This is the mail archive of the
mailing list for the GCC project.
Re: gcc cache misses [was: Re: OT: How is memory latency importanton AMD64 box while compiling large C/C++ sources]
- From: Nicholas Nethercote <njn at cs dot utexas dot edu>
- To: Karel Gardas <kgardas at objectsecurity dot com>
- Cc: Mike Stump <mrs at apple dot com>, GCC Mailing List <gcc at gcc dot gnu dot org>
- Date: Tue, 12 Apr 2005 19:12:37 -0500 (CDT)
- Subject: Re: gcc cache misses [was: Re: OT: How is memory latency importanton AMD64 box while compiling large C/C++ sources]
- References: <Pine.LNX.firstname.lastname@example.org>
On Tue, 12 Apr 2005, Karel Gardas wrote:
cachegrind can also be used to estimate the number (though, not sure
how accurate it is, possibly not very). I use Shark to actually get
the real number.
Perhaps it's possible that cachegrind is wrong or cache misses differ from
platform to platform, but I would tell that I get very good numbers for
gcc running on x86 platform:
In my experience Cachegrind can give pretty good numbers for L1 misses,
espcially D1, but the L2 misses tend to vary more. I saw this with
comparisons against the real numbers reported by the performance counters
on an Athlon. However, Cachegrind certainly makes a number of
approximations (see section 3.3.7 of
http://www.valgrind.org/docs/phd2004.pdf) and so you shouldn't trust it
too much. It should give reasonable numbers though.