#define MSR_IA32_VMX_TRUE_ENTRY_CTLS 0x00000490
+#define MSR_IA32_VMX_VMFUNC 0x00000491
+
+/* VMX_BASIC bits and bitmasks */
+#define VMX_BASIC_VMCS_SIZE_SHIFT 32
+#define VMX_BASIC_64 0x0001000000000000LLU
+#define VMX_BASIC_MEM_TYPE_SHIFT 50
+#define VMX_BASIC_MEM_TYPE_MASK 0x003c000000000000LLU
+#define VMX_BASIC_MEM_TYPE_WB 6LLU
+#define VMX_BASIC_INOUT 0x0040000000000000LLU
+
+/* MSR_IA32_VMX_MISC bits */
+#define MSR_IA32_VMX_MISC_VMWRITE_SHADOW_RO_FIELDS (1ULL << 29)
+#define MSR_IA32_VMX_MISC_PREEMPTION_TIMER_SCALE 0x1F
+/* AMD-V MSRs */
+
+#define MSR_VM_CR 0xc0010114
+#define MSR_VM_IGNNE 0xc0010115
+#define MSR_VM_HSAVE_PA 0xc0010117
+
+#endif /* _ASM_X86_MSR_INDEX_H */
--- /dev/null
+/*
+ * turbostat -- Log CPU frequency and C-state residency
+ * on modern Intel turbo-capable processors for collectd.
+ *
+ * Based on the kernel tools by:
+ * Copyright (c) 2013 Intel Corporation.
+ * Len Brown <len.brown@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Ported to collectd by Vincent Brillault <git@lerya.net>
+ */
+
+#define _GNU_SOURCE
+#include "msr-index.h"
+#include <stdarg.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <err.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <sys/stat.h>
+#include <sys/resource.h>
+#include <fcntl.h>
+#include <signal.h>
+#include <sys/time.h>
+#include <stdlib.h>
+#include <dirent.h>
+#include <string.h>
+#include <ctype.h>
+#include <sched.h>
+#include <cpuid.h>
+
+#include "collectd.h"
+#include "common.h"
+#include "plugin.h"
+
+#define PLUGIN_NAME "turbostat"
+
+static const char *proc_stat = "/proc/stat";
+static unsigned int interval_sec = 5; /* set with -i interval_sec */
+static unsigned int skip_c0;
+static unsigned int skip_c1;
+static unsigned int do_nhm_cstates;
+static unsigned int do_snb_cstates;
+static unsigned int do_c8_c9_c10;
+static unsigned int do_slm_cstates;
+static unsigned int has_aperf;
+static unsigned int has_epb;
+static unsigned int units = 1000000000; /* Ghz etc */
+static unsigned int genuine_intel;
+static unsigned int has_invariant_tsc;
+static unsigned int do_nehalem_platform_info;
+static int do_smi;
+static unsigned int show_pkg;
+static unsigned int show_core;
+static unsigned int show_cpu;
+static unsigned int do_rapl;
+static unsigned int do_dts;
+static unsigned int do_ptm;
+static unsigned int tcc_activation_temp;
+static unsigned int tcc_activation_temp_override;
+static double rapl_power_units, rapl_energy_units, rapl_time_units;
+static double rapl_joule_counter_range;
+
+#define RAPL_PKG (1 << 0)
+ /* 0x610 MSR_PKG_POWER_LIMIT */
+ /* 0x611 MSR_PKG_ENERGY_STATUS */
+#define RAPL_PKG_PERF_STATUS (1 << 1)
+ /* 0x613 MSR_PKG_PERF_STATUS */
+#define RAPL_PKG_POWER_INFO (1 << 2)
+ /* 0x614 MSR_PKG_POWER_INFO */
+
+#define RAPL_DRAM (1 << 3)
+ /* 0x618 MSR_DRAM_POWER_LIMIT */
+ /* 0x619 MSR_DRAM_ENERGY_STATUS */
+ /* 0x61c MSR_DRAM_POWER_INFO */
+#define RAPL_DRAM_PERF_STATUS (1 << 4)
+ /* 0x61b MSR_DRAM_PERF_STATUS */
+
+#define RAPL_CORES (1 << 5)
+ /* 0x638 MSR_PP0_POWER_LIMIT */
+ /* 0x639 MSR_PP0_ENERGY_STATUS */
+#define RAPL_CORE_POLICY (1 << 6)
+ /* 0x63a MSR_PP0_POLICY */
+
+
+#define RAPL_GFX (1 << 7)
+ /* 0x640 MSR_PP1_POWER_LIMIT */
+ /* 0x641 MSR_PP1_ENERGY_STATUS */
+ /* 0x642 MSR_PP1_POLICY */
+#define TJMAX_DEFAULT 100
+
+int aperf_mperf_unstable;
+int backwards_count;
+char *progname;
+
+cpu_set_t *cpu_present_set, *cpu_affinity_set;
+size_t cpu_present_setsize, cpu_affinity_setsize;
+
+struct thread_data {
+ unsigned long long tsc;
+ unsigned long long aperf;
+ unsigned long long mperf;
+ unsigned long long c1;
+ unsigned int smi_count;
+ unsigned int cpu_id;
+ unsigned int flags;
+#define CPU_IS_FIRST_THREAD_IN_CORE 0x2
+#define CPU_IS_FIRST_CORE_IN_PACKAGE 0x4
+} *thread_even, *thread_odd;
+
+struct core_data {
+ unsigned long long c3;
+ unsigned long long c6;
+ unsigned long long c7;
+ unsigned int core_temp_c;
+ unsigned int core_id;
+} *core_even, *core_odd;
+
+struct pkg_data {
+ unsigned long long pc2;
+ unsigned long long pc3;
+ unsigned long long pc6;
+ unsigned long long pc7;
+ unsigned long long pc8;
+ unsigned long long pc9;
+ unsigned long long pc10;
+ unsigned int package_id;
+ unsigned int energy_pkg; /* MSR_PKG_ENERGY_STATUS */
+ unsigned int energy_dram; /* MSR_DRAM_ENERGY_STATUS */
+ unsigned int energy_cores; /* MSR_PP0_ENERGY_STATUS */
+ unsigned int energy_gfx; /* MSR_PP1_ENERGY_STATUS */
+ unsigned int rapl_pkg_perf_status; /* MSR_PKG_PERF_STATUS */
+ unsigned int rapl_dram_perf_status; /* MSR_DRAM_PERF_STATUS */
+ unsigned int pkg_temp_c;
+
+} *package_even, *package_odd;
+
+#define ODD_COUNTERS thread_odd, core_odd, package_odd
+#define EVEN_COUNTERS thread_even, core_even, package_even
+static bool is_even = true;
+
+static bool allocated = false;
+static bool initialized = false;
+
+#define GET_THREAD(thread_base, thread_no, core_no, pkg_no) \
+ (thread_base + (pkg_no) * topo.num_cores_per_pkg * \
+ topo.num_threads_per_core + \
+ (core_no) * topo.num_threads_per_core + (thread_no))
+#define GET_CORE(core_base, core_no, pkg_no) \
+ (core_base + (pkg_no) * topo.num_cores_per_pkg + (core_no))
+#define GET_PKG(pkg_base, pkg_no) (pkg_base + pkg_no)
+
+struct topo_params {
+ int num_packages;
+ int num_cpus;
+ int num_cores;
+ int max_cpu_num;
+ int num_cores_per_pkg;
+ int num_threads_per_core;
+} topo;
+
+struct timeval tv_even, tv_odd, tv_delta;
+
+enum return_values {
+ OK = 0,
+ ERR_CPU_MIGRATE,
+ ERR_MSR_IA32_APERF,
+ ERR_MSR_IA32_MPERF,
+ ERR_MSR_SMI_COUNT,
+ ERR_MSR_CORE_C3_RESIDENCY,
+ ERR_MSR_CORE_C6_RESIDENCY,
+ ERR_MSR_CORE_C7_RESIDENCY,
+ ERR_MSR_IA32_THERM_STATUS,
+ ERR_MSR_PKG_C3_RESIDENCY,
+ ERR_MSR_PKG_C6_RESIDENCY,
+ ERR_MSR_PKG_C2_RESIDENCY,
+ ERR_MSR_PKG_C7_RESIDENCY,
+ ERR_MSR_PKG_C8_RESIDENCY,
+ ERR_MSR_PKG_C9_RESIDENCY,
+ ERR_MSR_PKG_C10_RESIDENCY,
+ ERR_MSR_PKG_ENERGY_STATUS,
+ ERR_MSR_PP0_ENERGY_STATUS,
+ ERR_MSR_DRAM_ENERGY_STATUS,
+ ERR_MSR_PP1_ENERGY_STATUS,
+ ERR_MSR_PKG_PERF_STATUS,
+ ERR_MSR_DRAM_PERF_STATUS,
+ ERR_MSR_IA32_PACKAGE_THERM_STATUS,
+ ERR_CPU_NOT_PRESENT,
+ ERR_NO_MSR,
+ ERR_CANT_OPEN_FILE,
+ ERR_CANT_READ_NUMBER,
+ ERR_CANT_READ_PROC_STAT,
+ ERR_NO_INVARIANT_TSC,
+ ERR_NO_APERF,
+ ERR_CALLOC,
+ ERR_CPU_ALLOC,
+ ERR_NOT_ROOT,
+};
+
+#define STATIC_MUST_CHECK(function) \
+function \
+ __attribute__((warn_unused_result)); \
+function
+
+static int setup_all_buffers(void);
+
+static int cpu_is_not_present(int cpu)
+{
+ return !CPU_ISSET_S(cpu, cpu_present_setsize, cpu_present_set);
+}
+/*
+ * run func(thread, core, package) in topology order
+ * skip non-present cpus
+ */
+
+STATIC_MUST_CHECK(static int for_all_cpus(int (func)(struct thread_data *, struct core_data *, struct pkg_data *),
+ struct thread_data *thread_base, struct core_data *core_base, struct pkg_data *pkg_base))
+{
+ int retval, pkg_no, core_no, thread_no;
+
+ for (pkg_no = 0; pkg_no < topo.num_packages; ++pkg_no) {
+ for (core_no = 0; core_no < topo.num_cores_per_pkg; ++core_no) {
+ for (thread_no = 0; thread_no <
+ topo.num_threads_per_core; ++thread_no) {
+ struct thread_data *t;
+ struct core_data *c;
+ struct pkg_data *p;
+
+ t = GET_THREAD(thread_base, thread_no, core_no, pkg_no);
+
+ if (cpu_is_not_present(t->cpu_id))
+ continue;
+
+ c = GET_CORE(core_base, core_no, pkg_no);
+ p = GET_PKG(pkg_base, pkg_no);
+
+ retval = func(t, c, p);
+ if (retval)
+ return retval;
+ }
+ }
+ }
+ return 0;
+}
+
+STATIC_MUST_CHECK(static int cpu_migrate(int cpu))
+{
+ CPU_ZERO_S(cpu_affinity_setsize, cpu_affinity_set);
+ CPU_SET_S(cpu, cpu_affinity_setsize, cpu_affinity_set);
+ if (sched_setaffinity(0, cpu_affinity_setsize, cpu_affinity_set) == -1)
+ return -ERR_CPU_MIGRATE;
+ else
+ return 0;
+}
+
+STATIC_MUST_CHECK(static int get_msr(int cpu, off_t offset, unsigned long long *msr))
+{
+ ssize_t retval;
+ char pathname[32];
+ int fd;
+
+ sprintf(pathname, "/dev/cpu/%d/msr", cpu);
+ fd = open(pathname, O_RDONLY);
+ if (fd < 0)
+ return -1;
+
+ retval = pread(fd, msr, sizeof *msr, offset);
+ close(fd);
+
+ if (retval != sizeof *msr) {
+ ERROR ("%s offset 0x%llx read failed\n", pathname, (unsigned long long)offset);
+ return -1;
+ }
+
+ return 0;
+}
+
+#define DELTA_WRAP32(new, old) \
+ if (new > old) { \
+ old = new - old; \
+ } else { \
+ old = 0x100000000 + new - old; \
+ }
+
+static void
+delta_package(struct pkg_data *new, struct pkg_data *old)
+{
+ old->pc2 = new->pc2 - old->pc2;
+ old->pc3 = new->pc3 - old->pc3;
+ old->pc6 = new->pc6 - old->pc6;
+ old->pc7 = new->pc7 - old->pc7;
+ old->pc8 = new->pc8 - old->pc8;
+ old->pc9 = new->pc9 - old->pc9;
+ old->pc10 = new->pc10 - old->pc10;
+ old->pkg_temp_c = new->pkg_temp_c;
+
+ DELTA_WRAP32(new->energy_pkg, old->energy_pkg);
+ DELTA_WRAP32(new->energy_cores, old->energy_cores);
+ DELTA_WRAP32(new->energy_gfx, old->energy_gfx);
+ DELTA_WRAP32(new->energy_dram, old->energy_dram);
+ DELTA_WRAP32(new->rapl_pkg_perf_status, old->rapl_pkg_perf_status);
+ DELTA_WRAP32(new->rapl_dram_perf_status, old->rapl_dram_perf_status);
+}
+
+static void
+delta_core(struct core_data *new, struct core_data *old)
+{
+ old->c3 = new->c3 - old->c3;
+ old->c6 = new->c6 - old->c6;
+ old->c7 = new->c7 - old->c7;
+ old->core_temp_c = new->core_temp_c;
+}
+
+/*
+ * old = new - old
+ */
+STATIC_MUST_CHECK(static int
+delta_thread(struct thread_data *new, struct thread_data *old,
+ struct core_data *core_delta))
+{
+ old->tsc = new->tsc - old->tsc;
+
+ /* check for TSC < 1 Mcycles over interval */
+ if (old->tsc < (1000 * 1000)) {
+ WARNING("Insanely slow TSC rate, TSC stops in idle?\n"
+ "You can disable all c-states by booting with \"idle=poll\"\n"
+ "or just the deep ones with \"processor.max_cstate=1\"");
+ return -1;
+ }
+
+ old->c1 = new->c1 - old->c1;
+
+ if ((new->aperf > old->aperf) && (new->mperf > old->mperf)) {
+ old->aperf = new->aperf - old->aperf;
+ old->mperf = new->mperf - old->mperf;
+ } else {
+
+ if (!aperf_mperf_unstable) {
+ WARNING("%s: APERF or MPERF went backwards *\n", progname);
+ WARNING("* Frequency results do not cover entire interval *\n");
+ WARNING("* fix this by running Linux-2.6.30 or later *\n");
+
+ aperf_mperf_unstable = 1;
+ }
+ /*
+ * mperf delta is likely a huge "positive" number
+ * can not use it for calculating c0 time
+ */
+ skip_c0 = 1;
+ skip_c1 = 1;
+ }
+
+
+ /*
+ * As counter collection is not atomic,
+ * it is possible for mperf's non-halted cycles + idle states
+ * to exceed TSC's all cycles: show c1 = 0% in that case.
+ */
+ if ((old->mperf + core_delta->c3 + core_delta->c6 + core_delta->c7) > old->tsc)
+ old->c1 = 0;
+ else {
+ /* normal case, derive c1 */
+ old->c1 = old->tsc - old->mperf - core_delta->c3
+ - core_delta->c6 - core_delta->c7;
+ }
+
+ if (old->mperf == 0) {
+ WARNING("cpu%d MPERF 0!\n", old->cpu_id);
+ old->mperf = 1; /* divide by 0 protection */
+ }
+
+ if (do_smi)
+ old->smi_count = new->smi_count - old->smi_count;
+
+ return 0;
+}
+
+STATIC_MUST_CHECK(static int delta_cpu(struct thread_data *t, struct core_data *c,
+ struct pkg_data *p, struct thread_data *t2,
+ struct core_data *c2, struct pkg_data *p2))
+{
+ int ret;
+
+ /* calculate core delta only for 1st thread in core */
+ if (t->flags & CPU_IS_FIRST_THREAD_IN_CORE)
+ delta_core(c, c2);
+
+ /* always calculate thread delta */
+ ret = delta_thread(t, t2, c2); /* c2 is core delta */
+ if (ret != 0)
+ return ret;
+
+ /* calculate package delta only for 1st core in package */
+ if (t->flags & CPU_IS_FIRST_CORE_IN_PACKAGE)
+ delta_package(p, p2);
+
+ return 0;
+}
+
+static unsigned long long rdtsc(void)
+{
+ unsigned int low, high;
+
+ asm volatile("rdtsc" : "=a" (low), "=d" (high));
+
+ return low | ((unsigned long long)high) << 32;
+}
+
+
+/*
+ * get_counters(...)
+ * migrate to cpu
+ * acquire and record local counters for that cpu
+ */
+STATIC_MUST_CHECK(static int get_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p))
+{
+ int cpu = t->cpu_id;
+ unsigned long long msr;
+
+ if (cpu_migrate(cpu)) {
+ WARNING("Could not migrate to CPU %d\n", cpu);
+ return -ERR_CPU_MIGRATE;
+ }
+
+ t->tsc = rdtsc(); /* we are running on local CPU of interest */
+
+ if (has_aperf) {
+ if (get_msr(cpu, MSR_IA32_APERF, &t->aperf))
+ return -ERR_MSR_IA32_APERF;
+ if (get_msr(cpu, MSR_IA32_MPERF, &t->mperf))
+ return -ERR_MSR_IA32_MPERF;
+ }
+
+ if (do_smi) {
+ if (get_msr(cpu, MSR_SMI_COUNT, &msr))
+ return -ERR_MSR_SMI_COUNT;
+ t->smi_count = msr & 0xFFFFFFFF;
+ }
+
+ /* collect core counters only for 1st thread in core */
+ if (!(t->flags & CPU_IS_FIRST_THREAD_IN_CORE))
+ return 0;
+
+ if (do_nhm_cstates && !do_slm_cstates) {
+ if (get_msr(cpu, MSR_CORE_C3_RESIDENCY, &c->c3))
+ return -ERR_MSR_CORE_C3_RESIDENCY;
+ }
+
+ if (do_nhm_cstates) {
+ if (get_msr(cpu, MSR_CORE_C6_RESIDENCY, &c->c6))
+ return -ERR_MSR_CORE_C6_RESIDENCY;
+ }
+
+ if (do_snb_cstates)
+ if (get_msr(cpu, MSR_CORE_C7_RESIDENCY, &c->c7))
+ return -ERR_MSR_CORE_C7_RESIDENCY;
+
+ if (do_dts) {
+ if (get_msr(cpu, MSR_IA32_THERM_STATUS, &msr))
+ return -ERR_MSR_IA32_THERM_STATUS;
+ c->core_temp_c = tcc_activation_temp - ((msr >> 16) & 0x7F);
+ }
+
+
+ /* collect package counters only for 1st core in package */
+ if (!(t->flags & CPU_IS_FIRST_CORE_IN_PACKAGE))
+ return 0;
+
+ if (do_nhm_cstates && !do_slm_cstates) {
+ if (get_msr(cpu, MSR_PKG_C3_RESIDENCY, &p->pc3))
+ return -ERR_MSR_PKG_C3_RESIDENCY;
+ if (get_msr(cpu, MSR_PKG_C6_RESIDENCY, &p->pc6))
+ return -ERR_MSR_PKG_C6_RESIDENCY;
+ }
+ if (do_snb_cstates) {
+ if (get_msr(cpu, MSR_PKG_C2_RESIDENCY, &p->pc2))
+ return -ERR_MSR_PKG_C2_RESIDENCY;
+ if (get_msr(cpu, MSR_PKG_C7_RESIDENCY, &p->pc7))
+ return -ERR_MSR_PKG_C7_RESIDENCY;
+ }
+ if (do_c8_c9_c10) {
+ if (get_msr(cpu, MSR_PKG_C8_RESIDENCY, &p->pc8))
+ return -ERR_MSR_PKG_C8_RESIDENCY;
+ if (get_msr(cpu, MSR_PKG_C9_RESIDENCY, &p->pc9))
+ return -ERR_MSR_PKG_C9_RESIDENCY;
+ if (get_msr(cpu, MSR_PKG_C10_RESIDENCY, &p->pc10))
+ return -ERR_MSR_PKG_C10_RESIDENCY;
+ }
+ if (do_rapl & RAPL_PKG) {
+ if (get_msr(cpu, MSR_PKG_ENERGY_STATUS, &msr))
+ return -ERR_MSR_PKG_ENERGY_STATUS;
+ p->energy_pkg = msr & 0xFFFFFFFF;
+ }
+ if (do_rapl & RAPL_CORES) {
+ if (get_msr(cpu, MSR_PP0_ENERGY_STATUS, &msr))
+ return MSR_PP0_ENERGY_STATUS;
+ p->energy_cores = msr & 0xFFFFFFFF;
+ }
+ if (do_rapl & RAPL_DRAM) {
+ if (get_msr(cpu, MSR_DRAM_ENERGY_STATUS, &msr))
+ return -ERR_MSR_DRAM_ENERGY_STATUS;
+ p->energy_dram = msr & 0xFFFFFFFF;
+ }
+ if (do_rapl & RAPL_GFX) {
+ if (get_msr(cpu, MSR_PP1_ENERGY_STATUS, &msr))
+ return -ERR_MSR_PP1_ENERGY_STATUS;
+ p->energy_gfx = msr & 0xFFFFFFFF;
+ }
+ if (do_rapl & RAPL_PKG_PERF_STATUS) {
+ if (get_msr(cpu, MSR_PKG_PERF_STATUS, &msr))
+ return -ERR_MSR_PKG_PERF_STATUS;
+ p->rapl_pkg_perf_status = msr & 0xFFFFFFFF;
+ }
+ if (do_rapl & RAPL_DRAM_PERF_STATUS) {
+ if (get_msr(cpu, MSR_DRAM_PERF_STATUS, &msr))
+ return -ERR_MSR_DRAM_PERF_STATUS;
+ p->rapl_dram_perf_status = msr & 0xFFFFFFFF;
+ }
+ if (do_ptm) {
+ if (get_msr(cpu, MSR_IA32_PACKAGE_THERM_STATUS, &msr))
+ return -ERR_MSR_IA32_PACKAGE_THERM_STATUS;
+ p->pkg_temp_c = tcc_activation_temp - ((msr >> 16) & 0x7F);
+ }
+ return 0;
+}
+
+static void free_all_buffers(void)
+{
+ allocated = false;
+ initialized = false;
+
+ CPU_FREE(cpu_present_set);
+ cpu_present_set = NULL;
+ cpu_present_set = 0;
+
+ CPU_FREE(cpu_affinity_set);
+ cpu_affinity_set = NULL;
+ cpu_affinity_setsize = 0;
+
+ free(thread_even);
+ free(core_even);
+ free(package_even);
+
+ thread_even = NULL;
+ core_even = NULL;
+ package_even = NULL;
+
+ free(thread_odd);
+ free(core_odd);
+ free(package_odd);
+
+ thread_odd = NULL;
+ core_odd = NULL;
+ package_odd = NULL;
+}
+
+/*
+ * Parse a file containing a single int.
+ */
+static int parse_int_file(const char *fmt, ...)
+{
+ va_list args;
+ char path[PATH_MAX];
+ FILE *filep;
+ int value;
+
+ va_start(args, fmt);
+ vsnprintf(path, sizeof(path), fmt, args);
+ va_end(args);
+ filep = fopen(path, "r");
+ if (!filep) {
+ ERROR("%s: open failed", path);
+ return -ERR_CANT_OPEN_FILE;
+ }
+ if (fscanf(filep, "%d", &value) != 1) {
+ ERROR("%s: failed to parse number from file", path);
+ return -ERR_CANT_READ_NUMBER;
+ }
+ fclose(filep);
+ return value;
+}
+
+/*
+ * cpu_is_first_sibling_in_core(cpu)
+ * return 1 if given CPU is 1st HT sibling in the core
+ */
+static int cpu_is_first_sibling_in_core(int cpu)
+{
+ return cpu == parse_int_file("/sys/devices/system/cpu/cpu%d/topology/thread_siblings_list", cpu);
+}
+
+/*
+ * cpu_is_first_core_in_package(cpu)
+ * return 1 if given CPU is 1st core in package
+ */
+static int cpu_is_first_core_in_package(int cpu)
+{
+ return cpu == parse_int_file("/sys/devices/system/cpu/cpu%d/topology/core_siblings_list", cpu);
+}
+
+static int get_physical_package_id(int cpu)
+{
+ return parse_int_file("/sys/devices/system/cpu/cpu%d/topology/physical_package_id", cpu);
+}
+
+static int get_core_id(int cpu)
+{
+ return parse_int_file("/sys/devices/system/cpu/cpu%d/topology/core_id", cpu);
+}
+
+static int get_num_ht_siblings(int cpu)
+{
+ char path[80];
+ FILE *filep;
+ int sib1, sib2;
+ int matches;
+ char character;
+
+ sprintf(path, "/sys/devices/system/cpu/cpu%d/topology/thread_siblings_list", cpu);
+ filep = fopen(path, "r");
+ if (!filep) {
+ ERROR("%s: open failed", path);
+ return -ERR_CANT_OPEN_FILE;
+ }
+ /*
+ * file format:
+ * if a pair of number with a character between: 2 siblings (eg. 1-2, or 1,4)
+ * otherwinse 1 sibling (self).
+ */
+ matches = fscanf(filep, "%d%c%d\n", &sib1, &character, &sib2);
+
+ fclose(filep);
+
+ if (matches == 3)
+ return 2;
+ else
+ return 1;
+}
+
+/*
+ * run func(thread, core, package) in topology order
+ * skip non-present cpus
+ */
+
+STATIC_MUST_CHECK(
+static int for_all_cpus_2(int (func)(struct thread_data *, struct core_data *,
+ struct pkg_data *, struct thread_data *, struct core_data *,
+ struct pkg_data *), struct thread_data *thread_base,
+ struct core_data *core_base, struct pkg_data *pkg_base,
+ struct thread_data *thread_base2, struct core_data *core_base2,
+ struct pkg_data *pkg_base2))
+{
+ int retval, pkg_no, core_no, thread_no;
+
+ for (pkg_no = 0; pkg_no < topo.num_packages; ++pkg_no) {
+ for (core_no = 0; core_no < topo.num_cores_per_pkg; ++core_no) {
+ for (thread_no = 0; thread_no <
+ topo.num_threads_per_core; ++thread_no) {
+ struct thread_data *t, *t2;
+ struct core_data *c, *c2;
+ struct pkg_data *p, *p2;
+
+ t = GET_THREAD(thread_base, thread_no, core_no, pkg_no);
+
+ if (cpu_is_not_present(t->cpu_id))
+ continue;
+
+ t2 = GET_THREAD(thread_base2, thread_no, core_no, pkg_no);
+
+ c = GET_CORE(core_base, core_no, pkg_no);
+ c2 = GET_CORE(core_base2, core_no, pkg_no);
+
+ p = GET_PKG(pkg_base, pkg_no);
+ p2 = GET_PKG(pkg_base2, pkg_no);
+
+ retval = func(t, c, p, t2, c2, p2);
+ if (retval)
+ return retval;
+ }
+ }
+ }
+ return 0;
+}
+
+/*
+ * run func(cpu) on every cpu in /proc/stat
+ * return max_cpu number
+ */
+STATIC_MUST_CHECK(static int for_all_proc_cpus(int (func)(int)))
+{
+ FILE *fp;
+ int cpu_num;
+ int retval;
+
+ fp = fopen(proc_stat, "r");
+ if (!fp) {
+ ERROR("%s: open failed", proc_stat);
+ return -ERR_CANT_OPEN_FILE;
+ }
+
+ retval = fscanf(fp, "cpu %*d %*d %*d %*d %*d %*d %*d %*d %*d %*d\n");
+ if (retval != 0) {
+ ERROR("%s: failed to parse format", proc_stat);
+ return -ERR_CANT_READ_PROC_STAT;
+ }
+
+ while (1) {
+ retval = fscanf(fp, "cpu%u %*d %*d %*d %*d %*d %*d %*d %*d %*d %*d\n", &cpu_num);
+ if (retval != 1)
+ break;
+
+ retval = func(cpu_num);
+ if (retval) {
+ fclose(fp);
+ return(retval);
+ }
+ }
+ fclose(fp);
+ return 0;
+}
+
+/*
+ * count_cpus()
+ * remember the last one seen, it will be the max
+ */
+static int count_cpus(int cpu)
+{
+ if (topo.max_cpu_num < cpu)
+ topo.max_cpu_num = cpu;
+
+ topo.num_cpus += 1;
+ return 0;
+}
+static int mark_cpu_present(int cpu)
+{
+ CPU_SET_S(cpu, cpu_present_setsize, cpu_present_set);
+ return 0;
+}
+
+
+static void turbostat_submit (const char *plugin_instance,
+ const char *type, const char *type_instance,
+ gauge_t value)
+{
+ value_list_t vl = VALUE_LIST_INIT;
+ value_t v;
+
+ v.gauge = value;
+ vl.values = &v;
+ vl.values_len = 1;
+ sstrncpy (vl.host, hostname_g, sizeof (vl.host));
+ sstrncpy (vl.plugin, PLUGIN_NAME, sizeof (vl.plugin));
+ if (plugin_instance != NULL)
+ sstrncpy (vl.plugin_instance, plugin_instance, sizeof (vl.plugin_instance));
+ sstrncpy (vl.type, type, sizeof (vl.type));
+ if (type_instance != NULL)
+ sstrncpy (vl.type_instance, type_instance, sizeof (vl.type_instance));
+
+ plugin_dispatch_values (&vl);
+}
+
+/*
+ * column formatting convention & formats
+ * package: "pk" 2 columns %2d
+ * core: "cor" 3 columns %3d
+ * CPU: "CPU" 3 columns %3d
+ * Pkg_W: %6.2
+ * Cor_W: %6.2
+ * GFX_W: %5.2
+ * RAM_W: %5.2
+ * GHz: "GHz" 3 columns %3.2
+ * TSC: "TSC" 3 columns %3.2
+ * SMI: "SMI" 4 columns %4d
+ * percentage " %pc3" %6.2
+ * Perf Status percentage: %5.2
+ * "CTMP" 4 columns %4d
+ */
+#define NAME_LEN 12
+static int submit_counters(struct thread_data *t, struct core_data *c,
+ struct pkg_data *p)
+{
+ char name[NAME_LEN];
+ double interval_float;
+
+ interval_float = tv_delta.tv_sec + tv_delta.tv_usec/1000000.0;
+
+ snprintf(name, NAME_LEN, "cpu%02d", t->cpu_id);
+
+ if (do_nhm_cstates) {
+ if (!skip_c0)
+ turbostat_submit(name, "percent", "c0", 100.0 * t->mperf/t->tsc);
+ if (!skip_c1)
+ turbostat_submit(name, "percent", "c1", 100.0 * t->c1/t->tsc);
+ }
+
+ /* GHz */
+ if (has_aperf && ((!aperf_mperf_unstable) || (!(t->aperf > t->tsc || t->mperf > t->tsc))))
+ turbostat_submit(NULL, "frequency", name, 1.0 * t->tsc / units * t->aperf / t->mperf / interval_float);
+
+ /* SMI */
+ if (do_smi)
+ turbostat_submit(NULL, "current", name, t->smi_count);
+
+ /* print per-core data only for 1st thread in core */
+ if (!(t->flags & CPU_IS_FIRST_THREAD_IN_CORE))
+ goto done;
+
+ snprintf(name, NAME_LEN, "core%02d", c->core_id);
+
+ if (do_nhm_cstates && !do_slm_cstates)
+ turbostat_submit(name, "percent", "c3", 100.0 * c->c3/t->tsc);
+ if (do_nhm_cstates)
+ turbostat_submit(name, "percent", "c6", 100.0 * c->c6/t->tsc);
+ if (do_snb_cstates)
+ turbostat_submit(name, "percent", "c7", 100.0 * c->c7/t->tsc);
+
+ if (do_dts)
+ turbostat_submit(NULL, "temperature", name, c->core_temp_c);
+
+ /* print per-package data only for 1st core in package */
+ if (!(t->flags & CPU_IS_FIRST_CORE_IN_PACKAGE))
+ goto done;
+
+ snprintf(name, NAME_LEN, "pc%02d", p->package_id);
+
+ if (do_ptm)
+ turbostat_submit(NULL, "temperature", name, p->pkg_temp_c);
+
+ if (do_snb_cstates)
+ turbostat_submit(name, "percent", "pc2", 100.0 * p->pc2/t->tsc);
+ if (do_nhm_cstates && !do_slm_cstates)
+ turbostat_submit(name, "percent", "pc3", 100.0 * p->pc3/t->tsc);
+ if (do_nhm_cstates && !do_slm_cstates)
+ turbostat_submit(name, "percent", "pc6", 100.0 * p->pc6/t->tsc);
+ if (do_snb_cstates)
+ turbostat_submit(name, "percent", "pc7", 100.0 * p->pc7/t->tsc);
+ if (do_c8_c9_c10) {
+ turbostat_submit(name, "percent", "pc8", 100.0 * p->pc8/t->tsc);
+ turbostat_submit(name, "percent", "pc9", 100.0 * p->pc9/t->tsc);
+ turbostat_submit(name, "percent", "pc10", 100.0 * p->pc10/t->tsc);
+ }
+
+ if (do_rapl) {
+ if (do_rapl & RAPL_PKG)
+ turbostat_submit(name, "power", "Pkg_W", p->energy_pkg * rapl_energy_units / interval_float);
+ if (do_rapl & RAPL_CORES)
+ turbostat_submit(name, "power", "Cor_W", p->energy_cores * rapl_energy_units / interval_float);
+ if (do_rapl & RAPL_GFX)
+ turbostat_submit(name, "power", "GFX_W", p->energy_gfx * rapl_energy_units / interval_float);
+ if (do_rapl & RAPL_DRAM)
+ turbostat_submit(name, "power", "RAM_W", p->energy_dram * rapl_energy_units / interval_float);
+ }
+done:
+ return 0;
+}
+
+static int turbostat_read (user_data_t * not_used)
+{
+ int ret;
+
+ if (!allocated) {
+ if ((ret = setup_all_buffers()) < 0)
+ return ret;
+ }
+
+ if (for_all_proc_cpus(cpu_is_not_present)) {
+ free_all_buffers();
+ if ((ret = setup_all_buffers()) < 0)
+ return ret;
+ if (for_all_proc_cpus(cpu_is_not_present))
+ return -ERR_CPU_NOT_PRESENT;
+ }
+
+ if (!initialized) {
+ if ((ret = for_all_cpus(get_counters, EVEN_COUNTERS)) < 0)
+ return ret;
+ gettimeofday(&tv_even, (struct timezone *)NULL);
+ is_even = true;
+ initialized = true;
+ return 0;
+ }
+
+ if (is_even) {
+ if ((ret = for_all_cpus(get_counters, ODD_COUNTERS)) < 0)
+ return ret;
+ gettimeofday(&tv_odd, (struct timezone *)NULL);
+ is_even = false;
+ timersub(&tv_odd, &tv_even, &tv_delta);
+ if ((ret = for_all_cpus_2(delta_cpu, ODD_COUNTERS, EVEN_COUNTERS)) < 0)
+ return ret;
+ if ((ret = for_all_cpus(submit_counters, EVEN_COUNTERS)) < 0)
+ return ret;
+ } else {
+ if ((ret = for_all_cpus(get_counters, EVEN_COUNTERS)) < 0)
+ return ret;
+ gettimeofday(&tv_even, (struct timezone *)NULL);
+ is_even = true;
+ timersub(&tv_even, &tv_odd, &tv_delta);
+ if ((ret = for_all_cpus_2(delta_cpu, EVEN_COUNTERS, ODD_COUNTERS)) < 0)
+ return ret;
+ if ((ret = for_all_cpus(submit_counters, ODD_COUNTERS)) < 0)
+ return ret;
+ }
+ return 0;
+}
+
+STATIC_MUST_CHECK(static int check_dev_msr())
+{
+ struct stat sb;
+
+ if (stat("/dev/cpu/0/msr", &sb)) {
+ ERROR("no /dev/cpu/0/msr\n"
+ "Try \"# modprobe msr\"");
+ return -ERR_NO_MSR;
+ }
+ return 0;
+}
+
+STATIC_MUST_CHECK(static int check_super_user())
+{
+ if (getuid() != 0) {
+ ERROR("must be root");
+ return -ERR_NOT_ROOT;
+ }
+ return 0;
+}
+
+
+#define RAPL_POWER_GRANULARITY 0x7FFF /* 15 bit power granularity */
+#define RAPL_TIME_GRANULARITY 0x3F /* 6 bit time granularity */
+
+static double get_tdp(unsigned int model)
+{
+ unsigned long long msr;
+
+ if (do_rapl & RAPL_PKG_POWER_INFO)
+ if (!get_msr(0, MSR_PKG_POWER_INFO, &msr))
+ return ((msr >> 0) & RAPL_POWER_GRANULARITY) * rapl_power_units;
+
+ switch (model) {
+ case 0x37:
+ case 0x4D:
+ return 30.0;
+ default:
+ return 135.0;
+ }
+}
+
+
+/*
+ * rapl_probe()
+ *
+ * sets do_rapl, rapl_power_units, rapl_energy_units, rapl_time_units
+ */
+static void rapl_probe(unsigned int family, unsigned int model)
+{
+ unsigned long long msr;
+ unsigned int time_unit;
+ double tdp;
+
+ if (!genuine_intel)
+ return;
+
+ if (family != 6)
+ return;
+
+ switch (model) {
+ case 0x2A:
+ case 0x3A:
+ case 0x3C: /* HSW */
+ case 0x45: /* HSW */
+ case 0x46: /* HSW */
+ do_rapl = RAPL_PKG | RAPL_CORES | RAPL_CORE_POLICY | RAPL_GFX | RAPL_PKG_POWER_INFO;
+ break;
+ case 0x3F: /* HSX */
+ do_rapl = RAPL_PKG | RAPL_DRAM | RAPL_DRAM_PERF_STATUS | RAPL_PKG_PERF_STATUS | RAPL_PKG_POWER_INFO;
+ break;
+ case 0x2D:
+ case 0x3E:
+ do_rapl = RAPL_PKG | RAPL_CORES | RAPL_CORE_POLICY | RAPL_DRAM | RAPL_PKG_PERF_STATUS | RAPL_DRAM_PERF_STATUS | RAPL_PKG_POWER_INFO;
+ break;
+ case 0x37: /* BYT */
+ case 0x4D: /* AVN */
+ do_rapl = RAPL_PKG | RAPL_CORES ;
+ break;
+ default:
+ return;
+ }
+
+ /* units on package 0, verify later other packages match */
+ if (get_msr(0, MSR_RAPL_POWER_UNIT, &msr))
+ return;
+
+ rapl_power_units = 1.0 / (1 << (msr & 0xF));
+ if (model == 0x37)
+ rapl_energy_units = 1.0 * (1 << (msr >> 8 & 0x1F)) / 1000000;
+ else
+ rapl_energy_units = 1.0 / (1 << (msr >> 8 & 0x1F));
+
+ time_unit = msr >> 16 & 0xF;
+ if (time_unit == 0)
+ time_unit = 0xA;
+
+ rapl_time_units = 1.0 / (1 << (time_unit));
+
+ tdp = get_tdp(model);
+
+ rapl_joule_counter_range = 0xFFFFFFFF * rapl_energy_units / tdp;
+// if (verbose)
+// fprintf(stderr, "RAPL: %.0f sec. Joule Counter Range, at %.0f Watts\n", rapl_joule_counter_range, tdp);
+
+ return;
+}
+
+static int is_snb(unsigned int family, unsigned int model)
+{
+ if (!genuine_intel)
+ return 0;
+
+ switch (model) {
+ case 0x2A:
+ case 0x2D:
+ case 0x3A: /* IVB */
+ case 0x3E: /* IVB Xeon */
+ case 0x3C: /* HSW */
+ case 0x3F: /* HSW */
+ case 0x45: /* HSW */
+ case 0x46: /* HSW */
+ return 1;
+ }
+ return 0;
+}
+
+static int has_c8_c9_c10(unsigned int family, unsigned int model)
+{
+ if (!genuine_intel)
+ return 0;
+
+ switch (model) {
+ case 0x45:
+ return 1;
+ }
+ return 0;
+}
+
+
+static int is_slm(unsigned int family, unsigned int model)
+{
+ if (!genuine_intel)
+ return 0;
+ switch (model) {
+ case 0x37: /* BYT */
+ case 0x4D: /* AVN */
+ return 1;
+ }
+ return 0;
+}
+
+/*
+ * MSR_IA32_TEMPERATURE_TARGET indicates the temperature where
+ * the Thermal Control Circuit (TCC) activates.
+ * This is usually equal to tjMax.
+ *
+ * Older processors do not have this MSR, so there we guess,
+ * but also allow cmdline over-ride with -T.
+ *
+ * Several MSR temperature values are in units of degrees-C
+ * below this value, including the Digital Thermal Sensor (DTS),
+ * Package Thermal Management Sensor (PTM), and thermal event thresholds.
+ */
+STATIC_MUST_CHECK(static int set_temperature_target(struct thread_data *t, struct core_data *c, struct pkg_data *p))
+{
+ unsigned long long msr;
+ unsigned int target_c_local;
+ int cpu;
+
+ /* tcc_activation_temp is used only for dts or ptm */
+ if (!(do_dts || do_ptm))
+ return 0;
+
+ /* this is a per-package concept */
+ if (!(t->flags & CPU_IS_FIRST_THREAD_IN_CORE) || !(t->flags & CPU_IS_FIRST_CORE_IN_PACKAGE))
+ return 0;
+
+ cpu = t->cpu_id;
+ if (cpu_migrate(cpu)) {
+ ERROR("Could not migrate to CPU %d\n", cpu);
+ return -ERR_CPU_MIGRATE;
+ }
+
+ if (tcc_activation_temp_override != 0) {
+ tcc_activation_temp = tcc_activation_temp_override;
+ ERROR("cpu%d: Using cmdline TCC Target (%d C)\n",
+ cpu, tcc_activation_temp);
+ return 0;
+ }
+
+ /* Temperature Target MSR is Nehalem and newer only */
+ if (!do_nehalem_platform_info)
+ goto guess;
+
+ if (get_msr(0, MSR_IA32_TEMPERATURE_TARGET, &msr))
+ goto guess;
+
+ target_c_local = (msr >> 16) & 0x7F;
+
+ if (target_c_local < 85 || target_c_local > 127)
+ goto guess;
+
+ tcc_activation_temp = target_c_local;
+
+ return 0;
+
+guess:
+ tcc_activation_temp = TJMAX_DEFAULT;
+ WARNING("cpu%d: Guessing tjMax %d C, Please use -T to specify\n",
+ cpu, tcc_activation_temp);
+
+ return 0;
+}
+
+STATIC_MUST_CHECK(static int check_cpuid())
+{
+ unsigned int eax, ebx, ecx, edx, max_level;
+ unsigned int fms, family, model;
+
+ eax = ebx = ecx = edx = 0;
+
+ __get_cpuid(0, &max_level, &ebx, &ecx, &edx);
+
+ if (ebx == 0x756e6547 && edx == 0x49656e69 && ecx == 0x6c65746e)
+ genuine_intel = 1;
+
+ fms = 0;
+ __get_cpuid(1, &fms, &ebx, &ecx, &edx);
+ family = (fms >> 8) & 0xf;
+ model = (fms >> 4) & 0xf;
+ if (family == 6 || family == 0xf)
+ model += ((fms >> 16) & 0xf) << 4;
+
+ if (!(edx & (1 << 5))) {
+ ERROR("CPUID: no MSR");
+ return -ERR_NO_MSR;
+ }
+
+ /*
+ * check max extended function levels of CPUID.
+ * This is needed to check for invariant TSC.
+ * This check is valid for both Intel and AMD.
+ */
+ ebx = ecx = edx = 0;
+ __get_cpuid(0x80000000, &max_level, &ebx, &ecx, &edx);
+
+ if (max_level < 0x80000007) {
+ ERROR("CPUID: no invariant TSC (max_level 0x%x)", max_level);
+ return -ERR_NO_INVARIANT_TSC;
+ }
+
+ /*
+ * Non-Stop TSC is advertised by CPUID.EAX=0x80000007: EDX.bit8
+ * this check is valid for both Intel and AMD
+ */
+ __get_cpuid(0x80000007, &eax, &ebx, &ecx, &edx);
+ has_invariant_tsc = edx & (1 << 8);
+
+ if (!has_invariant_tsc) {
+ ERROR("No invariant TSC");
+ return -ERR_NO_INVARIANT_TSC;
+ }
+
+ /*
+ * APERF/MPERF is advertised by CPUID.EAX=0x6: ECX.bit0
+ * this check is valid for both Intel and AMD
+ */
+
+ __get_cpuid(0x6, &eax, &ebx, &ecx, &edx);
+ has_aperf = ecx & (1 << 0);
+ do_dts = eax & (1 << 0);
+ do_ptm = eax & (1 << 6);
+ has_epb = ecx & (1 << 3);
+
+ if (!has_aperf) {
+ ERROR("No APERF");
+ return -ERR_NO_APERF;
+ }
+
+ do_nehalem_platform_info = genuine_intel && has_invariant_tsc;
+ do_nhm_cstates = genuine_intel; /* all Intel w/ non-stop TSC have NHM counters */
+ do_smi = do_nhm_cstates;
+ do_snb_cstates = is_snb(family, model);
+ do_c8_c9_c10 = has_c8_c9_c10(family, model);
+ do_slm_cstates = is_slm(family, model);
+
+ rapl_probe(family, model);
+
+ return 0;
+}
+
+
+
+STATIC_MUST_CHECK(static int topology_probe())
+{
+ int i;
+ int ret;
+ int max_core_id = 0;
+ int max_package_id = 0;
+ int max_siblings = 0;
+ struct cpu_topology {
+ int core_id;
+ int physical_package_id;
+ } *cpus;
+
+ /* Initialize num_cpus, max_cpu_num */
+ topo.num_cpus = 0;
+ topo.max_cpu_num = 0;
+ ret = for_all_proc_cpus(count_cpus);
+ if (ret < 0)
+ return ret;
+ if (topo.num_cpus > 1)
+ show_cpu = 1;
+
+ DEBUG("num_cpus %d max_cpu_num %d\n", topo.num_cpus, topo.max_cpu_num);
+
+ cpus = calloc(1, (topo.max_cpu_num + 1) * sizeof(struct cpu_topology));
+ if (cpus == NULL) {
+ ERROR("calloc cpus");
+ return -ERR_CALLOC;
+ }
+
+ /*
+ * Allocate and initialize cpu_present_set
+ */
+ cpu_present_set = CPU_ALLOC((topo.max_cpu_num + 1));
+ if (cpu_present_set == NULL) {
+ free(cpus);
+ ERROR("CPU_ALLOC");
+ return -ERR_CPU_ALLOC;
+ }
+ cpu_present_setsize = CPU_ALLOC_SIZE((topo.max_cpu_num + 1));
+ CPU_ZERO_S(cpu_present_setsize, cpu_present_set);
+ ret = for_all_proc_cpus(mark_cpu_present);
+ if (ret < 0) {
+ free(cpus);
+ return ret;
+ }
+
+ /*
+ * Allocate and initialize cpu_affinity_set
+ */
+ cpu_affinity_set = CPU_ALLOC((topo.max_cpu_num + 1));
+ if (cpu_affinity_set == NULL) {
+ free(cpus);
+ ERROR("CPU_ALLOC");
+ return -ERR_CPU_ALLOC;
+ }
+ cpu_affinity_setsize = CPU_ALLOC_SIZE((topo.max_cpu_num + 1));
+ CPU_ZERO_S(cpu_affinity_setsize, cpu_affinity_set);
+
+
+ /*
+ * For online cpus
+ * find max_core_id, max_package_id
+ */
+ for (i = 0; i <= topo.max_cpu_num; ++i) {
+ int siblings;
+
+ if (cpu_is_not_present(i)) {
+ //if (verbose > 1)
+ fprintf(stderr, "cpu%d NOT PRESENT\n", i);
+ continue;
+ }
+ cpus[i].core_id = get_core_id(i);
+ if (cpus[i].core_id < 0)
+ return cpus[i].core_id;
+ if (cpus[i].core_id > max_core_id)
+ max_core_id = cpus[i].core_id;
+
+ cpus[i].physical_package_id = get_physical_package_id(i);
+ if (cpus[i].physical_package_id < 0)
+ return cpus[i].physical_package_id;
+ if (cpus[i].physical_package_id > max_package_id)
+ max_package_id = cpus[i].physical_package_id;
+
+ siblings = get_num_ht_siblings(i);
+ if (siblings < 0)
+ return siblings;
+ if (siblings > max_siblings)
+ max_siblings = siblings;
+ DEBUG("cpu %d pkg %d core %d\n",
+ i, cpus[i].physical_package_id, cpus[i].core_id);
+ }
+ topo.num_cores_per_pkg = max_core_id + 1;
+ DEBUG("max_core_id %d, sizing for %d cores per package\n",
+ max_core_id, topo.num_cores_per_pkg);
+ if (topo.num_cores_per_pkg > 1)
+ show_core = 1;
+
+ topo.num_packages = max_package_id + 1;
+ DEBUG("max_package_id %d, sizing for %d packages\n",
+ max_package_id, topo.num_packages);
+ if (topo.num_packages > 1)
+ show_pkg = 1;
+
+ topo.num_threads_per_core = max_siblings;
+ DEBUG("max_siblings %d\n", max_siblings);
+
+ free(cpus);
+ return 0;
+}
+
+static int
+allocate_counters(struct thread_data **t, struct core_data **c, struct pkg_data **p)
+{
+ int i;
+
+ *t = calloc(topo.num_threads_per_core * topo.num_cores_per_pkg *
+ topo.num_packages, sizeof(struct thread_data));
+ if (*t == NULL)
+ goto error;
+
+ for (i = 0; i < topo.num_threads_per_core *
+ topo.num_cores_per_pkg * topo.num_packages; i++)
+ (*t)[i].cpu_id = -1;
+
+ *c = calloc(topo.num_cores_per_pkg * topo.num_packages,
+ sizeof(struct core_data));
+ if (*c == NULL)
+ goto error;
+
+ for (i = 0; i < topo.num_cores_per_pkg * topo.num_packages; i++)
+ (*c)[i].core_id = -1;
+
+ *p = calloc(topo.num_packages, sizeof(struct pkg_data));
+ if (*p == NULL)
+ goto error;
+
+ for (i = 0; i < topo.num_packages; i++)
+ (*p)[i].package_id = i;
+
+ return 0;
+error:
+ ERROR("calloc counters");
+ return -ERR_CALLOC;
+}
+/*
+ * init_counter()
+ *
+ * set cpu_id, core_num, pkg_num
+ * set FIRST_THREAD_IN_CORE and FIRST_CORE_IN_PACKAGE
+ *
+ * increment topo.num_cores when 1st core in pkg seen
+ */
+static int init_counter(struct thread_data *thread_base, struct core_data *core_base,
+ struct pkg_data *pkg_base, int thread_num, int core_num,
+ int pkg_num, int cpu_id)
+{
+ int ret;
+ struct thread_data *t;
+ struct core_data *c;
+ struct pkg_data *p;
+
+ t = GET_THREAD(thread_base, thread_num, core_num, pkg_num);
+ c = GET_CORE(core_base, core_num, pkg_num);
+ p = GET_PKG(pkg_base, pkg_num);
+
+ t->cpu_id = cpu_id;
+ if (thread_num == 0) {
+ t->flags |= CPU_IS_FIRST_THREAD_IN_CORE;
+ if ((ret = cpu_is_first_core_in_package(cpu_id)) < 0) {
+ return ret;
+ } else if (ret != 0) {
+ t->flags |= CPU_IS_FIRST_CORE_IN_PACKAGE;
+ }
+ }
+
+ c->core_id = core_num;
+ p->package_id = pkg_num;
+
+ return 0;
+}
+
+
+static int initialize_counters(int cpu_id)
+{
+ int my_thread_id, my_core_id, my_package_id;
+ int ret;
+
+ my_package_id = get_physical_package_id(cpu_id);
+ if (my_package_id < 0)
+ return my_package_id;
+ my_core_id = get_core_id(cpu_id);
+ if (my_core_id < 0)
+ return my_core_id;
+
+ if ((ret = cpu_is_first_sibling_in_core(cpu_id)) < 0) {
+ return ret;
+ } else if (ret != 0) {
+ my_thread_id = 0;
+ topo.num_cores++;
+ } else {
+ my_thread_id = 1;
+ }
+
+ ret = init_counter(EVEN_COUNTERS, my_thread_id, my_core_id, my_package_id, cpu_id);
+ if (ret < 0)
+ return ret;
+ ret = init_counter(ODD_COUNTERS, my_thread_id, my_core_id, my_package_id, cpu_id);
+ if (ret < 0)
+ return ret;
+ return 0;
+}
+
+#define DO_OR_GOTO_ERR(something) \
+do { \
+ ret = something; \
+ if (ret < 0) \
+ goto err; \
+} while (0);
+
+static int setup_all_buffers(void)
+{
+ int ret;
+
+ DO_OR_GOTO_ERR(topology_probe())
+ DO_OR_GOTO_ERR(allocate_counters(&thread_even, &core_even, &package_even))
+ DO_OR_GOTO_ERR(allocate_counters(&thread_odd, &core_odd, &package_odd))
+ DO_OR_GOTO_ERR(for_all_proc_cpus(initialize_counters))
+
+ allocated = true;
+ return 0;
+err:
+ free_all_buffers();
+ return ret;
+}
+
+static int turbostat_init(void)
+{
+ int ret;
+ struct timespec ts;
+
+ DO_OR_GOTO_ERR(check_cpuid())
+ DO_OR_GOTO_ERR(check_dev_msr())
+ DO_OR_GOTO_ERR(check_super_user())
+ DO_OR_GOTO_ERR(setup_all_buffers())
+ DO_OR_GOTO_ERR(for_all_cpus(set_temperature_target, EVEN_COUNTERS))
+
+ ts.tv_sec = interval_sec;
+ ts.tv_nsec = 0;
+
+ plugin_register_complex_read(NULL, PLUGIN_NAME, turbostat_read, &ts, NULL);
+
+ return 0;
+err:
+ free_all_buffers();
+ return ret;
+}
+
+static const char *config_keys[] =
+{
+ "Interval",
+};
+static int config_keys_num = STATIC_ARRAY_SIZE (config_keys);
+
+static int turbostat_config (const char *key, const char *value)
+{
+ if (strcasecmp("Interval", key) == 0)
+ interval_sec = atoi(value);
+ else
+ return -1;
+ return 0;
+}
+
+void module_register (void);
+void module_register (void)
+{
+ plugin_register_init(PLUGIN_NAME, turbostat_init);
+ plugin_register_config(PLUGIN_NAME, turbostat_config, config_keys, config_keys_num);
+}
projects / collectd.git / commitdiff