cpupower-monitor.1 (6619B)
1.TH CPUPOWER\-MONITOR "1" "22/02/2011" "" "cpupower Manual" 2.SH NAME 3cpupower\-monitor \- Report processor frequency and idle statistics 4.SH SYNOPSIS 5.ft B 6.B cpupower monitor 7.RB "\-l" 8 9.B cpupower monitor 10.RB [ -c ] [ "\-m <mon1>," [ "<mon2>,..." ] ] 11.RB [ "\-i seconds" ] 12.br 13.B cpupower monitor 14.RB [ -c ][ "\-m <mon1>," [ "<mon2>,..." ] ] 15.RB command 16.br 17.SH DESCRIPTION 18\fBcpupower-monitor \fP reports processor topology, frequency and idle power 19state statistics. Either \fBcommand\fP is forked and 20statistics are printed upon its completion, or statistics are printed periodically. 21 22\fBcpupower-monitor \fP implements independent processor sleep state and 23frequency counters. Some are retrieved from kernel statistics, some are 24directly reading out hardware registers. Use \-l to get an overview which are 25supported on your system. 26 27.SH Options 28.PP 29\-l 30.RS 4 31List available monitors on your system. Additional details about each monitor 32are shown: 33.RS 2 34.IP \(bu 35The name in quotation marks which can be passed to the \-m parameter. 36.IP \(bu 37The number of different counters the monitor supports in brackets. 38.IP \(bu 39The amount of time in seconds the counters might overflow, due to 40implementation constraints. 41.IP \(bu 42The name and a description of each counter and its processor hierarchy level 43coverage in square brackets: 44.RS 4 45.IP \(bu 46[T] \-> Thread 47.IP \(bu 48[C] \-> Core 49.IP \(bu 50[P] \-> Processor Package (Socket) 51.IP \(bu 52[M] \-> Machine/Platform wide counter 53.RE 54.RE 55.RE 56.PP 57\-m <mon1>,<mon2>,... 58.RS 4 59Only display specific monitors. Use the monitor string(s) provided by \-l option. 60.RE 61.PP 62\-i seconds 63.RS 4 64Measure interval. 65.RE 66.PP 67\-c 68.RS 4 69Schedule the process on every core before starting and ending measuring. 70This could be needed for the Idle_Stats monitor when no other MSR based 71monitor (has to be run on the core that is measured) is run in parallel. 72This is to wake up the processors from deeper sleep states and let the 73kernel re 74-account its cpuidle (C-state) information before reading the 75cpuidle timings from sysfs. 76.RE 77.PP 78command 79.RS 4 80Measure idle and frequency characteristics of an arbitrary command/workload. 81The executable \fBcommand\fP is forked and upon its exit, statistics gathered since it was 82forked are displayed. 83.RE 84.PP 85\-v 86.RS 4 87Increase verbosity if the binary was compiled with the DEBUG option set. 88.RE 89 90.SH MONITOR DESCRIPTIONS 91.SS "Idle_Stats" 92Shows statistics of the cpuidle kernel subsystem. Values are retrieved from 93/sys/devices/system/cpu/cpu*/cpuidle/state*/. 94The kernel updates these values every time an idle state is entered or 95left. Therefore there can be some inaccuracy when cores are in an idle 96state for some time when the measure starts or ends. In worst case it can happen 97that one core stayed in an idle state for the whole measure time and the idle 98state usage time as exported by the kernel did not get updated. In this case 99a state residency of 0 percent is shown while it was 100. 100 101.SS "Mperf" 102The name comes from the aperf/mperf (average and maximum) MSR registers used 103which are available on recent X86 processors. It shows the average frequency 104(including boost frequencies). 105The fact that on all recent hardware the mperf timer stops ticking in any idle 106state it is also used to show C0 (processor is active) and Cx (processor is in 107any sleep state) times. These counters do not have the inaccuracy restrictions 108the "Idle_Stats" counters may show. 109May work poorly on Linux-2.6.20 through 2.6.29, as the \fBacpi-cpufreq \fP 110kernel frequency driver periodically cleared aperf/mperf registers in those 111kernels. 112 113.SS "Nehalem" "SandyBridge" "HaswellExtended" 114Intel Core and Package sleep state counters. 115Threads (hyperthreaded cores) may not be able to enter deeper core states if 116its sibling is utilized. 117Deepest package sleep states may in reality show up as machine/platform wide 118sleep states and can only be entered if all cores are idle. Look up Intel 119manuals (some are provided in the References section) for further details. 120The monitors are named after the CPU family where the sleep state capabilities 121got introduced and may not match exactly the CPU name of the platform. 122For example an IvyBridge processor has sleep state capabilities which got 123introduced in Nehalem and SandyBridge processor families. 124Thus on an IvyBridge processor one will get Nehalem and SandyBridge sleep 125state monitors. 126HaswellExtended extra package sleep state capabilities are available only in a 127specific Haswell (family 0x45) and probably also other future processors. 128 129.SS "Fam_12h" "Fam_14h" 130AMD laptop and desktop processor (family 12h and 14h) sleep state counters. 131The registers are accessed via PCI and therefore can still be read out while 132cores have been offlined. 133 134There is one special counter: NBP1 (North Bridge P1). 135This one always returns 0 or 1, depending on whether the North Bridge P1 136power state got entered at least once during measure time. 137Being able to enter NBP1 state also depends on graphics power management. 138Therefore this counter can be used to verify whether the graphics' driver 139power management is working as expected. 140 141.SH EXAMPLES 142 143cpupower monitor -l" may show: 144.RS 4 145Monitor "Mperf" (3 states) \- Might overflow after 922000000 s 146 147 ... 148 149Monitor "Idle_Stats" (3 states) \- Might overflow after 4294967295 s 150 151 ... 152 153.RE 154cpupower monitor \-m "Idle_Stats,Mperf" scp /tmp/test /nfs/tmp 155 156Monitor the scp command, show both Mperf and Idle_Stats states counter 157statistics, but in exchanged order. 158 159 160 161.RE 162Be careful that the typical command to fully utilize one CPU by doing: 163 164cpupower monitor cat /dev/zero >/dev/null 165 166Does not work as expected, because the measured output is redirected to 167/dev/null. This could get workarounded by putting the line into an own, tiny 168shell script. Hit CTRL\-c to terminate the command and get the measure output 169displayed. 170 171.SH REFERENCES 172"BIOS and Kernel Developer’s Guide (BKDG) for AMD Family 14h Processors" 173https://support.amd.com/us/Processor_TechDocs/43170.pdf 174 175"Intel® Turbo Boost Technology 176in Intel® Core™ Microarchitecture (Nehalem) Based Processors" 177http://download.intel.com/design/processor/applnots/320354.pdf 178 179"Intel® 64 and IA-32 Architectures Software Developer's Manual 180Volume 3B: System Programming Guide" 181https://www.intel.com/products/processor/manuals 182 183.SH FILES 184.ta 185.nf 186/dev/cpu/*/msr 187/sys/devices/system/cpu/cpu*/cpuidle/state*/. 188.fi 189 190.SH "SEE ALSO" 191powertop(8), msr(4), vmstat(8) 192.PP 193.SH AUTHORS 194.nf 195Written by Thomas Renninger <trenn@suse.de> 196 197Nehalem, SandyBridge monitors and command passing 198based on turbostat.8 from Len Brown <len.brown@intel.com>