[kde-solaris] Re: fibonacci

[Sunil]

that's bizzare!! a performance regression in 3.4.2?

--- Stefan Teleman <steleman at nyc.rr.com> wrote:

>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 22:01:54][196]>> 
> /usr/local/bin/gcc -m32 -mtune=v9 -O3 fibonacci.c -o
> 
> fibonacci.gcc342.v9
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 22:04:28][197]>>  
> ./fibonacci.gcc342.v9 36
> ./fibonacci.gcc342.v9: calculated 10 fibonacci(36)
> in 19.000000 
> seconds
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 22:05:02][198]>> 
> ./fibonacci.gcc342.v9 36
> ./fibonacci.gcc342.v9: calculated 10 fibonacci(36)
> in 19.000000 
> seconds
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 22:07:36][199]>> 
> ./fibonacci.gcc342.v9 36
> ./fibonacci.gcc342.v9: calculated 10 fibonacci(36)
> in 19.000000 
> seconds
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 22:07:58][200]>> 
> 
> It's actually consistently (3 times in a row)
> significantly worse.
> 
> --Stefan
> 
> ------
> 
> On Friday 05 November 2004 21:58, Sunil wrote:
> > the same code and same compile arguments, I get
> around
> > 24% boost with gcc with -mtune=v9 instead of
> -mv8plus.
> > it is with gcc 3.3.4, gcc3.4.2 probably should do
> > better than that in bridging the gap. Can you
> please
> > post gcc-3.4.2 result with -mtune=v9?
> >
> > And I agree that it will still not be enough to
> > account for all of the difference in this
> benchmark.
> >
> > Thanks,
> > Sunil
> 
> -- 
> Stefan Teleman          'Nobody Expects the Spanish
> Inquisition'
> steleman at nyc.rr.com                          -Monty
> Python
> 
> 



		
__________________________________ 
Do you Yahoo!? 
Check out the new Yahoo! Front Page. 
www.yahoo.com