1 files changed, 56 insertions, 45 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index f049e0ac308a..c9d308e91965 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -31,7 +31,7 @@
  */
 
 #include <linux/bpf_trace.h>
-#include <net/xdp_sock.h>
+#include <net/xdp_sock_drv.h>
 #include "en/xdp.h"
 #include "en/params.h"
 
@@ -64,14 +64,14 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 	struct xdp_frame *xdpf;
 	dma_addr_t dma_addr;
 
-	xdpf = convert_to_xdp_frame(xdp);
+	xdpf = xdp_convert_buff_to_frame(xdp);
 	if (unlikely(!xdpf))
 		return false;
 
 	xdptxd.data = xdpf->data;
 	xdptxd.len  = xdpf->len;
 
-	if (xdp->rxq->mem.type == MEM_TYPE_ZERO_COPY) {
+	if (xdp->rxq->mem.type == MEM_TYPE_XSK_BUFF_POOL) {
 		/* The xdp_buff was in the UMEM and was copied into a newly
 		 * allocated page. The UMEM page was returned via the ZCA, and
 		 * this new page has to be mapped at this point and has to be
@@ -97,10 +97,10 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 		xdpi.frame.xdpf     = xdpf;
 		xdpi.frame.dma_addr = dma_addr;
 	} else {
-		/* Driver assumes that convert_to_xdp_frame returns an xdp_frame
-		 * that points to the same memory region as the original
-		 * xdp_buff. It allows to map the memory only once and to use
-		 * the DMA_BIDIRECTIONAL mode.
+		/* Driver assumes that xdp_convert_buff_to_frame returns
+		 * an xdp_frame that points to the same memory region as
+		 * the original xdp_buff. It allows to map the memory only
+		 * once and to use the DMA_BIDIRECTIONAL mode.
 		 */
 
 		xdpi.mode = MLX5E_XDP_XMIT_MODE_PAGE;
@@ -119,49 +119,33 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 
 /* returns true if packet was consumed by xdp */
 bool mlx5e_xdp_handle(struct mlx5e_rq *rq, struct mlx5e_dma_info *di,
-		      void *va, u16 *rx_headroom, u32 *len, bool xsk)
+		      u32 *len, struct xdp_buff *xdp)
 {
 	struct bpf_prog *prog = READ_ONCE(rq->xdp_prog);
-	struct xdp_umem *umem = rq->umem;
-	struct xdp_buff xdp;
 	u32 act;
 	int err;
 
 	if (!prog)
 		return false;
 
-	xdp.data = va + *rx_headroom;
-	xdp_set_data_meta_invalid(&xdp);
-	xdp.data_end = xdp.data + *len;
-	xdp.data_hard_start = va;
-	if (xsk)
-		xdp.handle = di->xsk.handle;
-	xdp.rxq = &rq->xdp_rxq;
-
-	act = bpf_prog_run_xdp(prog, &xdp);
-	if (xsk) {
-		u64 off = xdp.data - xdp.data_hard_start;
-
-		xdp.handle = xsk_umem_adjust_offset(umem, xdp.handle, off);
-	}
+	act = bpf_prog_run_xdp(prog, xdp);
 	switch (act) {
 	case XDP_PASS:
-		*rx_headroom = xdp.data - xdp.data_hard_start;
-		*len = xdp.data_end - xdp.data;
+		*len = xdp->data_end - xdp->data;
 		return false;
 	case XDP_TX:
-		if (unlikely(!mlx5e_xmit_xdp_buff(rq->xdpsq, rq, di, &xdp)))
+		if (unlikely(!mlx5e_xmit_xdp_buff(rq->xdpsq, rq, di, xdp)))
 			goto xdp_abort;
 		__set_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags); /* non-atomic */
 		return true;
 	case XDP_REDIRECT:
 		/* When XDP enabled then page-refcnt==1 here */
-		err = xdp_do_redirect(rq->netdev, &xdp, prog);
+		err = xdp_do_redirect(rq->netdev, xdp, prog);
 		if (unlikely(err))
 			goto xdp_abort;
 		__set_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags);
 		__set_bit(MLX5E_RQ_FLAG_XDP_REDIRECT, rq->flags);
-		if (!xsk)
+		if (xdp->rxq->mem.type != MEM_TYPE_XSK_BUFF_POOL)
 			mlx5e_page_dma_unmap(rq, di);
 		rq->stats->xdp_redirect++;
 		return true;
@@ -178,20 +162,43 @@ xdp_abort:
 	}
 }
 
-static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq)
+static u16 mlx5e_xdpsq_get_next_pi(struct mlx5e_xdpsq *sq, u16 size)
 {
-	struct mlx5e_xdp_mpwqe *session = &sq->mpwqe;
-	struct mlx5e_xdpsq_stats *stats = sq->stats;
 	struct mlx5_wq_cyc *wq = &sq->wq;
 	u16 pi, contig_wqebbs;
 
 	pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc);
 	contig_wqebbs = mlx5_wq_cyc_get_contig_wqebbs(wq, pi);
+	if (unlikely(contig_wqebbs < size)) {
+		struct mlx5e_xdp_wqe_info *wi, *edge_wi;
+
+		wi = &sq->db.wqe_info[pi];
+		edge_wi = wi + contig_wqebbs;
+
+		/* Fill SQ frag edge with NOPs to avoid WQE wrapping two pages. */
+		for (; wi < edge_wi; wi++) {
+			*wi = (struct mlx5e_xdp_wqe_info) {
+				.num_wqebbs = 1,
+				.num_pkts = 0,
+			};
+			mlx5e_post_nop(wq, sq->sqn, &sq->pc);
+		}
+		sq->stats->nops += contig_wqebbs;
 
-	if (unlikely(contig_wqebbs < MLX5_SEND_WQE_MAX_WQEBBS))
-		mlx5e_fill_xdpsq_frag_edge(sq, wq, pi, contig_wqebbs);
+		pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc);
+	}
+
+	return pi;
+}
 
-	session->wqe = mlx5e_xdpsq_fetch_wqe(sq, &pi);
+static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq)
+{
+	struct mlx5e_xdp_mpwqe *session = &sq->mpwqe;
+	struct mlx5e_xdpsq_stats *stats = sq->stats;
+	u16 pi;
+
+	pi = mlx5e_xdpsq_get_next_pi(sq, MLX5_SEND_WQE_MAX_WQEBBS);
+	session->wqe = MLX5E_TX_FETCH_WQE(sq, pi);
 
 	prefetchw(session->wqe->data);
 	session->ds_count  = MLX5E_XDP_TX_EMPTY_DS_COUNT;
@@ -233,8 +240,10 @@ enum {
 static int mlx5e_xmit_xdp_frame_check_mpwqe(struct mlx5e_xdpsq *sq)
 {
 	if (unlikely(!sq->mpwqe.wqe)) {
+		const u16 stop_room = mlx5e_stop_room_for_wqe(MLX5_SEND_WQE_MAX_WQEBBS);
+
 		if (unlikely(!mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc,
-						     MLX5E_XDPSQ_STOP_ROOM))) {
+						     stop_room))) {
 			/* SQ is full, ring doorbell */
 			mlx5e_xmit_xdp_doorbell(sq);
 			sq->stats->full++;
@@ -408,22 +417,15 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq)
 
 	i = 0;
 	do {
-		u16 wqe_counter;
+		struct mlx5e_xdp_wqe_info *wi;
+		u16 wqe_counter, ci;
 		bool last_wqe;
 
 		mlx5_cqwq_pop(&cq->wq);
 
 		wqe_counter = be16_to_cpu(cqe->wqe_counter);
 
-		if (unlikely(get_cqe_opcode(cqe) != MLX5_CQE_REQ))
-			netdev_WARN_ONCE(sq->channel->netdev,
-					 "Bad OP in XDPSQ CQE: 0x%x\n",
-					 get_cqe_opcode(cqe));
-
 		do {
-			struct mlx5e_xdp_wqe_info *wi;
-			u16 ci;
-
 			last_wqe = (sqcc == wqe_counter);
 			ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc);
 			wi = &sq->db.wqe_info[ci];
@@ -432,6 +434,15 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq)
 
 			mlx5e_free_xdpsq_desc(sq, wi, &xsk_frames, true);
 		} while (!last_wqe);
+
+		if (unlikely(get_cqe_opcode(cqe) != MLX5_CQE_REQ)) {
+			netdev_WARN_ONCE(sq->channel->netdev,
+					 "Bad OP in XDPSQ CQE: 0x%x\n",
+					 get_cqe_opcode(cqe));
+			mlx5e_dump_error_cqe(&sq->cq, sq->sqn,
+					     (struct mlx5_err_cqe *)cqe);
+			mlx5_wq_cyc_wqe_dump(&sq->wq, ci, wi->num_wqebbs);
+		}
 	} while ((++i < MLX5E_TX_CQ_POLL_BUDGET) && (cqe = mlx5_cqwq_get_cqe(&cq->wq)));
 
 	if (xsk_frames)