This is the mail archive of the gcc-patches@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]
Other format: [Raw text]

Re: Regressions g++.dg/init/new19.C g++.dg/tree-ssa/pr27090.C due to your fix for 32540 and 31651


Actually, I doubt they should be xfailed, my guess is that PRE now
does the optimization it was looking for to happen in LIM (for new19),
and that the code just looks slightly different for pr27090.C
I'll look into it.


On 6/30/07, Hans-Peter Nilsson <hans-peter.nilsson@axis.com> wrote:
> Date: Sat, 30 Jun 2007 14:15:27 -0000
> From: dberlin@gcc.gnu.org

gcc/ChangeLog
>       Fix PR tree-optimization/32540
>       Fix PR tree-optimization/31651

The change above is in the range 126148:126153, and is the main
suspect for being the cause of these regressions:

g++.sum g++.dg/init/new19.C
g++.sum g++.dg/tree-ssa/pr27090.C

g++.log says:

Executing on host: /tmp/hpautotest-gcc1/cris-elf/gccobj/gcc/testsuite/g++/../../g++ -B/tmp/hpautotest-gcc1/cris-elf/gccobj/gcc/testsuite/g++/../../ /tmp/hpautotest-gcc1/gcc/gcc/testsuite/g++.dg/init/new19.C  -nostdinc++ -I/tmp/hpautotest-gcc1/cris-elf/gccobj/cris-elf/libstdc++-v3/include/cris-elf -I/tmp/hpautotest-gcc1/cris-elf/gccobj/cris-elf/libstdc++-v3/include -I/tmp/hpautotest-gcc1/gcc/libstdc++-v3/libsupc++ -I/tmp/hpautotest-gcc1/gcc/libstdc++-v3/include/backward -I/tmp/hpautotest-gcc1/gcc/libstdc++-v3/testsuite/util -fmessage-length=0  -O2 -fstrict-aliasing -fdump-tree-lim-details  -S   -isystem /tmp/hpautotest-gcc1/cris-elf/gccobj/cris-elf/./newlib/targ-include -isystem /tmp/hpautotest-gcc1/gcc/newlib/libc/include  -o new19.s    (timeout = 300)
PASS: g++.dg/init/new19.C (test for excess errors)
FAIL: g++.dg/init/new19.C scan-tree-dump Moving statement\n.*->ai\[0\];\n.*out of loop

Executing on host: /tmp/hpautotest-gcc1/cris-elf/gccobj/gcc/testsuite/g++/../../g++ -B/tmp/hpautotest-gcc1/cris-elf/gccobj/gcc/testsuite/g++/../../ /tmp/hpautotest-gcc1/gcc/gcc/testsuite/g++.dg/tree-ssa/pr27090.C  -nostdinc++ -I/tmp/hpautotest-gcc1/cris-elf/gccobj/cris-elf/libstdc++-v3/include/cris-elf -I/tmp/hpautotest-gcc1/cris-elf/gccobj/cris-elf/libstdc++-v3/include -I/tmp/hpautotest-gcc1/gcc/libstdc++-v3/libsupc++ -I/tmp/hpautotest-gcc1/gcc/libstdc++-v3/include/backward -I/tmp/hpautotest-gcc1/gcc/libstdc++-v3/testsuite/util -fmessage-length=0  -O -fdump-tree-optimized  -S   -isystem /tmp/hpautotest-gcc1/cris-elf/gccobj/cris-elf/./newlib/targ-include -isystem /tmp/hpautotest-gcc1/gcc/newlib/libc/include  -o pr27090.s    (timeout = 300)
PASS: g++.dg/tree-ssa/pr27090.C (test for excess errors)
FAIL: g++.dg/tree-ssa/pr27090.C scan-tree-dump return f->x;

Presumably just a case of missing " { xfail *-*-* }"; I see
several other test-files had that added (but with no ChangeLog
entry, tsk tsk) and pinskia says on IRC that this was indeed the
case.  Not obvious enough for me to add them, though.

brgds, H-P



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