|
Home > Archive > Unix Programming > October 2006 > Looking for C Compiler Performance Info
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Looking for C Compiler Performance Info
|
|
| lxrocks 2006-10-13, 7:33 am |
| I am looking for comparitive performance information for C-Compilers.
More especially in the area of Float/Integer computes.
Any information would be appreciated.
tks
| |
| Rick Jones 2006-10-13, 7:46 pm |
| lxrocks <lxrocks@hotmail.com> wrote:
> I am looking for comparitive performance information for C-Compilers.
> More especially in the area of Float/Integer computes.
performance measured how? time to compile? speed of resulting code
regardless of compile time? ostensibly, with a large enough based on
submittals, one might make use of the SPECcpu benchmarks with the
results archived on www.spec.org. Not that there are always
overlapping sets of results posted where the hardware remains fixed
while the compiler changes, but there are some overlaps I believe.
rick jones
--
denial, anger, bargaining, depression, acceptance, rebirth...
where do you want to be today?
these opinions are mine, all mine; HP might not want them anyway... 
feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...
| |
| lxrocks 2006-10-14, 1:31 am |
| Rick Jones wrote:
> lxrocks <lxrocks@hotmail.com> wrote:
>
> performance measured how? time to compile? speed of resulting code
Apologies, I should have been clearer here. Compile time is not the
issue - I only am interested in resulting code.
> regardless of compile time? ostensibly, with a large enough based on
> submittals, one might make use of the SPECcpu benchmarks with the
> results archived on www.spec.org. Not that there are always
> overlapping sets of results posted where the hardware remains fixed
> while the compiler changes, but there are some overlaps I believe.
I presumed spec.org to be more about hardware than compiler, I had not
thought of checking to see if there were different compilers used for
the same hardware and compare the specCPU / FPU results - thats a good
point you make, I'll take a look.
In respect of float calculations, I have noticed that there some huge
performance differences for different compilers. For example Intel C is
much faster than GCC on Linux. In some cases Visual C on XP is faster
than GCC on Linux. Of course, my testing is probably faulted in many
ways and the environments less controlled in terms of testing. Which is
why I was looking for any info in this area, carried out by people with
more experience than me.
I know there was a legal battle some time back between AMD and Intel
where Intel were accused of compiling slower objects if they detected
AMD CPU's - which is kind of sneaky, but I don't know if that is still
relevant.
| |
|
|
|
|
| Mark Hobley 2006-10-17, 7:37 pm |
| lxrocks <lxrocks@hotmail.com> wrote:
> Apologies, I should have been clearer here. Compile time is not the
> issue - I only am interested in resulting code.
The size of the compiled binary may also be an issue to consider. I use Tiny C
Compiler.
http://fabrice.bellard.free.fr/tcc/
Regards,
Mark.
--
Mark Hobley
393 Quinton Road West
QUINTON
Birmingham
B32 1QE
Telephone: (0121) 247 1596
International: 0044 121 247 1596
Email: markhobley at hotpop dot donottypethisbit com
http://markhobley.yi.org/
|
|
|
|
|