Skip to contentSkip to conversation

virtio: reset device on bad guest index in virtio_load()

PATCH In review John Levon 2 participants 4 messages Nov 20, 2020 – Dec 2, 2020

lore
Share & tools
lore

John Levon

author
#0 lore raw

PATCH 1/1 virtio: reset device on bad guest index in virtio_load()

If we find a queue with an inconsistent guest index value, explicitly mark the device as needing a reset - and broken - via virtio_error().

There's at least one driver implementation - the virtio-win NetKVM driver - that is able to handle a VIRTIO_CONFIG_S_NEEDS_RESET notification and successfully restore the device to a working state. Other implementations do not correctly handle this, but as the VQ is not in a functional state anyway, this is still worth doing.

---
 hw/virtio/virtio.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)
hw/virtio/virtio.c +9 -6
index ceb58fda6c..eff35fab7c 100644
@@ -3161,12 +3161,15 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f, int version_id)
             nheads = vring_avail_idx(&vdev->vq[i]) - vdev->vq[i].last_avail_idx;
             /* Check it isn't doing strange things with descriptor numbers. */
             if (nheads > vdev->vq[i].vring.num) {
-                qemu_log_mask(LOG_GUEST_ERROR,
-                              "VQ %d size 0x%x Guest index 0x%x "
-                              "inconsistent with Host index 0x%x: delta 0x%x",
-                              i, vdev->vq[i].vring.num,
-                              vring_avail_idx(&vdev->vq[i]),
-                              vdev->vq[i].last_avail_idx, nheads);
+                virtio_error(vdev, "VQ %d size 0x%x Guest index 0x%x "
+                             "inconsistent with Host index 0x%x: delta 0x%x",
+                             i, vdev->vq[i].vring.num,
+                             vring_avail_idx(&vdev->vq[i]),
+                             vdev->vq[i].last_avail_idx, nheads);
+                vdev->vq[i].used_idx = 0;
+                vdev->vq[i].shadow_avail_idx = 0;
+                vdev->vq[i].inuse = 0;
+                continue;
             }
             vdev->vq[i].used_idx = vring_used_idx(&vdev->vq[i]);
             vdev->vq[i].shadow_avail_idx = vring_avail_idx(&vdev->vq[i]);
Signature · 1 line
2.22.3
author
#1 lore raw
Quoted 8 lines · John Levon

If we find a queue with an inconsistent guest index value, explicitly mark the device as needing a reset - and broken - via virtio_error().

There's at least one driver implementation - the virtio-win NetKVM driver - that is able to handle a VIRTIO_CONFIG_S_NEEDS_RESET notification and successfully restore the device to a working state. Other implementations do not correctly handle this, but as the VQ is not in a functional state anyway, this is still worth doing.

Ping, anyone have issues with doing this?

cheers
john

Quoted 31 lines
---
 hw/virtio/virtio.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)
hw/virtio/virtio.c +9 -6
index ceb58fda6c..eff35fab7c 100644
@@ -3161,12 +3161,15 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f, int version_id)
             nheads = vring_avail_idx(&vdev->vq[i]) - vdev->vq[i].last_avail_idx;
             /* Check it isn't doing strange things with descriptor numbers. */
             if (nheads > vdev->vq[i].vring.num) {
-                qemu_log_mask(LOG_GUEST_ERROR,
-                              "VQ %d size 0x%x Guest index 0x%x "
-                              "inconsistent with Host index 0x%x: delta 0x%x",
-                              i, vdev->vq[i].vring.num,
-                              vring_avail_idx(&vdev->vq[i]),
-                              vdev->vq[i].last_avail_idx, nheads);
+                virtio_error(vdev, "VQ %d size 0x%x Guest index 0x%x "
+                             "inconsistent with Host index 0x%x: delta 0x%x",
+                             i, vdev->vq[i].vring.num,
+                             vring_avail_idx(&vdev->vq[i]),
+                             vdev->vq[i].last_avail_idx, nheads);
+                vdev->vq[i].used_idx = 0;
+                vdev->vq[i].shadow_avail_idx = 0;
+                vdev->vq[i].inuse = 0;
+                continue;
             }
             vdev->vq[i].used_idx = vring_used_idx(&vdev->vq[i]);
             vdev->vq[i].shadow_avail_idx = vring_avail_idx(&vdev->vq[i]);
#2 lore raw

No, but how about sending the patch to me and the mailing list? I didn't get it through either channel.

Quoted 47 lines · John Levon
Quoted 8 lines · John Levon

If we find a queue with an inconsistent guest index value, explicitly mark the device as needing a reset - and broken - via virtio_error().

There's at least one driver implementation - the virtio-win NetKVM driver - that is able to handle a VIRTIO_CONFIG_S_NEEDS_RESET notification and successfully restore the device to a working state. Other implementations do not correctly handle this, but as the VQ is not in a functional state anyway, this is still worth doing.

Ping, anyone have issues with doing this?

cheers
john

Quoted 31 lines
---
 hw/virtio/virtio.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)
hw/virtio/virtio.c +9 -6
index ceb58fda6c..eff35fab7c 100644
@@ -3161,12 +3161,15 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f, int version_id)
             nheads = vring_avail_idx(&vdev->vq[i]) - vdev->vq[i].last_avail_idx;
             /* Check it isn't doing strange things with descriptor numbers. */
             if (nheads > vdev->vq[i].vring.num) {
-                qemu_log_mask(LOG_GUEST_ERROR,
-                              "VQ %d size 0x%x Guest index 0x%x "
-                              "inconsistent with Host index 0x%x: delta 0x%x",
-                              i, vdev->vq[i].vring.num,
-                              vring_avail_idx(&vdev->vq[i]),
-                              vdev->vq[i].last_avail_idx, nheads);
+                virtio_error(vdev, "VQ %d size 0x%x Guest index 0x%x "
+                             "inconsistent with Host index 0x%x: delta 0x%x",
+                             i, vdev->vq[i].vring.num,
+                             vring_avail_idx(&vdev->vq[i]),
+                             vdev->vq[i].last_avail_idx, nheads);
+                vdev->vq[i].used_idx = 0;
+                vdev->vq[i].shadow_avail_idx = 0;
+                vdev->vq[i].inuse = 0;
+                continue;
             }
             vdev->vq[i].used_idx = vring_used_idx(&vdev->vq[i]);
             vdev->vq[i].shadow_avail_idx = vring_avail_idx(&vdev->vq[i]);
#3 lore raw
In reply to John Levon
Quoted 11 lines · John Levon

If we find a queue with an inconsistent guest index value, explicitly mark the device as needing a reset - and broken - via virtio_error().

There's at least one driver implementation - the virtio-win NetKVM driver - that is able to handle a VIRTIO_CONFIG_S_NEEDS_RESET notification and successfully restore the device to a working state. Other implementations do not correctly handle this, but as the VQ is not in a functional state anyway, this is still worth doing.

I tagged this for after the release. pls ping me after the release to help make sure it does not get lost.

Quoted 33 lines
---
 hw/virtio/virtio.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)
hw/virtio/virtio.c +9 -6
index ceb58fda6c..eff35fab7c 100644
@@ -3161,12 +3161,15 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f, int version_id)
             nheads = vring_avail_idx(&vdev->vq[i]) - vdev->vq[i].last_avail_idx;
             /* Check it isn't doing strange things with descriptor numbers. */
             if (nheads > vdev->vq[i].vring.num) {
-                qemu_log_mask(LOG_GUEST_ERROR,
-                              "VQ %d size 0x%x Guest index 0x%x "
-                              "inconsistent with Host index 0x%x: delta 0x%x",
-                              i, vdev->vq[i].vring.num,
-                              vring_avail_idx(&vdev->vq[i]),
-                              vdev->vq[i].last_avail_idx, nheads);
+                virtio_error(vdev, "VQ %d size 0x%x Guest index 0x%x "
+                             "inconsistent with Host index 0x%x: delta 0x%x",
+                             i, vdev->vq[i].vring.num,
+                             vring_avail_idx(&vdev->vq[i]),
+                             vdev->vq[i].last_avail_idx, nheads);
+                vdev->vq[i].used_idx = 0;
+                vdev->vq[i].shadow_avail_idx = 0;
+                vdev->vq[i].inuse = 0;
+                continue;
             }
             vdev->vq[i].used_idx = vring_used_idx(&vdev->vq[i]);
             vdev->vq[i].shadow_avail_idx = vring_avail_idx(&vdev->vq[i]);
Signature · 1 line
2.22.3