[dpdk-dev] app/testpmd: fix io fwd mode to forward VLAN packets

Message ID 20170629030615.20098-1-ajit.khaparde@broadcom.com (mailing list archive)
State Rejected, archived
Headers

Checks

Context Check Description
ci/Intel-compilation success Compilation OK
ci/checkpatch success coding style OK

Commit Message

Ajit Khaparde June 29, 2017, 3:06 a.m. UTC
  In io fwd mode, while the received VLAN packets are forwarded
for transmit, the ol_flags are not updated appropriately to
indicate presence of VLAN tags in the Tx path. This patch fixes that.

Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
---
 app/test-pmd/iofwd.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
  

Comments

Jingjing Wu June 29, 2017, 5:36 a.m. UTC | #1
> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Ajit Khaparde
> Sent: Thursday, June 29, 2017 11:06 AM
> To: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH] app/testpmd: fix io fwd mode to forward VLAN
> packets
> 
> In io fwd mode, while the received VLAN packets are forwarded for transmit,
> the ol_flags are not updated appropriately to indicate presence of VLAN tags in
> the Tx path. This patch fixes that.
> 
> Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
> ---
>  app/test-pmd/iofwd.c | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/app/test-pmd/iofwd.c b/app/test-pmd/iofwd.c index
> 15cb4a2..997d3c3 100644
> --- a/app/test-pmd/iofwd.c
> +++ b/app/test-pmd/iofwd.c
> @@ -91,6 +91,7 @@ pkt_burst_io_forward(struct fwd_stream *fs)  #ifdef
> RTE_TEST_PMD_RECORD_CORE_CYCLES
>  	start_tsc = rte_rdtsc();
>  #endif
> +	int i;
> 
>  	/*
>  	 * Receive a burst of packets and forward them.
> @@ -104,6 +105,20 @@ pkt_burst_io_forward(struct fwd_stream *fs)  #ifdef
> RTE_TEST_PMD_RECORD_BURST_STATS
>  	fs->rx_burst_stats.pkt_burst_spread[nb_rx]++;
>  #endif
> +
> +	for (i = 0; i < nb_rx; i++) {
> +		uint64_t tx_ol_flags = 0, rx_ol_flags = pkts_burst[i]->ol_flags;
> +
> +		if (rx_ol_flags & PKT_RX_VLAN_PKT)
> +			tx_ol_flags |= PKT_TX_VLAN_PKT;
> +		else if (rx_ol_flags & PKT_RX_QINQ_PKT)
> +			tx_ol_flags |= PKT_TX_QINQ_PKT;
> +
> +		pkts_burst[i]->ol_flags = 0;
> +		pkts_burst[i]->ol_flags = tx_ol_flags;
> +
> +	}
> +

The iofwd engine is an fwd engine what just calls tx_burst and rx_burst,
it is purely fwd without carrying any offload features. 
If you'd like to demonstrate the vlan offload feature, you can other fwd engine
Such as txonly, macfwd......

Thanks
Jingjing
  

Patch

diff --git a/app/test-pmd/iofwd.c b/app/test-pmd/iofwd.c
index 15cb4a2..997d3c3 100644
--- a/app/test-pmd/iofwd.c
+++ b/app/test-pmd/iofwd.c
@@ -91,6 +91,7 @@  pkt_burst_io_forward(struct fwd_stream *fs)
 #ifdef RTE_TEST_PMD_RECORD_CORE_CYCLES
 	start_tsc = rte_rdtsc();
 #endif
+	int i;
 
 	/*
 	 * Receive a burst of packets and forward them.
@@ -104,6 +105,20 @@  pkt_burst_io_forward(struct fwd_stream *fs)
 #ifdef RTE_TEST_PMD_RECORD_BURST_STATS
 	fs->rx_burst_stats.pkt_burst_spread[nb_rx]++;
 #endif
+
+	for (i = 0; i < nb_rx; i++) {
+		uint64_t tx_ol_flags = 0, rx_ol_flags = pkts_burst[i]->ol_flags;
+
+		if (rx_ol_flags & PKT_RX_VLAN_PKT)
+			tx_ol_flags |= PKT_TX_VLAN_PKT;
+		else if (rx_ol_flags & PKT_RX_QINQ_PKT)
+			tx_ol_flags |= PKT_TX_QINQ_PKT;
+
+		pkts_burst[i]->ol_flags = 0;
+		pkts_burst[i]->ol_flags = tx_ol_flags;
+
+	}
+
 	nb_tx = rte_eth_tx_burst(fs->tx_port, fs->tx_queue,
 			pkts_burst, nb_rx);
 	/*