kvm: mmu: Used range based flushing in slot_handle_level_range
authorBen Gardon <bgardon@google.com>
Tue, 12 Mar 2019 18:45:59 +0000 (11:45 -0700)
committerPaolo Bonzini <pbonzini@redhat.com>
Thu, 28 Mar 2019 16:28:57 +0000 (17:28 +0100)
commitf285c633cb6d68d2bf3a8ad65bee3835aac9886c
tree58657b50af4d6435d38d76cafdfa9a233cbe26e5
parent3d9683cf3bfb6d4e4605a153958dfca7e18b52f2
kvm: mmu: Used range based flushing in slot_handle_level_range

Replace kvm_flush_remote_tlbs with kvm_flush_remote_tlbs_with_address
in slot_handle_level_range. When range based flushes are not enabled
kvm_flush_remote_tlbs_with_address falls back to kvm_flush_remote_tlbs.

This changes the behavior of many functions that indirectly use
slot_handle_level_range, iff the range based flushes are enabled. The
only potential problem I see with this is that kvm->tlbs_dirty will be
cleared less often, however the only caller of slot_handle_level_range that
checks tlbs_dirty is kvm_mmu_notifier_invalidate_range_start which
checks it and does a kvm_flush_remote_tlbs after calling
kvm_unmap_hva_range anyway.

Tested: Ran all kvm-unit-tests on a Intel Haswell machine with and
without this patch. The patch introduced no new failures.

Signed-off-by: Ben Gardon <bgardon@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
arch/x86/kvm/mmu.c