From: Dongdong Liu <liudongdong3@huawei.com>
To: <dev@dpdk.org>, <andrew.rybchenko@oktetlabs.ru>,
<ferruh.yigit@amd.com>, <thomas@monjalon.net>
Cc: <stable@dpdk.org>, <fengchengwen@huawei.com>,
<yisen.zhuang@huawei.com>, <liudongdong3@huawei.com>,
<lihuisong@huawei.com>, <haijie1@huawei.com>
Subject: [PATCH 19/19] net/hns3: fix the obtaination of minimum Tx frame length
Date: Fri, 30 Sep 2022 15:22:20 +0800 [thread overview]
Message-ID: <20220930072220.20753-20-liudongdong3@huawei.com> (raw)
In-Reply-To: <20220930072220.20753-1-liudongdong3@huawei.com>
From: Jie Hai <haijie1@huawei.com>
When packet length in Tx is less than length hardware supported,
the minimum frame length in hns3 is used to do padding to avoid
hardware error. Currently, this length is fixed by macro, which
is very unfavorable for subsequent hardware evolution. So fix it
as firmware report.
Fixes: 395b5e08ef8d ("net/hns3: add Tx short frame padding compatibility")
Cc: stable@dpdk.org
Signed-off-by: Jie Hai <haijie1@huawei.com>
Signed-off-by: Dongdong Liu <liudongdong3@huawei.com>
---
drivers/net/hns3/hns3_cmd.h | 6 ++++++
drivers/net/hns3/hns3_ethdev.c | 4 +++-
drivers/net/hns3/hns3_ethdev.h | 3 +--
drivers/net/hns3/hns3_ethdev_vf.c | 4 +++-
4 files changed, 13 insertions(+), 4 deletions(-)
diff --git a/drivers/net/hns3/hns3_cmd.h b/drivers/net/hns3/hns3_cmd.h
index 8ac8b45819..994dfc48cc 100644
--- a/drivers/net/hns3/hns3_cmd.h
+++ b/drivers/net/hns3/hns3_cmd.h
@@ -967,6 +967,12 @@ struct hns3_dev_specs_0_cmd {
uint32_t max_tm_rate;
};
+struct hns3_dev_specs_1_cmd {
+ uint8_t rsv0[12];
+ uint8_t min_tx_pkt_len;
+ uint8_t rsv1[11];
+};
+
struct hns3_query_rpu_cmd {
uint32_t tc_queue_num;
uint32_t rsv1[2];
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index 60e933998a..813fcedc6a 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -2661,14 +2661,17 @@ static void
hns3_parse_dev_specifications(struct hns3_hw *hw, struct hns3_cmd_desc *desc)
{
struct hns3_dev_specs_0_cmd *req0;
+ struct hns3_dev_specs_1_cmd *req1;
req0 = (struct hns3_dev_specs_0_cmd *)desc[0].data;
+ req1 = (struct hns3_dev_specs_1_cmd *)desc[1].data;
hw->max_non_tso_bd_num = req0->max_non_tso_bd_num;
hw->rss_ind_tbl_size = rte_le_to_cpu_16(req0->rss_ind_tbl_size);
hw->rss_key_size = rte_le_to_cpu_16(req0->rss_key_size);
hw->max_tm_rate = rte_le_to_cpu_32(req0->max_tm_rate);
hw->intr.int_ql_max = rte_le_to_cpu_16(req0->intr_ql_max);
+ hw->min_tx_pkt_len = req1->min_tx_pkt_len;
}
static int
@@ -2763,7 +2766,6 @@ hns3_get_capability(struct hns3_hw *hw)
hw->tso_mode = HNS3_TSO_HW_CAL_PSEUDO_H_CSUM;
hw->vlan_mode = HNS3_HW_SHIFT_AND_DISCARD_MODE;
hw->drop_stats_mode = HNS3_PKTS_DROP_STATS_MODE2;
- hw->min_tx_pkt_len = HNS3_HIP09_MIN_TX_PKT_LEN;
pf->tqp_config_mode = HNS3_FLEX_MAX_TQP_NUM_MODE;
hw->rss_info.ipv6_sctp_offload_supported = true;
hw->udp_cksum_mode = HNS3_SPECIAL_PORT_HW_CKSUM_MODE;
diff --git a/drivers/net/hns3/hns3_ethdev.h b/drivers/net/hns3/hns3_ethdev.h
index 9fe9766736..2457754b3d 100644
--- a/drivers/net/hns3/hns3_ethdev.h
+++ b/drivers/net/hns3/hns3_ethdev.h
@@ -75,7 +75,6 @@
#define HNS3_DEFAULT_MTU 1500UL
#define HNS3_DEFAULT_FRAME_LEN (HNS3_DEFAULT_MTU + HNS3_ETH_OVERHEAD)
#define HNS3_HIP08_MIN_TX_PKT_LEN 33
-#define HNS3_HIP09_MIN_TX_PKT_LEN 9
#define HNS3_BITS_PER_BYTE 8
@@ -550,7 +549,7 @@ struct hns3_hw {
* The minimum length of the packet supported by hardware in the Tx
* direction.
*/
- uint32_t min_tx_pkt_len;
+ uint8_t min_tx_pkt_len;
struct hns3_queue_intr intr;
/*
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c b/drivers/net/hns3/hns3_ethdev_vf.c
index 446a0cdbc7..c1bbcf42b1 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -701,13 +701,16 @@ static void
hns3vf_parse_dev_specifications(struct hns3_hw *hw, struct hns3_cmd_desc *desc)
{
struct hns3_dev_specs_0_cmd *req0;
+ struct hns3_dev_specs_1_cmd *req1;
req0 = (struct hns3_dev_specs_0_cmd *)desc[0].data;
+ req1 = (struct hns3_dev_specs_1_cmd *)desc[1].data;
hw->max_non_tso_bd_num = req0->max_non_tso_bd_num;
hw->rss_ind_tbl_size = rte_le_to_cpu_16(req0->rss_ind_tbl_size);
hw->rss_key_size = rte_le_to_cpu_16(req0->rss_key_size);
hw->intr.int_ql_max = rte_le_to_cpu_16(req0->intr_ql_max);
+ hw->min_tx_pkt_len = req1->min_tx_pkt_len;
}
static int
@@ -846,7 +849,6 @@ hns3vf_get_capability(struct hns3_hw *hw)
hw->intr.gl_unit = HNS3_INTR_COALESCE_GL_UINT_1US;
hw->tso_mode = HNS3_TSO_HW_CAL_PSEUDO_H_CSUM;
hw->drop_stats_mode = HNS3_PKTS_DROP_STATS_MODE2;
- hw->min_tx_pkt_len = HNS3_HIP09_MIN_TX_PKT_LEN;
hw->rss_info.ipv6_sctp_offload_supported = true;
hw->promisc_mode = HNS3_LIMIT_PROMISC_MODE;
--
2.22.0
next prev parent reply other threads:[~2022-09-30 7:24 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-30 7:22 [PATCH 00/19] some bugfixes and clean code for hns3 - part2 Dongdong Liu
2022-09-30 7:22 ` [PATCH 01/19] net/hns3: fix have no valid RSS rule Dongdong Liu
2022-09-30 7:22 ` [PATCH 02/19] net/hns3: fix RSS filter restore Dongdong Liu
2022-09-30 7:22 ` [PATCH 03/19] net/hns3: fix the lock protection of RSS flow rule Dongdong Liu
2022-09-30 7:22 ` [PATCH 04/19] net/hns3: fix RSS flow rule restore failed Dongdong Liu
2022-09-30 7:22 ` [PATCH 05/19] net/hns3: move flow direction rule recovery position Dongdong Liu
2022-09-30 7:22 ` [PATCH 06/19] net/hns3: fix hns3 restore filter function input Dongdong Liu
2022-09-30 7:22 ` [PATCH 07/19] net/hns3: fix gcov compile warning Dongdong Liu
2022-09-30 7:22 ` [PATCH 08/19] net/hns3: fix incorrect packet type report for GENEVE Dongdong Liu
2022-09-30 7:22 ` [PATCH 09/19] net/hns3: fix magic numbers Dongdong Liu
2022-09-30 7:22 ` [PATCH 10/19] net/hns3: fix header file self contained Dongdong Liu
2022-09-30 7:22 ` [PATCH 11/19] net/hns3: remove unused structure definition Dongdong Liu
2022-09-30 7:22 ` [PATCH 12/19] net/hns3: do not use reserved identifier macro Dongdong Liu
2022-09-30 7:22 ` [PATCH 13/19] net/hns3: fix fail to enable IPV4 or IPV6 packet RSS Dongdong Liu
2022-09-30 7:22 ` [PATCH 14/19] net/hns3: fix some spelling errors Dongdong Liu
2022-09-30 7:22 ` [PATCH 15/19] net/hns3: fix inconsistent RSS behavior Dongdong Liu
2022-09-30 7:22 ` [PATCH 16/19] net/hns3: add check for L3 and L4 type Dongdong Liu
2022-09-30 7:22 ` [PATCH 17/19] net/hns3: revert fix mailbox communication with HW Dongdong Liu
2022-09-30 7:22 ` [PATCH 18/19] net/hns3: fix VF mark wrong message processed Dongdong Liu
2022-09-30 7:22 ` Dongdong Liu [this message]
2022-10-05 6:18 ` [PATCH 00/19] some bugfixes and clean code for hns3 - part2 Andrew Rybchenko
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=20220930072220.20753-20-liudongdong3@huawei.com \
--to=liudongdong3@huawei.com \
--cc=andrew.rybchenko@oktetlabs.ru \
--cc=dev@dpdk.org \
--cc=fengchengwen@huawei.com \
--cc=ferruh.yigit@amd.com \
--cc=haijie1@huawei.com \
--cc=lihuisong@huawei.com \
--cc=stable@dpdk.org \
--cc=thomas@monjalon.net \
--cc=yisen.zhuang@huawei.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).