From: Nithin Dabilpuram <ndabilpuram@marvell.com>
To: <jerinj@marvell.com>, <konstantin.ananyev@intel.com>,
Radu Nicolau <radu.nicolau@intel.com>,
Akhil Goyal <gakhil@marvell.com>
Cc: <dev@dpdk.org>, <anoobj@marvell.com>,
Nithin Dabilpuram <ndabilpuram@marvell.com>
Subject: [PATCH v3 2/7] examples/ipsec-secgw: disable Tx chksum offload for inline
Date: Thu, 28 Apr 2022 20:34:54 +0530 [thread overview]
Message-ID: <20220428150459.23950-2-ndabilpuram@marvell.com> (raw)
In-Reply-To: <20220428150459.23950-1-ndabilpuram@marvell.com>
Enable Tx IPv4 checksum offload only when Tx inline crypto, lookaside
crypto/protocol or cpu crypto is needed.
For Tx Inline protocol offload, checksum computation
is implicitly taken care by HW.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
---
examples/ipsec-secgw/ipsec-secgw.c | 3 ---
examples/ipsec-secgw/sa.c | 46 ++++++++++++++++++++++++++++++++------
2 files changed, 39 insertions(+), 10 deletions(-)
diff --git a/examples/ipsec-secgw/ipsec-secgw.c b/examples/ipsec-secgw/ipsec-secgw.c
index 959a20b..5fe5eee 100644
--- a/examples/ipsec-secgw/ipsec-secgw.c
+++ b/examples/ipsec-secgw/ipsec-secgw.c
@@ -1761,9 +1761,6 @@ port_init(uint16_t portid, uint64_t req_rx_offloads, uint64_t req_tx_offloads)
local_port_conf.txmode.offloads |=
RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE;
- if (dev_info.tx_offload_capa & RTE_ETH_TX_OFFLOAD_IPV4_CKSUM)
- local_port_conf.txmode.offloads |= RTE_ETH_TX_OFFLOAD_IPV4_CKSUM;
-
printf("port %u configuring rx_offloads=0x%" PRIx64
", tx_offloads=0x%" PRIx64 "\n",
portid, local_port_conf.rxmode.offloads,
diff --git a/examples/ipsec-secgw/sa.c b/examples/ipsec-secgw/sa.c
index 1839ac7..e8f2598 100644
--- a/examples/ipsec-secgw/sa.c
+++ b/examples/ipsec-secgw/sa.c
@@ -1766,10 +1766,18 @@ sa_check_offloads(uint16_t port_id, uint64_t *rx_offloads,
struct ipsec_sa *rule;
uint32_t idx_sa;
enum rte_security_session_action_type rule_type;
+ struct rte_eth_dev_info dev_info;
+ int ret;
*rx_offloads = 0;
*tx_offloads = 0;
+ ret = rte_eth_dev_info_get(port_id, &dev_info);
+ if (ret != 0)
+ rte_exit(EXIT_FAILURE,
+ "Error during getting device (port %u) info: %s\n",
+ port_id, strerror(-ret));
+
/* Check for inbound rules that use offloads and use this port */
for (idx_sa = 0; idx_sa < nb_sa_in; idx_sa++) {
rule = &sa_in[idx_sa];
@@ -1785,13 +1793,37 @@ sa_check_offloads(uint16_t port_id, uint64_t *rx_offloads,
for (idx_sa = 0; idx_sa < nb_sa_out; idx_sa++) {
rule = &sa_out[idx_sa];
rule_type = ipsec_get_action_type(rule);
- if ((rule_type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO ||
- rule_type ==
- RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL)
- && rule->portid == port_id) {
- *tx_offloads |= RTE_ETH_TX_OFFLOAD_SECURITY;
- if (rule->mss)
- *tx_offloads |= RTE_ETH_TX_OFFLOAD_TCP_TSO;
+ switch (rule_type) {
+ case RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL:
+ /* Checksum offload is not needed for inline protocol as
+ * all processing for Outbound IPSec packets will be
+ * implicitly taken care and for non-IPSec packets,
+ * there is no need of IPv4 Checksum offload.
+ */
+ if (rule->portid == port_id) {
+ *tx_offloads |= RTE_ETH_TX_OFFLOAD_SECURITY;
+ if (rule->mss)
+ *tx_offloads |= (RTE_ETH_TX_OFFLOAD_TCP_TSO |
+ RTE_ETH_TX_OFFLOAD_IPV4_CKSUM);
+ }
+ break;
+ case RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO:
+ if (rule->portid == port_id) {
+ *tx_offloads |= RTE_ETH_TX_OFFLOAD_SECURITY;
+ if (rule->mss)
+ *tx_offloads |=
+ RTE_ETH_TX_OFFLOAD_TCP_TSO;
+ *tx_offloads |= RTE_ETH_TX_OFFLOAD_IPV4_CKSUM;
+ }
+ break;
+ default:
+ /* Enable IPv4 checksum offload even if one of lookaside
+ * SA's are present.
+ */
+ if (dev_info.tx_offload_capa &
+ RTE_ETH_TX_OFFLOAD_IPV4_CKSUM)
+ *tx_offloads |= RTE_ETH_TX_OFFLOAD_IPV4_CKSUM;
+ break;
}
}
return 0;
--
2.8.4
next prev parent reply other threads:[~2022-04-28 15:05 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-03-22 17:58 [PATCH 1/7] " Nithin Dabilpuram
2022-03-22 17:58 ` [PATCH 2/7] examples/ipsec-secgw: use HW parsed packet type in poll mode Nithin Dabilpuram
2022-04-14 15:43 ` Ananyev, Konstantin
2022-03-22 17:58 ` [PATCH 3/7] examples/ipsec-secgw: allow larger burst size for vectors Nithin Dabilpuram
2022-03-22 17:58 ` [PATCH 4/7] examples/ipsec-secgw: move fast path helper functions Nithin Dabilpuram
2022-03-22 17:58 ` [PATCH 5/7] examples/ipsec-secgw: get security context from lcore conf Nithin Dabilpuram
2022-03-22 17:58 ` [PATCH 6/7] examples/ipsec-secgw: update eth header during route lookup Nithin Dabilpuram
2022-03-22 17:58 ` [PATCH 7/7] examples/ipsec-secgw: add poll mode worker for inline proto Nithin Dabilpuram
2022-04-13 6:13 ` [PATCH 1/7] examples/ipsec-secgw: disable Tx chksum offload for inline Nithin Kumar Dabilpuram
2022-04-14 14:07 ` Ananyev, Konstantin
2022-04-19 13:56 ` Nithin Kumar Dabilpuram
2022-04-20 10:42 ` Ananyev, Konstantin
2022-04-21 13:31 ` [PATCH v2 1/7] examples/ipsec-secgw: move fast path helper functions Nithin Dabilpuram
2022-04-21 13:31 ` [PATCH v2 2/7] examples/ipsec-secgw: disable Tx chksum offload for inline Nithin Dabilpuram
2022-04-21 13:31 ` [PATCH v2 3/7] examples/ipsec-secgw: use HW parsed packet type in poll mode Nithin Dabilpuram
2022-04-21 13:31 ` [PATCH v2 4/7] examples/ipsec-secgw: allow larger burst size for vectors Nithin Dabilpuram
2022-04-21 13:31 ` [PATCH v2 5/7] examples/ipsec-secgw: get security context from lcore conf Nithin Dabilpuram
2022-04-21 13:31 ` [PATCH v2 6/7] examples/ipsec-secgw: update eth header during route lookup Nithin Dabilpuram
2022-04-21 13:31 ` [PATCH v2 7/7] examples/ipsec-secgw: add poll mode worker for inline proto Nithin Dabilpuram
2022-04-28 15:04 ` [PATCH v3 1/7] examples/ipsec-secgw: move fast path helper functions Nithin Dabilpuram
2022-04-28 15:04 ` Nithin Dabilpuram [this message]
2022-04-28 15:04 ` [PATCH v3 3/7] examples/ipsec-secgw: use HW parsed packet type in poll mode Nithin Dabilpuram
2022-04-28 15:04 ` [PATCH v3 4/7] examples/ipsec-secgw: allow larger burst size for vectors Nithin Dabilpuram
2022-04-28 15:04 ` [PATCH v3 5/7] examples/ipsec-secgw: get security context from lcore conf Nithin Dabilpuram
2022-04-28 15:04 ` [PATCH v3 6/7] examples/ipsec-secgw: update eth header during route lookup Nithin Dabilpuram
2022-04-28 15:04 ` [PATCH v3 7/7] examples/ipsec-secgw: add poll mode worker for inline proto Nithin Dabilpuram
2022-04-29 10:23 ` [PATCH v3 1/7] examples/ipsec-secgw: move fast path helper functions Nithin Kumar Dabilpuram
2022-04-29 10:29 ` Akhil Goyal
2022-04-29 20:44 ` [PATCH v4 " Nithin Dabilpuram
2022-04-29 20:44 ` [PATCH v4 2/7] examples/ipsec-secgw: disable Tx chksum offload for inline Nithin Dabilpuram
2022-05-01 17:10 ` Konstantin Ananyev
2022-04-29 20:44 ` [PATCH v4 3/7] examples/ipsec-secgw: use HW parsed packet type in poll mode Nithin Dabilpuram
2022-04-29 20:44 ` [PATCH v4 4/7] examples/ipsec-secgw: allow larger burst size for vectors Nithin Dabilpuram
2022-04-29 20:44 ` [PATCH v4 5/7] examples/ipsec-secgw: get security context from lcore conf Nithin Dabilpuram
2022-04-29 20:44 ` [PATCH v4 6/7] examples/ipsec-secgw: update eth header during route lookup Nithin Dabilpuram
2022-04-29 20:44 ` [PATCH v4 7/7] examples/ipsec-secgw: add poll mode worker for inline proto Nithin Dabilpuram
2022-05-11 19:34 ` [PATCH v4 1/7] examples/ipsec-secgw: move fast path helper functions Akhil Goyal
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=20220428150459.23950-2-ndabilpuram@marvell.com \
--to=ndabilpuram@marvell.com \
--cc=anoobj@marvell.com \
--cc=dev@dpdk.org \
--cc=gakhil@marvell.com \
--cc=jerinj@marvell.com \
--cc=konstantin.ananyev@intel.com \
--cc=radu.nicolau@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).