| turbostat.8 (bf61c8840efe60fd8f91446860b63338fb424158) | turbostat.8 (fc04cc67ea8f44124f048832a745a24bc2fa12fa) |
|---|---|
| 1.TH TURBOSTAT 8 2.SH NAME 3turbostat \- Report processor frequency and idle statistics 4.SH SYNOPSIS 5.ft B 6.B turbostat 7.RB [ Options ] 8.RB command --- 33 unchanged lines hidden (view full) --- 42The \fB-i interval_sec\fP option prints statistics every \fiinterval_sec\fP seconds. 43The default is 5 seconds. 44.PP 45The \fBcommand\fP parameter forks \fBcommand\fP and upon its exit, 46displays the statistics gathered since it was forked. 47.PP 48.SH FIELD DESCRIPTIONS 49.nf | 1.TH TURBOSTAT 8 2.SH NAME 3turbostat \- Report processor frequency and idle statistics 4.SH SYNOPSIS 5.ft B 6.B turbostat 7.RB [ Options ] 8.RB command --- 33 unchanged lines hidden (view full) --- 42The \fB-i interval_sec\fP option prints statistics every \fiinterval_sec\fP seconds. 43The default is 5 seconds. 44.PP 45The \fBcommand\fP parameter forks \fBcommand\fP and upon its exit, 46displays the statistics gathered since it was forked. 47.PP 48.SH FIELD DESCRIPTIONS 49.nf |
| 50\fBpk\fP processor package number. 51\fBcor\fP processor core number. | 50\fBPackage\fP processor package number. 51\fBCore\fP processor core number. |
| 52\fBCPU\fP Linux CPU (logical processor) number. 53Note that multiple CPUs per core indicate support for Intel(R) Hyper-Threading Technology. | 52\fBCPU\fP Linux CPU (logical processor) number. 53Note that multiple CPUs per core indicate support for Intel(R) Hyper-Threading Technology. |
| 54\fB%c0\fP percent of the interval that the CPU retired instructions. 55\fBGHz\fP average clock rate while the CPU was in c0 state. 56\fBTSC\fP average GHz that the TSC ran during the entire interval. 57\fB%c1, %c3, %c6, %c7\fP show the percentage residency in hardware core idle states. 58\fBCTMP\fP Degrees Celsius reported by the per-core Digital Thermal Sensor. 59\fBPTMP\fP Degrees Celsius reported by the per-package Package Thermal Monitor. 60\fB%pc2, %pc3, %pc6, %pc7\fP percentage residency in hardware package idle states. 61\fBPkg_W\fP Watts consumed by the whole package. 62\fBCor_W\fP Watts consumed by the core part of the package. 63\fBGFX_W\fP Watts consumed by the Graphics part of the package -- available only on client processors. 64\fBRAM_W\fP Watts consumed by the DRAM DIMMS -- available only on server processors. | 54\fBAVG_MHz\fP number of cycles executed divided by time elapsed. 55\fB%Buzy\fP percent of the interval that the CPU retired instructions, aka. % of time in "C0" state. 56\fBBzy_MHz\fP average clock rate while the CPU was busy (in "c0" state). 57\fBTSC_MHz\fP average MHz that the TSC ran during the entire interval. 58\fBCPU%c1, CPU%c3, CPU%c6, CPU%c7\fP show the percentage residency in hardware core idle states. 59\fBCoreTmp\fP Degrees Celsius reported by the per-core Digital Thermal Sensor. 60\fBPkgTtmp\fP Degrees Celsius reported by the per-package Package Thermal Monitor. 61\fBPkg%pc2, Pkg%pc3, Pkg%pc6, Pkg%pc7\fP percentage residency in hardware package idle states. 62\fBPkgWatt\fP Watts consumed by the whole package. 63\fBCorWatt\fP Watts consumed by the core part of the package. 64\fBGFXWatt\fP Watts consumed by the Graphics part of the package -- available only on client processors. 65\fBRAMWatt\fP Watts consumed by the DRAM DIMMS -- available only on server processors. |
| 65\fBPKG_%\fP percent of the interval that RAPL throttling was active on the Package. 66\fBRAM_%\fP percent of the interval that RAPL throttling was active on DRAM. 67.fi 68.PP 69.SH EXAMPLE 70Without any parameters, turbostat prints out counters ever 5 seconds. 71(override interval with "-i sec" option, or specify a command 72for turbostat to fork). 73 74The first row of statistics is a summary for the entire system. 75For residency % columns, the summary is a weighted average. 76For Temperature columns, the summary is the column maximum. 77For Watts columns, the summary is a system total. 78Subsequent rows show per-CPU statistics. 79 80.nf | 66\fBPKG_%\fP percent of the interval that RAPL throttling was active on the Package. 67\fBRAM_%\fP percent of the interval that RAPL throttling was active on DRAM. 68.fi 69.PP 70.SH EXAMPLE 71Without any parameters, turbostat prints out counters ever 5 seconds. 72(override interval with "-i sec" option, or specify a command 73for turbostat to fork). 74 75The first row of statistics is a summary for the entire system. 76For residency % columns, the summary is a weighted average. 77For Temperature columns, the summary is the column maximum. 78For Watts columns, the summary is a system total. 79Subsequent rows show per-CPU statistics. 80 81.nf |
| 81[root@sandy]# ./turbostat 82cor CPU %c0 GHz TSC %c1 %c3 %c6 %c7 CTMP PTMP %pc2 %pc3 %pc6 %pc7 Pkg_W Cor_W GFX_W 83 0.06 0.80 2.29 0.11 0.00 0.00 99.83 47 40 0.26 0.01 0.44 98.78 3.49 0.12 0.14 84 0 0 0.07 0.80 2.29 0.07 0.00 0.00 99.86 40 40 0.26 0.01 0.44 98.78 3.49 0.12 0.14 85 0 4 0.03 0.80 2.29 0.12 86 1 1 0.04 0.80 2.29 0.25 0.01 0.00 99.71 40 87 1 5 0.16 0.80 2.29 0.13 88 2 2 0.05 0.80 2.29 0.06 0.01 0.00 99.88 40 89 2 6 0.03 0.80 2.29 0.08 90 3 3 0.05 0.80 2.29 0.08 0.00 0.00 99.87 47 91 3 7 0.04 0.84 2.29 0.09 | 82[root@ivy]# ./turbostat 83 Core CPU Avg_MHz %Busy Bzy_MHz TSC_MHz SMI CPU%c1 CPU%c3 CPU%c6 CPU%c7 CoreTmp PkgTmp Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 PkgWatt CorWatt GFXWatt 84 - - 6 0.36 1596 3492 0 0.59 0.01 99.04 0.00 23 24 23.82 0.01 72.47 0.00 6.40 1.01 0.00 85 0 0 9 0.58 1596 3492 0 0.28 0.01 99.13 0.00 23 24 23.82 0.01 72.47 0.00 6.40 1.01 0.00 86 0 4 1 0.07 1596 3492 0 0.79 87 1 1 10 0.65 1596 3492 0 0.59 0.00 98.76 0.00 23 88 1 5 5 0.28 1596 3492 0 0.95 89 2 2 10 0.66 1596 3492 0 0.41 0.01 98.92 0.00 23 90 2 6 2 0.10 1597 3492 0 0.97 91 3 3 3 0.20 1596 3492 0 0.44 0.00 99.37 0.00 23 92 3 7 5 0.31 1596 3492 0 0.33 |
| 92.fi | 93.fi |
| 93.SH SUMMARY EXAMPLE 94The "-s" option prints the column headers just once, 95and then the one line system summary for each sample interval. 96 97.nf 98[root@wsm]# turbostat -S 99 %c0 GHz TSC %c1 %c3 %c6 CTMP %pc3 %pc6 100 1.40 2.81 3.38 10.78 43.47 44.35 42 13.67 2.09 101 1.34 2.90 3.38 11.48 58.96 28.23 41 19.89 0.15 102 1.55 2.72 3.38 26.73 37.66 34.07 42 2.53 2.80 103 1.37 2.83 3.38 16.95 60.05 21.63 42 5.76 0.20 104.fi | |
| 105.SH VERBOSE EXAMPLE 106The "-v" option adds verbosity to the output: 107 108.nf 109[root@ivy]# turbostat -v 110turbostat v3.0 November 23, 2012 - Len Brown <lenb@kernel.org> 111CPUID(0): GenuineIntel 13 CPUID levels; family:model:stepping 0x6:3a:9 (6:58:9) 112CPUID(6): APERF, DTS, PTM, EPB --- 36 unchanged lines hidden (view full) --- 149not available on all processors. 150.SH FORK EXAMPLE 151If turbostat is invoked with a command, it will fork that command 152and output the statistics gathered when the command exits. 153eg. Here a cycle soaker is run on 1 CPU (see %c0) for a few seconds 154until ^C while the other CPUs are mostly idle: 155 156.nf | 94.SH VERBOSE EXAMPLE 95The "-v" option adds verbosity to the output: 96 97.nf 98[root@ivy]# turbostat -v 99turbostat v3.0 November 23, 2012 - Len Brown <lenb@kernel.org> 100CPUID(0): GenuineIntel 13 CPUID levels; family:model:stepping 0x6:3a:9 (6:58:9) 101CPUID(6): APERF, DTS, PTM, EPB --- 36 unchanged lines hidden (view full) --- 138not available on all processors. 139.SH FORK EXAMPLE 140If turbostat is invoked with a command, it will fork that command 141and output the statistics gathered when the command exits. 142eg. Here a cycle soaker is run on 1 CPU (see %c0) for a few seconds 143until ^C while the other CPUs are mostly idle: 144 145.nf |
| 157[root@x980 lenb]# ./turbostat cat /dev/zero > /dev/null | 146root@ivy: turbostat cat /dev/zero > /dev/null |
| 158^C | 147^C |
| 159cor CPU %c0 GHz TSC %c1 %c3 %c6 %pc3 %pc6 160 8.86 3.61 3.38 15.06 31.19 44.89 0.00 0.00 161 0 0 1.46 3.22 3.38 16.84 29.48 52.22 0.00 0.00 162 0 6 0.21 3.06 3.38 18.09 163 1 2 0.53 3.33 3.38 2.80 46.40 50.27 164 1 8 0.89 3.47 3.38 2.44 165 2 4 1.36 3.43 3.38 9.04 23.71 65.89 166 2 10 0.18 2.86 3.38 10.22 167 8 1 0.04 2.87 3.38 99.96 0.01 0.00 168 8 7 99.72 3.63 3.38 0.27 169 9 3 0.31 3.21 3.38 7.64 56.55 35.50 170 9 9 0.08 2.95 3.38 7.88 171 10 5 1.42 3.43 3.38 2.14 30.99 65.44 172 10 11 0.16 2.88 3.38 3.40 | 148 Core CPU Avg_MHz %Busy Bzy_MHz TSC_MHz SMI CPU%c1 CPU%c3 CPU%c6 CPU%c7 CoreTmp PkgTmp Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 PkgWatt CorWatt GFXWatt 149 - - 496 12.75 3886 3492 0 13.16 0.04 74.04 0.00 36 36 0.00 0.00 0.00 0.00 23.15 17.65 0.00 150 0 0 22 0.57 3830 3492 0 0.83 0.02 98.59 0.00 27 36 0.00 0.00 0.00 0.00 23.15 17.65 0.00 151 0 4 9 0.24 3829 3492 0 1.15 152 1 1 4 0.09 3783 3492 0 99.91 0.00 0.00 0.00 36 153 1 5 3880 99.82 3888 3492 0 0.18 154 2 2 17 0.44 3813 3492 0 0.77 0.04 98.75 0.00 28 155 2 6 12 0.32 3823 3492 0 0.89 156 3 3 16 0.43 3844 3492 0 0.63 0.11 98.84 0.00 30 157 3 7 4 0.11 3827 3492 0 0.94 15830.372243 sec 159 |
| 173.fi | 160.fi |
| 174Above the cycle soaker drives cpu7 up its 3.6 GHz turbo limit | 161Above the cycle soaker drives cpu5 up its 3.8 GHz turbo limit |
| 175while the other processors are generally in various states of idle. 176 | 162while the other processors are generally in various states of idle. 163 |
| 177Note that cpu1 and cpu7 are HT siblings within core8. 178As cpu7 is very busy, it prevents its sibling, cpu1, | 164Note that cpu1 and cpu5 are HT siblings within core1. 165As cpu5 is very busy, it prevents its sibling, cpu1, |
| 179from entering a c-state deeper than c1. 180 | 166from entering a c-state deeper than c1. 167 |
| 181Note that turbostat reports average GHz of 3.63, while 182the arithmetic average of the GHz column above is lower. 183This is a weighted average, where the weight is %c0. ie. it is the total number of 184un-halted cycles elapsed per time divided by the number of CPUs. 185.SH SMI COUNTING EXAMPLE 186On Intel Nehalem and newer processors, MSR 0x34 is a System Management Mode Interrupt (SMI) counter. 187This counter is shown by default under the "SMI" column. 188.nf 189[root@x980 ~]# turbostat 190cor CPU %c0 GHz TSC SMI %c1 %c3 %c6 CTMP %pc3 %pc6 191 0.11 1.91 3.38 0 1.84 0.26 97.79 29 0.82 83.87 192 0 0 0.40 1.63 3.38 0 10.27 0.12 89.20 20 0.82 83.88 193 0 6 0.06 1.63 3.38 0 10.61 194 1 2 0.37 2.63 3.38 0 0.02 0.10 99.51 22 195 1 8 0.01 1.62 3.38 0 0.39 196 2 4 0.07 1.62 3.38 0 0.04 0.07 99.82 23 197 2 10 0.02 1.62 3.38 0 0.09 198 8 1 0.23 1.64 3.38 0 0.10 1.07 98.60 24 199 8 7 0.02 1.64 3.38 0 0.31 200 9 3 0.03 1.62 3.38 0 0.03 0.05 99.89 29 201 9 9 0.02 1.62 3.38 0 0.05 202 10 5 0.07 1.62 3.38 0 0.08 0.12 99.73 27 203 10 11 0.03 1.62 3.38 0 0.13 204^C 205.fi | 168Note that the Avg_MHz column reflects the total number of cycles executed 169divided by the measurement interval. If the %Busy column is 100%, 170then the processor was running at that speed the entire interval. 171The Avg_MHz multiplied by the %Busy results in the Bzy_MHz -- 172which is the average frequency while the processor was executing -- 173not including any non-busy idle time. 174 |
| 206.SH NOTES 207 208.B "turbostat " 209must be run as root. 210 211.B "turbostat " 212reads hardware counters, but doesn't write them. 213So it will not interfere with the OS or other programs, including --- 39 unchanged lines hidden --- | 175.SH NOTES 176 177.B "turbostat " 178must be run as root. 179 180.B "turbostat " 181reads hardware counters, but doesn't write them. 182So it will not interfere with the OS or other programs, including --- 39 unchanged lines hidden --- |