On Donnerstag, 8. Mai 2003 15:19, Eduard Bloch wrote: > #include <hallo.h> > * Ralf Nolden [Wed, May 07 2003, 11:40:56PM]: > Content-Description: signed data > > > > According to Ralf Noden, USB is compiled into the kernel. > > > > Well, I originally *assumed* that would be the problem - I just could see > > that the hanging process is rmmod - which brought me to the quick > > assumption that the module can't be removed (and is therefore compiled > > in). But I'm > > I cannot reproduce this. And even if I could, file a bug against > modutils. Rmmod should exit immediately if the module is not loaded. > > > apparently wrong - if you load the module with hotplug then you'll > > naturally also unload it again. So it seems the bug is in the kernel > > module. > > Module? Which module? > > > OTOH, I don't know where the exact difference in the code is between the > > 2.4.20-bf2.4 kernel and the kernel-image-2.4.20-1-i386 (your one Herbert > > from unstable; I backported it to woody). As kernel-image-2.4.20-1-i386 > > works perfectly well with hotplug, the question is rather what the > > difference in the USB setup is. The best idea is probably to build bf2.4 > > from the kernel-image-2.4.20-1-i386, no ? > > Huch? That are different binary packages built from different source > packages but both using same source from the kernel-source packages. > The difference is in the config and a small patch, iirc. Ok, I tested the installation with regard to modutils. Using the version in woody (2.4.15-1) works but the version in unstable (2.4.21-2) triggers the hang in rmmod. Ralf > > MfG, > Eduard. -- We're not a company, we just produce better code at less costs. -------------------------------------------------------------------- Ralf Nolden nolden@kde.org The K Desktop Environment The KDevelop Project http://www.kde.org http://www.kdevelop.org
Attachment:
pgpUUDpOHCwKW.pgp
Description: signature