Hi Volker,
Let's change the meaning of this. There is nothing like a reproducible
testcase. Noone was ever able to fix this because it is not reproducible.
(No, the testcase is not: Build OOo with tcsh because that just doesn't
hang on all the systems that were accessible.
My money is still on some drivers/services, see the links in this snippet
from the cygwin ml with respect to a maybe related topic:
Quote from cygwin ml with respect to fork problems/hangs:
> It's crappy coding on the part of the Logitech module. Ross Ridge gives
> a convincing argument for what's going on:
> <http://sourceforge.net/mailarchive/message.php?msg_id=17999960>. More
> info on the loader lock issue is in this paper:
> <http://www.microsoft.com/whdc/driver/kernel/DLL_bestprac.mspx>. Now if
> only someone could forward these clues to the idiots that wrote that
> Logitech crapware, it would save a whole lot of frustration.
I can reproduce this bug in two Windows XP machines, but very
different hardware: a laptop Dell Latitude D600 (Windowx XP
Professional English) and a Dell Dimension 9100 (Windows XP Home
Edition Spanish) .
A win32-tcsh build will freeze (apparently) executing the shell script
at "extras\source\misc_config\lang" 90% of the time. This is the
easiest way to reproduce the freeze in both computers. I'm not using
any Logitech software at all.
I will try to do some serious investigation on that topic this weekend.
--
Jesús Corrius
http://www.corrius.org/bloc/
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
