[PATCH 0/2] KVM enablement for facility bit 81 and 82

[Joseph Salisbury]

On 02/08/2018 01:55 PM, Khaled Elmously wrote:
> On 2018-02-02 17:58:22 , Joseph Salisbury wrote:
>> On 02/02/2018 05:54 PM, Joseph Salisbury wrote:
>>> From: Christian Borntraeger <borntraeger at de.ibm.com>
>>>
>>> BugLink: http://bugs.launchpad.net/bugs/1747090
>>>
>>> == SRU Justification ==
>>> Mainline commit 35b3fde6203b9 is a KVM patch for s390x to provide
>>> facility bits 81 (ppa15) and 82 (bpb).  This is required for branch prediction
>>> behaviour changes.
>>>
>>> This is the public bug for SRU.  There is also a priave bug report:
>>> http://bugs.launchpad.net/bugs/1743560
>>>
>>> There is a qemu portion to this fix:
>>> http://patchwork.ozlabs.org/cover/862801/
>>>
>>> This fix is also requred in Artful and Bionic, but Xenial requires a prereq commit,
>>> so it is being SRU'd separatly.
>>>
>>> == Fixes ==
>>> ed8dda0bf74b ("Enable all facility bits that are known good for passthrough")
>>> 35b3fde6203b ("KVM: s390: wire up bpb feature")
>>>
>>>
>>> == Regression Potential ==
>>> Low, this fix is limited to s390.
>>>
>>> == Test Case ==
>>> A test kernel was built with these patches and tested by the original bug
>>> reporter.  The bug reporter states the test kernel resolved the bug.
>>>
>>>
>>>
>>> Alexander Yarygin (1):
>>>   KVM: s390: Enable all facility bits that are known good for
>>>     passthrough
>>>
>>> Christian Borntraeger (1):
>>>   KVM: s390: wire up bpb feature
>>>
>>>  arch/s390/include/asm/kvm_host.h |  3 ++-
>>>  arch/s390/include/uapi/asm/kvm.h |  3 +++
>>>  arch/s390/kvm/kvm-s390.c         | 16 ++++++++++++++--
>>>  include/uapi/linux/kvm.h         |  1 +
>>>  4 files changed, 20 insertions(+), 3 deletions(-)
>>>
>> This request is for Xenial.  I missed adding that to the subject. 
>> Artful and Bionic are in a seperate SRU request.
>>
> I'm curious how you realized that you need a pre-requisite commit (given that the desired commit would have applied more-or-less cleanly anyway as it did for artful)? Was that by testing?
>
> The patchset doesn't look wrong to me at all, I'm just asking for my general knowledge.
The prereq commit was identified by IBM internally.  They posted the
first version of the backports in the private bug[0], in comment #4.

[0] http://pad.lv/1743560

>
>
>> -- 
>> kernel-team mailing list
>> kernel-team at lists.ubuntu.com
>> https://lists.ubuntu.com/mailman/listinfo/kernel-team