Re: [PATCH 0/2] s390x: kvm: Honor storage keys during emulation

[Thomas Huth <thuth@redhat.com>]

On 10/05/2022 15.43, Cornelia Huck wrote:
> On Tue, May 10 2022, Janis Schoetterl-Glausch <scgl@linux.ibm.com> wrote:
> 
>> On 5/9/22 10:06, Cornelia Huck wrote:
>>> On Fri, May 06 2022, Janis Schoetterl-Glausch <scgl@linux.ibm.com> wrote:
>>>
>>>> Make use of the storage key support of the MEMOP ioctl, if available,
>>>> in order to support storage key checking during emulation.
>>>>
>>>> I did not update all the headers, since that broke the build,
>>>> not sure what the best way of dealing with that is.
>>>
>>> Yeah, the vfio change is expected to break the build; the fix should be
>>> easy (simple rename), and the code affected is deprecated anyway (there
>>> hasn't been any upstream implementation that actually exposed the
>>> interfaces). I think we should do that in a single commit to preserve
>>> bisectability; I have not seen any patches posted yet to actually use
>>> the new vfio migration interface, so a simple compile fixup should be
>>> all that is needed.
>>
>> So basically this patch (pasted below)
>> https://lore.kernel.org/qemu-devel/20220404181726.60291-3-mjrosato@linux.ibm.com/
>> squashed with the updated headers.
> 
> Yes. We should probably queue that seperately, just to disarm that trap
> for everyone; unless there's already a vfio update in flight? (Sorry, I've
> lost track a bit.)

Unless somebody else has queued this already, I can try to come up with a 
separate pull request for the header update today or tomorrow.

  Thomas