[patch] Merge to trunk from Graphite branch

[Toon Moene]

[ Somewhat belatedly, as I was in Oslo for the work-for-hire
   experience (meteorology) last week ]

Tobias Burnus wrote:

> On 03/15/2010 04:12 PM, Jack Howarth wrote:

>> On Mon, Mar 15, 2010 at 03:11:21PM +0100, Richard Guenther wrote:
>>> Updating the default x86 arch does bring you nothing (well, if
>>> you're not clueless in case you shouldn't build GCC yourself).
>>    I mentioned that because of the 2% improvement in the Polyhedron
>> 2005 benchmarks which coincided with the x86 arch changes which
>> enabled -msse2 across the board...
>>
>> http://gcc.gnu.org/ml/gcc/2010-03/msg00075.html
> 
> I somehow doubt that the changing of the default has anything to do with
> it. I could well imagine that Toon uses -march=native or similar.

I changed to using --with-arch-64=native --with-tune-64=native instead 
of --with-arch-64=core2 --with-tune-64=core2 after the default was 
changed.  Before that the --with-{blah}=native wasn't accepted.

This of course (I hope :-) affected the build of the various run-time 
libraries and hence the performance of code ...

At least, that was what *I* thought had brought me 2 % performance 
improvement.

But as Richard Guenther writes: Really figuring out what 
enables/inhibits performance improvements is *hard work*, and has to be 
balanced against other hard work (like fixing P1 bugs :-)

[ As an aside: the representative of Intel on the Fortran
   Standardization Committee was aghast that the Polyhedron suite
   "suddenly" started to be important in evaluating the performance of
   compiler-generated code, because that meant they had to track *yet
   another* suite of programs for performance regressions.

   You know, I agreed ... ]

Cheers,

-- 
Toon Moene - e-mail: toon@moene.org - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
At home: http://moene.org/~toon/
Progress of GNU Fortran: http://gcc.gnu.org/gcc-4.5/changes.html