tools/power/x86/turbostat/turbostat.8

   1 .TH TURBOSTAT 8
   2 .SH NAME
   3 turbostat \- Report processor frequency and idle statistics
   4 .SH SYNOPSIS
   5 .ft B
   6 .B turbostat
   7 .RB [ Options ]
   8 .RB command
   9 .br
  10 .B turbostat
  11 .RB [ Options ]
  12 .RB [ "\--interval seconds" ]
  13 .SH DESCRIPTION
  14 \fBturbostat \fP reports processor topology, frequency,
  15 idle power-state statistics, temperature and power on X86 processors.
  16 There are two ways to invoke turbostat.
  17 The first method is to supply a
  18 \fBcommand\fP, which is forked and statistics are printed
  19 in one-shot upon its completion.
  20 The second method is to omit the command,
  21 and turbostat displays statistics every 5 seconds interval.
  22 The 5-second interval can be changed using the --interval option.
  23 .PP
  24 Some information is not available on older processors.
  25 .SS Options
  26 Options can be specified with a single or double '-', and only as much of the option
  27 name as necessary to disambiguate it from others is necessary.  Note that options are case-sensitive.
  28 .PP
  29 \fB--add attributes\fP add column with counter having specified 'attributes'.  The 'location' attribute is required, all others are optional.
  30 .nf
  31         location: {\fBmsrDDD\fP | \fBmsr0xXXX\fP | \fB/sys/path...\fP}
  32                 msrDDD is a decimal offset, eg. msr16
  33                 msr0xXXX is a hex offset, eg. msr0x10
  34                 /sys/path... is an absolute path to a sysfs attribute
  35
  36         scope: {\fBcpu\fP | \fBcore\fP | \fBpackage\fP}
  37                 sample and print the counter for every cpu, core, or package.
  38                 default: cpu
  39
  40         size: {\fBu32\fP | \fBu64\fP }
  41                 MSRs are read as 64-bits, u32 truncates the displayed value to 32-bits.
  42                 default: u64
  43
  44         format: {\fBraw\fP | \fBdelta\fP | \fBpercent\fP}
  45                 'raw' shows the MSR contents in hex.
  46                 'delta' shows the difference in values during the measurement interval.
  47                 'percent' shows the delta as a percentage of the cycles elapsed.
  48                 default: delta
  49
  50         name: "name_string"
  51                 Any string that does not match a key-word above is used
  52                 as the column header.
  53 .fi
  54 .PP
  55 \fB--cpu cpu-set\fP limit output to system summary plus the specified cpu-set.  If cpu-set is the string "core", then the system summary plus the first CPU in each core are printed -- eg. subsequent HT siblings are not printed.  Or if cpu-set is the string "package", then the system summary plus the first CPU in each package is printed.  Otherwise, the system summary plus the specified set of CPUs are printed.  The cpu-set is ordered from low to high, comma delimited with ".." and "-" permitted to denote a range. eg. 1,2,8,14..17,21-44
  56 .PP
  57 \fB--hide column\fP do not show the specified columns.  May be invoked multiple times, or with a comma-separated list of column names.  Use "--hide sysfs" to hide the sysfs statistics columns as a group.
  58 .PP
  59 \fB--show column\fP show only the specified columns.  May be invoked multiple times, or with a comma-separated list of column names.  Use "--show sysfs" to show the sysfs statistics columns as a group.
  60 .PP
  61 \fB--Dump\fP displays the raw counter values.
  62 .PP
  63 \fB--quiet\fP Do not decode and print the system configuration header information.
  64 .PP
  65 \fB--interval seconds\fP overrides the default 5.0 second measurement interval.
  66 .PP
  67 \fB--out output_file\fP turbostat output is written to the specified output_file.
  68 The file is truncated if it already exists, and it is created if it does not exist.
  69 .PP
  70 \fB--help\fP displays usage for the most common parameters.
  71 .PP
  72 \fB--Joules\fP displays energy in Joules, rather than dividing Joules by time to print power in Watts.
  73 .PP
  74 \fB--list\fP display column header names available for use by --show and --hide, then exit.
  75 .PP
  76 \fB--Summary\fP limits output to a 1-line System Summary for each interval.
  77 .PP
  78 \fB--TCC temperature\fP sets the Thermal Control Circuit temperature for systems which do not export that value.  This is used for making sense of the Digital Thermal Sensor outputs, as they return degrees Celsius below the TCC activation temperature.
  79 .PP
  80 \fB--version\fP displays the version.
  81 .PP
  82 The \fBcommand\fP parameter forks \fBcommand\fP, and upon its exit,
  83 displays the statistics gathered since it was forked.
  84 .PP
  85 .SH ROW DESCRIPTIONS
  86 The system configuration dump (if --quiet is not used) is followed by statistics.  The first row of the statistics labels the content of each column (below).  The second row of statistics is the system summary line.  The system summary line has a '-' in the columns for the Package, Core, and CPU.  The contents of the system summary line depends on the type of column.  Columns that count items (eg. IRQ) show the sum across all CPUs in the system.  Columns that show a percentage show the average across all CPUs in the system.  Columns that dump raw MSR values simply show 0 in the summary.  After the system summary row, each row describes a specific Package/Core/CPU.  Note that if the --cpu parameter is used to limit which specific CPUs are displayed, turbostat will still collect statistics for all CPUs in the system and will still show the system summary for all CPUs in the system.
  87 .SH COLUMN DESCRIPTIONS
  88 .nf
  89 \fBCore\fP processor core number.  Note that multiple CPUs per core indicate support for Intel(R) Hyper-Threading Technology (HT).
  90 \fBCPU\fP Linux CPU (logical processor) number.  Yes, it is okay that on many systems the CPUs are not listed in numerical order -- for efficiency reasons, turbostat runs in topology order, so HT siblings appear together.
  91 \fBPackage\fP processor package number -- not present on systems with a single processor package.
  92 \fBAvg_MHz\fP number of cycles executed divided by time elapsed.  Note that this includes idle-time when 0 instructions are executed.
  93 \fBBusy%\fP percent of the measurement interval that the CPU executes instructions, aka. % of time in "C0" state.
  94 \fBBzy_MHz\fP average clock rate while the CPU was not idle (ie. in "c0" state).
  95 \fBTSC_MHz\fP average MHz that the TSC ran during the entire interval.
  96 \fBIRQ\fP The number of interrupts serviced by that CPU during the measurement interval.  The system total line is the sum of interrupts serviced across all CPUs.  turbostat parses /proc/interrupts to generate this summary.
  97 \fBSMI\fP The number of System Management Interrupts  serviced CPU during the measurement interval.  While this counter is actually per-CPU, SMI are triggered on all processors, so the number should be the same for all CPUs.
  98 \fBC1, C2, C3...\fP The number times Linux requested the C1, C2, C3 idle state during the measurement interval.  The system summary line shows the sum for all CPUs.  These are C-state names as exported in /sys/devices/system/cpu/cpu*/cpuidle/state*/name.  While their names are generic, their attributes are processor specific. They the system description section of output shows what MWAIT sub-states they are mapped to on each system.
  99 \fBC1%, C2%, C3%\fP The residency percentage that Linux requested C1, C2, C3....  The system summary is the average of all CPUs in the system.  Note that these are software, reflecting what was requested.  The hardware counters reflect what was actually achieved.
 100 \fBCPU%c1, CPU%c3, CPU%c6, CPU%c7\fP show the percentage residency in hardware core idle states.  These numbers are from hardware residency counters.
 101 \fBCoreTmp\fP Degrees Celsius reported by the per-core Digital Thermal Sensor.
 102 \fBPkgTtmp\fP Degrees Celsius reported by the per-package Package Thermal Monitor.
 103 \fBGFX%rc6\fP The percentage of time the GPU is in the "render C6" state, rc6, during the measurement interval. From /sys/class/drm/card0/power/rc6_residency_ms.
 104 \fBGFXMHz\fP Instantaneous snapshot of what sysfs presents at the end of the measurement interval. From /sys/class/graphics/fb0/device/drm/card0/gt_cur_freq_mhz.
 105 \fBPkg%pc2, Pkg%pc3, Pkg%pc6, Pkg%pc7\fP percentage residency in hardware package idle states.  These numbers are from hardware residency counters.
 106 \fBPkgWatt\fP Watts consumed by the whole package.
 107 \fBCorWatt\fP Watts consumed by the core part of the package.
 108 \fBGFXWatt\fP Watts consumed by the Graphics part of the package -- available only on client processors.
 109 \fBRAMWatt\fP Watts consumed by the DRAM DIMMS -- available only on server processors.
 110 \fBPKG_%\fP percent of the interval that RAPL throttling was active on the Package.
 111 \fBRAM_%\fP percent of the interval that RAPL throttling was active on DRAM.
 112 .fi
 113 .SH TOO MUCH INFORMATION EXAMPLE
 114 By default, turbostat dumps all possible information -- a system configuration header, followed by columns for all counters.
 115 This is ideal for remote debugging, use the "--out" option to save everything to a text file, and get that file to the expert helping you debug.
 116 .PP
 117 When you are not interested in all that information, and there are several ways to see only what you want.  First the "--quiet" option will skip the configuration information, and turbostat will show only the counter columns.  Second, you can reduce the columns with the "--hide" and "--show" options.  If you use the "--show" option, then turbostat will show only the columns you list.  If you use the "--hide" option, turbostat will show all columns, except the ones you list.
 118 .PP
 119 To find out what columns are available for --show and --hide, the "--list" option is available.  For convenience, the special strings "sysfs" can be used to refer to all of the sysfs C-state counters at once:
 120 .nf
 121 sudo ./turbostat --show sysfs --quiet sleep 10
 122 10.003837 sec
 123         C1      C1E     C3      C6      C7s     C1%     C1E%    C3%     C6%     C7s%
 124         4       21      2       2       459     0.14    0.82    0.00    0.00    98.93
 125         1       17      2       2       130     0.00    0.02    0.00    0.00    99.80
 126         0       0       0       0       31      0.00    0.00    0.00    0.00    99.95
 127         2       1       0       0       52      1.14    6.49    0.00    0.00    92.21
 128         1       2       0       0       52      0.00    0.08    0.00    0.00    99.86
 129         0       0       0       0       71      0.00    0.00    0.00    0.00    99.89
 130         0       0       0       0       25      0.00    0.00    0.00    0.00    99.96
 131         0       0       0       0       74      0.00    0.00    0.00    0.00    99.94
 132         0       1       0       0       24      0.00    0.00    0.00    0.00    99.84
 133 .fi
 134 .PP
 135 .SH ONE SHOT COMMAND EXAMPLE
 136 If turbostat is invoked with a command, it will fork that command
 137 and output the statistics gathered after the command exits.
 138 In this case, turbostat output goes to stderr, by default.
 139 Output can instead be saved to a file using the --out option.
 140 In this example, the "sleep 10" command is forked, and turbostat waits for it to complete before saving all statistics into "ts.out".  Note that "sleep 10" is not part of turbostat, but is simply an example of a command that turbostat can fork.  The "ts.out" file is what you want to edit in a very wide window, paste into a spreadsheet, or attach to a bugzilla entry.
 141
 142 .nf
 143 [root@hsw]# ./turbostat -o ts.out sleep 10
 144 [root@hsw]#
 145 .fi
 146
 147 .SH PERIODIC INTERVAL EXAMPLE
 148 Without a command to fork, turbostat displays statistics ever 5 seconds.
 149 Periodic output goes to stdout, by default, unless --out is used to specify an output file.
 150 The 5-second interval can be changed with the "-i sec" option.
 151 .nf
 152 sudo ./turbostat --quiet --hide sysfs,IRQ,SMI,CoreTmp,PkgTmp,GFX%rc6,GFXMHz,PkgWatt,CorWatt,GFXWatt
 153         Core    CPU     Avg_MHz Busy%   Bzy_MHz TSC_MHz CPU%c1  CPU%c3  CPU%c6  CPU%c7
 154         -       -       488     12.52   3900    3498    12.50   0.00    0.00    74.98
 155         0       0       5       0.13    3900    3498    99.87   0.00    0.00    0.00
 156         0       4       3897    99.99   3900    3498    0.01
 157         1       1       0       0.00    3856    3498    0.01    0.00    0.00    99.98
 158         1       5       0       0.00    3861    3498    0.01
 159         2       2       1       0.02    3889    3498    0.03    0.00    0.00    99.95
 160         2       6       0       0.00    3863    3498    0.05
 161         3       3       0       0.01    3869    3498    0.02    0.00    0.00    99.97
 162         3       7       0       0.00    3878    3498    0.03
 163         Core    CPU     Avg_MHz Busy%   Bzy_MHz TSC_MHz CPU%c1  CPU%c3  CPU%c6  CPU%c7
 164         -       -       491     12.59   3900    3498    12.42   0.00    0.00    74.99
 165         0       0       27      0.69    3900    3498    99.31   0.00    0.00    0.00
 166         0       4       3898    99.99   3900    3498    0.01
 167         1       1       0       0.00    3883    3498    0.01    0.00    0.00    99.99
 168         1       5       0       0.00    3898    3498    0.01
 169         2       2       0       0.01    3889    3498    0.02    0.00    0.00    99.98
 170         2       6       0       0.00    3889    3498    0.02
 171         3       3       0       0.00    3856    3498    0.01    0.00    0.00    99.99
 172         3       7       0       0.00    3897    3498    0.01
 173 .fi
 174 This example also shows the use of the --hide option to skip columns that are not wanted.
 175 Note that cpu4 in this example is 99.99% busy, while the other CPUs are all under 1% busy.
 176 Notice that cpu4's HT sibling is cpu0, which is under 1% busy, but can get into CPU%c1 only,
 177 because its cpu4's activity on shared hardware keeps it from entering a deeper C-state.
 178
 179 .SH SYSTEM CONFIGURATION INFORMATION EXAMPLE
 180
 181 By default, turbostat always dumps system configuration information
 182 before taking measurements.  In the example above, "--quiet" is used
 183 to suppress that output.  Here is an example of the configuration information:
 184 .nf
 185 turbostat version 2017.02.15 - Len Brown <lenb@kernel.org>
 186 CPUID(0): GenuineIntel 13 CPUID levels; family:model:stepping 0x6:3c:3 (6:60:3)
 187 CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM TM
 188 CPUID(6): APERF, TURBO, DTS, PTM, No-HWP, No-HWPnotify, No-HWPwindow, No-HWPepp, No-HWPpkg, EPB
 189 cpu4: MSR_IA32_MISC_ENABLE: 0x00850089 (TCC EIST No-MWAIT PREFETCH TURBO)
 190 CPUID(7): No-SGX
 191 cpu4: MSR_MISC_PWR_MGMT: 0x00400000 (ENable-EIST_Coordination DISable-EPB DISable-OOB)
 192 RAPL: 3121 sec. Joule Counter Range, at 84 Watts
 193 cpu4: MSR_PLATFORM_INFO: 0x80838f3012300
 194 8 * 100.0 = 800.0 MHz max efficiency frequency
 195 35 * 100.0 = 3500.0 MHz base frequency
 196 cpu4: MSR_IA32_POWER_CTL: 0x0004005d (C1E auto-promotion: DISabled)
 197 cpu4: MSR_TURBO_RATIO_LIMIT: 0x25262727
 198 37 * 100.0 = 3700.0 MHz max turbo 4 active cores
 199 38 * 100.0 = 3800.0 MHz max turbo 3 active cores
 200 39 * 100.0 = 3900.0 MHz max turbo 2 active cores
 201 39 * 100.0 = 3900.0 MHz max turbo 1 active cores
 202 cpu4: MSR_CONFIG_TDP_NOMINAL: 0x00000023 (base_ratio=35)
 203 cpu4: MSR_CONFIG_TDP_LEVEL_1: 0x00000000 ()
 204 cpu4: MSR_CONFIG_TDP_LEVEL_2: 0x00000000 ()
 205 cpu4: MSR_CONFIG_TDP_CONTROL: 0x80000000 ( lock=1)
 206 cpu4: MSR_TURBO_ACTIVATION_RATIO: 0x00000000 (MAX_NON_TURBO_RATIO=0 lock=0)
 207 cpu4: MSR_PKG_CST_CONFIG_CONTROL: 0x1e000400 (UNdemote-C3, UNdemote-C1, demote-C3, demote-C1, UNlocked: pkg-cstate-limit=0: pc0)
 208 cpu4: POLL: CPUIDLE CORE POLL IDLE
 209 cpu4: C1: MWAIT 0x00
 210 cpu4: C1E: MWAIT 0x01
 211 cpu4: C3: MWAIT 0x10
 212 cpu4: C6: MWAIT 0x20
 213 cpu4: C7s: MWAIT 0x32
 214 cpu4: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch)
 215 cpu0: MSR_IA32_ENERGY_PERF_BIAS: 0x00000006 (balanced)
 216 cpu0: MSR_CORE_PERF_LIMIT_REASONS, 0x31200000 (Active: ) (Logged: Transitions, MultiCoreTurbo, Amps, Auto-HWP, )
 217 cpu0: MSR_GFX_PERF_LIMIT_REASONS, 0x00000000 (Active: ) (Logged: )
 218 cpu0: MSR_RING_PERF_LIMIT_REASONS, 0x0d000000 (Active: ) (Logged: Amps, PkgPwrL1, PkgPwrL2, )
 219 cpu0: MSR_RAPL_POWER_UNIT: 0x000a0e03 (0.125000 Watts, 0.000061 Joules, 0.000977 sec.)
 220 cpu0: MSR_PKG_POWER_INFO: 0x000002a0 (84 W TDP, RAPL 0 - 0 W, 0.000000 sec.)
 221 cpu0: MSR_PKG_POWER_LIMIT: 0x428348001a82a0 (UNlocked)
 222 cpu0: PKG Limit #1: ENabled (84.000000 Watts, 8.000000 sec, clamp DISabled)
 223 cpu0: PKG Limit #2: ENabled (105.000000 Watts, 0.002441* sec, clamp DISabled)
 224 cpu0: MSR_PP0_POLICY: 0
 225 cpu0: MSR_PP0_POWER_LIMIT: 0x00000000 (UNlocked)
 226 cpu0: Cores Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled)
 227 cpu0: MSR_PP1_POLICY: 0
 228 cpu0: MSR_PP1_POWER_LIMIT: 0x00000000 (UNlocked)
 229 cpu0: GFX Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled)
 230 cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x00641400 (100 C)
 231 cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x884c0800 (24 C)
 232 cpu0: MSR_IA32_THERM_STATUS: 0x884c0000 (24 C +/- 1)
 233 cpu1: MSR_IA32_THERM_STATUS: 0x88510000 (19 C +/- 1)
 234 cpu2: MSR_IA32_THERM_STATUS: 0x884e0000 (22 C +/- 1)
 235 cpu3: MSR_IA32_THERM_STATUS: 0x88510000 (19 C +/- 1)
 236 cpu4: MSR_PKGC3_IRTL: 0x00008842 (valid, 67584 ns)
 237 cpu4: MSR_PKGC6_IRTL: 0x00008873 (valid, 117760 ns)
 238 cpu4: MSR_PKGC7_IRTL: 0x00008891 (valid, 148480 ns)
 239 .fi
 240 The \fBmax efficiency\fP frequency, a.k.a. Low Frequency Mode, is the frequency
 241 available at the minimum package voltage.  The \fBTSC frequency\fP is the base
 242 frequency of the processor -- this should match the brand string
 243 in /proc/cpuinfo.  This base frequency
 244 should be sustainable on all CPUs indefinitely, given nominal power and cooling.
 245 The remaining rows show what maximum turbo frequency is possible
 246 depending on the number of idle cores.  Note that not all information is
 247 available on all processors.
 248 .SH ADD COUNTER EXAMPLE
 249 Here we limit turbostat to showing just the CPU number for cpu0 - cpu3.
 250 We add a counter showing the 32-bit raw value of MSR 0x199 (MSR_IA32_PERF_CTL),
 251 labeling it with the column header, "PRF_CTRL", and display it only once,
 252 afte the conclusion of a 0.1 second sleep.
 253 .nf
 254 sudo ./turbostat --quiet --cpu 0-3 --show CPU --add msr0x199,u32,raw,PRF_CTRL sleep .1
 255 0.101604 sec
 256 CPU       PRF_CTRL
 257 -       0x00000000
 258 0       0x00000c00
 259 1       0x00000800
 260 2       0x00000a00
 261 3       0x00000800
 262
 263 .fi
 264
 265 .SH NOTES
 266
 267 .B "turbostat "
 268 must be run as root.
 269 Alternatively, non-root users can be enabled to run turbostat this way:
 270
 271 # setcap cap_sys_rawio=ep ./turbostat
 272
 273 # chmod +r /dev/cpu/*/msr
 274
 275 .B "turbostat "
 276 reads hardware counters, but doesn't write them.
 277 So it will not interfere with the OS or other programs, including
 278 multiple invocations of itself.
 279
 280 \fBturbostat \fP
 281 may work poorly on Linux-2.6.20 through 2.6.29,
 282 as \fBacpi-cpufreq \fPperiodically cleared the APERF and MPERF MSRs
 283 in those kernels.
 284
 285 AVG_MHz = APERF_delta/measurement_interval.  This is the actual
 286 number of elapsed cycles divided by the entire sample interval --
 287 including idle time.  Note that this calculation is resilient
 288 to systems lacking a non-stop TSC.
 289
 290 TSC_MHz = TSC_delta/measurement_interval.
 291 On a system with an invariant TSC, this value will be constant
 292 and will closely match the base frequency value shown
 293 in the brand string in /proc/cpuinfo.  On a system where
 294 the TSC stops in idle, TSC_MHz will drop
 295 below the processor's base frequency.
 296
 297 Busy% = MPERF_delta/TSC_delta
 298
 299 Bzy_MHz = TSC_delta/APERF_delta/MPERF_delta/measurement_interval
 300
 301 Note that these calculations depend on TSC_delta, so they
 302 are not reliable during intervals when TSC_MHz is not running at the base frequency.
 303
 304 Turbostat data collection is not atomic.
 305 Extremely short measurement intervals (much less than 1 second),
 306 or system activity that prevents turbostat from being able
 307 to run on all CPUS to quickly collect data, will result in
 308 inconsistent results.
 309
 310 The APERF, MPERF MSRs are defined to count non-halted cycles.
 311 Although it is not guaranteed by the architecture, turbostat assumes
 312 that they count at TSC rate, which is true on all processors tested to date.
 313
 314 .SH REFERENCES
 315 Volume 3B: System Programming Guide"
 316 http://www.intel.com/products/processor/manuals/
 317
 318 .SH FILES
 319 .ta
 320 .nf
 321 /dev/cpu/*/msr
 322 .fi
 323
 324 .SH "SEE ALSO"
 325 msr(4), vmstat(8)
 326 .PP
 327 .SH AUTHOR
 328 .nf
 329 Written by Len Brown <len.brown@intel.com>