KVM: VMX: Tell the nested hypervisor to skip L1D flush on vmentry
authorPaolo Bonzini <pbonzini@redhat.com>
Sun, 5 Aug 2018 14:07:47 +0000 (16:07 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 15 Aug 2018 16:14:53 +0000 (18:14 +0200)
commitf56c8ee659c926bdba42c0d45405433e1a00eb2e
tree2c913d171a8c02b0d4749fe3815341b0026002c2
parent383f160027af7f3e3c32c2988980652e708a2119
KVM: VMX: Tell the nested hypervisor to skip L1D flush on vmentry

commit 5b76a3cff011df2dcb6186c965a2e4d809a05ad4 upstream

When nested virtualization is in use, VMENTER operations from the nested
hypervisor into the nested guest will always be processed by the bare metal
hypervisor, and KVM's "conditional cache flushes" mode in particular does a
flush on nested vmentry.  Therefore, include the "skip L1D flush on
vmentry" bit in KVM's suggested ARCH_CAPABILITIES setting.

Add the relevant Documentation.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Documentation/l1tf.rst
arch/x86/include/asm/kvm_host.h
arch/x86/kvm/vmx.c
arch/x86/kvm/x86.c