1 # SPDX-License-Identifier: GPL-2.0
6 source "virt/kvm/Kconfig"
8 menuconfig VIRTUALIZATION
11 Say Y here to get to see options for using your Linux host to run
12 other operating systems inside virtual machines (guests).
13 This option alone does not add any kernel code.
15 If you say N, all options in this submenu will be skipped and
22 select PREEMPT_NOTIFIERS
23 select HAVE_KVM_EVENTFD
24 select HAVE_KVM_VCPU_ASYNC_IOCTL
27 select IRQ_BYPASS_MANAGER
28 select HAVE_KVM_IRQ_BYPASS
31 config KVM_BOOK3S_HANDLER
34 config KVM_BOOK3S_32_HANDLER
36 select KVM_BOOK3S_HANDLER
39 config KVM_BOOK3S_64_HANDLER
41 select KVM_BOOK3S_HANDLER
43 config KVM_BOOK3S_PR_POSSIBLE
48 config KVM_BOOK3S_HV_POSSIBLE
52 tristate "KVM support for PowerPC book3s_32 processors"
53 depends on PPC_BOOK3S_32 && !SMP && !PTE_64BIT
54 depends on !CONTEXT_TRACKING_USER
56 select KVM_BOOK3S_32_HANDLER
57 select KVM_BOOK3S_PR_POSSIBLE
60 Support running unmodified book3s_32 guest kernels
61 in virtual machines on book3s_32 host processors.
63 This module provides access to the hardware capabilities through
64 a character device node named /dev/kvm.
69 tristate "KVM support for PowerPC book3s_64 processors"
70 depends on PPC_BOOK3S_64
71 select KVM_BOOK3S_64_HANDLER
73 select KVM_BOOK3S_PR_POSSIBLE if !KVM_BOOK3S_HV_POSSIBLE
74 select PPC_64S_HASH_MMU
75 select SPAPR_TCE_IOMMU if IOMMU_SUPPORT && (PPC_PSERIES || PPC_POWERNV)
77 Support running unmodified book3s_64 and book3s_32 guest kernels
78 in virtual machines on book3s_64 host processors.
80 This module provides access to the hardware capabilities through
81 a character device node named /dev/kvm.
85 config KVM_BOOK3S_64_HV
86 tristate "KVM for POWER7 and later using hypervisor mode in host"
87 depends on KVM_BOOK3S_64 && PPC_POWERNV
88 select KVM_BOOK3S_HV_POSSIBLE
92 Support running unmodified book3s_64 guest kernels in
93 virtual machines on POWER7 and newer processors that have
94 hypervisor mode available to the host.
96 If you say Y here, KVM will use the hardware virtualization
97 facilities of POWER7 (and later) processors, meaning that
98 guest operating systems will run at full hardware speed
99 using supervisor and user modes. However, this also means
100 that KVM is not usable under PowerVM (pHyp), is only usable
101 on POWER7 or later processors, and cannot emulate a
102 different processor from the host processor.
106 config KVM_BOOK3S_64_PR
107 tristate "KVM support without using hypervisor mode in host"
108 depends on KVM_BOOK3S_64
109 depends on !CONTEXT_TRACKING_USER
110 select KVM_BOOK3S_PR_POSSIBLE
112 Support running guest kernels in virtual machines on processors
113 without using hypervisor mode in the host, by running the
114 guest in user mode (problem state) and emulating all
115 privileged instructions and registers.
117 This is only available for hash MMU mode and only supports
118 guests that use hash MMU mode.
120 This is not as fast as using hypervisor mode, but works on
121 machines where hypervisor mode is not available or not usable,
122 and can emulate processors that are different from the host
123 processor, including emulating 32-bit processors on a 64-bit
126 Selecting this option will cause the SCV facility to be
127 disabled when the kernel is booted on the pseries platform in
128 hash MMU mode (regardless of PR VMs running). When any PR VMs
129 are running, "AIL" mode is disabled which may slow interrupts
130 and system calls on the host.
132 config KVM_BOOK3S_HV_EXIT_TIMING
135 config KVM_BOOK3S_HV_P9_TIMING
136 bool "Detailed timing for the P9 entry point"
137 select KVM_BOOK3S_HV_EXIT_TIMING
138 depends on KVM_BOOK3S_HV_POSSIBLE && DEBUG_FS
140 Calculate time taken for each vcpu during vcpu entry and
141 exit, time spent inside the guest and time spent handling
142 hypercalls and page faults. The total, minimum and maximum
143 times in nanoseconds together with the number of executions
144 are reported in debugfs in kvm/vm#/vcpu#/timings.
148 config KVM_BOOK3S_HV_P8_TIMING
149 bool "Detailed timing for hypervisor real-mode code (for POWER8)"
150 select KVM_BOOK3S_HV_EXIT_TIMING
151 depends on KVM_BOOK3S_HV_POSSIBLE && DEBUG_FS && !KVM_BOOK3S_HV_P9_TIMING
153 Calculate time taken for each vcpu in the real-mode guest entry,
154 exit, and interrupt handling code, plus time spent in the guest
155 and in nap mode due to idle (cede) while other threads are still
156 in the guest. The total, minimum and maximum times in nanoseconds
157 together with the number of executions are reported in debugfs in
158 kvm/vm#/vcpu#/timings. The overhead is of the order of 30 - 40
159 ns per exit on POWER8.
163 config KVM_BOOK3S_HV_NESTED_PMU_WORKAROUND
164 bool "Nested L0 host workaround for L1 KVM host PMU handling bug" if EXPERT
165 depends on KVM_BOOK3S_HV_POSSIBLE
168 Old nested HV capable Linux guests have a bug where they don't
169 reflect the PMU in-use status of their L2 guest to the L0 host
170 while the L2 PMU registers are live. This can result in loss
171 of L2 PMU register state, causing perf to not work correctly in
174 Selecting this option for the L0 host implements a workaround for
175 those buggy L1s which saves the L2 state, at the cost of performance
176 in all nested-capable guest entry/exit.
181 config KVM_EXIT_TIMING
182 bool "Detailed exit timing"
183 depends on KVM_E500V2 || KVM_E500MC
185 Calculate elapsed time for every exit/enter cycle. A per-vcpu
186 report is available in debugfs kvm/vm#_vcpu#_timing.
187 The overhead is relatively small, however it is not recommended for
188 production environments.
193 bool "KVM support for PowerPC E500v2 processors"
194 depends on PPC_E500 && !PPC_E500MC
195 depends on !CONTEXT_TRACKING_USER
200 Support running unmodified E500 guest kernels in virtual machines on
201 E500v2 host processors.
203 This module provides access to the hardware capabilities through
204 a character device node named /dev/kvm.
209 bool "KVM support for PowerPC E500MC/E5500/E6500 processors"
210 depends on PPC_E500MC
211 depends on !CONTEXT_TRACKING_USER
217 Support running unmodified E500MC/E5500/E6500 guest kernels in
218 virtual machines on E500MC/E5500/E6500 host processors.
220 This module provides access to the hardware capabilities through
221 a character device node named /dev/kvm.
226 bool "KVM in-kernel MPIC emulation"
227 depends on KVM && PPC_E500
228 select HAVE_KVM_IRQCHIP
229 select HAVE_KVM_IRQFD
230 select HAVE_KVM_IRQ_ROUTING
233 Enable support for emulating MPIC devices inside the
234 host kernel, rather than relying on userspace to emulate.
235 Currently, support is limited to certain versions of
236 Freescale's MPIC implementation.
239 bool "KVM in-kernel XICS emulation"
240 depends on KVM_BOOK3S_64 && !KVM_MPIC
241 select HAVE_KVM_IRQCHIP
242 select HAVE_KVM_IRQFD
245 Include support for the XICS (eXternal Interrupt Controller
246 Specification) interrupt controller architecture used on
247 IBM POWER (pSeries) servers.
252 depends on KVM_XICS && PPC_XIVE_NATIVE && KVM_BOOK3S_HV_POSSIBLE
254 endif # VIRTUALIZATION