In message <4ad44ab0.4050...@gmail.com>, Regis writes:
>
> Regis wrote:
> > regi...@apache.org wrote:
> >
> > This commit is incorrect and causes SelectorTest failed on Windows,
> > I'll fix it soon.
>
>
> Fixed at r824652. I found SelectorTest is in exclude list of linux,
> but it passes on my linux machine. It excluded in HARMONY-4116, does
> anyone can reproduce this issue?
I ran the test 100 times on linux/x86_64 with no failures.
Wondering if it was a port selection problem (it is using the
doomed-to-intermittent-failure Support_PortManager), I ran 100 times in
parallel on two sessions on linux/x86. I saw no failures. If I'd have
stopped here I'd probably have been happy. But a further 100 runs on
x86 on debian oldstable (etch) had one crash out of 100 runs with the
following stack trace:
Program terminated with signal 11, Segmentation fault.
#0 0xf7f986e7 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2
(gdb) bt
#0 0xf7f986e7 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2
#1 0xf7f98b07 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2
#2 0xf7f9c2f9 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2
#3 0xf7f9c090 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2
#4 0xf7f4a7f1 in port_thread_set_priority (os_thread=3592702896, priority=5)
at
/home/hindessm/hy/fed.x86/working_vm/vm/port/src/thread/linux/thread_os.c:519
#5 0xf7f47ba1 in hythread_set_priority (thread=0x8f32ae8, priority=5)
at
/home/hindessm/hy/fed.x86/working_vm/vm/thread/src/thread_native_attrs.c:41
#6 0xf7f44a1b in hythread_wrapper_start_proc (arg=0x8f34c40)
at
/home/hindessm/hy/fed.x86/working_vm/vm/thread/src/thread_native_basic.c:804
#7 0xf7f49d0c in thread_start_func (arg=0x8f34c40)
at
/home/hindessm/hy/fed.x86/working_vm/vm/port/src/thread/linux/thread_os.c:106
#8 0xf7cc20bd in start_thread () from /lib/tls/libpthread.so.0
#9 0xf7eb201e in clone () from /lib/tls/libc.so.6
This is unlikely to relate to the HARMONY-4116 issue.
Regards,
-Mark.
