net/bnxt: fix Rx buffer posting

[ upstream commit bb4f70d ]

Remove early buffer posting logic from burst receive loop to address
several issues:
   - Posting receive descriptors without first posting completion
     entries risks overflowing the completion queue.
   - Posting receive descriptors without updating rx_raw_prod
     creates the possibility that the receive descriptor doorbell
     can be written twice with the same value.
   - Having this logic in the inner descriptor processing loop
     can impact performance.

Fixes: 637e34b ("net/bnxt: optimize Rx processing")
Fixes: 0406784 ("net/bnxt: reduce CQ queue size without aggregation ring")

Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>

drivers/net/bnxt/bnxt_rxr.c

@@ -654,9 +654,6 @@ uint16_t bnxt_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
 		raw_cons = NEXT_RAW_CMP(raw_cons);
 		if (nb_rx_pkts == nb_pkts || evt)
 			break;
-		/* Post some Rx buf early in case of larger burst processing */
-		if (nb_rx_pkts == BNXT_RX_POST_THRESH)
-			bnxt_db_write(&rxr->rx_db, rxr->rx_prod);
 	}
 
 	cpr->cp_raw_cons = raw_cons;

drivers/net/bnxt/bnxt_rxr.h

@@ -169,8 +169,6 @@ static inline uint16_t bnxt_tpa_start_agg_id(struct bnxt *bp,
 	 ((rxcmp1)->errors_v2 & \
 	  rte_cpu_to_le_32(RX_PKT_CMPL_ERRORS_L4_CS_ERROR))
 
-#define BNXT_RX_POST_THRESH	32
-
 enum pkt_hash_types {
 	PKT_HASH_TYPE_NONE,	/* Undefined type */
 	PKT_HASH_TYPE_L2,	/* Input: src_MAC, dest_MAC */