Call for hardware (pld3.0 Wc)

[Arkadiusz Miskiewicz]

On Tuesday 15 of June 2004 14:04, Bartosz Taudul wrote:
> On Tue, Jun 15, 2004 at 01:56:28PM +0200, Artur Flinta wrote:
> > Zresztą, skoro --march=i686 -mtune=pentium4 działa na athlonach równie
> > szybko jak pakiety dedykowane to dla mnie wątek się zakończył.
>
> JEŻELI działa równie szybko. Na razie opieramy się na tym że ktośtam
> cośtam gdzieśtam kiedyśtam przeczytał.
Jakub Jelinek:
packages will be compiled
with -march{3,6}86 -mtune=pentium4, so although they will run on
i386 (resp. i686), they will be optimized for P4 (Athlons and newer AMD
chips run P4 optimized code without noticeable performance hit).

Jakub Jelinek:
> Is there difference betwwen -mtune=pentium4 and -mtune=athlon ?
> Pentium4 sounds intel-ish centric but what about AMD ?.

Yes, there is.  The thing is, Intel chips are far more sensitive to
instruction scheduling than AMD chips.
So, if you run code tuned for AMD on P4, the performance hit is big,
if you run P4 optimized code on AMD, the performance hit compared to
running AMD tuned code on AMD is really small.

 Arjan van de Ven o -mtune=pentium4 vs -mtune=athlon:
The biggest change is NOT using a few forms of instructions (INC/DEC
hurt p4's while ADD/SUB by one are fast), AMD cpus seem to not care
really which form is chosen, it seens AMD optimizes in the cpu beyond
this level. You'll be *really* hard pressed to find (artificial)
examples where this would slow down AMD cpus even 0.1%....

Alan Cox:
Likewise my own testing has always found that the Athlon really doesn't care
much how you order instructions. If you think about it AMD have spent years
dealing with everyone optimising for random intel processor of the year and
adapted appropriately. Except on things like 3dnow and prefetch stuff the
AMD really doesn't seem to care.


Testów sensownych nie widziałem tak jak i nie widziałem sensownych testów 
-march=i686 vs -march=athlon.

> wolf

-- 
Arkadiusz Miśkiewicz     CS at FoE, Wroclaw University of Technology
arekm.pld-linux.org, 1024/3DB19BBD, JID: arekm.jabber.org, PLD/Linux