Re: [PATCH 0/7] i386: Add `machine` parameter to query-cpu-definitions

From: David Hildenbrand <david@redhat.com>
To: Eduardo Habkost <ehabkost@redhat.com>
Cc: Thomas Huth <thuth@redhat.com>,
	"Daniel P. Berrange" <berrange@redhat.com>,
	Janosch Frank <frankja@linux.ibm.com>,
	libvir-list@redhat.com, Cornelia Huck <cohuck@redhat.com>,
	Michal Skrivanek <michal.skrivanek@redhat.com>,
	qemu-devel@nongnu.org, Markus Armbruster <armbru@redhat.com>,
	Christian Borntraeger <borntraeger@de.ibm.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Igor Mammedov <imammedo@redhat.com>,
	Jiri Denemark <jdenemar@redhat.com>,
	Richard Henderson <rth@twiddle.net>
Subject: Re: [PATCH 0/7] i386: Add `machine` parameter to query-cpu-definitions
Date: Fri, 25 Oct 2019 16:10:10 +0200	[thread overview]
Message-ID: <25e33d78-3d25-7ca9-6922-920846d0d98b@redhat.com> (raw)
In-Reply-To: <20191025133846.GD6744@habkost.net>

On 25.10.19 15:38, Eduardo Habkost wrote:
> CCing danpb, libvir-list, mskrivanek.
> 
> On Fri, Oct 25, 2019 at 09:17:46AM +0200, David Hildenbrand wrote:
>> On 25.10.19 04:25, Eduardo Habkost wrote:
>>> We had introduced versioned CPU models in QEMU 4.1, including a
>>> method for querying for CPU model versions using
>>
>> Interesting, I was not aware of that.
>>
>> On s390x, we somewhat have versioned CPU models, but we decided against
>> giving them explicit names (e.g., z13-v1 or z13-4.1.0), because it didn't
>> really seem to be necessary. (and we often implement/add features for older
>> CPU models, there is a lot of fluctuation) Actually, you would have had to
>> add "z13-z/VM-x.x.x" or "z13-LPAR-x.x.x" or "z13-KVM-x.x.x" to model the
>> features you actually see in all the different virtual environments ("what a
>> CPU looks like"). Not to talk about QEMU versions in addition to that. So we
>> decided to always model what you would see under LPAR and are able to
>> specify for a KVM guest. So you can use "z13" in an up-to-date LPAR
>> environment, but not in a z/VM environment (you would have to disable
>> features).
>>
>> Each (!base) CPU model has a specific feature set per machine. Libvirt uses
>> query-cpu-model-expansion() to convert this model+machine to a
>> machine-independent representation. That representation is sufficient for
>> all use cases we were aware of (esp. "virsh domcapabilities", the host CPU
>> model, migration).
>>
>> While s390x has versioned CPU models, we have no explicit way of specifying
>> them for older machines, besides going over query-cpu-model-expansion and
>> using expanded "base model + features".
>>
>> I can see that this might make sense on x86-64, where you only have maybe 3
>> versions of a CPU (e.g., the one Intel messed up first - Haswell, the one
>> Intel messed up next - Haswell-noTSX, and the one that Intel eventually did
>> right - Haswell-noTSX-IBRS) and you might want to specify "Haswell" vs.
>> "Haswell-IBRS" vs. "Haswell-noTSX-IBRS". But actually, you will always want
>> to go for "Haswell-noTSX-IBRS", because you can expect to run in updated
>> environments if I am not wrong, everything else are corner cases.
>>
>> Of course, versioned CPU model are neat to avoid "base model + list of
>> features", but at least for expanding the host model on s390x, it is not
>> really helpful. When migrating, the model expansion does the trick.
>>
>> I haven't looked into details of "how to specify or model versions". Maybe
>> IBM wants to look into creating versions for all the old models we had. But
>> again, not sure if that is of any help for s390x. CCing IBM.
> 
> I'm not sure yet if there are the x86-specific goals and
> constraints that would make it difficult to follow the same
> approach followed by s390x.  I have the impression we do,
> but I need to think more carefully about it.
> 
> Let's discuss that during KVM Forum?

I won't be attending KVM Forum this year, but Christian should be around.

> 
> The two main goals of versioned CPU models in x86 are:
> 1) Decoupling CPU model updates from machine-types (users should be
>     able to update a CPU model definition without changing machine
>     type).
> 2) Letting management software automate CPU model updates.
>     Normally this is necessary when bare metal microcode or
>     software updates add/remove features from CPUs.  Sometimes the
>     Virtual CPU model update needs to be performed before the host
>     updates are applied (if features are being removed)

We have 2) on s390x after a QEMU machine update. You need a QEMU update 
usually either way to support the new CPU feature. But I can see how 
decoupling the CPU model definition from the machine makes this easier. 
It does introduce complexity. It applies only when running older QEMU 
machines, or if we have to update a CPU model before we get a new QEMU 
machine.

But we do have versioned CPU models already, so the discussion is 
already over :)

> 
> The main constraint that makes it difficult to address the above
> without a new API is:
> A) Every CPU model in x86 except "host" is already expected to
>     be migration-safe (I don't know how this compares to s390x).

I would describe that not a bug but a feature :) It does come with the 
price that only using the newest QEMU machine results in the newest CPU 
model, that part I understand.

-- 

Thanks,

David / dhildenb