Re: Comparison of GCC-4.9 and LLVM-3.4 performance on SPECInt2000 for x86-64 and ARM

[Vladimir Makarov <vmakarov at redhat dot com>]

On 06/24/2014 10:42 AM, Renato Golin wrote:
> On 24 June 2014 15:11, Vladimir Makarov <vmakarov@redhat.com> 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
>>
>> http://vmakarov.fedorapeople.org/spec/
>>
>>   The comparison is achievable from 2014 link and links under it in
>> the left frame.
> Hi Vladimir,
>
> Nice comparison!
>
> It's on the same ballpark of my own findings (with SPEC and other
> benchmarks) on ARM: +10% performance, but also code size for GCC over
> LLVM, and massive compilation time savings on LLVM.
>
> I wonder how much of that is due to auto-vectorization (on LLVM, -O2+
> turns it on, I suppose GCC is only on -O3?). From Ramana's point,
> there may be nothing serious if you haven't enabled NEON, though.
I guess eon could benefit from autovectorization a lot.  May be crafty
too.  I don't think it is true for other SPECint benchmarks.
> Also interesting to see the impact of LTO being a major drive in
> recent performance improvements on both compilers.
>
I did not manage to use LTO for LLVM as it requires GOLD and GOLD
crashes on my machine (even with fresher built versions).

Comparison and benchmarking is really time consuming even on x86-64 as
you need to solve wrong code generation (e.g. LLVM always generates
wrong code for gap with any optimization options).  So I gave up to make
LTO working with LLVM on ARM.