>>> On 23.04.15 at 13:07, <vkuzn...@redhat.com> wrote:
> "Jan Beulich" <jbeul...@suse.com> writes:
>
>>>>> On 23.04.15 at 12:39, <vkuzn...@redhat.com> wrote:
>>> --- a/xen/common/shutdown.c
>>> +++ b/xen/common/shutdown.c
>>> @@ -37,46 +37,36 @@ void hwdom_shutdown(u8 reason)
>>> switch ( reason )
>>> {
>>> case SHUTDOWN_poweroff:
>>> - {
>>> - printk("Domain 0 halted: halting machine.\n");
>>> + printk("Hardware domain halted: halting machine.\n");
>>> machine_halt();
>>> break; /* not reached */
>>
>> Would it perhaps make sense (for the avoidance of ambiguity/
>> confusion) to also print the hardware domain's domain ID?
>
> Sure, why not. Did I get it right that we need to print
> hardware_domain->domain_id and not hardware_domid here to support hwdom
> kexec case?
Irrespective of the kexec case I think you should print the actual
state of things, not the ID that was requested to become the
hardware domain (think e.g. also of an early death of Dom0
before the actual HW domain got created).
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
