Skip to contentSkip to conversation

virtio-iommu: use-after-free fix

PATCH In review Michael S. Tsirkin 2 participants 4 messages Apr 7, 2022 – Apr 9, 2022

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 virtio-iommu: use-after-free fix

From: Wentao Liang <[email protected]>

A potential Use-after-free was reported in virtio_iommu_handle_command when using virtio-iommu:

Quoted 32 lines

I find a potential Use-after-free in QEMU 6.2.0, which is in virtio_iommu_handle_command() (./hw/virtio/virtio-iommu.c).

Specifically, in the loop body, the variable 'buf' allocated at line 639 can be freed by g_free() at line 659. However, if the execution path enters the loop body again and the if branch takes true at line 616, the control will directly jump to 'out' at line 651. At this time, 'buf' is a freed pointer, which is not assigned with an allocated memory but used at line 653. As a result, a UAF bug is triggered.

599 for (;;) {
...
615 sz = iov_to_buf(iov, iov_cnt, 0, &head, sizeof(head));
616 if (unlikely(sz != sizeof(head))) {
617 tail.status = VIRTIO_IOMMU_S_DEVERR;
618 goto out;
619 }
...
639 buf = g_malloc0(output_size);
...
651 out:
652 sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
653 buf ? buf : &tail, output_size);
...
659 g_free(buf);

We can fix it by set ‘buf‘ to NULL after freeing it:

Quoted 7 lines

651 out:
652 sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
653 buf ? buf : &tail, output_size);
...
659 g_free(buf);
+++ buf = NULL;
660 }

Fix as suggested by the reporter.

Message-ID: <[email protected]>

---
 hw/virtio/virtio-iommu.c | 1 +
 1 file changed, 1 insertion(+)
hw/virtio/virtio-iommu.c +1 -0
index 239fe97b12..2b1d21edd1 100644
@@ -683,6 +683,7 @@ out:
         virtio_notify(vdev, vq);
         g_free(elem);
         g_free(buf);
+        buf = NULL;
     }
 }
 
Signature · 1 line
MST
#1 lore raw
Quoted 8 lines · Michael S. Tsirkin

From: Wentao Liang <[email protected]>

A potential Use-after-free was reported in virtio_iommu_handle_command when using virtio-iommu:

Quoted 2 lines

I find a potential Use-after-free in QEMU 6.2.0, which is in virtio_iommu_handle_command() (./hw/virtio/virtio-iommu.c).

So, this isn't a regression. Do you think it's critically necessary it goes in 7.0, or is it in the category "put it into 7.0 if we need an rc4 for some other reason anyway" ?

(I have a feeling we'll need an rc4, but we'll see.)

thanks
-- PMM

#2 lore raw
Quoted 10 lines · Peter Maydell
Quoted 8 lines · Michael S. Tsirkin

From: Wentao Liang <[email protected]>

A potential Use-after-free was reported in virtio_iommu_handle_command when using virtio-iommu:

Quoted 2 lines

I find a potential Use-after-free in QEMU 6.2.0, which is in virtio_iommu_handle_command() (./hw/virtio/virtio-iommu.c).

Quoted 8 lines

So, this isn't a regression. Do you think it's critically necessary it goes in 7.0, or is it in the category "put it into 7.0 if we need an rc4 for some other reason anyway" ?

(I have a feeling we'll need an rc4, but we'll see.)

thanks
-- PMM

I am concerned it can be used to trigger a CVE but I could not find a way. So I would say if there's an rc4 pls include it but if not then we can pick it up in stable.

Signature · 1 line
MST
#3 lore raw
Quoted 16 lines · Michael S. Tsirkin
Quoted 14 lines · Peter Maydell
Quoted 8 lines · Michael S. Tsirkin

From: Wentao Liang <[email protected]>

A potential Use-after-free was reported in virtio_iommu_handle_command when using virtio-iommu:

Quoted 2 lines

I find a potential Use-after-free in QEMU 6.2.0, which is in virtio_iommu_handle_command() (./hw/virtio/virtio-iommu.c).

So, this isn't a regression. Do you think it's critically necessary it goes in 7.0, or is it in the category "put it into 7.0 if we need an rc4 for some other reason anyway" ?

Quoted 8 lines
Quoted 4 lines

(I have a feeling we'll need an rc4, but we'll see.)

thanks
-- PMM

I am concerned it can be used to trigger a CVE but I could not find a way. So I would say if there's an rc4 pls include it but if not then we can pick it up in stable.

We needed an rc4 for a couple of other security fixes, so I've applied this to master; thanks.

-- PMM