bnxt_en: Support TX coalesced completion on 5760X chips

TX coalesced completions are supported on newer chips to provide
one TX completion record for multiple TX packets up to the
sq_cons_idx in the completion record.  This method saves PCIe
bandwidth by reducing the number of TX completions.

Only very minor changes are now required to support this mode
with the new framework that handles TX completions based on
the consumer indices.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Link: https://lore.kernel.org/r/20231212005122.2401-6-michael.chan@broadcom.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

drivers/net/ethernet/broadcom/bnxt/bnxt.c

@@ -2785,14 +2785,18 @@ static int __bnxt_poll_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 		 */
 		dma_rmb();
 		cmp_type = TX_CMP_TYPE(txcmp);
-		if (cmp_type == CMP_TYPE_TX_L2_CMP) {
+		if (cmp_type == CMP_TYPE_TX_L2_CMP ||
+		    cmp_type == CMP_TYPE_TX_L2_COAL_CMP) {
 			u32 opaque = txcmp->tx_cmp_opaque;
 			struct bnxt_tx_ring_info *txr;
 			u16 tx_freed;
 
 			txr = bnapi->tx_ring[TX_OPAQUE_RING(opaque)];
 			event |= BNXT_TX_CMP_EVENT;
-			txr->tx_hw_cons = TX_OPAQUE_PROD(bp, opaque);
+			if (cmp_type == CMP_TYPE_TX_L2_COAL_CMP)
+				txr->tx_hw_cons = TX_CMP_SQ_CONS_IDX(txcmp);
+			else
+				txr->tx_hw_cons = TX_OPAQUE_PROD(bp, opaque);
 			tx_freed = (txr->tx_hw_cons - txr->tx_cons) &
 				   bp->tx_ring_mask;
 			/* return full budget so NAPI will complete. */
@@ -6068,6 +6072,9 @@ static int hwrm_ring_alloc_send_msg(struct bnxt *bp,
 		req->length = cpu_to_le32(bp->tx_ring_mask + 1);
 		req->stat_ctx_id = cpu_to_le32(grp_info->fw_stats_ctx);
 		req->queue_id = cpu_to_le16(ring->queue_id);
+		if (bp->flags & BNXT_FLAG_TX_COAL_CMPL)
+			req->cmpl_coal_cnt =
+				RING_ALLOC_REQ_CMPL_COAL_CNT_COAL_64;
 		break;
 	}
 	case HWRM_RING_ALLOC_RX:
@@ -8279,6 +8286,8 @@ static int __bnxt_hwrm_func_qcaps(struct bnxt *bp)
 		bp->fw_cap |= BNXT_FW_CAP_LIVEPATCH;
 	if (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_BS_V2_SUPPORTED)
 		bp->fw_cap |= BNXT_FW_CAP_BACKING_STORE_V2;
+	if (flags_ext & FUNC_QCAPS_RESP_FLAGS_EXT_TX_COAL_CMPL_CAP)
+		bp->flags |= BNXT_FLAG_TX_COAL_CMPL;
 
 	flags_ext2 = le32_to_cpu(resp->flags_ext2);
 	if (flags_ext2 & FUNC_QCAPS_RESP_FLAGS_EXT2_RX_ALL_PKTS_TIMESTAMPS_SUPPORTED)

drivers/net/ethernet/broadcom/bnxt/bnxt.h

@@ -2047,6 +2047,7 @@ struct bnxt {
 	#define BNXT_FLAG_CHIP_NITRO_A0	0x1000000
 	#define BNXT_FLAG_DIM		0x2000000
 	#define BNXT_FLAG_ROCE_MIRROR_CAP	0x4000000
+	#define BNXT_FLAG_TX_COAL_CMPL	0x8000000
 	#define BNXT_FLAG_PORT_STATS_EXT	0x10000000
 
 	#define BNXT_FLAG_ALL_CONFIG_FEATS (BNXT_FLAG_TPA |		\