linux-kernel - [PATCH v2 13/16] KVM: MMU: break sptes write-protect if gfn is writable

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <4F87FCDB.3050008@linux.vnet.ibm.com>
Date:	Fri, 13 Apr 2012 18:15:55 +0800
From:	Xiao Guangrong <xiaoguangrong@...ux.vnet.ibm.com>
To:	Xiao Guangrong <xiaoguangrong@...ux.vnet.ibm.com>
CC:	Avi Kivity <avi@...hat.com>, Marcelo Tosatti <mtosatti@...hat.com>,
	LKML <linux-kernel@...r.kernel.org>, KVM <kvm@...r.kernel.org>
Subject: [PATCH v2 13/16] KVM: MMU: break sptes write-protect if gfn is writable

Make all sptes to be writable if the gfn become write-free to reduce
the later page fault

The idea is from Avi

Signed-off-by: Xiao Guangrong <xiaoguangrong@...ux.vnet.ibm.com>
---
 arch/x86/kvm/mmu.c |   34 +++++++++++++++++++++++++++++++++-
 1 files changed, 33 insertions(+), 1 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 578a1e2..efa5d59 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2323,15 +2323,45 @@ static void kvm_unsync_pages(struct kvm_vcpu *vcpu,  gfn_t gfn)
 	}
 }

+/*
+ * If the gfn become write-free, we make all sptes which point to this
+ * gfn to be writable.
+ * Note: we should call mark_page_dirty for the gfn later.
+ */
+static void rmap_break_page_table_wp(struct kvm_memory_slot *slot, gfn_t gfn)
+{
+	struct spte_iterator iter;
+	u64 *sptep;
+	int i;
+
+	for (i = PT_PAGE_TABLE_LEVEL;
+	      i < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; i++) {
+		unsigned long *rmap = __gfn_to_rmap(gfn, i, slot);
+
+		for_each_rmap_spte(rmap, &iter, sptep) {
+			u64 spte = *sptep;
+
+			if (!is_writable_pte(spte) &&
+			      (spte & SPTE_ALLOW_WRITE)) {
+				spte &= ~SPTE_WRITE_PROTECT;
+				spte |= PT_WRITABLE_MASK;
+				mmu_spte_update(sptep, spte);
+			}
+		}
+	}
+}
+
 static int mmu_need_write_protect(struct kvm_vcpu *vcpu, gfn_t gfn,
 				  bool can_unsync)
 {
+	struct kvm_memory_slot *slot;
 	struct kvm_mmu_page *s;
 	struct hlist_node *node;
 	unsigned long *rmap;
 	bool need_unsync = false;

-	rmap = gfn_to_rmap(vcpu->kvm, gfn, PT_PAGE_TABLE_LEVEL);
+	slot = gfn_to_memslot(vcpu->kvm, gfn);
+	rmap =	__gfn_to_rmap(gfn, PT_PAGE_TABLE_LEVEL, slot);

 	if (!vcpu->kvm->arch.indirect_shadow_pages)
 		goto write_free;
@@ -2353,6 +2383,8 @@ static int mmu_need_write_protect(struct kvm_vcpu *vcpu, gfn_t gfn,
 	if (need_unsync)
 		kvm_unsync_pages(vcpu, gfn);

+	rmap_break_page_table_wp(slot, gfn);
+
 write_free:
 	__clear_bit(PTE_LIST_WP_BIT, rmap);
 	return 0;
-- 
1.7.7.6

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/