net/ena: do not set Tx L4 offloads in Rx path
authorRafal Kozik <rk@semihalf.com>
Thu, 25 Jan 2018 15:27:43 +0000 (16:27 +0100)
committerFerruh Yigit <ferruh.yigit@intel.com>
Mon, 29 Jan 2018 09:04:28 +0000 (10:04 +0100)
Information about received packet type detected by NIC should be
stored in packet_type field of rte_mbuf. TX L4 offload flags should
not be set in RX path. Only fields that could be set in of_flags
during packet receiving are information if L4 and L3 checksum is
correct.

Fixes: 1173fca25af9 ("ena: add polling-mode driver")
Cc: stable@dpdk.org
Reported-by: Matthew Smith <mgsmith@netgate.com>
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
drivers/net/ena/ena_ethdev.c

index 83e0ae2..1e2af80 100644 (file)
@@ -275,16 +275,17 @@ static inline void ena_rx_mbuf_prepare(struct rte_mbuf *mbuf,
                                       struct ena_com_rx_ctx *ena_rx_ctx)
 {
        uint64_t ol_flags = 0;
+       uint32_t packet_type = 0;
 
        if (ena_rx_ctx->l4_proto == ENA_ETH_IO_L4_PROTO_TCP)
-               ol_flags |= PKT_TX_TCP_CKSUM;
+               packet_type |= RTE_PTYPE_L4_TCP;
        else if (ena_rx_ctx->l4_proto == ENA_ETH_IO_L4_PROTO_UDP)
-               ol_flags |= PKT_TX_UDP_CKSUM;
+               packet_type |= RTE_PTYPE_L4_UDP;
 
        if (ena_rx_ctx->l3_proto == ENA_ETH_IO_L3_PROTO_IPV4)
-               ol_flags |= PKT_TX_IPV4;
+               packet_type |= RTE_PTYPE_L3_IPV4;
        else if (ena_rx_ctx->l3_proto == ENA_ETH_IO_L3_PROTO_IPV6)
-               ol_flags |= PKT_TX_IPV6;
+               packet_type |= RTE_PTYPE_L3_IPV6;
 
        if (unlikely(ena_rx_ctx->l4_csum_err))
                ol_flags |= PKT_RX_L4_CKSUM_BAD;
@@ -292,6 +293,7 @@ static inline void ena_rx_mbuf_prepare(struct rte_mbuf *mbuf,
                ol_flags |= PKT_RX_IP_CKSUM_BAD;
 
        mbuf->ol_flags = ol_flags;
+       mbuf->packet_type = packet_type;
 }
 
 static inline void ena_tx_mbuf_prepare(struct rte_mbuf *mbuf,