This series includes some updates for the HNS3 ethernet driver.
Guangbin Huang (1):
net: hns3: add query vf ring and vector map relation
Hao Chen (1):
net: hns3: fix access null pointer issue when set tx-buf-size as 0
Jie Wang (2):
net: hns3: add byte order conversion for PF to VF mailbox message
net: hns3: add byte order conversion for VF to PF mailbox message
Yufeng Mo (1):
net: hns3: remove the affinity settings of vector0
.../net/ethernet/hisilicon/hns3/hclge_mbx.h | 60 +++++-
.../ethernet/hisilicon/hns3/hns3_ethtool.c | 7 +-
.../hisilicon/hns3/hns3pf/hclge_main.c | 27 +--
.../hisilicon/hns3/hns3pf/hclge_main.h | 2 -
.../hisilicon/hns3/hns3pf/hclge_mbx.c | 193 +++++++++++++-----
.../hisilicon/hns3/hns3vf/hclgevf_main.c | 58 +++---
.../hisilicon/hns3/hns3vf/hclgevf_main.h | 2 +-
.../hisilicon/hns3/hns3vf/hclgevf_mbx.c | 80 +++++---
8 files changed, 269 insertions(+), 160 deletions(-)
--
2.33.0
This patch adds a new mailbox opcode to query map relation between
vf ring and vector.
Signed-off-by: Guangbin Huang <[email protected]>
---
.../net/ethernet/hisilicon/hns3/hclge_mbx.h | 1 +
.../hisilicon/hns3/hns3pf/hclge_mbx.c | 83 +++++++++++++++++++
2 files changed, 84 insertions(+)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h
index 9001a3abc26c..0bff2d834c4b 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h
@@ -46,6 +46,7 @@ enum HCLGE_MBX_OPCODE {
HCLGE_MBX_PUSH_PROMISC_INFO, /* (PF -> VF) push vf promisc info */
HCLGE_MBX_VF_UNINIT, /* (VF -> PF) vf is unintializing */
HCLGE_MBX_HANDLE_VF_TBL, /* (VF -> PF) store/clear hw table */
+ HCLGE_MBX_GET_RING_VECTOR_MAP, /* (VF -> PF) get ring-to-vector map */
HCLGE_MBX_GET_VF_FLR_STATUS = 200, /* (M7 -> PF) get vf flr status */
HCLGE_MBX_PUSH_LINK_STATUS, /* (M7 -> PF) get port link status */
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
index 6957b5e158c9..e1012f7f9b73 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
@@ -251,6 +251,81 @@ static int hclge_map_unmap_ring_to_vf_vector(struct hclge_vport *vport, bool en,
return ret;
}
+static int hclge_query_ring_vector_map(struct hclge_vport *vport,
+ struct hnae3_ring_chain_node *ring_chain,
+ struct hclge_desc *desc)
+{
+ struct hclge_ctrl_vector_chain_cmd *req =
+ (struct hclge_ctrl_vector_chain_cmd *)desc->data;
+ struct hclge_dev *hdev = vport->back;
+ u16 tqp_type_and_id;
+ int status;
+
+ hclge_cmd_setup_basic_desc(desc, HCLGE_OPC_ADD_RING_TO_VECTOR, true);
+
+ tqp_type_and_id = le16_to_cpu(req->tqp_type_and_id[0]);
+ hnae3_set_field(tqp_type_and_id, HCLGE_INT_TYPE_M, HCLGE_INT_TYPE_S,
+ hnae3_get_bit(ring_chain->flag, HNAE3_RING_TYPE_B));
+ hnae3_set_field(tqp_type_and_id, HCLGE_TQP_ID_M, HCLGE_TQP_ID_S,
+ ring_chain->tqp_index);
+ req->tqp_type_and_id[0] = cpu_to_le16(tqp_type_and_id);
+ req->vfid = vport->vport_id;
+
+ status = hclge_cmd_send(&hdev->hw, desc, 1);
+ if (status)
+ dev_err(&hdev->pdev->dev,
+ "Get VF ring vector map info fail, status is %d.\n",
+ status);
+
+ return status;
+}
+
+static int hclge_get_vf_ring_vector_map(struct hclge_vport *vport,
+ struct hclge_mbx_vf_to_pf_cmd *req,
+ struct hclge_respond_to_vf_msg *resp)
+{
+#define HCLGE_LIMIT_RING_NUM 1
+#define HCLGE_RING_TYPE_OFFSET 0
+#define HCLGE_TQP_INDEX_OFFSET 1
+#define HCLGE_INT_GL_INDEX_OFFSET 2
+#define HCLGE_VECTOR_ID_OFFSET 3
+#define HCLGE_RING_VECTOR_MAP_INFO_LEN 4
+ struct hnae3_ring_chain_node ring_chain;
+ struct hclge_desc desc;
+ struct hclge_ctrl_vector_chain_cmd *data =
+ (struct hclge_ctrl_vector_chain_cmd *)desc.data;
+ u16 tqp_type_and_id;
+ u8 int_gl_index;
+ int ret;
+
+ req->msg.ring_num = HCLGE_LIMIT_RING_NUM;
+
+ memset(&ring_chain, 0, sizeof(ring_chain));
+ ret = hclge_get_ring_chain_from_mbx(req, &ring_chain, vport);
+ if (ret)
+ return ret;
+
+ ret = hclge_query_ring_vector_map(vport, &ring_chain, &desc);
+ if (ret) {
+ hclge_free_vector_ring_chain(&ring_chain);
+ return ret;
+ }
+
+ tqp_type_and_id = le16_to_cpu(data->tqp_type_and_id[0]);
+ int_gl_index = hnae3_get_field(tqp_type_and_id,
+ HCLGE_INT_GL_IDX_M, HCLGE_INT_GL_IDX_S);
+
+ resp->data[HCLGE_RING_TYPE_OFFSET] = req->msg.param[0].ring_type;
+ resp->data[HCLGE_TQP_INDEX_OFFSET] = req->msg.param[0].tqp_index;
+ resp->data[HCLGE_INT_GL_INDEX_OFFSET] = int_gl_index;
+ resp->data[HCLGE_VECTOR_ID_OFFSET] = data->int_vector_id_l;
+ resp->len = HCLGE_RING_VECTOR_MAP_INFO_LEN;
+
+ hclge_free_vector_ring_chain(&ring_chain);
+
+ return ret;
+}
+
static void hclge_set_vf_promisc_mode(struct hclge_vport *vport,
struct hclge_mbx_vf_to_pf_cmd *req)
{
@@ -755,6 +830,14 @@ void hclge_mbx_handler(struct hclge_dev *hdev)
ret = hclge_map_unmap_ring_to_vf_vector(vport, false,
req);
break;
+ case HCLGE_MBX_GET_RING_VECTOR_MAP:
+ ret = hclge_get_vf_ring_vector_map(vport, req,
+ &resp_msg);
+ if (ret)
+ dev_err(&hdev->pdev->dev,
+ "PF fail(%d) to get VF ring vector map\n",
+ ret);
+ break;
case HCLGE_MBX_SET_PROMISC_MODE:
hclge_set_vf_promisc_mode(vport, req);
break;
--
2.33.0
From: Jie Wang <[email protected]>
Currently, hns3 mailbox processing between PF and VF missed to convert
message byte order and use data type u16 instead of __le16 for mailbox
data process. These processes may cause problems between different
architectures.
So this patch uses __le16/__le32 data type to define mailbox data
structures. To be compatible with old hns3 driver, these structures use
one-byte alignment. Then byte order conversions are added to mailbox
messages from PF to VF.
Signed-off-by: Jie Wang <[email protected]>
Signed-off-by: Guangbin Huang <[email protected]>
---
.../net/ethernet/hisilicon/hns3/hclge_mbx.h | 34 ++++++--
.../hisilicon/hns3/hns3pf/hclge_mbx.c | 60 +++++++-------
.../hisilicon/hns3/hns3vf/hclgevf_main.c | 4 +-
.../hisilicon/hns3/hns3vf/hclgevf_main.h | 2 +-
.../hisilicon/hns3/hns3vf/hclgevf_mbx.c | 80 +++++++++++--------
5 files changed, 106 insertions(+), 74 deletions(-)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h
index 8c7fadf2b734..b29474f13b23 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h
@@ -134,13 +134,13 @@ struct hclge_vf_to_pf_msg {
};
struct hclge_pf_to_vf_msg {
- u16 code;
+ __le16 code;
union {
/* used for mbx response */
struct {
- u16 vf_mbx_msg_code;
- u16 vf_mbx_msg_subcode;
- u16 resp_status;
+ __le16 vf_mbx_msg_code;
+ __le16 vf_mbx_msg_subcode;
+ __le16 resp_status;
u8 resp_data[HCLGE_MBX_MAX_RESP_DATA_SIZE];
};
/* used for general mbx */
@@ -157,7 +157,7 @@ struct hclge_mbx_vf_to_pf_cmd {
u8 rsv1[1];
u8 msg_len;
u8 rsv2;
- u16 match_id;
+ __le16 match_id;
struct hclge_vf_to_pf_msg msg;
};
@@ -168,7 +168,7 @@ struct hclge_mbx_pf_to_vf_cmd {
u8 rsv[3];
u8 msg_len;
u8 rsv1;
- u16 match_id;
+ __le16 match_id;
struct hclge_pf_to_vf_msg msg;
};
@@ -178,6 +178,28 @@ struct hclge_vf_rst_cmd {
u8 rsv[22];
};
+#pragma pack(1)
+struct hclge_mbx_link_status {
+ __le16 link_status;
+ __le32 speed;
+ __le16 duplex;
+ u8 flag;
+};
+
+struct hclge_mbx_link_mode {
+ __le16 idx;
+ __le64 link_mode;
+};
+
+struct hclge_mbx_port_base_vlan {
+ __le16 state;
+ __le16 vlan_proto;
+ __le16 qos;
+ __le16 vlan_tag;
+};
+
+#pragma pack()
+
/* used by VF to store the received Async responses from PF */
struct hclgevf_mbx_arq_ring {
#define HCLGE_MBX_MAX_ARQ_MSG_SIZE 8
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
index 49c40744cda5..d3c4d2c4a5ba 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
@@ -57,17 +57,19 @@ static int hclge_gen_resp_to_vf(struct hclge_vport *vport,
resp_pf_to_vf->msg_len = vf_to_pf_req->msg_len;
resp_pf_to_vf->match_id = vf_to_pf_req->match_id;
- resp_pf_to_vf->msg.code = HCLGE_MBX_PF_VF_RESP;
- resp_pf_to_vf->msg.vf_mbx_msg_code = vf_to_pf_req->msg.code;
- resp_pf_to_vf->msg.vf_mbx_msg_subcode = vf_to_pf_req->msg.subcode;
+ resp_pf_to_vf->msg.code = cpu_to_le16(HCLGE_MBX_PF_VF_RESP);
+ resp_pf_to_vf->msg.vf_mbx_msg_code =
+ cpu_to_le16(vf_to_pf_req->msg.code);
+ resp_pf_to_vf->msg.vf_mbx_msg_subcode =
+ cpu_to_le16(vf_to_pf_req->msg.subcode);
resp = hclge_errno_to_resp(resp_msg->status);
if (resp < SHRT_MAX) {
- resp_pf_to_vf->msg.resp_status = resp;
+ resp_pf_to_vf->msg.resp_status = cpu_to_le16(resp);
} else {
dev_warn(&hdev->pdev->dev,
"failed to send response to VF, response status %u is out-of-bound\n",
resp);
- resp_pf_to_vf->msg.resp_status = EIO;
+ resp_pf_to_vf->msg.resp_status = cpu_to_le16(EIO);
}
if (resp_msg->len > 0)
@@ -107,7 +109,7 @@ static int hclge_send_mbx_msg(struct hclge_vport *vport, u8 *msg, u16 msg_len,
resp_pf_to_vf->dest_vfid = dest_vfid;
resp_pf_to_vf->msg_len = msg_len;
- resp_pf_to_vf->msg.code = mbx_opcode;
+ resp_pf_to_vf->msg.code = cpu_to_le16(mbx_opcode);
memcpy(resp_pf_to_vf->msg.msg_data, msg, msg_len);
@@ -125,8 +127,8 @@ static int hclge_send_mbx_msg(struct hclge_vport *vport, u8 *msg, u16 msg_len,
int hclge_inform_reset_assert_to_vf(struct hclge_vport *vport)
{
struct hclge_dev *hdev = vport->back;
+ __le16 msg_data;
u16 reset_type;
- u8 msg_data[2];
u8 dest_vfid;
BUILD_BUG_ON(HNAE3_MAX_RESET > U16_MAX);
@@ -140,10 +142,10 @@ int hclge_inform_reset_assert_to_vf(struct hclge_vport *vport)
else
reset_type = HNAE3_VF_FUNC_RESET;
- memcpy(&msg_data[0], &reset_type, sizeof(u16));
+ msg_data = cpu_to_le16(reset_type);
/* send this requested info to VF */
- return hclge_send_mbx_msg(vport, msg_data, sizeof(msg_data),
+ return hclge_send_mbx_msg(vport, (u8 *)&msg_data, sizeof(msg_data),
HCLGE_MBX_ASSERTING_RESET, dest_vfid);
}
@@ -339,16 +341,14 @@ int hclge_push_vf_port_base_vlan_info(struct hclge_vport *vport, u8 vfid,
u16 state,
struct hclge_vlan_info *vlan_info)
{
-#define MSG_DATA_SIZE 8
+ struct hclge_mbx_port_base_vlan base_vlan;
- u8 msg_data[MSG_DATA_SIZE];
+ base_vlan.state = cpu_to_le16(state);
+ base_vlan.vlan_proto = cpu_to_le16(vlan_info->vlan_proto);
+ base_vlan.qos = cpu_to_le16(vlan_info->qos);
+ base_vlan.vlan_tag = cpu_to_le16(vlan_info->vlan_tag);
- memcpy(&msg_data[0], &state, sizeof(u16));
- memcpy(&msg_data[2], &vlan_info->vlan_proto, sizeof(u16));
- memcpy(&msg_data[4], &vlan_info->qos, sizeof(u16));
- memcpy(&msg_data[6], &vlan_info->vlan_tag, sizeof(u16));
-
- return hclge_send_mbx_msg(vport, msg_data, sizeof(msg_data),
+ return hclge_send_mbx_msg(vport, (u8 *)&base_vlan, sizeof(base_vlan),
HCLGE_MBX_PUSH_VLAN_INFO, vfid);
}
@@ -488,10 +488,9 @@ int hclge_push_vf_link_status(struct hclge_vport *vport)
#define HCLGE_VF_LINK_STATE_UP 1U
#define HCLGE_VF_LINK_STATE_DOWN 0U
+ struct hclge_mbx_link_status link_info;
struct hclge_dev *hdev = vport->back;
u16 link_status;
- u8 msg_data[9];
- u16 duplex;
/* mac.link can only be 0 or 1 */
switch (vport->vf_info.link_state) {
@@ -507,14 +506,13 @@ int hclge_push_vf_link_status(struct hclge_vport *vport)
break;
}
- duplex = hdev->hw.mac.duplex;
- memcpy(&msg_data[0], &link_status, sizeof(u16));
- memcpy(&msg_data[2], &hdev->hw.mac.speed, sizeof(u32));
- memcpy(&msg_data[6], &duplex, sizeof(u16));
- msg_data[8] = HCLGE_MBX_PUSH_LINK_STATUS_EN;
+ link_info.link_status = cpu_to_le16(link_status);
+ link_info.speed = cpu_to_le32(hdev->hw.mac.speed);
+ link_info.duplex = cpu_to_le16(hdev->hw.mac.duplex);
+ link_info.flag = HCLGE_MBX_PUSH_LINK_STATUS_EN;
/* send this requested info to VF */
- return hclge_send_mbx_msg(vport, msg_data, sizeof(msg_data),
+ return hclge_send_mbx_msg(vport, (u8 *)&link_info, sizeof(link_info),
HCLGE_MBX_LINK_STAT_CHANGE, vport->vport_id);
}
@@ -522,22 +520,22 @@ static void hclge_get_link_mode(struct hclge_vport *vport,
struct hclge_mbx_vf_to_pf_cmd *mbx_req)
{
#define HCLGE_SUPPORTED 1
+ struct hclge_mbx_link_mode link_mode;
struct hclge_dev *hdev = vport->back;
unsigned long advertising;
unsigned long supported;
unsigned long send_data;
- u8 msg_data[10] = {};
u8 dest_vfid;
advertising = hdev->hw.mac.advertising[0];
supported = hdev->hw.mac.supported[0];
dest_vfid = mbx_req->mbx_src_vfid;
- msg_data[0] = mbx_req->msg.data[0];
-
- send_data = msg_data[0] == HCLGE_SUPPORTED ? supported : advertising;
+ send_data = mbx_req->msg.data[0] == HCLGE_SUPPORTED ? supported :
+ advertising;
+ link_mode.idx = cpu_to_le16((u16)mbx_req->msg.data[0]);
+ link_mode.link_mode = cpu_to_le64(send_data);
- memcpy(&msg_data[2], &send_data, sizeof(unsigned long));
- hclge_send_mbx_msg(vport, msg_data, sizeof(msg_data),
+ hclge_send_mbx_msg(vport, (u8 *)&link_mode, sizeof(link_mode),
HCLGE_MBX_LINK_STAT_MODE, dest_vfid);
}
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index e13d71abd9f7..38a26797c9d4 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -3333,7 +3333,7 @@ static void hclgevf_get_regs(struct hnae3_handle *handle, u32 *version,
}
void hclgevf_update_port_base_vlan_info(struct hclgevf_dev *hdev, u16 state,
- u8 *port_base_vlan_info, u8 data_size)
+ struct hclge_mbx_port_base_vlan *port_base_vlan)
{
struct hnae3_handle *nic = &hdev->nic;
struct hclge_vf_to_pf_msg send_msg;
@@ -3358,7 +3358,7 @@ void hclgevf_update_port_base_vlan_info(struct hclgevf_dev *hdev, u16 state,
/* send msg to PF and wait update port based vlan info */
hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN,
HCLGE_MBX_PORT_BASE_VLAN_CFG);
- memcpy(send_msg.data, port_base_vlan_info, data_size);
+ memcpy(send_msg.data, port_base_vlan, sizeof(*port_base_vlan));
ret = hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
if (!ret) {
if (state == HNAE3_PORT_BASE_VLAN_DISABLE)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
index 4b00fd44f118..59ca6c794d6d 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
@@ -293,5 +293,5 @@ void hclgevf_update_speed_duplex(struct hclgevf_dev *hdev, u32 speed,
void hclgevf_reset_task_schedule(struct hclgevf_dev *hdev);
void hclgevf_mbx_task_schedule(struct hclgevf_dev *hdev);
void hclgevf_update_port_base_vlan_info(struct hclgevf_dev *hdev, u16 state,
- u8 *port_base_vlan_info, u8 data_size);
+ struct hclge_mbx_port_base_vlan *port_base_vlan);
#endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
index c8055d69255c..a291d022777b 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
@@ -124,7 +124,7 @@ int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev,
if (need_resp) {
mutex_lock(&hdev->mbx_resp.mbx_mutex);
hclgevf_reset_mbx_resp_status(hdev);
- req->match_id = hdev->mbx_resp.match_id;
+ req->match_id = cpu_to_le16(hdev->mbx_resp.match_id);
status = hclgevf_cmd_send(&hdev->hw, &desc, 1);
if (status) {
dev_err(&hdev->pdev->dev,
@@ -162,27 +162,29 @@ static bool hclgevf_cmd_crq_empty(struct hclgevf_hw *hw)
static void hclgevf_handle_mbx_response(struct hclgevf_dev *hdev,
struct hclge_mbx_pf_to_vf_cmd *req)
{
+ u16 vf_mbx_msg_subcode = le16_to_cpu(req->msg.vf_mbx_msg_subcode);
+ u16 vf_mbx_msg_code = le16_to_cpu(req->msg.vf_mbx_msg_code);
struct hclgevf_mbx_resp_status *resp = &hdev->mbx_resp;
+ u16 resp_status = le16_to_cpu(req->msg.resp_status);
+ u16 match_id = le16_to_cpu(req->match_id);
if (resp->received_resp)
dev_warn(&hdev->pdev->dev,
- "VF mbx resp flag not clear(%u)\n",
- req->msg.vf_mbx_msg_code);
-
- resp->origin_mbx_msg =
- (req->msg.vf_mbx_msg_code << 16);
- resp->origin_mbx_msg |= req->msg.vf_mbx_msg_subcode;
- resp->resp_status =
- hclgevf_resp_to_errno(req->msg.resp_status);
+ "VF mbx resp flag not clear(%u)\n",
+ vf_mbx_msg_code);
+
+ resp->origin_mbx_msg = (vf_mbx_msg_code << 16);
+ resp->origin_mbx_msg |= vf_mbx_msg_subcode;
+ resp->resp_status = hclgevf_resp_to_errno(resp_status);
memcpy(resp->additional_info, req->msg.resp_data,
HCLGE_MBX_MAX_RESP_DATA_SIZE * sizeof(u8));
- if (req->match_id) {
+ if (match_id) {
/* If match_id is not zero, it means PF support match_id.
* if the match_id is right, VF get the right response, or
* ignore the response. and driver will clear hdev->mbx_resp
* when send next message which need response.
*/
- if (req->match_id == resp->match_id)
+ if (match_id == resp->match_id)
resp->received_resp = true;
} else {
resp->received_resp = true;
@@ -199,7 +201,7 @@ static void hclgevf_handle_mbx_msg(struct hclgevf_dev *hdev,
HCLGE_MBX_MAX_ARQ_MSG_NUM) {
dev_warn(&hdev->pdev->dev,
"Async Q full, dropping msg(%u)\n",
- req->msg.code);
+ le16_to_cpu(req->msg.code));
return;
}
@@ -218,6 +220,7 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev)
struct hclge_comm_cmq_ring *crq;
struct hclge_desc *desc;
u16 flag;
+ u16 code;
crq = &hdev->hw.hw.cmq.crq;
@@ -232,10 +235,11 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev)
req = (struct hclge_mbx_pf_to_vf_cmd *)desc->data;
flag = le16_to_cpu(crq->desc[crq->next_to_use].flag);
+ code = le16_to_cpu(req->msg.code);
if (unlikely(!hnae3_get_bit(flag, HCLGEVF_CMDQ_RX_OUTVLD_B))) {
dev_warn(&hdev->pdev->dev,
"dropped invalid mailbox message, code = %u\n",
- req->msg.code);
+ code);
/* dropping/not processing this invalid message */
crq->desc[crq->next_to_use].flag = 0;
@@ -251,7 +255,7 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev)
* timeout and simultaneously queue the async messages for later
* prcessing in context of mailbox task i.e. the slow path.
*/
- switch (req->msg.code) {
+ switch (code) {
case HCLGE_MBX_PF_VF_RESP:
hclgevf_handle_mbx_response(hdev, req);
break;
@@ -265,7 +269,7 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev)
default:
dev_err(&hdev->pdev->dev,
"VF received unsupported(%u) mbx msg from PF\n",
- req->msg.code);
+ code);
break;
}
crq->desc[crq->next_to_use].flag = 0;
@@ -287,14 +291,18 @@ static void hclgevf_parse_promisc_info(struct hclgevf_dev *hdev,
void hclgevf_mbx_async_handler(struct hclgevf_dev *hdev)
{
+ struct hclge_mbx_port_base_vlan *vlan_info;
+ struct hclge_mbx_link_status *link_info;
+ struct hclge_mbx_link_mode *link_mode;
enum hnae3_reset_type reset_type;
u16 link_status, state;
- u16 *msg_q, *vlan_info;
+ u16 *msg_q;
+ u16 opcode;
u8 duplex;
u32 speed;
u32 tail;
u8 flag;
- u8 idx;
+ u16 idx;
tail = hdev->arq.tail;
@@ -308,13 +316,14 @@ void hclgevf_mbx_async_handler(struct hclgevf_dev *hdev)
}
msg_q = hdev->arq.msg_q[hdev->arq.head];
-
- switch (msg_q[0]) {
+ opcode = le16_to_cpu(msg_q[0]);
+ switch (opcode) {
case HCLGE_MBX_LINK_STAT_CHANGE:
- link_status = msg_q[1];
- memcpy(&speed, &msg_q[2], sizeof(speed));
- duplex = (u8)msg_q[4];
- flag = (u8)msg_q[5];
+ link_info = (struct hclge_mbx_link_status *)(msg_q + 1);
+ link_status = le16_to_cpu(link_info->link_status);
+ speed = le32_to_cpu(link_info->speed);
+ duplex = (u8)le16_to_cpu(link_info->duplex);
+ flag = link_info->flag;
/* update upper layer with new link link status */
hclgevf_update_speed_duplex(hdev, speed, duplex);
@@ -326,13 +335,14 @@ void hclgevf_mbx_async_handler(struct hclgevf_dev *hdev)
break;
case HCLGE_MBX_LINK_STAT_MODE:
- idx = (u8)msg_q[1];
+ link_mode = (struct hclge_mbx_link_mode *)(msg_q + 1);
+ idx = le16_to_cpu(link_mode->idx);
if (idx)
- memcpy(&hdev->hw.mac.supported, &msg_q[2],
- sizeof(unsigned long));
+ hdev->hw.mac.supported =
+ le64_to_cpu(link_mode->link_mode);
else
- memcpy(&hdev->hw.mac.advertising, &msg_q[2],
- sizeof(unsigned long));
+ hdev->hw.mac.advertising =
+ le64_to_cpu(link_mode->link_mode);
break;
case HCLGE_MBX_ASSERTING_RESET:
/* PF has asserted reset hence VF should go in pending
@@ -340,25 +350,27 @@ void hclgevf_mbx_async_handler(struct hclgevf_dev *hdev)
* has been completely reset. After this stack should
* eventually be re-initialized.
*/
- reset_type = (enum hnae3_reset_type)msg_q[1];
+ reset_type =
+ (enum hnae3_reset_type)le16_to_cpu(msg_q[1]);
set_bit(reset_type, &hdev->reset_pending);
set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state);
hclgevf_reset_task_schedule(hdev);
break;
case HCLGE_MBX_PUSH_VLAN_INFO:
- state = msg_q[1];
- vlan_info = &msg_q[1];
+ vlan_info =
+ (struct hclge_mbx_port_base_vlan *)(msg_q + 1);
+ state = le16_to_cpu(vlan_info->state);
hclgevf_update_port_base_vlan_info(hdev, state,
- (u8 *)vlan_info, 8);
+ vlan_info);
break;
case HCLGE_MBX_PUSH_PROMISC_INFO:
- hclgevf_parse_promisc_info(hdev, msg_q[1]);
+ hclgevf_parse_promisc_info(hdev, le16_to_cpu(msg_q[1]));
break;
default:
dev_err(&hdev->pdev->dev,
"fetched unsupported(%u) message from arq\n",
- msg_q[0]);
+ opcode);
break;
}
--
2.33.0
From: Yufeng Mo <[email protected]>
Vector0 is used for common interrupt control events and is
irrelevant to performance. Currently, the driver sets the
default affinity of vector0 to NUMA nodes, which is unnecessary.
Therefore, the default setting is removed, and the driver does
not set the affinity for vector0.
Signed-off-by: Yufeng Mo <[email protected]>
Signed-off-by: Guangbin Huang <[email protected]>
---
.../hisilicon/hns3/hns3pf/hclge_main.c | 27 +------------------
.../hisilicon/hns3/hns3pf/hclge_main.h | 2 --
2 files changed, 1 insertion(+), 28 deletions(-)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index a5dd2c8c244a..1ebad0e50e6a 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -1546,9 +1546,8 @@ static void hclge_init_tc_config(struct hclge_dev *hdev)
static int hclge_configure(struct hclge_dev *hdev)
{
struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
- const struct cpumask *cpumask = cpu_online_mask;
struct hclge_cfg cfg;
- int node, ret;
+ int ret;
ret = hclge_get_cfg(hdev, &cfg);
if (ret)
@@ -1594,13 +1593,6 @@ static int hclge_configure(struct hclge_dev *hdev)
hclge_init_tc_config(hdev);
hclge_init_kdump_kernel_config(hdev);
- /* Set the affinity based on numa node */
- node = dev_to_node(&hdev->pdev->dev);
- if (node != NUMA_NO_NODE)
- cpumask = cpumask_of_node(node);
-
- cpumask_copy(&hdev->affinity_mask, cpumask);
-
return ret;
}
@@ -3564,17 +3556,6 @@ static void hclge_get_misc_vector(struct hclge_dev *hdev)
hdev->num_msi_used += 1;
}
-static void hclge_misc_affinity_setup(struct hclge_dev *hdev)
-{
- irq_set_affinity_hint(hdev->misc_vector.vector_irq,
- &hdev->affinity_mask);
-}
-
-static void hclge_misc_affinity_teardown(struct hclge_dev *hdev)
-{
- irq_set_affinity_hint(hdev->misc_vector.vector_irq, NULL);
-}
-
static int hclge_misc_irq_init(struct hclge_dev *hdev)
{
int ret;
@@ -11457,11 +11438,6 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
timer_setup(&hdev->reset_timer, hclge_reset_timer, 0);
INIT_DELAYED_WORK(&hdev->service_task, hclge_service_task);
- /* Setup affinity after service timer setup because add_timer_on
- * is called in affinity notify.
- */
- hclge_misc_affinity_setup(hdev);
-
hclge_clear_all_event_cause(hdev);
hclge_clear_resetting_state(hdev);
@@ -11879,7 +11855,6 @@ static void hclge_uninit_ae_dev(struct hnae3_ae_dev *ae_dev)
hclge_reset_vf_rate(hdev);
hclge_clear_vf_vlan(hdev);
- hclge_misc_affinity_teardown(hdev);
hclge_state_uninit(hdev);
hclge_ptp_uninit(hdev);
hclge_uninit_rxd_adv_layout(hdev);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
index c70239758bb2..ab5c37848a7b 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
@@ -938,8 +938,6 @@ struct hclge_dev {
DECLARE_KFIFO(mac_tnl_log, struct hclge_mac_tnl_stats,
HCLGE_MAC_TNL_LOG_SIZE);
- /* affinity mask and notify for misc interrupt */
- cpumask_t affinity_mask;
struct hclge_ptp *ptp;
struct devlink *devlink;
struct hclge_comm_rss_cfg rss_cfg;
--
2.33.0
On Thu, 5 May 2022 20:44:42 +0800 Guangbin Huang wrote:
> From: Jie Wang <[email protected]>
>
> Currently, hns3 mailbox processing between PF and VF missed to convert
> message byte order and use data type u16 instead of __le16 for mailbox
> data process. These processes may cause problems between different
> architectures.
>
> So this patch uses __le16/__le32 data type to define mailbox data
> structures. To be compatible with old hns3 driver, these structures use
> one-byte alignment. Then byte order conversions are added to mailbox
> messages from PF to VF.
This adds a few sparse [1] warnings, you must have missed a few
conversions.
Please wait at least 24h for additional feedback before reposting.
[1] https://www.kernel.org/doc/html/latest/dev-tools/sparse.html
On 2022/5/6 9:47, Jakub Kicinski wrote:
> On Thu, 5 May 2022 20:44:42 +0800 Guangbin Huang wrote:
>> From: Jie Wang <[email protected]>
>>
>> Currently, hns3 mailbox processing between PF and VF missed to convert
>> message byte order and use data type u16 instead of __le16 for mailbox
>> data process. These processes may cause problems between different
>> architectures.
>>
>> So this patch uses __le16/__le32 data type to define mailbox data
>> structures. To be compatible with old hns3 driver, these structures use
>> one-byte alignment. Then byte order conversions are added to mailbox
>> messages from PF to VF.
>
> This adds a few sparse [1] warnings, you must have missed a few
> conversions.
>
> Please wait at least 24h for additional feedback before reposting.
>
> [1] https://www.kernel.org/doc/html/latest/dev-tools/sparse.html
>
> .
>
Ok, thanks