![pmu-tools] (http://halobates.de/pmu-tools.png)

pmu tools is a collection of tools for profile collection and performance analysis on Intel CPUs on top of Linux perf.

• A wrapper to "perf" that provides a full core event list for common Intel CPUs. This allows to use all the Intel events, not just the builtin events of perf.
• Support for Intel "offcore" events on older systems that do not have support for this in the Intel. Offcore events allow to profile the location of a memory access outside the CPU's caches. This also implements a workaround for some issues with offcore events on Sandy Bridge EP (Intel Xeon E5 first generation) This is automatically enabled for the respective events, and also available as a standalone program.
• A "toplev.py" tool to do cycle composition for a workload, that is measure where in the CPU pipe line the bottleneck occurs.
• A tool to manage and compute uncore events on Intel Xeon E5 2600 series (SandyBridge EP) cpus. This can be useful to monitor power management, IO bandwidth, memory bandwidth, QPI (interconnect) traffic, cache hit rates and other metrics. ucevent automatically generates event descriptions for the perf uncore driver and pretty print the output.
• A variety of tools for plotting and post processing perf stat -I1000 -x, or toplev.py -I1000 -x, interval measurements.
• A plot tool to plot perf stat -Ixxx -x, or toplev.py -Ixxx -x, output
• Some utility programs to access pci space or msrs on the command line
• A utility program to program the PMU directly from user space (pmumon.py) for counting. This is mainly useful for testing and experimental purposes.
• A library for self profiling with Linux since Linux 3.3 Note for self-profiling on older kernels you can use [simple-pmu] (http://halobates.de/simple-pmu)
• An example program for address profiling on Nehalem and later Intel CPUs (addr)
• A program to print the currently running events (event-rmap)
• Support for analyzing the raw PEBS records with perf.

Check out the repository. Run the tools from the directory you checked out (but it does not need to be the current directory) They automatically search for other modules and data files in the same directory the script was located in.

A more detailed introduction is in Andi's blog

ocperf.py list List all the events perf and ocperf supports on the current CPU

ocperf.py stat -e eventname ... 

ocperf.py record -e eventname ...

ocperf.py report --stdio

The translation back from the raw event name for report is only supported for the --stdio mode. The interactive browser can be also used, but will display raw perf event names.

When a older kernel is used with offcore events, that does not support offcore events natively, ocperf has to run as root and only one such profiling can be active on a machine.

When -c default is specified, the default sampling overflow value will be filled in for the sampling period. This option needs to be specified before the events and is not supported for all CPUs.

tester provides a simple test suite.

The latego.py, msr.py, pci.py modules can be also used as standalone programs to enable the offcore workaround, change MSRs or change PCI config space respectively.

Do cycle decomposition on a workload: estimate on which part of the CPU pipeline the bottle neck occurs. The bottlenecks are expressed as a tree with different levels (max 4). Each bottleneck is only meaningful if the parent higher level crossed the threshold (it acts similar to a binary search). The tool automatically only prints meaningful ratios, unless -v is specified.

This follows the "Top Down" methology described in B.3.2 of the Intel optimization manual and in this article and in Ahmad Yasin's presentation at a ISCA workshop. I didn't invent it, I'm just implementing it.

A more gentle introduction is in andi's blog

toplev.py only supports counting, that is it cannot tell you where in the program the problem occurred, just what happened.

Requires an Intel Sandy, Ivy Bridge, Haswell CPU.

It works best on Ivy Bridge currently. By default the simple high level model is used. The detailed model is selected with -d, and the max level specified with -l (max 4, default 2).

[IVB model] (http://halobates.de/ivb-hierarchy.svg) [Simple model] (http://halobates.de/simple-hierarchy.svg)

Usage:

./toplev.py [-lX] [-v] [-d] [-o logfile] program
measure program
./toplev.py [-lX] [-v] [-d] [-o logfile] -a sleep X
measure whole system for X seconds
./toplev.py [-lX] [-v] [-d] [-o logfile] -p PID
measure pid PID

-o set output file
-v print everything
-d use detailed model if available (only Ivy Bridge currently)
-lLEVEL only use events upto max level (max 4)
-x,     Enable CSV mode with separator ,
-Ixxx   Enable interval mode, measure every xxx ms

Other perf arguments allowed (see the perf documentation) After -- perf arguments conflicting with toplevel can be used.

The lower levels of the measurement tree are less reliable than the higher levels. They also rely on counter multi-plexing and cannot use groups, which can cause larger measurement errors with non steady state workloads.

(If you don't understand this terminology; it means measurements are much less accurate and it works best with programs that primarily do the same thing over and over)

It's recommended to measure the work load only after the startup phase by using interval mode or attaching later. level 1 or running without -d is generally the most reliable.

One of the events (even used by level 1) requires a recent enough kernel that understands its counter constraints. 3.10+ is safe.

Please see the [ucevent documentation] (http://github.com/andikleen/pmu-tools/tree/master/ucevent/#ucevent-uncore-monitoring)

interval-plot.py can plot the output of perf stat -I1000 -x, or or toplev.py -I1000 -x,

Requires pyplot to be installed.

Below is the level 2 toplev measurement of a Linux kernel compile.

![plot-example] (http://halobates.de/interval.plot.l2.png)

This converts the output of perf stat -Ixxx -x, / toplev.py -Ixxx -x, to a normalized output (one column for each event). This allows easier plotting and processing with other tools (spreadsheets, R, JMP, gnuplot etc.)

Plot an already normalized data file. Requires pyplot to be installed.

Self is a simple library to support self-profiling of programs, that is programs that measure their own execution.

cd self make

Read the documentation.

[rdpmc] (http://htmlpreview.github.com/?https://github.com/andikleen/pmu-tools/blob/master/self/rdpmc.html) is the basic facility to access raw counters.

ocperf can be used to generate raw perf numbers for your CPU to pass to rdpmc_open()

ocperf list | less

![example] (http://halobates.de/pmutools-event.png)

XXX is the needed event number in hex. Note that self does not support offcore or uncore events.

Also the event numbers are CPU specific, so you may need a /proc/cpuinfo model check for portable programs (see the ocperf source for example)

Example (replace EVENTNUMBER with your intended event from above or a perf event like PERF_COUNT_HW_CPU_CYCLES).

This is very simplified, for a real benchmark you almost certainly want some warmup, multiple iterations, possibly context switch filtering and some filler code to avoid cache effects.

#include "rdpmc.h"

struct rdpmc_ctx ctx;
unsigned long long start, end;

if (rdpmc_open(EVENTNUMBER, &ctx) < 0) ... error ...
start = rdpmc_read(&ctx);
... your workload ...
end = rdpmc_read(&ctx);

[measure] (http://htmlpreview.github.com/?https://github.com/andikleen/pmu-tools/blob/master/self/measure.html) supports event group profiling.
[interrupts] (http://htmlpreview.github.com/?https://github.com/andikleen/pmu-tools/blob/master/self/interrupts.html) provides functions for a common use case of filtering out context switches and interrupts from micro benchmarks. These only work on Intel Ivy and Sandy Bridge CPUs.

Link the object files and include the header files in your program

/sys/devices/cpu/rdpmc must be 1.

rtest.c and test2.c provide examples. http://halobates.de/modern-pmus-yokohama.pdf provides some additional general information on cycle counting. The techniques used with simple-pmu described there can be used with self too.

Perf doesn't export the raw PEBS output, which contains a lot of useful information. PEBS is a sampling format generated by Intel CPUs for some events.

pebs-grabber gabs PEBS data from perf. This assumes the perf pebs handler is running, we just also do trace points with the raw data.

May need some minor tweaks as kernel interface change, and will also not likely work on very old kernels.

This will create two new trace points trace_pebs_v1 and trace_pebs_v2 that log the complete PEBS record. When the CPU supports PEBSv2 (Haswell) the additional fields will be logged in pebs_v2.

  make [KDIR=/my/kernel/build/dir]
  insmod pebs-grabber.ko 
  # needs to record as root
  perf record -e cycles:p,pebs_v1,pebs_v2 [command, -a for all etc.]
  perf report
  perf script to display pebs data
  # alternatively trace-cmd and kernelshark can be also used to dump
  # the pebs data

See http://download.intel.com/products/processor/manual/253669.pdf 18.10.2 for a description of the PEBS fields.

Note this doesn't work with standard FC18 kernels, as they broke trace points in modules. It works with later and earlier kernels.

addr is a basic frame work for self profiling of memory addresses accessed by the program. Requires a Linux 3.10+ kernel and a supported CPU.

event-rmap [cpu] prints the currently running events. This provides an easier answer to question Q2j in Vince Weaver's perf events FAQ.

Please post to the linux-perf-users@vger.kernel.org mailing list.

ocperf is under GPLv2, self and addr under the modified BSD license.

Andi Kleen pmu-tools@halobates.de