This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Long term code compactness regression


>> Jan, sorry for my pedantry, but I have ready at hand a very short and
basic
>> c++ testcode (reduced from the Haney tests) which today is 2 times
(!) slower
>> than one week ago on i686 (-O2)... Do you have time to look a little
bit at
>> it? Here it is (PR4714 for some more details):
>Sure!
>I will try to figure out what has changed.
>
>Thanks a lot for nice testcase.  I need to read gcc-bugs more
curefully.
>Honza

Very good!
Now that the annoying regression for Intel ia32 in Oopack's Max  is
finally solved (thanks again to you and Richard Henderson for the
beautiful work) it would be nice if GCC3.1 could show also nice Haney
numbers. For sure it has the potential for this since the "vector" test
has improved recently and both the "Complex Matrix" and "Real Matrix"
tests were very good until June.

I noticed that indeed something interesting happened on 23-24 October:
look at Andreas Jaeger's SPECfp2000 plots, in particular 179.art and
183.equake.

Cheers,
Paolo.



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]