Skip to content

Latest commit

 

History

History
133 lines (94 loc) · 3.05 KB

File metadata and controls

133 lines (94 loc) · 3.05 KB

pa.py

The PerformanceAnalyst stores information in a database. This command can be used to query and manage this database and to create plots of the performance results.

The commandline syntax is similar to tools like svn and git. The command uses multiple sub-commands with their own set of commandline options.

Type:

pa.py --help
pa.py help

for the general commandline usage.

Type:

pa.py help <command>
pa.py <command> --help

For the commandline usage of a sub-command.

Here is an example that creates a pdf with a plot with three graphs:

pa.py plot test_timings.pdf test_timings.sqlite \
    my_module_timers_time_a \
    my_module_timers_time_b \
    my_module_timers_time_c

Now follows a short rundown of the usage of the sub-commands of the pa.py command. Check the tool's usage for further details.

The high-level usage of the pa.py command is as folows:

pa.py [options] <command>

Command is one of the sub-commands:

##ls List the contents of the performance database:

pa.py ls [options] <database> {timer}+
argument description
database Name of performance database
timer Name of timer

##mv Move (rename) timers stored in the performance database:

pa.py mv [options] <database> <current timer> <new timer>
argument description
database Name of performance database
current timer Current name of timer
new timer New name of timer

##plot Plot timers stored in the performance database:

pa.py plot [options] <output> <database> <timer>+

The plot will contain two lines for each timer: a solid line for the cpu-time and a dashed line for the real-time.

argument description
output Name of output pdf
database Name of performance database
timer Name of timer

##rm Remove timers stored in the performance database:

pa.py rm [options] <database> <timer>+
argument description
database Name of performance database
timer Name of timer

##stat Calculate statistics of timers stored in the performance database:

pa.py stat [options] <database> <timer>+
argument description
database Name of performance database
timer Name of timer

##history Print history of the database:

pa.py stat [options] <database>
argument description
database Name of performance database