Re: [Gomp-discuss] Re: Implementing OpenMP pragmas for the C frontend

[Steven Bosscher]

Op ma 10-02-2003, om 09:48 schreef Lars Segerlund:
> > 
> > This is is not to be taken as official - it is just my
> > opinion, which is uninformed (and I don't have time enough
> > to learn very much about it).
> 
>   Perhaps you should take the time to get informed, before wasting other 
> peoples time and perhaps consideer that some of these people are quite 
> well informed, also conisder that here are people willing to make an 
> effort to support 'multi threading' or 'native parallallism' in gcc, 
> which if you had a look at most modern systems might be of major 
> importance to gcc and the free community.

I wouldn't have put it in such strong words ;-)

Anyway, PB said that his "rebuttable assumption" is that
most members of the committee were numerical people or hardware
people.  I don't know if that's true or not, but they were people with
knowledge of high performance computing, and they knew at least
something about compiler technology.  

Maybe the resulting design is less than optimal from a language-design
point of view.  To me, it's just an easy interface to make my numerical
applications run fast on my SMP box without significant changes to my
code.

Fact is that GCC is not a high-performance compiler for numerical
applications: It does not do autovectorizing, and it's not easy to
create parallel programs, etc.  Result is that when I compile a simple
1D Riemann solver with Intel C++, it runs more than twice as fast on my
dual-P4 compared with GCC.  Differences: Vectorizing compiler with
OpenMP support.

Maybe one of the reasons for this is because people involved in GCC
development mostly are computer scientists, and that such people are not
well known for understanding computational scientists?

The picture at for example SGI or Fujitsu is probably different.  With
such companies, optimal performance of numerical applications is the
primary goal, so you'd expect to see numerical experts in the compiler
team as well.
And those numerical experts saw a trend: Everybody goes multi-processor!
So they identified a *need* for an easy-to-use interface to create
explicitly parallel software for people who are not computer
scientists.  OpenMP was born, and is now widely used and well
established.

If GCC does not want to be useful in a hpc environment, that's fine. 
But I would like to hear that *before* we start an effort to implement
OpenMP in GCC.  If the GCC community has an attitude like, "just put it
in a branch and we'll see what we do with it," then I can think of more
important things that I actually should put time in.

Greetz
Steven



>   Perhaps something as simple as reading openmp's homepage should give a 
> hint as to which institutions have contributed to the standard, ( which 
> I believe is quite minimalistic and good as far as standards go ).
>   You would discover that a lot of people working on scientific 
> computing have been involved, and they can not all be idiots , ( exept 
> Meee ! ;-) ).
> 
>   Lighten up, more speed is good news !
> 
>   / cheerio, Lars Segerlund
> 
> 
> 
> _______________________________________________
> Gomp-discuss mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/gomp-discuss