Lines Matching +full:cpu +full:- +full:offset

3 turbostat \- Report processor frequency and idle statistics
12 .RB [ "\--interval seconds" ]
15 idle power-state statistics, temperature and power on X86 processors.
19 in one-shot upon its completion.
22 The 5-second interval can be changed using the --interval option.
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-sensitiv…
29 \fB--add attributes\fP add column with counter having specified 'attributes'. The 'location' attri…
32 msrDDD is a decimal offset, eg. msr16
33 msr0xXXX is a hex offset, eg. msr0x10
36 …On Intel hybrid platforms, instead of one "cpu" perf device there are two, "cpu_core" and "cpu_ato…
37 …Turbostat, in this case, allow user to use "cpu" device and will automatically detect the type of …
38 For a complete example see "ADD PERF COUNTER EXAMPLE #2 (using virtual "cpu" device)".
39 … event for given device from /sys/bus/event_source/devices/<device>/events/<event> eg. c1-residency
40 …perf/cstate_core/c1-residency would then use /sys/bus/event_source/devices/cstate_core/events/c1-r…
42 scope: {\fBcpu\fP | \fBcore\fP | \fBpackage\fP}
43 sample and print the counter for every cpu, core, or package.
44 default: cpu
47 MSRs are read as 64-bits, u32 truncates the displayed value to 32-bits.
54 'average' similar to raw, but also averaged for node/package summaries (or when using -S).
58 Any string that does not match a key-word above is used
62--add pmt,[attr_name=attr_value, ...]\fP add column with a PMT (Intel Platform Monitoring Technolo…
76 domain={\fBcpu%u\fP | \fBcore%u\fP | \fBpackage%u\fP}
77 'cpu' per cpu/thread counter.
83 '%u' offset within the PMT MMIO region.
86 …'%u' least significant bit within the 64 bit value read from 'offset'. Together with 'msb', used t…
89 …'%u' most significant bit within the 64 bit value read from 'offset'. Together with 'lsb', used to…
95--cpu cpu-set\fP limit output to system summary plus the specified cpu-set. If cpu-set is the str…
97 \fB--hide column\fP do not show the specified built-in columns. May be invoked multiple times, or …
99 \fB--enable column\fP show the specified built-in columns, which are otherwise disabled, by default…
100 The column name "all" can be used to enable all disabled-by-default built-in counters.
102 \fB--show column\fP show only the specified built-in columns. May be invoked multiple times, or wi…
104 \fB--show CATEGORY --hide CATEGORY\fP Show and hide also accept a single CATEGORY of columns: "all…
106 \fB--Dump\fP displays the raw counter values.
108 \fB--quiet\fP Do not decode and print the system configuration header information.
110 \fB--no-msr\fP Disable all the uses of the MSR driver.
112 \fB--no-perf\fP Disable all the uses of the perf API.
114 \fB--interval seconds\fP overrides the default 5.0 second measurement interval.
116 \fB--num_iterations num\fP number of the measurement iterations.
118 \fB--out output_file\fP turbostat output is written to the specified output_file.
121 \fB--help\fP displays usage for the most common parameters.
123 \fB--Joules\fP displays energy in Joules, rather than dividing Joules by time to print power in Wat…
125 \fB--list\fP display column header names available for use by --show and --hide, then exit.
127 \fB--Summary\fP limits output to a 1-line System Summary for each interval.
129 \fB--TCC temperature\fP sets the Thermal Control Circuit temperature for systems which do not expor…
131 \fB--version\fP displays the version.
137--quiet is not used) is followed by statistics. The first row of the statistics labels the conten…
140CPU, the number of microseconds elapsed during counter collection, including thread migration -- i…
142CPU, the gettimeofday(2) value (seconds.subsec since Epoch) when the counters ending the measureme…
144 …umber. Note that multiple CPUs per core indicate support for Intel(R) Hyper-Threading Technology …
146 \fBCPU\fP Linux CPU (logical processor) number. Yes, it is okay that on many systems the CPUs are …
148 \fBPackage\fP processor package number -- not present on systems with a single processor package.
150 \fBAvg_MHz\fP number of cycles executed divided by time elapsed. Note that this includes idle-time…
152 \fBBusy%\fP percent of the measurement interval that the CPU executes instructions, aka. % of time …
154 \fBBzy_MHz\fP average clock rate while the CPU was not idle (ie. in "c0" state).
158 \fBIRQ\fP The number of interrupts serviced by that CPU during the measurement interval. The syste…
160 …ystem Management Interrupts serviced CPU during the measurement interval. While this counter is …
162-state names as exported in /sys/devices/system/cpu/cpu*/cpuidle/state*/name. While their names a…
164 … exists and enabled). These statistics come from the /sys/devices/system/cpu/cpu*/cpuidle/state*/b…
166-, C2-, C3-...\fP The idle governor idle state misprediction statistics. Inidcates the number time…
170 \fBCPU%c1, CPU%c3, CPU%c6, CPU%c7\fP show the percentage residency in hardware core idle states. T…
172 \fBCoreTmp\fP Degrees Celsius reported by the per-core Digital Thermal Sensor.
174 \fBPkgTmp\fP Degrees Celsius reported by the per-package Package Thermal Monitor.
176 …nterval. Note that events since boot can be find in /sys/devices/system/cpu/cpu*/thermal_throttle…
192 \fBAny%C0\fP Percentage of time that at least one CPU is busy.
196 \fBCPUGFX%\fP Percentage of time that at least one CPU is busy at the same time as at least one Gra…
204 \fBGFXWatt\fP Watts consumed by the Graphics part of the package -- available only on client proces…
206 \fBRAMWatt\fP Watts consumed by the DRAM DIMMS -- available only on server processors.
210 … of the package throttling time, and thus may be higher than 100% on a multi-package system. Note…
214 \fBUncMHz\fP per-package uncore MHz, instantaneous sample.
216 \fBUMHz1.0\fP per-package uncore MHz for pm_domain=1 and fabric_cluster=0, instantaneous sample. S…
217 Intel Granite Rapids systems use pm_domains 0-2 for CPUs, and 3-4 for IO, with cluster always 0.
218 For the "--show" and "--hide" options, use "UncMHz" to operate on all UMHz*.* as a group.
220 By default, turbostat dumps all possible information -- a system configuration header, followed by …
221 This is ideal for remote debugging, use the "--out" option to save everything to a text file, and g…
223--quiet" option will skip the configuration information, and turbostat will show only the counter …
225--show and --hide, the "--list" option is available. Usually, the CATEGORY names above are used t…
228 sudo ./turbostat --show sysfs --quiet sleep 10
246 Output can instead be saved to a file using the --out option.
250 [root@hsw]# ./turbostat -o ts.out sleep 10
256 Periodic output goes to stdout, by default, unless --out is used to specify an output file.
257 The 5-second interval can be changed with the "-i sec" option.
259 sudo turbostat --quiet --show CPU,frequency
260 Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz CPU%c7 UncMhz
261 - - 524 12.48 4198 3096 74.53 3800
270 Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz CPU%c7 UncMhz
271 - - 525 12.52 4198 3096 74.54 3800
281 This example also shows the use of the --show option to show only the desired columns.
286 before taking measurements. In the example above, "--quiet" is used
289 turbostat version 2022.04.16 - Len Brown <lenb@kernel.org>
290 Kernel command line: BOOT_IMAGE=/boot/vmlinuz-5.18.0-rc6-00001-ge6891250e3b5 ...
294 CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM HT TM
295 CPUID(6): APERF, TURBO, DTS, PTM, HWP, HWPnotify, HWPwindow, HWPepp, No-HWPpkg, EPB
302 cpu7: MSR_MISC_PWR_MGMT: 0x00401cc0 (ENable-EIST_Coordination DISable-EPB DISable-OOB)
307 cpu7: MSR_IA32_POWER_CTL: 0x002c005d (C1E auto-promotion: DISabled)
318 …SR_PKG_CST_CONFIG_CONTROL: 0x1e008008 (UNdemote-C3, UNdemote-C1, demote-C3, demote-C1, locked, pkg
319 Uncore Frequency pkg0 die0: 800 - 3900 MHz (800 - 3900 MHz)
336 cpu7: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch)
341 cpu0: MSR_HWP_STATUS: 0x00000004 (No-Guaranteed_Perf_Change, No-Excursion_Min)
344 cpu0: MSR_PKG_POWER_INFO: 0x00000168 (45 W TDP, RAPL 0 - 0 W, 0.000000 sec.)
358 cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x00640000 (100 C) (100 default - 0 offset)
371 frequency of the processor -- this should match the brand string
378 Here we limit turbostat to showing just the CPU number for cpu0 - cpu3.
379 We add a counter showing the 32-bit raw value of MSR 0x199 (MSR_IA32_PERF_CTL),
383 sudo ./turbostat --quiet --cpu 0-3 --show CPU --add msr0x199,u32,raw,PRF_CTRL sleep .1
386 - 0x00000000
395 Here we limit turbostat to showing just the CPU number for cpu0 - cpu3.
399 We also show CPU%c1 built-in counter that should show similar values.
401 sudo ./turbostat --quiet --cpu 0-3 --show CPU,CPU%c1 --add perf/cstate_core/c1-residency,cpu,delta,…
403 CPU pCPU%c1 CPU%c1
404 - 34.89 34.89
412 .SH ADD PERF COUNTER EXAMPLE #2 (using virtual cpu device)
415 We add a counter showing number of L3 cache misses, using virtual "cpu" device,
423--quiet --cpu 0,12 --show CPU --add perf/cpu/cache-misses,cpu,delta,raw,VCMISS --add perf/cpu_core…
424 turbostat: added_perf_counters_init_: perf/cpu_atom/cache-misses: failed to open counter on cpu0
425 turbostat: added_perf_counters_init_: perf/cpu_core/cache-misses: failed to open counter on cpu12
428 - 0x0000000000000000 0x0000000000000000 0x0000000000000000
435 Here we limit turbostat to showing just the CPU number 0.
444 we set the offset to '0', which is a offset of the counter within the PMT MMIO region,
446 … guid to '0x1a067102', that identifies the PMT MMIO region to which the 'offset' is applied to rea…
453 we set the offset to '0', which is a offset of the counter within the PMT MMIO region,
455 … guid to '0x1a067102', that identifies the PMT MMIO region to which the 'offset' is applied to rea…
458--quiet --cpu 0 --show CPU --add pmt,name=XTAL,type=raw,format=delta,domain=package0,offset=0,lsb=…
461 - 0x0000006d4d957ca7 0.00
468 For interval-mode, turbostat will immediately end the current interval
471 Control-C will be send a SIGINT to turbostat,
475 SIGINT will interrupt interval-mode.
476 The end-of-interval data will be collected and displayed before turbostat exits.
479 end-of-interval data will be collected and displayed before turbostat
485 Alternatively, non-root users can be enabled to run turbostat this way:
489 # chmod +r /dev/cpu/*/msr
499 may work poorly on Linux-2.6.20 through 2.6.29,
500 as \fBacpi-cpufreq \fPperiodically cleared the APERF and MPERF MSRs
504 number of elapsed cycles divided by the entire sample interval --
506 to systems lacking a non-stop TSC.
528 The APERF, MPERF MSRs are defined to count non-halted cycles.
563 /dev/cpu/*/msr
565 /sys/devices/system/cpu/