C Optimization Tests, 1 May 2004, tree-ssa/3.5/3.4/icc

H. J. Lu hjl@lucon.org
Sun May 2 20:39:00 GMT 2004


On Sun, May 02, 2004 at 04:00:03PM -0400, Scott Robert Ladd wrote:
> Kaveh R. Ghazi wrote:
> >Scott -- I'm curious how mainline performs on all these tests with the
> >patch below.  Would you please try it out?  If we can demonstrate an
> >overall win, I'll submit it for installation.
> 
> GCC Mainline, Pentium 4
> 
>          default math     with -mfpmath=sse
> test  no patch  patched   no patch  patched
> ----  --------  --------  --------  --------
> alma    64.3      42.4      68.2      44.7
>  evo    52.9      52.8      79.5      79.8
>  fft    34.6      32.0      34.5      33.4
> huff    16.0      16.0      16.7      16.3
>  lin    26.2      22.9      25.9      23.7
> mat1     7.5       7.5       8.0       7.9
> mole    32.8      31.3      30.6      22.4
> tree    25.0      24.5      25.7      25.6
>       --------  --------  --------  --------
> total  259.3     229.5     289.0     253.9
> 
> I'd say your patch is a winner, at least as far as the Pentium 4 is 
> concerned. On the Opteron, your patch (and -mfpmath=sse, BTW) made no 
> difference in generated code speed.
> 
> I'm certain other people will notice the -mfpmath is pessimistic on the 
> Pentium 4. Being somewhat pressed for time today, I haven't figured out 
> why yet,
> 

Have you added -march=pentium4 to -mfpmath?


H.J.



More information about the Gcc mailing list