[dpdk-dev] [PATCH 1/3] nfp: clean tx descriptor flags

[dpdk-dev] [PATCH 1/3] nfp: clean tx descriptor flags

[Alejandro Lucero]

When LSO, not doing this can led to firmware disruption. It does
not show as error because TCP ends up sending data again later on.

Fixes: 9ba3d0ae2090 (\"net/nfp: add TSO support\")

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 drivers/net/nfp/nfp_net.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index f8ed976..9d9420d 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -1660,16 +1660,22 @@ static void nfp_net_read_mac(struct nfp_net_hw *hw)
 	struct nfp_net_hw *hw = txq->hw;
 
 	if (!(hw->cap & NFP_NET_CFG_CTRL_LSO))
-		return;
+		goto clean_txd;
 
 	ol_flags = mb->ol_flags;
 
 	if (!(ol_flags & PKT_TX_TCP_SEG))
-		return;
+		goto clean_txd;
 
 	txd->l4_offset = mb->l2_len + mb->l3_len + mb->l4_len;
 	txd->lso = rte_cpu_to_le_16(mb->tso_segsz);
-	txd->flags |= PCIE_DESC_TX_LSO;
+	txd->flags = PCIE_DESC_TX_LSO;
+	return;
+
+clean_txd:
+	txd->flags = 0;
+	txd->l4_offset = 0;
+	txd->lso = 0;
 }
 
 /* nfp_net_tx_cksum - Set TX CSUM offload flags in TX descriptor */
-- 
1.9.1

[dpdk-dev] [PATCH 2/3] nfp: fix problem with pkt_len to data_len conversion

[Alejandro Lucero]

Chained mbufs hold data_len as the length of that particular mbuf
and pkt_len as the full packet length including all the chained
mbufs. It is not clear from the mbuf definition if pkt_len should
be set for all the mbufs in a chain, but code there for handling
mbufs suggests just the first mbuf requires to have pkt_len set.

NFP PMD was assuming pkt_len is set in all the chained mbufs and
unit tests for gather dma were building mbufs with pkt_len always
set. This patch gets rid of that assumption.

Fixes: b812daadad0d (\"nfp: add Rx and Tx\")

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 drivers/net/nfp/nfp_net.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 9d9420d..6155d7c 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -2097,6 +2097,7 @@ uint32_t nfp_net_txq_full(struct nfp_net_txq *txq)
 		 * Checksum and VLAN flags just in the first descriptor for a
 		 * multisegment packet, but TSO info needs to be in all of them.
 		 */
+		txd.data_len = pkt->pkt_len;
 		nfp_net_tx_tso(txq, &txd, pkt);
 		nfp_net_tx_cksum(txq, &txd, pkt);
 
@@ -2132,7 +2133,7 @@ uint32_t nfp_net_txq_full(struct nfp_net_txq *txq)
 
 			/* Filling descriptors fields */
 			txds->dma_len = dma_size;
-			txds->data_len = pkt->pkt_len;
+			txds->data_len = txd.data_len;
 			txds->dma_addr_hi = (dma_addr >> 32) & 0xff;
 			txds->dma_addr_lo = (dma_addr & 0xffffffff);
 			ASSERT(free_descs > 0);
-- 
1.9.1

[dpdk-dev] [PATCH 3/3] nfp: fix problem with rx interrupts

[Alejandro Lucero]

Current code enables RX interrupts even if this it not
requested.

Fixes: ea121b28316d (\"net/nfp: add Rx interrupts\")

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 drivers/net/nfp/nfp_net.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 6155d7c..a1ad97a 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -708,7 +708,8 @@ static void nfp_net_read_mac(struct nfp_net_hw *hw)
 			return -1;
 	}
 
-	nfp_configure_rx_interrupt(dev, intr_handle);
+	if (rte_intr_dp_is_en(intr_handle))
+		nfp_configure_rx_interrupt(dev, intr_handle);
 
 	rte_intr_enable(intr_handle);
 
-- 
1.9.1

Re: [dpdk-dev] [PATCH 1/3] nfp: clean tx descriptor flags

[Ferruh Yigit]

On 3/21/2017 10:43 AM, Alejandro Lucero wrote:
> When LSO, not doing this can led to firmware disruption. It does
> not show as error because TCP ends up sending data again later on.
> 
> Fixes: 9ba3d0ae2090 (\"net/nfp: add TSO support\")
> 
> Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>

Series applied to dpdk-next-net/master, thanks.