Skip to content

Commit

Permalink
gh-118518: Improve perf docs (#118708)
Browse files Browse the repository at this point in the history
  • Loading branch information
pablogsal authored May 7, 2024
1 parent a94ac56 commit b9caa09
Showing 1 changed file with 48 additions and 22 deletions.
70 changes: 48 additions & 22 deletions Doc/howto/perf_profiling.rst
Original file line number Diff line number Diff line change
Expand Up @@ -162,12 +162,12 @@ the :option:`!-X` option takes precedence over the environment variable.

Example, using the environment variable::

$ PYTHONPERFSUPPORT=1 python script.py
$ PYTHONPERFSUPPORT=1 perf record -F 9999 -g -o perf.data python script.py
$ perf report -g -i perf.data

Example, using the :option:`!-X` option::

$ python -X perf script.py
$ perf record -F 9999 -g -o perf.data python -X perf script.py
$ perf report -g -i perf.data

Example, using the :mod:`sys` APIs in file :file:`example.py`:
Expand All @@ -184,7 +184,7 @@ Example, using the :mod:`sys` APIs in file :file:`example.py`:
...then::

$ python ./example.py
$ perf record -F 9999 -g -o perf.data python ./example.py
$ perf report -g -i perf.data


Expand All @@ -210,31 +210,57 @@ of ``perf``.
How to work without frame pointers
----------------------------------

If you are working with a Python interpreter that has been compiled without frame pointers
you can still use the ``perf`` profiler but the overhead will be a bit higher because Python
needs to generate unwinding information for every Python function call on the fly. Additionally,
``perf`` will take more time to process the data because it will need to use the DWARF debugging
information to unwind the stack and this is a slow process.
If you are working with a Python interpreter that has been compiled without
frame pointers, you can still use the ``perf`` profiler, but the overhead will be
a bit higher because Python needs to generate unwinding information for every
Python function call on the fly. Additionally, ``perf`` will take more time to
process the data because it will need to use the DWARF debugging information to
unwind the stack and this is a slow process.

To enable this mode, you can use the environment variable :envvar:`PYTHON_PERF_JIT_SUPPORT` or the
:option:`-X perf_jit <-X>` option, which will enable the JIT mode for the ``perf`` profiler.
To enable this mode, you can use the environment variable
:envvar:`PYTHON_PERF_JIT_SUPPORT` or the :option:`-X perf_jit <-X>` option,
which will enable the JIT mode for the ``perf`` profiler.

When using the perf JIT mode, you need an extra step before you can run ``perf report``. You need to
call the ``perf inject`` command to inject the JIT information into the ``perf.data`` file.
.. note::

Due to a bug in the ``perf`` tool, only ``perf`` versions higher than v6.8
will work with the JIT mode. The fix was also backported to the v6.7.2
version of the tool.

Note that when checking the version of the ``perf`` tool (which can be done
by running ``perf version``) you must take into account that some distros
add some custom version numbers including a ``-`` character. This means
that ``perf 6.7-3`` is not necessarily ``perf 6.7.3``.

When using the perf JIT mode, you need an extra step before you can run ``perf
report``. You need to call the ``perf inject`` command to inject the JIT
information into the ``perf.data`` file.::

$ perf record -F 9999 -g --call-graph dwarf -o perf.data python -Xperf_jit my_script.py
$ perf inject -i perf.data --jit
$ perf report -g -i perf.data
$ perf inject -i perf.data --jit --output perf.jit.data
$ perf report -g -i perf.jit.data

or using the environment variable::

$ PYTHON_PERF_JIT_SUPPORT=1 perf record -F 9999 -g --call-graph dwarf -o perf.data python my_script.py
$ perf inject -i perf.data --jit
$ perf report -g -i perf.data

Notice that when using ``--call-graph dwarf`` the ``perf`` tool will take snapshots of the stack of
the process being profiled and save the information in the ``perf.data`` file. By default the size of
the stack dump is 8192 bytes but the user can change the size by passing the size after comma like
``--call-graph dwarf,4096``. The size of the stack dump is important because if the size is too small
``perf`` will not be able to unwind the stack and the output will be incomplete.
$ perf inject -i perf.data --jit --output perf.jit.data
$ perf report -g -i perf.jit.data

``perf inject --jit`` command will read ``perf.data``,
automatically pick up the perf dump file that Python creates (in
``/tmp/perf-$PID.dump``), and then create ``perf.jit.data`` which merges all the
JIT information together. It should also create a lot of ``jitted-XXXX-N.so``
files in the current directory which are ELF images for all the JIT trampolines
that were created by Python.

.. warning::
Notice that when using ``--call-graph dwarf`` the ``perf`` tool will take
snapshots of the stack of the process being profiled and save the
information in the ``perf.data`` file. By default the size of the stack dump
is 8192 bytes but the user can change the size by passing the size after
comma like ``--call-graph dwarf,4096``. The size of the stack dump is
important because if the size is too small ``perf`` will not be able to
unwind the stack and the output will be incomplete. On the other hand, if
the size is too big, then ``perf`` won't be able to sample the process as
frequently as it would like as the overhead will be higher.

0 comments on commit b9caa09

Please sign in to comment.