Marc, On Tue, Sep 23, 2008 at 2:51 PM, <[EMAIL PROTECTED]> wrote: > Hey all, > > for the last hours I tried to count the retired instruction during a single > pthread_mutex_lock() call. I tried quadrillion different runtime options of > pfmon, but wasn't able to accomplish this. Currently I am trying with > pfmon --resolve-addresses --trigger-code-start-address=__pthread_mutex_lock > --trigger-code-stop-address=__pthread_mutex_lock -einstructions_retired > --follow-pthread my_program > > However, I always get just 32 retired instructions. I also wrote my own > pthread_mutex_lock wrapper using the LD_PRELOAD approach and libpfm. This > reports about 830 retired instructions per call. > > I do not know what's wrong. Any hint is highly appreciated. I tried pfmon 3.4 > and 3.5. >
Use the -v option and check that the addresses of the start trigger and stop trigger match your expectation. those routines may be implemented in a way which could defeat the start = stop trick used by pfmon to dynamically detect the return address. ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ perfmon2-devel mailing list perfmon2-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/perfmon2-devel
