This is the mail archive of the
mailing list for the GCC project.
Re: Comparison of GCC-4.9 and LLVM-3.4 performance on SPECInt2000 for x86-64 and ARM
- From: Vladimir Makarov <vmakarov at redhat dot com>
- To: Ramana Radhakrishnan <ramana dot radhakrishnan at arm dot com>, "gcc.gcc.gnu.org" <gcc at gcc dot gnu dot org>
- Date: Tue, 24 Jun 2014 10:45:30 -0400
- Subject: Re: Comparison of GCC-4.9 and LLVM-3.4 performance on SPECInt2000 for x86-64 and ARM
- Authentication-results: sourceware.org; auth=none
- References: <53A98705 dot 10909 at redhat dot com> <53A98CE2 dot 9080108 at arm dot com>
On 06/24/2014 10:36 AM, Ramana Radhakrishnan wrote:
> On 24/06/14 15:11, Vladimir Makarov wrote:
>> A few people asked me about new performance comparison of latest GCC
>> and LLVM. So I've finished it and put it on my site
>> The comparison is achievable from 2014 link and links under it in
>> the left frame.
>> These pages are also achievable as
>> http://vmakarov.fedorapeople.org/spec/2014/2014.html -- what I
>> compare, my motivation, major results, and my conclusions.
>> http://vmakarov.fedorapeople.org/spec/2014/llvmgcc64.html -- The
>> results of comparison on x86-64 SPECInt2000
>> http://vmakarov.fedorapeople.org/spec/2014/arm.html -- The
>> results of comparison on ARM SPECInt2000
> Interesting ...
> On ARM, I note that by default you are generating code for
> fpu=vfpv3-d16 and armv7-a. Of interest is the fact that you aren't
> using the Neon unit at all if you configure the compiler with
Unfortunately, that is the configuration I can use on my system because
of lack of libraries for other configurations.
I don't think Neon can improve score for SPECInt2000 significantly but
may be I am wrong.
> Did you add any other architecture specific options to your SPEC2k runs ?
No. The only options I used are -Ofast.
Could you recommend me what best options you think I should use for this
As I wrote, I am more interesting in aarch64 which can be used in a
server (as you know RedHat works on a server market) but unfortunately I
have no such machine for SPEC benchmarking.