Keir Fraser wrote:
>
> A null pointer dereference in evtchn_set_pending can't be that hard to
> track down, if it's crashing every time in this way with suse 10.1.
It's vcpu being NULL, %esi register here:
ff1055f0 <evtchn_set_pending>:
ff1055f0: 83 ec 1c sub 0ドルx1c,%esp
ff1055f3: 89 74 24 14 mov %esi,0x14(%esp)
ff1055f7: 8b 74 24 20 mov 0x20(%esp),%esi
ff1055fb: 89 5c 24 10 mov %ebx,0x10(%esp)
ff1055ff: 8b 4c 24 24 mov 0x24(%esp),%ecx
ff105603: 89 7c 24 18 mov %edi,0x18(%esp)
ff105607: 8b 5e 0c mov 0xc(%esi),%ebx <==
ff10560a: 8b 53 04 mov 0x4(%ebx),%edx
The stack trace looks suspious to me though, none of the functions in
the trace calls evtchn_set_pending() ...
cheers,
 Gerd
-- 
Gerd Hoffmann <kraxel@xxxxxxx>
http://www.suse.de/~kraxel/julika-dora.jpeg
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel