aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/net/ethernet/intel/i40e
diff options
context:
space:
mode:
authorMagnus Karlsson <magnus.karlsson@intel.com>2018-08-28 14:44:34 +0200
committerAlexei Starovoitov <ast@kernel.org>2018-08-29 12:25:53 -0700
commit1328dcddbd53900481994f4c4d8b79e64477965e (patch)
tree545afb0764c2764022bd0a215e91a4ebf0b28e0f /drivers/net/ethernet/intel/i40e
parenti40e: move common Tx functions to i40e_txrx_common.h (diff)
downloadlinux-dev-1328dcddbd53900481994f4c4d8b79e64477965e.tar.xz
linux-dev-1328dcddbd53900481994f4c4d8b79e64477965e.zip
i40e: add AF_XDP zero-copy Tx support
This patch adds zero-copy Tx support for AF_XDP sockets. It implements the ndo_xsk_async_xmit netdev ndo and performs all the Tx logic from a NAPI context. This means pulling egress packets from the Tx ring, placing the frames on the NIC HW descriptor ring and completing sent frames back to the application via the completion ring. The regular XDP Tx ring is used for AF_XDP as well. This rationale for this is as follows: XDP_REDIRECT guarantees mutual exclusion between different NAPI contexts based on CPU id. In other words, a netdev can XDP_REDIRECT to another netdev with a different NAPI context, since the operation is bound to a specific core and each core has its own hardware ring. As the AF_XDP Tx action is running in the same NAPI context and using the same ring, it will also be protected from XDP_REDIRECT actions with the exact same mechanism. As with AF_XDP Rx, all AF_XDP Tx specific functions are added to i40e_xsk.c. Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Diffstat (limited to 'drivers/net/ethernet/intel/i40e')
-rw-r--r--drivers/net/ethernet/intel/i40e/i40e_main.c4
-rw-r--r--drivers/net/ethernet/intel/i40e/i40e_txrx.c6
-rw-r--r--drivers/net/ethernet/intel/i40e/i40e_xsk.c173
-rw-r--r--drivers/net/ethernet/intel/i40e/i40e_xsk.h4
4 files changed, 186 insertions, 1 deletions
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 848eea7c84db..5da7eb0fe4ae 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3074,6 +3074,9 @@ static int i40e_configure_tx_ring(struct i40e_ring *ring)
i40e_status err = 0;
u32 qtx_ctl = 0;
+ if (ring_is_xdp(ring))
+ ring->xsk_umem = i40e_xsk_umem(ring);
+
/* some ATR related tx ring init */
if (vsi->back->flags & I40E_FLAG_FD_ATR_ENABLED) {
ring->atr_sample_rate = vsi->back->atr_sample_rate;
@@ -12185,6 +12188,7 @@ static const struct net_device_ops i40e_netdev_ops = {
.ndo_bridge_setlink = i40e_ndo_bridge_setlink,
.ndo_bpf = i40e_xdp,
.ndo_xdp_xmit = i40e_xdp_xmit,
+ .ndo_xsk_async_xmit = i40e_xsk_async_xmit,
};
/**
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 11e201fcb57a..37bd4e50ccde 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2597,7 +2597,11 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)
* budget and be more aggressive about cleaning up the Tx descriptors.
*/
i40e_for_each_ring(ring, q_vector->tx) {
- if (!i40e_clean_tx_irq(vsi, ring, budget)) {
+ bool wd = ring->xsk_umem ?
+ i40e_clean_xdp_tx_irq(vsi, ring, budget) :
+ i40e_clean_tx_irq(vsi, ring, budget);
+
+ if (!wd) {
clean_complete = false;
continue;
}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c b/drivers/net/ethernet/intel/i40e/i40e_xsk.c
index bf502f2307c2..94947a826bc3 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c
@@ -659,3 +659,176 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget)
return failure ? budget : (int)total_rx_packets;
}
+/**
+ * i40e_xmit_zc - Performs zero-copy Tx AF_XDP
+ * @xdp_ring: XDP Tx ring
+ * @budget: NAPI budget
+ *
+ * Returns true if the work is finished.
+ **/
+static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget)
+{
+ unsigned int total_packets = 0;
+ struct i40e_tx_buffer *tx_bi;
+ struct i40e_tx_desc *tx_desc;
+ bool work_done = true;
+ dma_addr_t dma;
+ u32 len;
+
+ while (budget-- > 0) {
+ if (!unlikely(I40E_DESC_UNUSED(xdp_ring))) {
+ xdp_ring->tx_stats.tx_busy++;
+ work_done = false;
+ break;
+ }
+
+ if (!xsk_umem_consume_tx(xdp_ring->xsk_umem, &dma, &len))
+ break;
+
+ dma_sync_single_for_device(xdp_ring->dev, dma, len,
+ DMA_BIDIRECTIONAL);
+
+ tx_bi = &xdp_ring->tx_bi[xdp_ring->next_to_use];
+ tx_bi->bytecount = len;
+
+ tx_desc = I40E_TX_DESC(xdp_ring, xdp_ring->next_to_use);
+ tx_desc->buffer_addr = cpu_to_le64(dma);
+ tx_desc->cmd_type_offset_bsz =
+ build_ctob(I40E_TX_DESC_CMD_ICRC
+ | I40E_TX_DESC_CMD_EOP,
+ 0, len, 0);
+ total_packets++;
+
+ xdp_ring->next_to_use++;
+ if (xdp_ring->next_to_use == xdp_ring->count)
+ xdp_ring->next_to_use = 0;
+ }
+
+ if (total_packets > 0) {
+ /* Request an interrupt for the last frame and bump tail ptr. */
+ tx_desc->cmd_type_offset_bsz |= (I40E_TX_DESC_CMD_RS <<
+ I40E_TXD_QW1_CMD_SHIFT);
+ i40e_xdp_ring_update_tail(xdp_ring);
+
+ xsk_umem_consume_tx_done(xdp_ring->xsk_umem);
+ }
+
+ return !!budget && work_done;
+}
+
+/**
+ * i40e_clean_xdp_tx_buffer - Frees and unmaps an XDP Tx entry
+ * @tx_ring: XDP Tx ring
+ * @tx_bi: Tx buffer info to clean
+ **/
+static void i40e_clean_xdp_tx_buffer(struct i40e_ring *tx_ring,
+ struct i40e_tx_buffer *tx_bi)
+{
+ xdp_return_frame(tx_bi->xdpf);
+ dma_unmap_single(tx_ring->dev,
+ dma_unmap_addr(tx_bi, dma),
+ dma_unmap_len(tx_bi, len), DMA_TO_DEVICE);
+ dma_unmap_len_set(tx_bi, len, 0);
+}
+
+/**
+ * i40e_clean_xdp_tx_irq - Completes AF_XDP entries, and cleans XDP entries
+ * @tx_ring: XDP Tx ring
+ * @tx_bi: Tx buffer info to clean
+ *
+ * Returns true if cleanup/tranmission is done.
+ **/
+bool i40e_clean_xdp_tx_irq(struct i40e_vsi *vsi,
+ struct i40e_ring *tx_ring, int napi_budget)
+{
+ unsigned int ntc, total_bytes = 0, budget = vsi->work_limit;
+ u32 i, completed_frames, frames_ready, xsk_frames = 0;
+ struct xdp_umem *umem = tx_ring->xsk_umem;
+ u32 head_idx = i40e_get_head(tx_ring);
+ bool work_done = true, xmit_done;
+ struct i40e_tx_buffer *tx_bi;
+
+ if (head_idx < tx_ring->next_to_clean)
+ head_idx += tx_ring->count;
+ frames_ready = head_idx - tx_ring->next_to_clean;
+
+ if (frames_ready == 0) {
+ goto out_xmit;
+ } else if (frames_ready > budget) {
+ completed_frames = budget;
+ work_done = false;
+ } else {
+ completed_frames = frames_ready;
+ }
+
+ ntc = tx_ring->next_to_clean;
+
+ for (i = 0; i < completed_frames; i++) {
+ tx_bi = &tx_ring->tx_bi[ntc];
+
+ if (tx_bi->xdpf)
+ i40e_clean_xdp_tx_buffer(tx_ring, tx_bi);
+ else
+ xsk_frames++;
+
+ tx_bi->xdpf = NULL;
+ total_bytes += tx_bi->bytecount;
+
+ if (++ntc >= tx_ring->count)
+ ntc = 0;
+ }
+
+ tx_ring->next_to_clean += completed_frames;
+ if (unlikely(tx_ring->next_to_clean >= tx_ring->count))
+ tx_ring->next_to_clean -= tx_ring->count;
+
+ if (xsk_frames)
+ xsk_umem_complete_tx(umem, xsk_frames);
+
+ i40e_arm_wb(tx_ring, vsi, budget);
+ i40e_update_tx_stats(tx_ring, completed_frames, total_bytes);
+
+out_xmit:
+ xmit_done = i40e_xmit_zc(tx_ring, budget);
+
+ return work_done && xmit_done;
+}
+
+/**
+ * i40e_xsk_async_xmit - Implements the ndo_xsk_async_xmit
+ * @dev: the netdevice
+ * @queue_id: queue id to wake up
+ *
+ * Returns <0 for errors, 0 otherwise.
+ **/
+int i40e_xsk_async_xmit(struct net_device *dev, u32 queue_id)
+{
+ struct i40e_netdev_priv *np = netdev_priv(dev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_ring *ring;
+
+ if (test_bit(__I40E_VSI_DOWN, vsi->state))
+ return -ENETDOWN;
+
+ if (!i40e_enabled_xdp_vsi(vsi))
+ return -ENXIO;
+
+ if (queue_id >= vsi->num_queue_pairs)
+ return -ENXIO;
+
+ if (!vsi->xdp_rings[queue_id]->xsk_umem)
+ return -ENXIO;
+
+ ring = vsi->xdp_rings[queue_id];
+
+ /* The idea here is that if NAPI is running, mark a miss, so
+ * it will run again. If not, trigger an interrupt and
+ * schedule the NAPI from interrupt context. If NAPI would be
+ * scheduled here, the interrupt affinity would not be
+ * honored.
+ */
+ if (!napi_if_scheduled_mark_missed(&ring->q_vector->napi))
+ i40e_force_wb(vsi, ring->q_vector);
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.h b/drivers/net/ethernet/intel/i40e/i40e_xsk.h
index 427a844f78a7..9038c5d5cf08 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_xsk.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.h
@@ -18,4 +18,8 @@ void i40e_zca_free(struct zero_copy_allocator *alloc, unsigned long handle);
bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 cleaned_count);
int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget);
+bool i40e_clean_xdp_tx_irq(struct i40e_vsi *vsi,
+ struct i40e_ring *tx_ring, int napi_budget);
+int i40e_xsk_async_xmit(struct net_device *dev, u32 queue_id);
+
#endif /* _I40E_XSK_H_ */