| Kconfig (ff0d6be4bf9ad4daba024ba0157b97750c7ad1fb) | Kconfig (ed922739c9199bf515a3e7fec3e319ce1edeef2a) |
|---|---|
| 1# SPDX-License-Identifier: GPL-2.0 2# 3# KVM configuration 4# 5 6source "virt/kvm/Kconfig" 7 8menuconfig VIRTUALIZATION --- 12 unchanged lines hidden (view full) --- 21 bool 22 select PREEMPT_NOTIFIERS 23 select HAVE_KVM_EVENTFD 24 select HAVE_KVM_VCPU_ASYNC_IOCTL 25 select SRCU 26 select KVM_VFIO 27 select IRQ_BYPASS_MANAGER 28 select HAVE_KVM_IRQ_BYPASS | 1# SPDX-License-Identifier: GPL-2.0 2# 3# KVM configuration 4# 5 6source "virt/kvm/Kconfig" 7 8menuconfig VIRTUALIZATION --- 12 unchanged lines hidden (view full) --- 21 bool 22 select PREEMPT_NOTIFIERS 23 select HAVE_KVM_EVENTFD 24 select HAVE_KVM_VCPU_ASYNC_IOCTL 25 select SRCU 26 select KVM_VFIO 27 select IRQ_BYPASS_MANAGER 28 select HAVE_KVM_IRQ_BYPASS |
| 29 select INTERVAL_TREE |
|
| 29 30config KVM_BOOK3S_HANDLER 31 bool 32 33config KVM_BOOK3S_32_HANDLER 34 bool 35 select KVM_BOOK3S_HANDLER 36 select KVM_MMIO --- 88 unchanged lines hidden (view full) --- 125 and in nap mode due to idle (cede) while other threads are still 126 in the guest. The total, minimum and maximum times in nanoseconds 127 together with the number of executions are reported in debugfs in 128 kvm/vm#/vcpu#/timings. The overhead is of the order of 30 - 40 129 ns per exit on POWER8. 130 131 If unsure, say N. 132 | 30 31config KVM_BOOK3S_HANDLER 32 bool 33 34config KVM_BOOK3S_32_HANDLER 35 bool 36 select KVM_BOOK3S_HANDLER 37 select KVM_MMIO --- 88 unchanged lines hidden (view full) --- 126 and in nap mode due to idle (cede) while other threads are still 127 in the guest. The total, minimum and maximum times in nanoseconds 128 together with the number of executions are reported in debugfs in 129 kvm/vm#/vcpu#/timings. The overhead is of the order of 30 - 40 130 ns per exit on POWER8. 131 132 If unsure, say N. 133 |
| 133config KVM_BOOK3S_HV_NESTED_PMU_WORKAROUND 134 bool "Nested L0 host workaround for L1 KVM host PMU handling bug" if EXPERT 135 depends on KVM_BOOK3S_HV_POSSIBLE 136 default !EXPERT 137 help 138 Old nested HV capable Linux guests have a bug where they don't 139 reflect the PMU in-use status of their L2 guest to the L0 host 140 while the L2 PMU registers are live. This can result in loss 141 of L2 PMU register state, causing perf to not work correctly in 142 L2 guests. 143 144 Selecting this option for the L0 host implements a workaround for 145 those buggy L1s which saves the L2 state, at the cost of performance 146 in all nested-capable guest entry/exit. 147 | |
| 148config KVM_BOOKE_HV 149 bool 150 151config KVM_EXIT_TIMING 152 bool "Detailed exit timing" 153 depends on KVM_E500V2 || KVM_E500MC 154 help 155 Calculate elapsed time for every exit/enter cycle. A per-vcpu --- 67 unchanged lines hidden --- | 134config KVM_BOOKE_HV 135 bool 136 137config KVM_EXIT_TIMING 138 bool "Detailed exit timing" 139 depends on KVM_E500V2 || KVM_E500MC 140 help 141 Calculate elapsed time for every exit/enter cycle. A per-vcpu --- 67 unchanged lines hidden --- |