Re: odd crash in jabberd2

[Greg Troxel <gdt%ir.bbn.com@localhost>]

David Laight <david%l8s.co.uk@localhost> writes:

> On Wed, Jan 28, 2009 at 06:40:23PM -0500, Greg Troxel wrote:
>> 
>> The problem seems to be dlopening authreg_sqlite3 which is linked
>> with pthread, but the base program is not.
>
> That can't be supported ...

I see your point, but does anyone know what happens on linux and
solaris?

>> What I don't understand is
>> why the unlock is blowing up but the corresponding lock did not - it
>> doesn't seem that pthreads were initialized during the malloc call -
>> there is only one thread.  I also had to do this with sm.
>
> pthreads have to be initialised at program startup, since some
> standard parts of libc have to behave differently in a threaded program.
>
>> Is this a bug in our libc?  Or jabberd2? 
>
> I'd vote for jaBberd2...

Is there a spec that says this?  Should we have a better/earlier abort
at dlopen time?