[kde-solaris] Re: fibonacci
Sunil
funtoos at yahoo.com
Sat Nov 6 03:58:47 CET 2004
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 <steleman at nyc.rr.com> wrote:
> On Friday 05 November 2004 08:07, Eva Brucherseifer
> wrote:
> > ... which doesn't really give a good number ;-)
> > CPU architectures and compilers give very
> different time results,
> > if it comes to integer vs. float calculation. KDE
> is a gui and thus
> > you mainly have integer calculation - here
> calculating PI certainly
> > isn't a good benchmark. And don't ask me for a
> proposal an how to
> > measure speed on KDE...
> >
> > Greetings,
> > eva
>
> ------
>
> :-)
>
> No doubles, no floating point:
>
> /* fibonacci.c */
>
> #include <stdio.h>
> #include <math.h>
> #include <time.h>
>
> static const unsigned int iterations = 10;
>
> unsigned int fibonacci (unsigned int n)
> {
> return ((n < 2) ? 1 : fibonacci(n-1) +
> fibonacci(n-2));
> }
>
> int
> main (int argc, char* argv[])
> {
> int i = 0;
> unsigned int j = 0;
> unsigned int result;
> clock_t start;
> clock_t end;
> double elapsed;
>
> if (argc != 2)
> {
> (void) fprintf (stderr, "Usage: %s
> <fibonacci>\n",
> argv[0]);
> return (-1);
> }
> i = atoi(argv[1]);
> start = clock();
> for (j = 0; j < iterations; j++)
> {
> result = fibonacci((unsigned int)
> i);
> }
>
> end = clock();
> elapsed = ((double) (end / CLOCKS_PER_SEC) -
> (double) (start /
> CLOCKS_PER_SEC));
> (void) fprintf (stderr, "%s: calculated %u
> fibonacci(%ld) in
> %lf seconds\n", argv[0], iterations, i, elapsed);
> return (0);
> }
>
> /* fibonacci.c */
>
> Downloaded from sunfreeware: GCC 3.4.2
>
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:46:08][178]>>
> /usr/local/bin/gcc --version
> gcc (GCC) 3.4.2
> Copyright (C) 2004 Free Software Foundation, Inc.
> Copyright (C) 2004 Free Software Foundation, Inc.
> This is free software; see the source for copying
> conditions. There
> is NO warranty; not even for MERCHANTABILITY or
> FITNESS FOR A
> PARTICULAR PURPOSE.
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:46:08][179]>>
>
> [ ... figure out gcc 3.4.2's flags ... ]
>
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:49:41][182]>>
> /usr/local/bin/gcc -m32 -mv8plus -O3 fibonacci.c -o
> fibonacci.gcc342
> -lc -lm
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:49:51][183]>>
> ./fibonacci.gcc342 36
> ./fibonacci.gcc342: calculated 10 fibonacci(36) in
> 17.000000 seconds
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:50:48][184]>>
> source .cflags
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:51:23][185]>>
> echo $CC
> /opt/Forte-9.0/SUNWspro/prod/bin/cc
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:51:36][186]>>
> echo $CFLAGS
> -erroff=%all -errshort=full -errfmt=error
> -errwarn=%none -xO4 -s
> -xc99=%all,no%lib -xjobs=2 -xtarget=ultra2 -dalign
> -xprefetch=auto
> -xprefetch_level=3 -xbuiltin=%all
> -xregs=appl,no%float
> -xcache=32/64/4:2048/64/1 -xcsi -xcode=pic32
> -xinline=%auto
> -xustr=ascii_utf16_ushort -xvis=yes -z now -z rescan
> -z absexec
> -xildoff -xldscope=symbolic -xpagesize=default
> -xF=%all
> -xalias_level=std -xmemalign=8i -xsafe=mem
> -xthreadvar -lpthread
> -lposix4 -lrt -mt -D_REENTRANT -D__EXTENSIONS__
> -D_XPG4_2 -DSOLARIS
> -DSOLARIS9 -DUSE_SOLARIS -DNDEBUG -KPIC
> -xarch=v8plusa
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:51:38][187]>>
> $CC $CFLAGS fibonacci.c -o fibonacci.forte9 -lc -lm
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:53:11][188]>>
> > ./fibonacci.forte9 36
> ./fibonacci.forte9: calculated 10 fibonacci(36) in
> 9.000000 seconds
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:53:51][189]>>
> uname -a
> SunOS obiwan 5.9 Generic_117171-05 sun4u sparc
> SUNW,Ultra-60
>
[steleman at obiwan][~/programming/fibonacci][11/05/2004
> 19:59:28][190]>>
>
> --Stefan
>
> --
> Stefan Teleman 'Nobody Expects the Spanish
> Inquisition'
> steleman at nyc.rr.com -Monty
> Python
>
> ___________________________________________________
> This message is from the kde-solaris mailing list.
> Account management:
> https://mail.kde.org/mailman/listinfo/kde-solaris.
> Archives: http://lists.kde.org/.
> More info: http://www.kde.org/faq.html.
>
__________________________________
Do you Yahoo!?
Check out the new Yahoo! Front Page.
www.yahoo.com
More information about the kde-solaris
mailing list