Re: [Xen-devel] Xen-unstable: xen panic RIP: dpci_softirq

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 14.11.14 at 14:11, <linux@xxxxxxxxxxxxxx> wrote:
> (XEN) [2014-11-14 13:00:06.810] ----[ Xen-4.5.0-rc  x86_64  debug=y  Not 
> tainted ]----
> (XEN) [2014-11-14 13:00:06.810] CPU:    3
> (XEN) [2014-11-14 13:00:06.810] RIP:    e008:[<ffff82d080148f14>] 
> dpci_softirq+0x9c/0x231
> (XEN) [2014-11-14 13:00:06.810] RFLAGS: 0000000000010283   CONTEXT: hypervisor
> (XEN) [2014-11-14 13:00:06.811] rax: 0000000000100100   rbx: ffff830515c86d90 
>  rcx: 0000000000000001

This and ...

> (XEN) [2014-11-14 13:00:06.811] Pagetable walk from 0000000000100108:
> (XEN) [2014-11-14 13:00:06.811]  L4[0x000] = 000000055d09a063 ffffffffffffffff
> (XEN) [2014-11-14 13:00:06.811]  L3[0x000] = 000000055d099063 ffffffffffffffff
> (XEN) [2014-11-14 13:00:06.811]  L2[0x000] = 000000055d098063 
> ffffffffffffffff 

... this suggest it hit a poisoned list entry, due to

#define LIST_POISON1  ((void *) 0x00100100)

But without you helping out with associating the crash location with
a source line (or allowing us to do so by making xen-syms available
somewhere) this is going to remain guesswork.

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel