This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
vortex peak performance regression
- From: Jan Hubicka <jh at suse dot cz>
- To: hzoli at hzoli dot 2y dot net, gcc at gcc dot gnu dot org
- Date: Wed, 2 Jan 2002 13:05:43 +0100
- Subject: vortex peak performance regression
Zoltan,
It looks like your patch:
+ 2001-11-30 Zoltan Hidvegi <hzoli@hzoli.2y.net>
+
+ * unroll.c (unroll_loop): Correct special exit cases.
+
causes regression in the spec2000 tester - most importantly the vortex
benchmark (- see www.suse.de/~aj/SPEC
It is not obvious for me why, but can you please double check
whether the regression is unavoidable.
The patches from same testing that does not look like real cause are:
+ 2001-11-30 Stephane Carrez <Stephane.Carrez@sun.com>
+
+ * config/sparc/sparc.c (DF_MODES_NO_S): Fix pr/3623, define to
+ accept DFmode and DImode only.
+
+ 2001-11-30 Aldy Hernandez <aldyh@redhat.com>
+
+ * config/rs6000/rs6000.h (STARTING_FRAME_OFFSET): Adjust for altivec.
+ (STACK_DYNAMIC_OFFSET): Same.
+
+ 2001-11-30 Richard Henderson <rth@redhat.com>
+
+ * c-decl.c (duplicate_decls): Copy DECL_SAVED_TREE.
+
+ 2001-11-30 Aldy Hernandez <aldyh@redhat.com>
+
+ * config/rs6000/rs6000.h (CALL_REALLY_USED_REGISTERS): New.
+ (CONDITIONAL_REGISTER_USAGE): Set call_really_used_registers to
+ the values in call_used_registers.
+
+ Fri Nov 30 12:48:26 2001 Jeffrey A Law (law@cygnus.com)
+
+ * gthr-dce.h: If _DCE_THREADS is not defined, then just include
+ gthr-single.
Honza