This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: gcc compile-time performance

   From: Marc Espie <>
   Date: Sun, 19 May 2002 07:59:13 +0200

   On Sat, May 18, 2002 at 09:24:25PM -0700, David S. Miller wrote:
   > Hey Marc, could you run your tests with this patch applied
   > to gcc-3.1?  It should kill ~10% or more of the compile time
   > overhead when optimizations are enabled.
   Not significant on x86... not more than 2 seconds difference, out of
   several minutes, both in -O2 and -O1 mode.

Thanks for the test.

I wonder how hard it would be for you to build 'cc1' with
the gcov options enabled ("-fprofile-arcs -ftest-coverage")
and then tar up the foo.gcov files after building each part
of the BSD kernel.

I suppose one could write a shell script that did:

1) Run GCC with $* as options.

2) Tar up *.gcov files into uniquely named tar file.

and then you could set "CC" to this script during your
build to generate all the profiling data.

Make sure you have a lot of disk space available if you
attempt this :-)

Something similar can be done to generate gprof information

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]