Call Graph for Linux Discussion

jeffrey-gallagher · ‎05-16-2003

Hi gang,

This thread is an extension of the poll posted this week. If you haven't answered it, bump up a level to the answer the poll. Then, come back if you want to discuss the following:

1) Are you currently using callgraph activities to analyzer your Linux code?

2) If no, why not?

3) If yes, do you prefer the character output that you are currently viewing? Or would you prefer a graphical viewer instead if you could get one?

All comments welcomed, team.

Cheers

jdg

Daniel_B_Intel2 · ‎05-19-2003

Jeff,

I do use the call graph to Analyze my Linux applications :-)!

Yet I'd like you to see how "host" can manage the Poll.

Do you want to make a try?

-Daniel

Message Edited by intel.software.network.support on 12-09-2005 10:39 AM

Konstantin_L_Intel · ‎05-20-2003

I regularly use Call Graph on Linux and Windows to get a general notion about my code bottlenecks or call flow of the code written by other people. Call Graph session is my usual start point in the optimization process. The most useful thing in CG for me is various information it provides.

Of course it is always nice to have convenient GUI interface like in the VTune analyzer for Windows and it will be a good addition if it is available. This would help to avoid writing special scripts for better representation of vtl output or/and importing vtl output to Excel-like program. In general, from engineering point of view I found VTune analyzer for Linux pretty useful even w/o its own GUI.

TimP · ‎05-21-2003

We've gone so far as to use gcc to get gprof call graph information on IA64 linux and IA32 Windows. gprof is supported reasonably well by Intel compilers on IA32 linux. Format becomes more of a consideration when comparing results among various compilers and operating systems, and there the gprof format is the only one likely to permit easy comparison. Even the HPUX-ia64 Caliper profiler has a gprof call graph format. Granted, the data have to be copied into a spread sheet before comparisons are feasible, let alone admissible in staff meetings.