Stacey and I have been fighting with pythn27 building ports for
ARM(11-current), and it seems we have a detectable regression in unwind.

oot@beaglebone:/usr/src/tools/regression/pthread/unwind # make
c++ -o main_thread_exit main_thread_exit.cpp -lpthread
c++ -o sem_wait_cancel sem_wait_cancel.cpp -lpthread
c++ -o cond_wait_cancel cond_wait_cancel.cpp -lpthread
c++ -o cond_wait_cancel2 cond_wait_cancel2.cpp -lpthread
c++ -o catch_pthread_exit catch_pthread_exit.cpp -lpthread
root@beaglebone:/usr/src/tools/regression/pthread/unwind # ./main_thread_exit 
Test::Test()
Bug, object destructor is not called
root@beaglebone:/usr/src/tools/regression/pthread/unwind # ./sem_wait_cancel
Test::Test()
Bug, object destructor is not called
root@beaglebone:/usr/src/tools/regression/pthread/unwind # ./cond_wait_cancel
Test::Test()
Bug, object destructor is not called
root@beaglebone:/usr/src/tools/regression/pthread/unwind # ./cond_wait_cancel2
Test::Test()
cleanup_handler()
Bug, object destructor is not called
root@beaglebone:/usr/src/tools/regression/pthread/unwind # ./catch_pthread_exit 
failure
root@beaglebone:/usr/src/tools/regression/pthread/unwind # 

Attachment: signature.asc
Description: This is a digitally signed message part

Reply via email to