Skip to contentSkip to conversation

bugfix for hns3 PMD

PATCH In review Min Hu (Connor) 4 participants 38 messages May 19, 2022 – Jun 9, 2022 Network drivers

lore
Share & tools
lore
#0 lore raw

cover bugfix for hns3 PMD

This patch set contains 12 bugfix for hns3 PMD.

Chengwen Feng (1):
net/hns3: fix TM capability incorrectly defined

Huisong Li (10):
net/hns3: fix spelling errors
net/hns3: add check for deferred start queue when rollback net/hns3: remove redundant parentheses
net/hns3: adjust the data type of some variables net/hns3: fix an unreasonable memset
net/hns3: remove duplicate definition
net/hns3: fix return value for unsupported tuple net/hns3: modify a function name
net/hns3: unify the code wrap style
net/hns3: fix a segfault from secondary process

Min Hu (Connor) (1):
net/hns3: fix coverity codecheck

 drivers/net/hns3/hns3_cmd.c       |  6 +--
 drivers/net/hns3/hns3_cmd.h       |  2 +-
 drivers/net/hns3/hns3_common.c    | 14 +++----
 drivers/net/hns3/hns3_dcb.c       |  7 ++--
 drivers/net/hns3/hns3_ethdev.c    | 20 +++++-----
 drivers/net/hns3/hns3_ethdev_vf.c | 23 ++++++------
 drivers/net/hns3/hns3_fdir.c      | 30 +++++++--------
 drivers/net/hns3/hns3_fdir.h      |  2 +-
 drivers/net/hns3/hns3_flow.c      | 25 ++++---------
 drivers/net/hns3/hns3_ptp.c       |  2 +-
 drivers/net/hns3/hns3_regs.c      |  2 +-
 drivers/net/hns3/hns3_rss.c       |  8 ++--
 drivers/net/hns3/hns3_rss.h       |  2 +
 drivers/net/hns3/hns3_rxtx.c      | 62 ++++++++++++++++---------------
 drivers/net/hns3/hns3_rxtx_vec.h  |  4 +-
 drivers/net/hns3/hns3_stats.c     | 28 +++++++-------
 16 files changed, 113 insertions(+), 124 deletions(-)
Signature · 1 line
2.33.0
#1 lore raw

PATCH 1/12 net/hns3: fix spelling errors

From: Huisong Li <[email protected]>

This patch fix spelling errors.

Fixes 70791213242e ("net/hns3: support masking device capability")
Fixes 2192c428f9a6 ("net/hns3: fix firmware compatibility configuration")
---
 drivers/net/hns3/hns3_cmd.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/hns3/hns3_cmd.c +2 -2
index 96f8f38cbb..5dc874fd7a 100644
@@ -462,7 +462,7 @@ hns3_mask_capability(struct hns3_hw *hw,
 	for (i = 0; i < MAX_CAPS_BIT; i++) {
 		if (!(caps_masked & BIT_ULL(i)))
 			continue;
-		hns3_info(hw, "mask capabiliy: id-%u, name-%s.",
+		hns3_info(hw, "mask capability: id-%u, name-%s.",
 			  i, hns3_get_caps_name(i));
 	}
 }
@@ -732,7 +732,7 @@ hns3_cmd_init(struct hns3_hw *hw)
 		return 0;
 
 	/*
-	 * Requiring firmware to enable some features, firber port can still
+	 * Requiring firmware to enable some features, fiber port can still
 	 * work without it, but copper port can't work because the firmware
 	 * fails to take over the PHY.
 	 */
Signature · 1 line
2.33.0
#2 lore raw

PATCH 2/12 net/hns3: add check for deferred start queue when rollback

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

Driver doesn't allocate mbufs for the deferred start queues, so no need to free it when rollback.

---
 drivers/net/hns3/hns3_rxtx.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/hns3/hns3_rxtx.c +3 -0
index d3fa4889d2..a9b997d32e 100644
@@ -1202,6 +1202,9 @@ hns3_init_rx_queues(struct hns3_adapter *hns)
 out:
 	for (j = 0; j < i; j++) {
 		rxq = (struct hns3_rx_queue *)hw->data->rx_queues[j];
+		if (rxq->rx_deferred_start)
+			continue;
+
 		hns3_rx_queue_release_mbufs(rxq);
 	}
 
Signature · 1 line
2.33.0
#3 lore raw

PATCH 3/12 net/hns3: remove redundant parentheses

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

Remove redundant parentheses.

---
 drivers/net/hns3/hns3_rxtx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/hns3/hns3_rxtx.c +2 -2
index a9b997d32e..ee0aaaf7fc 100644
@@ -170,7 +170,7 @@ hns3_fake_rx_queue_release(struct hns3_rx_queue *queue)
 	}
 
 	/* free fake rx queue arrays */
-	if (idx == (hw->fkq_data.nb_fake_rx_queues - 1)) {
+	if (idx == hw->fkq_data.nb_fake_rx_queues - 1) {
 		hw->fkq_data.nb_fake_rx_queues = 0;
 		rte_free(hw->fkq_data.rx_queues);
 		hw->fkq_data.rx_queues = NULL;
@@ -197,7 +197,7 @@ hns3_fake_tx_queue_release(struct hns3_tx_queue *queue)
 	}
 
 	/* free fake tx queue arrays */
-	if (idx == (hw->fkq_data.nb_fake_tx_queues - 1)) {
+	if (idx == hw->fkq_data.nb_fake_tx_queues - 1) {
 		hw->fkq_data.nb_fake_tx_queues = 0;
 		rte_free(hw->fkq_data.tx_queues);
 		hw->fkq_data.tx_queues = NULL;
Signature · 1 line
2.33.0
#4 lore raw

PATCH 4/12 net/hns3: adjust the data type of some variables

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

Using the 'int' type and 'uint16_t' type to compare is insecure. This patch makes them consistent.

---
 drivers/net/hns3/hns3_common.c   |  4 ++--
 drivers/net/hns3/hns3_dcb.c      |  2 +-
 drivers/net/hns3/hns3_ethdev.c   |  2 +-
 drivers/net/hns3/hns3_regs.c     |  2 +-
 drivers/net/hns3/hns3_rss.c      |  2 +-
 drivers/net/hns3/hns3_rxtx.c     | 23 ++++++++++++-----------
 drivers/net/hns3/hns3_rxtx_vec.h |  4 ++--
 7 files changed, 20 insertions(+), 19 deletions(-)
drivers/net/hns3/hns3_common.c +2 -2
index 9c86c00a04..edd16d8076 100644
@@ -475,7 +475,7 @@ hns3_configure_all_mac_addr(struct hns3_adapter *hns, bool del)
 	struct rte_ether_addr *addr;
 	uint16_t mac_addrs_capa;
 	int ret = 0;
-	int i;
+	uint16_t i;
 
 	mac_addrs_capa =
 		hns->is_vf ? HNS3_VF_UC_MACADDR_NUM : HNS3_UC_MACADDR_NUM;
@@ -645,8 +645,8 @@ int
 hns3_init_ring_with_vector(struct hns3_hw *hw)
 {
 	uint16_t vec;
+	uint16_t i;
 	int ret;
-	int i;
 
 	/*
 	 * In hns3 network engine, vector 0 is always the misc interrupt of this
drivers/net/hns3/hns3_dcb.c +1 -1
index 136ada626b..d88757611c 100644
@@ -628,7 +628,7 @@ hns3_set_rss_size(struct hns3_hw *hw, uint16_t nb_rx_q)
 	struct hns3_rss_conf *rss_cfg = &hw->rss_info;
 	uint16_t rx_qnum_per_tc;
 	uint16_t used_rx_queues;
-	int i;
+	uint16_t i;
 
 	rx_qnum_per_tc = nb_rx_q / hw->num_tc;
 	if (rx_qnum_per_tc > hw->rss_size_max) {
drivers/net/hns3/hns3_ethdev.c +1 -1
index 29c9f96c05..97cf27d2a1 100644
@@ -2929,8 +2929,8 @@ hns3_map_tqps_to_func(struct hns3_hw *hw, uint16_t func_id, uint16_t tqp_pid,
 static int
 hns3_map_tqp(struct hns3_hw *hw)
 {
+	uint16_t i;
 	int ret;
-	int i;
 
 	/*
 	 * In current version, VF is not supported when PF is driven by DPDK
drivers/net/hns3/hns3_regs.c +1 -1
index 86a4cf74d5..6778e4cfc2 100644
@@ -294,8 +294,8 @@ hns3_direct_access_regs(struct hns3_hw *hw, uint32_t *data)
 	struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw);
 	uint32_t *origin_data_ptr = data;
 	uint32_t reg_offset;
+	uint16_t i, j;
 	int reg_num;
-	int i, j;
 
 	/* fetching per-PF registers values from PF PCIe register space */
 	reg_num = sizeof(cmdq_reg_addrs) / sizeof(uint32_t);
drivers/net/hns3/hns3_rss.c +1 -1
index d376486a1d..4c546c9363 100644
@@ -631,7 +631,7 @@ hns3_rss_set_default_args(struct hns3_hw *hw)
 {
 	struct hns3_rss_conf *rss_cfg = &hw->rss_info;
 	uint16_t queue_num = hw->alloc_rss_size;
-	int i;
+	uint16_t i;
 
 	/* Default hash algorithm */
 	rss_cfg->conf.func = RTE_ETH_HASH_FUNCTION_TOEPLITZ;
drivers/net/hns3/hns3_rxtx.c +12 -11
index ee0aaaf7fc..510802be05 100644
@@ -390,7 +390,7 @@ hns3_enable_all_queues(struct hns3_hw *hw, bool en)
 	struct hns3_tx_queue *txq;
 	uint32_t rcb_reg;
 	void *tqp_base;
-	int i;
+	uint16_t i;
 
 	for (i = 0; i < hw->cfg_max_queues; i++) {
 		if (hns3_dev_get_support(hw, INDEP_TXRX)) {
@@ -736,8 +736,8 @@ hns3pf_reset_all_tqps(struct hns3_hw *hw)
 #define HNS3_RESET_RCB_NOT_SUPPORT	0U
 #define HNS3_RESET_ALL_TQP_SUCCESS	1U
 	uint8_t reset_status;
+	uint16_t i;
 	int ret;
-	int i;
 
 	ret = hns3_reset_rcb_cmd(hw, &reset_status);
 	if (ret)
@@ -774,7 +774,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
 	uint8_t reset_status;
 	uint8_t msg_data[2];
 	int ret;
-	int i;
+	uint16_t i;
 
 	memset(msg_data, 0, sizeof(uint16_t));
 	ret = hns3_send_mbx_msg(hw, HNS3_MBX_QUEUE_RESET, 0, msg_data,
@@ -806,7 +806,8 @@ int
 hns3_reset_all_tqps(struct hns3_adapter *hns)
 {
 	struct hns3_hw *hw = &hns->hw;
-	int ret, i;
+	uint16_t i;
+	int ret;
 
 	/* Disable all queues before reset all queues */
 	for (i = 0; i < hw->cfg_max_queues; i++) {
@@ -1037,7 +1038,7 @@ hns3_dev_all_rx_queue_intr_enable(struct hns3_hw *hw, bool en)
 {
 	struct rte_eth_dev *dev = &rte_eth_devices[hw->data->port_id];
 	uint16_t nb_rx_q = hw->data->nb_rx_queues;
-	int i;
+	uint16_t i;
 
 	if (dev->data->dev_conf.intr_conf.rxq == 0)
 		return;
@@ -1121,7 +1122,7 @@ static void
 hns3_init_txq(struct hns3_tx_queue *txq)
 {
 	struct hns3_desc *desc;
-	int i;
+	uint16_t i;
 
 	/* Clear tx bd */
 	desc = txq->tx_ring;
@@ -1145,7 +1146,7 @@ hns3_init_tx_ring_tc(struct hns3_adapter *hns)
 
 	for (i = 0; i < HNS3_MAX_TC_NUM; i++) {
 		struct hns3_tc_queue_info *tc_queue = &hw->tc_queue[i];
-		int j;
+		uint16_t j;
 
 		if (!tc_queue->enable)
 			continue;
@@ -1442,7 +1443,7 @@ hns3_alloc_txq_and_dma_zone(struct rte_eth_dev *dev,
 	struct hns3_tx_queue *txq;
 	struct hns3_desc *desc;
 	unsigned int tx_desc;
-	int i;
+	uint16_t i;
 
 	txq = rte_zmalloc_socket(q_info->type, sizeof(struct hns3_tx_queue),
 				 RTE_CACHE_LINE_SIZE, q_info->socket_id);
@@ -1679,7 +1680,7 @@ hns3_dev_release_mbufs(struct hns3_adapter *hns)
 	struct rte_eth_dev_data *dev_data = hns->hw.data;
 	struct hns3_rx_queue *rxq;
 	struct hns3_tx_queue *txq;
-	int i;
+	uint16_t i;
 
 	if (dev_data->rx_queues)
 		for (i = 0; i < dev_data->nb_rx_queues; i++) {
@@ -3086,7 +3087,7 @@ hns3_tx_free_useless_buffer(struct hns3_tx_queue *txq)
 	uint16_t tx_next_use = txq->next_to_use;
 	struct hns3_entry *tx_entry = &txq->sw_ring[tx_next_clean];
 	struct hns3_desc *desc = &txq->tx_ring[tx_next_clean];
-	int i;
+	uint16_t i;
 
 	if (tx_next_use >= tx_next_clean &&
 	    tx_next_use < tx_next_clean + txq->tx_rs_thresh)
@@ -3984,7 +3985,7 @@ hns3_tx_free_buffer_simple(struct hns3_tx_queue *txq)
 	struct hns3_entry *tx_entry;
 	struct hns3_desc *desc;
 	uint16_t tx_next_clean;
-	int i;
+	uint16_t i;
 
 	while (1) {
 		if (HNS3_GET_TX_QUEUE_PEND_BD_NUM(txq) < txq->tx_rs_thresh)
drivers/net/hns3/hns3_rxtx_vec.h +2 -2
index 4985a7cae8..d13f18627d 100644
@@ -15,7 +15,7 @@ hns3_tx_bulk_free_buffers(struct hns3_tx_queue *txq)
 	struct hns3_entry *tx_entry;
 	struct rte_mbuf *m;
 	int nb_free = 0;
-	int i;
+	uint16_t i;
 
 	tx_entry = &txq->sw_ring[txq->next_to_clean];
 	if (txq->mbuf_fast_free_en) {
@@ -56,7 +56,7 @@ static inline void
 hns3_tx_free_buffers(struct hns3_tx_queue *txq)
 {
 	struct hns3_desc *tx_desc;
-	int i;
+	uint16_t i;
 
 	/*
 	 * All mbufs can be released only when the VLD bits of all
Signature · 1 line
2.33.0
#5 lore raw

PATCH 5/12 net/hns3: fix an unreasonable memset

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch fixes an unreasonable memset.

Fixes bba636698316 ("net/hns3: support Rx/Tx and related operations")
---
 drivers/net/hns3/hns3_rxtx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_rxtx.c +1 -1
index 510802be05..5a2cfe5a54 100644
@@ -776,7 +776,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
 	int ret;
 	uint16_t i;
 
-	memset(msg_data, 0, sizeof(uint16_t));
+	memset(msg_data, 0, sizeof(msg_data));
 	ret = hns3_send_mbx_msg(hw, HNS3_MBX_QUEUE_RESET, 0, msg_data,
 				sizeof(msg_data), true, &reset_status,
 				sizeof(reset_status));
Signature · 1 line
2.33.0
#6 lore raw

Re: [PATCH 05/12] net/hns3: fix an unreasonable memset

Quoted 18 lines · Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch fixes an unreasonable memset.

Fixes bba636698316 ("net/hns3: support Rx/Tx and related operations")
---
  drivers/net/hns3/hns3_rxtx.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_rxtx.c +0 -0
index 510802be05..5a2cfe5a54 100644

drivers/net/hns3/hns3_rxtx.c hns3vf_reset_all_tqps(struct hns3_hw *hw) · line 779

Quoted 5 lines
drivers/net/hns3/hns3_rxtx.c +1 -1
@@ -776,7 +776,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
  	int ret;
  	uint16_t i;
 
-	memset(msg_data, 0, sizeof(uint16_t));
+	memset(msg_data, 0, sizeof(msg_data));

It looks a bit suspicious. May be it is better to do:

    memset(&msg_data, 0, sizeof(msg_data));
It is pretty common mistake to do:
    memset(p, 0, sizeof(p));
instead of
    memset(p, 0, sizeof(*p));

when p is a pointer.
I realize that msg_data is an array in this case, but I think it is better to avoid bad pattern in the code.

Quoted 3 lines
drivers/net/hns3/hns3_rxtx.c +0 -0
  	ret = hns3_send_mbx_msg(hw, HNS3_MBX_QUEUE_RESET, 0, msg_data,
  				sizeof(msg_data), true, &reset_status,
  				sizeof(reset_status));
#7 lore raw

Re: [PATCH 05/12] net/hns3: fix an unreasonable memset

Hi, Andrew ,

Quoted 14 lines · Andrew Rybchenko
Quoted 13 lines · Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch fixes an unreasonable memset.

Fixes bba636698316 ("net/hns3: support Rx/Tx and related operations")

---
  drivers/net/hns3/hns3_rxtx.c | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 12 lines

drivers/net/hns3/hns3_rxtx.c hns3vf_reset_all_tqps(struct hns3_hw *hw) · line 778

Quoted 9 lines
drivers/net/hns3/hns3_rxtx.c +1 -1
index 510802be05..5a2cfe5a54 100644
@@ -776,7 +776,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
       int ret;
       uint16_t i;
-    memset(msg_data, 0, sizeof(uint16_t));
+    memset(msg_data, 0, sizeof(msg_data));

It looks a bit suspicious. May be it is better to do:
    memset(&msg_data, 0, sizeof(msg_data));

I think this is too hard to understand for &msg_data is **p. maybe it confuse others when use memset to operate level-2 pointer.

Quoted 13 lines

It is pretty common mistake to do:
    memset(p, 0, sizeof(p));
instead of
    memset(p, 0, sizeof(*p));
when p is a pointer.
I realize that msg_data is an array in this case, but I think it is better to avoid bad pattern in the code.

drivers/net/hns3/hns3_rxtx.c hns3vf_reset_all_tqps(struct hns3_hw *hw) · line 782

Quoted 3 lines
drivers/net/hns3/hns3_rxtx.c +0 -0
       ret = hns3_send_mbx_msg(hw, HNS3_MBX_QUEUE_RESET, 0, msg_data,
                   sizeof(msg_data), true, &reset_status,
                   sizeof(reset_status));

.

#8 lore raw

Re: [PATCH 05/12] net/hns3: fix an unreasonable memset

Quoted 29 lines · Min Hu (Connor)

Hi, Andrew ,

Quoted 14 lines · Andrew Rybchenko
Quoted 13 lines · Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch fixes an unreasonable memset.

Fixes bba636698316 ("net/hns3: support Rx/Tx and related operations")

---
  drivers/net/hns3/hns3_rxtx.c | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 12 lines

drivers/net/hns3/hns3_rxtx.c hns3vf_reset_all_tqps(struct hns3_hw *hw) · line 778

Quoted 9 lines
drivers/net/hns3/hns3_rxtx.c +1 -1
index 510802be05..5a2cfe5a54 100644
@@ -776,7 +776,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
       int ret;
       uint16_t i;
-    memset(msg_data, 0, sizeof(uint16_t));
+    memset(msg_data, 0, sizeof(msg_data));

It looks a bit suspicious. May be it is better to do:
     memset(&msg_data, 0, sizeof(msg_data));

Quoted 2 lines

I think this is too hard to understand for &msg_data is **p. maybe it confuse others when use memset to operate level-2 pointer.

msg_data == &amsg_data
since it is an array

Quoted 15 lines
Quoted 15 lines

It is pretty common mistake to do:
     memset(p, 0, sizeof(p));
instead of
     memset(p, 0, sizeof(*p));
when p is a pointer.
I realize that msg_data is an array in this case, but I think it is better to avoid bad pattern in the code.

drivers/net/hns3/hns3_rxtx.c hns3vf_reset_all_tqps(struct hns3_hw *hw) · line 782

Quoted 3 lines
drivers/net/hns3/hns3_rxtx.c +0 -0
       ret = hns3_send_mbx_msg(hw, HNS3_MBX_QUEUE_RESET, 0, msg_data,
                   sizeof(msg_data), true, &reset_status,
                   sizeof(reset_status));

.

#9 lore raw

Re: [PATCH 05/12] net/hns3: fix an unreasonable memset

Hi,Andrew,

Quoted 30 lines · Andrew Rybchenko
Quoted 29 lines · Min Hu (Connor)

Hi, Andrew ,

Quoted 26 lines · Andrew Rybchenko
Quoted 18 lines · Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch fixes an unreasonable memset.

Fixes bba636698316 ("net/hns3: support Rx/Tx and related operations")

---
  drivers/net/hns3/hns3_rxtx.c | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_rxtx.c +0 -0
index 510802be05..5a2cfe5a54 100644

drivers/net/hns3/hns3_rxtx.c hns3vf_reset_all_tqps(struct hns3_hw *hw) · line 778

Quoted 4 lines
drivers/net/hns3/hns3_rxtx.c +1 -1
@@ -776,7 +776,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
       int ret;
       uint16_t i;
-    memset(msg_data, 0, sizeof(uint16_t));
+    memset(msg_data, 0, sizeof(msg_data));

It looks a bit suspicious. May be it is better to do:
     memset(&msg_data, 0, sizeof(msg_data));

Quoted 5 lines
Quoted 2 lines

I think this is too hard to understand for &msg_data is **p. maybe it confuse others when use memset to operate level-2 pointer.

msg_data == &amsg_data
since it is an array

Yes, you are right, the address of 'msg_data ' is the same as the address of '&amsg_data'.

But I still think this usage is not common.
like the current code in DPDK:
*****
static int
app_parse_flow_conf(const char *conf_str)

{
	int ret;
	uint32_t vals[5];
	struct flow_conf *pconf;
	uint64_t mask;

	memset(vals, 0, sizeof(vals));

...
*****

how about others's opinion? @Ferruh, @Thomas.

Quoted 18 lines
Quoted 11 lines
Quoted 11 lines

It is pretty common mistake to do:
     memset(p, 0, sizeof(p));
instead of
     memset(p, 0, sizeof(*p));
when p is a pointer.
I realize that msg_data is an array in this case, but I think it is better to avoid bad pattern in the code.

Quoted 5 lines
Quoted 5 lines

drivers/net/hns3/hns3_rxtx.c hns3vf_reset_all_tqps(struct hns3_hw *hw) · line 782

Quoted 3 lines
drivers/net/hns3/hns3_rxtx.c +0 -0
       ret = hns3_send_mbx_msg(hw, HNS3_MBX_QUEUE_RESET, 0, msg_data,
                   sizeof(msg_data), true, &reset_status,
                   sizeof(reset_status));

.

.

#10 lore raw

Re: [PATCH 05/12] net/hns3: fix an unreasonable memset

Quoted 55 lines · Min Hu (Connor)

[CAUTION: External Email]

Hi,Andrew,

Quoted 31 lines · Andrew Rybchenko
Quoted 30 lines · Min Hu (Connor)

Hi, Andrew ,

Quoted 27 lines · Andrew Rybchenko
Quoted 19 lines · Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch fixes an unreasonable memset.

Fixes bba636698316 ("net/hns3: support Rx/Tx and related operations")

---
  drivers/net/hns3/hns3_rxtx.c | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_rxtx.c +0 -0
index 510802be05..5a2cfe5a54 100644

drivers/net/hns3/hns3_rxtx.c hns3vf_reset_all_tqps(struct hns3_hw *hw) · line 778

Quoted 4 lines
drivers/net/hns3/hns3_rxtx.c +1 -1
@@ -776,7 +776,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
       int ret;
       uint16_t i;
-    memset(msg_data, 0, sizeof(uint16_t));
+    memset(msg_data, 0, sizeof(msg_data));

It looks a bit suspicious. May be it is better to do:
     memset(&msg_data, 0, sizeof(msg_data));

Quoted 5 lines
Quoted 2 lines

I think this is too hard to understand for &msg_data is **p. maybe it confuse others when use memset to operate level-2 pointer.

msg_data == &amsg_data
since it is an array

Yes, you are right, the address of 'msg_data ' is the same as the address of '&amsg_data'.

But I still think this usage is not common.
like the current code in DPDK:
*****
static int
app_parse_flow_conf(const char *conf_str)
{
       int ret;
       uint32_t vals[5];
       struct flow_conf *pconf;
       uint64_t mask;

Quoted 6 lines

       memset(vals, 0, sizeof(vals));
...
*****

how about others's opinion? @Ferruh, @Thomas.

I would use array without address operator, so original patch lgtm.

+1 to main purpose of the patch (to replace 'sizeof(uint16_t)').

#11 lore raw

Re: [PATCH 05/12] net/hns3: fix an unreasonable memset

Hi,

Quoted 56 lines · Ferruh Yigit
Quoted 55 lines · Min Hu (Connor)

[CAUTION: External Email]

Hi,Andrew,

Quoted 31 lines · Andrew Rybchenko
Quoted 30 lines · Min Hu (Connor)

Hi, Andrew ,

Quoted 27 lines · Andrew Rybchenko
Quoted 19 lines · Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch fixes an unreasonable memset.

Fixes bba636698316 ("net/hns3: support Rx/Tx and related operations")

---
  drivers/net/hns3/hns3_rxtx.c | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_rxtx.c +0 -0
index 510802be05..5a2cfe5a54 100644

drivers/net/hns3/hns3_rxtx.c hns3vf_reset_all_tqps(struct hns3_hw *hw) · line 778

Quoted 4 lines
drivers/net/hns3/hns3_rxtx.c +1 -1
@@ -776,7 +776,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
       int ret;
       uint16_t i;
-    memset(msg_data, 0, sizeof(uint16_t));
+    memset(msg_data, 0, sizeof(msg_data));

It looks a bit suspicious. May be it is better to do:
     memset(&msg_data, 0, sizeof(msg_data));

Quoted 5 lines
Quoted 2 lines

I think this is too hard to understand for &msg_data is **p. maybe it confuse others when use memset to operate level-2 pointer.

msg_data == &amsg_data
since it is an array

Yes, you are right, the address of 'msg_data ' is the same as the address of '&amsg_data'.

But I still think this usage is not common.
like the current code in DPDK:
*****
static int
app_parse_flow_conf(const char *conf_str)
{
        int ret;
        uint32_t vals[5];
        struct flow_conf *pconf;
        uint64_t mask;

Quoted 8 lines
Quoted 6 lines

        memset(vals, 0, sizeof(vals));
...
*****

how about others's opinion? @Ferruh, @Thomas.

I would use array without address operator, so original patch lgtm.

thanks Ferruh.

Quoted 3 lines

+1 to main purpose of the patch (to replace 'sizeof(uint16_t)'). .

#12 lore raw

PATCH 6/12 net/hns3: remove duplicate definition

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

The default hash key array is defined twice. Let's remove the extra one.

Fixes c37ca66f2b27 ("net/hns3: support RSS")
---
 drivers/net/hns3/hns3_flow.c | 9 ---------
 drivers/net/hns3/hns3_rss.c  | 6 ++----
 drivers/net/hns3/hns3_rss.h  | 2 ++
 3 files changed, 4 insertions(+), 13 deletions(-)
drivers/net/hns3/hns3_flow.c +0 -9
index 12afc24910..e994cac314 100644
@@ -10,15 +10,6 @@
 #include "hns3_logs.h"
 #include "hns3_flow.h"
 
-/* Default default keys */
-static uint8_t hns3_hash_key[] = {
-	0x6D, 0x5A, 0x56, 0xDA, 0x25, 0x5B, 0x0E, 0xC2,
-	0x41, 0x67, 0x25, 0x3D, 0x43, 0xA3, 0x8F, 0xB0,
-	0xD0, 0xCA, 0x2B, 0xCB, 0xAE, 0x7B, 0x30, 0xB4,
-	0x77, 0xCB, 0x2D, 0xA3, 0x80, 0x30, 0xF2, 0x0C,
-	0x6A, 0x42, 0xB7, 0x3B, 0xBE, 0xAC, 0x01, 0xFA
-};
-
 static const uint8_t full_mask[VNI_OR_TNI_LEN] = { 0xFF, 0xFF, 0xFF };
 static const uint8_t zero_mask[VNI_OR_TNI_LEN] = { 0x00, 0x00, 0x00 };
 
drivers/net/hns3/hns3_rss.c +2 -4
index 4c546c9363..1003daf03e 100644
@@ -9,10 +9,8 @@
 #include "hns3_ethdev.h"
 #include "hns3_logs.h"
 
-/*
- * The hash key used for rss initialization.
- */
-static const uint8_t hns3_hash_key[] = {
+/* Default hash keys */
+const uint8_t hns3_hash_key[] = {
 	0x6D, 0x5A, 0x56, 0xDA, 0x25, 0x5B, 0x0E, 0xC2,
 	0x41, 0x67, 0x25, 0x3D, 0x43, 0xA3, 0x8F, 0xB0,
 	0xD0, 0xCA, 0x2B, 0xCB, 0xAE, 0x7B, 0x30, 0xB4,
drivers/net/hns3/hns3_rss.h +2 -0
index 7789f02a08..5b90d3a628 100644
@@ -88,6 +88,8 @@ static inline uint32_t roundup_pow_of_two(uint32_t x)
 	return 1UL << fls(x - 1);
 }
 
+extern const uint8_t hns3_hash_key[];
+
 struct hns3_adapter;
 
 int hns3_dev_rss_hash_update(struct rte_eth_dev *dev,
Signature · 1 line
2.33.0
#13 lore raw

PATCH 7/12 net/hns3: fix coverity codecheck

In reply to Min Hu (Connor)

In bitwise operation, "val" should be an unsigned type. This patch fixed it.

Fixes 38b539d96eb6 ("net/hns3: support IEEE 1588 PTP")
---
 drivers/net/hns3/hns3_ptp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_ptp.c +1 -1
index 1442241a4e..0b0061bba5 100644
@@ -81,7 +81,7 @@ hns3_timesync_configure(struct hns3_adapter *hns, bool en)
 	struct hns3_hw *hw = &hns->hw;
 	struct hns3_pf *pf = &hns->pf;
 	struct hns3_cmd_desc desc;
-	int val;
+	uint32_t val;
 	int ret;
 
 	hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_CFG_PTP_MODE, false);
Signature · 1 line
2.33.0
#14 lore raw

PATCH 8/12 net/hns3: fix return value for unsupported tuple

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

Driver should return false for unsupported tuple. This patch fixes it.

Fixes 18a4b4c3fa80 ("net/hns3: add default to switch when parsing fd tuple")
---
 drivers/net/hns3/hns3_fdir.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_fdir.c +1 -1
index 2a7978ac07..a0d6598e57 100644
@@ -631,7 +631,7 @@ static bool hns3_fd_convert_tuple(struct hns3_hw *hw,
 		break;
 	default:
 		hns3_warn(hw, "not support tuple of (%u)", tuple);
-		break;
+		return false;
 	}
 	return true;
 }
Signature · 1 line
2.33.0
#15 lore raw

PATCH 9/12 net/hns3: modify a function name

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

The meaning of the "hns3_get_count" function is not precise enough. This patch changes from "hns3_get_count" to "hns3_fd_get_count".

---
 drivers/net/hns3/hns3_fdir.c | 2 +-
 drivers/net/hns3/hns3_fdir.h | 2 +-
 drivers/net/hns3/hns3_flow.c | 6 +++---
 3 files changed, 5 insertions(+), 5 deletions(-)
drivers/net/hns3/hns3_fdir.c +1 -1
index a0d6598e57..762b89a51e 100644
@@ -1099,7 +1099,7 @@ int hns3_restore_all_fdir_filter(struct hns3_adapter *hns)
 	return 0;
 }
 
-int hns3_get_count(struct hns3_hw *hw, uint32_t id, uint64_t *value)
+int hns3_fd_get_count(struct hns3_hw *hw, uint32_t id, uint64_t *value)
 {
 	struct hns3_fd_get_cnt_cmd *req;
 	struct hns3_cmd_desc desc;
drivers/net/hns3/hns3_fdir.h +1 -1
index 3376c40c8e..4d18759160 100644
@@ -184,7 +184,7 @@ void hns3_fdir_filter_uninit(struct hns3_adapter *hns);
 int hns3_fdir_filter_program(struct hns3_adapter *hns,
 			     struct hns3_fdir_rule *rule, bool del);
 int hns3_clear_all_fdir_filter(struct hns3_adapter *hns);
-int hns3_get_count(struct hns3_hw *hw, uint32_t id, uint64_t *value);
+int hns3_fd_get_count(struct hns3_hw *hw, uint32_t id, uint64_t *value);
 int hns3_restore_all_fdir_filter(struct hns3_adapter *hns);
 
 #endif /* _HNS3_FDIR_H_ */
drivers/net/hns3/hns3_flow.c +3 -3
index e994cac314..b60ad596dc 100644
@@ -164,13 +164,13 @@ hns3_counter_new(struct rte_eth_dev *dev, uint32_t indirect, uint32_t id,
 				"Counter id is used, indirect flag not match");
 		/* Clear the indirect counter on first use. */
 		if (cnt->indirect && cnt->ref_cnt == 1)
-			(void)hns3_get_count(hw, id, &value);
+			(void)hns3_fd_get_count(hw, id, &value);
 		cnt->ref_cnt++;
 		return 0;
 	}
 
 	/* Clear the counter by read ops because the counter is read-clear */
-	ret = hns3_get_count(hw, id, &value);
+	ret = hns3_fd_get_count(hw, id, &value);
 	if (ret)
 		return rte_flow_error_set(error, EIO,
 					  RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
@@ -210,7 +210,7 @@ hns3_counter_query(struct rte_eth_dev *dev, struct rte_flow *flow,
 					  RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
 					  "Can't find counter id");
 
-	ret = hns3_get_count(&hns->hw, flow->counter_id, &value);
+	ret = hns3_fd_get_count(&hns->hw, flow->counter_id, &value);
 	if (ret) {
 		rte_flow_error_set(error, -ret, RTE_FLOW_ERROR_TYPE_HANDLE,
 				   NULL, "Read counter fail.");
Signature · 1 line
2.33.0
#16 lore raw

PATCH 10/12 net/hns3: unify the code wrap style

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch unifies the code wrap style.

---
 drivers/net/hns3/hns3_cmd.c       |  2 +-
 drivers/net/hns3/hns3_common.c    | 10 +++++-----
 drivers/net/hns3/hns3_dcb.c       |  5 ++---
 drivers/net/hns3/hns3_ethdev.c    | 18 ++++++++----------
 drivers/net/hns3/hns3_ethdev_vf.c | 23 +++++++++++------------
 drivers/net/hns3/hns3_fdir.c      | 26 +++++++++++++-------------
 drivers/net/hns3/hns3_flow.c      | 10 ++++------
 drivers/net/hns3/hns3_rxtx.c      | 28 +++++++++++++---------------
 drivers/net/hns3/hns3_stats.c     | 28 ++++++++++++++--------------
 9 files changed, 71 insertions(+), 79 deletions(-)
drivers/net/hns3/hns3_cmd.c +1 -1
index 5dc874fd7a..eadc15093f 100644
@@ -108,7 +108,7 @@ hns3_alloc_cmd_queue(struct hns3_hw *hw, int ring_type)
 	ret = hns3_alloc_cmd_desc(hw, ring);
 	if (ret)
 		hns3_err(hw, "descriptor %s alloc error %d",
-			    (ring_type == HNS3_TYPE_CSQ) ? "CSQ" : "CRQ", ret);
+			 (ring_type == HNS3_TYPE_CSQ) ? "CSQ" : "CRQ", ret);
 
 	return ret;
 }
drivers/net/hns3/hns3_common.c +5 -5
index edd16d8076..7a65db907e 100644
@@ -604,7 +604,7 @@ hns3_init_mac_addrs(struct rte_eth_dev *dev)
 				0);
 	if (dev->data->mac_addrs == NULL) {
 		hns3_err(hw, "failed to allocate %zx bytes needed to store MAC addresses",
-			     sizeof(struct rte_ether_addr) * mac_addrs_capa);
+			 sizeof(struct rte_ether_addr) * mac_addrs_capa);
 		return -ENOMEM;
 	}
 
@@ -680,16 +680,16 @@ hns3_init_ring_with_vector(struct hns3_hw *hw)
 		ret = hw->ops.bind_ring_with_vector(hw, vec, false,
 						    HNS3_RING_TYPE_TX, i);
 		if (ret) {
-			PMD_INIT_LOG(ERR, "fail to unbind TX ring(%d) with "
-					  "vector: %u, ret=%d", i, vec, ret);
+			PMD_INIT_LOG(ERR, "fail to unbind TX ring(%d) with vector: %u, ret=%d",
+				     i, vec, ret);
 			return ret;
 		}
 
 		ret = hw->ops.bind_ring_with_vector(hw, vec, false,
 						    HNS3_RING_TYPE_RX, i);
 		if (ret) {
-			PMD_INIT_LOG(ERR, "fail to unbind RX ring(%d) with "
-					  "vector: %u, ret=%d", i, vec, ret);
+			PMD_INIT_LOG(ERR, "fail to unbind RX ring(%d) with vector: %u, ret=%d",
+				     i, vec, ret);
 			return ret;
 		}
 	}
drivers/net/hns3/hns3_dcb.c +2 -3
index d88757611c..b22f618e0a 100644
@@ -876,9 +876,8 @@ hns3_dcb_pri_tc_base_dwrr_cfg(struct hns3_hw *hw)
 
 		ret = hns3_dcb_pri_weight_cfg(hw, i, dwrr);
 		if (ret) {
-			hns3_err(hw,
-			       "fail to send priority weight cmd: %d, ret = %d",
-			       i, ret);
+			hns3_err(hw, "fail to send priority weight cmd: %d, ret = %d",
+				 i, ret);
 			return ret;
 		}
 
drivers/net/hns3/hns3_ethdev.c +8 -10
index 97cf27d2a1..8a8f3f1950 100644
@@ -1627,7 +1627,7 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
 	ret = hw->ops.del_uc_mac_addr(hw, oaddr);
 	if (ret) {
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      oaddr);
+				       oaddr);
 		hns3_warn(hw, "Remove old uc mac address(%s) fail: %d",
 			  mac_str, ret);
 
@@ -1659,7 +1659,7 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
 	ret_val = hw->ops.del_uc_mac_addr(hw, mac_addr);
 	if (ret_val) {
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      mac_addr);
+				       mac_addr);
 		hns3_warn(hw,
 			  "Failed to roll back to del setted mac addr(%s): %d",
 			  mac_str, ret_val);
@@ -1670,7 +1670,7 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
 	if (ret_val) {
 		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);
+			  mac_str, ret_val);
 	}
 	rte_spinlock_unlock(&hw->lock);
 
@@ -1747,7 +1747,7 @@ hns3_add_mc_mac_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
 		if (ret == -ENOSPC)
 			hns3_err(hw, "mc mac vlan table is full");
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      mac_addr);
+				       mac_addr);
 		hns3_err(hw, "failed to add mc mac addr(%s): %d", mac_str, ret);
 	}
 
@@ -2676,9 +2676,8 @@ hns3_check_dev_specifications(struct hns3_hw *hw)
 {
 	if (hw->rss_ind_tbl_size == 0 ||
 	    hw->rss_ind_tbl_size > HNS3_RSS_IND_TBL_SIZE_MAX) {
-		hns3_err(hw, "the size of hash lookup table configured (%u)"
-			      " exceeds the maximum(%u)", hw->rss_ind_tbl_size,
-			      HNS3_RSS_IND_TBL_SIZE_MAX);
+		hns3_err(hw, "the size of hash lookup table configured (%u) exceeds the maximum(%u)",
+			 hw->rss_ind_tbl_size, HNS3_RSS_IND_TBL_SIZE_MAX);
 		return -EINVAL;
 	}
 
@@ -3916,7 +3915,7 @@ hns3_dev_promiscuous_enable(struct rte_eth_dev *dev)
 		ret = hns3_enable_vlan_filter(hns, false);
 		if (ret) {
 			hns3_err(hw, "failed to enable promiscuous mode due to "
-				     "failure to disable vlan filter, ret = %d",
+				 "failure to disable vlan filter, ret = %d",
 				 ret);
 			err = hns3_set_promisc_mode(hw, false, allmulti);
 			if (err)
@@ -5993,8 +5992,7 @@ hns3_reset_service(void *param)
 		timersub(&tv, &tv_start, &tv_delta);
 		msec = hns3_clock_calctime_ms(&tv_delta);
 		if (msec > HNS3_RESET_PROCESS_MS)
-			hns3_err(hw, "%d handle long time delta %" PRIu64
-				     " ms time=%ld.%.6ld",
+			hns3_err(hw, "%d handle long time delta %" PRIu64 " ms time=%ld.%.6ld",
 				 hw->reset.level, msec,
 				 tv.tv_sec, tv.tv_usec);
 		if (ret == -EAGAIN)
drivers/net/hns3/hns3_ethdev_vf.c +11 -12
index 0c170797f4..7da6e21b0f 100644
@@ -142,7 +142,7 @@ hns3vf_enable_msix(const struct rte_pci_device *device, bool op)
 	pos = hns3vf_find_pci_capability(device, PCI_CAP_ID_MSIX);
 	if (pos) {
 		ret = rte_pci_read_config(device, &control, sizeof(control),
-				    (pos + PCI_MSIX_FLAGS));
+					  (pos + PCI_MSIX_FLAGS));
 		if (ret < 0) {
 			PMD_INIT_LOG(ERR, "Failed to read PCI offset 0x%x",
 				     (pos + PCI_MSIX_FLAGS));
@@ -154,10 +154,10 @@ hns3vf_enable_msix(const struct rte_pci_device *device, bool op)
 		else
 			control &= ~PCI_MSIX_FLAGS_ENABLE;
 		ret = rte_pci_write_config(device, &control, sizeof(control),
-					  (pos + PCI_MSIX_FLAGS));
+					   (pos + PCI_MSIX_FLAGS));
 		if (ret < 0) {
 			PMD_INIT_LOG(ERR, "failed to write PCI offset 0x%x",
-				    (pos + PCI_MSIX_FLAGS));
+				     (pos + PCI_MSIX_FLAGS));
 			return -ENXIO;
 		}
 
@@ -199,7 +199,7 @@ hns3vf_remove_uc_mac_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
 				false, NULL, 0);
 	if (ret) {
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      mac_addr);
+				       mac_addr);
 		hns3_err(hw, "failed to add uc mac addr(%s), ret = %d",
 			 mac_str, ret);
 	}
@@ -241,12 +241,12 @@ hns3vf_set_default_mac_addr(struct rte_eth_dev *dev,
 		 */
 		if (ret == -EPERM) {
 			hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-					      old_addr);
+					       old_addr);
 			hns3_warn(hw, "Has permanet mac addr(%s) for vf",
 				  mac_str);
 		} else {
 			hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-					      mac_addr);
+					       mac_addr);
 			hns3_err(hw, "Failed to set mac addr(%s) for vf: %d",
 				 mac_str, ret);
 		}
@@ -293,7 +293,7 @@ hns3vf_remove_mc_mac_addr(struct hns3_hw *hw,
 				NULL, 0);
 	if (ret) {
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      mac_addr);
+				       mac_addr);
 		hns3_err(hw, "Failed to remove mc mac addr(%s) for vf: %d",
 			 mac_str, ret);
 	}
@@ -715,9 +715,8 @@ hns3vf_check_dev_specifications(struct hns3_hw *hw)
 {
 	if (hw->rss_ind_tbl_size == 0 ||
 	    hw->rss_ind_tbl_size > HNS3_RSS_IND_TBL_SIZE_MAX) {
-		hns3_warn(hw, "the size of hash lookup table configured (%u)"
-			      " exceeds the maximum(%u)", hw->rss_ind_tbl_size,
-			      HNS3_RSS_IND_TBL_SIZE_MAX);
+		hns3_warn(hw, "the size of hash lookup table configured (%u) exceeds the maximum(%u)",
+			  hw->rss_ind_tbl_size, HNS3_RSS_IND_TBL_SIZE_MAX);
 		return -EINVAL;
 	}
 
@@ -1168,8 +1167,8 @@ hns3vf_vlan_offload_set(struct rte_eth_dev *dev, int mask)
 	int ret = 0;
 
 	if (__atomic_load_n(&hw->reset.resetting, __ATOMIC_RELAXED)) {
-		hns3_err(hw, "vf set vlan offload failed during resetting, "
-			     "mask = 0x%x", mask);
+		hns3_err(hw, "vf set vlan offload failed during resetting, mask = 0x%x",
+			 mask);
 		return -EIO;
 	}
 
drivers/net/hns3/hns3_fdir.c +13 -13
index 762b89a51e..30e5e66772 100644
@@ -321,7 +321,7 @@ int hns3_init_fd_config(struct hns3_adapter *hns)
 		break;
 	default:
 		hns3_err(hw, "Unsupported flow director mode %u",
-			    pf->fdir.fd_cfg.fd_mode);
+			 pf->fdir.fd_cfg.fd_mode);
 		return -EOPNOTSUPP;
 	}
 
@@ -337,7 +337,7 @@ int hns3_init_fd_config(struct hns3_adapter *hns)
 	    BIT(INNER_SRC_IP) | BIT(INNER_DST_IP) |
 	    BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT);
 	hns3_dbg(hw, "fdir tuple: inner<vlan_tag1 eth_type ip_src ip_dst "
-		  "ip_proto ip_tos l4_src_port l4_dst_port>");
+		 "ip_proto ip_tos l4_src_port l4_dst_port>");
 
 	/* If use max 400bit key, we can support tuples for ether type */
 	if (pf->fdir.fd_cfg.max_key_length == MAX_KEY_LENGTH) {
@@ -348,8 +348,8 @@ int hns3_init_fd_config(struct hns3_adapter *hns)
 		    BIT(OUTER_TUN_VNI) | BIT(OUTER_TUN_FLOW_ID) |
 		    BIT(OUTER_ETH_TYPE) | BIT(OUTER_IP_PROTO);
 		hns3_dbg(hw, "fdir tuple more: inner<dst_mac src_mac "
-			  "vlan_tag2 sctp_tag> outer<eth_type ip_proto "
-			  "l4_src_port l4_dst_port tun_vni tun_flow_id>");
+			 "vlan_tag2 sctp_tag> outer<eth_type ip_proto "
+			 "l4_src_port l4_dst_port tun_vni tun_flow_id>");
 	}
 
 	/* roce_type is used to filter roce frames
@@ -367,12 +367,11 @@ int hns3_init_fd_config(struct hns3_adapter *hns)
 	if (ret)
 		return ret;
 
-	hns3_dbg(hw, "fdir: stage1<rules-%u counters-%u> stage2<rules-%u "
-		  "counters=%u>",
-		  pf->fdir.fd_cfg.rule_num[HNS3_FD_STAGE_1],
-		  pf->fdir.fd_cfg.cnt_num[HNS3_FD_STAGE_1],
-		  pf->fdir.fd_cfg.rule_num[HNS3_FD_STAGE_2],
-		  pf->fdir.fd_cfg.cnt_num[HNS3_FD_STAGE_2]);
+	hns3_dbg(hw, "fdir: stage1<rules-%u counters-%u> stage2<rules-%u counters=%u>",
+		 pf->fdir.fd_cfg.rule_num[HNS3_FD_STAGE_1],
+		 pf->fdir.fd_cfg.cnt_num[HNS3_FD_STAGE_1],
+		 pf->fdir.fd_cfg.rule_num[HNS3_FD_STAGE_2],
+		 pf->fdir.fd_cfg.cnt_num[HNS3_FD_STAGE_2]);
 
 	return hns3_set_fd_key_config(hns);
 }
@@ -420,7 +419,7 @@ static int hns3_fd_tcam_config(struct hns3_hw *hw, bool sel_x, int loc,
 	ret = hns3_cmd_send(hw, desc, FD_TCAM_CMD_NUM);
 	if (ret)
 		hns3_err(hw, "Config tcam key fail, ret=%d loc=%d add=%d",
-			    ret, loc, is_add);
+			 ret, loc, is_add);
 	return ret;
 }
 
@@ -673,6 +672,7 @@ static void hns3_fd_convert_meta_data(struct hns3_fd_key_cfg *cfg,
 		} else if (i == VLAN_NUMBER) {
 			uint32_t vlan_tag;
 			uint8_t vlan_num;
+
 			if (rule->key_conf.spec.tunnel_type == 0)
 				vlan_num = rule->key_conf.vlan_num;
 			else
@@ -758,14 +758,14 @@ static int hns3_config_key(struct hns3_adapter *hns,
 	ret = hns3_fd_tcam_config(hw, false, rule->location, key_y, true);
 	if (ret) {
 		hns3_err(hw, "Config fd key_y fail, loc=%u, ret=%d",
-			    rule->queue_id, ret);
+			 rule->queue_id, ret);
 		return ret;
 	}
 
 	ret = hns3_fd_tcam_config(hw, true, rule->location, key_x, true);
 	if (ret)
 		hns3_err(hw, "Config fd key_x fail, loc=%u, ret=%d",
-			    rule->queue_id, ret);
+			 rule->queue_id, ret);
 	return ret;
 }
 
drivers/net/hns3/hns3_flow.c +4 -6
index b60ad596dc..5e0a9bc93f 100644
@@ -285,9 +285,8 @@ hns3_handle_action_queue(struct rte_eth_dev *dev,
 
 	queue = (const struct rte_flow_action_queue *)action->conf;
 	if (queue->index >= hw->data->nb_rx_queues) {
-		hns3_err(hw, "queue ID(%u) is greater than number of "
-			  "available queue (%u) in driver.",
-			  queue->index, hw->data->nb_rx_queues);
+		hns3_err(hw, "queue ID(%u) is greater than number of available queue (%u) in driver.",
+			 queue->index, hw->data->nb_rx_queues);
 		return rte_flow_error_set(error, EINVAL,
 					  RTE_FLOW_ERROR_TYPE_ACTION_CONF,
 					  action, "Invalid queue ID in PF");
@@ -1656,9 +1655,8 @@ hns3_clear_rss_filter(struct rte_eth_dev *dev)
 	}
 
 	if (rss_rule_fail_cnt) {
-		hns3_err(hw, "fail to delete all RSS filters, success num = %d "
-			     "fail num = %d", rss_rule_succ_cnt,
-			     rss_rule_fail_cnt);
+		hns3_err(hw, "fail to delete all RSS filters, success num = %d fail num = %d",
+			 rss_rule_succ_cnt, rss_rule_fail_cnt);
 		ret = -EIO;
 	}
 
drivers/net/hns3/hns3_rxtx.c +13 -15
index 5a2cfe5a54..3f576fbf4b 100644
@@ -606,8 +606,8 @@ hns3_send_reset_tqp_cmd(struct hns3_hw *hw, uint16_t queue_id, bool enable)
 	hns3_set_bit(req->reset_req, HNS3_TQP_RESET_B, enable ? 1 : 0);
 	ret = hns3_cmd_send(hw, &desc, 1);
 	if (ret)
-		hns3_err(hw, "send tqp reset cmd error, queue_id = %u, "
-			     "ret = %d", queue_id, ret);
+		hns3_err(hw, "send tqp reset cmd error, queue_id = %u, ret = %d",
+			 queue_id, ret);
 
 	return ret;
 }
@@ -627,8 +627,8 @@ hns3_get_tqp_reset_status(struct hns3_hw *hw, uint16_t queue_id,
 
 	ret = hns3_cmd_send(hw, &desc, 1);
 	if (ret) {
-		hns3_err(hw, "get tqp reset status error, queue_id = %u, "
-			     "ret = %d.", queue_id, ret);
+		hns3_err(hw, "get tqp reset status error, queue_id = %u, ret = %d.",
+			 queue_id, ret);
 		return ret;
 	}
 	*reset_status = hns3_get_bit(req->ready_to_reset, HNS3_TQP_RESET_B);
@@ -669,7 +669,7 @@ hns3pf_reset_tqp(struct hns3_hw *hw, uint16_t queue_id)
 	if (!reset_status) {
 		ret = -ETIMEDOUT;
 		hns3_err(hw, "reset tqp timeout, queue_id = %u, ret = %d",
-			     queue_id, ret);
+			 queue_id, ret);
 		goto tqp_reset_fail;
 	}
 
@@ -752,15 +752,14 @@ hns3pf_reset_all_tqps(struct hns3_hw *hw)
 		for (i = 0; i < hw->cfg_max_queues; i++) {
 			ret = hns3pf_reset_tqp(hw, i);
 			if (ret) {
-				hns3_err(hw,
-				  "fail to reset tqp, queue_id = %d, ret = %d.",
-				  i, ret);
+				hns3_err(hw, "fail to reset tqp, queue_id = %d, ret = %d.",
+					 i, ret);
 				return ret;
 			}
 		}
 	} else if (reset_status != HNS3_RESET_ALL_TQP_SUCCESS) {
 		hns3_err(hw, "fail to reset all tqps, reset_status = %u.",
-				reset_status);
+			 reset_status);
 		return -EIO;
 	}
 
@@ -813,9 +812,8 @@ hns3_reset_all_tqps(struct hns3_adapter *hns)
 	for (i = 0; i < hw->cfg_max_queues; i++) {
 		ret = hns3_tqp_enable(hw, i, false);
 		if (ret) {
-			hns3_err(hw,
-			    "fail to disable tqps before tqps reset, ret = %d.",
-			    ret);
+			hns3_err(hw, "fail to disable tqps before tqps reset, ret = %d.",
+				 ret);
 			return ret;
 		}
 	}
@@ -922,9 +920,9 @@ hns3_reset_queue(struct hns3_hw *hw, uint16_t queue_id,
 	}
 
 	if (!reset_status) {
-		hns3_err(hw, "reset queue timeout, queue_id = %u, "
-			     "queue_type = %s", queue_id,
-			     queue_type == HNS3_RING_TYPE_TX ? "Tx" : "Rx");
+		hns3_err(hw, "reset queue timeout, queue_id = %u, queue_type = %s",
+			 queue_id,
+			 queue_type == HNS3_RING_TYPE_TX ? "Tx" : "Rx");
 		ret = -ETIMEDOUT;
 		goto queue_reset_fail;
 	}
drivers/net/hns3/hns3_stats.c +14 -14
index 9b7ad067aa..83725ecf20 100644
@@ -507,8 +507,8 @@ hns3_update_port_rx_ssu_drop_stats(struct hns3_hw *hw)
 
 	req = (struct hns3_query_ssu_cmd *)desc[0].data;
 	cnt = rte_le_to_cpu_32(req->oq_drop_cnt) +
-		rte_le_to_cpu_32(req->full_drop_cnt) +
-		rte_le_to_cpu_32(req->part_drop_cnt);
+	      rte_le_to_cpu_32(req->full_drop_cnt) +
+	      rte_le_to_cpu_32(req->part_drop_cnt);
 
 	stats->ssu_rx_drop_cnt += cnt;
 
@@ -532,8 +532,8 @@ hns3_update_port_tx_ssu_drop_stats(struct hns3_hw *hw)
 
 	req = (struct hns3_query_ssu_cmd *)desc[0].data;
 	cnt = rte_le_to_cpu_32(req->oq_drop_cnt) +
-		rte_le_to_cpu_32(req->full_drop_cnt) +
-		rte_le_to_cpu_32(req->part_drop_cnt);
+	      rte_le_to_cpu_32(req->full_drop_cnt) +
+	      rte_le_to_cpu_32(req->part_drop_cnt);
 
 	hw->oerror_stats += cnt;
 
@@ -1336,8 +1336,8 @@ hns3_dev_xstats_get_by_id(struct rte_eth_dev *dev, const uint64_t *ids,
 	len = cnt_stats * sizeof(struct rte_eth_xstat);
 	values_copy = rte_zmalloc("hns3_xstats_values", len, 0);
 	if (values_copy == NULL) {
-		hns3_err(hw, "Failed to allocate 0x%" PRIx64 " bytes needed "
-			     "to store statistics values", len);
+		hns3_err(hw, "Failed to allocate 0x%" PRIx64 " bytes needed to store statistics values",
+			 len);
 		return -ENOMEM;
 	}
 
@@ -1358,8 +1358,8 @@ hns3_dev_xstats_get_by_id(struct rte_eth_dev *dev, const uint64_t *ids,
 
 	for (i = 0; i < size; i++) {
 		if (ids[i] >= cnt_stats) {
-			hns3_err(hw, "ids[%u] (%" PRIu64 ") is invalid, "
-				     "should < %u", i, ids[i], cnt_stats);
+			hns3_err(hw, "ids[%u] (%" PRIu64 ") is invalid, should < %u",
+				 i, ids[i], cnt_stats);
 			rte_free(values_copy);
 			return -EINVAL;
 		}
@@ -1418,8 +1418,8 @@ hns3_dev_xstats_get_names_by_id(struct rte_eth_dev *dev,
 	len = cnt_stats * sizeof(struct rte_eth_xstat_name);
 	names_copy = rte_zmalloc("hns3_xstats_names", len, 0);
 	if (names_copy == NULL) {
-		hns3_err(hw, "Failed to allocate 0x%" PRIx64 " bytes needed "
-			     "to store statistics names", len);
+		hns3_err(hw, "Failed to allocate 0x%" PRIx64 " bytes needed to store statistics names",
+			 len);
 		return -ENOMEM;
 	}
 
@@ -1427,8 +1427,8 @@ hns3_dev_xstats_get_names_by_id(struct rte_eth_dev *dev,
 
 	for (i = 0; i < size; i++) {
 		if (ids[i] >= cnt_stats) {
-			hns3_err(hw, "ids[%u] (%" PRIu64 ") is invalid, "
-				     "should < %u", i, ids[i], cnt_stats);
+			hns3_err(hw, "ids[%u] (%" PRIu64 ") is invalid, should < %u",
+				 i, ids[i], cnt_stats);
 			rte_free(names_copy);
 			return -EINVAL;
 		}
@@ -1500,14 +1500,14 @@ hns3_tqp_stats_init(struct hns3_hw *hw)
 	struct hns3_tqp_stats *tqp_stats = &hw->tqp_stats;
 
 	tqp_stats->rcb_rx_ring_pktnum = rte_zmalloc("hns3_rx_ring_pkt_num",
-					 sizeof(uint64_t) * hw->tqps_num, 0);
+					sizeof(uint64_t) * hw->tqps_num, 0);
 	if (tqp_stats->rcb_rx_ring_pktnum == NULL) {
 		hns3_err(hw, "failed to allocate rx_ring pkt_num.");
 		return -ENOMEM;
 	}
 
 	tqp_stats->rcb_tx_ring_pktnum = rte_zmalloc("hns3_tx_ring_pkt_num",
-					 sizeof(uint64_t) * hw->tqps_num, 0);
+					sizeof(uint64_t) * hw->tqps_num, 0);
 	if (tqp_stats->rcb_tx_ring_pktnum == NULL) {
 		hns3_err(hw, "failed to allocate tx_ring pkt_num.");
 		rte_free(tqp_stats->rcb_rx_ring_pktnum);
Signature · 1 line
2.33.0
#17 lore raw

PATCH 11/12 net/hns3: fix a segfault from secondary process

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

If a hns3 device in the secondary process is attached to do probing operation, 'rx_queues' and 'tx_queues' in dev->data are null in eth_dev_fp_ops_setup when calling rte_eth_dev_probing_finish. The primary process calls dev_start to re-setup their fp_ops. But the secondary process can't call dev_start and has no chance to do it. If the application sends and receives packets at this time, a segfault will occur. So this patch uses the MP communication of the PMD to update the fp_ops of the device in the secondary process.

Fixes 9e311d8f93b9 ("net/hns3: fix Rx/Tx when fast path operation introduced")
---
 drivers/net/hns3/hns3_rxtx.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/hns3/hns3_rxtx.c +2 -0
index 3f576fbf4b..0dc1d8cb60 100644
@@ -4420,6 +4420,8 @@ hns3_eth_dev_fp_ops_config(const struct rte_eth_dev *dev)
 	fpo[port_id].tx_pkt_prepare = dev->tx_pkt_prepare;
 	fpo[port_id].rx_descriptor_status = dev->rx_descriptor_status;
 	fpo[port_id].tx_descriptor_status = dev->tx_descriptor_status;
+	fpo[port_id].rxq.data = dev->data->rx_queues;
+	fpo[port_id].txq.data = dev->data->tx_queues;
 }
 
 void
Signature · 1 line
2.33.0
#18 lore raw

PATCH 12/12 net/hns3: fix TM capability incorrectly defined

In reply to Min Hu (Connor)

From: Chengwen Feng <[email protected]>

The TM capability should be bit-19 according to the user manual of firmware. This patch fixes it.

Fixes fc18d1b4b85f ("net/hns3: fix traffic management")
---
 drivers/net/hns3/hns3_cmd.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_cmd.h +1 -1
index f9addc6069..82c999061d 100644
@@ -323,7 +323,7 @@ enum HNS3_CAPS_BITS {
 	HNS3_CAPS_UDP_TUNNEL_CSUM_B,
 	HNS3_CAPS_RAS_IMP_B,
 	HNS3_CAPS_RXD_ADV_LAYOUT_B = 15,
-	HNS3_CAPS_TM_B = 17,
+	HNS3_CAPS_TM_B = 19,
 };
 
 /* Capabilities of VF dependent on the PF */
Signature · 1 line
2.33.0
#20 lore raw

Hi, Andrew ,
v2 has been sent, thanks.

Quoted 6 lines · Andrew Rybchenko
Quoted 1 line · Min Hu (Connor)

This patch set contains 12 bugfix for hns3 PMD.

The patch series fails to apply to dpdk-next-net/main. Please, rebase and send v2.
.

#21 lore raw

v2 cover bugfix for hns3 PMD

In reply to Min Hu (Connor)

This patch set contains 11 bugfix for hns3 PMD.

Chengwen Feng (1):
net/hns3: fix TM capability incorrectly defined

Huisong Li (9):
net/hns3: add check for deferred start queue when rollback net/hns3: remove redundant parentheses
net/hns3: adjust the data type of some variables net/hns3: fix an unreasonable memset
net/hns3: remove duplicate definition
net/hns3: fix return value for unsupported tuple net/hns3: modify a function name
net/hns3: unify the code wrap style
net/hns3: fix a segfault from secondary process

Min Hu (Connor) (1):
net/hns3: fix coverity codecheck

 drivers/net/hns3/hns3_cmd.c       |  2 +-
 drivers/net/hns3/hns3_cmd.h       |  2 +-
 drivers/net/hns3/hns3_common.c    | 14 +++----
 drivers/net/hns3/hns3_dcb.c       |  7 ++--
 drivers/net/hns3/hns3_ethdev.c    | 20 +++++-----
 drivers/net/hns3/hns3_ethdev_vf.c | 23 ++++++------
 drivers/net/hns3/hns3_fdir.c      | 30 +++++++--------
 drivers/net/hns3/hns3_fdir.h      |  2 +-
 drivers/net/hns3/hns3_flow.c      | 25 ++++---------
 drivers/net/hns3/hns3_ptp.c       |  2 +-
 drivers/net/hns3/hns3_regs.c      |  2 +-
 drivers/net/hns3/hns3_rss.c       |  8 ++--
 drivers/net/hns3/hns3_rss.h       |  2 +
 drivers/net/hns3/hns3_rxtx.c      | 62 ++++++++++++++++---------------
 drivers/net/hns3/hns3_rxtx_vec.h  |  4 +-
 drivers/net/hns3/hns3_stats.c     | 28 +++++++-------
 16 files changed, 111 insertions(+), 122 deletions(-)

---
v2:
* delete one patch which is fixed by others.
refer to:
https://patches.dpdk.org/project/dpdk/cover/[email protected]/

Signature · 1 line
2.33.0
#22 lore raw

PATCH v2 1/11 net/hns3: add check for deferred start queue when rollback

From: Huisong Li <[email protected]>

Driver doesn't allocate mbufs for the deferred start queues, so no need to free it when rollback.

---
 drivers/net/hns3/hns3_rxtx.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/hns3/hns3_rxtx.c +3 -0
index 0c91e4721e..1c9648bd52 100644
@@ -1199,6 +1199,9 @@ hns3_init_rx_queues(struct hns3_adapter *hns)
 out:
 	for (j = 0; j < i; j++) {
 		rxq = (struct hns3_rx_queue *)hw->data->rx_queues[j];
+		if (rxq->rx_deferred_start)
+			continue;
+
 		hns3_rx_queue_release_mbufs(rxq);
 	}
 
Signature · 1 line
2.33.0
#23 lore raw

PATCH v2 2/11 net/hns3: remove redundant parentheses

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

Remove redundant parentheses.

---
 drivers/net/hns3/hns3_rxtx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/hns3/hns3_rxtx.c +2 -2
index 1c9648bd52..4cba4fd13e 100644
@@ -167,7 +167,7 @@ hns3_fake_rx_queue_release(struct hns3_rx_queue *queue)
 	}
 
 	/* free fake rx queue arrays */
-	if (idx == (hw->fkq_data.nb_fake_rx_queues - 1)) {
+	if (idx == hw->fkq_data.nb_fake_rx_queues - 1) {
 		hw->fkq_data.nb_fake_rx_queues = 0;
 		rte_free(hw->fkq_data.rx_queues);
 		hw->fkq_data.rx_queues = NULL;
@@ -194,7 +194,7 @@ hns3_fake_tx_queue_release(struct hns3_tx_queue *queue)
 	}
 
 	/* free fake tx queue arrays */
-	if (idx == (hw->fkq_data.nb_fake_tx_queues - 1)) {
+	if (idx == hw->fkq_data.nb_fake_tx_queues - 1) {
 		hw->fkq_data.nb_fake_tx_queues = 0;
 		rte_free(hw->fkq_data.tx_queues);
 		hw->fkq_data.tx_queues = NULL;
Signature · 1 line
2.33.0
#24 lore raw

PATCH v2 3/11 net/hns3: adjust the data type of some variables

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

Using the 'int' type and 'uint16_t' type to compare is insecure. This patch makes them consistent.

---
 drivers/net/hns3/hns3_common.c   |  4 ++--
 drivers/net/hns3/hns3_dcb.c      |  2 +-
 drivers/net/hns3/hns3_ethdev.c   |  2 +-
 drivers/net/hns3/hns3_regs.c     |  2 +-
 drivers/net/hns3/hns3_rss.c      |  2 +-
 drivers/net/hns3/hns3_rxtx.c     | 23 ++++++++++++-----------
 drivers/net/hns3/hns3_rxtx_vec.h |  4 ++--
 7 files changed, 20 insertions(+), 19 deletions(-)
drivers/net/hns3/hns3_common.c +2 -2
index dcdc609654..221d4ccec3 100644
@@ -475,7 +475,7 @@ hns3_configure_all_mac_addr(struct hns3_adapter *hns, bool del)
 	struct rte_ether_addr *addr;
 	uint16_t mac_addrs_capa;
 	int ret = 0;
-	int i;
+	uint16_t i;
 
 	mac_addrs_capa =
 		hns->is_vf ? HNS3_VF_UC_MACADDR_NUM : HNS3_UC_MACADDR_NUM;
@@ -645,8 +645,8 @@ int
 hns3_init_ring_with_vector(struct hns3_hw *hw)
 {
 	uint16_t vec;
+	uint16_t i;
 	int ret;
-	int i;
 
 	/*
 	 * In hns3 network engine, vector 0 is always the misc interrupt of this
drivers/net/hns3/hns3_dcb.c +1 -1
index 73693786d1..04d7f530ef 100644
@@ -628,7 +628,7 @@ hns3_set_rss_size(struct hns3_hw *hw, uint16_t nb_rx_q)
 	struct hns3_rss_conf *rss_cfg = &hw->rss_info;
 	uint16_t rx_qnum_per_tc;
 	uint16_t used_rx_queues;
-	int i;
+	uint16_t i;
 
 	rx_qnum_per_tc = nb_rx_q / hw->num_tc;
 	if (rx_qnum_per_tc > hw->rss_size_max) {
drivers/net/hns3/hns3_ethdev.c +1 -1
index 0b565a5614..3cffacdd6a 100644
@@ -2928,8 +2928,8 @@ hns3_map_tqps_to_func(struct hns3_hw *hw, uint16_t func_id, uint16_t tqp_pid,
 static int
 hns3_map_tqp(struct hns3_hw *hw)
 {
+	uint16_t i;
 	int ret;
-	int i;
 
 	/*
 	 * In current version, VF is not supported when PF is driven by DPDK
drivers/net/hns3/hns3_regs.c +1 -1
index 86a4cf74d5..6778e4cfc2 100644
@@ -294,8 +294,8 @@ hns3_direct_access_regs(struct hns3_hw *hw, uint32_t *data)
 	struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw);
 	uint32_t *origin_data_ptr = data;
 	uint32_t reg_offset;
+	uint16_t i, j;
 	int reg_num;
-	int i, j;
 
 	/* fetching per-PF registers values from PF PCIe register space */
 	reg_num = sizeof(cmdq_reg_addrs) / sizeof(uint32_t);
drivers/net/hns3/hns3_rss.c +1 -1
index d376486a1d..4c546c9363 100644
@@ -631,7 +631,7 @@ hns3_rss_set_default_args(struct hns3_hw *hw)
 {
 	struct hns3_rss_conf *rss_cfg = &hw->rss_info;
 	uint16_t queue_num = hw->alloc_rss_size;
-	int i;
+	uint16_t i;
 
 	/* Default hash algorithm */
 	rss_cfg->conf.func = RTE_ETH_HASH_FUNCTION_TOEPLITZ;
drivers/net/hns3/hns3_rxtx.c +12 -11
index 4cba4fd13e..be4ab16dcc 100644
@@ -387,7 +387,7 @@ hns3_enable_all_queues(struct hns3_hw *hw, bool en)
 	struct hns3_tx_queue *txq;
 	uint32_t rcb_reg;
 	void *tqp_base;
-	int i;
+	uint16_t i;
 
 	for (i = 0; i < hw->cfg_max_queues; i++) {
 		if (hns3_dev_get_support(hw, INDEP_TXRX)) {
@@ -733,8 +733,8 @@ hns3pf_reset_all_tqps(struct hns3_hw *hw)
 #define HNS3_RESET_RCB_NOT_SUPPORT	0U
 #define HNS3_RESET_ALL_TQP_SUCCESS	1U
 	uint8_t reset_status;
+	uint16_t i;
 	int ret;
-	int i;
 
 	ret = hns3_reset_rcb_cmd(hw, &reset_status);
 	if (ret)
@@ -771,7 +771,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
 	uint8_t reset_status;
 	uint8_t msg_data[2];
 	int ret;
-	int i;
+	uint16_t i;
 
 	memset(msg_data, 0, sizeof(uint16_t));
 	ret = hns3_send_mbx_msg(hw, HNS3_MBX_QUEUE_RESET, 0, msg_data,
@@ -803,7 +803,8 @@ int
 hns3_reset_all_tqps(struct hns3_adapter *hns)
 {
 	struct hns3_hw *hw = &hns->hw;
-	int ret, i;
+	uint16_t i;
+	int ret;
 
 	/* Disable all queues before reset all queues */
 	for (i = 0; i < hw->cfg_max_queues; i++) {
@@ -1034,7 +1035,7 @@ hns3_dev_all_rx_queue_intr_enable(struct hns3_hw *hw, bool en)
 {
 	struct rte_eth_dev *dev = &rte_eth_devices[hw->data->port_id];
 	uint16_t nb_rx_q = hw->data->nb_rx_queues;
-	int i;
+	uint16_t i;
 
 	if (dev->data->dev_conf.intr_conf.rxq == 0)
 		return;
@@ -1118,7 +1119,7 @@ static void
 hns3_init_txq(struct hns3_tx_queue *txq)
 {
 	struct hns3_desc *desc;
-	int i;
+	uint16_t i;
 
 	/* Clear tx bd */
 	desc = txq->tx_ring;
@@ -1142,7 +1143,7 @@ hns3_init_tx_ring_tc(struct hns3_adapter *hns)
 
 	for (i = 0; i < HNS3_MAX_TC_NUM; i++) {
 		struct hns3_tc_queue_info *tc_queue = &hw->tc_queue[i];
-		int j;
+		uint16_t j;
 
 		if (!tc_queue->enable)
 			continue;
@@ -1439,7 +1440,7 @@ hns3_alloc_txq_and_dma_zone(struct rte_eth_dev *dev,
 	struct hns3_tx_queue *txq;
 	struct hns3_desc *desc;
 	unsigned int tx_desc;
-	int i;
+	uint16_t i;
 
 	txq = rte_zmalloc_socket(q_info->type, sizeof(struct hns3_tx_queue),
 				 RTE_CACHE_LINE_SIZE, q_info->socket_id);
@@ -1676,7 +1677,7 @@ hns3_dev_release_mbufs(struct hns3_adapter *hns)
 	struct rte_eth_dev_data *dev_data = hns->hw.data;
 	struct hns3_rx_queue *rxq;
 	struct hns3_tx_queue *txq;
-	int i;
+	uint16_t i;
 
 	if (dev_data->rx_queues)
 		for (i = 0; i < dev_data->nb_rx_queues; i++) {
@@ -3083,7 +3084,7 @@ hns3_tx_free_useless_buffer(struct hns3_tx_queue *txq)
 	uint16_t tx_next_use = txq->next_to_use;
 	struct hns3_entry *tx_entry = &txq->sw_ring[tx_next_clean];
 	struct hns3_desc *desc = &txq->tx_ring[tx_next_clean];
-	int i;
+	uint16_t i;
 
 	if (tx_next_use >= tx_next_clean &&
 	    tx_next_use < tx_next_clean + txq->tx_rs_thresh)
@@ -3981,7 +3982,7 @@ hns3_tx_free_buffer_simple(struct hns3_tx_queue *txq)
 	struct hns3_entry *tx_entry;
 	struct hns3_desc *desc;
 	uint16_t tx_next_clean;
-	int i;
+	uint16_t i;
 
 	while (1) {
 		if (HNS3_GET_TX_QUEUE_PEND_BD_NUM(txq) < txq->tx_rs_thresh)
drivers/net/hns3/hns3_rxtx_vec.h +2 -2
index 4985a7cae8..d13f18627d 100644
@@ -15,7 +15,7 @@ hns3_tx_bulk_free_buffers(struct hns3_tx_queue *txq)
 	struct hns3_entry *tx_entry;
 	struct rte_mbuf *m;
 	int nb_free = 0;
-	int i;
+	uint16_t i;
 
 	tx_entry = &txq->sw_ring[txq->next_to_clean];
 	if (txq->mbuf_fast_free_en) {
@@ -56,7 +56,7 @@ static inline void
 hns3_tx_free_buffers(struct hns3_tx_queue *txq)
 {
 	struct hns3_desc *tx_desc;
-	int i;
+	uint16_t i;
 
 	/*
 	 * All mbufs can be released only when the VLD bits of all
Signature · 1 line
2.33.0
#25 lore raw

PATCH v2 4/11 net/hns3: fix an unreasonable memset

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch fixes an unreasonable memset.

Fixes bba636698316 ("net/hns3: support Rx/Tx and related operations")
---
 drivers/net/hns3/hns3_rxtx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_rxtx.c +1 -1
index be4ab16dcc..dcb1cdf168 100644
@@ -773,7 +773,7 @@ hns3vf_reset_all_tqps(struct hns3_hw *hw)
 	int ret;
 	uint16_t i;
 
-	memset(msg_data, 0, sizeof(uint16_t));
+	memset(msg_data, 0, sizeof(msg_data));
 	ret = hns3_send_mbx_msg(hw, HNS3_MBX_QUEUE_RESET, 0, msg_data,
 				sizeof(msg_data), true, &reset_status,
 				sizeof(reset_status));
Signature · 1 line
2.33.0
#26 lore raw

PATCH v2 5/11 net/hns3: remove duplicate definition

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

The default hash key array is defined twice. Let's remove the extra one.

Fixes c37ca66f2b27 ("net/hns3: support RSS")
---
 drivers/net/hns3/hns3_flow.c | 9 ---------
 drivers/net/hns3/hns3_rss.c  | 6 ++----
 drivers/net/hns3/hns3_rss.h  | 2 ++
 3 files changed, 4 insertions(+), 13 deletions(-)
drivers/net/hns3/hns3_flow.c +0 -9
index 86ebbf69b6..ca9edc5244 100644
@@ -10,15 +10,6 @@
 #include "hns3_logs.h"
 #include "hns3_flow.h"
 
-/* Default default keys */
-static uint8_t hns3_hash_key[] = {
-	0x6D, 0x5A, 0x56, 0xDA, 0x25, 0x5B, 0x0E, 0xC2,
-	0x41, 0x67, 0x25, 0x3D, 0x43, 0xA3, 0x8F, 0xB0,
-	0xD0, 0xCA, 0x2B, 0xCB, 0xAE, 0x7B, 0x30, 0xB4,
-	0x77, 0xCB, 0x2D, 0xA3, 0x80, 0x30, 0xF2, 0x0C,
-	0x6A, 0x42, 0xB7, 0x3B, 0xBE, 0xAC, 0x01, 0xFA
-};
-
 static const uint8_t full_mask[VNI_OR_TNI_LEN] = { 0xFF, 0xFF, 0xFF };
 static const uint8_t zero_mask[VNI_OR_TNI_LEN] = { 0x00, 0x00, 0x00 };
 
drivers/net/hns3/hns3_rss.c +2 -4
index 4c546c9363..1003daf03e 100644
@@ -9,10 +9,8 @@
 #include "hns3_ethdev.h"
 #include "hns3_logs.h"
 
-/*
- * The hash key used for rss initialization.
- */
-static const uint8_t hns3_hash_key[] = {
+/* Default hash keys */
+const uint8_t hns3_hash_key[] = {
 	0x6D, 0x5A, 0x56, 0xDA, 0x25, 0x5B, 0x0E, 0xC2,
 	0x41, 0x67, 0x25, 0x3D, 0x43, 0xA3, 0x8F, 0xB0,
 	0xD0, 0xCA, 0x2B, 0xCB, 0xAE, 0x7B, 0x30, 0xB4,
drivers/net/hns3/hns3_rss.h +2 -0
index 55d5718ffc..56627cbd4c 100644
@@ -88,6 +88,8 @@ static inline uint32_t roundup_pow_of_two(uint32_t x)
 	return 1UL << fls(x - 1);
 }
 
+extern const uint8_t hns3_hash_key[];
+
 struct hns3_adapter;
 
 int hns3_dev_rss_hash_update(struct rte_eth_dev *dev,
Signature · 1 line
2.33.0
#27 lore raw

PATCH v2 6/11 net/hns3: fix coverity codecheck

In reply to Min Hu (Connor)

In bitwise operation, "val" should be an unsigned type. This patch fixed it.

Fixes 38b539d96eb6 ("net/hns3: support IEEE 1588 PTP")
---
 drivers/net/hns3/hns3_ptp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_ptp.c +1 -1
index 1442241a4e..0b0061bba5 100644
@@ -81,7 +81,7 @@ hns3_timesync_configure(struct hns3_adapter *hns, bool en)
 	struct hns3_hw *hw = &hns->hw;
 	struct hns3_pf *pf = &hns->pf;
 	struct hns3_cmd_desc desc;
-	int val;
+	uint32_t val;
 	int ret;
 
 	hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_CFG_PTP_MODE, false);
Signature · 1 line
2.33.0
#28 lore raw

Re: [PATCH v2 06/11] net/hns3: fix coverity codecheck

Quoted 5 lines · Min Hu (Connor)

In bitwise operation, "val" should be an unsigned type. This patch fixed it.

Fixes 38b539d96eb6 ("net/hns3: support IEEE 1588 PTP")

Please, let me know coverity ID. No necessity to resend. I'll add it on applying.

Quoted 19 lines
---
  drivers/net/hns3/hns3_ptp.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_ptp.c +1 -1
index 1442241a4e..0b0061bba5 100644
@@ -81,7 +81,7 @@ hns3_timesync_configure(struct hns3_adapter *hns, bool en)
  	struct hns3_hw *hw = &hns->hw;
  	struct hns3_pf *pf = &hns->pf;
  	struct hns3_cmd_desc desc;
-	int val;
+	uint32_t val;
  	int ret;
 
  	hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_CFG_PTP_MODE, false);
#29 lore raw

Re: [PATCH v2 06/11] net/hns3: fix coverity codecheck

Hi Andrew

Many thanks for your review.

I will take over Hu Min's current work and continue to do the upstream work.

Quoted 9 lines · Andrew Rybchenko
Quoted 5 lines · Min Hu (Connor)

In bitwise operation, "val" should be an unsigned type. This patch fixed it.

Fixes 38b539d96eb6 ("net/hns3: support IEEE 1588 PTP")

Please, let me know coverity ID. No necessity to resend. I'll add it on applying.

The code warning is checked by our internal code check tool, does not have the coverity ID.

Thanks,
Dongdong

Quoted 23 lines
Quoted 12 lines
---
  drivers/net/hns3/hns3_ptp.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_ptp.c +0 -0
index 1442241a4e..0b0061bba5 100644

drivers/net/hns3/hns3_ptp.c hns3_timesync_configure(struct hns3_adapter *hns, · line 87

Quoted 8 lines
drivers/net/hns3/hns3_ptp.c +1 -1
@@ -81,7 +81,7 @@ hns3_timesync_configure(struct hns3_adapter *hns,
 bool en)
      struct hns3_hw *hw = &hns->hw;
      struct hns3_pf *pf = &hns->pf;
      struct hns3_cmd_desc desc;
-    int val;
+    uint32_t val;
      int ret;
        hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_CFG_PTP_MODE, false);

.

#30 lore raw

Re: [PATCH v2 06/11] net/hns3: fix coverity codecheck

Hi Dongdong,

Quoted 11 lines · Dongdong Liu

Hi Andrew

Many thanks for your review.

I will take over Hu Min's current work and continue to do the upstream work.

Quoted 4 lines · Andrew Rybchenko
Quoted 3 lines · Min Hu (Connor)

In bitwise operation, "val" should be an unsigned type. This patch fixed it.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines
Fixes 38b539d96eb6 ("net/hns3: support IEEE 1588 PTP")

Please, let me know coverity ID. No necessity to resend. I'll add it on applying.

The code warning is checked by our internal code check tool, does not have the coverity ID.

Hm, the summary says coverity. Could you suggest how to rephrase it.

Thanks,
Andrew.

#31 lore raw

Re: [PATCH v2 06/11] net/hns3: fix coverity codecheck

Hi Andrew

Quoted 18 lines · Andrew Rybchenko

Hi Dongdong,

Quoted 15 lines · Dongdong Liu

Hi Andrew

Many thanks for your review.

I will take over Hu Min's current work and continue to do the upstream work.

Quoted 7 lines · Andrew Rybchenko
Quoted 5 lines · Min Hu (Connor)

In bitwise operation, "val" should be an unsigned type. This patch fixed it.

Fixes 38b539d96eb6 ("net/hns3: support IEEE 1588 PTP")
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

Please, let me know coverity ID. No necessity to resend. I'll add it on applying.

The code warning is checked by our internal code check tool, does not have the coverity ID.

Hm, the summary says coverity. Could you suggest how to rephrase it.

Aha, the title is inaccurate.
How about change to "net/hns3: fix code check warning".

Thanks,
Dongdong

Quoted 6 lines

Thanks,
Andrew.

.

#32 lore raw

PATCH v2 7/11 net/hns3: fix return value for unsupported tuple

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

Driver should return false for unsupported tuple. This patch fixes it.

Fixes 18a4b4c3fa80 ("net/hns3: add default to switch when parsing fd tuple")
---
 drivers/net/hns3/hns3_fdir.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_fdir.c +1 -1
index 2a7978ac07..a0d6598e57 100644
@@ -631,7 +631,7 @@ static bool hns3_fd_convert_tuple(struct hns3_hw *hw,
 		break;
 	default:
 		hns3_warn(hw, "not support tuple of (%u)", tuple);
-		break;
+		return false;
 	}
 	return true;
 }
Signature · 1 line
2.33.0
#33 lore raw

PATCH v2 8/11 net/hns3: modify a function name

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

The meaning of the "hns3_get_count" function is not precise enough. This patch changes from "hns3_get_count" to "hns3_fd_get_count".

---
 drivers/net/hns3/hns3_fdir.c | 2 +-
 drivers/net/hns3/hns3_fdir.h | 2 +-
 drivers/net/hns3/hns3_flow.c | 6 +++---
 3 files changed, 5 insertions(+), 5 deletions(-)
drivers/net/hns3/hns3_fdir.c +1 -1
index a0d6598e57..762b89a51e 100644
@@ -1099,7 +1099,7 @@ int hns3_restore_all_fdir_filter(struct hns3_adapter *hns)
 	return 0;
 }
 
-int hns3_get_count(struct hns3_hw *hw, uint32_t id, uint64_t *value)
+int hns3_fd_get_count(struct hns3_hw *hw, uint32_t id, uint64_t *value)
 {
 	struct hns3_fd_get_cnt_cmd *req;
 	struct hns3_cmd_desc desc;
drivers/net/hns3/hns3_fdir.h +1 -1
index 8d588ffef3..d81f04a3f3 100644
@@ -184,7 +184,7 @@ void hns3_fdir_filter_uninit(struct hns3_adapter *hns);
 int hns3_fdir_filter_program(struct hns3_adapter *hns,
 			     struct hns3_fdir_rule *rule, bool del);
 int hns3_clear_all_fdir_filter(struct hns3_adapter *hns);
-int hns3_get_count(struct hns3_hw *hw, uint32_t id, uint64_t *value);
+int hns3_fd_get_count(struct hns3_hw *hw, uint32_t id, uint64_t *value);
 int hns3_restore_all_fdir_filter(struct hns3_adapter *hns);
 
 #endif /* _HNS3_FDIR_H_ */
drivers/net/hns3/hns3_flow.c +3 -3
index ca9edc5244..1bd4ac160d 100644
@@ -164,13 +164,13 @@ hns3_counter_new(struct rte_eth_dev *dev, uint32_t indirect, uint32_t id,
 				"Counter id is used, indirect flag not match");
 		/* Clear the indirect counter on first use. */
 		if (cnt->indirect && cnt->ref_cnt == 1)
-			(void)hns3_get_count(hw, id, &value);
+			(void)hns3_fd_get_count(hw, id, &value);
 		cnt->ref_cnt++;
 		return 0;
 	}
 
 	/* Clear the counter by read ops because the counter is read-clear */
-	ret = hns3_get_count(hw, id, &value);
+	ret = hns3_fd_get_count(hw, id, &value);
 	if (ret)
 		return rte_flow_error_set(error, EIO,
 					  RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
@@ -210,7 +210,7 @@ hns3_counter_query(struct rte_eth_dev *dev, struct rte_flow *flow,
 					  RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
 					  "Can't find counter id");
 
-	ret = hns3_get_count(&hns->hw, flow->counter_id, &value);
+	ret = hns3_fd_get_count(&hns->hw, flow->counter_id, &value);
 	if (ret) {
 		rte_flow_error_set(error, -ret, RTE_FLOW_ERROR_TYPE_HANDLE,
 				   NULL, "Read counter fail.");
Signature · 1 line
2.33.0
#34 lore raw

PATCH v2 9/11 net/hns3: unify the code wrap style

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

This patch unifies the code wrap style.

---
 drivers/net/hns3/hns3_cmd.c       |  2 +-
 drivers/net/hns3/hns3_common.c    | 10 +++++-----
 drivers/net/hns3/hns3_dcb.c       |  5 ++---
 drivers/net/hns3/hns3_ethdev.c    | 18 ++++++++----------
 drivers/net/hns3/hns3_ethdev_vf.c | 23 +++++++++++------------
 drivers/net/hns3/hns3_fdir.c      | 26 +++++++++++++-------------
 drivers/net/hns3/hns3_flow.c      | 10 ++++------
 drivers/net/hns3/hns3_rxtx.c      | 28 +++++++++++++---------------
 drivers/net/hns3/hns3_stats.c     | 28 ++++++++++++++--------------
 9 files changed, 71 insertions(+), 79 deletions(-)
drivers/net/hns3/hns3_cmd.c +1 -1
index 5dc874fd7a..eadc15093f 100644
@@ -108,7 +108,7 @@ hns3_alloc_cmd_queue(struct hns3_hw *hw, int ring_type)
 	ret = hns3_alloc_cmd_desc(hw, ring);
 	if (ret)
 		hns3_err(hw, "descriptor %s alloc error %d",
-			    (ring_type == HNS3_TYPE_CSQ) ? "CSQ" : "CRQ", ret);
+			 (ring_type == HNS3_TYPE_CSQ) ? "CSQ" : "CRQ", ret);
 
 	return ret;
 }
drivers/net/hns3/hns3_common.c +5 -5
index 221d4ccec3..424205356e 100644
@@ -604,7 +604,7 @@ hns3_init_mac_addrs(struct rte_eth_dev *dev)
 				0);
 	if (dev->data->mac_addrs == NULL) {
 		hns3_err(hw, "failed to allocate %zx bytes needed to store MAC addresses",
-			     sizeof(struct rte_ether_addr) * mac_addrs_capa);
+			 sizeof(struct rte_ether_addr) * mac_addrs_capa);
 		return -ENOMEM;
 	}
 
@@ -680,16 +680,16 @@ hns3_init_ring_with_vector(struct hns3_hw *hw)
 		ret = hw->ops.bind_ring_with_vector(hw, vec, false,
 						    HNS3_RING_TYPE_TX, i);
 		if (ret) {
-			PMD_INIT_LOG(ERR, "fail to unbind TX ring(%d) with "
-					  "vector: %u, ret=%d", i, vec, ret);
+			PMD_INIT_LOG(ERR, "fail to unbind TX ring(%d) with vector: %u, ret=%d",
+				     i, vec, ret);
 			return ret;
 		}
 
 		ret = hw->ops.bind_ring_with_vector(hw, vec, false,
 						    HNS3_RING_TYPE_RX, i);
 		if (ret) {
-			PMD_INIT_LOG(ERR, "fail to unbind RX ring(%d) with "
-					  "vector: %u, ret=%d", i, vec, ret);
+			PMD_INIT_LOG(ERR, "fail to unbind RX ring(%d) with vector: %u, ret=%d",
+				     i, vec, ret);
 			return ret;
 		}
 	}
drivers/net/hns3/hns3_dcb.c +2 -3
index 04d7f530ef..af045b22f7 100644
@@ -876,9 +876,8 @@ hns3_dcb_pri_tc_base_dwrr_cfg(struct hns3_hw *hw)
 
 		ret = hns3_dcb_pri_weight_cfg(hw, i, dwrr);
 		if (ret) {
-			hns3_err(hw,
-			       "fail to send priority weight cmd: %d, ret = %d",
-			       i, ret);
+			hns3_err(hw, "fail to send priority weight cmd: %d, ret = %d",
+				 i, ret);
 			return ret;
 		}
 
drivers/net/hns3/hns3_ethdev.c +8 -10
index 3cffacdd6a..6c77c8b964 100644
@@ -1626,7 +1626,7 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
 	ret = hw->ops.del_uc_mac_addr(hw, oaddr);
 	if (ret) {
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      oaddr);
+				       oaddr);
 		hns3_warn(hw, "Remove old uc mac address(%s) fail: %d",
 			  mac_str, ret);
 
@@ -1658,7 +1658,7 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
 	ret_val = hw->ops.del_uc_mac_addr(hw, mac_addr);
 	if (ret_val) {
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      mac_addr);
+				       mac_addr);
 		hns3_warn(hw,
 			  "Failed to roll back to del setted mac addr(%s): %d",
 			  mac_str, ret_val);
@@ -1669,7 +1669,7 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
 	if (ret_val) {
 		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);
+			  mac_str, ret_val);
 	}
 	rte_spinlock_unlock(&hw->lock);
 
@@ -1746,7 +1746,7 @@ hns3_add_mc_mac_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
 		if (ret == -ENOSPC)
 			hns3_err(hw, "mc mac vlan table is full");
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      mac_addr);
+				       mac_addr);
 		hns3_err(hw, "failed to add mc mac addr(%s): %d", mac_str, ret);
 	}
 
@@ -2675,9 +2675,8 @@ hns3_check_dev_specifications(struct hns3_hw *hw)
 {
 	if (hw->rss_ind_tbl_size == 0 ||
 	    hw->rss_ind_tbl_size > HNS3_RSS_IND_TBL_SIZE_MAX) {
-		hns3_err(hw, "the size of hash lookup table configured (%u)"
-			      " exceeds the maximum(%u)", hw->rss_ind_tbl_size,
-			      HNS3_RSS_IND_TBL_SIZE_MAX);
+		hns3_err(hw, "the size of hash lookup table configured (%u) exceeds the maximum(%u)",
+			 hw->rss_ind_tbl_size, HNS3_RSS_IND_TBL_SIZE_MAX);
 		return -EINVAL;
 	}
 
@@ -3915,7 +3914,7 @@ hns3_dev_promiscuous_enable(struct rte_eth_dev *dev)
 		ret = hns3_enable_vlan_filter(hns, false);
 		if (ret) {
 			hns3_err(hw, "failed to enable promiscuous mode due to "
-				     "failure to disable vlan filter, ret = %d",
+				 "failure to disable vlan filter, ret = %d",
 				 ret);
 			err = hns3_set_promisc_mode(hw, false, allmulti);
 			if (err)
@@ -5992,8 +5991,7 @@ hns3_reset_service(void *param)
 		timersub(&tv, &tv_start, &tv_delta);
 		msec = hns3_clock_calctime_ms(&tv_delta);
 		if (msec > HNS3_RESET_PROCESS_MS)
-			hns3_err(hw, "%d handle long time delta %" PRIu64
-				     " ms time=%ld.%.6ld",
+			hns3_err(hw, "%d handle long time delta %" PRIu64 " ms time=%ld.%.6ld",
 				 hw->reset.level, msec,
 				 tv.tv_sec, tv.tv_usec);
 		if (ret == -EAGAIN)
drivers/net/hns3/hns3_ethdev_vf.c +11 -12
index 589de0ab3a..bebfaa6417 100644
@@ -141,7 +141,7 @@ hns3vf_enable_msix(const struct rte_pci_device *device, bool op)
 	pos = hns3vf_find_pci_capability(device, PCI_CAP_ID_MSIX);
 	if (pos) {
 		ret = rte_pci_read_config(device, &control, sizeof(control),
-				    (pos + PCI_MSIX_FLAGS));
+					  (pos + PCI_MSIX_FLAGS));
 		if (ret < 0) {
 			PMD_INIT_LOG(ERR, "Failed to read PCI offset 0x%x",
 				     (pos + PCI_MSIX_FLAGS));
@@ -153,10 +153,10 @@ hns3vf_enable_msix(const struct rte_pci_device *device, bool op)
 		else
 			control &= ~PCI_MSIX_FLAGS_ENABLE;
 		ret = rte_pci_write_config(device, &control, sizeof(control),
-					  (pos + PCI_MSIX_FLAGS));
+					   (pos + PCI_MSIX_FLAGS));
 		if (ret < 0) {
 			PMD_INIT_LOG(ERR, "failed to write PCI offset 0x%x",
-				    (pos + PCI_MSIX_FLAGS));
+				     (pos + PCI_MSIX_FLAGS));
 			return -ENXIO;
 		}
 
@@ -198,7 +198,7 @@ hns3vf_remove_uc_mac_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr)
 				false, NULL, 0);
 	if (ret) {
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      mac_addr);
+				       mac_addr);
 		hns3_err(hw, "failed to add uc mac addr(%s), ret = %d",
 			 mac_str, ret);
 	}
@@ -240,12 +240,12 @@ hns3vf_set_default_mac_addr(struct rte_eth_dev *dev,
 		 */
 		if (ret == -EPERM) {
 			hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-					      old_addr);
+					       old_addr);
 			hns3_warn(hw, "Has permanent mac addr(%s) for vf",
 				  mac_str);
 		} else {
 			hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-					      mac_addr);
+					       mac_addr);
 			hns3_err(hw, "Failed to set mac addr(%s) for vf: %d",
 				 mac_str, ret);
 		}
@@ -292,7 +292,7 @@ hns3vf_remove_mc_mac_addr(struct hns3_hw *hw,
 				NULL, 0);
 	if (ret) {
 		hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-				      mac_addr);
+				       mac_addr);
 		hns3_err(hw, "Failed to remove mc mac addr(%s) for vf: %d",
 			 mac_str, ret);
 	}
@@ -714,9 +714,8 @@ hns3vf_check_dev_specifications(struct hns3_hw *hw)
 {
 	if (hw->rss_ind_tbl_size == 0 ||
 	    hw->rss_ind_tbl_size > HNS3_RSS_IND_TBL_SIZE_MAX) {
-		hns3_warn(hw, "the size of hash lookup table configured (%u)"
-			      " exceeds the maximum(%u)", hw->rss_ind_tbl_size,
-			      HNS3_RSS_IND_TBL_SIZE_MAX);
+		hns3_warn(hw, "the size of hash lookup table configured (%u) exceeds the maximum(%u)",
+			  hw->rss_ind_tbl_size, HNS3_RSS_IND_TBL_SIZE_MAX);
 		return -EINVAL;
 	}
 
@@ -1167,8 +1166,8 @@ hns3vf_vlan_offload_set(struct rte_eth_dev *dev, int mask)
 	int ret = 0;
 
 	if (__atomic_load_n(&hw->reset.resetting, __ATOMIC_RELAXED)) {
-		hns3_err(hw, "vf set vlan offload failed during resetting, "
-			     "mask = 0x%x", mask);
+		hns3_err(hw, "vf set vlan offload failed during resetting, mask = 0x%x",
+			 mask);
 		return -EIO;
 	}
 
drivers/net/hns3/hns3_fdir.c +13 -13
index 762b89a51e..30e5e66772 100644
@@ -321,7 +321,7 @@ int hns3_init_fd_config(struct hns3_adapter *hns)
 		break;
 	default:
 		hns3_err(hw, "Unsupported flow director mode %u",
-			    pf->fdir.fd_cfg.fd_mode);
+			 pf->fdir.fd_cfg.fd_mode);
 		return -EOPNOTSUPP;
 	}
 
@@ -337,7 +337,7 @@ int hns3_init_fd_config(struct hns3_adapter *hns)
 	    BIT(INNER_SRC_IP) | BIT(INNER_DST_IP) |
 	    BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT);
 	hns3_dbg(hw, "fdir tuple: inner<vlan_tag1 eth_type ip_src ip_dst "
-		  "ip_proto ip_tos l4_src_port l4_dst_port>");
+		 "ip_proto ip_tos l4_src_port l4_dst_port>");
 
 	/* If use max 400bit key, we can support tuples for ether type */
 	if (pf->fdir.fd_cfg.max_key_length == MAX_KEY_LENGTH) {
@@ -348,8 +348,8 @@ int hns3_init_fd_config(struct hns3_adapter *hns)
 		    BIT(OUTER_TUN_VNI) | BIT(OUTER_TUN_FLOW_ID) |
 		    BIT(OUTER_ETH_TYPE) | BIT(OUTER_IP_PROTO);
 		hns3_dbg(hw, "fdir tuple more: inner<dst_mac src_mac "
-			  "vlan_tag2 sctp_tag> outer<eth_type ip_proto "
-			  "l4_src_port l4_dst_port tun_vni tun_flow_id>");
+			 "vlan_tag2 sctp_tag> outer<eth_type ip_proto "
+			 "l4_src_port l4_dst_port tun_vni tun_flow_id>");
 	}
 
 	/* roce_type is used to filter roce frames
@@ -367,12 +367,11 @@ int hns3_init_fd_config(struct hns3_adapter *hns)
 	if (ret)
 		return ret;
 
-	hns3_dbg(hw, "fdir: stage1<rules-%u counters-%u> stage2<rules-%u "
-		  "counters=%u>",
-		  pf->fdir.fd_cfg.rule_num[HNS3_FD_STAGE_1],
-		  pf->fdir.fd_cfg.cnt_num[HNS3_FD_STAGE_1],
-		  pf->fdir.fd_cfg.rule_num[HNS3_FD_STAGE_2],
-		  pf->fdir.fd_cfg.cnt_num[HNS3_FD_STAGE_2]);
+	hns3_dbg(hw, "fdir: stage1<rules-%u counters-%u> stage2<rules-%u counters=%u>",
+		 pf->fdir.fd_cfg.rule_num[HNS3_FD_STAGE_1],
+		 pf->fdir.fd_cfg.cnt_num[HNS3_FD_STAGE_1],
+		 pf->fdir.fd_cfg.rule_num[HNS3_FD_STAGE_2],
+		 pf->fdir.fd_cfg.cnt_num[HNS3_FD_STAGE_2]);
 
 	return hns3_set_fd_key_config(hns);
 }
@@ -420,7 +419,7 @@ static int hns3_fd_tcam_config(struct hns3_hw *hw, bool sel_x, int loc,
 	ret = hns3_cmd_send(hw, desc, FD_TCAM_CMD_NUM);
 	if (ret)
 		hns3_err(hw, "Config tcam key fail, ret=%d loc=%d add=%d",
-			    ret, loc, is_add);
+			 ret, loc, is_add);
 	return ret;
 }
 
@@ -673,6 +672,7 @@ static void hns3_fd_convert_meta_data(struct hns3_fd_key_cfg *cfg,
 		} else if (i == VLAN_NUMBER) {
 			uint32_t vlan_tag;
 			uint8_t vlan_num;
+
 			if (rule->key_conf.spec.tunnel_type == 0)
 				vlan_num = rule->key_conf.vlan_num;
 			else
@@ -758,14 +758,14 @@ static int hns3_config_key(struct hns3_adapter *hns,
 	ret = hns3_fd_tcam_config(hw, false, rule->location, key_y, true);
 	if (ret) {
 		hns3_err(hw, "Config fd key_y fail, loc=%u, ret=%d",
-			    rule->queue_id, ret);
+			 rule->queue_id, ret);
 		return ret;
 	}
 
 	ret = hns3_fd_tcam_config(hw, true, rule->location, key_x, true);
 	if (ret)
 		hns3_err(hw, "Config fd key_x fail, loc=%u, ret=%d",
-			    rule->queue_id, ret);
+			 rule->queue_id, ret);
 	return ret;
 }
 
drivers/net/hns3/hns3_flow.c +4 -6
index 1bd4ac160d..fb9354fe7f 100644
@@ -285,9 +285,8 @@ hns3_handle_action_queue(struct rte_eth_dev *dev,
 
 	queue = (const struct rte_flow_action_queue *)action->conf;
 	if (queue->index >= hw->data->nb_rx_queues) {
-		hns3_err(hw, "queue ID(%u) is greater than number of "
-			  "available queue (%u) in driver.",
-			  queue->index, hw->data->nb_rx_queues);
+		hns3_err(hw, "queue ID(%u) is greater than number of available queue (%u) in driver.",
+			 queue->index, hw->data->nb_rx_queues);
 		return rte_flow_error_set(error, EINVAL,
 					  RTE_FLOW_ERROR_TYPE_ACTION_CONF,
 					  action, "Invalid queue ID in PF");
@@ -1656,9 +1655,8 @@ hns3_clear_rss_filter(struct rte_eth_dev *dev)
 	}
 
 	if (rss_rule_fail_cnt) {
-		hns3_err(hw, "fail to delete all RSS filters, success num = %d "
-			     "fail num = %d", rss_rule_succ_cnt,
-			     rss_rule_fail_cnt);
+		hns3_err(hw, "fail to delete all RSS filters, success num = %d fail num = %d",
+			 rss_rule_succ_cnt, rss_rule_fail_cnt);
 		ret = -EIO;
 	}
 
drivers/net/hns3/hns3_rxtx.c +13 -15
index dcb1cdf168..bbf98fc90e 100644
@@ -603,8 +603,8 @@ hns3_send_reset_tqp_cmd(struct hns3_hw *hw, uint16_t queue_id, bool enable)
 	hns3_set_bit(req->reset_req, HNS3_TQP_RESET_B, enable ? 1 : 0);
 	ret = hns3_cmd_send(hw, &desc, 1);
 	if (ret)
-		hns3_err(hw, "send tqp reset cmd error, queue_id = %u, "
-			     "ret = %d", queue_id, ret);
+		hns3_err(hw, "send tqp reset cmd error, queue_id = %u, ret = %d",
+			 queue_id, ret);
 
 	return ret;
 }
@@ -624,8 +624,8 @@ hns3_get_tqp_reset_status(struct hns3_hw *hw, uint16_t queue_id,
 
 	ret = hns3_cmd_send(hw, &desc, 1);
 	if (ret) {
-		hns3_err(hw, "get tqp reset status error, queue_id = %u, "
-			     "ret = %d.", queue_id, ret);
+		hns3_err(hw, "get tqp reset status error, queue_id = %u, ret = %d.",
+			 queue_id, ret);
 		return ret;
 	}
 	*reset_status = hns3_get_bit(req->ready_to_reset, HNS3_TQP_RESET_B);
@@ -666,7 +666,7 @@ hns3pf_reset_tqp(struct hns3_hw *hw, uint16_t queue_id)
 	if (!reset_status) {
 		ret = -ETIMEDOUT;
 		hns3_err(hw, "reset tqp timeout, queue_id = %u, ret = %d",
-			     queue_id, ret);
+			 queue_id, ret);
 		goto tqp_reset_fail;
 	}
 
@@ -749,15 +749,14 @@ hns3pf_reset_all_tqps(struct hns3_hw *hw)
 		for (i = 0; i < hw->cfg_max_queues; i++) {
 			ret = hns3pf_reset_tqp(hw, i);
 			if (ret) {
-				hns3_err(hw,
-				  "fail to reset tqp, queue_id = %d, ret = %d.",
-				  i, ret);
+				hns3_err(hw, "fail to reset tqp, queue_id = %d, ret = %d.",
+					 i, ret);
 				return ret;
 			}
 		}
 	} else if (reset_status != HNS3_RESET_ALL_TQP_SUCCESS) {
 		hns3_err(hw, "fail to reset all tqps, reset_status = %u.",
-				reset_status);
+			 reset_status);
 		return -EIO;
 	}
 
@@ -810,9 +809,8 @@ hns3_reset_all_tqps(struct hns3_adapter *hns)
 	for (i = 0; i < hw->cfg_max_queues; i++) {
 		ret = hns3_tqp_enable(hw, i, false);
 		if (ret) {
-			hns3_err(hw,
-			    "fail to disable tqps before tqps reset, ret = %d.",
-			    ret);
+			hns3_err(hw, "fail to disable tqps before tqps reset, ret = %d.",
+				 ret);
 			return ret;
 		}
 	}
@@ -919,9 +917,9 @@ hns3_reset_queue(struct hns3_hw *hw, uint16_t queue_id,
 	}
 
 	if (!reset_status) {
-		hns3_err(hw, "reset queue timeout, queue_id = %u, "
-			     "queue_type = %s", queue_id,
-			     queue_type == HNS3_RING_TYPE_TX ? "Tx" : "Rx");
+		hns3_err(hw, "reset queue timeout, queue_id = %u, queue_type = %s",
+			 queue_id,
+			 queue_type == HNS3_RING_TYPE_TX ? "Tx" : "Rx");
 		ret = -ETIMEDOUT;
 		goto queue_reset_fail;
 	}
drivers/net/hns3/hns3_stats.c +14 -14
index 07fad03485..a71356934a 100644
@@ -507,8 +507,8 @@ hns3_update_port_rx_ssu_drop_stats(struct hns3_hw *hw)
 
 	req = (struct hns3_query_ssu_cmd *)desc[0].data;
 	cnt = rte_le_to_cpu_32(req->oq_drop_cnt) +
-		rte_le_to_cpu_32(req->full_drop_cnt) +
-		rte_le_to_cpu_32(req->part_drop_cnt);
+	      rte_le_to_cpu_32(req->full_drop_cnt) +
+	      rte_le_to_cpu_32(req->part_drop_cnt);
 
 	stats->ssu_rx_drop_cnt += cnt;
 
@@ -532,8 +532,8 @@ hns3_update_port_tx_ssu_drop_stats(struct hns3_hw *hw)
 
 	req = (struct hns3_query_ssu_cmd *)desc[0].data;
 	cnt = rte_le_to_cpu_32(req->oq_drop_cnt) +
-		rte_le_to_cpu_32(req->full_drop_cnt) +
-		rte_le_to_cpu_32(req->part_drop_cnt);
+	      rte_le_to_cpu_32(req->full_drop_cnt) +
+	      rte_le_to_cpu_32(req->part_drop_cnt);
 
 	hw->oerror_stats += cnt;
 
@@ -1337,8 +1337,8 @@ hns3_dev_xstats_get_by_id(struct rte_eth_dev *dev, const uint64_t *ids,
 	len = cnt_stats * sizeof(struct rte_eth_xstat);
 	values_copy = rte_zmalloc("hns3_xstats_values", len, 0);
 	if (values_copy == NULL) {
-		hns3_err(hw, "Failed to allocate 0x%" PRIx64 " bytes needed "
-			     "to store statistics values", len);
+		hns3_err(hw, "Failed to allocate 0x%" PRIx64 " bytes needed to store statistics values",
+			 len);
 		return -ENOMEM;
 	}
 
@@ -1359,8 +1359,8 @@ hns3_dev_xstats_get_by_id(struct rte_eth_dev *dev, const uint64_t *ids,
 
 	for (i = 0; i < size; i++) {
 		if (ids[i] >= cnt_stats) {
-			hns3_err(hw, "ids[%u] (%" PRIu64 ") is invalid, "
-				     "should < %u", i, ids[i], cnt_stats);
+			hns3_err(hw, "ids[%u] (%" PRIu64 ") is invalid, should < %u",
+				 i, ids[i], cnt_stats);
 			rte_free(values_copy);
 			return -EINVAL;
 		}
@@ -1419,8 +1419,8 @@ hns3_dev_xstats_get_names_by_id(struct rte_eth_dev *dev,
 	len = cnt_stats * sizeof(struct rte_eth_xstat_name);
 	names_copy = rte_zmalloc("hns3_xstats_names", len, 0);
 	if (names_copy == NULL) {
-		hns3_err(hw, "Failed to allocate 0x%" PRIx64 " bytes needed "
-			     "to store statistics names", len);
+		hns3_err(hw, "Failed to allocate 0x%" PRIx64 " bytes needed to store statistics names",
+			 len);
 		return -ENOMEM;
 	}
 
@@ -1428,8 +1428,8 @@ hns3_dev_xstats_get_names_by_id(struct rte_eth_dev *dev,
 
 	for (i = 0; i < size; i++) {
 		if (ids[i] >= cnt_stats) {
-			hns3_err(hw, "ids[%u] (%" PRIu64 ") is invalid, "
-				     "should < %u", i, ids[i], cnt_stats);
+			hns3_err(hw, "ids[%u] (%" PRIu64 ") is invalid, should < %u",
+				 i, ids[i], cnt_stats);
 			rte_free(names_copy);
 			return -EINVAL;
 		}
@@ -1501,14 +1501,14 @@ hns3_tqp_stats_init(struct hns3_hw *hw)
 	struct hns3_tqp_stats *tqp_stats = &hw->tqp_stats;
 
 	tqp_stats->rcb_rx_ring_pktnum = rte_zmalloc("hns3_rx_ring_pkt_num",
-					 sizeof(uint64_t) * hw->tqps_num, 0);
+					sizeof(uint64_t) * hw->tqps_num, 0);
 	if (tqp_stats->rcb_rx_ring_pktnum == NULL) {
 		hns3_err(hw, "failed to allocate rx_ring pkt_num.");
 		return -ENOMEM;
 	}
 
 	tqp_stats->rcb_tx_ring_pktnum = rte_zmalloc("hns3_tx_ring_pkt_num",
-					 sizeof(uint64_t) * hw->tqps_num, 0);
+					sizeof(uint64_t) * hw->tqps_num, 0);
 	if (tqp_stats->rcb_tx_ring_pktnum == NULL) {
 		hns3_err(hw, "failed to allocate tx_ring pkt_num.");
 		rte_free(tqp_stats->rcb_rx_ring_pktnum);
Signature · 1 line
2.33.0
#35 lore raw

PATCH v2 10/11 net/hns3: fix a segfault from secondary process

In reply to Min Hu (Connor)

From: Huisong Li <[email protected]>

If a hns3 device in the secondary process is attached to do probing operation, 'rx_queues' and 'tx_queues' in dev->data are null in eth_dev_fp_ops_setup when calling rte_eth_dev_probing_finish. The primary process calls dev_start to re-setup their fp_ops. But the secondary process can't call dev_start and has no chance to do it. If the application sends and receives packets at this time, a segfault will occur. So this patch uses the MP communication of the PMD to update the fp_ops of the device in the secondary process.

Fixes 96c33cfb06cf ("net/hns3: fix Rx/Tx functions update")
---
 drivers/net/hns3/hns3_rxtx.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/hns3/hns3_rxtx.c +2 -0
index bbf98fc90e..e25a91becf 100644
@@ -4409,6 +4409,8 @@ hns3_eth_dev_fp_ops_config(const struct rte_eth_dev *dev)
 	fpo[port_id].tx_pkt_prepare = dev->tx_pkt_prepare;
 	fpo[port_id].rx_descriptor_status = dev->rx_descriptor_status;
 	fpo[port_id].tx_descriptor_status = dev->tx_descriptor_status;
+	fpo[port_id].rxq.data = dev->data->rx_queues;
+	fpo[port_id].txq.data = dev->data->tx_queues;
 }
 
 void
Signature · 1 line
2.33.0
#36 lore raw

PATCH v2 11/11 net/hns3: fix TM capability incorrectly defined

In reply to Min Hu (Connor)

From: Chengwen Feng <[email protected]>

The TM capability should be bit-19 according to the user manual of firmware. This patch fixes it.

Fixes fc18d1b4b85f ("net/hns3: fix traffic management")
---
 drivers/net/hns3/hns3_cmd.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/hns3/hns3_cmd.h +1 -1
index f9addc6069..82c999061d 100644
@@ -323,7 +323,7 @@ enum HNS3_CAPS_BITS {
 	HNS3_CAPS_UDP_TUNNEL_CSUM_B,
 	HNS3_CAPS_RAS_IMP_B,
 	HNS3_CAPS_RXD_ADV_LAYOUT_B = 15,
-	HNS3_CAPS_TM_B = 17,
+	HNS3_CAPS_TM_B = 19,
 };
 
 /* Capabilities of VF dependent on the PF */
Signature · 1 line
2.33.0
#37 lore raw
In reply to Min Hu (Connor)
Quoted 36 lines · Min Hu (Connor)

This patch set contains 11 bugfix for hns3 PMD.

Chengwen Feng (1):
net/hns3: fix TM capability incorrectly defined

Huisong Li (9):
net/hns3: add check for deferred start queue when rollback net/hns3: remove redundant parentheses
net/hns3: adjust the data type of some variables net/hns3: fix an unreasonable memset
net/hns3: remove duplicate definition
net/hns3: fix return value for unsupported tuple net/hns3: modify a function name
net/hns3: unify the code wrap style
net/hns3: fix a segfault from secondary process

Min Hu (Connor) (1):
net/hns3: fix coverity codecheck

  drivers/net/hns3/hns3_cmd.c       |  2 +-
  drivers/net/hns3/hns3_cmd.h       |  2 +-
  drivers/net/hns3/hns3_common.c    | 14 +++----
  drivers/net/hns3/hns3_dcb.c       |  7 ++--
  drivers/net/hns3/hns3_ethdev.c    | 20 +++++-----
  drivers/net/hns3/hns3_ethdev_vf.c | 23 ++++++------
  drivers/net/hns3/hns3_fdir.c      | 30 +++++++--------
  drivers/net/hns3/hns3_fdir.h      |  2 +-
  drivers/net/hns3/hns3_flow.c      | 25 ++++---------
  drivers/net/hns3/hns3_ptp.c       |  2 +-
  drivers/net/hns3/hns3_regs.c      |  2 +-
  drivers/net/hns3/hns3_rss.c       |  8 ++--
  drivers/net/hns3/hns3_rss.h       |  2 +
  drivers/net/hns3/hns3_rxtx.c      | 62 ++++++++++++++++---------------
  drivers/net/hns3/hns3_rxtx_vec.h  |  4 +-
  drivers/net/hns3/hns3_stats.c     | 28 +++++++-------
  16 files changed, 111 insertions(+), 122 deletions(-)
Quoted 5 lines

---
v2:
* delete one patch which is fixed by others.
refer to:
https://patches.dpdk.org/project/dpdk/cover/[email protected]/

Applied to dpdk-next-net/main, thanks.