[Numpy-discussion] String sort

[Bruce Southey]

Hi,
I added gcc 4.2 from the openSUSE 10.1 repository so I now have both
the 4.1.2 and 4.2.1 compilers installed. But still have glibc-2.4-31.1
installed. I see your result with 4.2.1 but not with 4.1.2 so I think
that there could be a difference in the compiler flags. I don't know
enough about those to help but I can test any suggestions.

$ gcc --version
gcc (GCC) 4.1.2 20070115 (prerelease) (SUSE Linux)
$ gcc -O3 sort-string-bench.c -o sort412
$ ./sort412
Benchmark with 1000000 strings of size 15
C qsort with C style compare: 0.630000
C qsort with Python style compare: 0.640000
NumPy newqsort: 0.360000

$ gcc-4.2 --version
gcc-4.2 (GCC) 4.2.1 (SUSE Linux)
$ gcc-4.2 -O3 sort-string-bench.c -o sort421
$ ./sort421
Benchmark with 1000000 strings of size 15
C qsort with C style compare: 0.620000
C qsort with Python style compare: 0.610000
NumPy newqsort: 0.550000

This is  the same as:
$ gcc-4.2 -O2 -finline-functions sort-string-bench.c -o sort421
$ ./sort421
Benchmark with 1000000 strings of size 15
C qsort with C style compare: 0.710000
C qsort with Python style compare: 0.700000
NumPy newqsort: 0.550000

(NumPy newqsort with -O2 alone is 0.60000)

For completeness, 4.1.2 using '-O2' versus '-O2 -finline-functions' is
NumPy newqsort: 0.620000 vs NumPy newqsort: 0.500000

Regards
Bruce


On Feb 13, 2008 4:35 AM, Francesc Altet <faltet at carabos.com> wrote:
> A Tuesday 12 February 2008, Bruce Southey escrigué:
> > Hi,
> >
> > I have a Opteron 248 (2.66GHz) that with gcc 4.1.0 (SUSE10.1?) that
> > gives C qsort with C style compare: 0.650000
> > C qsort with Python style compare: 0.640000
> > NumPy newqsort: 0.360000
>
> That's very intersting.  In a similar configuration, but using SuSe 10.3
> (Enterprise) instead of 10.1, I don't see this factor of almost 2 of
> difference in performance (in fact, both performances, C qsort and
> NumPy newqsort, are very similar).
>
> This seems to confirm that the GNU glibc crew has fixed the qsort
> performance very recently (i.e. I hope it is not only a fix in SuSe
> 10.3 Enterprise), and this is why most of current distros are seeing
> the poor performance in qsort.
>
>
> Cheers,
>
> --
> >0,0<   Francesc Altet     http://www.carabos.com/
> V   V   Cárabos Coop. V.   Enjoy Data
>  "-"
> _______________________________________________
> Numpy-discussion mailing list
> Numpy-discussion at scipy.org
> http://projects.scipy.org/mailman/listinfo/numpy-discussion
>