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
26 select IRQ_BYPASS_MANAGER
27 select HAVE_KVM_IRQ_BYPASS
30 config KVM_BOOK3S_HANDLER
33 config KVM_BOOK3S_32_HANDLER
35 select KVM_BOOK3S_HANDLER
38 config KVM_BOOK3S_64_HANDLER
40 select KVM_BOOK3S_HANDLER
42 config KVM_BOOK3S_PR_POSSIBLE
47 config KVM_BOOK3S_HV_POSSIBLE
51 tristate "KVM support for PowerPC book3s_32 processors"
52 depends on PPC_BOOK3S_32 && !SMP && !PTE_64BIT
53 depends on !CONTEXT_TRACKING_USER
55 select KVM_BOOK3S_32_HANDLER
56 select KVM_BOOK3S_PR_POSSIBLE
59 Support running unmodified book3s_32 guest kernels
60 in virtual machines on book3s_32 host processors.
62 This module provides access to the hardware capabilities through
63 a character device node named /dev/kvm.
68 tristate "KVM support for PowerPC book3s_64 processors"
69 depends on PPC_BOOK3S_64
70 select KVM_BOOK3S_64_HANDLER
72 select KVM_BOOK3S_PR_POSSIBLE if !KVM_BOOK3S_HV_POSSIBLE
73 select PPC_64S_HASH_MMU
74 select SPAPR_TCE_IOMMU if IOMMU_SUPPORT && (PPC_PSERIES || PPC_POWERNV)
76 Support running unmodified book3s_64 and book3s_32 guest kernels
77 in virtual machines on book3s_64 host processors.
79 This module provides access to the hardware capabilities through
80 a character device node named /dev/kvm.
84 config KVM_BOOK3S_64_HV
85 tristate "KVM for POWER7 and later using hypervisor mode in host"
86 depends on KVM_BOOK3S_64 && PPC_POWERNV
87 select KVM_BOOK3S_HV_POSSIBLE
91 Support running unmodified book3s_64 guest kernels in
92 virtual machines on POWER7 and newer processors that have
93 hypervisor mode available to the host.
95 If you say Y here, KVM will use the hardware virtualization
96 facilities of POWER7 (and later) processors, meaning that
97 guest operating systems will run at full hardware speed
98 using supervisor and user modes. However, this also means
99 that KVM is not usable under PowerVM (pHyp), is only usable
100 on POWER7 or later processors, and cannot emulate a
101 different processor from the host processor.
105 config KVM_BOOK3S_64_PR
106 tristate "KVM support without using hypervisor mode in host"
107 depends on KVM_BOOK3S_64
108 depends on !CONTEXT_TRACKING_USER
109 select KVM_BOOK3S_PR_POSSIBLE
111 Support running guest kernels in virtual machines on processors
112 without using hypervisor mode in the host, by running the
113 guest in user mode (problem state) and emulating all
114 privileged instructions and registers.
116 This is only available for hash MMU mode and only supports
117 guests that use hash MMU mode.
119 This is not as fast as using hypervisor mode, but works on
120 machines where hypervisor mode is not available or not usable,
121 and can emulate processors that are different from the host
122 processor, including emulating 32-bit processors on a 64-bit
125 Selecting this option will cause the SCV facility to be
126 disabled when the kernel is booted on the pseries platform in
127 hash MMU mode (regardless of PR VMs running). When any PR VMs
128 are running, "AIL" mode is disabled which may slow interrupts
129 and system calls on the host.
131 config KVM_BOOK3S_HV_EXIT_TIMING
134 config KVM_BOOK3S_HV_P9_TIMING
135 bool "Detailed timing for the P9 entry point"
136 select KVM_BOOK3S_HV_EXIT_TIMING
137 depends on KVM_BOOK3S_HV_POSSIBLE && DEBUG_FS
139 Calculate time taken for each vcpu during vcpu entry and
140 exit, time spent inside the guest and time spent handling
141 hypercalls and page faults. The total, minimum and maximum
142 times in nanoseconds together with the number of executions
143 are reported in debugfs in kvm/vm#/vcpu#/timings.
147 config KVM_BOOK3S_HV_P8_TIMING
148 bool "Detailed timing for hypervisor real-mode code (for POWER8)"
149 select KVM_BOOK3S_HV_EXIT_TIMING
150 depends on KVM_BOOK3S_HV_POSSIBLE && DEBUG_FS && !KVM_BOOK3S_HV_P9_TIMING
152 Calculate time taken for each vcpu in the real-mode guest entry,
153 exit, and interrupt handling code, plus time spent in the guest
154 and in nap mode due to idle (cede) while other threads are still
155 in the guest. The total, minimum and maximum times in nanoseconds
156 together with the number of executions are reported in debugfs in
157 kvm/vm#/vcpu#/timings. The overhead is of the order of 30 - 40
158 ns per exit on POWER8.
162 config KVM_BOOK3S_HV_NESTED_PMU_WORKAROUND
163 bool "Nested L0 host workaround for L1 KVM host PMU handling bug" if EXPERT
164 depends on KVM_BOOK3S_HV_POSSIBLE
167 Old nested HV capable Linux guests have a bug where they don't
168 reflect the PMU in-use status of their L2 guest to the L0 host
169 while the L2 PMU registers are live. This can result in loss
170 of L2 PMU register state, causing perf to not work correctly in
173 Selecting this option for the L0 host implements a workaround for
174 those buggy L1s which saves the L2 state, at the cost of performance
175 in all nested-capable guest entry/exit.
180 config KVM_EXIT_TIMING
181 bool "Detailed exit timing"
182 depends on KVM_E500V2 || KVM_E500MC
184 Calculate elapsed time for every exit/enter cycle. A per-vcpu
185 report is available in debugfs kvm/vm#_vcpu#_timing.
186 The overhead is relatively small, however it is not recommended for
187 production environments.
192 bool "KVM support for PowerPC E500v2 processors"
193 depends on PPC_E500 && !PPC_E500MC
194 depends on !CONTEXT_TRACKING_USER
199 Support running unmodified E500 guest kernels in virtual machines on
200 E500v2 host processors.
202 This module provides access to the hardware capabilities through
203 a character device node named /dev/kvm.
208 bool "KVM support for PowerPC E500MC/E5500/E6500 processors"
209 depends on PPC_E500MC
210 depends on !CONTEXT_TRACKING_USER
216 Support running unmodified E500MC/E5500/E6500 guest kernels in
217 virtual machines on E500MC/E5500/E6500 host processors.
219 This module provides access to the hardware capabilities through
220 a character device node named /dev/kvm.
225 bool "KVM in-kernel MPIC emulation"
226 depends on KVM && PPC_E500
227 select HAVE_KVM_IRQCHIP
228 select HAVE_KVM_IRQFD
229 select HAVE_KVM_IRQ_ROUTING
232 Enable support for emulating MPIC devices inside the
233 host kernel, rather than relying on userspace to emulate.
234 Currently, support is limited to certain versions of
235 Freescale's MPIC implementation.
238 bool "KVM in-kernel XICS emulation"
239 depends on KVM_BOOK3S_64 && !KVM_MPIC
240 select HAVE_KVM_IRQCHIP
241 select HAVE_KVM_IRQFD
244 Include support for the XICS (eXternal Interrupt Controller
245 Specification) interrupt controller architecture used on
246 IBM POWER (pSeries) servers.
251 depends on KVM_XICS && PPC_XIVE_NATIVE && KVM_BOOK3S_HV_POSSIBLE
253 endif # VIRTUALIZATION