Merge branch 'mlx5-xdp-perf-optimizations'

Saeed Mahameed says:

====================
Mellanox mlx5e XDP performance optimization

This series provides some preformancee optimizations for mlx5e
driver, especially for XDP TX flows.

1st patch is a simple change of rmb to dma_rmb in CQE fetch routine
which shows a huge gain for both RX and TX packet rates.

2nd patch removes write combining logic from the driver TX handler
and simplifies the TX logic while improving TX CPU utilization.

All other patches combined provide some refactoring to the driver TX
flows to allow some significant XDP TX improvements.

More details and performance numbers per patch can be found in each patch
commit message compared to the preceding patch.

Overall performance improvemnets
  System: Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz

Test case                   Baseline      Now      improvement
---------------------------------------------------------------
TX packets (24 threads)     45Mpps        54Mpps      20%
TC stack Drop (1 core)      3.45Mpps      3.6Mpps     5%
XDP Drop      (1 core)      14Mpps        16.9Mpps    20%
XDP TX        (1 core)      10.4Mpps      13.7Mpps    31%
====================
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

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

@@ -111,18 +111,13 @@
 #define MLX5E_MAX_NUM_SQS              (MLX5E_MAX_NUM_CHANNELS * MLX5E_MAX_NUM_TC)
 #define MLX5E_TX_CQ_POLL_BUDGET        128
 #define MLX5E_UPDATE_STATS_INTERVAL    200 /* msecs */
-#define MLX5E_SQ_BF_BUDGET             16
 
 #define MLX5E_ICOSQ_MAX_WQEBBS \
 	(DIV_ROUND_UP(sizeof(struct mlx5e_umr_wqe), MLX5_SEND_WQE_BB))
 
 #define MLX5E_XDP_MIN_INLINE (ETH_HLEN + VLAN_HLEN)
-#define MLX5E_XDP_IHS_DS_COUNT \
-	DIV_ROUND_UP(MLX5E_XDP_MIN_INLINE - 2, MLX5_SEND_WQE_DS)
 #define MLX5E_XDP_TX_DS_COUNT \
 	((sizeof(struct mlx5e_tx_wqe) / MLX5_SEND_WQE_DS) + 1 /* SG DS */)
-#define MLX5E_XDP_TX_WQEBBS \
-	DIV_ROUND_UP(MLX5E_XDP_TX_DS_COUNT, MLX5_SEND_WQEBB_NUM_DS)
 
 #define MLX5E_NUM_MAIN_GROUPS 9
 
@@ -298,19 +293,159 @@ struct mlx5e_cq {
 	struct mlx5_frag_wq_ctrl   wq_ctrl;
 } ____cacheline_aligned_in_smp;
 
-struct mlx5e_rq;
-typedef void (*mlx5e_fp_handle_rx_cqe)(struct mlx5e_rq *rq,
-				       struct mlx5_cqe64 *cqe);
-typedef int (*mlx5e_fp_alloc_wqe)(struct mlx5e_rq *rq, struct mlx5e_rx_wqe *wqe,
-				  u16 ix);
+struct mlx5e_tx_wqe_info {
+	u32 num_bytes;
+	u8  num_wqebbs;
+	u8  num_dma;
+};
+
+enum mlx5e_dma_map_type {
+	MLX5E_DMA_MAP_SINGLE,
+	MLX5E_DMA_MAP_PAGE
+};
+
+struct mlx5e_sq_dma {
+	dma_addr_t              addr;
+	u32                     size;
+	enum mlx5e_dma_map_type type;
+};
+
+enum {
+	MLX5E_SQ_STATE_ENABLED,
+};
+
+struct mlx5e_sq_wqe_info {
+	u8  opcode;
+	u8  num_wqebbs;
+};
+
+struct mlx5e_txqsq {
+	/* data path */
+
+	/* dirtied @completion */
+	u16                        cc;
+	u32                        dma_fifo_cc;
+
+	/* dirtied @xmit */
+	u16                        pc ____cacheline_aligned_in_smp;
+	u32                        dma_fifo_pc;
+	struct mlx5e_sq_stats      stats;
+
+	struct mlx5e_cq            cq;
+
+	/* write@xmit, read@completion */
+	struct {
+		struct sk_buff           **skb;
+		struct mlx5e_sq_dma       *dma_fifo;
+		struct mlx5e_tx_wqe_info  *wqe_info;
+	} db;
+
+	/* read only */
+	struct mlx5_wq_cyc         wq;
+	u32                        dma_fifo_mask;
+	void __iomem              *uar_map;
+	struct netdev_queue       *txq;
+	u32                        sqn;
+	u16                        max_inline;
+	u8                         min_inline_mode;
+	u16                        edge;
+	struct device             *pdev;
+	struct mlx5e_tstamp       *tstamp;
+	__be32                     mkey_be;
+	unsigned long              state;
+
+	/* control path */
+	struct mlx5_wq_ctrl        wq_ctrl;
+	struct mlx5e_channel      *channel;
+	int                        tc;
+	u32                        rate_limit;
+} ____cacheline_aligned_in_smp;
+
+struct mlx5e_xdpsq {
+	/* data path */
+
+	/* dirtied @rx completion */
+	u16                        cc;
+	u16                        pc;
+
+	struct mlx5e_cq            cq;
+
+	/* write@xmit, read@completion */
+	struct {
+		struct mlx5e_dma_info     *di;
+		bool                       doorbell;
+	} db;
+
+	/* read only */
+	struct mlx5_wq_cyc         wq;
+	void __iomem              *uar_map;
+	u32                        sqn;
+	struct device             *pdev;
+	__be32                     mkey_be;
+	u8                         min_inline_mode;
+	unsigned long              state;
+
+	/* control path */
+	struct mlx5_wq_ctrl        wq_ctrl;
+	struct mlx5e_channel      *channel;
+} ____cacheline_aligned_in_smp;
+
+struct mlx5e_icosq {
+	/* data path */
+
+	/* dirtied @completion */
+	u16                        cc;
+
+	/* dirtied @xmit */
+	u16                        pc ____cacheline_aligned_in_smp;
+	u32                        dma_fifo_pc;
+	u16                        prev_cc;
+
+	struct mlx5e_cq            cq;
+
+	/* write@xmit, read@completion */
+	struct {
+		struct mlx5e_sq_wqe_info *ico_wqe;
+	} db;
+
+	/* read only */
+	struct mlx5_wq_cyc         wq;
+	void __iomem              *uar_map;
+	u32                        sqn;
+	u16                        edge;
+	struct device             *pdev;
+	__be32                     mkey_be;
+	unsigned long              state;
+
+	/* control path */
+	struct mlx5_wq_ctrl        wq_ctrl;
+	struct mlx5e_channel      *channel;
+} ____cacheline_aligned_in_smp;
 
-typedef void (*mlx5e_fp_dealloc_wqe)(struct mlx5e_rq *rq, u16 ix);
+static inline bool
+mlx5e_wqc_has_room_for(struct mlx5_wq_cyc *wq, u16 cc, u16 pc, u16 n)
+{
+	return (((wq->sz_m1 & (cc - pc)) >= n) || (cc == pc));
+}
 
 struct mlx5e_dma_info {
 	struct page	*page;
 	dma_addr_t	addr;
 };
 
+struct mlx5e_umr_dma_info {
+	__be64                *mtt;
+	dma_addr_t             mtt_addr;
+	struct mlx5e_dma_info  dma_info[MLX5_MPWRQ_PAGES_PER_WQE];
+	struct mlx5e_umr_wqe   wqe;
+};
+
+struct mlx5e_mpw_info {
+	struct mlx5e_umr_dma_info umr;
+	u16 consumed_strides;
+	u16 skbs_frags[MLX5_MPWRQ_PAGES_PER_WQE];
+};
+
 struct mlx5e_rx_am_stats {
 	int ppms; /* packets per msec */
 	int epms; /* events per msec */
@@ -347,6 +482,11 @@ struct mlx5e_page_cache {
 	struct mlx5e_dma_info page_cache[MLX5E_CACHE_SIZE];
 };
 
+struct mlx5e_rq;
+typedef void (*mlx5e_fp_handle_rx_cqe)(struct mlx5e_rq*, struct mlx5_cqe64*);
+typedef int (*mlx5e_fp_alloc_wqe)(struct mlx5e_rq*, struct mlx5e_rx_wqe*, u16);
+typedef void (*mlx5e_fp_dealloc_wqe)(struct mlx5e_rq*, u16);
+
 struct mlx5e_rq {
 	/* data path */
 	struct mlx5_wq_ll      wq;
@@ -381,7 +521,10 @@ struct mlx5e_rq {
 	u16                    rx_headroom;
 
 	struct mlx5e_rx_am     am; /* Adaptive Moderation */
+
+	/* XDP */
 	struct bpf_prog       *xdp_prog;
+	struct mlx5e_xdpsq     xdpsq;
 
 	/* control */
 	struct mlx5_wq_ctrl    wq_ctrl;
@@ -394,114 +537,6 @@ struct mlx5e_rq {
 	struct mlx5_core_mkey  umr_mkey;
 } ____cacheline_aligned_in_smp;
 
-struct mlx5e_umr_dma_info {
-	__be64                *mtt;
-	dma_addr_t             mtt_addr;
-	struct mlx5e_dma_info  dma_info[MLX5_MPWRQ_PAGES_PER_WQE];
-	struct mlx5e_umr_wqe   wqe;
-};
-
-struct mlx5e_mpw_info {
-	struct mlx5e_umr_dma_info umr;
-	u16 consumed_strides;
-	u16 skbs_frags[MLX5_MPWRQ_PAGES_PER_WQE];
-};
-
-struct mlx5e_tx_wqe_info {
-	u32 num_bytes;
-	u8  num_wqebbs;
-	u8  num_dma;
-};
-
-enum mlx5e_dma_map_type {
-	MLX5E_DMA_MAP_SINGLE,
-	MLX5E_DMA_MAP_PAGE
-};
-
-struct mlx5e_sq_dma {
-	dma_addr_t              addr;
-	u32                     size;
-	enum mlx5e_dma_map_type type;
-};
-
-enum {
-	MLX5E_SQ_STATE_ENABLED,
-	MLX5E_SQ_STATE_BF_ENABLE,
-};
-
-struct mlx5e_sq_wqe_info {
-	u8  opcode;
-	u8  num_wqebbs;
-};
-
-enum mlx5e_sq_type {
-	MLX5E_SQ_TXQ,
-	MLX5E_SQ_ICO,
-	MLX5E_SQ_XDP
-};
-
-struct mlx5e_sq {
-	/* data path */
-
-	/* dirtied @completion */
-	u16                        cc;
-	u32                        dma_fifo_cc;
-
-	/* dirtied @xmit */
-	u16                        pc ____cacheline_aligned_in_smp;
-	u32                        dma_fifo_pc;
-	u16                        bf_offset;
-	u16                        prev_cc;
-	u8                         bf_budget;
-	struct mlx5e_sq_stats      stats;
-
-	struct mlx5e_cq            cq;
-
-	/* pointers to per tx element info: write@xmit, read@completion */
-	union {
-		struct {
-			struct sk_buff           **skb;
-			struct mlx5e_sq_dma       *dma_fifo;
-			struct mlx5e_tx_wqe_info  *wqe_info;
-		} txq;
-		struct mlx5e_sq_wqe_info *ico_wqe;
-		struct {
-			struct mlx5e_sq_wqe_info  *wqe_info;
-			struct mlx5e_dma_info     *di;
-			bool                       doorbell;
-		} xdp;
-	} db;
-
-	/* read only */
-	struct mlx5_wq_cyc         wq;
-	u32                        dma_fifo_mask;
-	void __iomem              *uar_map;
-	struct netdev_queue       *txq;
-	u32                        sqn;
-	u16                        bf_buf_size;
-	u16                        max_inline;
-	u8                         min_inline_mode;
-	u16                        edge;
-	struct device             *pdev;
-	struct mlx5e_tstamp       *tstamp;
-	__be32                     mkey_be;
-	unsigned long              state;
-
-	/* control path */
-	struct mlx5_wq_ctrl        wq_ctrl;
-	struct mlx5_sq_bfreg	   bfreg;
-	struct mlx5e_channel      *channel;
-	int                        tc;
-	u32                        rate_limit;
-	u8                         type;
-} ____cacheline_aligned_in_smp;
-
-static inline bool mlx5e_sq_has_room_for(struct mlx5e_sq *sq, u16 n)
-{
-	return (((sq->wq.sz_m1 & (sq->cc - sq->pc)) >= n) ||
-		(sq->cc  == sq->pc));
-}
-
 enum channel_flags {
 	MLX5E_CHANNEL_NAPI_SCHED = 1,
 };
@@ -509,9 +544,8 @@ enum channel_flags {
 struct mlx5e_channel {
 	/* data path */
 	struct mlx5e_rq            rq;
-	struct mlx5e_sq            xdp_sq;
-	struct mlx5e_sq            sq[MLX5E_MAX_NUM_TC];
-	struct mlx5e_sq            icosq;   /* internal control operations */
+	struct mlx5e_txqsq         sq[MLX5E_MAX_NUM_TC];
+	struct mlx5e_icosq         icosq;   /* internal control operations */
 	bool                       xdp;
 	struct napi_struct         napi;
 	struct device             *pdev;
@@ -693,7 +727,7 @@ struct mlx5e_profile {
 
 struct mlx5e_priv {
 	/* priv data path fields - start */
-	struct mlx5e_sq            **txq_to_sq_map;
+	struct mlx5e_txqsq         **txq_to_sq_map;
 	int channeltc_to_txq_map[MLX5E_MAX_NUM_CHANNELS][MLX5E_MAX_NUM_TC];
 	struct bpf_prog *xdp_prog;
 	/* priv data path fields - end */
@@ -734,7 +768,6 @@ struct mlx5e_priv {
 
 void mlx5e_build_ptys2ethtool_map(void);
 
-void mlx5e_send_nop(struct mlx5e_sq *sq, bool notify_hw);
 u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb,
 		       void *accel_priv, select_queue_fallback_t fallback);
 netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev);
@@ -744,7 +777,9 @@ void mlx5e_cq_error_event(struct mlx5_core_cq *mcq, enum mlx5_event event);
 int mlx5e_napi_poll(struct napi_struct *napi, int budget);
 bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget);
 int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget);
-void mlx5e_free_sq_descs(struct mlx5e_sq *sq);
+bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq);
+void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq);
+void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq);
 
 void mlx5e_page_release(struct mlx5e_rq *rq, struct mlx5e_dma_info *dma_info,
 			bool recycle);
@@ -818,28 +853,40 @@ void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params,
 				 u8 cq_period_mode);
 void mlx5e_set_rq_type_params(struct mlx5e_priv *priv, u8 rq_type);
 
-static inline void mlx5e_tx_notify_hw(struct mlx5e_sq *sq,
-				      struct mlx5_wqe_ctrl_seg *ctrl, int bf_sz)
+static inline
+struct mlx5e_tx_wqe *mlx5e_post_nop(struct mlx5_wq_cyc *wq, u32 sqn, u16 *pc)
 {
-	u16 ofst = sq->bf_offset;
+	u16                         pi   = *pc & wq->sz_m1;
+	struct mlx5e_tx_wqe        *wqe  = mlx5_wq_cyc_get_wqe(wq, pi);
+	struct mlx5_wqe_ctrl_seg   *cseg = &wqe->ctrl;
 
+	memset(cseg, 0, sizeof(*cseg));
+
+	cseg->opmod_idx_opcode = cpu_to_be32((*pc << 8) | MLX5_OPCODE_NOP);
+	cseg->qpn_ds           = cpu_to_be32((sqn << 8) | 0x01);
+
+	(*pc)++;
+
+	return wqe;
+}
+
+static inline
+void mlx5e_notify_hw(struct mlx5_wq_cyc *wq, u16 pc,
+		     void __iomem *uar_map,
+		     struct mlx5_wqe_ctrl_seg *ctrl)
+{
+	ctrl->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE;
 	/* ensure wqe is visible to device before updating doorbell record */
 	dma_wmb();
 
-	*sq->wq.db = cpu_to_be32(sq->pc);
+	*wq->db = cpu_to_be32(pc);
 
 	/* ensure doorbell record is visible to device before ringing the
 	 * doorbell
 	 */
 	wmb();
-	if (bf_sz)
-		__iowrite64_copy(sq->uar_map + ofst, ctrl, bf_sz);
-	else
-		mlx5_write64((__be32 *)ctrl, sq->uar_map + ofst, NULL);
-	/* flush the write-combining mapped buffer */
-	wmb();
 
-	sq->bf_offset ^= sq->bf_buf_size;
+	mlx5_write64((__be32 *)ctrl, uar_map, NULL);
 }
 
 static inline void mlx5e_cq_arm(struct mlx5e_cq *cq)

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

@@ -107,10 +107,18 @@ int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev)
 		goto err_dealloc_transport_domain;
 	}
 
+	err = mlx5_alloc_bfreg(mdev, &res->bfreg, false, false);
+	if (err) {
+		mlx5_core_err(mdev, "alloc bfreg failed, %d\n", err);
+		goto err_destroy_mkey;
+	}
+
 	INIT_LIST_HEAD(&mdev->mlx5e_res.td.tirs_list);
 
 	return 0;
 
+err_destroy_mkey:
+	mlx5_core_destroy_mkey(mdev, &res->mkey);
 err_dealloc_transport_domain:
 	mlx5_core_dealloc_transport_domain(mdev, res->td.tdn);
 err_dealloc_pd:
@@ -122,6 +130,7 @@ void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev)
 {
 	struct mlx5e_resources *res = &mdev->mlx5e_res;
 
+	mlx5_free_bfreg(mdev, &res->bfreg);
 	mlx5_core_destroy_mkey(mdev, &res->mkey);
 	mlx5_core_dealloc_transport_domain(mdev, res->td.tdn);
 	mlx5_core_dealloc_pd(mdev, res->pdn);

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

@@ -331,7 +331,7 @@ mlx5e_copy_skb_header_mpwqe(struct device *pdev,
 static inline void mlx5e_post_umr_wqe(struct mlx5e_rq *rq, u16 ix)
 {
 	struct mlx5e_mpw_info *wi = &rq->mpwqe.info[ix];
-	struct mlx5e_sq *sq = &rq->channel->icosq;
+	struct mlx5e_icosq *sq = &rq->channel->icosq;
 	struct mlx5_wq_cyc *wq = &sq->wq;
 	struct mlx5e_umr_wqe *wqe;
 	u8 num_wqebbs = DIV_ROUND_UP(sizeof(*wqe), MLX5_SEND_WQE_BB);
@@ -341,7 +341,7 @@ static inline void mlx5e_post_umr_wqe(struct mlx5e_rq *rq, u16 ix)
 	while ((pi = (sq->pc & wq->sz_m1)) > sq->edge) {
 		sq->db.ico_wqe[pi].opcode = MLX5_OPCODE_NOP;
 		sq->db.ico_wqe[pi].num_wqebbs = 1;
-		mlx5e_send_nop(sq, false);
+		mlx5e_post_nop(wq, sq->sqn, &sq->pc);
 	}
 
 	wqe = mlx5_wq_cyc_get_wqe(wq, pi);
@@ -353,7 +353,7 @@ static inline void mlx5e_post_umr_wqe(struct mlx5e_rq *rq, u16 ix)
 	sq->db.ico_wqe[pi].opcode = MLX5_OPCODE_UMR;
 	sq->db.ico_wqe[pi].num_wqebbs = num_wqebbs;
 	sq->pc += num_wqebbs;
-	mlx5e_tx_notify_hw(sq, &wqe->ctrl, 0);
+	mlx5e_notify_hw(&sq->wq, sq->pc, sq->uar_map, &wqe->ctrl);
 }
 
 static int mlx5e_alloc_rx_umr_mpwqe(struct mlx5e_rq *rq,
@@ -637,37 +637,36 @@ static inline void mlx5e_complete_rx_cqe(struct mlx5e_rq *rq,
 	mlx5e_build_rx_skb(cqe, cqe_bcnt, rq, skb);
 }
 
-static inline void mlx5e_xmit_xdp_doorbell(struct mlx5e_sq *sq)
+static inline void mlx5e_xmit_xdp_doorbell(struct mlx5e_xdpsq *sq)
 {
 	struct mlx5_wq_cyc *wq = &sq->wq;
 	struct mlx5e_tx_wqe *wqe;
-	u16 pi = (sq->pc - MLX5E_XDP_TX_WQEBBS) & wq->sz_m1; /* last pi */
+	u16 pi = (sq->pc - 1) & wq->sz_m1; /* last pi */
 
 	wqe  = mlx5_wq_cyc_get_wqe(wq, pi);
 
-	wqe->ctrl.fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE;
-	mlx5e_tx_notify_hw(sq, &wqe->ctrl, 0);
+	mlx5e_notify_hw(wq, sq->pc, sq->uar_map, &wqe->ctrl);
 }
 
 static inline bool mlx5e_xmit_xdp_frame(struct mlx5e_rq *rq,
 					struct mlx5e_dma_info *di,
 					const struct xdp_buff *xdp)
 {
-	struct mlx5e_sq          *sq   = &rq->channel->xdp_sq;
+	struct mlx5e_xdpsq       *sq   = &rq->xdpsq;
 	struct mlx5_wq_cyc       *wq   = &sq->wq;
-	u16                      pi    = sq->pc & wq->sz_m1;
+	u16                       pi   = sq->pc & wq->sz_m1;
 	struct mlx5e_tx_wqe      *wqe  = mlx5_wq_cyc_get_wqe(wq, pi);
-	struct mlx5e_sq_wqe_info *wi   = &sq->db.xdp.wqe_info[pi];
 
 	struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl;
 	struct mlx5_wqe_eth_seg  *eseg = &wqe->eth;
 	struct mlx5_wqe_data_seg *dseg;
-	u8 ds_cnt = MLX5E_XDP_TX_DS_COUNT;
 
 	ptrdiff_t data_offset = xdp->data - xdp->data_hard_start;
 	dma_addr_t dma_addr  = di->addr + data_offset;
 	unsigned int dma_len = xdp->data_end - xdp->data;
 
+	prefetchw(wqe);
+
 	if (unlikely(dma_len < MLX5E_XDP_MIN_INLINE ||
 		     MLX5E_SW2HW_MTU(rq->netdev->mtu) < dma_len)) {
 		rq->stats.xdp_drop++;
@@ -675,48 +674,42 @@ static inline bool mlx5e_xmit_xdp_frame(struct mlx5e_rq *rq,
 		return false;
 	}
 
-	if (unlikely(!mlx5e_sq_has_room_for(sq, MLX5E_XDP_TX_WQEBBS))) {
-		if (sq->db.xdp.doorbell) {
+	if (unlikely(!mlx5e_wqc_has_room_for(wq, sq->cc, sq->pc, 1))) {
+		if (sq->db.doorbell) {
 			/* SQ is full, ring doorbell */
 			mlx5e_xmit_xdp_doorbell(sq);
-			sq->db.xdp.doorbell = false;
+			sq->db.doorbell = false;
 		}
 		rq->stats.xdp_tx_full++;
 		mlx5e_page_release(rq, di, true);
 		return false;
 	}
 
-	dma_sync_single_for_device(sq->pdev, dma_addr, dma_len,
-				   PCI_DMA_TODEVICE);
+	dma_sync_single_for_device(sq->pdev, dma_addr, dma_len, PCI_DMA_TODEVICE);
 
-	memset(wqe, 0, sizeof(*wqe));
+	cseg->fm_ce_se = 0;
 
 	dseg = (struct mlx5_wqe_data_seg *)eseg + 1;
+
 	/* copy the inline part if required */
 	if (sq->min_inline_mode != MLX5_INLINE_MODE_NONE) {
 		memcpy(eseg->inline_hdr.start, xdp->data, MLX5E_XDP_MIN_INLINE);
 		eseg->inline_hdr.sz = cpu_to_be16(MLX5E_XDP_MIN_INLINE);
 		dma_len  -= MLX5E_XDP_MIN_INLINE;
 		dma_addr += MLX5E_XDP_MIN_INLINE;
-
-		ds_cnt   += MLX5E_XDP_IHS_DS_COUNT;
 		dseg++;
 	}
 
 	/* write the dma part */
 	dseg->addr       = cpu_to_be64(dma_addr);
 	dseg->byte_count = cpu_to_be32(dma_len);
-	dseg->lkey       = sq->mkey_be;
 
 	cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_SEND);
-	cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt);
 
-	sq->db.xdp.di[pi] = *di;
-	wi->opcode     = MLX5_OPCODE_SEND;
-	wi->num_wqebbs = MLX5E_XDP_TX_WQEBBS;
-	sq->pc += MLX5E_XDP_TX_WQEBBS;
+	sq->db.di[pi] = *di;
+	sq->pc++;
 
-	sq->db.xdp.doorbell = true;
+	sq->db.doorbell = true;
 	rq->stats.xdp_tx++;
 	return true;
 }
@@ -950,7 +943,7 @@ void mlx5e_handle_rx_cqe_mpwrq(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
 {
 	struct mlx5e_rq *rq = container_of(cq, struct mlx5e_rq, cq);
-	struct mlx5e_sq *xdp_sq = &rq->channel->xdp_sq;
+	struct mlx5e_xdpsq *xdpsq = &rq->xdpsq;
 	int work_done = 0;
 
 	if (unlikely(!test_bit(MLX5E_RQ_STATE_ENABLED, &rq->state)))
@@ -977,9 +970,9 @@ int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
 		rq->handle_rx_cqe(rq, cqe);
 	}
 
-	if (xdp_sq->db.xdp.doorbell) {
-		mlx5e_xmit_xdp_doorbell(xdp_sq);
-		xdp_sq->db.xdp.doorbell = false;
+	if (xdpsq->db.doorbell) {
+		mlx5e_xmit_xdp_doorbell(xdpsq);
+		xdpsq->db.doorbell = false;
 	}
 
 	mlx5_cqwq_update_db_record(&cq->wq);
@@ -989,3 +982,74 @@ int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
 
 	return work_done;
 }
+
+bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq)
+{
+	struct mlx5e_xdpsq *sq;
+	struct mlx5e_rq *rq;
+	u16 sqcc;
+	int i;
+
+	sq = container_of(cq, struct mlx5e_xdpsq, cq);
+
+	if (unlikely(!test_bit(MLX5E_SQ_STATE_ENABLED, &sq->state)))
+		return false;
+
+	rq = container_of(sq, struct mlx5e_rq, xdpsq);
+
+	/* sq->cc must be updated only after mlx5_cqwq_update_db_record(),
+	 * otherwise a cq overrun may occur
+	 */
+	sqcc = sq->cc;
+
+	for (i = 0; i < MLX5E_TX_CQ_POLL_BUDGET; i++) {
+		struct mlx5_cqe64 *cqe;
+		u16 wqe_counter;
+		bool last_wqe;
+
+		cqe = mlx5e_get_cqe(cq);
+		if (!cqe)
+			break;
+
+		mlx5_cqwq_pop(&cq->wq);
+
+		wqe_counter = be16_to_cpu(cqe->wqe_counter);
+
+		do {
+			struct mlx5e_dma_info *di;
+			u16 ci;
+
+			last_wqe = (sqcc == wqe_counter);
+
+			ci = sqcc & sq->wq.sz_m1;
+			di = &sq->db.di[ci];
+
+			sqcc++;
+			/* Recycle RX page */
+			mlx5e_page_release(rq, di, true);
+		} while (!last_wqe);
+	}
+
+	mlx5_cqwq_update_db_record(&cq->wq);
+
+	/* ensure cq space is freed before enabling more cqes */
+	wmb();
+
+	sq->cc = sqcc;
+	return (i == MLX5E_TX_CQ_POLL_BUDGET);
+}
+
+void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq)
+{
+	struct mlx5e_rq *rq = container_of(sq, struct mlx5e_rq, xdpsq);
+	struct mlx5e_dma_info *di;
+	u16 ci;
+
+	while (sq->cc != sq->pc) {
+		ci = sq->cc & sq->wq.sz_m1;
+		di = &sq->db.di[ci];
+		sq->cc++;
+
+		mlx5e_page_release(rq, di, false);
+	}
+}

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

@@ -38,29 +38,6 @@
 #define MLX5E_SQ_STOP_ROOM (MLX5_SEND_WQE_MAX_WQEBBS +\
 			    MLX5E_SQ_NOPS_ROOM)
 
-void mlx5e_send_nop(struct mlx5e_sq *sq, bool notify_hw)
-{
-	struct mlx5_wq_cyc                *wq  = &sq->wq;
-
-	u16 pi = sq->pc & wq->sz_m1;
-	struct mlx5e_tx_wqe              *wqe  = mlx5_wq_cyc_get_wqe(wq, pi);
-
-	struct mlx5_wqe_ctrl_seg         *cseg = &wqe->ctrl;
-
-	memset(cseg, 0, sizeof(*cseg));
-
-	cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_NOP);
-	cseg->qpn_ds           = cpu_to_be32((sq->sqn << 8) | 0x01);
-
-	sq->pc++;
-	sq->stats.nop++;
-
-	if (notify_hw) {
-		cseg->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE;
-		mlx5e_tx_notify_hw(sq, &wqe->ctrl, 0);
-	}
-}
-
 static inline void mlx5e_tx_dma_unmap(struct device *pdev,
 				      struct mlx5e_sq_dma *dma)
 {
@@ -76,25 +53,25 @@ static inline void mlx5e_tx_dma_unmap(struct device *pdev,
 	}
 }
 
-static inline void mlx5e_dma_push(struct mlx5e_sq *sq,
+static inline void mlx5e_dma_push(struct mlx5e_txqsq *sq,
 				  dma_addr_t addr,
 				  u32 size,
 				  enum mlx5e_dma_map_type map_type)
 {
 	u32 i = sq->dma_fifo_pc & sq->dma_fifo_mask;
 
-	sq->db.txq.dma_fifo[i].addr = addr;
-	sq->db.txq.dma_fifo[i].size = size;
-	sq->db.txq.dma_fifo[i].type = map_type;
+	sq->db.dma_fifo[i].addr = addr;
+	sq->db.dma_fifo[i].size = size;
+	sq->db.dma_fifo[i].type = map_type;
 	sq->dma_fifo_pc++;
 }
 
-static inline struct mlx5e_sq_dma *mlx5e_dma_get(struct mlx5e_sq *sq, u32 i)
+static inline struct mlx5e_sq_dma *mlx5e_dma_get(struct mlx5e_txqsq *sq, u32 i)
 {
-	return &sq->db.txq.dma_fifo[i & sq->dma_fifo_mask];
+	return &sq->db.dma_fifo[i & sq->dma_fifo_mask];
 }
 
-static void mlx5e_dma_unmap_wqe_err(struct mlx5e_sq *sq, u8 num_dma)
+static void mlx5e_dma_unmap_wqe_err(struct mlx5e_txqsq *sq, u8 num_dma)
 {
 	int i;
 
@@ -175,25 +152,6 @@ static inline unsigned int mlx5e_calc_min_inline(enum mlx5_inline_modes mode,
 	}
 }
 
-static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq,
-					    struct sk_buff *skb, bool bf)
-{
-	/* Some NIC TX decisions, e.g loopback, are based on the packet
-	 * headers and occur before the data gather.
-	 * Therefore these headers must be copied into the WQE
-	 */
-	if (bf) {
-		u16 ihs = skb_headlen(skb);
-
-		if (skb_vlan_tag_present(skb))
-			ihs += VLAN_HLEN;
-
-		if (ihs <= sq->max_inline)
-			return skb_headlen(skb);
-	}
-	return mlx5e_calc_min_inline(sq->min_inline_mode, skb);
-}
-
 static inline void mlx5e_tx_skb_pull_inline(unsigned char **skb_data,
 					    unsigned int *skb_len,
 					    unsigned int len)
@@ -218,13 +176,13 @@ static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs,
 	mlx5e_tx_skb_pull_inline(skb_data, skb_len, cpy2_sz);
 }
 
-static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
+static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb)
 {
 	struct mlx5_wq_cyc       *wq   = &sq->wq;
 
 	u16 pi = sq->pc & wq->sz_m1;
 	struct mlx5e_tx_wqe      *wqe  = mlx5_wq_cyc_get_wqe(wq, pi);
-	struct mlx5e_tx_wqe_info *wi   = &sq->db.txq.wqe_info[pi];
+	struct mlx5e_tx_wqe_info *wi   = &sq->db.wqe_info[pi];
 
 	struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl;
 	struct mlx5_wqe_eth_seg  *eseg = &wqe->eth;
@@ -235,7 +193,6 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 	u8  opcode = MLX5_OPCODE_SEND;
 	dma_addr_t dma_addr = 0;
 	unsigned int num_bytes;
-	bool bf = false;
 	u16 headlen;
 	u16 ds_cnt;
 	u16 ihs;
@@ -255,11 +212,6 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 	} else
 		sq->stats.csum_none++;
 
-	if (sq->cc != sq->prev_cc) {
-		sq->prev_cc = sq->cc;
-		sq->bf_budget = (sq->cc == sq->pc) ? MLX5E_SQ_BF_BUDGET : 0;
-	}
-
 	if (skb_is_gso(skb)) {
 		eseg->mss    = cpu_to_be16(skb_shinfo(skb)->gso_size);
 		opcode       = MLX5_OPCODE_LSO;
@@ -277,10 +229,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 		sq->stats.packets += skb_shinfo(skb)->gso_segs;
 		num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs;
 	} else {
-		bf = sq->bf_budget &&
-		     !skb->xmit_more &&
-		     !skb_shinfo(skb)->nr_frags;
-		ihs = mlx5e_get_inline_hdr_size(sq, skb, bf);
+		ihs = mlx5e_calc_min_inline(sq->min_inline_mode, skb);
 		sq->stats.packets++;
 		num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN);
 	}
@@ -349,7 +298,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 	cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | opcode);
 	cseg->qpn_ds           = cpu_to_be32((sq->sqn << 8) | ds_cnt);
 
-	sq->db.txq.skb[pi] = skb;
+	sq->db.skb[pi] = skb;
 
 	wi->num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS);
 	sq->pc += wi->num_wqebbs;
@@ -359,31 +308,23 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP))
 		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 
-	if (unlikely(!mlx5e_sq_has_room_for(sq, MLX5E_SQ_STOP_ROOM))) {
+	if (unlikely(!mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc,
+					     MLX5E_SQ_STOP_ROOM))) {
 		netif_tx_stop_queue(sq->txq);
 		sq->stats.stopped++;
 	}
 
 	sq->stats.xmit_more += skb->xmit_more;
-	if (!skb->xmit_more || netif_xmit_stopped(sq->txq)) {
-		int bf_sz = 0;
-
-		if (bf && test_bit(MLX5E_SQ_STATE_BF_ENABLE, &sq->state))
-			bf_sz = wi->num_wqebbs << 3;
-
-		cseg->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE;
-		mlx5e_tx_notify_hw(sq, &wqe->ctrl, bf_sz);
-	}
+	if (!skb->xmit_more || netif_xmit_stopped(sq->txq))
+		mlx5e_notify_hw(wq, sq->pc, sq->uar_map, cseg);
 
 	/* fill sq edge with nops to avoid wqe wrap around */
 	while ((pi = (sq->pc & wq->sz_m1)) > sq->edge) {
-		sq->db.txq.skb[pi] = NULL;
-		mlx5e_send_nop(sq, false);
+		sq->db.skb[pi] = NULL;
+		mlx5e_post_nop(&sq->wq, sq->sqn, &sq->pc);
+		sq->stats.nop++;
 	}
 
-	if (bf)
-		sq->bf_budget--;
-
 	return NETDEV_TX_OK;
 
 dma_unmap_wqe_err:
@@ -398,21 +339,21 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct mlx5e_priv *priv = netdev_priv(dev);
-	struct mlx5e_sq *sq = priv->txq_to_sq_map[skb_get_queue_mapping(skb)];
+	struct mlx5e_txqsq *sq = priv->txq_to_sq_map[skb_get_queue_mapping(skb)];
 
 	return mlx5e_sq_xmit(sq, skb);
 }
 
 bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget)
 {
-	struct mlx5e_sq *sq;
+	struct mlx5e_txqsq *sq;
 	u32 dma_fifo_cc;
 	u32 nbytes;
 	u16 npkts;
 	u16 sqcc;
 	int i;
 
-	sq = container_of(cq, struct mlx5e_sq, cq);
+	sq = container_of(cq, struct mlx5e_txqsq, cq);
 
 	if (unlikely(!test_bit(MLX5E_SQ_STATE_ENABLED, &sq->state)))
 		return false;
@@ -450,8 +391,8 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget)
 			last_wqe = (sqcc == wqe_counter);
 
 			ci = sqcc & sq->wq.sz_m1;
-			skb = sq->db.txq.skb[ci];
-			wi = &sq->db.txq.wqe_info[ci];
+			skb = sq->db.skb[ci];
+			wi = &sq->db.wqe_info[ci];
 
 			if (unlikely(!skb)) { /* nop */
 				sqcc++;
@@ -492,7 +433,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget)
 	netdev_tx_completed_queue(sq->txq, npkts, nbytes);
 
 	if (netif_tx_queue_stopped(sq->txq) &&
-	    mlx5e_sq_has_room_for(sq, MLX5E_SQ_STOP_ROOM)) {
+	    mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, MLX5E_SQ_STOP_ROOM)) {
 		netif_tx_wake_queue(sq->txq);
 		sq->stats.wake++;
 	}
@@ -500,7 +441,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget)
 	return (i == MLX5E_TX_CQ_POLL_BUDGET);
 }
 
-static void mlx5e_free_txq_sq_descs(struct mlx5e_sq *sq)
+void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq)
 {
 	struct mlx5e_tx_wqe_info *wi;
 	struct sk_buff *skb;
@@ -509,8 +450,8 @@ static void mlx5e_free_txq_sq_descs(struct mlx5e_sq *sq)
 
 	while (sq->cc != sq->pc) {
 		ci = sq->cc & sq->wq.sz_m1;
-		skb = sq->db.txq.skb[ci];
-		wi = &sq->db.txq.wqe_info[ci];
+		skb = sq->db.skb[ci];
+		wi = &sq->db.wqe_info[ci];
 
 		if (!skb) { /* nop */
 			sq->cc++;
@@ -528,37 +469,3 @@ static void mlx5e_free_txq_sq_descs(struct mlx5e_sq *sq)
 		sq->cc += wi->num_wqebbs;
 	}
 }
-
-static void mlx5e_free_xdp_sq_descs(struct mlx5e_sq *sq)
-{
-	struct mlx5e_sq_wqe_info *wi;
-	struct mlx5e_dma_info *di;
-	u16 ci;
-
-	while (sq->cc != sq->pc) {
-		ci = sq->cc & sq->wq.sz_m1;
-		di = &sq->db.xdp.di[ci];
-		wi = &sq->db.xdp.wqe_info[ci];
-
-		if (wi->opcode == MLX5_OPCODE_NOP) {
-			sq->cc++;
-			continue;
-		}
-
-		sq->cc += wi->num_wqebbs;
-
-		mlx5e_page_release(&sq->channel->rq, di, false);
-	}
-}
-
-void mlx5e_free_sq_descs(struct mlx5e_sq *sq)
-{
-	switch (sq->type) {
-	case MLX5E_SQ_TXQ:
-		mlx5e_free_txq_sq_descs(sq);
-		break;
-	case MLX5E_SQ_XDP:
-		mlx5e_free_xdp_sq_descs(sq);
-		break;
-	}
-}

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

@@ -44,14 +44,14 @@ struct mlx5_cqe64 *mlx5e_get_cqe(struct mlx5e_cq *cq)
 		return NULL;
 
 	/* ensure cqe content is read after cqe ownership bit */
-	rmb();
+	dma_rmb();
 
 	return cqe;
 }
 
 static void mlx5e_poll_ico_cq(struct mlx5e_cq *cq)
 {
-	struct mlx5e_sq *sq = container_of(cq, struct mlx5e_sq, cq);
+	struct mlx5e_icosq *sq = container_of(cq, struct mlx5e_icosq, cq);
 	struct mlx5_wq_cyc *wq;
 	struct mlx5_cqe64 *cqe;
 	u16 sqcc;
@@ -105,66 +105,6 @@ static void mlx5e_poll_ico_cq(struct mlx5e_cq *cq)
 	sq->cc = sqcc;
 }
 
-static inline bool mlx5e_poll_xdp_tx_cq(struct mlx5e_cq *cq)
-{
-	struct mlx5e_sq *sq;
-	u16 sqcc;
-	int i;
-
-	sq = container_of(cq, struct mlx5e_sq, cq);
-
-	if (unlikely(!test_bit(MLX5E_SQ_STATE_ENABLED, &sq->state)))
-		return false;
-
-	/* sq->cc must be updated only after mlx5_cqwq_update_db_record(),
-	 * otherwise a cq overrun may occur
-	 */
-	sqcc = sq->cc;
-
-	for (i = 0; i < MLX5E_TX_CQ_POLL_BUDGET; i++) {
-		struct mlx5_cqe64 *cqe;
-		u16 wqe_counter;
-		bool last_wqe;
-
-		cqe = mlx5e_get_cqe(cq);
-		if (!cqe)
-			break;
-
-		mlx5_cqwq_pop(&cq->wq);
-
-		wqe_counter = be16_to_cpu(cqe->wqe_counter);
-
-		do {
-			struct mlx5e_sq_wqe_info *wi;
-			struct mlx5e_dma_info *di;
-			u16 ci;
-
-			last_wqe = (sqcc == wqe_counter);
-
-			ci = sqcc & sq->wq.sz_m1;
-			di = &sq->db.xdp.di[ci];
-			wi = &sq->db.xdp.wqe_info[ci];
-
-			if (unlikely(wi->opcode == MLX5_OPCODE_NOP)) {
-				sqcc++;
-				continue;
-			}
-
-			sqcc += wi->num_wqebbs;
-			/* Recycle RX page */
-			mlx5e_page_release(&sq->channel->rq, di, true);
-		} while (!last_wqe);
-	}
-
-	mlx5_cqwq_update_db_record(&cq->wq);
-
-	/* ensure cq space is freed before enabling more cqes */
-	wmb();
-
-	sq->cc = sqcc;
-	return (i == MLX5E_TX_CQ_POLL_BUDGET);
-}
-
 int mlx5e_napi_poll(struct napi_struct *napi, int budget)
 {
 	struct mlx5e_channel *c = container_of(napi, struct mlx5e_channel,
@@ -178,12 +118,12 @@ 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);
 
+	if (c->xdp)
+		busy |= mlx5e_poll_xdpsq_cq(&c->rq.xdpsq.cq);
+
 	work_done = mlx5e_poll_rx_cq(&c->rq.cq, budget);
 	busy |= work_done == budget;
 
-	if (c->xdp)
-		busy |= mlx5e_poll_xdp_tx_cq(&c->xdp_sq.cq);
-
 	mlx5e_poll_ico_cq(&c->icosq.cq);
 
 	busy |= mlx5e_post_rx_wqes(&c->rq);

include/linux/mlx5/driver.h

@@ -728,6 +728,7 @@ struct mlx5e_resources {
 	u32                        pdn;
 	struct mlx5_td             td;
 	struct mlx5_core_mkey      mkey;
+	struct mlx5_sq_bfreg       bfreg;
 };
 
 struct mlx5_core_dev {