From mboxrd@z Thu Jan  1 00:00:00 1970
From: Gleb Natapov <gleb@redhat.com>
Subject: [PATCH] KVM: do not iterate over all VMs in mmu_shrink()
Date: Mon, 4 Jun 2012 14:53:23 +0300
Message-ID: <20120604115323.GG23670@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: avi@redhat.com, mtosatti@redhat.com
To: kvm@vger.kernel.org
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:48720 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1752778Ab2FDLxZ (ORCPT <rfc822;kvm@vger.kernel.org>);
	Mon, 4 Jun 2012 07:53:25 -0400
Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22])
	by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id q54BrOIu005347
	(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK)
	for <kvm@vger.kernel.org>; Mon, 4 Jun 2012 07:53:25 -0400
Content-Disposition: inline
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

mmu_shrink() needlessly iterates over all VMs even though it will not
attempt to free mmu pages from more than one on them. Fix that and also
check used mmu pages count outside of VM lock to skip inactive VMs faster.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index c2fef8e..d1d477a 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -3945,7 +3945,6 @@ static void kvm_mmu_remove_some_alloc_mmu_pages(struct kvm *kvm,
 static int mmu_shrink(struct shrinker *shrink, struct shrink_control *sc)
 {
 	struct kvm *kvm;
-	struct kvm *kvm_freed = NULL;
 	int nr_to_scan = sc->nr_to_scan;
 
 	if (nr_to_scan == 0)
@@ -3957,22 +3956,30 @@ static int mmu_shrink(struct shrinker *shrink, struct shrink_control *sc)
 		int idx;
 		LIST_HEAD(invalid_list);
 
+		/*
+		 * n_used_mmu_pages is accessed without holding kvm->mmu_lock
+		 * here. We may skip a VM instance errorneosly, but we do not
+		 * want to shrink a VM that only started to populate its MMU
+		 * anyway.
+		 */
+		if (kvm->arch.n_used_mmu_pages > 0) {
+			if (!nr_to_scan--)
+				break;
+			continue;
+		}
+
 		idx = srcu_read_lock(&kvm->srcu);
 		spin_lock(&kvm->mmu_lock);
-		if (!kvm_freed && nr_to_scan > 0 &&
-		    kvm->arch.n_used_mmu_pages > 0) {
-			kvm_mmu_remove_some_alloc_mmu_pages(kvm,
-							    &invalid_list);
-			kvm_freed = kvm;
-		}
-		nr_to_scan--;
 
+		kvm_mmu_remove_some_alloc_mmu_pages(kvm, &invalid_list);
 		kvm_mmu_commit_zap_page(kvm, &invalid_list);
+
 		spin_unlock(&kvm->mmu_lock);
 		srcu_read_unlock(&kvm->srcu, idx);
+
+		list_move_tail(&kvm->vm_list, &vm_list);
+		break;
 	}
-	if (kvm_freed)
-		list_move_tail(&kvm_freed->vm_list, &vm_list);
 
 	raw_spin_unlock(&kvm_lock);
 
--
			Gleb.