Re: GNU Guile 2.0.10 released

[Eli Zaretskii]

> Date: Wed, 26 Mar 2014 19:44:53 +0100
> From: Panicz Maciej Godek <address@hidden>
> Cc: Ludovic Courtès <address@hidden>, 
>       address@hidden, "address@hidden" <address@hidden>
> 
> 2014-03-26 19:24 GMT+01:00 Eli Zaretskii <address@hidden>:
> >> I had a "process hacker" tool installed, and it allows me to do some
> >> introspection (although I understand very little). [...]
> >
> > Isn't this the same problem I reported last year, starting here:
> >
> >   http://lists.gnu.org/archive/html/guile-user/2013-06/msg00022.html
> >
> > I suggest to configure Guile --without-threads, and see if that
> > resolves the issue.
> 
> I suspect that it is. But I wanted to rebuild my SLAYER framework for
> Windows and make a new release with Guile 2.0.10 and threads enabled
> -- because then it would make more sense to distribute it.

I know what you mean.  I also hate it when programs lose features when
ported.

> I wish I knew a way to find out whether the problem is on Guile's or
> GC's or mingw's or pthreads-win32's side.

Indeed.  At the time, no one here was able to help me locate the
problem.

> Some guys who had similar problem (i.e. their process also stopped at
> "ntdll!KiUserApcDispatcher+0x7") suggest that it is a busy resource:
> http://stackoverflow.com/questions/2214465/debugging-the-dreaded-application-has-failed-to-initialize-error

Last June I reported many details about the hang, you might find there
something about this stuff.  See, for example, these messages:

  http://lists.gnu.org/archive/html/guile-user/2013-06/msg00030.html
  http://lists.gnu.org/archive/html/guile-user/2013-06/msg00076.html
  http://lists.gnu.org/archive/html/guile-user/2013-06/msg00079.html

The last one leads to this:

  http://lists.gnu.org/archive/html/guile-devel/2011-06/msg00034.html

> So now, instead of just compiling --without-threads, I'd like to
> somehow get the threaded version working, even it takes much more time
> and more digging.

It would be wonderful if you succeed in solving this.  Mark told me
that someone succeeded in building MinGW Guile with pthreads, and he
built pthreads himself.  So maybe you should try that, among other
things.

Thanks.