With revision r149958 this test passed.
>From revision r149962 (i.e. the revision introducing the regression) and on,
this test has failed as follows:
Running /tmp/r149958-149962/gcc/gcc/testsuite/gcc.dg/dg.exp ...
...
FAIL: gcc.dg/pr28796-2.c execution test
With the message in the logfile being:
Executing on host: /tmp/r149958-149962/gccobj/gcc/xgcc
-B/tmp/r149958-149962/gccobj/gcc/ /tmp/r149958-149962/gcc/gcc/tes
tsuite/gcc.dg/pr28796-2.c -O2 -funsafe-math-optimizations
-fno-finite-math-only -DUNSAFE -isystem /tmp/r149958-14996
2/gccobj/cris-elf/./newlib/targ-include -isystem
/tmp/r149958-149962/gcc/newlib/libc/include -B/tmp/r149958-149962/gccob
j/cris-elf/./libgloss/cris/
-L/tmp/r149958-149962/gccobj/cris-elf/./libgloss/cris
-L/tmp/r149958-149962/gcc/libgloss/cri
s -B/tmp/r149958-149962/gccobj/cris-elf/./newlib/
-L/tmp/r149958-149962/gccobj/cris-elf/./newlib -sim3 -lm -o ./pr28
796-2.exe (timeout = 300)
PASS: gcc.dg/pr28796-2.c (test for excess errors)
program stopped with signal 6.
FAIL: gcc.dg/pr28796-2.c execution test
I'll attach a reduced test-case and a description of the wrong code.
Author of patch exposing the regression CC:ed.
--
Summary: [4.5 Regression]: cris-elf gcc.dg/pr28796-2.c
Product: gcc
Version: 4.5.0
Status: UNCONFIRMED
Keywords: wrong-code
Severity: normal
Priority: P3
Component: rtl-optimization
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: hp at gcc dot gnu dot org
GCC host triplet: x86_64-unknown-linux-gnu
GCC target triplet: cris-axis-elf
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41085
