1 Recommendations for KVM CPU model configuration on x86 hosts
2 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4 The information that follows provides recommendations for configuring
5 CPU models on x86 hosts. The goals are to maximise performance, while
6 protecting guest OS against various CPU hardware flaws, and optionally
7 enabling live migration between hosts with heterogeneous CPU models.
10 Two ways to configure CPU models with QEMU / KVM
11 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
13 (1) **Host passthrough**
15 This passes the host CPU model features, model, stepping, exactly to
16 the guest. Note that KVM may filter out some host CPU model features
17 if they cannot be supported with virtualization. Live migration is
18 unsafe when this mode is used as libvirt / QEMU cannot guarantee a
19 stable CPU is exposed to the guest across hosts. This is the
20 recommended CPU to use, provided live migration is not required.
24 QEMU comes with a number of predefined named CPU models, that
25 typically refer to specific generations of hardware released by
26 Intel and AMD. These allow the guest VMs to have a degree of
27 isolation from the host CPU, allowing greater flexibility in live
28 migrating between hosts with differing hardware. @end table
30 In both cases, it is possible to optionally add or remove individual CPU
31 features, to alter what is presented to the guest by default.
33 Libvirt supports a third way to configure CPU models known as "Host
34 model". This uses the QEMU "Named model" feature, automatically picking
35 a CPU model that is similar the host CPU, and then adding extra features
36 to approximate the host model as closely as possible. This does not
37 guarantee the CPU family, stepping, etc will precisely match the host
38 CPU, as they would with "Host passthrough", but gives much of the
39 benefit of passthrough, while making live migration safe.
42 Preferred CPU models for Intel x86 hosts
43 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
45 The following CPU models are preferred for use on Intel hosts.
46 Administrators / applications are recommended to use the CPU model that
47 matches the generation of the host CPUs in use. In a deployment with a
48 mixture of host CPU models between machines, if live migration
49 compatibility is required, use the newest CPU model that is compatible
50 across all desired hosts.
52 ``Cascadelake-Server``, ``Cascadelake-Server-noTSX``
53 Intel Xeon Processor (Cascade Lake, 2019), with "stepping" levels 6
54 or 7 only. (The Cascade Lake Xeon processor with *stepping 5 is
55 vulnerable to MDS variants*.)
57 ``Skylake-Server``, ``Skylake-Server-IBRS``, ``Skylake-Server-IBRS-noTSX``
58 Intel Xeon Processor (Skylake, 2016)
60 ``Skylake-Client``, ``Skylake-Client-IBRS``, ``Skylake-Client-noTSX-IBRS}``
61 Intel Core Processor (Skylake, 2015)
63 ``Broadwell``, ``Broadwell-IBRS``, ``Broadwell-noTSX``, ``Broadwell-noTSX-IBRS``
64 Intel Core Processor (Broadwell, 2014)
66 ``Haswell``, ``Haswell-IBRS``, ``Haswell-noTSX``, ``Haswell-noTSX-IBRS``
67 Intel Core Processor (Haswell, 2013)
69 ``IvyBridge``, ``IvyBridge-IBR``
70 Intel Xeon E3-12xx v2 (Ivy Bridge, 2012)
72 ``SandyBridge``, ``SandyBridge-IBRS``
73 Intel Xeon E312xx (Sandy Bridge, 2011)
75 ``Westmere``, ``Westmere-IBRS``
76 Westmere E56xx/L56xx/X56xx (Nehalem-C, 2010)
78 ``Nehalem``, ``Nehalem-IBRS``
79 Intel Core i7 9xx (Nehalem Class Core i7, 2008)
82 Intel Core 2 Duo P9xxx (Penryn Class Core 2, 2007)
85 Intel Celeron_4x0 (Conroe/Merom Class Core 2, 2006)
88 Important CPU features for Intel x86 hosts
89 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
91 The following are important CPU features that should be used on Intel
92 x86 hosts, when available in the host CPU. Some of them require explicit
93 configuration to enable, as they are not included by default in some, or
94 all, of the named CPU models listed above. In general all of these
95 features are included if using "Host passthrough" or "Host model".
98 Recommended to mitigate the cost of the Meltdown (CVE-2017-5754) fix.
100 Included by default in Haswell, Broadwell & Skylake Intel CPU models.
102 Should be explicitly turned on for Westmere, SandyBridge, and
103 IvyBridge Intel CPU models. Note that some desktop/mobile Westmere
104 CPUs cannot support this feature.
107 Required to enable the Spectre v2 (CVE-2017-5715) fix.
109 Included by default in Intel CPU models with -IBRS suffix.
111 Must be explicitly turned on for Intel CPU models without -IBRS
114 Requires the host CPU microcode to support this feature before it
115 can be used for guest CPUs.
118 Required to enable stronger Spectre v2 (CVE-2017-5715) fixes in some
121 Must be explicitly turned on for all Intel CPU models.
123 Requires the host CPU microcode to support this feature before it can
124 be used for guest CPUs.
127 Required to enable the CVE-2018-3639 fix.
129 Not included by default in any Intel CPU model.
131 Must be explicitly turned on for all Intel CPU models.
133 Requires the host CPU microcode to support this feature before it
134 can be used for guest CPUs.
137 Recommended to allow guest OS to use 1GB size pages.
139 Not included by default in any Intel CPU model.
141 Should be explicitly turned on for all Intel CPU models.
143 Note that not all CPU hardware will support this feature.
146 Required to confirm the MDS (CVE-2018-12126, CVE-2018-12127,
147 CVE-2018-12130, CVE-2019-11091) fixes.
149 Not included by default in any Intel CPU model.
151 Must be explicitly turned on for all Intel CPU models.
153 Requires the host CPU microcode to support this feature before it
154 can be used for guest CPUs.
157 Recommended to inform the guest OS that the host is *not* vulnerable
158 to any of the MDS variants ([MFBDS] CVE-2018-12130, [MLPDS]
159 CVE-2018-12127, [MSBDS] CVE-2018-12126).
161 This is an MSR (Model-Specific Register) feature rather than a CPUID feature,
162 so it will not appear in the Linux ``/proc/cpuinfo`` in the host or
163 guest. Instead, the host kernel uses it to populate the MDS
164 vulnerability file in ``sysfs``.
166 So it should only be enabled for VMs if the host reports @code{Not
167 affected} in the ``/sys/devices/system/cpu/vulnerabilities/mds`` file.
170 Recommended to inform that the guest that the host is ``not``
171 vulnerable to CVE-2019-11135, TSX Asynchronous Abort (TAA).
173 This too is an MSR feature, so it does not show up in the Linux
174 ``/proc/cpuinfo`` in the host or guest.
176 It should only be enabled for VMs if the host reports ``Not affected``
177 in the ``/sys/devices/system/cpu/vulnerabilities/tsx_async_abort``
181 Recommended to inform the guest that it can disable the Intel TSX
182 (Transactional Synchronization Extensions) feature; or, if the
183 processor is vulnerable, use the Intel VERW instruction (a
184 processor-level instruction that performs checks on memory access) as
185 a mitigation for the TAA vulnerability. (For details, refer to
186 Intel's `deep dive into MDS
187 <https://software.intel.com/security-software-guidance/insights/deep-dive-intel-analysis-microarchitectural-data-sampling>`_.)
189 Expose this to the guest OS if and only if: (a) the host has TSX
190 enabled; *and* (b) the guest has ``rtm`` CPU flag enabled.
192 By disabling TSX, KVM-based guests can avoid paying the price of
193 mitigating TSX-based attacks.
195 Note that ``tsx-ctrl`` too is an MSR feature, so it does not show
196 up in the Linux ``/proc/cpuinfo`` in the host or guest.
198 To validate that Intel TSX is indeed disabled for the guest, there are
199 two ways: (a) check for the *absence* of ``rtm`` in the guest's
200 ``/proc/cpuinfo``; or (b) the
201 ``/sys/devices/system/cpu/vulnerabilities/tsx_async_abort`` file in
202 the guest should report ``Mitigation: TSX disabled``.
205 Preferred CPU models for AMD x86 hosts
206 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
208 The following CPU models are preferred for use on Intel hosts.
209 Administrators / applications are recommended to use the CPU model that
210 matches the generation of the host CPUs in use. In a deployment with a
211 mixture of host CPU models between machines, if live migration
212 compatibility is required, use the newest CPU model that is compatible
213 across all desired hosts.
215 ``EPYC``, ``EPYC-IBPB``
216 AMD EPYC Processor (2017)
219 AMD Opteron 63xx class CPU (2012)
222 AMD Opteron 62xx class CPU (2011)
225 AMD Opteron 23xx (Gen 3 Class Opteron, 2009)
228 AMD Opteron 22xx (Gen 2 Class Opteron, 2006)
231 AMD Opteron 240 (Gen 1 Class Opteron, 2004)
234 Important CPU features for AMD x86 hosts
235 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
237 The following are important CPU features that should be used on AMD x86
238 hosts, when available in the host CPU. Some of them require explicit
239 configuration to enable, as they are not included by default in some, or
240 all, of the named CPU models listed above. In general all of these
241 features are included if using "Host passthrough" or "Host model".
244 Required to enable the Spectre v2 (CVE-2017-5715) fix.
246 Included by default in AMD CPU models with -IBPB suffix.
248 Must be explicitly turned on for AMD CPU models without -IBPB suffix.
250 Requires the host CPU microcode to support this feature before it
251 can be used for guest CPUs.
254 Required to enable stronger Spectre v2 (CVE-2017-5715) fixes in some
257 Must be explicitly turned on for all AMD CPU models.
259 Requires the host CPU microcode to support this feature before it
260 can be used for guest CPUs.
263 Required to enable the CVE-2018-3639 fix
265 Not included by default in any AMD CPU model.
267 Must be explicitly turned on for all AMD CPU models.
269 This should be provided to guests, even if amd-ssbd is also provided,
270 for maximum guest compatibility.
272 Note for some QEMU / libvirt versions, this must be force enabled when
273 when using "Host model", because this is a virtual feature that
274 doesn't exist in the physical host CPUs.
277 Required to enable the CVE-2018-3639 fix
279 Not included by default in any AMD CPU model.
281 Must be explicitly turned on for all AMD CPU models.
283 This provides higher performance than ``virt-ssbd`` so should be
284 exposed to guests whenever available in the host. ``virt-ssbd`` should
285 none the less also be exposed for maximum guest compatibility as some
286 kernels only know about ``virt-ssbd``.
289 Recommended to indicate the host is not vulnerable CVE-2018-3639
291 Not included by default in any AMD CPU model.
293 Future hardware generations of CPU will not be vulnerable to
294 CVE-2018-3639, and thus the guest should be told not to enable
295 its mitigations, by exposing amd-no-ssb. This is mutually
296 exclusive with virt-ssbd and amd-ssbd.
299 Recommended to allow guest OS to use 1GB size pages
301 Not included by default in any AMD CPU model.
303 Should be explicitly turned on for all AMD CPU models.
305 Note that not all CPU hardware will support this feature.
308 Default x86 CPU models
309 ^^^^^^^^^^^^^^^^^^^^^^
311 The default QEMU CPU models are designed such that they can run on all
312 hosts. If an application does not wish to do perform any host
313 compatibility checks before launching guests, the default is guaranteed
316 The default CPU models will, however, leave the guest OS vulnerable to
317 various CPU hardware flaws, so their use is strongly discouraged.
318 Applications should follow the earlier guidance to setup a better CPU
319 configuration, with host passthrough recommended if live migration is
322 ``qemu32``, ``qemu64``
323 QEMU Virtual CPU version 2.5+ (32 & 64 bit variants)
325 ``qemu64`` is used for x86_64 guests and ``qemu32`` is used for i686
326 guests, when no ``-cpu`` argument is given to QEMU, or no ``<cpu>`` is
327 provided in libvirt XML.
329 Other non-recommended x86 CPUs
330 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
332 The following CPUs models are compatible with most AMD and Intel x86
333 hosts, but their usage is discouraged, as they expose a very limited
334 featureset, which prevents guests having optimal performance.
337 Common KVM processor (32 & 64 bit variants).
339 Legacy models just for historical compatibility with ancient QEMU
342 ``486``, ``athlon``, ``phenom``, ``coreduo``, ``core2duo``, ``n270``, ``pentium``, ``pentium2``, ``pentium3``
343 Various very old x86 CPU models, mostly predating the introduction
344 of hardware assisted virtualization, that should thus not be
345 required for running virtual machines.
348 Syntax for configuring CPU models
349 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
351 The examples below illustrate the approach to configuring the various
352 CPU models / features in QEMU and libvirt.
361 |qemu_system| -cpu host
363 Host passthrough with feature customization:
367 |qemu_system| -cpu host,-vmx,...
373 |qemu_system| -cpu Westmere
375 Named CPU models with feature customization:
379 |qemu_system| -cpu Westmere,+pcid,...
386 <cpu mode='host-passthrough'/>
388 Host passthrough with feature customization::
390 <cpu mode='host-passthrough'>
391 <feature name="vmx" policy="disable"/>
397 <cpu mode='host-model'/>
399 Host model with feature customization::
401 <cpu mode='host-model'>
402 <feature name="vmx" policy="disable"/>
409 <model name="Westmere"/>
412 Named model with feature customization::
415 <model name="Westmere"/>
416 <feature name="pcid" policy="require"/>