Re: gprof Observations

["Bhattacharya, Soubhik" <soubhik_bhattacharya at mentorg dot com>]

John Levon wrote:

> > 1. if a function `foo()' is not compiled with -pg then gprof fails to 
> > determine the callers of and no. of calls to `foo()'. however, if 
> > `foo()' runs long enuff then it reliably estimates its self-time.
> >
> > 2. using -pg during linking ensures that the program is linked with a 
> > special gprof specific start file. this is the minimum requirement for 
> > gprof to be able to generate some useful information (a flat profile).
>
>
> These are both clearly documented AFAIK.

well, these observations are not novel, in the sense that they can be 
derived from whatever is written in the `gprof' documentation. but i 
still find one-line summaries useful, as they save the time spent 
groping around the manual, and using reasoning to draw a conclusion.

> > 3. gprof fails to gather profiling info (self-time, no. of calls, 
> > calling functions) of a function residing in a shared object. any 
> > suggestion? i searched the net to find a couple of mails on a similar 
> > topic (http://mail.gnu.org/archive/html/bug-gnu-utils/2001-07/msg00284.html
> > http://sources.redhat.com/ml/binutils/2003-03/msg00208.html), which, 
> > sadly, went unanswered.
>
>
> Use GNU libc's "sprof" (documented well enough if you google) or a more
> complete solution like OProfile on Linux etc.

thanks for the pointers. i'm reading your "Profiling in Linux HOWTO" and 
looking forward to using OProfile soon.

Regards,
Soubhik.

--
-0-1-2-3-4-5-6-7-8-9-0-1-2-3-4-5-6-7
Soubhik Bhattacharya
Member Technical Staff
Mentor Graphics
Hyderabad, India.
-0-1-2-3-4-5-6-7-8-9-0-1-2-3-4-5-6-7