DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Min Hu (Connor)" <humin29@huawei.com>
To: <dev@dpdk.org>
Cc: <ferruh.yigit@intel.com>
Subject: [dpdk-dev] [PATCH 5/5] net/hns3: select Tx prepare based on Tx offload
Date: Wed, 28 Apr 2021 15:20:55 +0800	[thread overview]
Message-ID: <1619594455-56787-6-git-send-email-humin29@huawei.com> (raw)
In-Reply-To: <1619594455-56787-1-git-send-email-humin29@huawei.com>

From: Chengchang Tang <tangchengchang@huawei.com>

Tx prepare should be called only when necessary to reduce the impact on
performance.

For partial TX offload, users need to call rte_eth_tx_prepare() to invoke
the tx_prepare callback of PMDs. In this callback, the PMDs adjust the
packet based on the offloading used by the user. (e.g. For some PMDs,
pseudo-headers need to be calculated when the TX cksum is offloaded.)

However, for the users, they cannot grasp all the hardware and PMDs
characteristics. As a result, users cannot decide when they need to
actually call tx_prepare. Therefore, we should assume that the user calls
rte_eth_tx_prepare() when using any Tx offloading to ensure that related
functions work properly. Whether packets need to be adjusted should be
determined by PMDs. They can make judgments in the dev_configure or
queue_setup phase. When the related function is not used, the pointer of
tx_prepare should be set to NULL to reduce the performance loss caused by
invoking rte_eth_tx_repare().

In this patch, if tx_prepare is not required for the offloading used by
the users, the tx_prepare pointer will be set to NULL.

Fixes: bba636698316 ("net/hns3: support Rx/Tx and related operations")
Cc: stable@dpdk.org

Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
 drivers/net/hns3/hns3_rxtx.c | 36 +++++++++++++++++++++++++++++++++---
 1 file changed, 33 insertions(+), 3 deletions(-)

diff --git a/drivers/net/hns3/hns3_rxtx.c b/drivers/net/hns3/hns3_rxtx.c
index 3881a72..7ac3a48 100644
--- a/drivers/net/hns3/hns3_rxtx.c
+++ b/drivers/net/hns3/hns3_rxtx.c
@@ -4203,17 +4203,45 @@ hns3_tx_check_simple_support(struct rte_eth_dev *dev)
 	return (offloads == (offloads & DEV_TX_OFFLOAD_MBUF_FAST_FREE));
 }
 
+static bool
+hns3_get_tx_prep_needed(struct rte_eth_dev *dev)
+{
+#ifdef RTE_LIBRTE_ETHDEV_DEBUG
+	/* always perform tx_prepare when debug */
+	return true;
+#else
+#define HNS3_DEV_TX_CSKUM_TSO_OFFLOAD_MASK (\
+		DEV_TX_OFFLOAD_IPV4_CKSUM | \
+		DEV_TX_OFFLOAD_TCP_CKSUM | \
+		DEV_TX_OFFLOAD_UDP_CKSUM | \
+		DEV_TX_OFFLOAD_SCTP_CKSUM | \
+		DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM | \
+		DEV_TX_OFFLOAD_OUTER_UDP_CKSUM | \
+		DEV_TX_OFFLOAD_TCP_TSO | \
+		DEV_TX_OFFLOAD_VXLAN_TNL_TSO | \
+		DEV_TX_OFFLOAD_GRE_TNL_TSO | \
+		DEV_TX_OFFLOAD_GENEVE_TNL_TSO)
+
+	uint64_t tx_offload = dev->data->dev_conf.txmode.offloads;
+	if (tx_offload & HNS3_DEV_TX_CSKUM_TSO_OFFLOAD_MASK)
+		return true;
+
+	return false;
+#endif
+}
+
 static eth_tx_burst_t
 hns3_get_tx_function(struct rte_eth_dev *dev, eth_tx_prep_t *prep)
 {
 	struct hns3_adapter *hns = dev->data->dev_private;
 	bool vec_allowed, sve_allowed, simple_allowed;
-	bool vec_support;
+	bool vec_support, tx_prepare_needed;
 
 	vec_support = hns3_tx_check_vec_support(dev) == 0;
 	vec_allowed = vec_support && hns3_get_default_vec_support();
 	sve_allowed = vec_support && hns3_get_sve_support();
 	simple_allowed = hns3_tx_check_simple_support(dev);
+	tx_prepare_needed = hns3_get_tx_prep_needed(dev);
 
 	*prep = NULL;
 
@@ -4224,7 +4252,8 @@ hns3_get_tx_function(struct rte_eth_dev *dev, eth_tx_prep_t *prep)
 	if (hns->tx_func_hint == HNS3_IO_FUNC_HINT_SIMPLE && simple_allowed)
 		return hns3_xmit_pkts_simple;
 	if (hns->tx_func_hint == HNS3_IO_FUNC_HINT_COMMON) {
-		*prep = hns3_prep_pkts;
+		if (tx_prepare_needed)
+			*prep = hns3_prep_pkts;
 		return hns3_xmit_pkts;
 	}
 
@@ -4233,7 +4262,8 @@ hns3_get_tx_function(struct rte_eth_dev *dev, eth_tx_prep_t *prep)
 	if (simple_allowed)
 		return hns3_xmit_pkts_simple;
 
-	*prep = hns3_prep_pkts;
+	if (tx_prepare_needed)
+		*prep = hns3_prep_pkts;
 	return hns3_xmit_pkts;
 }
 
-- 
2.7.4


  parent reply	other threads:[~2021-04-28  7:21 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-28  7:20 [dpdk-dev] [PATCH 0/5] Features and bugfix for hns3 PMD Min Hu (Connor)
2021-04-28  7:20 ` [dpdk-dev] [PATCH 1/5] net/hns3: support preferred burst size and queues in VF Min Hu (Connor)
2021-04-28  7:20 ` [dpdk-dev] [PATCH 2/5] net/hns3: log time delta in decimal format Min Hu (Connor)
2021-04-28  7:20 ` [dpdk-dev] [PATCH 3/5] net/hns3: fix use wrong time API Min Hu (Connor)
2021-04-28  7:20 ` [dpdk-dev] [PATCH 4/5] net/hns3: delete unused macro of cmd module Min Hu (Connor)
2021-04-28  7:20 ` Min Hu (Connor) [this message]
2021-05-07  9:26   ` [dpdk-dev] [PATCH 5/5] net/hns3: select Tx prepare based on Tx offload David Marchand
2021-05-07 10:23     ` Ferruh Yigit
2021-05-07 12:20       ` [dpdk-dev] [PATCH] net/hns3: fix debug build Thomas Monjalon
2021-05-07 12:25         ` David Marchand
2021-05-07 12:49           ` Thomas Monjalon
2021-05-08  0:47           ` Min Hu (Connor)
2021-04-29 16:27 ` [dpdk-dev] [PATCH 0/5] Features and bugfix for hns3 PMD Ferruh Yigit

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=1619594455-56787-6-git-send-email-humin29@huawei.com \
    --to=humin29@huawei.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).