On Mon, 2010-06-14 at 13:14 +0200, Peter Palfrader wrote: > On Mon, 14 Jun 2010, Ben Hutchings wrote: > > > On Sun, 2010-06-13 at 20:45 +0200, Peter Palfrader wrote: > > > cimarosa.d.o runs linux-image-2.6.32-bpo.3-amd64 (from 2.6.32-9) just > > > fine. > > > > > > However, it does not boot with linux-image-2.6.32-bpo.5-amd64 > > > (2.6.32-15): > > > > > > During boot up, when init brings up the network we get various different > > > kernel NULL pointer dereferences. Cimarosa has a lo, an eth0 and an > > > eth0.221 vlan device. If the eth0.221 is not configured to be brought > > > up on boot then the system boots correctly, and bringing up the > > > interface later also appears to work without any ill effects. So > > > there might be some racy thing going on. > > > > > > It'd be pretty sweet if cimarosa could be made to work with a recent > > > debian kernel. > > [...] > > > > Could you try reverting the patch > > bugfix/all/vlan-macvlan-propagate-transmission-state-to-upper-layer.patch ? > > That appears to have helped. [...] Was the crash reliably reproducible, so you can be confident that this change fixed it? Ben. -- Ben Hutchings Once a job is fouled up, anything done to improve it makes it worse.
Attachment:
signature.asc
Description: This is a digitally signed message part