There is a bug in the current interrupt handler code. I'm not 100% sure but I think we have a patch to that. The problem is that with Pistachio we run APIC interrupt mode and when IRQs are disabled they usually get re-injected with a self-IPI (I'm not arguing about the sensibility of that strategy though). Right now L4Linux doesn't do that and the IRQ gets lost.
- Volkmar
-- Mat/tia Belletti - Undergraduate student @ cs.unibo.it ICQ: 33292311 - email: mbellett@cs.unibo.it IRC: RedGlow - site(s): http://www.cs.unibo.it/~mbellett Linux registered user 299762 @ Linux registered machine 213003
l4-hackers mailing list l4-hackers@os.inf.tu-dresden.de http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers