From: Ajit Khaparde <ajit.khaparde@broadcom.com>
To: dev@dpdk.org
Cc: ferruh.yigit@intel.com
Subject: [dpdk-dev] [PATCH v3 09/24] net/bnxt: set checksum offload flags correctly
Date: Thu, 28 Sep 2017 09:13:14 -0500 [thread overview]
Message-ID: <20170928141329.73874-10-ajit.khaparde@broadcom.com> (raw)
In-Reply-To: <20170928141329.73874-1-ajit.khaparde@broadcom.com>
We are not correctly setting hw checksum offload for all the
offload flags. This patch fixes that.
Fixes: 6eb3cc2294fd ("net/bnxt: add initial Tx code")
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
---
drivers/net/bnxt/bnxt_txr.c | 32 +++++++++++++++++++++++++-------
drivers/net/bnxt/bnxt_txr.h | 21 +++++++++++++++++++++
2 files changed, 46 insertions(+), 7 deletions(-)
diff --git a/drivers/net/bnxt/bnxt_txr.c b/drivers/net/bnxt/bnxt_txr.c
index 6870b16d1..60cc17405 100644
--- a/drivers/net/bnxt/bnxt_txr.c
+++ b/drivers/net/bnxt/bnxt_txr.c
@@ -161,7 +161,7 @@ static uint16_t bnxt_start_xmit(struct rte_mbuf *tx_pkt,
if (tx_pkt->ol_flags & (PKT_TX_TCP_SEG | PKT_TX_TCP_CKSUM |
PKT_TX_UDP_CKSUM | PKT_TX_IP_CKSUM |
- PKT_TX_VLAN_PKT))
+ PKT_TX_VLAN_PKT | PKT_TX_OUTER_IP_CKSUM))
long_bd = true;
tx_buf = &txr->tx_buf_ring[txr->tx_prod];
@@ -211,21 +211,39 @@ static uint16_t bnxt_start_xmit(struct rte_mbuf *tx_pkt,
if (tx_pkt->ol_flags & PKT_TX_TCP_SEG) {
/* TSO */
- txbd1->lflags = TX_BD_LONG_LFLAGS_LSO;
+ txbd1->lflags |= TX_BD_LONG_LFLAGS_LSO;
txbd1->hdr_size = tx_pkt->l2_len + tx_pkt->l3_len +
tx_pkt->l4_len + tx_pkt->outer_l2_len +
tx_pkt->outer_l3_len;
txbd1->mss = tx_pkt->tso_segsz;
- } else if (tx_pkt->ol_flags & (PKT_TX_TCP_CKSUM |
- PKT_TX_UDP_CKSUM)) {
+ } else if (tx_pkt->ol_flags & PKT_TX_OIP_IIP_TCP_UDP_CKSUM) {
+ /* Outer IP, Inner IP, Inner TCP/UDP CSO */
+ txbd1->lflags |= TX_BD_FLG_TIP_IP_TCP_UDP_CHKSUM;
+ txbd1->mss = 0;
+ } else if (tx_pkt->ol_flags & PKT_TX_IIP_TCP_UDP_CKSUM) {
+ /* (Inner) IP, (Inner) TCP/UDP CSO */
+ txbd1->lflags |= TX_BD_FLG_IP_TCP_UDP_CHKSUM;
+ txbd1->mss = 0;
+ } else if (tx_pkt->ol_flags & PKT_TX_OIP_TCP_UDP_CKSUM) {
+ /* Outer IP, (Inner) TCP/UDP CSO */
+ txbd1->lflags |= TX_BD_FLG_TIP_TCP_UDP_CHKSUM;
+ txbd1->mss = 0;
+ } else if (tx_pkt->ol_flags & PKT_TX_OIP_IIP_CKSUM) {
+ /* Outer IP, Inner IP CSO */
+ txbd1->lflags |= TX_BD_FLG_TIP_IP_CHKSUM;
+ txbd1->mss = 0;
+ } else if (tx_pkt->ol_flags & PKT_TX_TCP_UDP_CKSUM) {
/* TCP/UDP CSO */
- txbd1->lflags = TX_BD_LONG_LFLAGS_TCP_UDP_CHKSUM;
+ txbd1->lflags |= TX_BD_LONG_LFLAGS_TCP_UDP_CHKSUM;
txbd1->mss = 0;
-
} else if (tx_pkt->ol_flags & PKT_TX_IP_CKSUM) {
/* IP CSO */
- txbd1->lflags = TX_BD_LONG_LFLAGS_IP_CHKSUM;
+ txbd1->lflags |= TX_BD_LONG_LFLAGS_IP_CHKSUM;
+ txbd1->mss = 0;
+ } else if (tx_pkt->ol_flags & PKT_TX_OUTER_IP_CKSUM) {
+ /* IP CSO */
+ txbd1->lflags |= TX_BD_LONG_LFLAGS_T_IP_CHKSUM;
txbd1->mss = 0;
}
} else {
diff --git a/drivers/net/bnxt/bnxt_txr.h b/drivers/net/bnxt/bnxt_txr.h
index 5b0971141..3f3eb312b 100644
--- a/drivers/net/bnxt/bnxt_txr.h
+++ b/drivers/net/bnxt/bnxt_txr.h
@@ -69,4 +69,25 @@ int bnxt_init_tx_ring_struct(struct bnxt_tx_queue *txq, unsigned int socket_id);
uint16_t bnxt_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
uint16_t nb_pkts);
+#define PKT_TX_OIP_IIP_TCP_UDP_CKSUM (PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM | \
+ PKT_TX_IP_CKSUM | PKT_TX_OUTER_IP_CKSUM)
+#define PKT_TX_IIP_TCP_UDP_CKSUM (PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM | \
+ PKT_TX_IP_CKSUM)
+#define PKT_TX_OIP_TCP_UDP_CKSUM (PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM | \
+ PKT_TX_OUTER_IP_CKSUM)
+#define PKT_TX_OIP_IIP_CKSUM (PKT_TX_IP_CKSUM | \
+ PKT_TX_OUTER_IP_CKSUM)
+#define PKT_TX_TCP_UDP_CKSUM (PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM)
+
+
+#define TX_BD_FLG_TIP_IP_TCP_UDP_CHKSUM (TX_BD_LONG_LFLAGS_TCP_UDP_CHKSUM | \
+ TX_BD_LONG_LFLAGS_T_IP_CHKSUM | \
+ TX_BD_LONG_LFLAGS_IP_CHKSUM)
+#define TX_BD_FLG_IP_TCP_UDP_CHKSUM (TX_BD_LONG_LFLAGS_TCP_UDP_CHKSUM | \
+ TX_BD_LONG_LFLAGS_IP_CHKSUM)
+#define TX_BD_FLG_TIP_IP_CHKSUM (TX_BD_LONG_LFLAGS_T_IP_CHKSUM | \
+ TX_BD_LONG_LFLAGS_IP_CHKSUM)
+#define TX_BD_FLG_TIP_TCP_UDP_CHKSUM (TX_BD_LONG_LFLAGS_TCP_UDP_CHKSUM | \
+ TX_BD_LONG_LFLAGS_T_IP_CHKSUM)
+
#endif
--
2.13.5 (Apple Git-94)
next prev parent reply other threads:[~2017-09-28 14:13 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-28 14:13 [dpdk-dev] [PATCH v3 00/24] bnxt patchset Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 01/24] net/bnxt: fix HWRM_*() macros and locking Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 02/24] net/bnxt: use 64-bits of address for vlan_table Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 03/24] net/bnxt: fix an issue with group id calculation Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 04/24] net/bnxt: fix calculation of number of pools Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 05/24] net/bnxt: handle multi queue mode properly Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 06/24] net/bnxt: fix rx handling and buffer allocation logic Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 07/24] net/bnxt: fix an issue with broadcast traffic Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 08/24] net/bnxt: fix usage of ETH_VMDQ_* flags Ajit Khaparde
2017-09-28 14:13 ` Ajit Khaparde [this message]
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 10/24] net/bnxt: update status of Rx IP/L4 CKSUM Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 11/24] net/bnxt: add support for xstats get by id Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 12/24] net/bnxt: fix config rss update Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 13/24] net/bnxt: set the hash_key_size Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 14/24] net/bnxt: add support for rx_queue_count Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 15/24] net/bnxt: add support for rx_descriptor_status Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 16/24] net/bnxt: add support for tx_descriptor_status Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 17/24] net/bnxt: add new HWRM structs to support flow filtering Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 18/24] net/bnxt: add support for flow filter ops Ajit Khaparde
2017-09-28 16:05 ` Ferruh Yigit
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 19/24] doc: update release notes Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 20/24] net/bnxt: fix per queue stats display in xstats Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 21/24] net/bnxt: prevent interrupt handler from accessing freed memory Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 22/24] net/bnxt: add dev_supported_ptypes_get dev_op Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 23/24] net/bnxt: add support for get/set EEPROM Ajit Khaparde
2017-09-28 14:13 ` [dpdk-dev] [PATCH v3 24/24] net/bnxt: add support for rx_queue_intr_enable/disable APIs Ajit Khaparde
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170928141329.73874-10-ajit.khaparde@broadcom.com \
--to=ajit.khaparde@broadcom.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).