[09/11] KVM: x86/mmu: Revert "KVM: x86/mmu: Remove is_obsolete() call"
diff mbox series

Message ID 20190913024612.28392-10-sean.j.christopherson@intel.com
State New
Headers show
Series
  • KVM: x86/mmu: Restore fast invalidate/zap flow
Related show

Commit Message

Sean Christopherson Sept. 13, 2019, 2:46 a.m. UTC
Now that the fast invalidate mechanism has been reintroduced, restore
the performance tweaks for fast invalidation that existed prior to its
removal.

Paraphrasing the original changelog (commit 5ff0568374ed2 was itself a
partial revert):

  Don't force reloading the remote mmu when zapping an obsolete page, as
  a MMU_RELOAD request has already been issued by kvm_mmu_zap_all_fast()
  immediately after incrementing mmu_valid_gen, i.e. after marking pages
  obsolete.

This reverts commit 5ff0568374ed2e585376a3832857ade5daccd381.

Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
---
 arch/x86/kvm/mmu.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

Patch
diff mbox series

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 84d916674529..bce19918ca5a 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2752,7 +2752,12 @@  static bool __kvm_mmu_prepare_zap_page(struct kvm *kvm,
 	} else {
 		list_move(&sp->link, &kvm->arch.active_mmu_pages);
 
-		if (!sp->role.invalid)
+		/*
+		 * Obsolete pages cannot be used on any vCPUs, see the comment
+		 * in kvm_mmu_zap_all_fast().  Note, is_obsolete_sp() also
+		 * treats invalid shadow pages as being obsolete.
+		 */
+		if (!is_obsolete_sp(kvm, sp))
 			kvm_reload_remote_mmus(kvm);
 	}