On Thu, Dec 28, 2006 at 01:31:34PM +0100, Jochen Kunz wrote:
> Same here. Prominent victim: fvwm2. I backtraced the core dumps with
> gdb, saw a fvwm internal GNOME_* function in the call trace, comented
> out that call and retried. Same error at a different position with a
> different GNOME_* function. Same procedure, comented out, retry, buss
> error, core dump and again a GNOME_* function in the way. So I disabled
> all GNONE support in fvwm (configure option) and since then fvwm is
> running well.

pwm is pretty simple (and the simplest reproduceable case I could find) -
no GNOME or GTK involved, just raw Xlib. When I get a chance this evening
I'll check and see if it is a bug in the code like Martin has suggested. I
remember running it successfully when I was running 3.99.x though (although
it's not my favourite wm, so it was a long time ago that I ran it).

> Also: Sometimes I get garbeld screen output in some window areas. It is
> easy to reproduce with sylpheed-claws, but I have seen it with xterm
> too. The xterm incident was related to konqueror. Some actions in
> konqueror garbeld the xterm window.

I haven't seen any screen oddities yet. I've actually been very impressed
with how well my dual Creator3D setup has been working with -current. Kudos
to those responsible!

> Maybe this is not a problem of NetBSD, but one of the libs from pkgsrc
> like gtk?

I haven't gotten as far as being able to build gtk - one of the other
problems I'm having is related to pango not building properly due to cairo
not building shared libs. But that's a problem for another day :)

> And: Do we have a swap problem? It seems that applications like to
> crash when they get paged / swaped out and back in.

Haven't seen this one, but my U60 hasn't swapped much. I know in the past
there were issues with applications that used native threads having issues
when they were paged out/in.

-mj
-- 
Michael-John Turner | http://mjturner.net/
mj@turner.org.za    | Open Source in WC ZA - http://www.clug.org.za/