Re: [Xen-devel] [PATCH RFC] hvm: Allow triple fault to imply crash rather than reboot

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

On 07/11/16 12:56, Xuquan (Quan Xu) wrote:
> On Feb 4, 2013, 6:25 AM, <andrew.cooper3@citrix> wrote:
>> While the triple fault action on native hardware will result in a system
>> reset, any modern operating system can and will make use of less violent
>> reboot methods. As a result, the most likely cause of a triple fault is a
>> fatal software bug.
>>
>> This patch allows the toolstack to indicate that a triple fault should mean a
>> crash rather than a reboot. The default of reboot still remains the same.
>>
> hi, Andrew
>
> this email has been ages ago. I am working for a triple fault related issues, 
> and I have some doubts:
>
> 1)  as you mentioned here, 'fatal software bug', is it to xen hypervisor or 
> guest os?

Guest

> 2)  why introduce 'crash'?  would the 'reboot' lead to a fatal problem ( i.e, 
> xen hypervisor panic) ?

The purpose was to change how the shutdown is reported to the toolstack,
and in particular whether the on_reboot or on_crash action gets called.

> 3)  are there any differences between __'crash the guest, destroy the guest , 
> create the guest again'__
>     And __'reboot guest'__ ?

Only which toolstack action gets invoked.

> 4)  any ideas to reproduce triple fault?

What do you mean by reproduce? Causing a triple fault is very easy from
the guest kernel.

~Andrew

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