Miro (Democracy) player port is here...

[Randy Pratt]

On Sun, 18 Nov 2007 15:28:39 -0600
"Jeremy Messenger" <mezz7 at cox.net> wrote:

> On Sun, 18 Nov 2007 12:06:26 -0600, Randy Pratt <bsd-unix at embarqmail.com>  
> wrote:
> 
> > On Sun, 18 Nov 2007 10:58:24 -0600
> > "Jeremy Messenger" <mezz7 at cox.net> wrote:
> >
> >> On Sun, 18 Nov 2007 10:50:17 -0600, Jeremy Messenger <mezz7 at cox.net>  
> >> wrote:
> >>
> >> > On Sun, 18 Nov 2007 10:08:20 -0600, Randy Pratt
> >> > <bsd-unix at embarqmail.com> wrote:
> >> >
> >> > <snip>
> >> >> That worked for the installation.  Thanks!
> >> >
> >> > Good, I have updated miro.shar in the same place. Thanks too!
> >> >
> >> >> Although, when Miro was started, it displayed and was loading
> >> >> then seg faulted with:
> >> >>
> >> >> ...
> >> >> INFO     First URL is https://www.miroguide.com/firsttime
> >> >> TIMING   Icon clear: 0.002
> >> >> INFO     Starting movie data updates
> >> >> INFO     Finished startup sequence
> >> >> TIMING   idle (finalizing startup) too slow (3.984 secs)
> >> >> INFO     *** Daemon ready ***
> >> >> INFO     got file:///tmp/tmpacEbWZ.html
> >> >> TIMING   gtkAsyncMethod: <function selectDisplay at 0x8bbe064> took  
> >> too
> >> >> long: 1.109 INFO     got file:///tmp/tmpA7pWuk.html
> >> >> WARNING  downloader: connection closed -- quitting
> >> >> INFO     Shutting down downloaders...
> >> >> Segmentation fault (core dumped)
> >> >
> >> > Ummm... It doesn't give enough info.. Can you see *.core file in  
> >> there?
> >> > If you do, then try to run this:
> >> >
> >> > 	# gdb foo foo.core
> >> > 	# bt
> >> > 	# bt full
> >> >
> >> > You might have to reinstall miro with debug (WITH_DEBUG=yes). I will
> >> > have to recheck in my build log to make sure everything compile with
> >> > -pthread.
> >>
> >> BTW: Try to run something like 'LD_PRELOAD=/usr/lib/libpthread.so miro'
> >> (or /usr/lib/libthr.so) to see if it works for you. If it does, then  
> >> will
> >> have to search somewhere that need to add -pthread. I don't have FreeBSD
> >> 6.x box anymore, so you or someone will have to keep test for me. ;-)
> >
> > There was no joy in trying either of those.  It produced the same
> > core dump.
> 
> You are right about your backtraces don't help in other email.
> 
> > These are the /var/db/ports/*/options settings I have pertaining to
> > threads:
> >
> >   boehm-gc/options:WITHOUT_THREADING=true
> >   boost/options:WITH_THREADS=true
> >   ilmbase/options:WITHOUT_THREAD=true
> >   python/options:WITH_THREADS=true
> >   python24/options:WITH_THREADS=true
> >   python25/options:WITH_THREADS=true
> >   ruby/options:WITHOUT_PTHREADS=true
> >   sqlite3/options:WITH_THREADS=true
> 
> That is only one different from mine. I left sqlite3 by default, no enable  
> thread. I have checked each files and build log, it looks like it has  
> nothing to do with missing -pthread. You can try to disable thread on  
> sqlite3 to see if it helps.

I actually had it at the default of no threads originally and rebuilt
it with threads to see if it helped (to no avail).  

> >   stunnel/options:WITH_PTHREAD=true
> >
> > Certainly I'll help debug this on 6.x.  During the ports freeze
> > there's not much ports excitement going on (for me ;-)
> 
> I am clueless at the moment, I need to dig in Miro's website to see if  
> there have any info about how to debug on miro.

I'll do some digging around myself although if it gets too deep I
tend to get dazed and confused.  If I find anyting, I'll advise.

If anyone has/gets Miro working on 6.x, I'd like to hear about it
and maybe do some comparisons.

Thanks!

Randy

--