Skip to content

Commit 889facb

Browse files
lenblenb
committed
tools/power turbostat: v3.0: monitor Watts and Temperature
Show power in Watts and temperature in Celsius when hardware support is present. Intel's Sandy Bridge and Ivy Bridge processor generations support RAPL (Run-Time-Average-Power-Limiting). Per the Intel SDM (Intel® 64 and IA-32 Architectures Software Developer Manual) RAPL provides hardware energy counters and power control MSRs (Model Specific Registers). RAPL MSRs are designed primarily as a method to implement power capping. However, they are useful for monitoring system power whether or not power capping is used. In addition, Turbostat now shows temperature from DTS (Digital Thermal Sensor) and PTM (Package Thermal Monitor) hardware, if present. As before, turbostat reads MSRs, and never writes MSRs. New columns are present in turbostat output: The Pkg_W column shows Watts for each package (socket) in the system. On multi-socket systems, the system summary on the 1st row shows the sum for all sockets together. The Cor_W column shows Watts due to processors cores. Note that Core_W is included in Pkg_W. The optional GFX_W column shows Watts due to the graphics "un-core". Note that GFX_W is included in Pkg_W. The optional RAM_W column on server processors shows Watts due to DRAM DIMMS. As DRAM DIMMs are outside the processor package, RAM_W is not included in Pkg_W. The optional PKG_% and RAM_% columns on server processors shows the % of time in the measurement interval that RAPL power limiting is in effect on the package and on DRAM. Note that the RAPL energy counters have some limitations. First, hardware updates the counters about once every milli-second. This is fine for typical turbostat measurement intervals > 1 sec. However, when turbostat is used to measure events that approach 1ms, the counters are less useful. Second, the 32-bit energy counters are subject to wrapping. For example, a counter incrementing 15 micro-Joule units on a 130 Watt TDP server processor could (in theory) roll over in about 9 minutes. Turbostat detects and handles up to 1 counter overflow per measurement interval. But when the measurement interval exceeds the guaranteed counter range, we can't detect if more than 1 overflow occured. So in this case turbostat indicates that the results are in question by replacing the fractional part of the Watts in the output with "**": Pkg_W Cor_W GFX_W 3** 0** 0** Third, the RAPL counters are energy (Joule) counters -- they sum up weighted events in the package to estimate energy consumed. They are not analong power (Watt) meters. In practice, they tend to under-count because they don't cover every possible use of energy in the package. The accuracy of the RAPL counters will vary between product generations, and between SKU's in the same product generation, and with temperature. turbostat's -v (verbose) option now displays more power and thermal configuration information -- as shown on the turbostat.8 manual page. For example, it now displays the Package and DRAM Thermal Design Power (TDP): cpu0: MSR_PKG_POWER_INFO: 0x2f064001980410 (130 W TDP, RAPL 51 - 200 W, 0.045898 sec.) cpu0: MSR_DRAM_POWER_INFO,: 0x28025800780118 (35 W TDP, RAPL 15 - 75 W, 0.039062 sec.) cpu8: MSR_PKG_POWER_INFO: 0x2f064001980410 (130 W TDP, RAPL 51 - 200 W, 0.045898 sec.) cpu8: MSR_DRAM_POWER_INFO,: 0x28025800780118 (35 W TDP, RAPL 15 - 75 W, 0.039062 sec.) Signed-off-by: Len Brown <len.brown@intel.com>
1 parent ddac0d6 commit 889facb

File tree

2 files changed

+690
-56
lines changed

2 files changed

+690
-56
lines changed

‎tools/power/x86/turbostat/turbostat.8

Lines changed: 67 additions & 36 deletions
Original file line numberDiff line numberDiff line change
@@ -11,16 +11,16 @@ turbostat \- Report processor frequency and idle statistics
1111
.RB [ Options ]
1212
.RB [ "\-i interval_sec" ]
1313
.SH DESCRIPTION
14-
\fBturbostat \fP reports processor topology, frequency
15-
and idle power state statistics on modern X86 processors.
14+
\fBturbostat \fP reports processor topology, frequency,
15+
idle power-state statistics, temperature and power on modern X86 processors.
1616
Either \fBcommand\fP is forked and statistics are printed
1717
upon its completion, or statistics are printed periodically.
1818

1919
\fBturbostat \fP
20-
requires that the processor
20+
must be run on root, and
21+
minimally requires that the processor
2122
supports an "invariant" TSC, plus the APERF and MPERF MSRs.
22-
\fBturbostat \fP will report idle cpu power state residency
23-
on processors that additionally support C-state residency counters.
23+
Additional information is reported depending on hardware counter support.
2424

2525
.SS Options
2626
The \fB-p\fP option limits output to the 1st thread in 1st core of each package.
@@ -57,7 +57,15 @@ Note that multiple CPUs per core indicate support for Intel(R) Hyper-Threading T
5757
\fBGHz\fP average clock rate while the CPU was in c0 state.
5858
\fBTSC\fP average GHz that the TSC ran during the entire interval.
5959
\fB%c1, %c3, %c6, %c7\fP show the percentage residency in hardware core idle states.
60+
\fBCTMP\fP Degrees Celsius reported by the per-core Digital Thermal Sensor.
61+
\fBPTMP\fP Degrees Celsius reported by the per-package Package Thermal Monitor.
6062
\fB%pc2, %pc3, %pc6, %pc7\fP percentage residency in hardware package idle states.
63+
\fBPkg_W\fP Watts consumed by the whole package.
64+
\fBCor_W\fP Watts consumed by the core part of the package.
65+
\fBGFX_W\fP Watts consumed by the Graphics part of the package -- available only on client processors.
66+
\fBRAM_W\fP Watts consumed by the DRAM DIMMS -- available only on server processors.
67+
\fBPKG_%\fP percent of the interval that RAPL throttling was active on the Package.
68+
\fBRAM_%\fP percent of the interval that RAPL throttling was active on DRAM.
6169
.fi
6270
.PP
6371
.SH EXAMPLE
@@ -66,50 +74,73 @@ Without any parameters, turbostat prints out counters ever 5 seconds.
6674
for turbostat to fork).
6775

6876
The first row of statistics is a summary for the entire system.
69-
Note that the summary is a weighted average.
77+
For residency % columns, the summary is a weighted average.
78+
For Temperature columns, the summary is the column maximum.
79+
For Watts columns, the summary is a system total.
7080
Subsequent rows show per-CPU statistics.
7181

7282
.nf
73-
[root@x980]# ./turbostat
74-
cor CPU %c0 GHz TSC %c1 %c3 %c6 %pc3 %pc6
75-
0.09 1.62 3.38 1.83 0.32 97.76 1.26 83.61
76-
0 0 0.15 1.62 3.38 10.23 0.05 89.56 1.26 83.61
77-
0 6 0.05 1.62 3.38 10.34
78-
1 2 0.03 1.62 3.38 0.07 0.05 99.86
79-
1 8 0.03 1.62 3.38 0.06
80-
2 4 0.21 1.62 3.38 0.10 1.49 98.21
81-
2 10 0.02 1.62 3.38 0.29
82-
8 1 0.04 1.62 3.38 0.04 0.08 99.84
83-
8 7 0.01 1.62 3.38 0.06
84-
9 3 0.53 1.62 3.38 0.10 0.20 99.17
85-
9 9 0.02 1.62 3.38 0.60
86-
10 5 0.01 1.62 3.38 0.02 0.04 99.92
87-
10 11 0.02 1.62 3.38 0.02
83+
[root@sandy]# ./turbostat
84+
cor CPU %c0 GHz TSC %c1 %c3 %c6 %c7 CTMP PTMP %pc2 %pc3 %pc6 %pc7 Pkg_W Cor_W GFX_W
85+
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
86+
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
87+
0 4 0.03 0.80 2.29 0.12
88+
1 1 0.04 0.80 2.29 0.25 0.01 0.00 99.71 40
89+
1 5 0.16 0.80 2.29 0.13
90+
2 2 0.05 0.80 2.29 0.06 0.01 0.00 99.88 40
91+
2 6 0.03 0.80 2.29 0.08
92+
3 3 0.05 0.80 2.29 0.08 0.00 0.00 99.87 47
93+
3 7 0.04 0.84 2.29 0.09
8894
.fi
8995
.SH SUMMARY EXAMPLE
9096
The "-s" option prints the column headers just once,
9197
and then the one line system summary for each sample interval.
9298

9399
.nf
94-
[root@x980]# ./turbostat -s
95-
%c0 GHz TSC %c1 %c3 %c6 %pc3 %pc6
96-
0.23 1.67 3.38 2.00 0.30 97.47 1.07 82.12
97-
0.10 1.62 3.38 1.87 2.25 95.77 12.02 72.60
98-
0.20 1.64 3.38 1.98 0.11 97.72 0.30 83.36
99-
0.11 1.70 3.38 1.86 1.81 96.22 9.71 74.90
100+
[root@wsm]# turbostat -S
101+
%c0 GHz TSC %c1 %c3 %c6 CTMP %pc3 %pc6
102+
1.40 2.81 3.38 10.78 43.47 44.35 42 13.67 2.09
103+
1.34 2.90 3.38 11.48 58.96 28.23 41 19.89 0.15
104+
1.55 2.72 3.38 26.73 37.66 34.07 42 2.53 2.80
105+
1.37 2.83 3.38 16.95 60.05 21.63 42 5.76 0.20
100106
.fi
101107
.SH VERBOSE EXAMPLE
102108
The "-v" option adds verbosity to the output:
103109

104110
.nf
105-
GenuineIntel 11 CPUID levels; family:model:stepping 0x6:2c:2 (6:44:2)
106-
12 * 133 = 1600 MHz max efficiency
107-
25 * 133 = 3333 MHz TSC frequency
108-
26 * 133 = 3467 MHz max turbo 4 active cores
109-
26 * 133 = 3467 MHz max turbo 3 active cores
110-
27 * 133 = 3600 MHz max turbo 2 active cores
111-
27 * 133 = 3600 MHz max turbo 1 active cores
112-
111+
[root@ivy]# turbostat -v
112+
turbostat v3.0 November 23, 2012 - Len Brown <lenb@kernel.org>
113+
CPUID(0): GenuineIntel 13 CPUID levels; family:model:stepping 0x6:3a:9 (6:58:9)
114+
CPUID(6): APERF, DTS, PTM, EPB
115+
RAPL: 851 sec. Joule Counter Range
116+
cpu0: MSR_NHM_PLATFORM_INFO: 0x81010f0012300
117+
16 * 100 = 1600 MHz max efficiency
118+
35 * 100 = 3500 MHz TSC frequency
119+
cpu0: MSR_NHM_SNB_PKG_CST_CFG_CTL: 0x1e008402 (UNdemote-C3, UNdemote-C1, demote-C3, demote-C1, locked: pkg-cstate-limit=2: pc6-noret)
120+
cpu0: MSR_NHM_TURBO_RATIO_LIMIT: 0x25262727
121+
37 * 100 = 3700 MHz max turbo 4 active cores
122+
38 * 100 = 3800 MHz max turbo 3 active cores
123+
39 * 100 = 3900 MHz max turbo 2 active cores
124+
39 * 100 = 3900 MHz max turbo 1 active cores
125+
cpu0: MSR_IA32_ENERGY_PERF_BIAS: 0x00000006 (balanced)
126+
cpu0: MSR_RAPL_POWER_UNIT: 0x000a1003 (0.125000 Watts, 0.000015 Joules, 0.000977 sec.)
127+
cpu0: MSR_PKG_POWER_INFO: 0x01e00268 (77 W TDP, RAPL 60 - 0 W, 0.000000 sec.)
128+
cpu0: MSR_PKG_POWER_LIMIT: 0x830000148268 (UNlocked)
129+
cpu0: PKG Limit #1: ENabled (77.000000 Watts, 1.000000 sec, clamp DISabled)
130+
cpu0: PKG Limit #2: ENabled (96.000000 Watts, 0.000977* sec, clamp DISabled)
131+
cpu0: MSR_PP0_POLICY: 0
132+
cpu0: MSR_PP0_POWER_LIMIT: 0x00000000 (UNlocked)
133+
cpu0: Cores Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled)
134+
cpu0: MSR_PP1_POLICY: 0
135+
cpu0: MSR_PP1_POWER_LIMIT: 0x00000000 (UNlocked)
136+
cpu0: GFX Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled)
137+
cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x00691400 (105 C)
138+
cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x884e0000 (27 C)
139+
cpu0: MSR_IA32_THERM_STATUS: 0x88560000 (19 C +/- 1)
140+
cpu1: MSR_IA32_THERM_STATUS: 0x88560000 (19 C +/- 1)
141+
cpu2: MSR_IA32_THERM_STATUS: 0x88540000 (21 C +/- 1)
142+
cpu3: MSR_IA32_THERM_STATUS: 0x884e0000 (27 C +/- 1)
143+
...
113144
.fi
114145
The \fBmax efficiency\fP frequency, a.k.a. Low Frequency Mode, is the frequency
115146
available at the minimum package voltage. The \fBTSC frequency\fP is the nominal
@@ -142,7 +173,7 @@ cor CPU %c0 GHz TSC %c1 %c3 %c6 %pc3 %pc6
142173
10 5 1.42 3.43 3.38 2.14 30.99 65.44
143174
10 11 0.16 2.88 3.38 3.40
144175
.fi
145-
Above the cycle soaker drives cpu7 up its 3.6 Ghz turbo limit
176+
Above the cycle soaker drives cpu7 up its 3.6 GHz turbo limit
146177
while the other processors are generally in various states of idle.
147178

148179
Note that cpu1 and cpu7 are HT siblings within core8.

0 commit comments

Comments
 (0)