net/virtio: support Tx checksum offload
authorOlivier Matz <olivier.matz@6wind.com>
Thu, 13 Oct 2016 14:16:09 +0000 (16:16 +0200)
committerThomas Monjalon <thomas.monjalon@6wind.com>
Thu, 13 Oct 2016 18:45:56 +0000 (20:45 +0200)
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
doc/guides/rel_notes/release_16_11.rst
drivers/net/virtio/virtio_ethdev.c
drivers/net/virtio/virtio_ethdev.h
drivers/net/virtio/virtio_rxtx.c

index 4805a1f..a6ff76d 100644 (file)
@@ -85,7 +85,7 @@ New Features
 
 * **Supported offloads with virtio.**
 
-  * Rx checksums
+  * Rx/Tx checksums
 
 * **Added virtio NEON support for ARM.**
 
index 6dff6e9..adae9a9 100644 (file)
@@ -1600,6 +1600,13 @@ virtio_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
        dev_info->rx_offload_capa =
                DEV_RX_OFFLOAD_TCP_CKSUM |
                DEV_RX_OFFLOAD_UDP_CKSUM;
+       dev_info->tx_offload_capa = 0;
+
+       if (hw->guest_features & (1ULL << VIRTIO_NET_F_CSUM)) {
+               dev_info->tx_offload_capa |=
+                       DEV_TX_OFFLOAD_UDP_CKSUM |
+                       DEV_TX_OFFLOAD_TCP_CKSUM;
+       }
 }
 
 /*
index 7043a68..2216305 100644 (file)
@@ -62,6 +62,7 @@
         1u << VIRTIO_NET_F_CTRL_VQ       |     \
         1u << VIRTIO_NET_F_CTRL_RX       |     \
         1u << VIRTIO_NET_F_CTRL_VLAN     |     \
+        1u << VIRTIO_NET_F_CSUM          |     \
         1u << VIRTIO_NET_F_MRG_RXBUF     |     \
         1u << VIRTIO_RING_F_INDIRECT_DESC |    \
         1ULL << VIRTIO_F_VERSION_1       |     \
index fc0d84b..675dc43 100644 (file)
@@ -53,6 +53,8 @@
 #include <rte_cpuflags.h>
 #include <rte_net.h>
 #include <rte_ip.h>
+#include <rte_udp.h>
+#include <rte_tcp.h>
 
 #include "virtio_logs.h"
 #include "virtio_ethdev.h"
@@ -207,18 +209,27 @@ virtqueue_enqueue_recv_refill(struct virtqueue *vq, struct rte_mbuf *cookie)
        return 0;
 }
 
+static inline int
+tx_offload_enabled(struct virtio_hw *hw)
+{
+       return vtpci_with_feature(hw, VIRTIO_NET_F_CSUM);
+}
+
 static inline void
 virtqueue_enqueue_xmit(struct virtnet_tx *txvq, struct rte_mbuf *cookie,
                       uint16_t needed, int use_indirect, int can_push)
 {
+       struct virtio_tx_region *txr = txvq->virtio_net_hdr_mz->addr;
        struct vq_desc_extra *dxp;
        struct virtqueue *vq = txvq->vq;
        struct vring_desc *start_dp;
        uint16_t seg_num = cookie->nb_segs;
        uint16_t head_idx, idx;
        uint16_t head_size = vq->hw->vtnet_hdr_size;
-       unsigned long offs;
+       struct virtio_net_hdr *hdr;
+       int offload;
 
+       offload = tx_offload_enabled(vq->hw);
        head_idx = vq->vq_desc_head_idx;
        idx = head_idx;
        dxp = &vq->vq_descx[idx];
@@ -228,10 +239,12 @@ virtqueue_enqueue_xmit(struct virtnet_tx *txvq, struct rte_mbuf *cookie,
        start_dp = vq->vq_ring.desc;
 
        if (can_push) {
-               /* put on zero'd transmit header (no offloads) */
-               void *hdr = rte_pktmbuf_prepend(cookie, head_size);
-
-               memset(hdr, 0, head_size);
+               /* prepend cannot fail, checked by caller */
+               hdr = (struct virtio_net_hdr *)
+                       rte_pktmbuf_prepend(cookie, head_size);
+               /* if offload disabled, it is not zeroed below, do it now */
+               if (offload == 0)
+                       memset(hdr, 0, head_size);
        } else if (use_indirect) {
                /* setup tx ring slot to point to indirect
                 * descriptor list stored in reserved region.
@@ -239,14 +252,11 @@ virtqueue_enqueue_xmit(struct virtnet_tx *txvq, struct rte_mbuf *cookie,
                 * the first slot in indirect ring is already preset
                 * to point to the header in reserved region
                 */
-               struct virtio_tx_region *txr = txvq->virtio_net_hdr_mz->addr;
-
-               offs = idx * sizeof(struct virtio_tx_region)
-                       + offsetof(struct virtio_tx_region, tx_indir);
-
-               start_dp[idx].addr  = txvq->virtio_net_hdr_mem + offs;
+               start_dp[idx].addr  = txvq->virtio_net_hdr_mem +
+                       RTE_PTR_DIFF(&txr[idx].tx_indir, txr);
                start_dp[idx].len   = (seg_num + 1) * sizeof(struct vring_desc);
                start_dp[idx].flags = VRING_DESC_F_INDIRECT;
+               hdr = (struct virtio_net_hdr *)&txr[idx].tx_hdr;
 
                /* loop below will fill in rest of the indirect elements */
                start_dp = txr[idx].tx_indir;
@@ -255,15 +265,43 @@ virtqueue_enqueue_xmit(struct virtnet_tx *txvq, struct rte_mbuf *cookie,
                /* setup first tx ring slot to point to header
                 * stored in reserved region.
                 */
-               offs = idx * sizeof(struct virtio_tx_region)
-                       + offsetof(struct virtio_tx_region, tx_hdr);
-
-               start_dp[idx].addr  = txvq->virtio_net_hdr_mem + offs;
+               start_dp[idx].addr  = txvq->virtio_net_hdr_mem +
+                       RTE_PTR_DIFF(&txr[idx].tx_hdr, txr);
                start_dp[idx].len   = vq->hw->vtnet_hdr_size;
                start_dp[idx].flags = VRING_DESC_F_NEXT;
+               hdr = (struct virtio_net_hdr *)&txr[idx].tx_hdr;
+
                idx = start_dp[idx].next;
        }
 
+       if (offload) {
+               /* Checksum Offload */
+               switch (cookie->ol_flags & PKT_TX_L4_MASK) {
+               case PKT_TX_UDP_CKSUM:
+                       hdr->csum_start = cookie->l2_len + cookie->l3_len;
+                       hdr->csum_offset = offsetof(struct udp_hdr,
+                               dgram_cksum);
+                       hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
+                       break;
+
+               case PKT_TX_TCP_CKSUM:
+                       hdr->csum_start = cookie->l2_len + cookie->l3_len;
+                       hdr->csum_offset = offsetof(struct tcp_hdr, cksum);
+                       hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
+                       break;
+
+               default:
+                       hdr->csum_start = 0;
+                       hdr->csum_offset = 0;
+                       hdr->flags = 0;
+                       break;
+               }
+
+               hdr->gso_type = 0;
+               hdr->gso_size = 0;
+               hdr->hdr_len = 0;
+       }
+
        do {
                start_dp[idx].addr  = VIRTIO_MBUF_DATA_DMA_ADDR(cookie, vq);
                start_dp[idx].len   = cookie->data_len;
@@ -527,11 +565,6 @@ virtio_dev_tx_queue_setup(struct rte_eth_dev *dev,
 
        PMD_INIT_FUNC_TRACE();
 
-       if ((tx_conf->txq_flags & ETH_TXQ_FLAGS_NOXSUMS)
-           != ETH_TXQ_FLAGS_NOXSUMS) {
-               PMD_INIT_LOG(ERR, "TX checksum offload not supported\n");
-               return -EINVAL;
-       }
 
        virtio_update_rxtx_handler(dev, tx_conf);