nfp: correct desc type when header dma len is 4096

When there's only one buffer to dma and its length is 4096, then
only one data descriptor is needed to carry it according to current
descriptor definition. So the descriptor type should be `simple`
instead of `gather`, the latter requires more than one descriptor,
otherwise it'll be dropped by application firmware.

Fixes: c10d12e3 ("nfp: add support for NFDK data path")
Fixes: d9d95049 ("nfp: nfdk: implement xdp tx path for NFDK")
Signed-off-by: Yinjun Zhang <yinjun.zhang@corigine.com>
Reviewed-by: Richard Donkin <richard.donkin@corigine.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@corigine.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Link: https://lore.kernel.org/r/20221202134646.311108-1-simon.horman@corigine.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

drivers/net/ethernet/netronome/nfp/nfdk/dp.c

@@ -282,7 +282,7 @@ netdev_tx_t nfp_nfdk_tx(struct sk_buff *skb, struct net_device *netdev)
 	dma_len = skb_headlen(skb);
 	if (skb_is_gso(skb))
 		type = NFDK_DESC_TX_TYPE_TSO;
-	else if (!nr_frags && dma_len < NFDK_TX_MAX_DATA_PER_HEAD)
+	else if (!nr_frags && dma_len <= NFDK_TX_MAX_DATA_PER_HEAD)
 		type = NFDK_DESC_TX_TYPE_SIMPLE;
 	else
 		type = NFDK_DESC_TX_TYPE_GATHER;
@@ -927,7 +927,7 @@ nfp_nfdk_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
 	dma_len = pkt_len;
 	dma_addr = rxbuf->dma_addr + dma_off;
 
-	if (dma_len < NFDK_TX_MAX_DATA_PER_HEAD)
+	if (dma_len <= NFDK_TX_MAX_DATA_PER_HEAD)
 		type = NFDK_DESC_TX_TYPE_SIMPLE;
 	else
 		type = NFDK_DESC_TX_TYPE_GATHER;
@@ -1325,7 +1325,7 @@ nfp_nfdk_ctrl_tx_one(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
 	txbuf = &tx_ring->ktxbufs[wr_idx];
 
 	dma_len = skb_headlen(skb);
-	if (dma_len < NFDK_TX_MAX_DATA_PER_HEAD)
+	if (dma_len <= NFDK_TX_MAX_DATA_PER_HEAD)
 		type = NFDK_DESC_TX_TYPE_SIMPLE;
 	else
 		type = NFDK_DESC_TX_TYPE_GATHER;