1QEMU-CPU-MODELS.7(7)                                      QEMU-CPU-MODELS.7(7)
2
3
4

NAME

6       qemu-cpu-models - QEMU / KVM CPU model configuration
7

SYNOPSIS

9       QEMU / KVM CPU model configuration
10

DESCRIPTION

12       QEMU / KVM virtualization supports two ways to configure CPU models
13
14       Host passthrough
15           This passes the host CPU model features, model, stepping, exactly
16           to the guest. Note that KVM may filter out some host CPU model
17           features if they cannot be supported with virtualization. Live
18           migration is unsafe when this mode is used as libvirt / QEMU cannot
19           guarantee a stable CPU is exposed to the guest across hosts. This
20           is the recommended CPU to use, provided live migration is not
21           required.
22
23       Named model
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.
29
30       In both cases, it is possible to optionally add or remove individual
31       CPU features, to alter what is presented to the guest by default.
32
33       Libvirt supports a third way to configure CPU models known as "Host
34       model".  This uses the QEMU "Named model" feature, automatically
35       picking a CPU model that is similar the host CPU, and then adding extra
36       features to approximate the host model as closely as possible. This
37       does not guarantee the CPU family, stepping, etc will precisely match
38       the host CPU, as they would with "Host passthrough", but gives much of
39       the benefit of passthrough, while making live migration safe.
40
41       Recommendations for KVM CPU model configuration on x86 hosts
42
43       The information that follows provides recommendations for configuring
44       CPU models on x86 hosts. The goals are to maximise performance, while
45       protecting guest OS against various CPU hardware flaws, and optionally
46       enabling live migration between hosts with heterogeneous CPU models.
47
48       Preferred CPU models for Intel x86 hosts
49
50       The following CPU models are preferred for use on Intel hosts.
51       Administrators / applications are recommended to use the CPU model that
52       matches the generation of the host CPUs in use. In a deployment with a
53       mixture of host CPU models between machines, if live migration
54       compatibility is required, use the newest CPU model that is compatible
55       across all desired hosts.
56
57       "Skylake-Server"
58       "Skylake-Server-IBRS"
59           Intel Xeon Processor (Skylake, 2016)
60
61       "Skylake-Client"
62       "Skylake-Client-IBRS"
63           Intel Core Processor (Skylake, 2015)
64
65       "Broadwell"
66       "Broadwell-IBRS"
67       "Broadwell-noTSX"
68       "Broadwell-noTSX-IBRS"
69           Intel Core Processor (Broadwell, 2014)
70
71       "Haswell"
72       "Haswell-IBRS"
73       "Haswell-noTSX"
74       "Haswell-noTSX-IBRS"
75           Intel Core Processor (Haswell, 2013)
76
77       "IvyBridge"
78       "IvyBridge-IBRS"
79           Intel Xeon E3-12xx v2 (Ivy Bridge, 2012)
80
81       "SandyBridge"
82       "SandyBridge-IBRS"
83           Intel Xeon E312xx (Sandy Bridge, 2011)
84
85       "Westmere"
86       "Westmere-IBRS"
87           Westmere E56xx/L56xx/X56xx (Nehalem-C, 2010)
88
89       "Nehalem"
90       "Nehalem-IBRS"
91           Intel Core i7 9xx (Nehalem Class Core i7, 2008)
92
93       "Penryn"
94           Intel Core 2 Duo P9xxx (Penryn Class Core 2, 2007)
95
96       "Conroe"
97           Intel Celeron_4x0 (Conroe/Merom Class Core 2, 2006)
98
99       Important CPU features for Intel x86 hosts
100
101       The following are important CPU features that should be used on Intel
102       x86 hosts, when available in the host CPU. Some of them require
103       explicit configuration to enable, as they are not included by default
104       in some, or all, of the named CPU models listed above. In general all
105       of these features are included if using "Host passthrough" or "Host
106       model".
107
108       "pcid"
109           Recommended to mitigate the cost of the Meltdown (CVE-2017-5754)
110           fix
111
112           Included by default in Haswell, Broadwell & Skylake Intel CPU
113           models.
114
115           Should be explicitly turned on for Westmere, SandyBridge, and
116           IvyBridge Intel CPU models. Note that some desktop/mobile Westmere
117           CPUs cannot support this feature.
118
119       "spec-ctrl"
120           Required to enable the Spectre v2 (CVE-2017-5715) fix.
121
122           Included by default in Intel CPU models with -IBRS suffix.
123
124           Must be explicitly turned on for Intel CPU models without -IBRS
125           suffix.
126
127           Requires the host CPU microcode to support this feature before it
128           can be used for guest CPUs.
129
130       "stibp"
131           Required to enable stronger Spectre v2 (CVE-2017-5715) fixes in
132           some operating systems.
133
134           Must be explicitly turned on for all Intel CPU models.
135
136           Requires the host CPU microcode to support this feature before it
137           can be used for guest CPUs.
138
139       "ssbd"
140           Required to enable the CVE-2018-3639 fix
141
142           Not included by default in any Intel CPU model.
143
144           Must be explicitly turned on for all Intel CPU models.
145
146           Requires the host CPU microcode to support this feature before it
147           can be used for guest CPUs.
148
149       "pdpe1gb"
150           Recommended to allow guest OS to use 1GB size pages
151
152           Not included by default in any Intel CPU model.
153
154           Should be explicitly turned on for all Intel CPU models.
155
156           Note that not all CPU hardware will support this feature.
157
158       "md-clear"
159           Required to confirm the MDS (CVE-2018-12126, CVE-2018-12127,
160           CVE-2018-12130, CVE-2019-11091) fixes.
161
162           Not included by default in any Intel CPU model.
163
164           Must be explicitly turned on for all Intel CPU models.
165
166           Requires the host CPU microcode to support this feature before it
167           can be used for guest CPUs.
168
169       Preferred CPU models for AMD x86 hosts
170
171       The following CPU models are preferred for use on Intel hosts.
172       Administrators / applications are recommended to use the CPU model that
173       matches the generation of the host CPUs in use. In a deployment with a
174       mixture of host CPU models between machines, if live migration
175       compatibility is required, use the newest CPU model that is compatible
176       across all desired hosts.
177
178       "EPYC"
179       "EPYC-IBPB"
180           AMD EPYC Processor (2017)
181
182       "Opteron_G5"
183           AMD Opteron 63xx class CPU (2012)
184
185       "Opteron_G4"
186           AMD Opteron 62xx class CPU (2011)
187
188       "Opteron_G3"
189           AMD Opteron 23xx (Gen 3 Class Opteron, 2009)
190
191       "Opteron_G2"
192           AMD Opteron 22xx (Gen 2 Class Opteron, 2006)
193
194       "Opteron_G1"
195           AMD Opteron 240 (Gen 1 Class Opteron, 2004)
196
197       Important CPU features for AMD x86 hosts
198
199       The following are important CPU features that should be used on AMD x86
200       hosts, when available in the host CPU. Some of them require explicit
201       configuration to enable, as they are not included by default in some,
202       or all, of the named CPU models listed above. In general all of these
203       features are included if using "Host passthrough" or "Host model".
204
205       "ibpb"
206           Required to enable the Spectre v2 (CVE-2017-5715) fix.
207
208           Included by default in AMD CPU models with -IBPB suffix.
209
210           Must be explicitly turned on for AMD CPU models without -IBPB
211           suffix.
212
213           Requires the host CPU microcode to support this feature before it
214           can be used for guest CPUs.
215
216       "stibp"
217           Required to enable stronger Spectre v2 (CVE-2017-5715) fixes in
218           some operating systems.
219
220           Must be explicitly turned on for all AMD CPU models.
221
222           Requires the host CPU microcode to support this feature before it
223           can be used for guest CPUs.
224
225       "virt-ssbd"
226           Required to enable the CVE-2018-3639 fix
227
228           Not included by default in any AMD CPU model.
229
230           Must be explicitly turned on for all AMD CPU models.
231
232           This should be provided to guests, even if amd-ssbd is also
233           provided, for maximum guest compatibility.
234
235           Note for some QEMU / libvirt versions, this must be force enabled
236           when when using "Host model", because this is a virtual feature
237           that doesn't exist in the physical host CPUs.
238
239       "amd-ssbd"
240           Required to enable the CVE-2018-3639 fix
241
242           Not included by default in any AMD CPU model.
243
244           Must be explicitly turned on for all AMD CPU models.
245
246           This provides higher performance than virt-ssbd so should be
247           exposed to guests whenever available in the host. virt-ssbd should
248           none the less also be exposed for maximum guest compatibility as
249           some kernels only know about virt-ssbd.
250
251       "amd-no-ssb"
252           Recommended to indicate the host is not vulnerable CVE-2018-3639
253
254           Not included by default in any AMD CPU model.
255
256           Future hardware generations of CPU will not be vulnerable to
257           CVE-2018-3639, and thus the guest should be told not to enable its
258           mitigations, by exposing amd-no-ssb. This is mutually exclusive
259           with virt-ssbd and amd-ssbd.
260
261       "pdpe1gb"
262           Recommended to allow guest OS to use 1GB size pages
263
264           Not included by default in any AMD CPU model.
265
266           Should be explicitly turned on for all AMD CPU models.
267
268           Note that not all CPU hardware will support this feature.
269
270       Default x86 CPU models
271
272       The default QEMU CPU models are designed such that they can run on all
273       hosts.  If an application does not wish to do perform any host
274       compatibility checks before launching guests, the default is guaranteed
275       to work.
276
277       The default CPU models will, however, leave the guest OS vulnerable to
278       various CPU hardware flaws, so their use is strongly discouraged.
279       Applications should follow the earlier guidance to setup a better CPU
280       configuration, with host passthrough recommended if live migration is
281       not needed.
282
283       "qemu32"
284       "qemu64"
285           QEMU Virtual CPU version 2.5+ (32 & 64 bit variants)
286
287           qemu64 is used for x86_64 guests and qemu32 is used for i686
288           guests, when no -cpu argument is given to QEMU, or no <cpu> is
289           provided in libvirt XML.
290
291       Other non-recommended x86 CPUs
292
293       The following CPUs models are compatible with most AMD and Intel x86
294       hosts, but their usage is discouraged, as they expose a very limited
295       featureset, which prevents guests having optimal performance.
296
297       "kvm32"
298       "kvm64"
299           Common KVM processor (32 & 64 bit variants)
300
301           Legacy models just for historical compatibility with ancient QEMU
302           versions.
303
304       486
305       "athlon"
306       "phenom"
307       "coreduo"
308       "core2duo"
309       "n270"
310       "pentium"
311       "pentium2"
312       "pentium3"
313           Various very old x86 CPU models, mostly predating the introduction
314           of hardware assisted virtualization, that should thus not be
315           required for running virtual machines.
316
317       Supported CPU model configurations on MIPS hosts
318
319       QEMU supports variety of MIPS CPU models:
320
321       Supported CPU models for MIPS32 hosts
322
323       The following CPU models are supported for use on MIPS32 hosts.
324       Administrators / applications are recommended to use the CPU model that
325       matches the generation of the host CPUs in use. In a deployment with a
326       mixture of host CPU models between machines, if live migration
327       compatibility is required, use the newest CPU model that is compatible
328       across all desired hosts.
329
330       "mips32r6-generic"
331           MIPS32 Processor (Release 6, 2015)
332
333       "P5600"
334           MIPS32 Processor (P5600, 2014)
335
336       "M14K"
337       "M14Kc"
338           MIPS32 Processor (M14K, 2009)
339
340       "74Kf"
341           MIPS32 Processor (74K, 2007)
342
343       "34Kf"
344           MIPS32 Processor (34K, 2006)
345
346       "24Kc"
347       "24KEc"
348       "24Kf"
349           MIPS32 Processor (24K, 2003)
350
351       "4Kc"
352       "4Km"
353       "4KEcR1"
354       "4KEmR1"
355       "4KEc"
356       "4KEm"
357           MIPS32 Processor (4K, 1999)
358
359       Supported CPU models for MIPS64 hosts
360
361       The following CPU models are supported for use on MIPS64 hosts.
362       Administrators / applications are recommended to use the CPU model that
363       matches the generation of the host CPUs in use. In a deployment with a
364       mixture of host CPU models between machines, if live migration
365       compatibility is required, use the newest CPU model that is compatible
366       across all desired hosts.
367
368       "I6400"
369           MIPS64 Processor (Release 6, 2014)
370
371       "Loongson-2F"
372           MIPS64 Processor (Loongson 2, 2008)
373
374       "Loongson-2E"
375           MIPS64 Processor (Loongson 2, 2006)
376
377       "mips64dspr2"
378           MIPS64 Processor (Release 2, 2006)
379
380       "MIPS64R2-generic"
381       "5KEc"
382       "5KEf"
383           MIPS64 Processor (Release 2, 2002)
384
385       "20Kc"
386           MIPS64 Processor (20K, 2000)
387
388       "5Kc"
389       "5Kf"
390           MIPS64 Processor (5K, 1999)
391
392       "VR5432"
393           MIPS64 Processor (VR, 1998)
394
395       "R4000"
396           MIPS64 Processor (MIPS III, 1991)
397
398       Supported CPU models for nanoMIPS hosts
399
400       The following CPU models are supported for use on nanoMIPS hosts.
401       Administrators / applications are recommended to use the CPU model that
402       matches the generation of the host CPUs in use. In a deployment with a
403       mixture of host CPU models between machines, if live migration
404       compatibility is required, use the newest CPU model that is compatible
405       across all desired hosts.
406
407       "I7200"
408           MIPS I7200 (nanoMIPS, 2018)
409
410       Preferred CPU models for MIPS hosts
411
412       The following CPU models are preferred for use on different MIPS hosts:
413
414       "MIPS III"
415           R4000
416
417       "MIPS32R2"
418           34Kf
419
420       "MIPS64R6"
421           I6400
422
423       "nanoMIPS"
424           I7200
425
426       Syntax for configuring CPU models
427
428       The example below illustrate the approach to configuring the various
429       CPU models / features in QEMU and libvirt
430
431       QEMU command line
432
433       Host passthrough
434                      $ qemu-system-x86_64 -cpu host
435
436           With feature customization:
437
438                      $ qemu-system-x86_64 -cpu host,-vmx,...
439
440       Named CPU models
441                      $ qemu-system-x86_64 -cpu Westmere
442
443           With feature customization:
444
445                      $ qemu-system-x86_64 -cpu Westmere,+pcid,...
446
447       Libvirt guest XML
448
449       Host passthrough
450                      <cpu mode='host-passthrough'/>
451
452           With feature customization:
453
454                      <cpu mode='host-passthrough'>
455                          <feature name="vmx" policy="disable"/>
456                          ...
457                      </cpu>
458
459       Host model
460                      <cpu mode='host-model'/>
461
462           With feature customization:
463
464                      <cpu mode='host-model'>
465                          <feature name="vmx" policy="disable"/>
466                          ...
467                      </cpu>
468
469       Named model
470                      <cpu mode='custom'>
471                          <model name="Westmere"/>
472                      </cpu>
473
474           With feature customization:
475
476                      <cpu mode='custom'>
477                          <model name="Westmere"/>
478                          <feature name="pcid" policy="require"/>
479                          ...
480                      </cpu>
481

SEE ALSO

483       The HTML documentation of QEMU for more precise information and Linux
484       user mode emulator invocation.
485

AUTHOR

487       Daniel P. Berrange
488
489
490
491                                  2020-03-17              QEMU-CPU-MODELS.7(7)