Re: [Qemu-devel] [PATCH] migration: catch unknown flag combinations in ram_load

[Eric Blake <eblake@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 1524 bytes --]

On 07/09/2014 07:23 AM, Peter Lieven wrote:
> Am 09.07.2014 13:00, schrieb Peter Maydell:
>> On 9 July 2014 11:56, Amit Shah <amit.shah@redhat.com> wrote:
>>> On (Wed) 09 Jul 2014 [11:50:18], Peter Maydell wrote:
>>>> OK; we're treating those as bugs so yes, I think this is 2.1
>>>> material. Has somebody other than the original author tested
>>>> it? (That's a step that would usually be done by Juan as the
>>>> maintainer.) If somebody can provide a Tested-by: I'm happy
>>>> to apply it to master.
>>> Not really sure if Juan did that as part of his 'thanks, applied'
>>> workflow, but I'll run this through the autotest migration tests and
>>> report back.
>> Thanks; it seems better to double-check given where we are in
>> the release cycle.
> We have also an:
> 
> Reviewed-by: Eric Blake <eblake@redhat.com>
> 
> 
> Maybe he tested as well.

No, while I did a review, and can even do a compile test, I have not
done a migration test, so I'm not comfortable with a Tested-by yet. Is
it something you need me to try today?

> 
> If there are any doubts we should postpone this for 2.2. The preceding
> patch is in 2.1 and should catch already a lot of cases.

I'm 50-50 on whether this is 2.1 material - it is a bug fix for
hardening against malicious incoming migration streams, but without
testing that it doesn't break normal migration, it is rather late in the
game.

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 604 bytes --]