gcc-3.1 libcj issues

[Bryce McKinlay]

Adam Megacz wrote:

>Tom Tromey <tromey@redhat.com> writes:
>
>>Yes, please report these.  This gets a little out of our hands.  It
>>would be too bad if we couldn't support Darwin in this release, but we
>>might be forced that way.  Let's at least give it a try though.
>>
>
>Hey, sorry I haven't been tracking this; did we ever come up with a
>solution for how to let the GC stop all threads on Darwin? It had
>something to do with Darwin not having pthread_kill()...
>

Yes, I think it could be done pretty easily using mach's thread_suspend, 
etc. However, the next release of Darwin will have per-thread signal 
masks and a proper pthread_kill, so the effort is probibly not worth it. 
Once we have that stuff the generic posix threads code ought to work.

regards

Bryce.