[Xen-devel] [DESIGN] Feature Levelling improvements

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

All,

With migration v2 getting close to being done, I have had time to pick
back up with feature levelling improvements.  Presented here for review
is draft E.

A PDF version of the design is available here:

http://xenbits.xen.org/people/andrewcoop/feature-levelling/feature-levelling-E.pdf

Pandoc version as follows:

% VM CPU Feature Levelling Improvements
% Andrew Cooper <<andrew.cooper3@xxxxxxxxxx>>
% Draft E

Introduction
============

Revision History
----------------

------------------------------------------------------------------------------
Version  Date         Changes
-------  ----------- 
--------------------------------------------------------
Draft A  07 Feb 2014  Initial draft

Draft B  13 Feb 2014  More detail for proposed new implementation

Draft C  17 Feb 2014  Even more details for proposed new implementation

Draft D  11 Jun 2014  More background, having had time to hack around and
                      experiment

Draft E  15 Jun 2015  More details for the proposed implementation.
------------------------------------------------------------------------------

Background
----------

_CPU feature masking_ is a term used to mean altering the visible
feature-set
of a processor.  For single systems, this could be to hide certain features
from operating system software, for which support is buggy.

In the world of virtualisation, it is common to have non-identical
hardware in
a cluster but still want to migrate a virtual machine safely.  On regular
hardware, the kernel can safely assume that the feature-set as detected on
boot will remain the same.  Live migration invalidates this assumption when
moving between two non-identical pieces of hardware.

To migrate virtual machines in this fashion, orchestration software must
ensure that the available feature set remains consistent anywhere the
virtual
machine might end up.

The feature-set of a particular CPU can be obtained using the `CPUID`
instruction.  It was introduced as a forward compatible way of
advertising new
features which were detectable at runtime.  Information available includes
processor branding, available features, topology information and cache
details.

The `CPUID` instruction is an unprivileged instruction, usable from
user-mode
without interception from the kernel.  This makes it impossible to
paravirtualise using the standard trap-and-emulate method.

Purpose
-------

This project originally started to improve the way in which XenServer
performed heterogeneous pool levelling.  In the process of investigation, it
was discovered that the current implementation in Xen and libxc are in
need of
improvement, particularly in relation to PV guests.

This document describes:

* What properties are needed from a VM point of view
* What hardware features are available to aid with levelling
* What abilities are exposed by Xen and libxc for levelling
* How XenServer currently does pool levelling (and why it is in need of
improvements)

This document also proposes a new mechanism for VM feature levelling, taking
into account the information needed by orchestration software.


What a Virtual Machine cares about
==================================

On native hardware, a kernel, as well as certain userspace libraries
will use
the set of available features to tune themselves to run more efficiently.
Over a migrate, it is critical that features a VM is using do not disappear.
(In some cases it might be possible to trap-and-emulate missing
features, but
this would be an exceedingly high overhead and is not considered.)

When a VM is liable to migrate between hardware of differing
feature-sets, it
is important to ensure that the VM is strictly only using the common
subset of
features available on any potential destination.

This can be done either by hiding features outside of the common subset,
or in
some cases specifically instructing the kernel not to use a feature which it
can see.

Hardware features to aid levelling
==================================

HVM
---

HVM guests (using `Intel VT-x` or `AMD SVM`) will unconditionally exit
to Xen
on all `CPUID` instructions, allowing Xen full and complete control over all
leaves.

PV
--

The `CPUID` instruction is unprivileged, so executing it in a PV guest will
not trap, leaving Xen no direct ability to control the information returned.

Xen Forced Emulation Prefix
---------------------------

Xen-aware PV guest kernels and userspace can make use of the 'Forced
Emulation
Prefix'

> `ud2a; .byte 'x'; .byte 'e'; .byte 'n'; cpuid`

which Xen recognises as a deliberate attempt to get the fully-controlled
`CPUID` information rather than the hardware-reported information.  This
only
works with cooperative guests and guest userspace, so cannot be directly
relied upon.

Masking and Override MSRs
-------------------------

AMD CPUs from the `K8` onwards support _Feature Override_ MSRs, which
specify
the raw value returned for all `CPUID` instructions querying a specific
feature bitmap.  These MSRs allow any result to be returned, including the
ability to advertise features which are not actually supported.

Intel CPUs between `Nehalem` and `SandyBridge` have differing numbers of
_Feature Mask_ MSRs, which are a simple AND-mask applied to all `CPUID`
instructions requesting specific feature bitmap sets.  The exact MSRs, and
which feature bitmap sets they affect are hardware specific.  These MSRs
allow
features to be hidden by clearing the appropriate bit in the mask, but does
not allow unsupported features to be advertised.

CPUID Faulting
----------------

On newer Intel hardware, a feature known as _CPUID Faulting_ can allow
Xen to
cause `CPUID` instruction executed in PV guests to trap, which allows
Xen full
and complete control over all leaves (exactly like an HVM guest).  _CPUID
Faulting_ support is present in `IvyBridge` and newer CPUs, although not
architecturally guaranteed.


How Xen currently uses and exposes levelling support
====================================================

Libxc has a `CPUID` Policy API which can be set by the toolstack for a
domain.
Libxc performs some information gathering, and uses the `DOMCTL_set_cpuid`
hypercall to specify what information should be returned by Xen when the
domain requests specific `CPUID` leaves.

The user of the libxc `CPUID` Policy API may specify, for any leaf
whatsoever,
whether particular bits should be forced high, forced low, default (as
chosen
by libxc), specifically the same as hardware, or specifically the same
hardware and maintained consistently across migration.

The default `CPUID` Policy involves libxc trying to work out which features
should be set or cleared in the policy.  It does this with a mixture of
native
`CPUID` instructions, some switch statements choosing to enable/disable
certain features and hypercalls querying certain Xen state.

When Xen is servicing a `CPUID` instruction on behalf of a guest and ends up
using the policy provided by libxc, it subsequently edits certain fields,
particularly in the feature sets.

Support for the feature masking MSRs is available via the `cpuid_mask_*`
command line parameters which get applied at boot and reduce the visible
feature set to every subsequent `CPUID` instruction.

Support for enabling _CPUID Faulting_ exists, but it does nothing more than
defer back to the default policy.


How XenServer currently does levelling
======================================

The _Heterogeneous Pool Levelling_ support in XenServer appears to
predate the
libxc CPUID policy API, so does not currently use it.  The toolstack has a
table of CPU model numbers identifying whether levelling is supported.  It
then uses native `CPUID` instructions to look at the first four feature
masks,
and identifies the subset of features across the pool.
`cpuid_mask_{,extd_}{ecx,edx}` is then set on Xen's command line for
each host
in the pool, and all hosts rebooted.

This has several limitations:

* Xen and dom0 have a reduced feature set despite not needing to migrate
* There is only a single level for all VMs in the pool
* The toolstack only understands the first 4 of the possible masking
MSRs, and
  there are now feature maps in further `CPUID` leaves which have no masking
  MSRs


Notes and observations
======================

Experimentally, the masking MSRs can be context switched.  There is no
need to
force all PV guests to the same level, and no need to prevent dom0 or
Xen from
using certain features.  Context switching the masking MSRs will however
incur
an overhead, and should be avoided where possible.

The toolstack needs to know how much control Xen has over VM features. 
In the
case that there are insufficient masking MSRs, and no faulting support is
present, a PV VM can still potentially be made safe to migrate by explicitly
disabling features on the kernel command line.  As a result, there
should be a
new mechanism which reports the levelling controls Xen has available.

The features available to each type of guest is really only known to Xen.
Having libxc try to divine them is bogus (especially as libxc is subject to
the toolstack domains cpuid policy itself).  Therefore on boot, Xen should
work out the maximal feature set available to each type of guest and
make this
information available to the toolstack.


Design
======

`struct sysctl_physinfo.levelling_caps`
---------------------------------------

Xen shall gain a new physinfo field which reports the degree to which it can
influence `CPUID` executed by a PV guest.  This is a bitmap containing:

* `faulting`
    * CPUID Faulting is available, and full control can be exercised.
* `mask_ecx`
    * Leaf 0x00000001.ECX
* `mask_edx`
    * Leaf 0x00000001.EDX
* `mask_extd_ecx`
    * Leaf 0x80000001.ECX
* `mask_extd_edx`
    * Leaf 0x80000001.EDX
* `mask_xsave_eax`
    * Leaf 0x0000000D[ECX=1].EAX
* `mask_therm_ecx`
    * Leaf 0x00000006.ECX
* `mask_l7s0_eax`
    * Leaf 0x00000007[ECX=0].EAX
* `mask_l7s0_ebx`
    * Leaf 0x00000007[ECX=0].EBX

At the time of writing, these are all the masking MSRs known by Xen.  The
bitmap shall be extended as new MSRs become available.

New 'featureset' API for use by the toolstack
---------------------------------------------

A featureset is a defined as a collection of words covering the cpuid leaves
which report features to the caller.  It is variable length, and expected to
grow over time as processors gain more features, or Xen starts supporting
exposing more features to guests.

At the time of writing, the leaves containing feature bits are:

* 0x00000001.ECX
* 0x00000001.EDX
* 0x80000001.ECX
* 0x80000001.EDX
* 0x0000000D[ECX=1].EAX
* 0x00000007[ECX=0].EBX
* 0x00000006.EAX
* 0x00000006.ECX
* 0x0000000A.EAX
* 0x0000000A.EBX
* 0x0000000F[ECX=0].EDX
* 0x0000000F[ECX=1].EDX

XEN_SYSCTL_get_featureset
-------------------------

Xen shall on boot create a featureset for itself, and the maximum available
features for each type of guest, based on hardware features, command line
options etc.  A toolstack shall be able to query all of these.

Cpuid feature-verification library
----------------------------------

There shall be a new library (shared between Xen and libxc in the same
way as
libelf etc.) which can verify the a featureset.  In particular, it will
confirm that no features are enabled without their dependent features.

XEN_DOMCTL_set_cpuid
--------------------

This is an existing hypercall.  Currently it just stashes the policy from
userspace.  It shall be extended to provide verification of the policy, and
reject attempts to advertise features which Xen is incapable of providing
(via hardware or emulation support).

VCPU context switch
-------------------

Xen shall be updated to lazily context switch all available masking
MSRs.  It
is noted that this shall incur a performance overhead if restricted
featuresets are assigned to PV guests, and _CPUID Faulting_ is not
available.

It shall be the responsibility of the host administrator to avoid creating
such a scenario, if the performance overhead is a concern.


Future work
===========

The above is a minimum quantity of work to support feature levelling, but
further problems exist.  They are acknowledged as being issues, but are
not in
scope for fixing as part of feature levelling.

* Xen has no notion of per-cpu and per-package data in the cpuid policy.  In
  particular, this causes issues for VMs attempting to detect topology,
which
  find inconsistent/incorrect cache information.

* In the case that `domain_cpuid()` can't locate a leaf in the topology, it
  will fall back to issuing a plain `CPUID` instruction.  This breaks VM
  encapsulation, as a VM which has migrated can observe differences which
  should be hidden.

* There is currently a positioning issue with the domains cpuid policy.
  Verifying the register state requires the policy, but the policy is behind
  the register state in the migration stream.  The domains cpuid policy
should
  become an item in Xen's migration state for a VM.


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel