packages/dpdk
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

synchronize hns3 bugfixes from upstream

Browse Source 
synchronize hns3 bugfixes from upstream

Signed-off-by: speech_white <humin29@huawei.com>
This commit is contained in:
speech_white 2021-11-01 15:56:13 +08:00
parent 717f872e49
commit 6e262cb552
13 changed files with 2176 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
0220-examples-kni-close-port-before-exit.patch Normal file
View File

@ -0,0 +1,30 @@
From 96fdb80048e279289e012fafe762f5b53e2ecd23 Mon Sep 17 00:00:00 2001
From: Huisong Li <lihuisong@huawei.com>
Date: Tue, 14 Sep 2021 14:02:19 +0800
Subject: [PATCH 01/17] examples/kni: close port before exit
This patch adds dev_close() step to release network adapter resources
when kni free.
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
---
examples/kni/main.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/examples/kni/main.c b/examples/kni/main.c
index fe93b8618..40e1790c4 100644
--- a/examples/kni/main.c
+++ b/examples/kni/main.c
@@ -1031,6 +1031,7 @@ kni_free_kni(uint16_t port_id)
if (ret != 0)
RTE_LOG(ERR, APP, "Failed to stop port %d: %s\n",
port_id, rte_strerror(-ret));
+ rte_eth_dev_close(port_id);
return 0;
}
--
2.23.0

141
0221-net-hns3-fix-residual-MAC-after-setting-default-MAC.patch Normal file
View File

@ -0,0 +1,141 @@
From 5fcd00f784f9b984bf1f8a084a6be32816585717 Mon Sep 17 00:00:00 2001
From: Huisong Li <lihuisong@huawei.com>
Date: Wed, 22 Sep 2021 11:41:51 +0800
Subject: [PATCH 02/17] net/hns3: fix residual MAC after setting default MAC
This problem occurs in the following scenarios:
1) reset is encountered when the adapter is running.
2) set a new default MAC address
After the above two steps, the old default MAC address should be not
take effect. But the current behavior is contrary to that. This is due
to the change of the "default_addr_setted" in hw->mac from 'true' to
'false' after the reset. As a result, the old MAC address is not removed
when the new default MAC address is set. This variable controls whether
to delete the old default MAC address when setting the default MAC
address. It is only used when the mac_addr_set API is called for the
first time. In fact, when a unicast MAC address is deleted, if the
address isn't in the MAC address table, the driver doesn't return
failure. So this patch remove the redundant and troublesome variables to
resolve this problem.
Fixes: 7d7f9f80bbfb ("net/hns3: support MAC address related operations")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
drivers/net/hns3/hns3_ethdev.c | 38 ++++++++++------------------------
drivers/net/hns3/hns3_ethdev.h | 1 -
2 files changed, 11 insertions(+), 28 deletions(-)
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index 2fb0c466c..d6228601f 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -1651,7 +1651,7 @@ hns3_remove_mc_addr_common(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
static int
hns3_add_mac_addr(struct rte_eth_dev *dev, struct rte_ether_addr *mac_addr,
- uint32_t idx, __rte_unused uint32_t pool)
+ __rte_unused uint32_t idx, __rte_unused uint32_t pool)
{
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
@@ -1682,8 +1682,6 @@ hns3_add_mac_addr(struct rte_eth_dev *dev, struct rte_ether_addr *mac_addr,
return ret;
}
- if (idx == 0)
- hw->mac.default_addr_setted = true;
rte_spinlock_unlock(&hw->lock);
return ret;
@@ -1748,30 +1746,19 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct rte_ether_addr *oaddr;
char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
- bool default_addr_setted;
int ret, ret_val;
- /*
- * It has been guaranteed that input parameter named mac_addr is valid
- * address in the rte layer of DPDK framework.
- */
- oaddr = (struct rte_ether_addr *)hw->mac.mac_addr;
- default_addr_setted = hw->mac.default_addr_setted;
- if (default_addr_setted && !!rte_is_same_ether_addr(mac_addr, oaddr))
- return 0;
-
rte_spinlock_lock(&hw->lock);
- if (default_addr_setted) {
- ret = hns3_remove_uc_addr_common(hw, oaddr);
- if (ret) {
- hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
- oaddr);
- hns3_warn(hw, "Remove old uc mac address(%s) fail: %d",
- mac_str, ret);
+ oaddr = (struct rte_ether_addr *)hw->mac.mac_addr;
+ ret = hns3_remove_uc_addr_common(hw, oaddr);
+ if (ret) {
+ hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
+ oaddr);
+ hns3_warn(hw, "Remove old uc mac address(%s) fail: %d",
+ mac_str, ret);
- rte_spinlock_unlock(&hw->lock);
- return ret;
- }
+ rte_spinlock_unlock(&hw->lock);
+ return ret;
}
ret = hns3_add_uc_addr_common(hw, mac_addr);
@@ -1790,7 +1777,6 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
rte_ether_addr_copy(mac_addr,
(struct rte_ether_addr *)hw->mac.mac_addr);
- hw->mac.default_addr_setted = true;
rte_spinlock_unlock(&hw->lock);
return 0;
@@ -1811,7 +1797,6 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, oaddr);
hns3_warn(hw, "Failed to restore old uc mac addr(%s): %d",
mac_str, ret_val);
- hw->mac.default_addr_setted = false;
}
rte_spinlock_unlock(&hw->lock);
@@ -3470,7 +3455,6 @@ hns3_get_board_configuration(struct hns3_hw *hw)
hw->rss_dis_flag = false;
memcpy(hw->mac.mac_addr, cfg.mac_addr, RTE_ETHER_ADDR_LEN);
hw->mac.phy_addr = cfg.phy_addr;
- hw->mac.default_addr_setted = false;
hw->num_tx_desc = cfg.tqp_desc_num;
hw->num_rx_desc = cfg.tqp_desc_num;
hw->dcb_info.num_pg = 1;
@@ -5928,7 +5912,7 @@ hns3_do_stop(struct hns3_adapter *hns)
return ret;
}
}
- hw->mac.default_addr_setted = false;
+
return 0;
}
diff --git a/drivers/net/hns3/hns3_ethdev.h b/drivers/net/hns3/hns3_ethdev.h
index ab44894a8..57387e05b 100644
--- a/drivers/net/hns3/hns3_ethdev.h
+++ b/drivers/net/hns3/hns3_ethdev.h
@@ -188,7 +188,6 @@ enum hns3_media_type {
struct hns3_mac {
uint8_t mac_addr[RTE_ETHER_ADDR_LEN];
- bool default_addr_setted; /* whether default addr(mac_addr) is set */
uint8_t media_type;
uint8_t phy_addr;
uint8_t link_duplex : 1; /* ETH_LINK_[HALF/FULL]_DUPLEX */
--
2.23.0

210
0222-net-hns3-fix-input-parameters-of-MAC-functions.patch Normal file
View File

@ -0,0 +1,210 @@
From bc25acd9ac200067f0f1a68c192076a65e4c76e6 Mon Sep 17 00:00:00 2001
From: Huisong Li <lihuisong@huawei.com>
Date: Wed, 22 Sep 2021 11:41:52 +0800
Subject: [PATCH 03/17] net/hns3: fix input parameters of MAC functions
When adding multicast and unicast MAC addresses, three descriptors and
one descriptor are required for querying or adding MAC VLAN table,
respectively. This patch uses the number of descriptors as input
parameter to complete this task to make the function more secure.
Fixes: 7d7f9f80bbfb ("net/hns3: support MAC address related operations")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
drivers/net/hns3/hns3_cmd.h | 3 +-
drivers/net/hns3/hns3_ethdev.c | 88 +++++++++++++++++++---------------
2 files changed, 51 insertions(+), 40 deletions(-)
diff --git a/drivers/net/hns3/hns3_cmd.h b/drivers/net/hns3/hns3_cmd.h
index a4683de0a..81bc9e9d9 100644
--- a/drivers/net/hns3/hns3_cmd.h
+++ b/drivers/net/hns3/hns3_cmd.h
@@ -923,7 +923,8 @@ enum hns3_mac_vlan_add_resp_code {
HNS3_ADD_MC_OVERFLOW, /* ADD failed for MC overflow */
};
-#define HNS3_MC_MAC_VLAN_ADD_DESC_NUM 3
+#define HNS3_MC_MAC_VLAN_OPS_DESC_NUM 3
+#define HNS3_UC_MAC_VLAN_OPS_DESC_NUM 1
#define HNS3_MAC_VLAN_BIT0_EN_B 0
#define HNS3_MAC_VLAN_BIT1_EN_B 1
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index d6228601f..02d68e496 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -1427,28 +1427,31 @@ hns3_get_mac_vlan_cmd_status(struct hns3_hw *hw, uint16_t cmdq_resp,
static int
hns3_lookup_mac_vlan_tbl(struct hns3_hw *hw,
struct hns3_mac_vlan_tbl_entry_cmd *req,
- struct hns3_cmd_desc *desc, bool is_mc)
+ struct hns3_cmd_desc *desc, uint8_t desc_num)
{
uint8_t resp_code;
uint16_t retval;
int ret;
+ int i;
- hns3_cmd_setup_basic_desc(&desc[0], HNS3_OPC_MAC_VLAN_ADD, true);
- if (is_mc) {
- desc[0].flag |= rte_cpu_to_le_16(HNS3_CMD_FLAG_NEXT);
- memcpy(desc[0].data, req,
- sizeof(struct hns3_mac_vlan_tbl_entry_cmd));
- hns3_cmd_setup_basic_desc(&desc[1], HNS3_OPC_MAC_VLAN_ADD,
- true);
- desc[1].flag |= rte_cpu_to_le_16(HNS3_CMD_FLAG_NEXT);
- hns3_cmd_setup_basic_desc(&desc[2], HNS3_OPC_MAC_VLAN_ADD,
+ if (desc_num == HNS3_MC_MAC_VLAN_OPS_DESC_NUM) {
+ for (i = 0; i < desc_num - 1; i++) {
+ hns3_cmd_setup_basic_desc(&desc[i],
+ HNS3_OPC_MAC_VLAN_ADD, true);
+ desc[i].flag |= rte_cpu_to_le_16(HNS3_CMD_FLAG_NEXT);
+ if (i == 0)
+ memcpy(desc[i].data, req,
+ sizeof(struct hns3_mac_vlan_tbl_entry_cmd));
+ }
+ hns3_cmd_setup_basic_desc(&desc[i], HNS3_OPC_MAC_VLAN_ADD,
true);
- ret = hns3_cmd_send(hw, desc, HNS3_MC_MAC_VLAN_ADD_DESC_NUM);
} else {
+ hns3_cmd_setup_basic_desc(&desc[0], HNS3_OPC_MAC_VLAN_ADD,
+ true);
memcpy(desc[0].data, req,
sizeof(struct hns3_mac_vlan_tbl_entry_cmd));
- ret = hns3_cmd_send(hw, desc, 1);
}
+ ret = hns3_cmd_send(hw, desc, desc_num);
if (ret) {
hns3_err(hw, "lookup mac addr failed for cmd_send, ret =%d.",
ret);
@@ -1464,38 +1467,40 @@ hns3_lookup_mac_vlan_tbl(struct hns3_hw *hw,
static int
hns3_add_mac_vlan_tbl(struct hns3_hw *hw,
struct hns3_mac_vlan_tbl_entry_cmd *req,
- struct hns3_cmd_desc *mc_desc)
+ struct hns3_cmd_desc *desc, uint8_t desc_num)
{
uint8_t resp_code;
uint16_t retval;
int cfg_status;
int ret;
+ int i;
- if (mc_desc == NULL) {
- struct hns3_cmd_desc desc;
-
- hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_MAC_VLAN_ADD, false);
- memcpy(desc.data, req,
+ if (desc_num == HNS3_UC_MAC_VLAN_OPS_DESC_NUM) {
+ hns3_cmd_setup_basic_desc(desc, HNS3_OPC_MAC_VLAN_ADD, false);
+ memcpy(desc->data, req,
sizeof(struct hns3_mac_vlan_tbl_entry_cmd));
- ret = hns3_cmd_send(hw, &desc, 1);
- resp_code = (rte_le_to_cpu_32(desc.data[0]) >> 8) & 0xff;
- retval = rte_le_to_cpu_16(desc.retval);
+ ret = hns3_cmd_send(hw, desc, desc_num);
+ resp_code = (rte_le_to_cpu_32(desc->data[0]) >> 8) & 0xff;
+ retval = rte_le_to_cpu_16(desc->retval);
cfg_status = hns3_get_mac_vlan_cmd_status(hw, retval, resp_code,
HNS3_MAC_VLAN_ADD);
} else {
- hns3_cmd_reuse_desc(&mc_desc[0], false);
- mc_desc[0].flag |= rte_cpu_to_le_16(HNS3_CMD_FLAG_NEXT);
- hns3_cmd_reuse_desc(&mc_desc[1], false);
- mc_desc[1].flag |= rte_cpu_to_le_16(HNS3_CMD_FLAG_NEXT);
- hns3_cmd_reuse_desc(&mc_desc[2], false);
- mc_desc[2].flag &= rte_cpu_to_le_16(~HNS3_CMD_FLAG_NEXT);
- memcpy(mc_desc[0].data, req,
+ for (i = 0; i < desc_num; i++) {
+ hns3_cmd_reuse_desc(&desc[i], false);
+ if (i == desc_num - 1)
+ desc[i].flag &=
+ rte_cpu_to_le_16(~HNS3_CMD_FLAG_NEXT);
+ else
+ desc[i].flag |=
+ rte_cpu_to_le_16(HNS3_CMD_FLAG_NEXT);
+ }
+ memcpy(desc[0].data, req,
sizeof(struct hns3_mac_vlan_tbl_entry_cmd));
- mc_desc[0].retval = 0;
- ret = hns3_cmd_send(hw, mc_desc, HNS3_MC_MAC_VLAN_ADD_DESC_NUM);
- resp_code = (rte_le_to_cpu_32(mc_desc[0].data[0]) >> 8) & 0xff;
- retval = rte_le_to_cpu_16(mc_desc[0].retval);
+ desc[0].retval = 0;
+ ret = hns3_cmd_send(hw, desc, desc_num);
+ resp_code = (rte_le_to_cpu_32(desc[0].data[0]) >> 8) & 0xff;
+ retval = rte_le_to_cpu_16(desc[0].retval);
cfg_status = hns3_get_mac_vlan_cmd_status(hw, retval, resp_code,
HNS3_MAC_VLAN_ADD);
@@ -1540,7 +1545,7 @@ hns3_add_uc_addr_common(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw);
struct hns3_mac_vlan_tbl_entry_cmd req;
struct hns3_pf *pf = &hns->pf;
- struct hns3_cmd_desc desc[3];
+ struct hns3_cmd_desc desc;
char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
uint16_t egress_port = 0;
uint8_t vf_id;
@@ -1574,10 +1579,12 @@ hns3_add_uc_addr_common(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
* it if the entry is inexistent. Repeated unicast entry
* is not allowed in the mac vlan table.
*/
- ret = hns3_lookup_mac_vlan_tbl(hw, &req, desc, false);
+ ret = hns3_lookup_mac_vlan_tbl(hw, &req, &desc,
+ HNS3_UC_MAC_VLAN_OPS_DESC_NUM);
if (ret == -ENOENT) {
if (!hns3_is_umv_space_full(hw)) {
- ret = hns3_add_mac_vlan_tbl(hw, &req, NULL);
+ ret = hns3_add_mac_vlan_tbl(hw, &req, &desc,
+ HNS3_UC_MAC_VLAN_OPS_DESC_NUM);
if (!ret)
hns3_update_umv_space(hw, false);
return ret;
@@ -1867,8 +1874,8 @@ hns3_update_desc_vfid(struct hns3_cmd_desc *desc, uint8_t vfid, bool clr)
static int
hns3_add_mc_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
{
+ struct hns3_cmd_desc desc[HNS3_MC_MAC_VLAN_OPS_DESC_NUM];
struct hns3_mac_vlan_tbl_entry_cmd req;
- struct hns3_cmd_desc desc[3];
char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
uint8_t vf_id;
int ret;
@@ -1885,7 +1892,8 @@ hns3_add_mc_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
memset(&req, 0, sizeof(req));
hns3_set_bit(req.entry_type, HNS3_MAC_VLAN_BIT0_EN_B, 0);
hns3_prepare_mac_addr(&req, mac_addr->addr_bytes, true);
- ret = hns3_lookup_mac_vlan_tbl(hw, &req, desc, true);
+ ret = hns3_lookup_mac_vlan_tbl(hw, &req, desc,
+ HNS3_MC_MAC_VLAN_OPS_DESC_NUM);
if (ret) {
/* This mac addr do not exist, add new entry for it */
memset(desc[0].data, 0, sizeof(desc[0].data));
@@ -1899,7 +1907,8 @@ hns3_add_mc_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
*/
vf_id = HNS3_PF_FUNC_ID;
hns3_update_desc_vfid(desc, vf_id, false);
- ret = hns3_add_mac_vlan_tbl(hw, &req, desc);
+ ret = hns3_add_mac_vlan_tbl(hw, &req, desc,
+ HNS3_MC_MAC_VLAN_OPS_DESC_NUM);
if (ret) {
if (ret == -ENOSPC)
hns3_err(hw, "mc mac vlan table is full");
@@ -1932,7 +1941,8 @@ hns3_remove_mc_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
memset(&req, 0, sizeof(req));
hns3_set_bit(req.entry_type, HNS3_MAC_VLAN_BIT0_EN_B, 0);
hns3_prepare_mac_addr(&req, mac_addr->addr_bytes, true);
- ret = hns3_lookup_mac_vlan_tbl(hw, &req, desc, true);
+ ret = hns3_lookup_mac_vlan_tbl(hw, &req, desc,
+ HNS3_MC_MAC_VLAN_OPS_DESC_NUM);
if (ret == 0) {
/*
* This mac addr exist, remove this handle's VFID for it.
--
2.23.0

84
0223-net-bonding-fix-dedicated-queue-mode-in-vector-burst.patch Normal file
View File

@ -0,0 +1,84 @@
From 19dc6356916c60f282b6d3046f5d2f1d74d48d35 Mon Sep 17 00:00:00 2001
From: Chengchang Tang <tangchengchang@huawei.com>
Date: Wed, 22 Sep 2021 15:09:12 +0800
Subject: [PATCH 10/17] net/bonding: fix dedicated queue mode in vector burst
If the vector burst mode is selected, the dedicated queue mode will not
take effect on some PMDs because these PMDs may have some limitations
in vector burst mode. For example, the limit on burst size. Currently,
both hns3 and intel I40E require four alignments when receiving packets
in vector mode. As a result, they can't accept packets if burst size
below four. However, in dedicated queue mode, the burst size of periodic
packets processing is one.
This patch fixes the above problem by modifying the burst size to 32.
This approach also makes the packet processing of the dedicated queue
mode more reasonable. Currently, if multiple LACP protocol packets are
received in the hardware queue in a cycle, only one LACP packet will be
processed in this cycle, and the left packets will be processed in the
following cycle. After the modification, all the LACP packets will be
processed at one time, which seems more reasonable and closer to the
behavior of the bonding driver when the dedicated queue is not turned on.
Fixes: 112891cd27e5 ("net/bonding: add dedicated HW queues for LACP control")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 32 ++++++++++++++++-------
1 file changed, 23 insertions(+), 9 deletions(-)
diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c b/drivers/net/bonding/rte_eth_bond_8023ad.c
index 67ca0730f..0bcce6652 100644
--- a/drivers/net/bonding/rte_eth_bond_8023ad.c
+++ b/drivers/net/bonding/rte_eth_bond_8023ad.c
@@ -823,6 +823,27 @@ rx_machine_update(struct bond_dev_private *internals, uint16_t slave_id,
rx_machine(internals, slave_id, NULL);
}
+static void
+bond_mode_8023ad_dedicated_rxq_process(struct bond_dev_private *internals,
+ uint16_t slave_id)
+{
+#define DEDICATED_QUEUE_BURST_SIZE 32
+ struct rte_mbuf *lacp_pkt[DEDICATED_QUEUE_BURST_SIZE];
+ uint16_t rx_count = rte_eth_rx_burst(slave_id,
+ internals->mode4.dedicated_queues.rx_qid,
+ lacp_pkt, DEDICATED_QUEUE_BURST_SIZE);
+
+ if (rx_count) {
+ uint16_t i;
+
+ for (i = 0; i < rx_count; i++)
+ bond_mode_8023ad_handle_slow_pkt(internals, slave_id,
+ lacp_pkt[i]);
+ } else {
+ rx_machine_update(internals, slave_id, NULL);
+ }
+}
+
static void
bond_mode_8023ad_periodic_cb(void *arg)
{
@@ -911,15 +932,8 @@ bond_mode_8023ad_periodic_cb(void *arg)
rx_machine_update(internals, slave_id, lacp_pkt);
} else {
- uint16_t rx_count = rte_eth_rx_burst(slave_id,
- internals->mode4.dedicated_queues.rx_qid,
- &lacp_pkt, 1);
-
- if (rx_count == 1)
- bond_mode_8023ad_handle_slow_pkt(internals,
- slave_id, lacp_pkt);
- else
- rx_machine_update(internals, slave_id, NULL);
+ bond_mode_8023ad_dedicated_rxq_process(internals,
+ slave_id);
}
periodic_machine(internals, slave_id);
--
2.23.0

143
0224-net-bonding-fix-RSS-key-length.patch Normal file
View File

@ -0,0 +1,143 @@
From 464bfbd345224ddb04399297988c0d99cbe8acc6 Mon Sep 17 00:00:00 2001
From: Chengchang Tang <tangchengchang@huawei.com>
Date: Wed, 22 Sep 2021 15:09:13 +0800
Subject: [PATCH 11/17] net/bonding: fix RSS key length
Currently the hash_key_size information has not been set. So, apps can
not get the key size from dev_info(), this make some problem.
e.g, in testpmd, the hash_key_size will be checked before configure
or get the hash key:
testpmd> show port 4 rss-hash
dev_info did not provide a valid hash key size
testpmd> show port 4 rss-hash key
dev_info did not provide a valid hash key size
testpmd> port config 4 rss-hash-key ipv4 (hash key)
dev_info did not provide a valid hash key size
In this patch, the meaning of rss_key_len has been modified. It only
indicated the length of the configured hash key before. Therefore,
its value depends on the user's configuration. This seems unreasonable.
And now, it indicates the minimum hash key length required by the
bonded device. Its value will be the shortest hash key among all slave
drivers.
Fixes: 734ce47f71e0 ("bonding: support RSS dynamic configuration")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
drivers/net/bonding/rte_eth_bond_api.c | 6 ++++
drivers/net/bonding/rte_eth_bond_pmd.c | 44 ++++++++++++++++----------
2 files changed, 33 insertions(+), 17 deletions(-)
diff --git a/drivers/net/bonding/rte_eth_bond_api.c b/drivers/net/bonding/rte_eth_bond_api.c
index 44775f61e..c751a1242 100644
--- a/drivers/net/bonding/rte_eth_bond_api.c
+++ b/drivers/net/bonding/rte_eth_bond_api.c
@@ -290,6 +290,7 @@ eth_bond_slave_inherit_dev_info_rx_first(struct bond_dev_private *internals,
struct rte_eth_rxconf *rxconf_i = &internals->default_rxconf;
internals->reta_size = di->reta_size;
+ internals->rss_key_len = di->hash_key_size;
/* Inherit Rx offload capabilities from the first slave device */
internals->rx_offload_capa = di->rx_offload_capa;
@@ -385,6 +386,11 @@ eth_bond_slave_inherit_dev_info_rx_next(struct bond_dev_private *internals,
*/
if (internals->reta_size > di->reta_size)
internals->reta_size = di->reta_size;
+ if (internals->rss_key_len > di->hash_key_size) {
+ RTE_BOND_LOG(WARNING, "slave has different rss key size, "
+ "configuring rss may fail");
+ internals->rss_key_len = di->hash_key_size;
+ }
if (!internals->max_rx_pktlen &&
di->max_rx_pktlen < internals->candidate_max_rx_pktlen)
diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c b/drivers/net/bonding/rte_eth_bond_pmd.c
index 057b1ada5..c21df6d6f 100644
--- a/drivers/net/bonding/rte_eth_bond_pmd.c
+++ b/drivers/net/bonding/rte_eth_bond_pmd.c
@@ -1705,14 +1705,11 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev,
/* If RSS is enabled for bonding, try to enable it for slaves */
if (bonded_eth_dev->data->dev_conf.rxmode.mq_mode & ETH_MQ_RX_RSS_FLAG) {
- if (internals->rss_key_len != 0) {
- slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len =
+ /* rss_key won't be empty if RSS is configured in bonded dev */
+ slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len =
internals->rss_key_len;
- slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key =
+ slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key =
internals->rss_key;
- } else {
- slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key = NULL;
- }
slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_hf =
bonded_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_hf;
@@ -2234,6 +2231,7 @@ bond_ethdev_info(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
dev_info->flow_type_rss_offloads = internals->flow_type_rss_offloads;
dev_info->reta_size = internals->reta_size;
+ dev_info->hash_key_size = internals->rss_key_len;
return 0;
}
@@ -3023,13 +3021,15 @@ bond_ethdev_rss_hash_update(struct rte_eth_dev *dev,
if (bond_rss_conf.rss_hf != 0)
dev->data->dev_conf.rx_adv_conf.rss_conf.rss_hf = bond_rss_conf.rss_hf;
- if (bond_rss_conf.rss_key && bond_rss_conf.rss_key_len <
- sizeof(internals->rss_key)) {
- if (bond_rss_conf.rss_key_len == 0)
- bond_rss_conf.rss_key_len = 40;
- internals->rss_key_len = bond_rss_conf.rss_key_len;
+ if (bond_rss_conf.rss_key) {
+ if (bond_rss_conf.rss_key_len < internals->rss_key_len)
+ return -EINVAL;
+ else if (bond_rss_conf.rss_key_len > internals->rss_key_len)
+ RTE_BOND_LOG(WARNING, "rss_key will be truncated");
+
memcpy(internals->rss_key, bond_rss_conf.rss_key,
internals->rss_key_len);
+ bond_rss_conf.rss_key_len = internals->rss_key_len;
}
for (i = 0; i < internals->slave_count; i++) {
@@ -3491,14 +3491,24 @@ bond_ethdev_configure(struct rte_eth_dev *dev)
* Fall back to default RSS key if the key is not specified
*/
if (dev->data->dev_conf.rxmode.mq_mode & ETH_MQ_RX_RSS) {
- if (dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key != NULL) {
- internals->rss_key_len =
- dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len;
- memcpy(internals->rss_key,
- dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key,
+ struct rte_eth_rss_conf *rss_conf =
+ &dev->data->dev_conf.rx_adv_conf.rss_conf;
+ if (rss_conf->rss_key != NULL) {
+ if (internals->rss_key_len > rss_conf->rss_key_len) {
+ RTE_BOND_LOG(ERR, "Invalid rss key length(%u)",
+ rss_conf->rss_key_len);
+ return -EINVAL;
+ }
+
+ memcpy(internals->rss_key, rss_conf->rss_key,
internals->rss_key_len);
} else {
- internals->rss_key_len = sizeof(default_rss_key);
+ if (internals->rss_key_len > sizeof(default_rss_key)) {
+ RTE_BOND_LOG(ERR,
+ "There is no suitable default hash key");
+ return -EINVAL;
+ }
+
memcpy(internals->rss_key, default_rss_key,
internals->rss_key_len);
}
--
2.23.0

247
0225-app-testpmd-add-command-to-show-LACP-bonding-info.patch Normal file
View File

@ -0,0 +1,247 @@
From 992fb12f5a2061144190986a1a82c64f9b324e5b Mon Sep 17 00:00:00 2001
From: Chengchang Tang <tangchengchang@huawei.com>
Date: Fri, 24 Sep 2021 17:57:20 +0800
Subject: [PATCH 12/17] app/testpmd: add command to show LACP bonding info
Add a new cmdline to help diagnostic the bonding mode 4 in testpmd.
Show the lacp information about the bonded device and its slaves:
show bonding lacp info <bonded device port_id>
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Acked-by: Xiaoyun Li <xiaoyun.li@intel.com>
---
app/test-pmd/cmdline.c | 184 ++++++++++++++++++++
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 6 +
2 files changed, 190 insertions(+)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index b69c648bf..5691fab94 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -631,6 +631,9 @@ static void cmd_help_long_parsed(void *parsed_result,
"show bonding config (port_id)\n"
" Show the bonding config for port_id.\n\n"
+ "show bonding lacp info (port_id)\n"
+ " Show the bonding lacp information for port_id.\n\n"
+
"set bonding mac_addr (port_id) (address)\n"
" Set the MAC address of a bonded device.\n\n"
@@ -6040,6 +6043,186 @@ cmdline_parse_inst_t cmd_set_balance_xmit_policy = {
}
};
+/* *** SHOW IEEE802.3 BONDING INFORMATION *** */
+struct cmd_show_bonding_lacp_info_result {
+ cmdline_fixed_string_t show;
+ cmdline_fixed_string_t bonding;
+ cmdline_fixed_string_t lacp;
+ cmdline_fixed_string_t info;
+ portid_t port_id;
+};
+
+static void port_param_show(struct port_params *params)
+{
+ char buf[RTE_ETHER_ADDR_FMT_SIZE];
+
+ printf("\t\tsystem priority: %u\n", params->system_priority);
+ rte_ether_format_addr(buf, RTE_ETHER_ADDR_FMT_SIZE, &params->system);
+ printf("\t\tsystem mac address: %s\n", buf);
+ printf("\t\tport key: %u\n", params->key);
+ printf("\t\tport priority: %u\n", params->port_priority);
+ printf("\t\tport number: %u\n", params->port_number);
+}
+
+static void lacp_slave_info_show(struct rte_eth_bond_8023ad_slave_info *info)
+{
+ char a_state[256] = { 0 };
+ char p_state[256] = { 0 };
+ int a_len = 0;
+ int p_len = 0;
+ uint32_t i;
+
+ static const char * const state[] = {
+ "ACTIVE",
+ "TIMEOUT",
+ "AGGREGATION",
+ "SYNCHRONIZATION",
+ "COLLECTING",
+ "DISTRIBUTING",
+ "DEFAULTED",
+ "EXPIRED"
+ };
+ static const char * const selection[] = {
+ "UNSELECTED",
+ "STANDBY",
+ "SELECTED"
+ };
+
+ for (i = 0; i < RTE_DIM(state); i++) {
+ if ((info->actor_state >> i) & 1)
+ a_len += snprintf(&a_state[a_len],
+ RTE_DIM(a_state) - a_len, "%s ",
+ state[i]);
+
+ if ((info->partner_state >> i) & 1)
+ p_len += snprintf(&p_state[p_len],
+ RTE_DIM(p_state) - p_len, "%s ",
+ state[i]);
+ }
+ printf("\tAggregator port id: %u\n", info->agg_port_id);
+ printf("\tselection: %s\n", selection[info->selected]);
+ printf("\tActor detail info:\n");
+ port_param_show(&info->actor);
+ printf("\t\tport state: %s\n", a_state);
+ printf("\tPartner detail info:\n");
+ port_param_show(&info->partner);
+ printf("\t\tport state: %s\n", p_state);
+ printf("\n");
+}
+
+static void lacp_conf_show(struct rte_eth_bond_8023ad_conf *conf)
+{
+ printf("\tfast period: %u ms\n", conf->fast_periodic_ms);
+ printf("\tslow period: %u ms\n", conf->slow_periodic_ms);
+ printf("\tshort timeout: %u ms\n", conf->short_timeout_ms);
+ printf("\tlong timeout: %u ms\n", conf->long_timeout_ms);
+ printf("\taggregate wait timeout: %u ms\n",
+ conf->aggregate_wait_timeout_ms);
+ printf("\ttx period: %u ms\n", conf->tx_period_ms);
+ printf("\trx marker period: %u ms\n", conf->rx_marker_period_ms);
+ printf("\tupdate timeout: %u ms\n", conf->update_timeout_ms);
+ switch (conf->agg_selection) {
+ case AGG_BANDWIDTH:
+ printf("\taggregation mode: bandwidth\n");
+ break;
+ case AGG_STABLE:
+ printf("\taggregation mode: stable\n");
+ break;
+ case AGG_COUNT:
+ printf("\taggregation mode: count\n");
+ break;
+ default:
+ printf("\taggregation mode: invalid\n");
+ break;
+ }
+
+ printf("\n");
+}
+
+static void cmd_show_bonding_lacp_info_parsed(void *parsed_result,
+ __rte_unused struct cmdline *cl,
+ __rte_unused void *data)
+{
+ struct cmd_show_bonding_lacp_info_result *res = parsed_result;
+ struct rte_eth_bond_8023ad_slave_info slave_info;
+ struct rte_eth_bond_8023ad_conf port_conf;
+ portid_t slaves[RTE_MAX_ETHPORTS];
+ portid_t port_id = res->port_id;
+ int num_active_slaves;
+ int bonding_mode;
+ int i;
+ int ret;
+
+ bonding_mode = rte_eth_bond_mode_get(port_id);
+ if (bonding_mode != BONDING_MODE_8023AD) {
+ fprintf(stderr, "\tBonding mode is not mode 4\n");
+ return;
+ }
+
+ num_active_slaves = rte_eth_bond_active_slaves_get(port_id, slaves,
+ RTE_MAX_ETHPORTS);
+ if (num_active_slaves < 0) {
+ fprintf(stderr, "\tFailed to get active slave list for port = %u\n",
+ port_id);
+ return;
+ }
+ if (num_active_slaves == 0)
+ fprintf(stderr, "\tIEEE802.3 port %u has no active slave\n",
+ port_id);
+
+ printf("\tIEEE802.3 port: %u\n", port_id);
+ ret = rte_eth_bond_8023ad_conf_get(port_id, &port_conf);
+ if (ret) {
+ fprintf(stderr, "\tGet bonded device %u info failed\n",
+ port_id);
+ return;
+ }
+ lacp_conf_show(&port_conf);
+
+ for (i = 0; i < num_active_slaves; i++) {
+ ret = rte_eth_bond_8023ad_slave_info(port_id, slaves[i],
+ &slave_info);
+ if (ret) {
+ fprintf(stderr, "\tGet slave device %u info failed\n",
+ slaves[i]);
+ return;
+ }
+ printf("\tSlave Port: %u\n", slaves[i]);
+ lacp_slave_info_show(&slave_info);
+ }
+}
+
+cmdline_parse_token_string_t cmd_show_bonding_lacp_info_show =
+TOKEN_STRING_INITIALIZER(struct cmd_show_bonding_lacp_info_result,
+ show, "show");
+cmdline_parse_token_string_t cmd_show_bonding_lacp_info_bonding =
+TOKEN_STRING_INITIALIZER(struct cmd_show_bonding_lacp_info_result,
+ bonding, "bonding");
+cmdline_parse_token_string_t cmd_show_bonding_lacp_info_lacp =
+TOKEN_STRING_INITIALIZER(struct cmd_show_bonding_lacp_info_result,
+ bonding, "lacp");
+cmdline_parse_token_string_t cmd_show_bonding_lacp_info_info =
+TOKEN_STRING_INITIALIZER(struct cmd_show_bonding_lacp_info_result,
+ info, "info");
+cmdline_parse_token_num_t cmd_show_bonding_lacp_info_port_id =
+TOKEN_NUM_INITIALIZER(struct cmd_show_bonding_lacp_info_result,
+ port_id, RTE_UINT16);
+
+cmdline_parse_inst_t cmd_show_bonding_lacp_info = {
+ .f = cmd_show_bonding_lacp_info_parsed,
+ .help_str = "show bonding lacp info <port_id> : "
+ "Show bonding IEEE802.3 information for port_id",
+ .data = NULL,
+ .tokens = {
+ (void *)&cmd_show_bonding_lacp_info_show,
+ (void *)&cmd_show_bonding_lacp_info_bonding,
+ (void *)&cmd_show_bonding_lacp_info_lacp,
+ (void *)&cmd_show_bonding_lacp_info_info,
+ (void *)&cmd_show_bonding_lacp_info_port_id,
+ NULL
+ }
+};
+
/* *** SHOW NIC BONDING CONFIGURATION *** */
struct cmd_show_bonding_config_result {
cmdline_fixed_string_t show;
@@ -17027,6 +17210,7 @@ cmdline_parse_ctx_t main_ctx[] = {
#ifdef RTE_NET_BOND
(cmdline_parse_inst_t *) &cmd_set_bonding_mode,
(cmdline_parse_inst_t *) &cmd_show_bonding_config,
+ (cmdline_parse_inst_t *) &cmd_show_bonding_lacp_info,
(cmdline_parse_inst_t *) &cmd_set_bonding_primary,
(cmdline_parse_inst_t *) &cmd_add_bonding_slave,
(cmdline_parse_inst_t *) &cmd_remove_bonding_slave,
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index f0e04232a..d5e85b083 100644
--- a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
+++ b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
@@ -2609,6 +2609,12 @@ in balance mode with a transmission policy of layer 2+3::
Active Slaves (3): [1 3 4]
Primary: [3]
+show bonding lacp info
+~~~~~~~~~~~~~~~~~~~~~~
+
+Show information about the Link Bonding device in mode 4 (link-aggregation-802.3ad)::
+
+ testpmd> show bonding lacp info (port_id)
Register Functions
------------------
--
2.23.0

40
0226-app-testpmd-retain-all-original-dev-conf-when-config.patch Normal file
View File

@ -0,0 +1,40 @@
From 65de76da4d8fc270af6bce73334399d0d3c20fa3 Mon Sep 17 00:00:00 2001
From: Huisong Li <lihuisong@huawei.com>
Date: Mon, 11 Oct 2021 17:12:46 +0800
Subject: [PATCH 13/17] app/testpmd: retain all original dev conf when config
DCB
When configuring DCB, testpmd retains the rx_mode/tx_mode configuration in
rte_port->dev_conf. But some configurations, such as the link_speed, were
not saved if they were set before configuring DCB.
Fixes: 1a572499beb6 ("app/testpmd: setup DCB forwarding based on traffic class")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Acked-by: Xiaoyun Li <xiaoyun.li@intel.com>
---
app/test-pmd/testpmd.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index 3098df6c5..0eaa4852d 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -3484,10 +3484,8 @@ init_port_dcb_config(portid_t pid,
rte_port = &ports[pid];
- memset(&port_conf, 0, sizeof(struct rte_eth_conf));
-
- port_conf.rxmode = rte_port->dev_conf.rxmode;
- port_conf.txmode = rte_port->dev_conf.txmode;
+ /* retain the original device configuration. */
+ memcpy(&port_conf, &rte_port->dev_conf, sizeof(struct rte_eth_conf));
/*set configuration of DCB in vt mode and DCB in non-vt mode*/
retval = get_eth_dcb_conf(pid, &port_conf, dcb_mode, num_tcs, pfc_en);
--
2.23.0

593
0227-net-hns3-remove-similar-macro-function-definitions.patch Normal file
View File

@ -0,0 +1,593 @@
From 637fc8040d4aa52eba9c7c78a5c826f4a13c4da0 Mon Sep 17 00:00:00 2001
From: Chengchang Tang <tangchengchang@huawei.com>
Date: Sat, 9 Oct 2021 15:48:05 +0800
Subject: [PATCH 14/17] net/hns3: remove similar macro function definitions
For different capabilities, we declare different macro functions to
determine whether the capabilities are supported.
This patch declare a unified macro function to judge capabilities.
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
drivers/net/hns3/hns3_cmd.c | 6 ++---
drivers/net/hns3/hns3_dcb.c | 4 +--
drivers/net/hns3/hns3_ethdev.c | 24 +++++++++---------
drivers/net/hns3/hns3_ethdev.h | 41 ++-----------------------------
drivers/net/hns3/hns3_ethdev_vf.c | 6 ++---
drivers/net/hns3/hns3_flow.c | 2 +-
drivers/net/hns3/hns3_intr.c | 2 +-
drivers/net/hns3/hns3_ptp.c | 18 +++++++-------
drivers/net/hns3/hns3_rxtx.c | 32 ++++++++++++------------
drivers/net/hns3/hns3_rxtx_vec.c | 4 +--
drivers/net/hns3/hns3_tm.c | 10 ++++----
11 files changed, 56 insertions(+), 93 deletions(-)
diff --git a/drivers/net/hns3/hns3_cmd.c b/drivers/net/hns3/hns3_cmd.c
index cfa943523..6e49108d2 100644
--- a/drivers/net/hns3/hns3_cmd.c
+++ b/drivers/net/hns3/hns3_cmd.c
@@ -619,7 +619,7 @@ hns3_update_dev_lsc_cap(struct hns3_hw *hw, int fw_compact_cmd_result)
static int
hns3_apply_fw_compat_cmd_result(struct hns3_hw *hw, int result)
{
- if (result != 0 && hns3_dev_copper_supported(hw)) {
+ if (result != 0 && hns3_dev_get_support(hw, COPPER)) {
hns3_err(hw, "firmware fails to initialize the PHY, ret = %d.",
result);
return result;
@@ -658,7 +658,7 @@ hns3_firmware_compat_config(struct hns3_hw *hw, bool is_init)
}
if (revision == PCI_REVISION_ID_HIP09_A) {
struct hns3_pf *pf = HNS3_DEV_HW_TO_PF(hw);
- if (hns3_dev_copper_supported(hw) == 0 || pf->is_tmp_phy) {
+ if (hns3_dev_get_support(hw, COPPER) == 0 || pf->is_tmp_phy) {
PMD_INIT_LOG(ERR, "***use temp phy driver in dpdk***");
pf->is_tmp_phy = true;
hns3_set_bit(hw->capability,
@@ -676,7 +676,7 @@ hns3_firmware_compat_config(struct hns3_hw *hw, bool is_init)
if (is_init) {
hns3_set_bit(compat, HNS3_LINK_EVENT_REPORT_EN_B, 1);
hns3_set_bit(compat, HNS3_NCSI_ERROR_REPORT_EN_B, 0);
- if (hns3_dev_copper_supported(hw))
+ if (hns3_dev_get_support(hw, COPPER))
hns3_set_bit(compat, HNS3_FIRMWARE_PHY_DRIVER_EN_B, 1);
}
req->compat = rte_cpu_to_le_32(compat);
diff --git a/drivers/net/hns3/hns3_dcb.c b/drivers/net/hns3/hns3_dcb.c
index b71e2e9ea..8753c340e 100644
--- a/drivers/net/hns3/hns3_dcb.c
+++ b/drivers/net/hns3/hns3_dcb.c
@@ -918,7 +918,7 @@ hns3_dcb_pri_dwrr_cfg(struct hns3_hw *hw)
if (ret)
return ret;
- if (!hns3_dev_dcb_supported(hw))
+ if (!hns3_dev_get_support(hw, DCB))
return 0;
ret = hns3_dcb_ets_tc_dwrr_cfg(hw);
@@ -1368,7 +1368,7 @@ hns3_dcb_pause_setup_hw(struct hns3_hw *hw)
}
/* Only DCB-supported dev supports qset back pressure and pfc cmd */
- if (!hns3_dev_dcb_supported(hw))
+ if (!hns3_dev_get_support(hw, DCB))
return 0;
ret = hns3_pfc_setup_hw(hw);
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index 02d68e496..c5c355d95 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -2408,7 +2408,7 @@ hns3_setup_dcb(struct rte_eth_dev *dev)
struct hns3_hw *hw = &hns->hw;
int ret;
- if (!hns3_dev_dcb_supported(hw)) {
+ if (!hns3_dev_get_support(hw, DCB)) {
hns3_err(hw, "this port does not support dcb configurations.");
return -EOPNOTSUPP;
}
@@ -2746,14 +2746,14 @@ hns3_dev_infos_get(struct rte_eth_dev *eth_dev, struct rte_eth_dev_info *info)
DEV_TX_OFFLOAD_MBUF_FAST_FREE |
hns3_txvlan_cap_get(hw));
- if (hns3_dev_outer_udp_cksum_supported(hw))
+ if (hns3_dev_get_support(hw, OUTER_UDP_CKSUM))
info->tx_offload_capa |= DEV_TX_OFFLOAD_OUTER_UDP_CKSUM;
- if (hns3_dev_indep_txrx_supported(hw))
+ if (hns3_dev_get_support(hw, INDEP_TXRX))
info->dev_capa = RTE_ETH_DEV_CAPA_RUNTIME_RX_QUEUE_SETUP |
RTE_ETH_DEV_CAPA_RUNTIME_TX_QUEUE_SETUP;
- if (hns3_dev_ptp_supported(hw))
+ if (hns3_dev_get_support(hw, PTP))
info->rx_offload_capa |= DEV_RX_OFFLOAD_TIMESTAMP;
info->rx_desc_lim = (struct rte_eth_desc_lim) {
@@ -3418,7 +3418,7 @@ hns3_check_media_type(struct hns3_hw *hw, uint8_t media_type)
switch (media_type) {
case HNS3_MEDIA_TYPE_COPPER:
- if (!hns3_dev_copper_supported(hw)) {
+ if (!hns3_dev_get_support(hw, COPPER)) {
PMD_INIT_LOG(ERR,
"Media type is copper, not supported.");
ret = -EOPNOTSUPP;
@@ -3486,7 +3486,7 @@ hns3_get_board_configuration(struct hns3_hw *hw)
}
/* Dev does not support DCB */
- if (!hns3_dev_dcb_supported(hw)) {
+ if (!hns3_dev_get_support(hw, DCB)) {
pf->tc_max = 1;
pf->pfc_max = 0;
} else
@@ -3799,7 +3799,7 @@ hns3_is_rx_buf_ok(struct hns3_hw *hw, struct hns3_pkt_buf_alloc *buf_alloc,
tc_num = hns3_get_tc_num(hw);
aligned_mps = roundup(pf->mps, HNS3_BUF_SIZE_UNIT);
- if (hns3_dev_dcb_supported(hw))
+ if (hns3_dev_get_support(hw, DCB))
shared_buf_min = HNS3_BUF_MUL_BY * aligned_mps +
pf->dv_buf_size;
else
@@ -3816,7 +3816,7 @@ hns3_is_rx_buf_ok(struct hns3_hw *hw, struct hns3_pkt_buf_alloc *buf_alloc,
shared_buf = rounddown(rx_all - rx_priv, HNS3_BUF_SIZE_UNIT);
buf_alloc->s_buf.buf_size = shared_buf;
- if (hns3_dev_dcb_supported(hw)) {
+ if (hns3_dev_get_support(hw, DCB)) {
buf_alloc->s_buf.self.high = shared_buf - pf->dv_buf_size;
buf_alloc->s_buf.self.low = buf_alloc->s_buf.self.high
- roundup(aligned_mps / HNS3_BUF_DIV_BY,
@@ -3827,7 +3827,7 @@ hns3_is_rx_buf_ok(struct hns3_hw *hw, struct hns3_pkt_buf_alloc *buf_alloc,
buf_alloc->s_buf.self.low = aligned_mps;
}
- if (hns3_dev_dcb_supported(hw)) {
+ if (hns3_dev_get_support(hw, DCB)) {
hi_thrd = shared_buf - pf->dv_buf_size;
if (tc_num <= NEED_RESERVE_TC_NUM)
@@ -4033,7 +4033,7 @@ static int
hns3_rx_buffer_calc(struct hns3_hw *hw, struct hns3_pkt_buf_alloc *buf_alloc)
{
/* When DCB is not supported, rx private buffer is not allocated. */
- if (!hns3_dev_dcb_supported(hw)) {
+ if (!hns3_dev_get_support(hw, DCB)) {
struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw);
struct hns3_pf *pf = &hns->pf;
uint32_t rx_all = pf->pkt_buf_size;
@@ -4261,7 +4261,7 @@ hns3_buffer_alloc(struct hns3_hw *hw)
return ret;
}
- if (hns3_dev_dcb_supported(hw)) {
+ if (hns3_dev_get_support(hw, DCB)) {
ret = hns3_rx_priv_wl_config(hw, &pkt_buf);
if (ret) {
PMD_INIT_LOG(ERR,
@@ -6230,7 +6230,7 @@ hns3_priority_flow_ctrl_set(struct rte_eth_dev *dev,
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
int ret;
- if (!hns3_dev_dcb_supported(hw)) {
+ if (!hns3_dev_get_support(hw, DCB)) {
hns3_err(hw, "This port does not support dcb configurations.");
return -EOPNOTSUPP;
}
diff --git a/drivers/net/hns3/hns3_ethdev.h b/drivers/net/hns3/hns3_ethdev.h
index 57387e05b..94fd14bfc 100644
--- a/drivers/net/hns3/hns3_ethdev.h
+++ b/drivers/net/hns3/hns3_ethdev.h
@@ -883,45 +883,8 @@ enum {
HNS3_DEV_SUPPORT_VF_VLAN_FLT_MOD_B,
};
-#define hns3_dev_dcb_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_DCB_B)
-
-/* Support copper media type */
-#define hns3_dev_copper_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_COPPER_B)
-
-/* Support the queue region action rule of flow directory */
-#define hns3_dev_fd_queue_region_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_FD_QUEUE_REGION_B)
-
-/* Support PTP timestamp offload */
-#define hns3_dev_ptp_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_PTP_B)
-
-/* Support to Independently enable/disable/reset Tx or Rx queues */
-#define hns3_dev_indep_txrx_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_INDEP_TXRX_B)
-
-#define hns3_dev_stash_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_STASH_B)
-
-#define hns3_dev_rxd_adv_layout_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_RXD_ADV_LAYOUT_B)
-
-#define hns3_dev_outer_udp_cksum_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_OUTER_UDP_CKSUM_B)
-
-#define hns3_dev_ras_imp_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_RAS_IMP_B)
-
-#define hns3_dev_tx_push_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_TX_PUSH_B)
-
-#define hns3_dev_tm_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_TM_B)
-
-#define hns3_dev_vf_vlan_flt_supported(hw) \
- hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_VF_VLAN_FLT_MOD_B)
+#define hns3_dev_get_support(hw, _name) \
+ hns3_get_bit((hw)->capability, HNS3_DEV_SUPPORT_##_name##_B)
#define HNS3_DEV_PRIVATE_TO_HW(adapter) \
(&((struct hns3_adapter *)adapter)->hw)
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c b/drivers/net/hns3/hns3_ethdev_vf.c
index e07eb2088..d2895b140 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -988,10 +988,10 @@ hns3vf_dev_infos_get(struct rte_eth_dev *eth_dev, struct rte_eth_dev_info *info)
DEV_TX_OFFLOAD_MBUF_FAST_FREE |
hns3_txvlan_cap_get(hw));
- if (hns3_dev_outer_udp_cksum_supported(hw))
+ if (hns3_dev_get_support(hw, OUTER_UDP_CKSUM))
info->tx_offload_capa |= DEV_TX_OFFLOAD_OUTER_UDP_CKSUM;
- if (hns3_dev_indep_txrx_supported(hw))
+ if (hns3_dev_get_support(hw, INDEP_TXRX))
info->dev_capa = RTE_ETH_DEV_CAPA_RUNTIME_RX_QUEUE_SETUP |
RTE_ETH_DEV_CAPA_RUNTIME_TX_QUEUE_SETUP;
@@ -1623,7 +1623,7 @@ hns3vf_en_vlan_filter(struct hns3_hw *hw, bool enable)
uint8_t msg_data;
int ret;
- if (!hns3_dev_vf_vlan_flt_supported(hw))
+ if (!hns3_dev_get_support(hw, VF_VLAN_FLT_MOD))
return 0;
msg_data = enable ? 1 : 0;
diff --git a/drivers/net/hns3/hns3_flow.c b/drivers/net/hns3/hns3_flow.c
index 6844a5dbe..b25fccbca 100644
--- a/drivers/net/hns3/hns3_flow.c
+++ b/drivers/net/hns3/hns3_flow.c
@@ -301,7 +301,7 @@ hns3_handle_action_queue_region(struct rte_eth_dev *dev,
struct hns3_hw *hw = &hns->hw;
uint16_t idx;
- if (!hns3_dev_fd_queue_region_supported(hw))
+ if (!hns3_dev_get_support(hw, FD_QUEUE_REGION))
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ACTION, action,
"Not support config queue region!");
diff --git a/drivers/net/hns3/hns3_intr.c b/drivers/net/hns3/hns3_intr.c
index 0b307fdd1..3484c76d2 100644
--- a/drivers/net/hns3/hns3_intr.c
+++ b/drivers/net/hns3/hns3_intr.c
@@ -2368,7 +2368,7 @@ hns3_handle_error(struct hns3_adapter *hns)
{
struct hns3_hw *hw = &hns->hw;
- if (hns3_dev_ras_imp_supported(hw)) {
+ if (hns3_dev_get_support(hw, RAS_IMP)) {
hns3_handle_hw_error_v2(hw);
hns3_schedule_reset(hns);
} else {
diff --git a/drivers/net/hns3/hns3_ptp.c b/drivers/net/hns3/hns3_ptp.c
index 146b69db7..14c1ad3e4 100644
--- a/drivers/net/hns3/hns3_ptp.c
+++ b/drivers/net/hns3/hns3_ptp.c
@@ -61,7 +61,7 @@ hns3_ptp_init(struct hns3_hw *hw)
{
int ret;
- if (!hns3_dev_ptp_supported(hw))
+ if (!hns3_dev_get_support(hw, PTP))
return 0;
ret = hns3_ptp_int_en(hw, true);
@@ -120,7 +120,7 @@ hns3_timesync_enable(struct rte_eth_dev *dev)
struct hns3_pf *pf = &hns->pf;
int ret;
- if (!hns3_dev_ptp_supported(hw))
+ if (!hns3_dev_get_support(hw, PTP))
return -ENOTSUP;
if (pf->ptp_enable)
@@ -140,7 +140,7 @@ hns3_timesync_disable(struct rte_eth_dev *dev)
struct hns3_pf *pf = &hns->pf;
int ret;
- if (!hns3_dev_ptp_supported(hw))
+ if (!hns3_dev_get_support(hw, PTP))
return -ENOTSUP;
if (!pf->ptp_enable)
@@ -164,7 +164,7 @@ hns3_timesync_read_rx_timestamp(struct rte_eth_dev *dev,
struct hns3_pf *pf = &hns->pf;
uint64_t ns, sec;
- if (!hns3_dev_ptp_supported(hw))
+ if (!hns3_dev_get_support(hw, PTP))
return -ENOTSUP;
ns = pf->rx_timestamp & TIME_RX_STAMP_NS_MASK;
@@ -190,7 +190,7 @@ hns3_timesync_read_tx_timestamp(struct rte_eth_dev *dev,
uint64_t ns;
int ts_cnt;
- if (!hns3_dev_ptp_supported(hw))
+ if (!hns3_dev_get_support(hw, PTP))
return -ENOTSUP;
ts_cnt = hns3_read_dev(hw, HNS3_TX_1588_BACK_TSP_CNT) &
@@ -219,7 +219,7 @@ hns3_timesync_read_time(struct rte_eth_dev *dev, struct timespec *ts)
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint64_t ns, sec;
- if (!hns3_dev_ptp_supported(hw))
+ if (!hns3_dev_get_support(hw, PTP))
return -ENOTSUP;
sec = hns3_read_dev(hw, HNS3_CURR_TIME_OUT_L);
@@ -240,7 +240,7 @@ hns3_timesync_write_time(struct rte_eth_dev *dev, const struct timespec *ts)
uint64_t sec = ts->tv_sec;
uint64_t ns = ts->tv_nsec;
- if (!hns3_dev_ptp_supported(hw))
+ if (!hns3_dev_get_support(hw, PTP))
return -ENOTSUP;
/* Set the timecounters to a new value. */
@@ -261,7 +261,7 @@ hns3_timesync_adjust_time(struct rte_eth_dev *dev, int64_t delta)
struct timespec cur_time;
uint64_t ns;
- if (!hns3_dev_ptp_supported(hw))
+ if (!hns3_dev_get_support(hw, PTP))
return -ENOTSUP;
(void)hns3_timesync_read_time(dev, &cur_time);
@@ -280,7 +280,7 @@ hns3_restore_ptp(struct hns3_adapter *hns)
bool en = pf->ptp_enable;
int ret;
- if (!hns3_dev_ptp_supported(hw))
+ if (!hns3_dev_get_support(hw, PTP))
return 0;
ret = hns3_timesync_configure(hns, en);
diff --git a/drivers/net/hns3/hns3_rxtx.c b/drivers/net/hns3/hns3_rxtx.c
index 80d2614d2..bb1723e29 100644
--- a/drivers/net/hns3/hns3_rxtx.c
+++ b/drivers/net/hns3/hns3_rxtx.c
@@ -381,7 +381,7 @@ hns3_enable_all_queues(struct hns3_hw *hw, bool en)
int i;
for (i = 0; i < hw->cfg_max_queues; i++) {
- if (hns3_dev_indep_txrx_supported(hw)) {
+ if (hns3_dev_get_support(hw, INDEP_TXRX)) {
rxq = i < nb_rx_q ? hw->data->rx_queues[i] : NULL;
txq = i < nb_tx_q ? hw->data->tx_queues[i] : NULL;
@@ -426,7 +426,7 @@ hns3_enable_txq(struct hns3_tx_queue *txq, bool en)
struct hns3_hw *hw = &txq->hns->hw;
uint32_t reg;
- if (hns3_dev_indep_txrx_supported(hw)) {
+ if (hns3_dev_get_support(hw, INDEP_TXRX)) {
reg = hns3_read_dev(txq, HNS3_RING_TX_EN_REG);
if (en)
reg |= BIT(HNS3_RING_EN_B);
@@ -443,7 +443,7 @@ hns3_enable_rxq(struct hns3_rx_queue *rxq, bool en)
struct hns3_hw *hw = &rxq->hns->hw;
uint32_t reg;
- if (hns3_dev_indep_txrx_supported(hw)) {
+ if (hns3_dev_get_support(hw, INDEP_TXRX)) {
reg = hns3_read_dev(rxq, HNS3_RING_RX_EN_REG);
if (en)
reg |= BIT(HNS3_RING_EN_B);
@@ -1618,7 +1618,7 @@ hns3_set_fake_rx_or_tx_queues(struct rte_eth_dev *dev, uint16_t nb_rx_q,
uint16_t q;
int ret;
- if (hns3_dev_indep_txrx_supported(hw))
+ if (hns3_dev_get_support(hw, INDEP_TXRX))
return 0;
/* Setup new number of fake RX/TX queues and reconfigure device. */
@@ -1862,7 +1862,7 @@ hns3_rx_queue_setup(struct rte_eth_dev *dev, uint16_t idx, uint16_t nb_desc,
conf->rx_free_thresh : HNS3_DEFAULT_RX_FREE_THRESH;
rxq->rx_deferred_start = conf->rx_deferred_start;
- if (rxq->rx_deferred_start && !hns3_dev_indep_txrx_supported(hw)) {
+ if (rxq->rx_deferred_start && !hns3_dev_get_support(hw, INDEP_TXRX)) {
hns3_warn(hw, "deferred start is not supported.");
rxq->rx_deferred_start = false;
}
@@ -1898,7 +1898,7 @@ hns3_rx_queue_setup(struct rte_eth_dev *dev, uint16_t idx, uint16_t nb_desc,
HNS3_PORT_BASE_VLAN_ENABLE;
else
rxq->pvid_sw_discard_en = false;
- rxq->ptype_en = hns3_dev_rxd_adv_layout_supported(hw) ? true : false;
+ rxq->ptype_en = hns3_dev_get_support(hw, RXD_ADV_LAYOUT) ? true : false;
rxq->configured = true;
rxq->io_base = (void *)((char *)hw->io_base + HNS3_TQP_REG_OFFSET +
idx * HNS3_TQP_REG_SIZE);
@@ -2026,7 +2026,7 @@ hns3_dev_supported_ptypes_get(struct rte_eth_dev *dev)
dev->rx_pkt_burst == hns3_recv_scattered_pkts ||
dev->rx_pkt_burst == hns3_recv_pkts_vec ||
dev->rx_pkt_burst == hns3_recv_pkts_vec_sve) {
- if (hns3_dev_rxd_adv_layout_supported(hw))
+ if (hns3_dev_get_support(hw, RXD_ADV_LAYOUT))
return adv_layout_ptypes;
else
return ptypes;
@@ -2928,7 +2928,7 @@ hns3_tx_push_init(struct rte_eth_dev *dev)
volatile uint32_t *reg;
uint32_t val;
- if (!hns3_dev_tx_push_supported(hw))
+ if (!hns3_dev_get_support(hw, TX_PUSH))
return;
reg = (volatile uint32_t *)hns3_tx_push_get_queue_tail_reg(dev, 0);
@@ -2949,7 +2949,7 @@ hns3_tx_push_queue_init(struct rte_eth_dev *dev,
struct hns3_tx_queue *txq)
{
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
- if (!hns3_dev_tx_push_supported(hw)) {
+ if (!hns3_dev_get_support(hw, TX_PUSH)) {
txq->tx_push_enable = false;
return;
}
@@ -2994,7 +2994,7 @@ hns3_tx_queue_setup(struct rte_eth_dev *dev, uint16_t idx, uint16_t nb_desc,
}
txq->tx_deferred_start = conf->tx_deferred_start;
- if (txq->tx_deferred_start && !hns3_dev_indep_txrx_supported(hw)) {
+ if (txq->tx_deferred_start && !hns3_dev_get_support(hw, INDEP_TXRX)) {
hns3_warn(hw, "deferred start is not supported.");
txq->tx_deferred_start = false;
}
@@ -4276,7 +4276,7 @@ hns3_tx_check_simple_support(struct rte_eth_dev *dev)
uint64_t offloads = dev->data->dev_conf.txmode.offloads;
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
- if (hns3_dev_ptp_supported(hw))
+ if (hns3_dev_get_support(hw, PTP))
return false;
return (offloads == (offloads & DEV_TX_OFFLOAD_MBUF_FAST_FREE));
@@ -4437,7 +4437,7 @@ hns3_dev_rx_queue_start(struct rte_eth_dev *dev, uint16_t rx_queue_id)
struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw);
int ret;
- if (!hns3_dev_indep_txrx_supported(hw))
+ if (!hns3_dev_get_support(hw, INDEP_TXRX))
return -ENOTSUP;
rte_spinlock_lock(&hw->lock);
@@ -4483,7 +4483,7 @@ hns3_dev_rx_queue_stop(struct rte_eth_dev *dev, uint16_t rx_queue_id)
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct hns3_rx_queue *rxq = dev->data->rx_queues[rx_queue_id];
- if (!hns3_dev_indep_txrx_supported(hw))
+ if (!hns3_dev_get_support(hw, INDEP_TXRX))
return -ENOTSUP;
rte_spinlock_lock(&hw->lock);
@@ -4505,7 +4505,7 @@ hns3_dev_tx_queue_start(struct rte_eth_dev *dev, uint16_t tx_queue_id)
struct hns3_tx_queue *txq = dev->data->tx_queues[tx_queue_id];
int ret;
- if (!hns3_dev_indep_txrx_supported(hw))
+ if (!hns3_dev_get_support(hw, INDEP_TXRX))
return -ENOTSUP;
rte_spinlock_lock(&hw->lock);
@@ -4531,7 +4531,7 @@ hns3_dev_tx_queue_stop(struct rte_eth_dev *dev, uint16_t tx_queue_id)
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct hns3_tx_queue *txq = dev->data->tx_queues[tx_queue_id];
- if (!hns3_dev_indep_txrx_supported(hw))
+ if (!hns3_dev_get_support(hw, INDEP_TXRX))
return -ENOTSUP;
rte_spinlock_lock(&hw->lock);
@@ -4704,7 +4704,7 @@ hns3_enable_rxd_adv_layout(struct hns3_hw *hw)
* If the hardware support rxd advanced layout, then driver enable it
* default.
*/
- if (hns3_dev_rxd_adv_layout_supported(hw))
+ if (hns3_dev_get_support(hw, RXD_ADV_LAYOUT))
hns3_write_dev(hw, HNS3_RXD_ADV_LAYOUT_EN_REG, 1);
}
diff --git a/drivers/net/hns3/hns3_rxtx_vec.c b/drivers/net/hns3/hns3_rxtx_vec.c
index 15a0bd075..bfe84e833 100644
--- a/drivers/net/hns3/hns3_rxtx_vec.c
+++ b/drivers/net/hns3/hns3_rxtx_vec.c
@@ -19,7 +19,7 @@ hns3_tx_check_vec_support(struct rte_eth_dev *dev)
struct rte_eth_txmode *txmode = &dev->data->dev_conf.txmode;
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
- if (hns3_dev_ptp_supported(hw))
+ if (hns3_dev_get_support(hw, PTP))
return -ENOTSUP;
/* Only support DEV_TX_OFFLOAD_MBUF_FAST_FREE */
@@ -234,7 +234,7 @@ hns3_rx_check_vec_support(struct rte_eth_dev *dev)
DEV_RX_OFFLOAD_VLAN;
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
- if (hns3_dev_ptp_supported(hw))
+ if (hns3_dev_get_support(hw, PTP))
return -ENOTSUP;
if (dev->data->scattered_rx)
diff --git a/drivers/net/hns3/hns3_tm.c b/drivers/net/hns3/hns3_tm.c
index db5ac786c..44b607af7 100644
--- a/drivers/net/hns3/hns3_tm.c
+++ b/drivers/net/hns3/hns3_tm.c
@@ -31,7 +31,7 @@ hns3_tm_conf_init(struct rte_eth_dev *dev)
struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t max_tx_queues = hns3_tm_max_tx_queues_get(dev);
- if (!hns3_dev_tm_supported(hw))
+ if (!hns3_dev_get_support(hw, TM))
return;
pf->tm_conf.nb_leaf_nodes_max = max_tx_queues;
@@ -58,7 +58,7 @@ hns3_tm_conf_uninit(struct rte_eth_dev *dev)
struct hns3_tm_shaper_profile *shaper_profile;
struct hns3_tm_node *tm_node;
- if (!hns3_dev_tm_supported(hw))
+ if (!hns3_dev_get_support(hw, TM))
return;
if (pf->tm_conf.nb_queue_node > 0) {
@@ -1233,7 +1233,7 @@ hns3_tm_ops_get(struct rte_eth_dev *dev, void *arg)
if (arg == NULL)
return -EINVAL;
- if (!hns3_dev_tm_supported(hw))
+ if (!hns3_dev_get_support(hw, TM))
return -EOPNOTSUPP;
*(const void **)arg = &hns3_tm_ops;
@@ -1246,7 +1246,7 @@ hns3_tm_dev_start_proc(struct hns3_hw *hw)
{
struct hns3_pf *pf = HNS3_DEV_HW_TO_PF(hw);
- if (!hns3_dev_tm_supported(hw))
+ if (!hns3_dev_get_support(hw, TM))
return;
if (pf->tm_conf.root && !pf->tm_conf.committed)
@@ -1295,7 +1295,7 @@ hns3_tm_conf_update(struct hns3_hw *hw)
struct hns3_pf *pf = HNS3_DEV_HW_TO_PF(hw);
struct rte_tm_error error;
- if (!hns3_dev_tm_supported(hw))
+ if (!hns3_dev_get_support(hw, TM))
return 0;
if (pf->tm_conf.root == NULL || !pf->tm_conf.committed)
--
2.23.0

33
0228-net-hns3-fix-interrupt-vector-freeing.patch Normal file
View File

@ -0,0 +1,33 @@
From b00fefaae559f49fbbbc39ec6ec1aaa1f4f5ba39 Mon Sep 17 00:00:00 2001
From: Chengwen Feng <fengchengwen@huawei.com>
Date: Wed, 13 Oct 2021 16:09:08 +0800
Subject: [PATCH 15/17] net/hns3: fix interrupt vector freeing
The intr_handle->intr_vec is allocated by rte_zmalloc(), but freed by
free(), this patch fixes it.
Fixes: 02a7b55657b2 ("net/hns3: support Rx interrupt")
Cc: stable@dpdk.org
Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
---
drivers/net/hns3/hns3_ethdev_vf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c b/drivers/net/hns3/hns3_ethdev_vf.c
index d2895b140..9dfc22d2d 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -2355,7 +2355,7 @@ hns3vf_map_rx_interrupt(struct rte_eth_dev *dev)
return 0;
vf_bind_vector_error:
- free(intr_handle->intr_vec);
+ rte_free(intr_handle->intr_vec);
intr_handle->intr_vec = NULL;
vf_alloc_intr_vec_error:
rte_intr_efd_disable(intr_handle);
--
2.23.0

179
0229-net-hns3-add-runtime-config-for-mailbox-limit-time.patch Normal file
View File

@ -0,0 +1,179 @@
From 61c8349bbed069317c59da812598b74d2e076ced Mon Sep 17 00:00:00 2001
From: Chengchang Tang <tangchengchang@huawei.com>
Date: Fri, 22 Oct 2021 09:38:40 +0800
Subject: [PATCH 16/17] net/hns3: add runtime config for mailbox limit time
Current, the max waiting time for MBX response is 500ms, but in
some scenarios, it is not enough. Since it depends on the response
of the kernel mode driver, and its response time is related to the
scheduling of the system. In this special scenario, most of the
cores are isolated, and only a few cores are used for system
scheduling. When a large number of services are started, the
scheduling of the system will be very busy, and the reply of the
mbx message will time out, which will cause our PMD initialization
to fail.
This patch add a runtime config to set the max wait time. For the
above scenes, users can adjust the waiting time to a suitable value
by themselves.
Fixes: 463e748964f5 ("net/hns3: support mailbox")
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_ethdev.c | 32 ++++++++++++++++++++++++++++++-
drivers/net/hns3/hns3_ethdev.h | 3 +++
drivers/net/hns3/hns3_ethdev_vf.c | 3 ++-
drivers/net/hns3/hns3_mbx.c | 8 +++++---
drivers/net/hns3/hns3_mbx.h | 1 +
5 files changed, 42 insertions(+), 5 deletions(-)
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index c5c355d95..2ae4cb9b7 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -7348,9 +7348,30 @@ hns3_parse_dev_caps_mask(const char *key, const char *value, void *extra_args)
return 0;
}
+static int
+hns3_parse_mbx_time_limit(const char *key, const char *value, void *extra_args)
+{
+ uint32_t val;
+
+ RTE_SET_USED(key);
+
+ val = strtoul(value, NULL, 10);
+
+ /*
+ * 500ms is empirical value in process of mailbox communication. If
+ * the delay value is set to one lower thanthe empirical value, mailbox
+ * communication may fail.
+ */
+ if (val > HNS3_MBX_DEF_TIME_LIMIT_MS && val <= UINT16_MAX)
+ *(uint16_t *)extra_args = val;
+
+ return 0;
+}
+
void
hns3_parse_devargs(struct rte_eth_dev *dev)
{
+ uint16_t mbx_time_limit_ms = HNS3_MBX_DEF_TIME_LIMIT_MS;
struct hns3_adapter *hns = dev->data->dev_private;
uint32_t rx_func_hint = HNS3_IO_FUNC_HINT_NONE;
uint32_t tx_func_hint = HNS3_IO_FUNC_HINT_NONE;
@@ -7371,6 +7392,9 @@ hns3_parse_devargs(struct rte_eth_dev *dev)
&hns3_parse_io_hint_func, &tx_func_hint);
(void)rte_kvargs_process(kvlist, HNS3_DEVARG_DEV_CAPS_MASK,
&hns3_parse_dev_caps_mask, &dev_caps_mask);
+ (void)rte_kvargs_process(kvlist, HNS3_DEVARG_MBX_TIME_LIMIT_MS,
+ &hns3_parse_mbx_time_limit, &mbx_time_limit_ms);
+
rte_kvargs_free(kvlist);
if (rx_func_hint != HNS3_IO_FUNC_HINT_NONE)
@@ -7386,6 +7410,11 @@ hns3_parse_devargs(struct rte_eth_dev *dev)
hns3_warn(hw, "parsed %s = 0x%" PRIx64 ".",
HNS3_DEVARG_DEV_CAPS_MASK, dev_caps_mask);
hns->dev_caps_mask = dev_caps_mask;
+
+ if (mbx_time_limit_ms != HNS3_MBX_DEF_TIME_LIMIT_MS)
+ hns3_warn(hw, "parsed %s = %u.", HNS3_DEVARG_MBX_TIME_LIMIT_MS,
+ mbx_time_limit_ms);
+ hns->mbx_time_limit_ms = mbx_time_limit_ms;
}
static const struct eth_dev_ops hns3_eth_dev_ops = {
@@ -7642,6 +7671,7 @@ RTE_PMD_REGISTER_KMOD_DEP(net_hns3, "* igb_uio | vfio-pci");
RTE_PMD_REGISTER_PARAM_STRING(net_hns3,
HNS3_DEVARG_RX_FUNC_HINT "=vec|sve|simple|common "
HNS3_DEVARG_TX_FUNC_HINT "=vec|sve|simple|common "
- HNS3_DEVARG_DEV_CAPS_MASK "=<1-65535> ");
+ HNS3_DEVARG_DEV_CAPS_MASK "=<1-65535> "
+ HNS3_DEVARG_MBX_TIME_LIMIT_MS "=<uint16> ");
RTE_LOG_REGISTER(hns3_logtype_init, pmd.net.hns3.init, NOTICE);
RTE_LOG_REGISTER(hns3_logtype_driver, pmd.net.hns3.driver, NOTICE);
diff --git a/drivers/net/hns3/hns3_ethdev.h b/drivers/net/hns3/hns3_ethdev.h
index 94fd14bfc..84f5a9f29 100644
--- a/drivers/net/hns3/hns3_ethdev.h
+++ b/drivers/net/hns3/hns3_ethdev.h
@@ -851,6 +851,7 @@ struct hns3_adapter {
uint32_t tx_func_hint;
uint64_t dev_caps_mask;
+ uint16_t mbx_time_limit_ms; /* wait time for mbx message */
struct hns3_ptype_table ptype_tbl __rte_cache_aligned;
};
@@ -868,6 +869,8 @@ enum {
#define HNS3_DEVARG_DEV_CAPS_MASK "dev_caps_mask"
+#define HNS3_DEVARG_MBX_TIME_LIMIT_MS "mbx_time_limit_ms"
+
enum {
HNS3_DEV_SUPPORT_DCB_B,
HNS3_DEV_SUPPORT_COPPER_B,
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c b/drivers/net/hns3/hns3_ethdev_vf.c
index 9dfc22d2d..29313c2f7 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -3112,4 +3112,5 @@ RTE_PMD_REGISTER_KMOD_DEP(net_hns3_vf, "* igb_uio | vfio-pci");
RTE_PMD_REGISTER_PARAM_STRING(net_hns3_vf,
HNS3_DEVARG_RX_FUNC_HINT "=vec|sve|simple|common "
HNS3_DEVARG_TX_FUNC_HINT "=vec|sve|simple|common "
- HNS3_DEVARG_DEV_CAPS_MASK "=<1-65535> ");
+ HNS3_DEVARG_DEV_CAPS_MASK "=<1-65535> "
+ HNS3_DEVARG_MBX_TIME_LIMIT_MS "=<uint16_t> ");
diff --git a/drivers/net/hns3/hns3_mbx.c b/drivers/net/hns3/hns3_mbx.c
index 411c5ebe9..a4d9afc45 100644
--- a/drivers/net/hns3/hns3_mbx.c
+++ b/drivers/net/hns3/hns3_mbx.c
@@ -61,8 +61,9 @@ static int
hns3_get_mbx_resp(struct hns3_hw *hw, uint16_t code, uint16_t subcode,
uint8_t *resp_data, uint16_t resp_len)
{
-#define HNS3_MAX_RETRY_US 500000
#define HNS3_WAIT_RESP_US 100
+#define US_PER_MS 1000
+ uint32_t mbx_time_limit;
struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw);
struct hns3_mbx_resp_status *mbx_resp;
uint32_t wait_time = 0;
@@ -74,7 +75,8 @@ hns3_get_mbx_resp(struct hns3_hw *hw, uint16_t code, uint16_t subcode,
return -EINVAL;
}
- while (wait_time < HNS3_MAX_RETRY_US) {
+ mbx_time_limit = (uint32_t)hns->mbx_time_limit_ms * US_PER_MS;
+ while (wait_time < mbx_time_limit) {
if (__atomic_load_n(&hw->reset.disable_cmd, __ATOMIC_RELAXED)) {
hns3_err(hw, "Don't wait for mbx respone because of "
"disable_cmd");
@@ -103,7 +105,7 @@ hns3_get_mbx_resp(struct hns3_hw *hw, uint16_t code, uint16_t subcode,
wait_time += HNS3_WAIT_RESP_US;
}
hw->mbx_resp.req_msg_data = 0;
- if (wait_time >= HNS3_MAX_RETRY_US) {
+ if (wait_time >= mbx_time_limit) {
hns3_mbx_proc_timeout(hw, code, subcode);
return -ETIME;
}
diff --git a/drivers/net/hns3/hns3_mbx.h b/drivers/net/hns3/hns3_mbx.h
index f868e33a9..d637bd2b2 100644
--- a/drivers/net/hns3/hns3_mbx.h
+++ b/drivers/net/hns3/hns3_mbx.h
@@ -87,6 +87,7 @@ enum hns3_mbx_link_fail_subcode {
#define HNS3_MBX_MAX_MSG_SIZE 16
#define HNS3_MBX_MAX_RESP_DATA_SIZE 8
+#define HNS3_MBX_DEF_TIME_LIMIT_MS 500
enum {
HNS3_MBX_RESP_MATCHING_SCHEME_OF_ORIGINAL = 0,
--
2.23.0

39
0230-net-hns3-fix-mailbox-communication-with-HW.patch Normal file
View File

@ -0,0 +1,39 @@
From a277f7dbaa54e8ea10f41a7dc4dadec5f08b61b3 Mon Sep 17 00:00:00 2001
From: "Min Hu (Connor)" <humin29@huawei.com>
Date: Thu, 28 Oct 2021 19:52:30 +0800
Subject: [PATCH 17/17] net/hns3: fix mailbox communication with HW
Mailbox is the communication mechanism between SW and HW. There exist
two approaches for SW to recognize mailbox message from HW. One way is
using match_id, the other is to compare the message code. The two
approaches are independent and used in different scenarios.
But for the second approach, "next_to_use" should be updated and written
to HW register. If it not done, HW do not know the position SW steps,
then, the communication between SW and HW will turn to be failed.
Fixes: dbbbad23e380 ("net/hns3: fix VF handling LSC event in secondary process")
Cc: stable@dpdk.org
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
drivers/net/hns3/hns3_mbx.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/hns3/hns3_mbx.c b/drivers/net/hns3/hns3_mbx.c
index a4d9afc45..3ad85e721 100644
--- a/drivers/net/hns3/hns3_mbx.c
+++ b/drivers/net/hns3/hns3_mbx.c
@@ -435,6 +435,9 @@ hns3_handle_mbx_msg_out_intr(struct hns3_hw *hw)
scan_next:
next_to_use = (next_to_use + 1) % hw->cmq.crq.desc_num;
}
+
+ crq->next_to_use = next_to_use;
+ hns3_write_dev(hw, HNS3_CMDQ_RX_HEAD_REG, crq->next_to_use);
}
void
--
2.23.0

421
0231-app-testpmd-support-multi-process.patch Normal file
View File

@ -0,0 +1,421 @@
From c7a841ce328cda8338494640b103d5182268fd1e Mon Sep 17 00:00:00 2001
From: "Min Hu (Connor)" <humin29@huawei.com>
Date: Wed, 25 Aug 2021 10:06:38 +0800
Subject: [PATCH] app/testpmd: support multi-process
This patch adds multi-process support for testpmd.
For example the following commands run two testpmd
processes:
* the primary process:
./dpdk-testpmd --proc-type=auto -l 0-1 -- -i \
--rxq=4 --txq=4 --num-procs=2 --proc-id=0
* the secondary process:
./dpdk-testpmd --proc-type=auto -l 2-3 -- -i \
--rxq=4 --txq=4 --num-procs=2 --proc-id=1
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Xiaoyun Li <xiaoyun.li@intel.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Aman Deep Singh <aman.deep.singh@intel.com>
---
app/test-pmd/cmdline.c | 6 ++
app/test-pmd/config.c | 20 +++++-
app/test-pmd/parameters.c | 9 +++
app/test-pmd/testpmd.c | 92 +++++++++++++++++++++++----
app/test-pmd/testpmd.h | 11 ++++
doc/guides/testpmd_app_ug/run_app.rst | 84 ++++++++++++++++++++++++
6 files changed, 210 insertions(+), 12 deletions(-)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 5691fab94..b701129d8 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -5441,6 +5441,12 @@ cmd_set_flush_rx_parsed(void *parsed_result,
__rte_unused void *data)
{
struct cmd_set_flush_rx *res = parsed_result;
+
+ if (num_procs > 1 && (strcmp(res->mode, "on") == 0)) {
+ printf("multi-process doesn't support to flush Rx queues.\n");
+ return;
+ }
+
no_flush_rx = (uint8_t)((strcmp(res->mode, "on") == 0) ? 0 : 1);
}
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 7af13f65c..0d6639020 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -3117,6 +3117,8 @@ rss_fwd_config_setup(void)
queueid_t rxq;
queueid_t nb_q;
streamid_t sm_id;
+ int start;
+ int end;
nb_q = nb_rxq;
if (nb_q > nb_txq)
@@ -3134,7 +3136,21 @@ rss_fwd_config_setup(void)
init_fwd_streams();
setup_fwd_config_of_each_lcore(&cur_fwd_config);
- rxp = 0; rxq = 0;
+
+ if (proc_id > 0 && nb_q % num_procs != 0)
+ printf("Warning! queue numbers should be multiple of processes, or packet loss will happen.\n");
+
+ /**
+ * In multi-process, All queues are allocated to different
+ * processes based on num_procs and proc_id. For example:
+ * if supports 4 queues(nb_q), 2 processes(num_procs),
+ * the 0~1 queue for primary process.
+ * the 2~3 queue for secondary process.
+ */
+ start = proc_id * nb_q / num_procs;
+ end = start + nb_q / num_procs;
+ rxp = 0;
+ rxq = start;
for (sm_id = 0; sm_id < cur_fwd_config.nb_fwd_streams; sm_id++) {
struct fwd_stream *fs;
@@ -3151,6 +3167,8 @@ rss_fwd_config_setup(void)
continue;
rxp = 0;
rxq++;
+ if (rxq >= end)
+ rxq = start;
}
}
diff --git a/app/test-pmd/parameters.c b/app/test-pmd/parameters.c
index 414a0068f..c464c42f6 100644
--- a/app/test-pmd/parameters.c
+++ b/app/test-pmd/parameters.c
@@ -487,6 +487,9 @@ parse_event_printing_config(const char *optarg, int enable)
void
launch_args_parse(int argc, char** argv)
{
+#define PARAM_PROC_ID "proc-id"
+#define PARAM_NUM_PROCS "num-procs"
+
int n, opt;
char **argvopt;
int opt_idx;
@@ -603,6 +606,8 @@ launch_args_parse(int argc, char** argv)
{ "rx-mq-mode", 1, 0, 0 },
{ "record-core-cycles", 0, 0, 0 },
{ "record-burst-stats", 0, 0, 0 },
+ { PARAM_NUM_PROCS, 1, 0, 0 },
+ { PARAM_PROC_ID, 1, 0, 0 },
{ 0, 0, 0, 0 },
};
@@ -1359,6 +1364,10 @@ launch_args_parse(int argc, char** argv)
record_core_cycles = 1;
if (!strcmp(lgopts[opt_idx].name, "record-burst-stats"))
record_burst_stats = 1;
+ if (!strcmp(lgopts[opt_idx].name, PARAM_NUM_PROCS))
+ num_procs = atoi(optarg);
+ if (!strcmp(lgopts[opt_idx].name, PARAM_PROC_ID))
+ proc_id = atoi(optarg);
break;
case 'h':
usage(argv[0]);
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index 0eaa4852d..983d8827d 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -505,6 +505,61 @@ uint8_t gro_flush_cycles = GRO_DEFAULT_FLUSH_CYCLES;
* hexadecimal bitmask of RX mq mode can be enabled.
*/
enum rte_eth_rx_mq_mode rx_mq_mode = ETH_MQ_RX_VMDQ_DCB_RSS;
+/*
+ * ID of the current process in multi-process, used to
+ * configure the queues to be polled.
+ */
+int proc_id;
+
+/*
+ * Number of processes in multi-process, used to
+ * configure the queues to be polled.
+ */
+unsigned int num_procs = 1;
+
+static int
+eth_dev_configure_mp(uint16_t port_id, uint16_t nb_rx_q, uint16_t nb_tx_q,
+ const struct rte_eth_conf *dev_conf)
+{
+ if (is_proc_primary())
+ return rte_eth_dev_configure(port_id, nb_rx_q, nb_tx_q,
+ dev_conf);
+ return 0;
+}
+
+static int
+eth_dev_start_mp(uint16_t port_id)
+{
+ if (is_proc_primary())
+ return rte_eth_dev_start(port_id);
+
+ return 0;
+}
+
+static int
+eth_dev_stop_mp(uint16_t port_id)
+{
+ if (is_proc_primary())
+ return rte_eth_dev_stop(port_id);
+
+ return 0;
+}
+
+static void
+mempool_free_mp(struct rte_mempool *mp)
+{
+ if (is_proc_primary())
+ rte_mempool_free(mp);
+}
+
+static int
+eth_dev_set_mtu_mp(uint16_t port_id, uint16_t mtu)
+{
+ if (is_proc_primary())
+ return rte_eth_dev_set_mtu(port_id, mtu);
+
+ return 0;
+}
/* Forward function declarations */
static void setup_attached_port(portid_t pi);
@@ -964,6 +1019,14 @@ mbuf_pool_create(uint16_t mbuf_seg_size, unsigned nb_mbuf,
mb_size = sizeof(struct rte_mbuf) + mbuf_seg_size;
mbuf_poolname_build(socket_id, pool_name, sizeof(pool_name), size_idx);
+ if (!is_proc_primary()) {
+ rte_mp = rte_mempool_lookup(pool_name);
+ if (rte_mp == NULL)
+ rte_exit(EXIT_FAILURE,
+ "Get mbuf pool for socket %u failed: %s\n",
+ socket_id, rte_strerror(rte_errno));
+ return rte_mp;
+ }
TESTPMD_LOG(INFO,
"create a new mbuf pool <%s>: n=%u, size=%u, socket=%u\n",
@@ -1969,6 +2032,11 @@ flush_fwd_rx_queues(void)
uint64_t prev_tsc = 0, diff_tsc, cur_tsc, timer_tsc = 0;
uint64_t timer_period;
+ if (num_procs > 1) {
+ printf("multi-process not support for flushing fwd Rx queues, skip the below lines and return.\n");
+ return;
+ }
+
/* convert to number of cycles */
timer_period = rte_get_timer_hz(); /* 1 second timeout */
@@ -2456,7 +2524,7 @@ start_port(portid_t pid)
return -1;
}
/* configure port */
- diag = rte_eth_dev_configure(pi, nb_rxq + nb_hairpinq,
+ diag = eth_dev_configure_mp(pi, nb_rxq + nb_hairpinq,
nb_txq + nb_hairpinq,
&(port->dev_conf));
if (diag != 0) {
@@ -2470,7 +2538,7 @@ start_port(portid_t pid)
return -1;
}
}
- if (port->need_reconfig_queues > 0) {
+ if (port->need_reconfig_queues > 0 && is_proc_primary()) {
port->need_reconfig_queues = 0;
/* setup tx queues */
for (qi = 0; qi < nb_txq; qi++) {
@@ -2571,7 +2639,7 @@ start_port(portid_t pid)
cnt_pi++;
/* start port */
- if (rte_eth_dev_start(pi) < 0) {
+ if (eth_dev_start_mp(pi) < 0) {
printf("Fail to start port %d\n", pi);
/* Fail to setup rx queue, return */
@@ -2700,7 +2768,7 @@ stop_port(portid_t pid)
}
}
- if (rte_eth_dev_stop(pi) != 0)
+ if (eth_dev_stop_mp(pi) != 0)
RTE_LOG(ERR, EAL, "rte_eth_dev_stop failed for port %u\n",
pi);
@@ -2769,8 +2837,10 @@ close_port(portid_t pid)
continue;
}
- port_flow_flush(pi);
- rte_eth_dev_close(pi);
+ if (is_proc_primary()) {
+ port_flow_flush(pi);
+ rte_eth_dev_close(pi);
+ }
}
remove_invalid_ports();
@@ -3035,7 +3105,7 @@ pmd_test_exit(void)
}
for (i = 0 ; i < RTE_DIM(mempools) ; i++) {
if (mempools[i])
- rte_mempool_free(mempools[i]);
+ mempool_free_mp(mempools[i]);
}
printf("\nBye...\n");
@@ -3482,6 +3552,10 @@ init_port_dcb_config(portid_t pid,
int retval;
uint16_t i;
+ if (num_procs > 1) {
+ printf("The multi-process feature doesn't support dcb.\n");
+ return -ENOTSUP;
+ }
rte_port = &ports[pid];
/* retain the original device configuration. */
@@ -3646,10 +3720,6 @@ main(int argc, char** argv)
rte_exit(EXIT_FAILURE, "Cannot init EAL: %s\n",
rte_strerror(rte_errno));
- if (rte_eal_process_type() == RTE_PROC_SECONDARY)
- rte_exit(EXIT_FAILURE,
- "Secondary process type not supported.\n");
-
ret = register_eth_event_callback();
if (ret != 0)
rte_exit(EXIT_FAILURE, "Cannot register for ethdev events");
diff --git a/app/test-pmd/testpmd.h b/app/test-pmd/testpmd.h
index 303bed830..122fca29c 100644
--- a/app/test-pmd/testpmd.h
+++ b/app/test-pmd/testpmd.h
@@ -626,6 +626,17 @@ extern struct mplsoudp_decap_conf mplsoudp_decap_conf;
extern enum rte_eth_rx_mq_mode rx_mq_mode;
+extern struct rte_flow_action_conntrack conntrack_context;
+
+extern int proc_id;
+extern unsigned int num_procs;
+
+static inline bool
+is_proc_primary(void)
+{
+ return rte_eal_process_type() == RTE_PROC_PRIMARY;
+}
+
static inline unsigned int
lcore_num(void)
{
diff --git a/doc/guides/testpmd_app_ug/run_app.rst b/doc/guides/testpmd_app_ug/run_app.rst
index ca67105b7..098cbbd43 100644
--- a/doc/guides/testpmd_app_ug/run_app.rst
+++ b/doc/guides/testpmd_app_ug/run_app.rst
@@ -529,3 +529,87 @@ The command line options are:
bit 1 - two hairpin ports paired
bit 0 - two hairpin ports loop
The default value is 0. Hairpin will use single port mode and implicit Tx flow mode.
+
+
+Testpmd Multi-Process Command-line Options
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The following are the command-line options for testpmd multi-process support:
+
+* primary process:
+
+.. code-block:: console
+
+ sudo ./dpdk-testpmd -a xxx --proc-type=auto -l 0-1 -- -i --rxq=4 --txq=4 \
+ --num-procs=2 --proc-id=0
+
+* secondary process:
+
+.. code-block:: console
+
+ sudo ./dpdk-testpmd -a xxx --proc-type=auto -l 2-3 -- -i --rxq=4 --txq=4 \
+ --num-procs=2 --proc-id=1
+
+The command line options are:
+
+* ``--num-procs=N``
+
+ The number of processes which will be used.
+
+* ``--proc-id=ID``
+
+ The ID of the current process (ID < num-procs). ID should be different in
+ primary process and secondary process, which starts from '0'.
+
+Calculation rule for queue:
+All queues are allocated to different processes based on ``proc_num`` and
+``proc_id``.
+Calculation rule for the testpmd to allocate queues to each process:
+* start(queue start id) = proc_id * nb_q / num_procs£»
+
+* end(queue end id) = start + nb_q / num_procs£»
+
+For example, if testpmd is configured to have 4 Tx and Rx queues,
+queues 0 and 1 will be used by the primary process and
+queues 2 and 3 will be used by the secondary process.
+
+The number of queues should be a multiple of the number of processes. If not,
+redundant queues will exist after queues are allocated to processes. If RSS
+is enabled, packet loss occurs when traffic is sent to all processes at the same
+time. Some traffic goes to redundant queues and cannot be forwarded.
+
+All the dev ops is supported in primary process. While secondary process is
+not permitted to allocate or release shared memory, so some ops are not supported
+as follows:
+
+- ``dev_configure``
+- ``dev_start``
+- ``dev_stop``
+- ``rx_queue_setup``
+- ``tx_queue_setup``
+- ``rx_queue_release``
+- ``tx_queue_release``
+
+So, any command from testpmd which calls those APIs will not be supported in
+secondary process, like:
+
+.. code-block:: console
+
+ port config all rxq|txq|rxd|txd <value>
+ port config <port_id> rx_offload xxx on/off
+ port config <port_id> tx_offload xxx on/off
+
+etc.
+
+When secondary is running, port in primary is not permitted to be stopped.
+Reconfigure operation is only valid in primary.
+
+Stats is supported, stats will not change when one quits and starts, as they
+share the same buffer to store the stats. Flow rules are maintained in process
+level: primary and secondary has its own flow list (but one flow list in HW).
+The two can see all the queues, so setting the flow rules for the other is OK.
+But in the testpmd primary process receiving or transmitting packets from the
+queue allocated for secondary process is not permitted, and same for secondary
+process.
+
+Flow API and RSS are supported.
--
2.23.0

18
dpdk.spec
View File

@ -1,6 +1,6 @@
Name: dpdk
Version: 20.11
Release: 11
Release: 12
Packager: packaging@6wind.com
URL: http://dpdk.org
%global source_version 20.11
@ -226,7 +226,18 @@ Patch216: 0216-net-hns3-support-set-link-up-down-for-PF.patch
Patch217: 0217-net-hns3-fix-queue-flow-action-validation.patch
Patch218: 0218-net-hns3-fix-taskqueue-pair-reset-command.patch
Patch219: 0219-net-hns3-fix-Tx-push-capability.patch
Patch220: 0220-examples-kni-close-port-before-exit.patch
Patch221: 0221-net-hns3-fix-residual-MAC-after-setting-default-MAC.patch
Patch222: 0222-net-hns3-fix-input-parameters-of-MAC-functions.patch
Patch223: 0223-net-bonding-fix-dedicated-queue-mode-in-vector-burst.patch
Patch224: 0224-net-bonding-fix-RSS-key-length.patch
Patch225: 0225-app-testpmd-add-command-to-show-LACP-bonding-info.patch
Patch226: 0226-app-testpmd-retain-all-original-dev-conf-when-config.patch
Patch227: 0227-net-hns3-remove-similar-macro-function-definitions.patch
Patch228: 0228-net-hns3-fix-interrupt-vector-freeing.patch
Patch229: 0229-net-hns3-add-runtime-config-for-mailbox-limit-time.patch
Patch230: 0230-net-hns3-fix-mailbox-communication-with-HW.patch
Patch231: 0231-app-testpmd-support-multi-process.patch
Summary: Data Plane Development Kit core
Group: System Environment/Libraries
@ -365,6 +376,9 @@ strip -g $RPM_BUILD_ROOT/lib/modules/${namer}/extra/dpdk/rte_kni.ko
/usr/sbin/depmod
%changelog
* Mon Nov 01 2021 Min Hu <humin29@huawei.com> - 20.11-12
- synchronize dmadev and hns3 bugfixes from upstream
* Mon Sep 13 2021 chenchen <chen_aka_jan@163.com> - 20.11-11
- del rpath from some binaries and bin
- add debug package to strip