linux-kernel - [PATCH] KVM: coalesced_mmio: add bounds checking

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-next>] [day] [month] [year] [list]

Message-Id: <1568815302-21319-1-git-send-email-pbonzini@redhat.com>
Date:   Wed, 18 Sep 2019 16:01:42 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     linux-kernel@...r.kernel.org, kvm@...r.kernel.org
Cc:     will@...nel.org, kernellwp@...il.com, gregkh@...uxfoundation.org,
        Matt Delco <delco@...omium.org>, stable@...r.kernel.org,
        Jim Mattson <jmattson@...gle.com>
Subject: [PATCH] KVM: coalesced_mmio: add bounds checking

From: Matt Delco <delco@...omium.org>

The first/last indexes are typically shared with a user app.
The app can change the 'last' index that the kernel uses
to store the next result.  This change sanity checks the index
before using it for writing to a potentially arbitrary address.

This fixes CVE-2019-14821.

Cc: stable@...r.kernel.org
Fixes: 5f94c1741bdc ("KVM: Add coalesced MMIO support (common part)")
Signed-off-by: Matt Delco <delco@...omium.org>
Signed-off-by: Jim Mattson <jmattson@...gle.com>
Reported-by: syzbot+983c866c3dd6efa3662a@...kaller.appspotmail.com
[Use READ_ONCE. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@...hat.com>
---
 virt/kvm/coalesced_mmio.c | 19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)

diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c
index 5294abb3f178..8ffd07e2a160 100644
--- a/virt/kvm/coalesced_mmio.c
+++ b/virt/kvm/coalesced_mmio.c
@@ -40,7 +40,7 @@ static int coalesced_mmio_in_range(struct kvm_coalesced_mmio_dev *dev,
 	return 1;
 }
 
-static int coalesced_mmio_has_room(struct kvm_coalesced_mmio_dev *dev)
+static int coalesced_mmio_has_room(struct kvm_coalesced_mmio_dev *dev, u32 last)
 {
 	struct kvm_coalesced_mmio_ring *ring;
 	unsigned avail;
@@ -52,7 +52,7 @@ static int coalesced_mmio_has_room(struct kvm_coalesced_mmio_dev *dev)
 	 * there is always one unused entry in the buffer
 	 */
 	ring = dev->kvm->coalesced_mmio_ring;
-	avail = (ring->first - ring->last - 1) % KVM_COALESCED_MMIO_MAX;
+	avail = (ring->first - last - 1) % KVM_COALESCED_MMIO_MAX;
 	if (avail == 0) {
 		/* full */
 		return 0;
@@ -67,25 +67,28 @@ static int coalesced_mmio_write(struct kvm_vcpu *vcpu,
 {
 	struct kvm_coalesced_mmio_dev *dev = to_mmio(this);
 	struct kvm_coalesced_mmio_ring *ring = dev->kvm->coalesced_mmio_ring;
+	__u32 insert;
 
 	if (!coalesced_mmio_in_range(dev, addr, len))
 		return -EOPNOTSUPP;
 
 	spin_lock(&dev->kvm->ring_lock);
 
-	if (!coalesced_mmio_has_room(dev)) {
+	insert = READ_ONCE(ring->last);
+	if (!coalesced_mmio_has_room(dev, insert) ||
+	    insert >= KVM_COALESCED_MMIO_MAX) {
 		spin_unlock(&dev->kvm->ring_lock);
 		return -EOPNOTSUPP;
 	}
 
 	/* copy data in first free entry of the ring */
 
-	ring->coalesced_mmio[ring->last].phys_addr = addr;
-	ring->coalesced_mmio[ring->last].len = len;
-	memcpy(ring->coalesced_mmio[ring->last].data, val, len);
-	ring->coalesced_mmio[ring->last].pio = dev->zone.pio;
+	ring->coalesced_mmio[insert].phys_addr = addr;
+	ring->coalesced_mmio[insert].len = len;
+	memcpy(ring->coalesced_mmio[insert].data, val, len);
+	ring->coalesced_mmio[insert].pio = dev->zone.pio;
 	smp_wmb();
-	ring->last = (ring->last + 1) % KVM_COALESCED_MMIO_MAX;
+	ring->last = (insert + 1) % KVM_COALESCED_MMIO_MAX;
 	spin_unlock(&dev->kvm->ring_lock);
 	return 0;
 }
-- 
1.8.3.1