[Bug tree-optimization/30735] 50% slow down due to mem-ssa merge

2007-02-08 Thread hjl at lucon dot org 


--- Comment #1 from hjl at lucon dot org  2007-02-08 17:01 ---
Created an attachment (id=13021)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13021&action=view)
A testcase

# /usr/gcc-good/bin/gfortran -v
Using built-in specs.
Target: i686-pc-linux-gnu
Configured with: /net/gnu-13/export/gnu/src/gcc/gcc/configure
--enable-clocale=gnu --with-system-zlib --with-demangler-in-ld --enable-shared
--enable-threads=posix --enable-haifa --enable-checking=assert
--prefix=/usr/gcc-4.3 --with-local-prefix=/usr/local
Thread model: posix
gcc version 4.3.0 20061211 (experimental) [trunk revision 119728]
# /usr/gcc-4.3/bin/gfortran -v
Using built-in specs.
Target: i686-pc-linux-gnu
Configured with: /net/gnu-13/export/gnu/src/gcc/gcc/configure
--enable-clocale=gnu --with-system-zlib --enable-decimal-float=yes
--with-demangler-in-ld --enable-shared --enable-threads=posix --enable-haifa
--enable-checking=assert --prefix=/usr/gcc-4.3 --with-local-prefix=/usr/local
Thread model: posix
gcc version 4.3.0 20070206 (experimental) [trunk revision 121636]
# make
/usr/gcc-good/bin/gfortran -c -O2 -ffast-math -static -o old.o e_c3d.f
/usr/gcc-4.3/bin/gfortran -O2 -ffast-math -static  -c -o tr535.o tr535.f
/usr/gcc-4.3/bin/gfortran -O2 -ffast-math -static -o old old.o tr535.o
/usr/gcc-4.3/bin/gfortran -c -O2 -ffast-math -static -o new.o e_c3d.f
/usr/gcc-4.3/bin/gfortran -O2 -ffast-math -static -o new new.o tr535.o
time ./old
0.03user 0.00system 0:00.03elapsed 100%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (0major+82minor)pagefaults 0swaps
time ./new
0.10user 0.00system 0:00.10elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (0major+83minor)pagefaults 0swaps


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30735

[Bug tree-optimization/30735] 50% slow down due to mem-ssa merge

2007-02-08 Thread rguenth at gcc dot gnu dot org 


--- Comment #2 from rguenth at gcc dot gnu dot org  2007-02-08 17:07 ---
Try one of the partitioning patches I proposed.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30735

[Bug tree-optimization/30735] 50% slow down due to mem-ssa merge

2007-02-08 Thread hjl at lucon dot org 


--- Comment #3 from hjl at lucon dot org  2007-02-08 17:13 ---
(In reply to comment #2)
> Try one of the partitioning patches I proposed.
> 

Which one? Please provide a link or attach it here. Thanks.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30735

[Bug tree-optimization/30735] 50% slow down due to mem-ssa merge

2007-02-08 Thread dnovillo at gcc dot gnu dot org 


-- 

dnovillo at gcc dot gnu dot org changed:

   What|Removed |Added

 AssignedTo|unassigned at gcc dot gnu   |dnovillo at gcc dot gnu dot
   |dot org |org
 Status|UNCONFIRMED |ASSIGNED
 Ever Confirmed|0   |1
   Last reconfirmed|-00-00 00:00:00 |2007-02-08 18:42:31
   date||


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30735

[Bug tree-optimization/30735] 50% slow down due to mem-ssa merge

2007-02-12 Thread dnovillo at gcc dot gnu dot org 


--- Comment #4 from dnovillo at gcc dot gnu dot org  2007-02-13 00:59 
---

I have now reproduced this locally and I'm working on a fix.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30735

[Bug tree-optimization/30735] 50% slow down due to mem-ssa merge

2007-02-15 Thread pthaugen at us dot ibm dot com 


--- Comment #5 from pthaugen at us dot ibm dot com  2007-02-15 23:36 ---
Seeing similar behavior on PPC for CPU2000. Comparing revisions 119759 and
119760, 179.art degrades 24% and 200.sixtrack degrades 50%.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30735