http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52839
--- Comment #15 from Alan Modra <amodra at gmail dot com> 2012-04-05 08:06:30 UTC --- Many hours later one of my 32-bit tests failed, but I'm relieved to say it was only the pthread_once bug. #0 0x0fbd839c in raise () from /lib/power7/libc.so.6 #1 0x0fbda034 in abort () from /lib/power7/libc.so.6 #2 0x0fd9f480 in uw_init_context_1 (context=context@entry=0xf2dce8f8, outer_cfa=outer_cfa@entry=0xf2dcecf0, outer_ra=0xff20f98) at /home/amodra/src/gcc-current/libgcc/unwind-dw2.c:1502 #3 0x0fd9fb7c in _Unwind_RaiseException (exc=0x1002c360) at /home/amodra/src/gcc-current/libgcc/unwind.inc:88 #4 0x0ff20f98 in __cxxabiv1::__cxa_throw (obj=0x1002c380, tinfo=<optimized out>, dest=<optimized out>) at /home/amodra/src/gcc-current/libstdc++-v3/libsupc++/eh_throw.cc:78 #5 0x10000d24 in std::tr1::__throw_bad_weak_ptr () at /home/amodra/build/gcc-current/powerpc-linux/libstdc++-v3/include/tr1/shared_ptr.h:76 #6 0x10000e9c in _M_add_ref_lock (this=0x10021f60) at /home/amodra/build/gcc-current/powerpc-linux/libstdc++-v3/include/tr1/shared_ptr.h:244 #7 __shared_count (__r=..., this=<synthetic pointer>) at /home/amodra/build/gcc-current/powerpc-linux/libstdc++-v3/include/tr1/shared_ptr.h:494 #8 __shared_ptr<A> (__r=..., this=<synthetic pointer>) at /home/amodra/build/gcc-current/powerpc-linux/libstdc++-v3/include/tr1/shared_ptr.h:586 #9 shared_ptr<A> (__r=..., this=<synthetic pointer>) at /home/amodra/build/gcc-current/powerpc-linux/libstdc++-v3/include/tr1/shared_ptr.h:1015 #10 thread_hammer (opaque_weak=0x1002db44) at /home/amodra/src/gcc-current/libstdc++-v3/testsuite/tr1/2_general_utilities/shared_ptr/thread/default_weaktoshared.cc:132 #11 0x0fd5749c in start_thread () from /lib/power7/libpthread.so.0 #12 0x0fc92160 in clone () from /lib/power7/libc.so.6
