octeontx2-pf: Fix SQE threshold checking

Current way of checking available SQE count which is based on
HW updated SQB count could result in driver submitting an SQE
even before CQE for the previously transmitted SQE at the same
index is processed in NAPI resulting losing SKB pointers,
hence a leak. Fix this by checking a consumer index which
is updated once CQE is processed.

Fixes: 3ca6c4c882a7 ("octeontx2-pf: Add packet transmission support")
Signed-off-by: Ratheesh Kannoth <[e-mail]>
Reviewed-by: Sunil Kovvuri Goutham <[e-mail]>
Link: https://lore.kernel.org/r/[e-mail]
Signed-off-by: Paolo Abeni <[e-mail]>

drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c

···
 	}
 
 	sq->head = 0;
+	sq->cons_head = 0;
 	sq->sqe_per_sqb = (pfvf->hw.sqb_size / sq->sqe_size) - 1;
 	sq->num_sqbs = (qset->sqe_cnt + sq->sqe_per_sqb) / sq->sqe_per_sqb;
 	/* Set SQE threshold to 10% of total SQEs */

drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c

···
 				struct otx2_cq_queue *cq, int budget)
 {
 	int tx_pkts = 0, tx_bytes = 0, qidx;
+	struct otx2_snd_queue *sq;
 	struct nix_cqe_tx_s *cqe;
 	int processed_cqe = 0;
 
···
 		return 0;
 
 process_cqe:
+	qidx = cq->cq_idx - pfvf->hw.rx_queues;
+	sq = &pfvf->qset.sq[qidx];
+
 	while (likely(processed_cqe < budget) && cq->pend_cqe) {
 		cqe = (struct nix_cqe_tx_s *)otx2_get_next_cqe(cq);
 		if (unlikely(!cqe)) {
···
 				return 0;
 			break;
 		}
+
 		if (cq->cq_type == CQ_XDP) {
-			qidx = cq->cq_idx - pfvf->hw.rx_queues;
-			otx2_xdp_snd_pkt_handler(pfvf, &pfvf->qset.sq[qidx],
-						 cqe);
+			otx2_xdp_snd_pkt_handler(pfvf, sq, cqe);
 		} else {
-			otx2_snd_pkt_handler(pfvf, cq,
-					     &pfvf->qset.sq[cq->cint_idx],
-					     cqe, budget, &tx_pkts, &tx_bytes);
+			otx2_snd_pkt_handler(pfvf, cq, sq, cqe, budget,
+					     &tx_pkts, &tx_bytes);
 		}
+
 		cqe->hdr.cqe_type = NIX_XQE_TYPE_INVALID;
 		processed_cqe++;
 		cq->pend_cqe--;
+
+		sq->cons_head++;
+		sq->cons_head &= (sq->sqe_cnt - 1);
 	}
 
 	/* Free CQEs to HW */
···
 {
 	struct netdev_queue *txq = netdev_get_tx_queue(netdev, qidx);
 	struct otx2_nic *pfvf = netdev_priv(netdev);
-	int offset, num_segs, free_sqe;
+	int offset, num_segs, free_desc;
 	struct nix_sqe_hdr_s *sqe_hdr;
 
-	/* Check if there is room for new SQE.
-	 * 'Num of SQBs freed to SQ's pool - SQ's Aura count'
-	 * will give free SQE count.
+	/* Check if there is enough room between producer
+	 * and consumer index.
 	 */
-	free_sqe = (sq->num_sqbs - *sq->aura_fc_addr) * sq->sqe_per_sqb;
+	free_desc = (sq->cons_head - sq->head - 1 + sq->sqe_cnt) & (sq->sqe_cnt - 1);
+	if (free_desc < sq->sqe_thresh)
+		return false;
 
-	if (free_sqe < sq->sqe_thresh ||
-	    free_sqe < otx2_get_sqe_count(pfvf, skb))
+	if (free_desc < otx2_get_sqe_count(pfvf, skb))
 		return false;
 
 	num_segs = skb_shinfo(skb)->nr_frags + 1;

drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h

···
 struct otx2_snd_queue {
 	u8			aura_id;
 	u16			head;
+	u16			cons_head;
 	u16			sqe_size;
 	u32			sqe_cnt;
 	u16			num_sqbs;