This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: 10% compile time decrease!
- From: Andrew Pinski <pinskia at physics dot uc dot edu>
- To: Richard Guenther <rguenth at tat dot physik dot uni-tuebingen dot de>
- Cc: gcc at gcc dot gnu dot org, Andrew Pinski <pinskia at physics dot uc dot edu>
- Date: Thu, 3 Jun 2004 07:57:59 -0400
- Subject: Re: 10% compile time decrease!
- References: <Pine.LNX.4.53.0406031331580.2199@bellatrix.tat.physik.uni-tuebingen.de>
On Jun 3, 2004, at 07:35, Richard Guenther wrote:
Hi!
Today the automatic POOMA tramp3d-v3 tester caught a 10% compile time
decrease (congrats!) for the leafified version on ia64-linux! This
brings us halfway back to the compile time performance of mainline
before the tree-ssa merge.
Also for the first time since the tree-ssa merge, the leafified test
didn't segfault and its performance is on-par with before the tree-ssa
merge. This is great.
The tester is at
http://www.tat.physik.uni-tuebingen.de/~rguenth/gcc/monitor-
summary.html
In case you wondering this is all (including having leafified working)
caused by my little patch to tree-ssa.c:
2004-06-02 Andrew Pinski <pinskia@physics.uc.edu>
PR tree-optimization/14042
PR tree-optimization/14729
PR tree-optimization/14736
* tree-ssa.c (tree_ssa_useless_type_conversion_1):
Check the type which the pointer points to
instead of the pointer types.
Thanks,
Andrew Pinski