net/mlx5: fix LRO TCP checksum

[ upstream commit 776209c4b865ad3e0eaff202c5ff7358b8f5ab57 ]

The variable csum is the sum of three 16 bits integers, the max value
is 0x2FFFD. The corner case of sum of 3 is 0x1FFFF gives the wrong
result: 0x1 + 0xFFFF = 0x10000, the upper 16 bits are not 0.
It must be folded again to ensure that the upper 16 bits are 0.

Fixes: e4c2a16 ("net/mlx5: handle LRO packets in Rx queue")

Signed-off-by: Heng Jiang <jiangheng14@huawei.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>

.mailmap

@@ -482,6 +482,7 @@ Helin Zhang <helin.zhang@intel.com>
 Hemant Agrawal <hemant.agrawal@nxp.com> <hemant@freescale.com>
 Heng Ding <hengx.ding@intel.com>
 Hengjian Zhang <hengjianx.zhang@intel.com>
+Heng Jiang <jiangheng14@huawei.com>
 Heng Wang <heng.wang@ericsson.com>
 Henning Schild <henning.schild@siemens.com>
 Henry Cai <caihe@huawei.com>

drivers/net/mlx5/mlx5_rx.c

@@ -1027,6 +1027,7 @@ mlx5_lro_update_tcp_hdr(struct rte_tcp_hdr *__rte_restrict tcp,
 	tcp->cksum = 0;
 	csum += rte_raw_cksum(tcp, (tcp->data_off >> 4) * 4);
 	csum = ((csum & 0xffff0000) >> 16) + (csum & 0xffff);
+	csum = ((csum & 0xffff0000) >> 16) + (csum & 0xffff);
 	csum = (~csum) & 0xffff;
 	if (csum == 0)
 		csum = 0xffff;