When used with vhost kernel backend, we can offload at both directions.
- From vhost kernel to virtio_user, the offload is enabled so that
DPDK app can trust the flow is checksum-correct; and if DPDK app
sends it through another port, the checksum needs to be
recalculated or offloaded. It also applies to TSO.
- From virtio_user to vhost_kernel, the offload is enabled so that
kernel can trust the flow is L4-checksum-correct, no need to verify
it; if kernel will consume it, DPDK app should make sure the
l3-checksum is correctly set.
Signed-off-by: Jianfeng Tan <jianfeng....@intel.com>
---
drivers/net/virtio/virtio_user/vhost_kernel.c | 61 ++++++++++++++++++++++++++-
1 file changed, 59 insertions(+), 2 deletions(-)
diff --git a/drivers/net/virtio/virtio_user/vhost_kernel.c
b/drivers/net/virtio/virtio_user/vhost_kernel.c
index 1e7cdef..bdb4af2 100644
--- a/drivers/net/virtio/virtio_user/vhost_kernel.c
+++ b/drivers/net/virtio/virtio_user/vhost_kernel.c
@@ -91,6 +91,13 @@ struct vhost_memory_kernel {
#define IFF_ATTACH_QUEUE 0x0200
#define IFF_DETACH_QUEUE 0x0400
+/* Features for GSO (TUNSETOFFLOAD). */
+#define TUN_F_CSUM 0x01 /* You can hand me unchecksummed packets. */
+#define TUN_F_TSO4 0x02 /* I can handle TSO for IPv4 packets */
+#define TUN_F_TSO6 0x04 /* I can handle TSO for IPv6 packets */
+#define TUN_F_TSO_ECN 0x08 /* I can handle TSO with ECN bits. */
+#define TUN_F_UFO 0x10 /* I can handle UFO packets */
+
/* Constants */
#define TUN_DEF_SNDBUF (1ull << 20)
#define PATH_NET_TUN "/dev/net/tun"
@@ -176,6 +183,28 @@ prepare_vhost_memory_kernel(void)
return vm;
}
+/* with below features, vhost kernel does not need to do the checksum and TSO,
+ * these info will be passed to virtio_user through virtio net header.
+ */
+#define VHOST_KERNEL_GUEST_OFFLOADS_MASK \
+ ((1ULL << VIRTIO_NET_F_GUEST_CSUM) | \
+ (1ULL << VIRTIO_NET_F_GUEST_TSO4) | \
+ (1ULL << VIRTIO_NET_F_GUEST_TSO6) | \
+ (1ULL << VIRTIO_NET_F_GUEST_ECN) | \
+ (1ULL << VIRTIO_NET_F_GUEST_UFO))
+
+/* with below features, when flows from virtio_user to vhost kernel
+ * (1) if flows goes up through the kernel networking stack, it does not need
+ * to verify checksum, which can save CPU cycles;
+ * (2) if flows goes through a Linux bridge and outside from an interface
+ * (kernel driver), checksum and TSO will be done by GSO in kernel or even
+ * offloaded into real physical device.
+ */
+#define VHOST_KERNEL_HOST_OFFLOADS_MASK \
+ ((1ULL << VIRTIO_NET_F_HOST_TSO4) | \
+ (1ULL << VIRTIO_NET_F_HOST_TSO6) | \
+ (1ULL << VIRTIO_NET_F_CSUM))
+
static int
vhost_kernel_ioctl(struct virtio_user_dev *dev,
enum vhost_user_request req,
@@ -196,10 +225,15 @@ vhost_kernel_ioctl(struct virtio_user_dev *dev,
arg = (void *)vm;
}
- /* Does not work when VIRTIO_F_IOMMU_PLATFORM now, why? */
- if (req_kernel == VHOST_SET_FEATURES)
+ if (req_kernel == VHOST_SET_FEATURES) {
+ /* Does not work when VIRTIO_F_IOMMU_PLATFORM now, why? */
*(uint64_t *)arg &= ~(1ULL << VIRTIO_F_IOMMU_PLATFORM);
+ /* VHOST kernel does not know about below flags */
+ *(uint64_t *)arg &= ~VHOST_KERNEL_GUEST_OFFLOADS_MASK;
+ *(uint64_t *)arg &= ~VHOST_KERNEL_HOST_OFFLOADS_MASK;
+ }
+
for (i = 0; i < VHOST_KERNEL_MAX_QUEUES; ++i) {
if (dev->vhostfds[i] < 0)
continue;
@@ -209,6 +243,15 @@ vhost_kernel_ioctl(struct virtio_user_dev *dev,
break;
}
+ if (!ret && req_kernel == VHOST_GET_FEATURES) {
+ /* with tap as the backend, all these features are supported
+ * but not claimed by vhost-net, so we add them back when
+ * reporting to upper layer.
+ */
+ *((uint64_t *)arg) |= VHOST_KERNEL_GUEST_OFFLOADS_MASK;
+ *((uint64_t *)arg) |= VHOST_KERNEL_HOST_OFFLOADS_MASK;
+ }
+
if (vm)
free(vm);
@@ -280,6 +323,12 @@ vhost_kernel_enable_queue_pair(struct virtio_user_dev *dev,
int hdr_size;
int vhostfd;
int tapfd;
+ unsigned int offload =
+ TUN_F_CSUM |
+ TUN_F_TSO4 |
+ TUN_F_TSO6 |
+ TUN_F_TSO_ECN |
+ TUN_F_UFO;
vhostfd = dev->vhostfds[pair_idx];
@@ -354,6 +403,14 @@ vhost_kernel_enable_queue_pair(struct virtio_user_dev *dev,
goto error;
}
+ /* TODO: before set the offload capabilities, we'd better (1) check
+ * negotiated features to see if necessary to offload; (2) query tap
+ * to see if it supports the offload capabilities.
+ */
+ if (ioctl(tapfd, TUNSETOFFLOAD, offload) != 0)
+ PMD_DRV_LOG(ERR, "TUNSETOFFLOAD ioctl() failed: %s",
+ strerror(errno));
+
if (vhost_kernel_set_backend(vhostfd, tapfd) < 0)
goto error;
--
2.7.4
