net/mlx5e: Add support for XDP_REDIRECT in device-out side

Add implementation for the ndo_xdp_xmit callback.

Dedicate a new set of XDP-SQ instances to satisfy the XDP_REDIRECT
requests.  These instances are totally separated from the existing
XDP-SQ objects that satisfy local XDP_TX actions.

Performance tests:

xdp_redirect_map from ConnectX-5 to ConnectX-5.
CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
Packet-rate of 64B packets.

Single queue: 7 Mpps.
Multi queue: 55 Mpps.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

drivers/net/ethernet/mellanox/mlx5/core/en.h

@@ -344,6 +344,7 @@ enum {
 	MLX5E_SQ_STATE_IPSEC,
 	MLX5E_SQ_STATE_AM,
 	MLX5E_SQ_STATE_TLS,
+	MLX5E_SQ_STATE_REDIRECT,
 };
 
 struct mlx5e_sq_wqe_info {
@@ -598,6 +599,9 @@ struct mlx5e_channel {
 	__be32                     mkey_be;
 	u8                         num_tc;
 
+	/* XDP_REDIRECT */
+	struct mlx5e_xdpsq         xdpsq;
+
 	/* data path - accessed per napi poll */
 	struct irq_desc *irq_desc;
 	struct mlx5e_ch_stats     *stats;
@@ -621,6 +625,7 @@ struct mlx5e_channel_stats {
 	struct mlx5e_sq_stats sq[MLX5E_MAX_NUM_TC];
 	struct mlx5e_rq_stats rq;
 	struct mlx5e_xdpsq_stats rq_xdpsq;
+	struct mlx5e_xdpsq_stats xdpsq;
 } ____cacheline_aligned_in_smp;
 
 enum mlx5e_traffic_types {

drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c

@@ -167,6 +167,7 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq)
 	struct mlx5e_xdpsq *sq;
 	struct mlx5_cqe64 *cqe;
 	struct mlx5e_rq *rq;
+	bool is_redirect;
 	u16 sqcc;
 	int i;
 
@@ -179,6 +180,7 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq)
 	if (!cqe)
 		return false;
 
+	is_redirect = test_bit(MLX5E_SQ_STATE_REDIRECT, &sq->state);
 	rq = container_of(sq, struct mlx5e_rq, xdpsq);
 
 	/* sq->cc must be updated only after mlx5_cqwq_update_db_record(),
@@ -196,17 +198,20 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq)
 		wqe_counter = be16_to_cpu(cqe->wqe_counter);
 
 		do {
-			struct mlx5e_xdp_info *xdpi;
-			u16 ci;
+			u16 ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc);
+			struct mlx5e_xdp_info *xdpi = &sq->db.xdpi[ci];
 
 			last_wqe = (sqcc == wqe_counter);
-
-			ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc);
-			xdpi = &sq->db.xdpi[ci];
-
 			sqcc++;
+
+			if (is_redirect) {
+				xdp_return_frame(xdpi->xdpf);
+				dma_unmap_single(sq->pdev, xdpi->dma_addr,
+						 xdpi->xdpf->len, DMA_TO_DEVICE);
+			} else {
 				/* Recycle RX page */
 				mlx5e_page_release(rq, &xdpi->di, true);
+			}
 		} while (!last_wqe);
 	} while ((++i < MLX5E_TX_CQ_POLL_BUDGET) && (cqe = mlx5_cqwq_get_cqe(&cq->wq)));
 
@@ -223,16 +228,75 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq)
 
 void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq)
 {
-	struct mlx5e_rq *rq = container_of(sq, struct mlx5e_rq, xdpsq);
-	struct mlx5e_xdp_info *xdpi;
-	u16 ci;
+	struct mlx5e_rq *rq;
+	bool is_redirect;
+
+	is_redirect = test_bit(MLX5E_SQ_STATE_REDIRECT, &sq->state);
+	rq = is_redirect ? NULL : container_of(sq, struct mlx5e_rq, xdpsq);
 
 	while (sq->cc != sq->pc) {
-		ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->cc);
-		xdpi = &sq->db.xdpi[ci];
+		u16 ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->cc);
+		struct mlx5e_xdp_info *xdpi = &sq->db.xdpi[ci];
+
 		sq->cc++;
 
+		if (is_redirect) {
+			xdp_return_frame(xdpi->xdpf);
+			dma_unmap_single(sq->pdev, xdpi->dma_addr,
+					 xdpi->xdpf->len, DMA_TO_DEVICE);
+		} else {
+			/* Recycle RX page */
 			mlx5e_page_release(rq, &xdpi->di, false);
 		}
+	}
 }
 
+int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
+		   u32 flags)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5e_xdpsq *sq;
+	int drops = 0;
+	int sq_num;
+	int i;
+
+	if (unlikely(!test_bit(MLX5E_STATE_OPENED, &priv->state)))
+		return -ENETDOWN;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+		return -EINVAL;
+
+	sq_num = smp_processor_id();
+
+	if (unlikely(sq_num >= priv->channels.num))
+		return -ENXIO;
+
+	sq = &priv->channels.c[sq_num]->xdpsq;
+
+	if (unlikely(!test_bit(MLX5E_SQ_STATE_ENABLED, &sq->state)))
+		return -ENETDOWN;
+
+	for (i = 0; i < n; i++) {
+		struct xdp_frame *xdpf = frames[i];
+		struct mlx5e_xdp_info xdpi;
+
+		xdpi.dma_addr = dma_map_single(sq->pdev, xdpf->data, xdpf->len,
+					       DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(sq->pdev, xdpi.dma_addr))) {
+			drops++;
+			continue;
+		}
+
+		xdpi.xdpf = xdpf;
+
+		if (unlikely(!mlx5e_xmit_xdp_frame(sq, &xdpi))) {
+			xdp_return_frame_rx_napi(xdpf);
+			drops++;
+		}
+	}
+
+	if (flags & XDP_XMIT_FLUSH)
+		mlx5e_xmit_xdp_doorbell(sq);
+
+	return n - drops;
+}

drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h

@@ -46,6 +46,8 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq);
 void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq);
 
 bool mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xdp_info *xdpi);
+int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
+		   u32 flags);
 
 static inline void mlx5e_xmit_xdp_doorbell(struct mlx5e_xdpsq *sq)
 {

drivers/net/ethernet/mellanox/mlx5/core/en_main.c

@@ -988,7 +988,8 @@ static int mlx5e_alloc_xdpsq_db(struct mlx5e_xdpsq *sq, int numa)
 static int mlx5e_alloc_xdpsq(struct mlx5e_channel *c,
 			     struct mlx5e_params *params,
 			     struct mlx5e_sq_param *param,
-			     struct mlx5e_xdpsq *sq)
+			     struct mlx5e_xdpsq *sq,
+			     bool is_redirect)
 {
 	void *sqc_wq               = MLX5_ADDR_OF(sqc, param->sqc, wq);
 	struct mlx5_core_dev *mdev = c->mdev;
@@ -1001,7 +1002,9 @@ static int mlx5e_alloc_xdpsq(struct mlx5e_channel *c,
 	sq->uar_map   = mdev->mlx5e_res.bfreg.map;
 	sq->min_inline_mode = params->tx_min_inline_mode;
 	sq->hw_mtu    = MLX5E_SW2HW_MTU(params, params->sw_mtu);
-	sq->stats     = &c->priv->channel_stats[c->ix].rq_xdpsq;
+	sq->stats     = is_redirect ?
+		&c->priv->channel_stats[c->ix].xdpsq :
+		&c->priv->channel_stats[c->ix].rq_xdpsq;
 
 	param->wq.db_numa_node = cpu_to_node(c->cpu);
 	err = mlx5_wq_cyc_create(mdev, &param->wq, sqc_wq, wq, &sq->wq_ctrl);
@@ -1531,7 +1534,8 @@ static void mlx5e_close_icosq(struct mlx5e_icosq *sq)
 static int mlx5e_open_xdpsq(struct mlx5e_channel *c,
 			    struct mlx5e_params *params,
 			    struct mlx5e_sq_param *param,
-			    struct mlx5e_xdpsq *sq)
+			    struct mlx5e_xdpsq *sq,
+			    bool is_redirect)
 {
 	unsigned int ds_cnt = MLX5E_XDP_TX_DS_COUNT;
 	struct mlx5e_create_sq_param csp = {};
@@ -1539,7 +1543,7 @@ static int mlx5e_open_xdpsq(struct mlx5e_channel *c,
 	int err;
 	int i;
 
-	err = mlx5e_alloc_xdpsq(c, params, param, sq);
+	err = mlx5e_alloc_xdpsq(c, params, param, sq, is_redirect);
 	if (err)
 		return err;
 
@@ -1548,6 +1552,8 @@ static int mlx5e_open_xdpsq(struct mlx5e_channel *c,
 	csp.cqn             = sq->cq.mcq.cqn;
 	csp.wq_ctrl         = &sq->wq_ctrl;
 	csp.min_inline_mode = sq->min_inline_mode;
+	if (is_redirect)
+		set_bit(MLX5E_SQ_STATE_REDIRECT, &sq->state);
 	set_bit(MLX5E_SQ_STATE_ENABLED, &sq->state);
 	err = mlx5e_create_sq_rdy(c->mdev, param, &csp, &sq->sqn);
 	if (err)
@@ -1930,10 +1936,14 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix,
 	if (err)
 		goto err_close_icosq_cq;
 
-	err = mlx5e_open_cq(c, params->rx_cq_moderation, &cparam->rx_cq, &c->rq.cq);
+	err = mlx5e_open_cq(c, params->tx_cq_moderation, &cparam->tx_cq, &c->xdpsq.cq);
 	if (err)
 		goto err_close_tx_cqs;
 
+	err = mlx5e_open_cq(c, params->rx_cq_moderation, &cparam->rx_cq, &c->rq.cq);
+	if (err)
+		goto err_close_xdp_tx_cqs;
+
 	/* XDP SQ CQ params are same as normal TXQ sq CQ params */
 	err = c->xdp ? mlx5e_open_cq(c, params->tx_cq_moderation,
 				     &cparam->tx_cq, &c->rq.xdpsq.cq) : 0;
@@ -1950,7 +1960,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix,
 	if (err)
 		goto err_close_icosq;
 
-	err = c->xdp ? mlx5e_open_xdpsq(c, params, &cparam->xdp_sq, &c->rq.xdpsq) : 0;
+	err = c->xdp ? mlx5e_open_xdpsq(c, params, &cparam->xdp_sq, &c->rq.xdpsq, false) : 0;
 	if (err)
 		goto err_close_sqs;
 
@@ -1958,9 +1968,17 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix,
 	if (err)
 		goto err_close_xdp_sq;
 
+	err = mlx5e_open_xdpsq(c, params, &cparam->xdp_sq, &c->xdpsq, true);
+	if (err)
+		goto err_close_rq;
+
 	*cp = c;
 
 	return 0;
+
+err_close_rq:
+	mlx5e_close_rq(&c->rq);
+
 err_close_xdp_sq:
 	if (c->xdp)
 		mlx5e_close_xdpsq(&c->rq.xdpsq);
@@ -1979,6 +1997,9 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix,
 err_close_rx_cq:
 	mlx5e_close_cq(&c->rq.cq);
 
+err_close_xdp_tx_cqs:
+	mlx5e_close_cq(&c->xdpsq.cq);
+
 err_close_tx_cqs:
 	mlx5e_close_tx_cqs(c);
 
@@ -2013,6 +2034,7 @@ static void mlx5e_deactivate_channel(struct mlx5e_channel *c)
 
 static void mlx5e_close_channel(struct mlx5e_channel *c)
 {
+	mlx5e_close_xdpsq(&c->xdpsq);
 	mlx5e_close_rq(&c->rq);
 	if (c->xdp)
 		mlx5e_close_xdpsq(&c->rq.xdpsq);
@@ -2022,6 +2044,7 @@ static void mlx5e_close_channel(struct mlx5e_channel *c)
 	if (c->xdp)
 		mlx5e_close_cq(&c->rq.xdpsq.cq);
 	mlx5e_close_cq(&c->rq.cq);
+	mlx5e_close_cq(&c->xdpsq.cq);
 	mlx5e_close_tx_cqs(c);
 	mlx5e_close_cq(&c->icosq.cq);
 	netif_napi_del(&c->napi);
@@ -4278,6 +4301,7 @@ static const struct net_device_ops mlx5e_netdev_ops = {
 #endif
 	.ndo_tx_timeout          = mlx5e_tx_timeout,
 	.ndo_bpf		 = mlx5e_xdp,
+	.ndo_xdp_xmit            = mlx5e_xdp_xmit,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller     = mlx5e_netpoll,
 #endif

drivers/net/ethernet/mellanox/mlx5/core/en_stats.c

@@ -75,6 +75,10 @@ static const struct counter_desc sw_stats_desc[] = {
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_queue_wake) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_udp_seg_rem) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_cqe_err) },
+	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_xdp_xmit) },
+	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_xdp_full) },
+	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_xdp_err) },
+	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_xdp_cqes) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_wqe_err) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_mpwqe_filler_cqes) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_mpwqe_filler_strides) },
@@ -130,6 +134,7 @@ void mlx5e_grp_sw_update_stats(struct mlx5e_priv *priv)
 	for (i = 0; i < priv->profile->max_nch(priv->mdev); i++) {
 		struct mlx5e_channel_stats *channel_stats =
 			&priv->channel_stats[i];
+		struct mlx5e_xdpsq_stats *xdpsq_red_stats = &channel_stats->xdpsq;
 		struct mlx5e_xdpsq_stats *xdpsq_stats = &channel_stats->rq_xdpsq;
 		struct mlx5e_rq_stats *rq_stats = &channel_stats->rq;
 		struct mlx5e_ch_stats *ch_stats = &channel_stats->ch;
@@ -168,6 +173,11 @@ void mlx5e_grp_sw_update_stats(struct mlx5e_priv *priv)
 		s->ch_arm         += ch_stats->arm;
 		s->ch_aff_change  += ch_stats->aff_change;
 		s->ch_eq_rearm    += ch_stats->eq_rearm;
+		/* xdp redirect */
+		s->tx_xdp_xmit    += xdpsq_red_stats->xmit;
+		s->tx_xdp_full    += xdpsq_red_stats->full;
+		s->tx_xdp_err     += xdpsq_red_stats->err;
+		s->tx_xdp_cqes    += xdpsq_red_stats->cqes;
 
 		for (j = 0; j < priv->max_opened_tc; j++) {
 			struct mlx5e_sq_stats *sq_stats = &channel_stats->sq[j];
@@ -1178,6 +1188,13 @@ static const struct counter_desc rq_xdpsq_stats_desc[] = {
 	{ MLX5E_DECLARE_RQ_XDPSQ_STAT(struct mlx5e_xdpsq_stats, cqes) },
 };
 
+static const struct counter_desc xdpsq_stats_desc[] = {
+	{ MLX5E_DECLARE_XDPSQ_STAT(struct mlx5e_xdpsq_stats, xmit) },
+	{ MLX5E_DECLARE_XDPSQ_STAT(struct mlx5e_xdpsq_stats, full) },
+	{ MLX5E_DECLARE_XDPSQ_STAT(struct mlx5e_xdpsq_stats, err) },
+	{ MLX5E_DECLARE_XDPSQ_STAT(struct mlx5e_xdpsq_stats, cqes) },
+};
+
 static const struct counter_desc ch_stats_desc[] = {
 	{ MLX5E_DECLARE_CH_STAT(struct mlx5e_ch_stats, events) },
 	{ MLX5E_DECLARE_CH_STAT(struct mlx5e_ch_stats, poll) },
@@ -1188,6 +1205,7 @@ static const struct counter_desc ch_stats_desc[] = {
 
 #define NUM_RQ_STATS			ARRAY_SIZE(rq_stats_desc)
 #define NUM_SQ_STATS			ARRAY_SIZE(sq_stats_desc)
+#define NUM_XDPSQ_STATS			ARRAY_SIZE(xdpsq_stats_desc)
 #define NUM_RQ_XDPSQ_STATS		ARRAY_SIZE(rq_xdpsq_stats_desc)
 #define NUM_CH_STATS			ARRAY_SIZE(ch_stats_desc)
 
@@ -1198,7 +1216,8 @@ static int mlx5e_grp_channels_get_num_stats(struct mlx5e_priv *priv)
 	return (NUM_RQ_STATS * max_nch) +
 	       (NUM_CH_STATS * max_nch) +
 	       (NUM_SQ_STATS * max_nch * priv->max_opened_tc) +
-	       (NUM_RQ_XDPSQ_STATS * max_nch);
+	       (NUM_RQ_XDPSQ_STATS * max_nch) +
+	       (NUM_XDPSQ_STATS * max_nch);
 }
 
 static int mlx5e_grp_channels_fill_strings(struct mlx5e_priv *priv, u8 *data,
@@ -1228,6 +1247,11 @@ static int mlx5e_grp_channels_fill_strings(struct mlx5e_priv *priv, u8 *data,
 					sq_stats_desc[j].format,
 					priv->channel_tc2txq[i][tc]);
 
+	for (i = 0; i < max_nch; i++)
+		for (j = 0; j < NUM_XDPSQ_STATS; j++)
+			sprintf(data + (idx++) * ETH_GSTRING_LEN,
+				xdpsq_stats_desc[j].format, i);
+
 	return idx;
 }
 
@@ -1261,6 +1285,12 @@ static int mlx5e_grp_channels_fill_stats(struct mlx5e_priv *priv, u64 *data,
 					MLX5E_READ_CTR64_CPU(&priv->channel_stats[i].sq[tc],
 							     sq_stats_desc, j);
 
+	for (i = 0; i < max_nch; i++)
+		for (j = 0; j < NUM_XDPSQ_STATS; j++)
+			data[idx++] =
+				MLX5E_READ_CTR64_CPU(&priv->channel_stats[i].xdpsq,
+						     xdpsq_stats_desc, j);
+
 	return idx;
 }
 

drivers/net/ethernet/mellanox/mlx5/core/en_stats.h

@@ -44,6 +44,7 @@
 #define MLX5E_DECLARE_STAT(type, fld) #fld, offsetof(type, fld)
 #define MLX5E_DECLARE_RX_STAT(type, fld) "rx%d_"#fld, offsetof(type, fld)
 #define MLX5E_DECLARE_TX_STAT(type, fld) "tx%d_"#fld, offsetof(type, fld)
+#define MLX5E_DECLARE_XDPSQ_STAT(type, fld) "tx%d_xdp_"#fld, offsetof(type, fld)
 #define MLX5E_DECLARE_RQ_XDPSQ_STAT(type, fld) "rx%d_xdp_tx_"#fld, offsetof(type, fld)
 #define MLX5E_DECLARE_CH_STAT(type, fld) "ch%d_"#fld, offsetof(type, fld)
 
@@ -87,6 +88,10 @@ struct mlx5e_sw_stats {
 	u64 tx_queue_wake;
 	u64 tx_udp_seg_rem;
 	u64 tx_cqe_err;
+	u64 tx_xdp_xmit;
+	u64 tx_xdp_full;
+	u64 tx_xdp_err;
+	u64 tx_xdp_cqes;
 	u64 rx_wqe_err;
 	u64 rx_mpwqe_filler_cqes;
 	u64 rx_mpwqe_filler_strides;

drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c

@@ -85,6 +85,8 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget)
 	for (i = 0; i < c->num_tc; i++)
 		busy |= mlx5e_poll_tx_cq(&c->sq[i].cq, budget);
 
+	busy |= mlx5e_poll_xdpsq_cq(&c->xdpsq.cq);
+
 	if (c->xdp)
 		busy |= mlx5e_poll_xdpsq_cq(&c->rq.xdpsq.cq);
 
@@ -117,6 +119,7 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget)
 
 	mlx5e_cq_arm(&c->rq.cq);
 	mlx5e_cq_arm(&c->icosq.cq);
+	mlx5e_cq_arm(&c->xdpsq.cq);
 
 	return work_done;
 }