Re^3: Profiling the C side of an Inline::C module

Well, my intuition is telling me it will work, if you compile perl with the -pg option, and arrange for the final invocation of ld to use the profiling C runtime. You should end up with a perl that, when run (e.g. with perl -e 1), produces a gmon.out file.

Then, make sure that Inline::C is using the -pg option to compile your module, and when you run perl on your test script, you'll now get a gmon.out file that includes profiling details about your module.

@/=map{[/./g]}qw/.h_nJ Xapou cets krht ele_ r_ra/;
map{y/X_/\n /;print}map{pop@$_}@/for@/
[download]

Comment on Re^3: Profiling the C side of an Inline::C module Select or Download Code

Replies are listed 'Best First'.
Re^4: Profiling the C side of an Inline::C module by samtregar (Abbot) on Sep 05, 2005 at 23:03 UTC
Alright, I'll give it a try. I was won over when I found in reading perlhack that Perl has built-in support for gprof via the "perl.gprof" make target. -sam	[reply]
Re^5: Profiling the C side of an Inline::C module (update?) by Anonymous Monk on Sep 07, 2005 at 17:53 UTC
So, did it ever get it to work? If so, can you post a little bit of your recipe? If not, can you enlighten us as to where the process crapped out?	[reply]
Re^6: Profiling the C side of an Inline::C module (update?) by samtregar (Abbot) on Sep 08, 2005 at 21:30 UTC
Well, I did build a "perl.gprof" which worked fine. Then I got side-tracked and never got back to it. If/when I get something working I'll make a new post about it and link it from this one. -sam	[reply]