ptt_paje - PTT graphical decoder
ptt_paje [options] input_file
Its primary function is to decode
the binary data collected by ptt_trace(1)
into a Paje compliant format.
The output can be used with the Paje visualization tool
to get a graphical representation of the program execution.
Warning: Paje needs complete events, so splitting could break object initialisation.
The comma acts like an or.
Warning: in case of fork without an exec, the libpthread won't be reinitialised, and Paje output won't detect which thread is the main thread.
See also the ptt_view(1)
comments.
The comma acts like an or.
Warning: you should only try to select tid. Selecting mutex or other thing won't work because Paje needs information of all objects that interteract with the display ones (for example thread info for mutex).
See also the ptt_view(1)
comments.
end - start
events skipping start
events
The syntax -z :end or -z start:
is also supported and imply start=0 or end=infinity.
Warning: You should always set start=0 with Paje. Skipping events will result in missing information and it will be impossible to display the chart.
See also the ptt_view(1)
comments.
start
microseconds and before end
microseconds.
The syntax -Z :end or -Z start:
is also supported and imply start=0 or end=infinity.
time filter
Warning: You should always set start=0 with Paje. Skipping events will result in missing information and it will be impossible to display the chart.
Current version of Paje is time and memory aware, so the size of the trace should be small enough in order to be workable.
Next version of Paje should improve this by alllowing partial parsing when displaying, ...
ptt_view(1)
, ptt_trace(1)
, Paje(1)
Matthieu CASTET, Guillaume DURANCEAU