Explicitly document the ordering of vcpu->mutex being taken *outside* of kvm->slots_lock. While extremely unintuitive, and arguably wrong, both arm64 and x86 have gained flows that take kvm->slots_lock inside of vcpu->mutex. x86's kvm_inhibit_apic_access_page() is particularly nasty, as slots_lock is taken quite deep within KVM_RUN, i.e. simply swapping the ordering isn't an option. Commit to the vcpu->mutex => kvm->slots_lock ordering even though taking a VM-scoped lock inside a vCPU-scoped lock is odd, as vcpu->mutex really is intended to be a "top-level" lock in most respects, whereas kvm->slots_lock is "just" a helper lock. Cc: Oliver Upton Cc: Marc Zyngier Signed-off-by: Sean Christopherson --- Documentation/virt/kvm/locking.rst | 2 ++ 1 file changed, 2 insertions(+) diff --git a/Documentation/virt/kvm/locking.rst b/Documentation/virt/kvm/locking.rst index ae8bce7fecbe..2b4c786038be 100644 --- a/Documentation/virt/kvm/locking.rst +++ b/Documentation/virt/kvm/locking.rst @@ -17,6 +17,8 @@ The acquisition orders for mutexes are as follows: - kvm->lock is taken outside kvm->slots_lock and kvm->irq_lock +- vcpu->mutex is taken outside kvm->slots_lock + - kvm->slots_lock is taken outside kvm->irq_lock, though acquiring them together is quite rare. base-commit: 6b36119b94d0b2bb8cea9d512017efafd461d6ac -- 2.51.0.858.gf9c4a03a3a-goog