Re: [Xen-devel] Xen 4.11.1: guest uses altp2m to crash host

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 25.04.19 at 14:53, <andrew.cooper3@xxxxxxxxxx> wrote:
> On 25/04/2019 13:17, Razvan Cojocaru wrote:
>> On 4/25/19 1:04 PM, Jan Beulich wrote:
>>>>>> On 15.04.19 at 16:10, <rcojocaru@xxxxxxxxxxxxxxx> wrote:
>>>> On 4/15/19 5:04 PM, Matt Leinhos wrote:
>>>>> You are correct -- cherry picking that commit into RELEASE-4.11.1 
>>>>> addresses the crash. The problem cannot be reproduced on 4.12.
>>>>>
>>>>> I hope this helps.
>>>> Thank you for confirming the fix!
>>>>
>>>> To the deciders: would it be reasonable to ask patches like this to be 
>>>> backported to the future 4.11 release?
>>> Well, first of all you can always send a request. And yes, in general
>>> I'd consider this reasonable, but I say so without having looked at
>>> the actual commit(s) in question.
>> Thank you for the answer. This is the commit in question:
>>
>> 
> http://xenbits.xen.org/gitweb/?p=xen.git;a=commitdiff;h=1dddfff4c39d3db17dfa7 
> 09b1c57f44e3ed352e3;hp=68465944c81fa2fbc40bf29ec6084072af5e48ec
>>
>> AFAICT there's nothing controversial about it, and it should easily
>> apply to 4.11, so if possible I propose that we consider backporting it
>> to the next 4.11 release.
> 
> I have a rather large number of altp2m backports to Xen 4.11 in the
> XenServer patchqueue, which in practice

... ???

> It would certainly make my life easier if they were to go upstream.  I
> can put together a branch if that would be helpful.

What volume / count are we talking about? Anyway, if it's really
mostly altp2m code that gets touched, I don't mind, and having a
branch to pull from would be helpful. 4.11.2 is going to be delayed
anyway at least until the master branch advances again, as I've
meanwhile queued up quite a few bug fixes which all sit behind the
push gate there.

Jan



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel