Re: [liblo] recent fixes should be published, needs 64-bit testing

test_bidirectional_tcp seems to block forever on  Debian Wheezy x86_64:

./test_bidirectional_tcp
0xf33010.server fd: 3
0xf33010.receiving1..
0xf34720.sending thread
0xf34720.message sent
0xf34720.sending thread waiting
0xf33010./test, from TCP::
0xf33010.address fd: 0
0xf33010.reply sent
0xf33010.done receiving1
0xf33010.receiving2..

It's stuck in a poll loop:

poll([{fd=3, events=POLLIN|POLLPRI|POLLERR|POLLHUP}, {fd=6,
events=POLLIN|POLLPRI|POLLERR|POLLHUP}], 2, 10000) = 0 (Timeout)
poll([{fd=3, events=POLLIN|POLLPRI|POLLERR|POLLHUP}, {fd=6,
events=POLLIN|POLLPRI|POLLERR|POLLHUP}], 2, 10000) = 0 (Timeout)
poll([{fd=3, events=POLLIN|POLLPRI|POLLERR|POLLHUP}, {fd=6,
events=POLLIN|POLLPRI|POLLERR|POLLHUP}], 2, 10000


Simon


On Sun, Jun 23, 2013 at 9:36 AM, Stephen Sinclair <rad...@gm...> wrote:
> Hi,
>
> After my initial publication of 0.27, a couple of serious bugs in the
> TCP implementation were posted, which have now been fixed.
>
> I think it's worth publishing a bug-fix release just for this, before
> 0.27 gets into any linux distributions.  However, there was also a
> complaint on the LAD mailing list about "make test" failing for
> someone on an x86-64 Suse Linux machine.
>
> I don't know whether the recent fixes were related, so I'll contact
> the author and check with him whether it is now fixed.  However, can
> anyone confirm that "make test" is successful on an x86-64 machine?
> My laptop is 32-bit, so I don't currently have access to one.
>
> (I will note that I had tested it on a 64-bit Mac prior to the
> release.. I thought there were no issues.)
>
> Steve
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Windows:
>
> Build for Windows Store.
>
> http://p.sf.net/sfu/windows-dev2dev
> _______________________________________________
> liblo-devel mailing list
> lib...@li...
> https://lists.sourceforge.net/lists/listinfo/liblo-devel