[Bug other/25527] [gomp] segfault in main.omp_fn.0 with -O2

2006-01-27 Thread reichelt at gcc dot gnu dot org 


--- Comment #5 from reichelt at gcc dot gnu dot org  2006-01-27 16:25 
---
I can reproduce the problem on x86_64-unknown-linux-gnu
with the following testcase:

===
#includeiostream
#includecomplex

int main()
{
  std::complexint c;
  std::cout  c  std::endl;
}
===

It crashes when compiled with -fopenmp -O -m32,
but runs fine if I leave out any of these options.


-- 

reichelt at gcc dot gnu dot org changed:

   What|Removed |Added

 CC||reichelt at gcc dot gnu dot
   ||org
   Keywords||monitored, wrong-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25527

[Bug other/25527] [gomp] segfault in main.omp_fn.0 with -O2

2006-01-27 Thread reichelt at gcc dot gnu dot org 


--- Comment #6 from reichelt at gcc dot gnu dot org  2006-01-27 16:47 
---
Even shorter testcase:

=
#includesstream

int main()
{
std::ostringstream s;
return 0;
}
=


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25527

[Bug other/25527] [gomp] segfault in main.omp_fn.0 with -O2

2005-12-30 Thread dev at stuffit dot at 


--- Comment #4 from dev at stuffit dot at  2005-12-30 20:05 ---
gdb backtrace:

(gdb) run
Starting program: /home/dev2/tmp/test
[Thread debugging using libthread_db enabled]
[New Thread -1210665248 (LWP 8209)]
OPENMP Thread 0 finished
RHS 0

Program received signal SIGILL, Illegal instruction.
[Switching to Thread -1210665248 (LWP 8209)]
0xb7fc791d in typeinfo for std::locale::facet () from
/home/dev2/gcc-gomp/gomp-gcc-build/lib/libstdc++.so.6
(gdb) backtrace
#0  0xb7fc791d in typeinfo for std::locale::facet () from
/home/dev2/gcc-gomp/gomp-gcc-build/lib/libstdc++.so.6
#1  0x08049763 in std::operator double, char, std::char_traitschar  ()
#2  0x08049309 in main.omp_fn.0 ()
#3  0x080495c7 in main ()


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25527

[Bug other/25527] [gomp] segfault in main.omp_fn.0 with -O2

2005-12-29 Thread jakub at gcc dot gnu dot org 


--- Comment #3 from jakub at gcc dot gnu dot org  2005-12-29 10:04 ---
Couldn't reproduce, on neither 4way nor UP box, with various OMP_NUM_THREADS
settings in environment.  The valgrind output isn't really helpful, because
when it reaches unhandled insn, all bets are off what happens afterwards and
also because you must have libstdc++ stripped as it isn't showing location
within libstdc++ where it faulted.
You should instead run it under gdb and get a backtrace from it.


-- 

jakub at gcc dot gnu dot org changed:

   What|Removed |Added

 Status|UNCONFIRMED |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25527

[Bug other/25527] [gomp] segfault in main.omp_fn.0 with -O2

2005-12-21 Thread dev at stuffit dot at 


--- Comment #1 from dev at stuffit dot at  2005-12-21 23:55 ---
Created an attachment (id=10547)
 -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=10547action=view)
Sample Code


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25527

[Bug other/25527] [gomp] segfault in main.omp_fn.0 with -O2

2005-12-21 Thread dev at stuffit dot at 


--- Comment #2 from dev at stuffit dot at  2005-12-21 23:59 ---
svn revision 108861 of gomp-20050608-branch, i should probably add!


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25527