On Tue, Jun 19, 2007 at 08:53:18PM +0200, Manuel Bouyer wrote: > Is the first panic() also a "read_psl() == 0" one ? Just a note to confirm that Xen 3.1 hasn't changed the problem I see with this KASSERT and HVM (not surprising, since it's the same dom0 kernel). I also tried again the same test kernel with this line commented out as before, but on the Xen 3.1 hypervisor, just to see if there was any observable difference. This time, the HVM started and ran without causing the local problem with the X server, but instead the HVM domU ran very very slowly. From a limited sample size, I'm not sure if this represents a difference because of xen versions, or just different test results I would have seen with either version. I haven't used a non-HVM domU much (or at all?) since the 3.1 upgrade, I shall try that when I can to see if that also trips this issue now. This is on my laptop during daily work, so running repeated tests (and fsck's) has to wait for suitable opportunities. -- Dan.
Attachment:
pgpMERFa4f4Nk.pgp
Description: PGP signature