2.5 KiB
layout | title | permalink | redirect_from |
---|---|---|---|
doc | Profiling | /doc/Profiling/ | /wiki/Profiling/ |
Profiling
This is python profiling primer.
For the purpose of this document, qubes-dev
is name of the domain used for postprocessing profiling stats.
Requirements
{% highlight trac-wiki %} yum install gprof2dot graphviz git clone http://git.woju.eu/qubes/profiling.git {% endhighlight %}
If you profile something on dom0, move Upload.sh
from repository to dom0:
{% highlight trac-wiki %} mkdir -p ~/profiling qvm-run -p qubes-dev 'cat ~/profiling/Upload.sh' > ~/profiling/Upload.sh {% endhighlight %}
- WARNING: this will obviously be running third party code which is not signed by ITL nor Fedora. You have been warned.
Workflow
Identify function responsible for some slow action
You have to select area in which you suspect less than optimal performance. If you do not narrow the area, graphs may be unreadable.
Replace suspect function with probe
Replace
def foo(self, bar):
# function content
with
def foo(self, *args, **kwargs):
profile.runctx('self.real_foo(*args, **kwargs)', globals(), locals(),
time.strftime('/home/user/profiling/foo-%Y%m%d-%H%M%S.pstats'))
def real_foo(self, bar):
# function content
Run application
Beware that some functions may be called often. For example qubesmanager/main.py:update_table
gets run once per second. This will produce one pstat file per second.
Remember to revert your changes to application afterwards.
Upload statistics
If you are in dom0:
{% highlight trac-wiki %} cd ~/profiling ./Upload.sh {% endhighlight %}
Analyse
{% highlight trac-wiki %} make {% endhighlight %}
For every ${basename}.pstats
this will produce ${basename}.txt
and ${basename}.svg
. SVG contains call graph. Text file contains list of all functions sorted by cumulative execution time. You may also try make all-png
.
{% highlight trac-wiki %} make index.html {% endhighlight %}
This creates index.html
with all SVG graphics linked to TXT files. Ready for upload.
{% highlight trac-wiki %} make REMOTE=example.com:public_html/qubes/profiling/ upload {% endhighlight %}
Example
This example is from qubes-manager
(qubesmanager/main.py
).
It is apparent than problem is around get_disk_usage
which calls something via subprocess.call
. It does it 15 times, probably once per VM.