[Qemu-devel] [PATCH] vhost: do not verify ring mappings when IOMMU is enabled

Jason Wang posted 1 patch 23 weeks ago
Patches applied successfully (tree, apply log)
git fetch https://github.com/patchew-project/qemu tags/patchew/1523588509-4868-1-git-send-email-jasowang@redhat.com
Test checkpatch passed
Test docker-build@min-glib passed
Test docker-mingw@fedora passed
Test s390x passed
hw/virtio/vhost.c | 4 ++++
1 file changed, 4 insertions(+)

[Qemu-devel] [PATCH] vhost: do not verify ring mappings when IOMMU is enabled

Posted by Jason Wang 23 weeks ago
When IOMMU is enabled, we store virtqueue metadata as iova (though it
may has _phys suffix) and access them through dma helpers. Any
translation failures could be reported by IOMMU.

In this case, trying to validate iova against gpa won't work and will
cause a false error reporting. So this patch bypasses the ring
verification if IOMMU is enabled which is similar to the behavior
before 0ca1fd2d6878 that calls vhost_memory_map() which is a nop when
IOMMU is enabled.

Fixes: 0ca1fd2d6878 ("vhost: Simplify ring verification checks")
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/virtio/vhost.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index f51bf57..9d5850a 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -342,6 +342,10 @@ static int vhost_verify_ring_mappings(struct vhost_dev *dev,
         "used ring"
     };
 
+    if (vhost_dev_has_iommu(dev)) {
+        return 0;
+    }
+
     for (i = 0; i < dev->nvqs; ++i) {
         struct vhost_virtqueue *vq = dev->vqs + i;
 
-- 
2.7.4


Re: [Qemu-devel] [PATCH] vhost: do not verify ring mappings when IOMMU is enabled

Posted by Igor Mammedov 23 weeks ago
On Fri, 13 Apr 2018 11:01:49 +0800
Jason Wang <jasowang@redhat.com> wrote:

> When IOMMU is enabled, we store virtqueue metadata as iova (though it
> may has _phys suffix) and access them through dma helpers. Any
> translation failures could be reported by IOMMU.
> 
> In this case, trying to validate iova against gpa won't work and will
> cause a false error reporting. So this patch bypasses the ring
> verification if IOMMU is enabled which is similar to the behavior
> before 0ca1fd2d6878 that calls vhost_memory_map() which is a nop when
> IOMMU is enabled.
> 
> Fixes: 0ca1fd2d6878 ("vhost: Simplify ring verification checks")
> Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
> Cc: Igor Mammedov <imammedo@redhat.com>
> Cc: Michael S. Tsirkin <mst@redhat.com>
> Signed-off-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>


Looks like candidat for 2.12 if it's not too late?

> ---
>  hw/virtio/vhost.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
> index f51bf57..9d5850a 100644
> --- a/hw/virtio/vhost.c
> +++ b/hw/virtio/vhost.c
> @@ -342,6 +342,10 @@ static int vhost_verify_ring_mappings(struct vhost_dev *dev,
>          "used ring"
>      };
>  
> +    if (vhost_dev_has_iommu(dev)) {
> +        return 0;
> +    }
> +
>      for (i = 0; i < dev->nvqs; ++i) {
>          struct vhost_virtqueue *vq = dev->vqs + i;
>