KVM: arm/arm64: PMU: remove request-less vcpu kick
authorAndrew Jones <drjones@redhat.com>
Sun, 4 Jun 2017 12:44:00 +0000 (14:44 +0200)
committerChristoffer Dall <cdall@linaro.org>
Sun, 4 Jun 2017 14:53:56 +0000 (16:53 +0200)
commitb7484931e4a8b06fe734ac2b5134f09204bad11a
tree0606976263fb79133196e43387cc26bd40542fb8
parent325f9c649c8a4e447e4d3babacc7a60b75012d5d
KVM: arm/arm64: PMU: remove request-less vcpu kick

Refactor PMU overflow handling in order to remove the request-less
vcpu kick.  Now, since kvm_vgic_inject_irq() uses vcpu requests,
there should be no chance that a kick sent at just the wrong time
(between the VCPU's call to kvm_pmu_flush_hwstate() and before it
enters guest mode) results in a failure for the guest to see updated
GIC state until its next exit some time later for some other reason.

Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
virt/kvm/arm/pmu.c