[PATCH v3] virtio: new feature to detect IOMMU device quirk

[Michael S. Tsirkin]

The interaction between virtio and IOMMUs is messy.

On most systems with virtio, physical addresses match bus addresses,
and it doesn't particularly matter which one we use to program
the device.

On some systems, including Xen and any system with a physical device
that speaks virtio behind a physical IOMMU, we must program the IOMMU
for virtio DMA to work at all.

On other systems, including SPARC and PPC64, virtio-pci devices are
enumerated as though they are behind an IOMMU, but the virtio host
ignores the IOMMU, so we must either pretend that the IOMMU isn't
there or somehow map everything as the identity.

Add a feature bit to detect that quirk: VIRTIO_F_IOMMU_PLATFORM.

Any device with this feature bit set to 0 needs a quirk and has to be
passed physical addresses (as opposed to bus addresses) even though
the device is behind an IOMMU.

Note: it has to be a per-device quirk because for example, there could
be a mix of passed-through and virtual virtio devices. As another
example, some devices could be implemented by an out of process
hypervisor backend (in case of qemu vhost, or vhost-user) and so support
for an IOMMU needs to be coded up separately.

It would be cleanest to handle this in IOMMU core code, but that needs
per-device DMA ops. While we are waiting for that to be implemented, use
a work-around in virtio core.

Signed-off-by: Michael S. Tsirkin <mst@xxxxxxxxxx>
---

wanted to use per-device dma ops but that does not
seem to be ready. So let's put this in virtio
code for now, and move when it becomes possible.

 include/uapi/linux/virtio_config.h |  4 +++-
 drivers/virtio/virtio_ring.c       | 15 ++++++++++++++-
 2 files changed, 17 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/virtio_config.h b/include/uapi/linux/virtio_config.h
index 4cb65bb..3f6195e 100644
--- a/include/uapi/linux/virtio_config.h
+++ b/include/uapi/linux/virtio_config.h
@@ -49,7 +49,7 @@
  * transport being used (eg. virtio_ring), the rest are per-device feature
  * bits. */
 #define VIRTIO_TRANSPORT_F_START	28
-#define VIRTIO_TRANSPORT_F_END		33
+#define VIRTIO_TRANSPORT_F_END		34
 
 #ifndef VIRTIO_CONFIG_NO_LEGACY
 /* Do we get callbacks when the ring is completely used, even if we've
@@ -63,4 +63,6 @@
 /* v1.0 compliant. */
 #define VIRTIO_F_VERSION_1		32
 
+/* Do not bypass the IOMMU (if configured) */
+#define VIRTIO_F_IOMMU_PLATFORM		33
 #endif /* _UAPI_LINUX_VIRTIO_CONFIG_H */
diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index ca6bfdd..2a0c8bf 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -117,7 +117,10 @@ struct vring_virtqueue {
 #define to_vvq(_vq) container_of(_vq, struct vring_virtqueue, vq)
 
 /*
- * The interaction between virtio and a possible IOMMU is a mess.
+ * Modern virtio devices might set feature bits to specify whether
+ * they use the platform IOMMU. If there, just use the DMA API.
+ *
+ * If not there, the interaction between virtio and DMA API is messy.
  *
  * On most systems with virtio, physical addresses match bus addresses,
  * and it doesn't particularly matter whether we use the DMA API.
@@ -137,6 +140,10 @@ struct vring_virtqueue {
 
 static bool vring_use_dma_api(struct virtio_device *vdev)
 {
+	if (virtio_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM))
+		return true;
+
+	/* Otherwise, we are left to guess. */
 	/*
 	 * In theory, it's possible to have a buggy QEMU-supposed
 	 * emulated Q35 IOMMU and Xen enabled at the same time.  On
@@ -1099,6 +1106,12 @@ void vring_transport_features(struct virtio_device *vdev)
 			break;
 		case VIRTIO_F_VERSION_1:
 			break;
+		case VIRTIO_F_IOMMU_PASSTHROUGH:
+			break;
+		case VIRTIO_F_IOMMU_PLATFORM:
+			/* Ignore passthrough hint for now, obey kernel config. */
+			__virtio_clear_bit(vdev, VIRTIO_F_IOMMU_PASSTHROUGH);
+			break;
 		default:
 			/* We don't understand this bit. */
 			__virtio_clear_bit(vdev, i);
-- 
MST