* [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver
@ 2019-04-14 1:47 Huazhong Tan
2019-04-14 1:47 ` [PATCH V2 net-next 01/12] net: hns3: modify VLAN initialization to be compatible with port based VLAN Huazhong Tan
` (12 more replies)
0 siblings, 13 replies; 14+ messages in thread
From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw)
To: davem
Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Huazhong Tan
This patch-set includes code optimizations and bugfixes for the HNS3
ethernet controller driver.
[patch 1/12 - 4/12] optimizes the VLAN freature and adds support for port
based VLAN, fixes some related bugs about the current implementation.
[patch 5/12 - 12/12] includes some other code optimizations for the HNS3
ethernet controller driver.
Change log:
V1->V2: modifies some patches' commint log and code.
Jian Shen (5):
net: hns3: modify VLAN initialization to be compatible with port based
VLAN
net: hns3: fix VLAN offload handle for VLAN inserted by port
net: hns3: fix set port based VLAN for PF
net: hns3: fix set port based VLAN issue for VF
net: hns3: do not initialize MDIO bus when PHY is inexistent
Peng Li (2):
net: hns3: free the pending skb when clean RX ring
net: hns3: code optimization for command queue' spin lock
Weihang Li (1):
net: hns3: set dividual reset level for all RAS and MSI-X errors
Yunsheng Lin (4):
net: hns3: minor refactor for hns3_rx_checksum
net: hns3: add hns3_gro_complete for HW GRO process
net: hns3: always assume no drop TC for performance reason
net: hns3: divide shared buffer between TC
drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h | 3 +
drivers/net/ethernet/hisilicon/hns3/hnae3.h | 9 +
drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 256 ++++--
.../net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c | 4 +-
.../net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c | 956 +++++++++++++--------
.../net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h | 1 +
.../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 378 ++++++--
.../ethernet/hisilicon/hns3/hns3pf/hclge_main.h | 29 +-
.../net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c | 42 +-
.../ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c | 8 +-
.../ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c | 4 +-
.../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 51 ++
.../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h | 2 +
.../ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c | 11 +-
14 files changed, 1224 insertions(+), 530 deletions(-)
--
2.7.4
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH V2 net-next 01/12] net: hns3: modify VLAN initialization to be compatible with port based VLAN 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 02/12] net: hns3: fix VLAN offload handle for VLAN inserted by port Huazhong Tan ` (11 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Jian Shen, Huazhong Tan From: Jian Shen <shenjian15@huawei.com> Our hardware supports inserting a specified VLAN header for each function when sending packets. User can enable it with command "ip link set <devname> vf <vfid> vlan <vlan id>". For this VLAN header is inserted by hardware, not from stack, hardware also needs to strip it from received packets before sending to stack. In this case, driver needs to tell hardware which VLAN to insert or strip. The current VLAN initialization doesn't allow inserting VLAN header by hardware, this patch modifies it, in order be compatible with VLAN inserted base on port. Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hnae3.h | 7 ++ .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 80 +++++++++++++++------- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.h | 21 ++++-- 3 files changed, 78 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h index 38b430f..f21932c 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h +++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h @@ -147,6 +147,13 @@ enum hnae3_flr_state { HNAE3_FLR_DONE, }; +enum hnae3_port_base_vlan_state { + HNAE3_PORT_BASE_VLAN_DISABLE, + HNAE3_PORT_BASE_VLAN_ENABLE, + HNAE3_PORT_BASE_VLAN_MODIFY, + HNAE3_PORT_BASE_VLAN_NOCHANGE, +}; + struct hnae3_vector_info { u8 __iomem *io_addr; int vector; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 11d9739..630f7f88 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -1358,6 +1358,8 @@ static int hclge_alloc_vport(struct hclge_dev *hdev) vport->back = hdev; vport->vport_id = i; vport->mps = HCLGE_MAC_DEFAULT_FRAME; + vport->port_base_vlan_cfg.state = HNAE3_PORT_BASE_VLAN_DISABLE; + vport->rxvlan_cfg.rx_vlan_offload_en = true; INIT_LIST_HEAD(&vport->vlan_list); INIT_LIST_HEAD(&vport->uc_mac_list); INIT_LIST_HEAD(&vport->mc_mac_list); @@ -6680,6 +6682,52 @@ static int hclge_set_vlan_rx_offload_cfg(struct hclge_vport *vport) return status; } +static int hclge_vlan_offload_cfg(struct hclge_vport *vport, + u16 port_base_vlan_state, + u16 vlan_tag) +{ + int ret; + + if (port_base_vlan_state == HNAE3_PORT_BASE_VLAN_DISABLE) { + vport->txvlan_cfg.accept_tag1 = true; + vport->txvlan_cfg.insert_tag1_en = false; + vport->txvlan_cfg.default_tag1 = 0; + } else { + vport->txvlan_cfg.accept_tag1 = false; + vport->txvlan_cfg.insert_tag1_en = true; + vport->txvlan_cfg.default_tag1 = vlan_tag; + } + + vport->txvlan_cfg.accept_untag1 = true; + + /* accept_tag2 and accept_untag2 are not supported on + * pdev revision(0x20), new revision support them, + * this two fields can not be configured by user. + */ + vport->txvlan_cfg.accept_tag2 = true; + vport->txvlan_cfg.accept_untag2 = true; + vport->txvlan_cfg.insert_tag2_en = false; + vport->txvlan_cfg.default_tag2 = 0; + + if (port_base_vlan_state == HNAE3_PORT_BASE_VLAN_DISABLE) { + vport->rxvlan_cfg.strip_tag1_en = false; + vport->rxvlan_cfg.strip_tag2_en = + vport->rxvlan_cfg.rx_vlan_offload_en; + } else { + vport->rxvlan_cfg.strip_tag1_en = + vport->rxvlan_cfg.rx_vlan_offload_en; + vport->rxvlan_cfg.strip_tag2_en = true; + } + vport->rxvlan_cfg.vlan1_vlan_prionly = false; + vport->rxvlan_cfg.vlan2_vlan_prionly = false; + + ret = hclge_set_vlan_tx_offload_cfg(vport); + if (ret) + return ret; + + return hclge_set_vlan_rx_offload_cfg(vport); +} + static int hclge_set_vlan_protocol_type(struct hclge_dev *hdev) { struct hclge_rx_vlan_type_cfg_cmd *rx_req; @@ -6770,34 +6818,14 @@ static int hclge_init_vlan_config(struct hclge_dev *hdev) return ret; for (i = 0; i < hdev->num_alloc_vport; i++) { - vport = &hdev->vport[i]; - vport->txvlan_cfg.accept_tag1 = true; - vport->txvlan_cfg.accept_untag1 = true; - - /* accept_tag2 and accept_untag2 are not supported on - * pdev revision(0x20), new revision support them. The - * value of this two fields will not return error when driver - * send command to fireware in revision(0x20). - * This two fields can not configured by user. - */ - vport->txvlan_cfg.accept_tag2 = true; - vport->txvlan_cfg.accept_untag2 = true; + u16 vlan_tag; - vport->txvlan_cfg.insert_tag1_en = false; - vport->txvlan_cfg.insert_tag2_en = false; - vport->txvlan_cfg.default_tag1 = 0; - vport->txvlan_cfg.default_tag2 = 0; - - ret = hclge_set_vlan_tx_offload_cfg(vport); - if (ret) - return ret; - - vport->rxvlan_cfg.strip_tag1_en = false; - vport->rxvlan_cfg.strip_tag2_en = true; - vport->rxvlan_cfg.vlan1_vlan_prionly = false; - vport->rxvlan_cfg.vlan2_vlan_prionly = false; + vport = &hdev->vport[i]; + vlan_tag = vport->port_base_vlan_cfg.vlan_info.vlan_tag; - ret = hclge_set_vlan_rx_offload_cfg(vport); + ret = hclge_vlan_offload_cfg(vport, + vport->port_base_vlan_cfg.state, + vlan_tag); if (ret) return ret; } diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h index b57ac4b..4c7ea9f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h @@ -807,10 +807,11 @@ struct hclge_tx_vtag_cfg { /* VPort level vlan tag configuration for RX direction */ struct hclge_rx_vtag_cfg { - bool strip_tag1_en; /* Whether strip inner vlan tag */ - bool strip_tag2_en; /* Whether strip outer vlan tag */ - bool vlan1_vlan_prionly;/* Inner VLAN Tag up to descriptor Enable */ - bool vlan2_vlan_prionly;/* Outer VLAN Tag up to descriptor Enable */ + u8 rx_vlan_offload_en; /* Whether enable rx vlan offload */ + u8 strip_tag1_en; /* Whether strip inner vlan tag */ + u8 strip_tag2_en; /* Whether strip outer vlan tag */ + u8 vlan1_vlan_prionly; /* Inner VLAN Tag up to descriptor Enable */ + u8 vlan2_vlan_prionly; /* Outer VLAN Tag up to descriptor Enable */ }; struct hclge_rss_tuple_cfg { @@ -829,6 +830,17 @@ enum HCLGE_VPORT_STATE { HCLGE_VPORT_STATE_MAX }; +struct hclge_vlan_info { + u16 vlan_proto; /* so far support 802.1Q only */ + u16 qos; + u16 vlan_tag; +}; + +struct hclge_port_base_vlan_config { + u16 state; + struct hclge_vlan_info vlan_info; +}; + struct hclge_vport { u16 alloc_tqps; /* Allocated Tx/Rx queues */ @@ -845,6 +857,7 @@ struct hclge_vport { u16 bw_limit; /* VSI BW Limit (0 = disabled) */ u8 dwrr; + struct hclge_port_base_vlan_config port_base_vlan_cfg; struct hclge_tx_vtag_cfg txvlan_cfg; struct hclge_rx_vtag_cfg rxvlan_cfg; -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 02/12] net: hns3: fix VLAN offload handle for VLAN inserted by port 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 01/12] net: hns3: modify VLAN initialization to be compatible with port based VLAN Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 03/12] net: hns3: fix set port based VLAN for PF Huazhong Tan ` (10 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Jian Shen, Huazhong Tan From: Jian Shen <shenjian15@huawei.com> Currently, in TX direction, driver implements the TX VLAN offload by checking the VLAN header in skb, and filling it into TX descriptor. Usually it works well, but if enable inserting VLAN header based on port, it may conflict when out_tag field of TX descriptor is already used, and cause RAS error. In RX direction, hardware supports stripping max two VLAN headers. For vlan_tci in skb can only store one VLAN tag, when RX VLAN offload enabled, driver tells hardware to strip one VLAN header from RX packet; when RX VLAN offload disabled, driver tells hardware not to strip VLAN header from RX packet. Now if port based insert VLAN enabled, all RX packets will have the port based VLAN header. This header is useless for stack, driver needs to ask hardware to strip it. Unfortunately, hardware can't drop this VLAN header, and always fill it into RX descriptor, so driver has to identify and drop it. Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hnae3.h | 2 + drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 44 +++++++++++++++++++++- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 10 ++++- 3 files changed, 52 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h index f21932c..681c175 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h +++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h @@ -585,6 +585,8 @@ struct hnae3_handle { u32 numa_node_mask; /* for multi-chip support */ + enum hnae3_port_base_vlan_state port_base_vlan_state; + u8 netdev_flags; struct dentry *hnae3_dbgfs; }; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index b53b091..ec16b94 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -963,6 +963,16 @@ static int hns3_fill_desc_vtags(struct sk_buff *skb, { #define HNS3_TX_VLAN_PRIO_SHIFT 13 + struct hnae3_handle *handle = tx_ring->tqp->handle; + + /* Since HW limitation, if port based insert VLAN enabled, only one VLAN + * header is allowed in skb, otherwise it will cause RAS error. + */ + if (unlikely(skb_vlan_tagged_multi(skb) && + handle->port_base_vlan_state == + HNAE3_PORT_BASE_VLAN_ENABLE)) + return -EINVAL; + if (skb->protocol == htons(ETH_P_8021Q) && !(tx_ring->tqp->handle->kinfo.netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) { @@ -984,8 +994,16 @@ static int hns3_fill_desc_vtags(struct sk_buff *skb, * and use inner_vtag in one tag case. */ if (skb->protocol == htons(ETH_P_8021Q)) { - hns3_set_field(*out_vlan_flag, HNS3_TXD_OVLAN_B, 1); - *out_vtag = vlan_tag; + if (handle->port_base_vlan_state == + HNAE3_PORT_BASE_VLAN_DISABLE){ + hns3_set_field(*out_vlan_flag, + HNS3_TXD_OVLAN_B, 1); + *out_vtag = vlan_tag; + } else { + hns3_set_field(*inner_vlan_flag, + HNS3_TXD_VLAN_B, 1); + *inner_vtag = vlan_tag; + } } else { hns3_set_field(*inner_vlan_flag, HNS3_TXD_VLAN_B, 1); *inner_vtag = vlan_tag; @@ -2390,6 +2408,7 @@ static bool hns3_parse_vlan_tag(struct hns3_enet_ring *ring, struct hns3_desc *desc, u32 l234info, u16 *vlan_tag) { + struct hnae3_handle *handle = ring->tqp->handle; struct pci_dev *pdev = ring->tqp->handle->pdev; if (pdev->revision == 0x20) { @@ -2402,15 +2421,36 @@ static bool hns3_parse_vlan_tag(struct hns3_enet_ring *ring, #define HNS3_STRP_OUTER_VLAN 0x1 #define HNS3_STRP_INNER_VLAN 0x2 +#define HNS3_STRP_BOTH 0x3 + /* Hardware always insert VLAN tag into RX descriptor when + * remove the tag from packet, driver needs to determine + * reporting which tag to stack. + */ switch (hnae3_get_field(l234info, HNS3_RXD_STRP_TAGP_M, HNS3_RXD_STRP_TAGP_S)) { case HNS3_STRP_OUTER_VLAN: + if (handle->port_base_vlan_state != + HNAE3_PORT_BASE_VLAN_DISABLE) + return false; + *vlan_tag = le16_to_cpu(desc->rx.ot_vlan_tag); return true; case HNS3_STRP_INNER_VLAN: + if (handle->port_base_vlan_state != + HNAE3_PORT_BASE_VLAN_DISABLE) + return false; + *vlan_tag = le16_to_cpu(desc->rx.vlan_tag); return true; + case HNS3_STRP_BOTH: + if (handle->port_base_vlan_state == + HNAE3_PORT_BASE_VLAN_DISABLE) + *vlan_tag = le16_to_cpu(desc->rx.ot_vlan_tag); + else + *vlan_tag = le16_to_cpu(desc->rx.vlan_tag); + + return true; default: return false; } diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 630f7f88..4bc2c07 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -6915,10 +6915,16 @@ int hclge_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable) { struct hclge_vport *vport = hclge_get_vport(handle); - vport->rxvlan_cfg.strip_tag1_en = false; - vport->rxvlan_cfg.strip_tag2_en = enable; + if (vport->port_base_vlan_cfg.state == HNAE3_PORT_BASE_VLAN_DISABLE) { + vport->rxvlan_cfg.strip_tag1_en = false; + vport->rxvlan_cfg.strip_tag2_en = enable; + } else { + vport->rxvlan_cfg.strip_tag1_en = enable; + vport->rxvlan_cfg.strip_tag2_en = true; + } vport->rxvlan_cfg.vlan1_vlan_prionly = false; vport->rxvlan_cfg.vlan2_vlan_prionly = false; + vport->rxvlan_cfg.rx_vlan_offload_en = enable; return hclge_set_vlan_rx_offload_cfg(vport); } -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 03/12] net: hns3: fix set port based VLAN for PF 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 01/12] net: hns3: modify VLAN initialization to be compatible with port based VLAN Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 02/12] net: hns3: fix VLAN offload handle for VLAN inserted by port Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 04/12] net: hns3: fix set port based VLAN issue for VF Huazhong Tan ` (9 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Jian Shen, Huazhong Tan From: Jian Shen <shenjian15@huawei.com> In original codes, ndo_set_vf_vlan() in hns3 driver was implemented wrong. It adds or removes VLAN into VLAN filter for VF, but VF is unaware of it. Indeed, ndo_set_vf_vlan() is expected to enable or disable port based VLAN (hardware inserts a specified VLAN tag to all TX packets for a specified VF) . When enable port based VLAN, we use port based VLAN id as VLAN filter entry. When disable port based VLAN, we use VLAN id of VLAN device. This patch fixes it for PF, enable/disable port based VLAN when calls ndo_set_vf_vlan(). Fixes: 46a3df9f9718 ("net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support") Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 229 ++++++++++++++++++--- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.h | 5 +- .../net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c | 3 - 3 files changed, 203 insertions(+), 34 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 4bc2c07..0a99a97 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -6585,30 +6585,6 @@ static int hclge_set_vlan_filter_hw(struct hclge_dev *hdev, __be16 proto, return ret; } -int hclge_set_vlan_filter(struct hnae3_handle *handle, __be16 proto, - u16 vlan_id, bool is_kill) -{ - struct hclge_vport *vport = hclge_get_vport(handle); - struct hclge_dev *hdev = vport->back; - - return hclge_set_vlan_filter_hw(hdev, proto, vport->vport_id, vlan_id, - 0, is_kill); -} - -static int hclge_set_vf_vlan_filter(struct hnae3_handle *handle, int vfid, - u16 vlan, u8 qos, __be16 proto) -{ - struct hclge_vport *vport = hclge_get_vport(handle); - struct hclge_dev *hdev = vport->back; - - if ((vfid >= hdev->num_alloc_vfs) || (vlan > 4095) || (qos > 7)) - return -EINVAL; - if (proto != htons(ETH_P_8021Q)) - return -EPROTONOSUPPORT; - - return hclge_set_vlan_filter_hw(hdev, proto, vfid, vlan, qos, false); -} - static int hclge_set_vlan_tx_offload_cfg(struct hclge_vport *vport) { struct hclge_tx_vtag_cfg *vcfg = &vport->txvlan_cfg; @@ -6833,7 +6809,8 @@ static int hclge_init_vlan_config(struct hclge_dev *hdev) return hclge_set_vlan_filter(handle, htons(ETH_P_8021Q), 0, false); } -void hclge_add_vport_vlan_table(struct hclge_vport *vport, u16 vlan_id) +static void hclge_add_vport_vlan_table(struct hclge_vport *vport, u16 vlan_id, + bool writen_to_tbl) { struct hclge_vport_vlan_cfg *vlan; @@ -6845,14 +6822,38 @@ void hclge_add_vport_vlan_table(struct hclge_vport *vport, u16 vlan_id) if (!vlan) return; - vlan->hd_tbl_status = true; + vlan->hd_tbl_status = writen_to_tbl; vlan->vlan_id = vlan_id; list_add_tail(&vlan->node, &vport->vlan_list); } -void hclge_rm_vport_vlan_table(struct hclge_vport *vport, u16 vlan_id, - bool is_write_tbl) +static int hclge_add_vport_all_vlan_table(struct hclge_vport *vport) +{ + struct hclge_vport_vlan_cfg *vlan, *tmp; + struct hclge_dev *hdev = vport->back; + int ret; + + list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) { + if (!vlan->hd_tbl_status) { + ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q), + vport->vport_id, + vlan->vlan_id, 0, false); + if (ret) { + dev_err(&hdev->pdev->dev, + "restore vport vlan list failed, ret=%d\n", + ret); + return ret; + } + } + vlan->hd_tbl_status = true; + } + + return 0; +} + +static void hclge_rm_vport_vlan_table(struct hclge_vport *vport, u16 vlan_id, + bool is_write_tbl) { struct hclge_vport_vlan_cfg *vlan, *tmp; struct hclge_dev *hdev = vport->back; @@ -6929,6 +6930,178 @@ int hclge_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable) return hclge_set_vlan_rx_offload_cfg(vport); } +static int hclge_update_vlan_filter_entries(struct hclge_vport *vport, + u16 port_base_vlan_state, + struct hclge_vlan_info *new_info, + struct hclge_vlan_info *old_info) +{ + struct hclge_dev *hdev = vport->back; + int ret; + + if (port_base_vlan_state == HNAE3_PORT_BASE_VLAN_ENABLE) { + hclge_rm_vport_all_vlan_table(vport, false); + return hclge_set_vlan_filter_hw(hdev, + htons(new_info->vlan_proto), + vport->vport_id, + new_info->vlan_tag, + new_info->qos, false); + } + + ret = hclge_set_vlan_filter_hw(hdev, htons(old_info->vlan_proto), + vport->vport_id, old_info->vlan_tag, + old_info->qos, true); + if (ret) + return ret; + + return hclge_add_vport_all_vlan_table(vport); +} + +int hclge_update_port_base_vlan_cfg(struct hclge_vport *vport, u16 state, + struct hclge_vlan_info *vlan_info) +{ + struct hnae3_handle *nic = &vport->nic; + struct hclge_vlan_info *old_vlan_info; + struct hclge_dev *hdev = vport->back; + int ret; + + old_vlan_info = &vport->port_base_vlan_cfg.vlan_info; + + ret = hclge_vlan_offload_cfg(vport, state, vlan_info->vlan_tag); + if (ret) + return ret; + + if (state == HNAE3_PORT_BASE_VLAN_MODIFY) { + /* add new VLAN tag */ + ret = hclge_set_vlan_filter_hw(hdev, vlan_info->vlan_proto, + vport->vport_id, + vlan_info->vlan_tag, + vlan_info->qos, false); + if (ret) + return ret; + + /* remove old VLAN tag */ + ret = hclge_set_vlan_filter_hw(hdev, old_vlan_info->vlan_proto, + vport->vport_id, + old_vlan_info->vlan_tag, + old_vlan_info->qos, true); + if (ret) + return ret; + + goto update; + } + + ret = hclge_update_vlan_filter_entries(vport, state, vlan_info, + old_vlan_info); + if (ret) + return ret; + + /* update state only when disable/enable port based VLAN */ + vport->port_base_vlan_cfg.state = state; + if (state == HNAE3_PORT_BASE_VLAN_DISABLE) + nic->port_base_vlan_state = HNAE3_PORT_BASE_VLAN_DISABLE; + else + nic->port_base_vlan_state = HNAE3_PORT_BASE_VLAN_ENABLE; + +update: + vport->port_base_vlan_cfg.vlan_info.vlan_tag = vlan_info->vlan_tag; + vport->port_base_vlan_cfg.vlan_info.qos = vlan_info->qos; + vport->port_base_vlan_cfg.vlan_info.vlan_proto = vlan_info->vlan_proto; + + return 0; +} + +static u16 hclge_get_port_base_vlan_state(struct hclge_vport *vport, + enum hnae3_port_base_vlan_state state, + u16 vlan) +{ + if (state == HNAE3_PORT_BASE_VLAN_DISABLE) { + if (!vlan) + return HNAE3_PORT_BASE_VLAN_NOCHANGE; + else + return HNAE3_PORT_BASE_VLAN_ENABLE; + } else { + if (!vlan) + return HNAE3_PORT_BASE_VLAN_DISABLE; + else if (vport->port_base_vlan_cfg.vlan_info.vlan_tag == vlan) + return HNAE3_PORT_BASE_VLAN_NOCHANGE; + else + return HNAE3_PORT_BASE_VLAN_MODIFY; + } +} + +static int hclge_set_vf_vlan_filter(struct hnae3_handle *handle, int vfid, + u16 vlan, u8 qos, __be16 proto) +{ + struct hclge_vport *vport = hclge_get_vport(handle); + struct hclge_dev *hdev = vport->back; + struct hclge_vlan_info vlan_info; + u16 state; + int ret; + + if (hdev->pdev->revision == 0x20) + return -EOPNOTSUPP; + + /* qos is a 3 bits value, so can not be bigger than 7 */ + if (vfid >= hdev->num_alloc_vfs || vlan > VLAN_N_VID - 1 || qos > 7) + return -EINVAL; + if (proto != htons(ETH_P_8021Q)) + return -EPROTONOSUPPORT; + + vport = &hdev->vport[vfid]; + state = hclge_get_port_base_vlan_state(vport, + vport->port_base_vlan_cfg.state, + vlan); + if (state == HNAE3_PORT_BASE_VLAN_NOCHANGE) + return 0; + + vlan_info.vlan_tag = vlan; + vlan_info.qos = qos; + vlan_info.vlan_proto = ntohs(proto); + + /* update port based VLAN for PF */ + if (!vfid) { + hclge_notify_client(hdev, HNAE3_DOWN_CLIENT); + ret = hclge_update_port_base_vlan_cfg(vport, state, &vlan_info); + hclge_notify_client(hdev, HNAE3_UP_CLIENT); + + return ret; + } + + return -EOPNOTSUPP; +} + +int hclge_set_vlan_filter(struct hnae3_handle *handle, __be16 proto, + u16 vlan_id, bool is_kill) +{ + struct hclge_vport *vport = hclge_get_vport(handle); + struct hclge_dev *hdev = vport->back; + bool writen_to_tbl = false; + int ret = 0; + + /* when port based VLAN enabled, we use port based VLAN as the VLAN + * filter entry. In this case, we don't update VLAN filter table + * when user add new VLAN or remove exist VLAN, just update the vport + * VLAN list. The VLAN id in VLAN list won't be writen in VLAN filter + * table until port based VLAN disabled + */ + if (handle->port_base_vlan_state == HNAE3_PORT_BASE_VLAN_DISABLE) { + ret = hclge_set_vlan_filter_hw(hdev, proto, vport->vport_id, + vlan_id, 0, is_kill); + writen_to_tbl = true; + } + + if (ret) + return ret; + + if (is_kill) + hclge_rm_vport_vlan_table(vport, vlan_id, false); + else + hclge_add_vport_vlan_table(vport, vlan_id, + writen_to_tbl); + + return 0; +} + static int hclge_set_mac_mtu(struct hclge_dev *hdev, int new_mps) { struct hclge_config_max_frm_size_cmd *req; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h index 4c7ea9f..45c9467 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h @@ -937,9 +937,8 @@ void hclge_rm_vport_mac_table(struct hclge_vport *vport, const u8 *mac_addr, void hclge_rm_vport_all_mac_table(struct hclge_vport *vport, bool is_del_list, enum HCLGE_MAC_ADDR_TYPE mac_type); void hclge_uninit_vport_mac_table(struct hclge_dev *hdev); -void hclge_add_vport_vlan_table(struct hclge_vport *vport, u16 vlan_id); -void hclge_rm_vport_vlan_table(struct hclge_vport *vport, u16 vlan_id, - bool is_write_tbl); void hclge_rm_vport_all_vlan_table(struct hclge_vport *vport, bool is_del_list); void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev); +int hclge_update_port_base_vlan_cfg(struct hclge_vport *vport, u16 state, + struct hclge_vlan_info *vlan_info); #endif diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c index 9aa9c64..fddbbca 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c @@ -305,9 +305,6 @@ static int hclge_set_vf_vlan_cfg(struct hclge_vport *vport, memcpy(&proto, &mbx_req->msg[5], sizeof(proto)); status = hclge_set_vlan_filter(handle, cpu_to_be16(proto), vlan, is_kill); - if (!status) - is_kill ? hclge_rm_vport_vlan_table(vport, vlan, false) - : hclge_add_vport_vlan_table(vport, vlan); } else if (mbx_req->msg[1] == HCLGE_MBX_VLAN_RX_OFF_CFG) { struct hnae3_handle *handle = &vport->nic; bool en = mbx_req->msg[2] ? true : false; -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 04/12] net: hns3: fix set port based VLAN issue for VF 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (2 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 03/12] net: hns3: fix set port based VLAN for PF Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 05/12] net: hns3: minor refactor for hns3_rx_checksum Huazhong Tan ` (8 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Jian Shen, Huazhong Tan From: Jian Shen <shenjian15@huawei.com> In original codes, ndo_set_vf_vlan() in hns3 driver was implemented wrong. It adds or removes VLAN into VLAN filter for VF, but VF is unaware of it. This patch fixes it. When VF loads up, it firstly queries the port based VLAN state from PF. When user change port based VLAN state from PF, PF firstly checks whether the VF is alive. If the VF is alive, then PF notifies the VF the modification; otherwise PF configure the port based VLAN state directly. Fixes: 46a3df9f9718 ("net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support") Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h | 3 ++ .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 11 ++++- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.h | 3 ++ .../net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c | 39 ++++++++++++++--- .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 51 ++++++++++++++++++++++ .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h | 2 + .../ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c | 11 ++++- 7 files changed, 111 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h index 9d3d45f..360463a 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h +++ b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h @@ -43,6 +43,7 @@ enum HCLGE_MBX_OPCODE { HCLGE_MBX_GET_QID_IN_PF, /* (VF -> PF) get queue id in pf */ HCLGE_MBX_LINK_STAT_MODE, /* (PF -> VF) link mode has changed */ HCLGE_MBX_GET_LINK_MODE, /* (VF -> PF) get the link mode of pf */ + HLCGE_MBX_PUSH_VLAN_INFO, /* (PF -> VF) push port base vlan */ HCLGE_MBX_GET_MEDIA_TYPE, /* (VF -> PF) get media type */ HCLGE_MBX_GET_VF_FLR_STATUS = 200, /* (M7 -> PF) get vf reset status */ @@ -63,6 +64,8 @@ enum hclge_mbx_vlan_cfg_subcode { HCLGE_MBX_VLAN_FILTER = 0, /* set vlan filter */ HCLGE_MBX_VLAN_TX_OFF_CFG, /* set tx side vlan offload */ HCLGE_MBX_VLAN_RX_OFF_CFG, /* set rx side vlan offload */ + HCLGE_MBX_PORT_BASE_VLAN_CFG, /* set port based vlan configuration */ + HCLGE_MBX_GET_PORT_BASE_VLAN_STATE, /* get port based vlan state */ }; #define HCLGE_MBX_MAX_MSG_SIZE 16 diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 0a99a97..0d1cd3f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -7067,7 +7067,16 @@ static int hclge_set_vf_vlan_filter(struct hnae3_handle *handle, int vfid, return ret; } - return -EOPNOTSUPP; + if (!test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) { + return hclge_update_port_base_vlan_cfg(vport, state, + &vlan_info); + } else { + ret = hclge_push_vf_port_base_vlan_info(&hdev->vport[0], + (u8)vfid, state, + vlan, qos, + ntohs(proto)); + return ret; + } } int hclge_set_vlan_filter(struct hnae3_handle *handle, __be16 proto, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h index 45c9467..f04a52f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h @@ -941,4 +941,7 @@ void hclge_rm_vport_all_vlan_table(struct hclge_vport *vport, bool is_del_list); void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev); int hclge_update_port_base_vlan_cfg(struct hclge_vport *vport, u16 state, struct hclge_vlan_info *vlan_info); +int hclge_push_vf_port_base_vlan_info(struct hclge_vport *vport, u8 vfid, + u16 state, u16 vlan_tag, u16 qos, + u16 vlan_proto); #endif diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c index fddbbca..24386bd 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c @@ -289,9 +289,25 @@ static int hclge_set_vf_mc_mac_addr(struct hclge_vport *vport, return 0; } +int hclge_push_vf_port_base_vlan_info(struct hclge_vport *vport, u8 vfid, + u16 state, u16 vlan_tag, u16 qos, + u16 vlan_proto) +{ +#define MSG_DATA_SIZE 8 + + u8 msg_data[MSG_DATA_SIZE]; + + memcpy(&msg_data[0], &state, sizeof(u16)); + memcpy(&msg_data[2], &vlan_proto, sizeof(u16)); + memcpy(&msg_data[4], &qos, sizeof(u16)); + memcpy(&msg_data[6], &vlan_tag, sizeof(u16)); + + return hclge_send_mbx_msg(vport, msg_data, sizeof(msg_data), + HLCGE_MBX_PUSH_VLAN_INFO, vfid); +} + static int hclge_set_vf_vlan_cfg(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req, - bool gen_resp) + struct hclge_mbx_vf_to_pf_cmd *mbx_req) { int status = 0; @@ -310,11 +326,22 @@ static int hclge_set_vf_vlan_cfg(struct hclge_vport *vport, bool en = mbx_req->msg[2] ? true : false; status = hclge_en_hw_strip_rxvtag(handle, en); + } else if (mbx_req->msg[1] == HCLGE_MBX_PORT_BASE_VLAN_CFG) { + struct hclge_vlan_info *vlan_info; + u16 *state; + + state = (u16 *)&mbx_req->msg[2]; + vlan_info = (struct hclge_vlan_info *)&mbx_req->msg[4]; + status = hclge_update_port_base_vlan_cfg(vport, *state, + vlan_info); + } else if (mbx_req->msg[1] == HCLGE_MBX_GET_PORT_BASE_VLAN_STATE) { + u8 state; + + state = vport->port_base_vlan_cfg.state; + status = hclge_gen_resp_to_vf(vport, mbx_req, 0, &state, + sizeof(u8)); } - if (gen_resp) - status = hclge_gen_resp_to_vf(vport, mbx_req, status, NULL, 0); - return status; } @@ -584,7 +611,7 @@ void hclge_mbx_handler(struct hclge_dev *hdev) ret); break; case HCLGE_MBX_SET_VLAN: - ret = hclge_set_vf_vlan_cfg(vport, req, false); + ret = hclge_set_vf_vlan_cfg(vport, req); if (ret) dev_err(&hdev->pdev->dev, "PF failed(%d) to config VF's VLAN\n", diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index b917960..2e277c9 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -245,6 +245,27 @@ static int hclgevf_get_tc_info(struct hclgevf_dev *hdev) return 0; } +static int hclgevf_get_port_base_vlan_filter_state(struct hclgevf_dev *hdev) +{ + struct hnae3_handle *nic = &hdev->nic; + u8 resp_msg; + int ret; + + ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_VLAN, + HCLGE_MBX_GET_PORT_BASE_VLAN_STATE, + NULL, 0, true, &resp_msg, sizeof(u8)); + if (ret) { + dev_err(&hdev->pdev->dev, + "VF request to get port based vlan state failed %d", + ret); + return ret; + } + + nic->port_base_vlan_state = resp_msg; + + return 0; +} + static int hclgevf_get_queue_info(struct hclgevf_dev *hdev) { #define HCLGEVF_TQPS_RSS_INFO_LEN 6 @@ -1834,6 +1855,11 @@ static int hclgevf_configure(struct hclgevf_dev *hdev) { int ret; + /* get current port based vlan state from PF */ + ret = hclgevf_get_port_base_vlan_filter_state(hdev); + if (ret) + return ret; + /* get queue configuration from PF */ ret = hclgevf_get_queue_info(hdev); if (ret) @@ -2790,6 +2816,31 @@ 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 hnae3_handle *nic = &hdev->nic; + + rtnl_lock(); + hclgevf_notify_client(hdev, HNAE3_DOWN_CLIENT); + rtnl_unlock(); + + /* send msg to PF and wait update port based vlan info */ + hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_VLAN, + HCLGE_MBX_PORT_BASE_VLAN_CFG, + port_base_vlan_info, data_size, + false, NULL, 0); + + if (state == HNAE3_PORT_BASE_VLAN_DISABLE) + nic->port_base_vlan_state = HNAE3_PORT_BASE_VLAN_DISABLE; + else + nic->port_base_vlan_state = HNAE3_PORT_BASE_VLAN_ENABLE; + + rtnl_lock(); + hclgevf_notify_client(hdev, HNAE3_UP_CLIENT); + rtnl_unlock(); +} + static const struct hnae3_ae_ops hclgevf_ops = { .init_ae_dev = hclgevf_init_ae_dev, .uninit_ae_dev = hclgevf_uninit_ae_dev, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h index c128863..49e5bec 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h @@ -290,4 +290,6 @@ void hclgevf_update_speed_duplex(struct hclgevf_dev *hdev, u32 speed, u8 duplex); 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); #endif diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c index 3bcf49b..bf57084 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c @@ -198,6 +198,7 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev) case HCLGE_MBX_LINK_STAT_CHANGE: case HCLGE_MBX_ASSERTING_RESET: case HCLGE_MBX_LINK_STAT_MODE: + case HLCGE_MBX_PUSH_VLAN_INFO: /* set this mbx event as pending. This is required as we * might loose interrupt event when mbx task is busy * handling. This shall be cleared when mbx task just @@ -243,8 +244,8 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev) void hclgevf_mbx_async_handler(struct hclgevf_dev *hdev) { enum hnae3_reset_type reset_type; - u16 link_status; - u16 *msg_q; + u16 link_status, state; + u16 *msg_q, *vlan_info; u8 duplex; u32 speed; u32 tail; @@ -299,6 +300,12 @@ void hclgevf_mbx_async_handler(struct hclgevf_dev *hdev) hclgevf_reset_task_schedule(hdev); break; + case HLCGE_MBX_PUSH_VLAN_INFO: + state = le16_to_cpu(msg_q[1]); + vlan_info = &msg_q[1]; + hclgevf_update_port_base_vlan_info(hdev, state, + (u8 *)vlan_info, 8); + break; default: dev_err(&hdev->pdev->dev, "fetched unsupported(%d) message from arq\n", -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 05/12] net: hns3: minor refactor for hns3_rx_checksum 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (3 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 04/12] net: hns3: fix set port based VLAN issue for VF Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 06/12] net: hns3: add hns3_gro_complete for HW GRO process Huazhong Tan ` (7 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Yunsheng Lin, Huazhong Tan From: Yunsheng Lin <linyunsheng@huawei.com> Change the parameters of hns3_rx_checksum to be more specific to what is used internally, rather than passing in a pointer to the whole hns3_desc. Reduces duplicate code and bring this function inline with the approach used in hns3_set_gro_param. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index ec16b94..60afa15 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -2332,16 +2332,11 @@ static void hns3_nic_reuse_page(struct sk_buff *skb, int i, } static void hns3_rx_checksum(struct hns3_enet_ring *ring, struct sk_buff *skb, - struct hns3_desc *desc) + u32 l234info, u32 bd_base_info) { struct net_device *netdev = ring->tqp->handle->kinfo.netdev; int l3_type, l4_type; - u32 bd_base_info; int ol4_type; - u32 l234info; - - bd_base_info = le32_to_cpu(desc->rx.bd_base_info); - l234info = le32_to_cpu(desc->rx.l234_info); skb->ip_summed = CHECKSUM_NONE; @@ -2751,7 +2746,7 @@ static int hns3_handle_rx_bd(struct hns3_enet_ring *ring, /* This is needed in order to enable forwarding support */ hns3_set_gro_param(skb, l234info, bd_base_info); - hns3_rx_checksum(ring, skb, desc); + hns3_rx_checksum(ring, skb, l234info, bd_base_info); *out_skb = skb; hns3_set_rx_skb_rss_type(ring, skb); -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 06/12] net: hns3: add hns3_gro_complete for HW GRO process 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (4 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 05/12] net: hns3: minor refactor for hns3_rx_checksum Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 07/12] net: hns3: always assume no drop TC for performance reason Huazhong Tan ` (6 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Yunsheng Lin, Huazhong Tan From: Yunsheng Lin <linyunsheng@huawei.com> When a GRO packet is received by driver, the cwr field in the struct tcphdr needs to be checked to decide whether to set the SKB_GSO_TCP_ECN for skb_shinfo(skb)->gso_type. So this patch adds hns3_gro_complete to do that, and adds the hns3_handle_bdinfo to handle the hns3_gro_complete and hns3_rx_checksum. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 198 +++++++++++++++--------- 1 file changed, 123 insertions(+), 75 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index 60afa15..a67dfdd 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -2331,6 +2331,44 @@ static void hns3_nic_reuse_page(struct sk_buff *skb, int i, } } +static int hns3_gro_complete(struct sk_buff *skb) +{ + __be16 type = skb->protocol; + struct tcphdr *th; + int depth = 0; + + while (type == htons(ETH_P_8021Q)) { + struct vlan_hdr *vh; + + if ((depth + VLAN_HLEN) > skb_headlen(skb)) + return -EFAULT; + + vh = (struct vlan_hdr *)(skb->data + depth); + type = vh->h_vlan_encapsulated_proto; + depth += VLAN_HLEN; + } + + if (type == htons(ETH_P_IP)) { + depth += sizeof(struct iphdr); + } else if (type == htons(ETH_P_IPV6)) { + depth += sizeof(struct ipv6hdr); + } else { + netdev_err(skb->dev, + "Error: FW GRO supports only IPv4/IPv6, not 0x%04x, depth: %d\n", + be16_to_cpu(type), depth); + return -EFAULT; + } + + th = (struct tcphdr *)(skb->data + depth); + skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count; + if (th->cwr) + skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN; + + skb->ip_summed = CHECKSUM_UNNECESSARY; + + return 0; +} + static void hns3_rx_checksum(struct hns3_enet_ring *ring, struct sk_buff *skb, u32 l234info, u32 bd_base_info) { @@ -2345,12 +2383,6 @@ static void hns3_rx_checksum(struct hns3_enet_ring *ring, struct sk_buff *skb, if (!(netdev->features & NETIF_F_RXCSUM)) return; - /* We MUST enable hardware checksum before enabling hardware GRO */ - if (skb_shinfo(skb)->gso_size) { - skb->ip_summed = CHECKSUM_UNNECESSARY; - return; - } - /* check if hardware has done checksum */ if (!(bd_base_info & BIT(HNS3_RXD_L3L4P_B))) return; @@ -2567,8 +2599,9 @@ static int hns3_add_frag(struct hns3_enet_ring *ring, struct hns3_desc *desc, return 0; } -static void hns3_set_gro_param(struct sk_buff *skb, u32 l234info, - u32 bd_base_info) +static int hns3_set_gro_and_checksum(struct hns3_enet_ring *ring, + struct sk_buff *skb, u32 l234info, + u32 bd_base_info) { u16 gro_count; u32 l3_type; @@ -2576,12 +2609,11 @@ static void hns3_set_gro_param(struct sk_buff *skb, u32 l234info, gro_count = hnae3_get_field(l234info, HNS3_RXD_GRO_COUNT_M, HNS3_RXD_GRO_COUNT_S); /* if there is no HW GRO, do not set gro params */ - if (!gro_count) - return; + if (!gro_count) { + hns3_rx_checksum(ring, skb, l234info, bd_base_info); + return 0; + } - /* tcp_gro_complete() will copy NAPI_GRO_CB(skb)->count - * to skb_shinfo(skb)->gso_segs - */ NAPI_GRO_CB(skb)->count = gro_count; l3_type = hnae3_get_field(l234info, HNS3_RXD_L3ID_M, @@ -2591,13 +2623,13 @@ static void hns3_set_gro_param(struct sk_buff *skb, u32 l234info, else if (l3_type == HNS3_L3_TYPE_IPV6) skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6; else - return; + return -EFAULT; skb_shinfo(skb)->gso_size = hnae3_get_field(bd_base_info, HNS3_RXD_GRO_SIZE_M, HNS3_RXD_GRO_SIZE_S); - if (skb_shinfo(skb)->gso_size) - tcp_gro_complete(skb); + + return hns3_gro_complete(skb); } static void hns3_set_rx_skb_rss_type(struct hns3_enet_ring *ring, @@ -2622,16 +2654,85 @@ static void hns3_set_rx_skb_rss_type(struct hns3_enet_ring *ring, skb_set_hash(skb, le32_to_cpu(desc->rx.rss_hash), rss_type); } -static int hns3_handle_rx_bd(struct hns3_enet_ring *ring, - struct sk_buff **out_skb) +static int hns3_handle_bdinfo(struct hns3_enet_ring *ring, struct sk_buff *skb, + struct hns3_desc *desc) { struct net_device *netdev = ring->tqp->handle->kinfo.netdev; + u32 bd_base_info = le32_to_cpu(desc->rx.bd_base_info); + u32 l234info = le32_to_cpu(desc->rx.l234_info); enum hns3_pkt_l2t_type l2_frame_type; + unsigned int len; + int ret; + + /* Based on hw strategy, the tag offloaded will be stored at + * ot_vlan_tag in two layer tag case, and stored at vlan_tag + * in one layer tag case. + */ + if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX) { + u16 vlan_tag; + + if (hns3_parse_vlan_tag(ring, desc, l234info, &vlan_tag)) + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), + vlan_tag); + } + + if (unlikely(!(bd_base_info & BIT(HNS3_RXD_VLD_B)))) { + u64_stats_update_begin(&ring->syncp); + ring->stats.non_vld_descs++; + u64_stats_update_end(&ring->syncp); + + return -EINVAL; + } + + if (unlikely(!desc->rx.pkt_len || (l234info & (BIT(HNS3_RXD_TRUNCAT_B) | + BIT(HNS3_RXD_L2E_B))))) { + u64_stats_update_begin(&ring->syncp); + if (l234info & BIT(HNS3_RXD_L2E_B)) + ring->stats.l2_err++; + else + ring->stats.err_pkt_len++; + u64_stats_update_end(&ring->syncp); + + return -EFAULT; + } + + len = skb->len; + + /* Do update ip stack process */ + skb->protocol = eth_type_trans(skb, netdev); + + /* This is needed in order to enable forwarding support */ + ret = hns3_set_gro_and_checksum(ring, skb, l234info, bd_base_info); + if (unlikely(ret)) { + u64_stats_update_begin(&ring->syncp); + ring->stats.rx_err_cnt++; + u64_stats_update_end(&ring->syncp); + return ret; + } + + l2_frame_type = hnae3_get_field(l234info, HNS3_RXD_DMAC_M, + HNS3_RXD_DMAC_S); + + u64_stats_update_begin(&ring->syncp); + ring->stats.rx_pkts++; + ring->stats.rx_bytes += len; + + if (l2_frame_type == HNS3_L2_TYPE_MULTICAST) + ring->stats.rx_multicast++; + + u64_stats_update_end(&ring->syncp); + + ring->tqp_vector->rx_group.total_bytes += len; + return 0; +} + +static int hns3_handle_rx_bd(struct hns3_enet_ring *ring, + struct sk_buff **out_skb) +{ struct sk_buff *skb = ring->skb; struct hns3_desc_cb *desc_cb; struct hns3_desc *desc; u32 bd_base_info; - u32 l234info; int length; int ret; @@ -2691,62 +2792,12 @@ static int hns3_handle_rx_bd(struct hns3_enet_ring *ring, ALIGN(ring->pull_len, sizeof(long))); } - l234info = le32_to_cpu(desc->rx.l234_info); - bd_base_info = le32_to_cpu(desc->rx.bd_base_info); - - /* Based on hw strategy, the tag offloaded will be stored at - * ot_vlan_tag in two layer tag case, and stored at vlan_tag - * in one layer tag case. - */ - if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX) { - u16 vlan_tag; - - if (hns3_parse_vlan_tag(ring, desc, l234info, &vlan_tag)) - __vlan_hwaccel_put_tag(skb, - htons(ETH_P_8021Q), - vlan_tag); - } - - if (unlikely(!(bd_base_info & BIT(HNS3_RXD_VLD_B)))) { - u64_stats_update_begin(&ring->syncp); - ring->stats.non_vld_descs++; - u64_stats_update_end(&ring->syncp); - - dev_kfree_skb_any(skb); - return -EINVAL; - } - - if (unlikely((!desc->rx.pkt_len) || - (l234info & (BIT(HNS3_RXD_TRUNCAT_B) | - BIT(HNS3_RXD_L2E_B))))) { - u64_stats_update_begin(&ring->syncp); - if (l234info & BIT(HNS3_RXD_L2E_B)) - ring->stats.l2_err++; - else - ring->stats.err_pkt_len++; - u64_stats_update_end(&ring->syncp); - + ret = hns3_handle_bdinfo(ring, skb, desc); + if (unlikely(ret)) { dev_kfree_skb_any(skb); - return -EFAULT; + return ret; } - - l2_frame_type = hnae3_get_field(l234info, HNS3_RXD_DMAC_M, - HNS3_RXD_DMAC_S); - u64_stats_update_begin(&ring->syncp); - if (l2_frame_type == HNS3_L2_TYPE_MULTICAST) - ring->stats.rx_multicast++; - - ring->stats.rx_pkts++; - ring->stats.rx_bytes += skb->len; - u64_stats_update_end(&ring->syncp); - - ring->tqp_vector->rx_group.total_bytes += skb->len; - - /* This is needed in order to enable forwarding support */ - hns3_set_gro_param(skb, l234info, bd_base_info); - - hns3_rx_checksum(ring, skb, l234info, bd_base_info); *out_skb = skb; hns3_set_rx_skb_rss_type(ring, skb); @@ -2758,7 +2809,6 @@ int hns3_clean_rx_ring( void (*rx_fn)(struct hns3_enet_ring *, struct sk_buff *)) { #define RCB_NOF_ALLOC_RX_BUFF_ONCE 16 - struct net_device *netdev = ring->tqp->handle->kinfo.netdev; int recv_pkts, recv_bds, clean_count, err; int unused_count = hns3_desc_unused(ring) - ring->pending_buf; struct sk_buff *skb = ring->skb; @@ -2795,8 +2845,6 @@ int hns3_clean_rx_ring( continue; } - /* Do update ip stack process */ - skb->protocol = eth_type_trans(skb, netdev); rx_fn(ring, skb); recv_bds += ring->pending_buf; clean_count += ring->pending_buf; -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 07/12] net: hns3: always assume no drop TC for performance reason 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (5 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 06/12] net: hns3: add hns3_gro_complete for HW GRO process Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 08/12] net: hns3: divide shared buffer between TC Huazhong Tan ` (5 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Yunsheng Lin, Peng Li, Huazhong Tan From: Yunsheng Lin <linyunsheng@huawei.com> Currently RX shared buffer' threshold size for speific TC is set to smaller value when the TC's PFC is not enabled, which may cause performance problem because hardware may not have enough hardware buffer when PFC is not enabled. This patch sets the same threshold size for all TC no matter if the specific TC's PFC is enabled. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Peng Li <lipeng321@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 29 +++------------------- 1 file changed, 4 insertions(+), 25 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 0d1cd3f..7e7cdad 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -1432,17 +1432,6 @@ static int hclge_get_tc_num(struct hclge_dev *hdev) return cnt; } -static int hclge_get_pfc_enalbe_num(struct hclge_dev *hdev) -{ - int i, cnt = 0; - - for (i = 0; i < HCLGE_MAX_TC_NUM; i++) - if (hdev->hw_tc_map & BIT(i) && - hdev->tm_info.hw_pfc_map & BIT(i)) - cnt++; - return cnt; -} - /* Get the number of pfc enabled TCs, which have private buffer */ static int hclge_get_pfc_priv_num(struct hclge_dev *hdev, struct hclge_pkt_buf_alloc *buf_alloc) @@ -1507,13 +1496,11 @@ static bool hclge_is_rx_buf_ok(struct hclge_dev *hdev, u32 rx_all) { u32 shared_buf_min, shared_buf_tc, shared_std; - int tc_num, pfc_enable_num; + int tc_num = hclge_get_tc_num(hdev); u32 shared_buf, aligned_mps; u32 rx_priv; int i; - tc_num = hclge_get_tc_num(hdev); - pfc_enable_num = hclge_get_pfc_enalbe_num(hdev); aligned_mps = roundup(hdev->mps, HCLGE_BUF_SIZE_UNIT); if (hnae3_dev_dcb_supported(hdev)) @@ -1522,9 +1509,7 @@ static bool hclge_is_rx_buf_ok(struct hclge_dev *hdev, shared_buf_min = aligned_mps + HCLGE_NON_DCB_ADDITIONAL_BUF + hdev->dv_buf_size; - shared_buf_tc = pfc_enable_num * aligned_mps + - (tc_num - pfc_enable_num) * aligned_mps / 2 + - aligned_mps; + shared_buf_tc = tc_num * aligned_mps + aligned_mps; shared_std = roundup(max_t(u32, shared_buf_min, shared_buf_tc), HCLGE_BUF_SIZE_UNIT); @@ -1546,14 +1531,8 @@ static bool hclge_is_rx_buf_ok(struct hclge_dev *hdev, } for (i = 0; i < HCLGE_MAX_TC_NUM; i++) { - if ((hdev->hw_tc_map & BIT(i)) && - (hdev->tm_info.hw_pfc_map & BIT(i))) { - buf_alloc->s_buf.tc_thrd[i].low = aligned_mps; - buf_alloc->s_buf.tc_thrd[i].high = 2 * aligned_mps; - } else { - buf_alloc->s_buf.tc_thrd[i].low = 0; - buf_alloc->s_buf.tc_thrd[i].high = aligned_mps; - } + buf_alloc->s_buf.tc_thrd[i].low = aligned_mps; + buf_alloc->s_buf.tc_thrd[i].high = 2 * aligned_mps; } return true; -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 08/12] net: hns3: divide shared buffer between TC 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (6 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 07/12] net: hns3: always assume no drop TC for performance reason Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 09/12] net: hns3: set dividual reset level for all RAS and MSI-X errors Huazhong Tan ` (4 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Yunsheng Lin, Huazhong Tan From: Yunsheng Lin <linyunsheng@huawei.com> Currently hardware may have not enough buffer to receive packet when it has used more than two MPS(maximum packet size) of buffer, but there are still a lot of shared buffer left unused when TC num is small. This patch divides shared buffer to be used between TC when the port supports DCB, and adjusts the waterline and threshold according to user manual for the port that does not support DCB. This patch also change hclge_get_tc_num's return type to u32 to avoid signed-unsigned mix with divide. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 27 ++++++++++++++++------ 1 file changed, 20 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 7e7cdad..d2fb548 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -1422,7 +1422,7 @@ static int hclge_tx_buffer_alloc(struct hclge_dev *hdev, return ret; } -static int hclge_get_tc_num(struct hclge_dev *hdev) +static u32 hclge_get_tc_num(struct hclge_dev *hdev) { int i, cnt = 0; @@ -1495,8 +1495,8 @@ static bool hclge_is_rx_buf_ok(struct hclge_dev *hdev, struct hclge_pkt_buf_alloc *buf_alloc, u32 rx_all) { - u32 shared_buf_min, shared_buf_tc, shared_std; - int tc_num = hclge_get_tc_num(hdev); + u32 shared_buf_min, shared_buf_tc, shared_std, hi_thrd, lo_thrd; + u32 tc_num = hclge_get_tc_num(hdev); u32 shared_buf, aligned_mps; u32 rx_priv; int i; @@ -1526,13 +1526,26 @@ static bool hclge_is_rx_buf_ok(struct hclge_dev *hdev, } else { buf_alloc->s_buf.self.high = aligned_mps + HCLGE_NON_DCB_ADDITIONAL_BUF; - buf_alloc->s_buf.self.low = - roundup(aligned_mps / 2, HCLGE_BUF_SIZE_UNIT); + buf_alloc->s_buf.self.low = aligned_mps; + } + + if (hnae3_dev_dcb_supported(hdev)) { + if (tc_num) + hi_thrd = (shared_buf - hdev->dv_buf_size) / tc_num; + else + hi_thrd = shared_buf - hdev->dv_buf_size; + + hi_thrd = max_t(u32, hi_thrd, 2 * aligned_mps); + hi_thrd = rounddown(hi_thrd, HCLGE_BUF_SIZE_UNIT); + lo_thrd = hi_thrd - aligned_mps / 2; + } else { + hi_thrd = aligned_mps + HCLGE_NON_DCB_ADDITIONAL_BUF; + lo_thrd = aligned_mps; } for (i = 0; i < HCLGE_MAX_TC_NUM; i++) { - buf_alloc->s_buf.tc_thrd[i].low = aligned_mps; - buf_alloc->s_buf.tc_thrd[i].high = 2 * aligned_mps; + buf_alloc->s_buf.tc_thrd[i].low = lo_thrd; + buf_alloc->s_buf.tc_thrd[i].high = hi_thrd; } return true; -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 09/12] net: hns3: set dividual reset level for all RAS and MSI-X errors 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (7 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 08/12] net: hns3: divide shared buffer between TC Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 10/12] net: hns3: do not initialize MDIO bus when PHY is inexistent Huazhong Tan ` (3 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Weihang Li, Peng Li, Huazhong Tan From: Weihang Li <liweihang@hisilicon.com> According to hardware description, reset level that should be triggered are not consistent in a module. For example, in SSU common errors, the first two bits has no need to do reset, but the other bits need global reset. This patch sets separate reset level for all RAS and MSI-X interrupts by adding a reset_lvel field in struct hclge_hw_error, and fixes some incorrect reset level. Signed-off-by: Weihang Li <liweihang@hisilicon.com> Signed-off-by: Peng Li <lipeng321@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- .../net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c | 956 +++++++++++++-------- .../net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h | 1 + 2 files changed, 618 insertions(+), 339 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c index 1f52d11..62ef161 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c @@ -4,287 +4,468 @@ #include "hclge_err.h" static const struct hclge_hw_error hclge_imp_tcm_ecc_int[] = { - { .int_msk = BIT(1), .msg = "imp_itcm0_ecc_mbit_err" }, - { .int_msk = BIT(3), .msg = "imp_itcm1_ecc_mbit_err" }, - { .int_msk = BIT(5), .msg = "imp_itcm2_ecc_mbit_err" }, - { .int_msk = BIT(7), .msg = "imp_itcm3_ecc_mbit_err" }, - { .int_msk = BIT(9), .msg = "imp_dtcm0_mem0_ecc_mbit_err" }, - { .int_msk = BIT(11), .msg = "imp_dtcm0_mem1_ecc_mbit_err" }, - { .int_msk = BIT(13), .msg = "imp_dtcm1_mem0_ecc_mbit_err" }, - { .int_msk = BIT(15), .msg = "imp_dtcm1_mem1_ecc_mbit_err" }, - { .int_msk = BIT(17), .msg = "imp_itcm4_ecc_mbit_err" }, + { .int_msk = BIT(1), .msg = "imp_itcm0_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(3), .msg = "imp_itcm1_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(5), .msg = "imp_itcm2_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(7), .msg = "imp_itcm3_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(9), .msg = "imp_dtcm0_mem0_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(11), .msg = "imp_dtcm0_mem1_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(13), .msg = "imp_dtcm1_mem0_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(15), .msg = "imp_dtcm1_mem1_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(17), .msg = "imp_itcm4_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_cmdq_nic_mem_ecc_int[] = { - { .int_msk = BIT(1), .msg = "cmdq_nic_rx_depth_ecc_mbit_err" }, - { .int_msk = BIT(3), .msg = "cmdq_nic_tx_depth_ecc_mbit_err" }, - { .int_msk = BIT(5), .msg = "cmdq_nic_rx_tail_ecc_mbit_err" }, - { .int_msk = BIT(7), .msg = "cmdq_nic_tx_tail_ecc_mbit_err" }, - { .int_msk = BIT(9), .msg = "cmdq_nic_rx_head_ecc_mbit_err" }, - { .int_msk = BIT(11), .msg = "cmdq_nic_tx_head_ecc_mbit_err" }, - { .int_msk = BIT(13), .msg = "cmdq_nic_rx_addr_ecc_mbit_err" }, - { .int_msk = BIT(15), .msg = "cmdq_nic_tx_addr_ecc_mbit_err" }, - { .int_msk = BIT(17), .msg = "cmdq_rocee_rx_depth_ecc_mbit_err" }, - { .int_msk = BIT(19), .msg = "cmdq_rocee_tx_depth_ecc_mbit_err" }, - { .int_msk = BIT(21), .msg = "cmdq_rocee_rx_tail_ecc_mbit_err" }, - { .int_msk = BIT(23), .msg = "cmdq_rocee_tx_tail_ecc_mbit_err" }, - { .int_msk = BIT(25), .msg = "cmdq_rocee_rx_head_ecc_mbit_err" }, - { .int_msk = BIT(27), .msg = "cmdq_rocee_tx_head_ecc_mbit_err" }, - { .int_msk = BIT(29), .msg = "cmdq_rocee_rx_addr_ecc_mbit_err" }, - { .int_msk = BIT(31), .msg = "cmdq_rocee_tx_addr_ecc_mbit_err" }, + { .int_msk = BIT(1), .msg = "cmdq_nic_rx_depth_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(3), .msg = "cmdq_nic_tx_depth_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(5), .msg = "cmdq_nic_rx_tail_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(7), .msg = "cmdq_nic_tx_tail_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(9), .msg = "cmdq_nic_rx_head_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(11), .msg = "cmdq_nic_tx_head_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(13), .msg = "cmdq_nic_rx_addr_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(15), .msg = "cmdq_nic_tx_addr_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(17), .msg = "cmdq_rocee_rx_depth_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(19), .msg = "cmdq_rocee_tx_depth_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(21), .msg = "cmdq_rocee_rx_tail_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(23), .msg = "cmdq_rocee_tx_tail_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(25), .msg = "cmdq_rocee_rx_head_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(27), .msg = "cmdq_rocee_tx_head_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(29), .msg = "cmdq_rocee_rx_addr_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(31), .msg = "cmdq_rocee_tx_addr_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_tqp_int_ecc_int[] = { - { .int_msk = BIT(6), .msg = "tqp_int_cfg_even_ecc_mbit_err" }, - { .int_msk = BIT(7), .msg = "tqp_int_cfg_odd_ecc_mbit_err" }, - { .int_msk = BIT(8), .msg = "tqp_int_ctrl_even_ecc_mbit_err" }, - { .int_msk = BIT(9), .msg = "tqp_int_ctrl_odd_ecc_mbit_err" }, - { .int_msk = BIT(10), .msg = "tx_que_scan_int_ecc_mbit_err" }, - { .int_msk = BIT(11), .msg = "rx_que_scan_int_ecc_mbit_err" }, + { .int_msk = BIT(6), .msg = "tqp_int_cfg_even_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(7), .msg = "tqp_int_cfg_odd_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(8), .msg = "tqp_int_ctrl_even_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(9), .msg = "tqp_int_ctrl_odd_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(10), .msg = "tx_que_scan_int_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(11), .msg = "rx_que_scan_int_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_msix_sram_ecc_int[] = { - { .int_msk = BIT(1), .msg = "msix_nic_ecc_mbit_err" }, - { .int_msk = BIT(3), .msg = "msix_rocee_ecc_mbit_err" }, + { .int_msk = BIT(1), .msg = "msix_nic_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(3), .msg = "msix_rocee_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_igu_int[] = { - { .int_msk = BIT(0), .msg = "igu_rx_buf0_ecc_mbit_err" }, - { .int_msk = BIT(2), .msg = "igu_rx_buf1_ecc_mbit_err" }, + { .int_msk = BIT(0), .msg = "igu_rx_buf0_ecc_mbit_err", + .reset_level = HNAE3_CORE_RESET }, + { .int_msk = BIT(2), .msg = "igu_rx_buf1_ecc_mbit_err", + .reset_level = HNAE3_CORE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_igu_egu_tnl_int[] = { - { .int_msk = BIT(0), .msg = "rx_buf_overflow" }, - { .int_msk = BIT(1), .msg = "rx_stp_fifo_overflow" }, - { .int_msk = BIT(2), .msg = "rx_stp_fifo_undeflow" }, - { .int_msk = BIT(3), .msg = "tx_buf_overflow" }, - { .int_msk = BIT(4), .msg = "tx_buf_underrun" }, - { .int_msk = BIT(5), .msg = "rx_stp_buf_overflow" }, + { .int_msk = BIT(0), .msg = "rx_buf_overflow", + .reset_level = HNAE3_CORE_RESET }, + { .int_msk = BIT(1), .msg = "rx_stp_fifo_overflow", + .reset_level = HNAE3_CORE_RESET }, + { .int_msk = BIT(2), .msg = "rx_stp_fifo_undeflow", + .reset_level = HNAE3_CORE_RESET }, + { .int_msk = BIT(3), .msg = "tx_buf_overflow", + .reset_level = HNAE3_CORE_RESET }, + { .int_msk = BIT(4), .msg = "tx_buf_underrun", + .reset_level = HNAE3_CORE_RESET }, + { .int_msk = BIT(5), .msg = "rx_stp_buf_overflow", + .reset_level = HNAE3_CORE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ncsi_err_int[] = { - { .int_msk = BIT(1), .msg = "ncsi_tx_ecc_mbit_err" }, + { .int_msk = BIT(1), .msg = "ncsi_tx_ecc_mbit_err", + .reset_level = HNAE3_NONE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ppp_mpf_abnormal_int_st1[] = { - { .int_msk = BIT(0), .msg = "vf_vlan_ad_mem_ecc_mbit_err" }, - { .int_msk = BIT(1), .msg = "umv_mcast_group_mem_ecc_mbit_err" }, - { .int_msk = BIT(2), .msg = "umv_key_mem0_ecc_mbit_err" }, - { .int_msk = BIT(3), .msg = "umv_key_mem1_ecc_mbit_err" }, - { .int_msk = BIT(4), .msg = "umv_key_mem2_ecc_mbit_err" }, - { .int_msk = BIT(5), .msg = "umv_key_mem3_ecc_mbit_err" }, - { .int_msk = BIT(6), .msg = "umv_ad_mem_ecc_mbit_err" }, - { .int_msk = BIT(7), .msg = "rss_tc_mode_mem_ecc_mbit_err" }, - { .int_msk = BIT(8), .msg = "rss_idt_mem0_ecc_mbit_err" }, - { .int_msk = BIT(9), .msg = "rss_idt_mem1_ecc_mbit_err" }, - { .int_msk = BIT(10), .msg = "rss_idt_mem2_ecc_mbit_err" }, - { .int_msk = BIT(11), .msg = "rss_idt_mem3_ecc_mbit_err" }, - { .int_msk = BIT(12), .msg = "rss_idt_mem4_ecc_mbit_err" }, - { .int_msk = BIT(13), .msg = "rss_idt_mem5_ecc_mbit_err" }, - { .int_msk = BIT(14), .msg = "rss_idt_mem6_ecc_mbit_err" }, - { .int_msk = BIT(15), .msg = "rss_idt_mem7_ecc_mbit_err" }, - { .int_msk = BIT(16), .msg = "rss_idt_mem8_ecc_mbit_err" }, - { .int_msk = BIT(17), .msg = "rss_idt_mem9_ecc_mbit_err" }, - { .int_msk = BIT(18), .msg = "rss_idt_mem10_ecc_m1bit_err" }, - { .int_msk = BIT(19), .msg = "rss_idt_mem11_ecc_mbit_err" }, - { .int_msk = BIT(20), .msg = "rss_idt_mem12_ecc_mbit_err" }, - { .int_msk = BIT(21), .msg = "rss_idt_mem13_ecc_mbit_err" }, - { .int_msk = BIT(22), .msg = "rss_idt_mem14_ecc_mbit_err" }, - { .int_msk = BIT(23), .msg = "rss_idt_mem15_ecc_mbit_err" }, - { .int_msk = BIT(24), .msg = "port_vlan_mem_ecc_mbit_err" }, - { .int_msk = BIT(25), .msg = "mcast_linear_table_mem_ecc_mbit_err" }, - { .int_msk = BIT(26), .msg = "mcast_result_mem_ecc_mbit_err" }, - { .int_msk = BIT(27), - .msg = "flow_director_ad_mem0_ecc_mbit_err" }, - { .int_msk = BIT(28), - .msg = "flow_director_ad_mem1_ecc_mbit_err" }, - { .int_msk = BIT(29), - .msg = "rx_vlan_tag_memory_ecc_mbit_err" }, - { .int_msk = BIT(30), - .msg = "Tx_UP_mapping_config_mem_ecc_mbit_err" }, + { .int_msk = BIT(0), .msg = "vf_vlan_ad_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(1), .msg = "umv_mcast_group_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(2), .msg = "umv_key_mem0_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(3), .msg = "umv_key_mem1_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(4), .msg = "umv_key_mem2_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(5), .msg = "umv_key_mem3_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(6), .msg = "umv_ad_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(7), .msg = "rss_tc_mode_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(8), .msg = "rss_idt_mem0_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(9), .msg = "rss_idt_mem1_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(10), .msg = "rss_idt_mem2_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(11), .msg = "rss_idt_mem3_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(12), .msg = "rss_idt_mem4_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(13), .msg = "rss_idt_mem5_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(14), .msg = "rss_idt_mem6_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(15), .msg = "rss_idt_mem7_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(16), .msg = "rss_idt_mem8_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(17), .msg = "rss_idt_mem9_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(18), .msg = "rss_idt_mem10_ecc_m1bit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(19), .msg = "rss_idt_mem11_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(20), .msg = "rss_idt_mem12_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(21), .msg = "rss_idt_mem13_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(22), .msg = "rss_idt_mem14_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(23), .msg = "rss_idt_mem15_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(24), .msg = "port_vlan_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(25), .msg = "mcast_linear_table_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(26), .msg = "mcast_result_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(27), .msg = "flow_director_ad_mem0_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(28), .msg = "flow_director_ad_mem1_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(29), .msg = "rx_vlan_tag_memory_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(30), .msg = "Tx_UP_mapping_config_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ppp_pf_abnormal_int[] = { - { .int_msk = BIT(0), .msg = "tx_vlan_tag_err" }, - { .int_msk = BIT(1), .msg = "rss_list_tc_unassigned_queue_err" }, + { .int_msk = BIT(0), .msg = "tx_vlan_tag_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(1), .msg = "rss_list_tc_unassigned_queue_err", + .reset_level = HNAE3_NONE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ppp_mpf_abnormal_int_st3[] = { - { .int_msk = BIT(0), .msg = "hfs_fifo_mem_ecc_mbit_err" }, - { .int_msk = BIT(1), .msg = "rslt_descr_fifo_mem_ecc_mbit_err" }, - { .int_msk = BIT(2), .msg = "tx_vlan_tag_mem_ecc_mbit_err" }, - { .int_msk = BIT(3), .msg = "FD_CN0_memory_ecc_mbit_err" }, - { .int_msk = BIT(4), .msg = "FD_CN1_memory_ecc_mbit_err" }, - { .int_msk = BIT(5), .msg = "GRO_AD_memory_ecc_mbit_err" }, + { .int_msk = BIT(0), .msg = "hfs_fifo_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(1), .msg = "rslt_descr_fifo_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(2), .msg = "tx_vlan_tag_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(3), .msg = "FD_CN0_memory_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(4), .msg = "FD_CN1_memory_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(5), .msg = "GRO_AD_memory_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_tm_sch_rint[] = { - { .int_msk = BIT(1), .msg = "tm_sch_ecc_mbit_err" }, - { .int_msk = BIT(2), .msg = "tm_sch_port_shap_sub_fifo_wr_err" }, - { .int_msk = BIT(3), .msg = "tm_sch_port_shap_sub_fifo_rd_err" }, - { .int_msk = BIT(4), .msg = "tm_sch_pg_pshap_sub_fifo_wr_err" }, - { .int_msk = BIT(5), .msg = "tm_sch_pg_pshap_sub_fifo_rd_err" }, - { .int_msk = BIT(6), .msg = "tm_sch_pg_cshap_sub_fifo_wr_err" }, - { .int_msk = BIT(7), .msg = "tm_sch_pg_cshap_sub_fifo_rd_err" }, - { .int_msk = BIT(8), .msg = "tm_sch_pri_pshap_sub_fifo_wr_err" }, - { .int_msk = BIT(9), .msg = "tm_sch_pri_pshap_sub_fifo_rd_err" }, - { .int_msk = BIT(10), .msg = "tm_sch_pri_cshap_sub_fifo_wr_err" }, - { .int_msk = BIT(11), .msg = "tm_sch_pri_cshap_sub_fifo_rd_err" }, - { .int_msk = BIT(12), - .msg = "tm_sch_port_shap_offset_fifo_wr_err" }, - { .int_msk = BIT(13), - .msg = "tm_sch_port_shap_offset_fifo_rd_err" }, - { .int_msk = BIT(14), - .msg = "tm_sch_pg_pshap_offset_fifo_wr_err" }, - { .int_msk = BIT(15), - .msg = "tm_sch_pg_pshap_offset_fifo_rd_err" }, - { .int_msk = BIT(16), - .msg = "tm_sch_pg_cshap_offset_fifo_wr_err" }, - { .int_msk = BIT(17), - .msg = "tm_sch_pg_cshap_offset_fifo_rd_err" }, - { .int_msk = BIT(18), - .msg = "tm_sch_pri_pshap_offset_fifo_wr_err" }, - { .int_msk = BIT(19), - .msg = "tm_sch_pri_pshap_offset_fifo_rd_err" }, - { .int_msk = BIT(20), - .msg = "tm_sch_pri_cshap_offset_fifo_wr_err" }, - { .int_msk = BIT(21), - .msg = "tm_sch_pri_cshap_offset_fifo_rd_err" }, - { .int_msk = BIT(22), .msg = "tm_sch_rq_fifo_wr_err" }, - { .int_msk = BIT(23), .msg = "tm_sch_rq_fifo_rd_err" }, - { .int_msk = BIT(24), .msg = "tm_sch_nq_fifo_wr_err" }, - { .int_msk = BIT(25), .msg = "tm_sch_nq_fifo_rd_err" }, - { .int_msk = BIT(26), .msg = "tm_sch_roce_up_fifo_wr_err" }, - { .int_msk = BIT(27), .msg = "tm_sch_roce_up_fifo_rd_err" }, - { .int_msk = BIT(28), .msg = "tm_sch_rcb_byte_fifo_wr_err" }, - { .int_msk = BIT(29), .msg = "tm_sch_rcb_byte_fifo_rd_err" }, - { .int_msk = BIT(30), .msg = "tm_sch_ssu_byte_fifo_wr_err" }, - { .int_msk = BIT(31), .msg = "tm_sch_ssu_byte_fifo_rd_err" }, + { .int_msk = BIT(1), .msg = "tm_sch_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(2), .msg = "tm_sch_port_shap_sub_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(3), .msg = "tm_sch_port_shap_sub_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(4), .msg = "tm_sch_pg_pshap_sub_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(5), .msg = "tm_sch_pg_pshap_sub_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(6), .msg = "tm_sch_pg_cshap_sub_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(7), .msg = "tm_sch_pg_cshap_sub_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(8), .msg = "tm_sch_pri_pshap_sub_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(9), .msg = "tm_sch_pri_pshap_sub_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(10), .msg = "tm_sch_pri_cshap_sub_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(11), .msg = "tm_sch_pri_cshap_sub_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(12), .msg = "tm_sch_port_shap_offset_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(13), .msg = "tm_sch_port_shap_offset_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(14), .msg = "tm_sch_pg_pshap_offset_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(15), .msg = "tm_sch_pg_pshap_offset_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(16), .msg = "tm_sch_pg_cshap_offset_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(17), .msg = "tm_sch_pg_cshap_offset_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(18), .msg = "tm_sch_pri_pshap_offset_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(19), .msg = "tm_sch_pri_pshap_offset_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(20), .msg = "tm_sch_pri_cshap_offset_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(21), .msg = "tm_sch_pri_cshap_offset_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(22), .msg = "tm_sch_rq_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(23), .msg = "tm_sch_rq_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(24), .msg = "tm_sch_nq_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(25), .msg = "tm_sch_nq_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(26), .msg = "tm_sch_roce_up_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(27), .msg = "tm_sch_roce_up_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(28), .msg = "tm_sch_rcb_byte_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(29), .msg = "tm_sch_rcb_byte_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(30), .msg = "tm_sch_ssu_byte_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(31), .msg = "tm_sch_ssu_byte_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_qcn_fifo_rint[] = { - { .int_msk = BIT(0), .msg = "qcn_shap_gp0_sch_fifo_rd_err" }, - { .int_msk = BIT(1), .msg = "qcn_shap_gp0_sch_fifo_wr_err" }, - { .int_msk = BIT(2), .msg = "qcn_shap_gp1_sch_fifo_rd_err" }, - { .int_msk = BIT(3), .msg = "qcn_shap_gp1_sch_fifo_wr_err" }, - { .int_msk = BIT(4), .msg = "qcn_shap_gp2_sch_fifo_rd_err" }, - { .int_msk = BIT(5), .msg = "qcn_shap_gp2_sch_fifo_wr_err" }, - { .int_msk = BIT(6), .msg = "qcn_shap_gp3_sch_fifo_rd_err" }, - { .int_msk = BIT(7), .msg = "qcn_shap_gp3_sch_fifo_wr_err" }, - { .int_msk = BIT(8), .msg = "qcn_shap_gp0_offset_fifo_rd_err" }, - { .int_msk = BIT(9), .msg = "qcn_shap_gp0_offset_fifo_wr_err" }, - { .int_msk = BIT(10), .msg = "qcn_shap_gp1_offset_fifo_rd_err" }, - { .int_msk = BIT(11), .msg = "qcn_shap_gp1_offset_fifo_wr_err" }, - { .int_msk = BIT(12), .msg = "qcn_shap_gp2_offset_fifo_rd_err" }, - { .int_msk = BIT(13), .msg = "qcn_shap_gp2_offset_fifo_wr_err" }, - { .int_msk = BIT(14), .msg = "qcn_shap_gp3_offset_fifo_rd_err" }, - { .int_msk = BIT(15), .msg = "qcn_shap_gp3_offset_fifo_wr_err" }, - { .int_msk = BIT(16), .msg = "qcn_byte_info_fifo_rd_err" }, - { .int_msk = BIT(17), .msg = "qcn_byte_info_fifo_wr_err" }, + { .int_msk = BIT(0), .msg = "qcn_shap_gp0_sch_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(1), .msg = "qcn_shap_gp0_sch_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(2), .msg = "qcn_shap_gp1_sch_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(3), .msg = "qcn_shap_gp1_sch_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(4), .msg = "qcn_shap_gp2_sch_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(5), .msg = "qcn_shap_gp2_sch_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(6), .msg = "qcn_shap_gp3_sch_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(7), .msg = "qcn_shap_gp3_sch_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(8), .msg = "qcn_shap_gp0_offset_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(9), .msg = "qcn_shap_gp0_offset_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(10), .msg = "qcn_shap_gp1_offset_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(11), .msg = "qcn_shap_gp1_offset_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(12), .msg = "qcn_shap_gp2_offset_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(13), .msg = "qcn_shap_gp2_offset_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(14), .msg = "qcn_shap_gp3_offset_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(15), .msg = "qcn_shap_gp3_offset_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(16), .msg = "qcn_byte_info_fifo_rd_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(17), .msg = "qcn_byte_info_fifo_wr_err", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_qcn_ecc_rint[] = { - { .int_msk = BIT(1), .msg = "qcn_byte_mem_ecc_mbit_err" }, - { .int_msk = BIT(3), .msg = "qcn_time_mem_ecc_mbit_err" }, - { .int_msk = BIT(5), .msg = "qcn_fb_mem_ecc_mbit_err" }, - { .int_msk = BIT(7), .msg = "qcn_link_mem_ecc_mbit_err" }, - { .int_msk = BIT(9), .msg = "qcn_rate_mem_ecc_mbit_err" }, - { .int_msk = BIT(11), .msg = "qcn_tmplt_mem_ecc_mbit_err" }, - { .int_msk = BIT(13), .msg = "qcn_shap_cfg_mem_ecc_mbit_err" }, - { .int_msk = BIT(15), .msg = "qcn_gp0_barrel_mem_ecc_mbit_err" }, - { .int_msk = BIT(17), .msg = "qcn_gp1_barrel_mem_ecc_mbit_err" }, - { .int_msk = BIT(19), .msg = "qcn_gp2_barrel_mem_ecc_mbit_err" }, - { .int_msk = BIT(21), .msg = "qcn_gp3_barral_mem_ecc_mbit_err" }, + { .int_msk = BIT(1), .msg = "qcn_byte_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(3), .msg = "qcn_time_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(5), .msg = "qcn_fb_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(7), .msg = "qcn_link_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(9), .msg = "qcn_rate_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(11), .msg = "qcn_tmplt_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(13), .msg = "qcn_shap_cfg_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(15), .msg = "qcn_gp0_barrel_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(17), .msg = "qcn_gp1_barrel_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(19), .msg = "qcn_gp2_barrel_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(21), .msg = "qcn_gp3_barral_mem_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_mac_afifo_tnl_int[] = { - { .int_msk = BIT(0), .msg = "egu_cge_afifo_ecc_1bit_err" }, - { .int_msk = BIT(1), .msg = "egu_cge_afifo_ecc_mbit_err" }, - { .int_msk = BIT(2), .msg = "egu_lge_afifo_ecc_1bit_err" }, - { .int_msk = BIT(3), .msg = "egu_lge_afifo_ecc_mbit_err" }, - { .int_msk = BIT(4), .msg = "cge_igu_afifo_ecc_1bit_err" }, - { .int_msk = BIT(5), .msg = "cge_igu_afifo_ecc_mbit_err" }, - { .int_msk = BIT(6), .msg = "lge_igu_afifo_ecc_1bit_err" }, - { .int_msk = BIT(7), .msg = "lge_igu_afifo_ecc_mbit_err" }, - { .int_msk = BIT(8), .msg = "cge_igu_afifo_overflow_err" }, - { .int_msk = BIT(9), .msg = "lge_igu_afifo_overflow_err" }, - { .int_msk = BIT(10), .msg = "egu_cge_afifo_underrun_err" }, - { .int_msk = BIT(11), .msg = "egu_lge_afifo_underrun_err" }, - { .int_msk = BIT(12), .msg = "egu_ge_afifo_underrun_err" }, - { .int_msk = BIT(13), .msg = "ge_igu_afifo_overflow_err" }, + { .int_msk = BIT(0), .msg = "egu_cge_afifo_ecc_1bit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(1), .msg = "egu_cge_afifo_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(2), .msg = "egu_lge_afifo_ecc_1bit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(3), .msg = "egu_lge_afifo_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(4), .msg = "cge_igu_afifo_ecc_1bit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(5), .msg = "cge_igu_afifo_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(6), .msg = "lge_igu_afifo_ecc_1bit_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(7), .msg = "lge_igu_afifo_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(8), .msg = "cge_igu_afifo_overflow_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(9), .msg = "lge_igu_afifo_overflow_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(10), .msg = "egu_cge_afifo_underrun_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(11), .msg = "egu_lge_afifo_underrun_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(12), .msg = "egu_ge_afifo_underrun_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(13), .msg = "ge_igu_afifo_overflow_err", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ppu_mpf_abnormal_int_st2[] = { - { .int_msk = BIT(13), .msg = "rpu_rx_pkt_bit32_ecc_mbit_err" }, - { .int_msk = BIT(14), .msg = "rpu_rx_pkt_bit33_ecc_mbit_err" }, - { .int_msk = BIT(15), .msg = "rpu_rx_pkt_bit34_ecc_mbit_err" }, - { .int_msk = BIT(16), .msg = "rpu_rx_pkt_bit35_ecc_mbit_err" }, - { .int_msk = BIT(17), .msg = "rcb_tx_ring_ecc_mbit_err" }, - { .int_msk = BIT(18), .msg = "rcb_rx_ring_ecc_mbit_err" }, - { .int_msk = BIT(19), .msg = "rcb_tx_fbd_ecc_mbit_err" }, - { .int_msk = BIT(20), .msg = "rcb_rx_ebd_ecc_mbit_err" }, - { .int_msk = BIT(21), .msg = "rcb_tso_info_ecc_mbit_err" }, - { .int_msk = BIT(22), .msg = "rcb_tx_int_info_ecc_mbit_err" }, - { .int_msk = BIT(23), .msg = "rcb_rx_int_info_ecc_mbit_err" }, - { .int_msk = BIT(24), .msg = "tpu_tx_pkt_0_ecc_mbit_err" }, - { .int_msk = BIT(25), .msg = "tpu_tx_pkt_1_ecc_mbit_err" }, - { .int_msk = BIT(26), .msg = "rd_bus_err" }, - { .int_msk = BIT(27), .msg = "wr_bus_err" }, - { .int_msk = BIT(28), .msg = "reg_search_miss" }, - { .int_msk = BIT(29), .msg = "rx_q_search_miss" }, - { .int_msk = BIT(30), .msg = "ooo_ecc_err_detect" }, - { .int_msk = BIT(31), .msg = "ooo_ecc_err_multpl" }, + { .int_msk = BIT(13), .msg = "rpu_rx_pkt_bit32_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(14), .msg = "rpu_rx_pkt_bit33_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(15), .msg = "rpu_rx_pkt_bit34_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(16), .msg = "rpu_rx_pkt_bit35_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(17), .msg = "rcb_tx_ring_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(18), .msg = "rcb_rx_ring_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(19), .msg = "rcb_tx_fbd_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(20), .msg = "rcb_rx_ebd_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(21), .msg = "rcb_tso_info_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(22), .msg = "rcb_tx_int_info_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(23), .msg = "rcb_rx_int_info_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(24), .msg = "tpu_tx_pkt_0_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(25), .msg = "tpu_tx_pkt_1_ecc_mbit_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(26), .msg = "rd_bus_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(27), .msg = "wr_bus_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(28), .msg = "reg_search_miss", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(29), .msg = "rx_q_search_miss", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(30), .msg = "ooo_ecc_err_detect", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(31), .msg = "ooo_ecc_err_multpl", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ppu_mpf_abnormal_int_st3[] = { - { .int_msk = BIT(4), .msg = "gro_bd_ecc_mbit_err" }, - { .int_msk = BIT(5), .msg = "gro_context_ecc_mbit_err" }, - { .int_msk = BIT(6), .msg = "rx_stash_cfg_ecc_mbit_err" }, - { .int_msk = BIT(7), .msg = "axi_rd_fbd_ecc_mbit_err" }, + { .int_msk = BIT(4), .msg = "gro_bd_ecc_mbit_err", + .reset_level = HNAE3_CORE_RESET }, + { .int_msk = BIT(5), .msg = "gro_context_ecc_mbit_err", + .reset_level = HNAE3_CORE_RESET }, + { .int_msk = BIT(6), .msg = "rx_stash_cfg_ecc_mbit_err", + .reset_level = HNAE3_CORE_RESET }, + { .int_msk = BIT(7), .msg = "axi_rd_fbd_ecc_mbit_err", + .reset_level = HNAE3_CORE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ppu_pf_abnormal_int[] = { - { .int_msk = BIT(0), .msg = "over_8bd_no_fe" }, - { .int_msk = BIT(1), .msg = "tso_mss_cmp_min_err" }, - { .int_msk = BIT(2), .msg = "tso_mss_cmp_max_err" }, - { .int_msk = BIT(3), .msg = "tx_rd_fbd_poison" }, - { .int_msk = BIT(4), .msg = "rx_rd_ebd_poison" }, - { .int_msk = BIT(5), .msg = "buf_wait_timeout" }, + { .int_msk = BIT(0), .msg = "over_8bd_no_fe", + .reset_level = HNAE3_FUNC_RESET }, + { .int_msk = BIT(1), .msg = "tso_mss_cmp_min_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(2), .msg = "tso_mss_cmp_max_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(3), .msg = "tx_rd_fbd_poison", + .reset_level = HNAE3_FUNC_RESET }, + { .int_msk = BIT(4), .msg = "rx_rd_ebd_poison", + .reset_level = HNAE3_FUNC_RESET }, + { .int_msk = BIT(5), .msg = "buf_wait_timeout", + .reset_level = HNAE3_NONE_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ssu_com_err_int[] = { - { .int_msk = BIT(0), .msg = "buf_sum_err" }, - { .int_msk = BIT(1), .msg = "ppp_mb_num_err" }, - { .int_msk = BIT(2), .msg = "ppp_mbid_err" }, - { .int_msk = BIT(3), .msg = "ppp_rlt_mac_err" }, - { .int_msk = BIT(4), .msg = "ppp_rlt_host_err" }, - { .int_msk = BIT(5), .msg = "cks_edit_position_err" }, - { .int_msk = BIT(6), .msg = "cks_edit_condition_err" }, - { .int_msk = BIT(7), .msg = "vlan_edit_condition_err" }, - { .int_msk = BIT(8), .msg = "vlan_num_ot_err" }, - { .int_msk = BIT(9), .msg = "vlan_num_in_err" }, + { .int_msk = BIT(0), .msg = "buf_sum_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(1), .msg = "ppp_mb_num_err", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(2), .msg = "ppp_mbid_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(3), .msg = "ppp_rlt_mac_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(4), .msg = "ppp_rlt_host_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(5), .msg = "cks_edit_position_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(6), .msg = "cks_edit_condition_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(7), .msg = "vlan_edit_condition_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(8), .msg = "vlan_num_ot_err", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(9), .msg = "vlan_num_in_err", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; #define HCLGE_SSU_MEM_ECC_ERR(x) \ - { .int_msk = BIT(x), .msg = "ssu_mem" #x "_ecc_mbit_err" } + { .int_msk = BIT(x), .msg = "ssu_mem" #x "_ecc_mbit_err", \ + .reset_level = HNAE3_GLOBAL_RESET } static const struct hclge_hw_error hclge_ssu_mem_ecc_err_int[] = { HCLGE_SSU_MEM_ECC_ERR(0), @@ -323,62 +504,106 @@ static const struct hclge_hw_error hclge_ssu_mem_ecc_err_int[] = { }; static const struct hclge_hw_error hclge_ssu_port_based_err_int[] = { - { .int_msk = BIT(0), .msg = "roc_pkt_without_key_port" }, - { .int_msk = BIT(1), .msg = "tpu_pkt_without_key_port" }, - { .int_msk = BIT(2), .msg = "igu_pkt_without_key_port" }, - { .int_msk = BIT(3), .msg = "roc_eof_mis_match_port" }, - { .int_msk = BIT(4), .msg = "tpu_eof_mis_match_port" }, - { .int_msk = BIT(5), .msg = "igu_eof_mis_match_port" }, - { .int_msk = BIT(6), .msg = "roc_sof_mis_match_port" }, - { .int_msk = BIT(7), .msg = "tpu_sof_mis_match_port" }, - { .int_msk = BIT(8), .msg = "igu_sof_mis_match_port" }, - { .int_msk = BIT(11), .msg = "ets_rd_int_rx_port" }, - { .int_msk = BIT(12), .msg = "ets_wr_int_rx_port" }, - { .int_msk = BIT(13), .msg = "ets_rd_int_tx_port" }, - { .int_msk = BIT(14), .msg = "ets_wr_int_tx_port" }, + { .int_msk = BIT(0), .msg = "roc_pkt_without_key_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(1), .msg = "tpu_pkt_without_key_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(2), .msg = "igu_pkt_without_key_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(3), .msg = "roc_eof_mis_match_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(4), .msg = "tpu_eof_mis_match_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(5), .msg = "igu_eof_mis_match_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(6), .msg = "roc_sof_mis_match_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(7), .msg = "tpu_sof_mis_match_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(8), .msg = "igu_sof_mis_match_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(11), .msg = "ets_rd_int_rx_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(12), .msg = "ets_wr_int_rx_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(13), .msg = "ets_rd_int_tx_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(14), .msg = "ets_wr_int_tx_port", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ssu_fifo_overflow_int[] = { - { .int_msk = BIT(0), .msg = "ig_mac_inf_int" }, - { .int_msk = BIT(1), .msg = "ig_host_inf_int" }, - { .int_msk = BIT(2), .msg = "ig_roc_buf_int" }, - { .int_msk = BIT(3), .msg = "ig_host_data_fifo_int" }, - { .int_msk = BIT(4), .msg = "ig_host_key_fifo_int" }, - { .int_msk = BIT(5), .msg = "tx_qcn_fifo_int" }, - { .int_msk = BIT(6), .msg = "rx_qcn_fifo_int" }, - { .int_msk = BIT(7), .msg = "tx_pf_rd_fifo_int" }, - { .int_msk = BIT(8), .msg = "rx_pf_rd_fifo_int" }, - { .int_msk = BIT(9), .msg = "qm_eof_fifo_int" }, - { .int_msk = BIT(10), .msg = "mb_rlt_fifo_int" }, - { .int_msk = BIT(11), .msg = "dup_uncopy_fifo_int" }, - { .int_msk = BIT(12), .msg = "dup_cnt_rd_fifo_int" }, - { .int_msk = BIT(13), .msg = "dup_cnt_drop_fifo_int" }, - { .int_msk = BIT(14), .msg = "dup_cnt_wrb_fifo_int" }, - { .int_msk = BIT(15), .msg = "host_cmd_fifo_int" }, - { .int_msk = BIT(16), .msg = "mac_cmd_fifo_int" }, - { .int_msk = BIT(17), .msg = "host_cmd_bitmap_empty_int" }, - { .int_msk = BIT(18), .msg = "mac_cmd_bitmap_empty_int" }, - { .int_msk = BIT(19), .msg = "dup_bitmap_empty_int" }, - { .int_msk = BIT(20), .msg = "out_queue_bitmap_empty_int" }, - { .int_msk = BIT(21), .msg = "bank2_bitmap_empty_int" }, - { .int_msk = BIT(22), .msg = "bank1_bitmap_empty_int" }, - { .int_msk = BIT(23), .msg = "bank0_bitmap_empty_int" }, + { .int_msk = BIT(0), .msg = "ig_mac_inf_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(1), .msg = "ig_host_inf_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(2), .msg = "ig_roc_buf_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(3), .msg = "ig_host_data_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(4), .msg = "ig_host_key_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(5), .msg = "tx_qcn_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(6), .msg = "rx_qcn_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(7), .msg = "tx_pf_rd_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(8), .msg = "rx_pf_rd_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(9), .msg = "qm_eof_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(10), .msg = "mb_rlt_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(11), .msg = "dup_uncopy_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(12), .msg = "dup_cnt_rd_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(13), .msg = "dup_cnt_drop_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(14), .msg = "dup_cnt_wrb_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(15), .msg = "host_cmd_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(16), .msg = "mac_cmd_fifo_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(17), .msg = "host_cmd_bitmap_empty_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(18), .msg = "mac_cmd_bitmap_empty_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(19), .msg = "dup_bitmap_empty_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(20), .msg = "out_queue_bitmap_empty_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(21), .msg = "bank2_bitmap_empty_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(22), .msg = "bank1_bitmap_empty_int", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(23), .msg = "bank0_bitmap_empty_int", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ssu_ets_tcg_int[] = { - { .int_msk = BIT(0), .msg = "ets_rd_int_rx_tcg" }, - { .int_msk = BIT(1), .msg = "ets_wr_int_rx_tcg" }, - { .int_msk = BIT(2), .msg = "ets_rd_int_tx_tcg" }, - { .int_msk = BIT(3), .msg = "ets_wr_int_tx_tcg" }, + { .int_msk = BIT(0), .msg = "ets_rd_int_rx_tcg", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(1), .msg = "ets_wr_int_rx_tcg", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(2), .msg = "ets_rd_int_tx_tcg", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(3), .msg = "ets_wr_int_tx_tcg", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; static const struct hclge_hw_error hclge_ssu_port_based_pf_int[] = { - { .int_msk = BIT(0), .msg = "roc_pkt_without_key_port" }, - { .int_msk = BIT(9), .msg = "low_water_line_err_port" }, - { .int_msk = BIT(10), .msg = "hi_water_line_err_port" }, + { .int_msk = BIT(0), .msg = "roc_pkt_without_key_port", + .reset_level = HNAE3_GLOBAL_RESET }, + { .int_msk = BIT(9), .msg = "low_water_line_err_port", + .reset_level = HNAE3_NONE_RESET }, + { .int_msk = BIT(10), .msg = "hi_water_line_err_port", + .reset_level = HNAE3_GLOBAL_RESET }, { /* sentinel */ } }; @@ -406,16 +631,29 @@ static const struct hclge_hw_error hclge_rocee_qmm_ovf_err_int[] = { { /* sentinel */ } }; -static void hclge_log_error(struct device *dev, char *reg, - const struct hclge_hw_error *err, - u32 err_sts) +static enum hnae3_reset_type hclge_log_error(struct device *dev, char *reg, + const struct hclge_hw_error *err, + u32 err_sts) { + enum hnae3_reset_type reset_level = HNAE3_FUNC_RESET; + bool need_reset = false; + while (err->msg) { - if (err->int_msk & err_sts) + if (err->int_msk & err_sts) { dev_warn(dev, "%s %s found [error status=0x%x]\n", reg, err->msg, err_sts); + if (err->reset_level != HNAE3_NONE_RESET && + err->reset_level >= reset_level) { + reset_level = err->reset_level; + need_reset = true; + } + } err++; } + if (need_reset) + return reset_level; + else + return HNAE3_NONE_RESET; } /* hclge_cmd_query_error: read the error information @@ -826,6 +1064,7 @@ static int hclge_handle_mpf_ras_error(struct hclge_dev *hdev, int num) { struct hnae3_ae_dev *ae_dev = hdev->ae_dev; + enum hnae3_reset_type reset_level; struct device *dev = &hdev->pdev->dev; __le32 *desc_data; u32 status; @@ -845,78 +1084,94 @@ static int hclge_handle_mpf_ras_error(struct hclge_dev *hdev, /* log HNS common errors */ status = le32_to_cpu(desc[0].data[0]); if (status) { - hclge_log_error(dev, "IMP_TCM_ECC_INT_STS", - &hclge_imp_tcm_ecc_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_GLOBAL_RESET); + reset_level = hclge_log_error(dev, "IMP_TCM_ECC_INT_STS", + &hclge_imp_tcm_ecc_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } status = le32_to_cpu(desc[0].data[1]); if (status) { - hclge_log_error(dev, "CMDQ_MEM_ECC_INT_STS", - &hclge_cmdq_nic_mem_ecc_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_GLOBAL_RESET); + reset_level = hclge_log_error(dev, "CMDQ_MEM_ECC_INT_STS", + &hclge_cmdq_nic_mem_ecc_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } if ((le32_to_cpu(desc[0].data[2])) & BIT(0)) { dev_warn(dev, "imp_rd_data_poison_err found\n"); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_GLOBAL_RESET); + HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_NONE_RESET); } status = le32_to_cpu(desc[0].data[3]); if (status) { - hclge_log_error(dev, "TQP_INT_ECC_INT_STS", - &hclge_tqp_int_ecc_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + reset_level = hclge_log_error(dev, "TQP_INT_ECC_INT_STS", + &hclge_tqp_int_ecc_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } status = le32_to_cpu(desc[0].data[4]); if (status) { - hclge_log_error(dev, "MSIX_ECC_INT_STS", - &hclge_msix_sram_ecc_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + reset_level = hclge_log_error(dev, "MSIX_ECC_INT_STS", + &hclge_msix_sram_ecc_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } /* log SSU(Storage Switch Unit) errors */ desc_data = (__le32 *)&desc[2]; status = le32_to_cpu(*(desc_data + 2)); if (status) { - hclge_log_error(dev, "SSU_ECC_MULTI_BIT_INT_0", - &hclge_ssu_mem_ecc_err_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + reset_level = hclge_log_error(dev, "SSU_ECC_MULTI_BIT_INT_0", + &hclge_ssu_mem_ecc_err_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } status = le32_to_cpu(*(desc_data + 3)) & BIT(0); if (status) { dev_warn(dev, "SSU_ECC_MULTI_BIT_INT_1 ssu_mem32_ecc_mbit_err found [error status=0x%x]\n", status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_GLOBAL_RESET); } status = le32_to_cpu(*(desc_data + 4)) & HCLGE_SSU_COMMON_ERR_INT_MASK; if (status) { - hclge_log_error(dev, "SSU_COMMON_ERR_INT", - &hclge_ssu_com_err_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_GLOBAL_RESET); + reset_level = hclge_log_error(dev, "SSU_COMMON_ERR_INT", + &hclge_ssu_com_err_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } /* log IGU(Ingress Unit) errors */ desc_data = (__le32 *)&desc[3]; status = le32_to_cpu(*desc_data) & HCLGE_IGU_INT_MASK; - if (status) - hclge_log_error(dev, "IGU_INT_STS", - &hclge_igu_int[0], status); + if (status) { + reset_level = hclge_log_error(dev, "IGU_INT_STS", + &hclge_igu_int[0], status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); + } /* log PPP(Programmable Packet Process) errors */ desc_data = (__le32 *)&desc[4]; status = le32_to_cpu(*(desc_data + 1)); - if (status) - hclge_log_error(dev, "PPP_MPF_ABNORMAL_INT_ST1", - &hclge_ppp_mpf_abnormal_int_st1[0], status); + if (status) { + reset_level = + hclge_log_error(dev, "PPP_MPF_ABNORMAL_INT_ST1", + &hclge_ppp_mpf_abnormal_int_st1[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); + } status = le32_to_cpu(*(desc_data + 3)) & HCLGE_PPP_MPF_INT_ST3_MASK; - if (status) - hclge_log_error(dev, "PPP_MPF_ABNORMAL_INT_ST3", - &hclge_ppp_mpf_abnormal_int_st3[0], status); + if (status) { + reset_level = + hclge_log_error(dev, "PPP_MPF_ABNORMAL_INT_ST3", + &hclge_ppp_mpf_abnormal_int_st3[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); + } /* log PPU(RCB) errors */ desc_data = (__le32 *)&desc[5]; @@ -924,55 +1179,60 @@ static int hclge_handle_mpf_ras_error(struct hclge_dev *hdev, if (status) { dev_warn(dev, "PPU_MPF_ABNORMAL_INT_ST1 %s found\n", "rpu_rx_pkt_ecc_mbit_err"); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_GLOBAL_RESET); } status = le32_to_cpu(*(desc_data + 2)); if (status) { - hclge_log_error(dev, "PPU_MPF_ABNORMAL_INT_ST2", - &hclge_ppu_mpf_abnormal_int_st2[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + reset_level = + hclge_log_error(dev, "PPU_MPF_ABNORMAL_INT_ST2", + &hclge_ppu_mpf_abnormal_int_st2[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } status = le32_to_cpu(*(desc_data + 3)) & HCLGE_PPU_MPF_INT_ST3_MASK; if (status) { - hclge_log_error(dev, "PPU_MPF_ABNORMAL_INT_ST3", - &hclge_ppu_mpf_abnormal_int_st3[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + reset_level = + hclge_log_error(dev, "PPU_MPF_ABNORMAL_INT_ST3", + &hclge_ppu_mpf_abnormal_int_st3[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } /* log TM(Traffic Manager) errors */ desc_data = (__le32 *)&desc[6]; status = le32_to_cpu(*desc_data); if (status) { - hclge_log_error(dev, "TM_SCH_RINT", - &hclge_tm_sch_rint[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + reset_level = hclge_log_error(dev, "TM_SCH_RINT", + &hclge_tm_sch_rint[0], status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } /* log QCN(Quantized Congestion Control) errors */ desc_data = (__le32 *)&desc[7]; status = le32_to_cpu(*desc_data) & HCLGE_QCN_FIFO_INT_MASK; if (status) { - hclge_log_error(dev, "QCN_FIFO_RINT", - &hclge_qcn_fifo_rint[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + reset_level = hclge_log_error(dev, "QCN_FIFO_RINT", + &hclge_qcn_fifo_rint[0], status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } status = le32_to_cpu(*(desc_data + 1)) & HCLGE_QCN_ECC_INT_MASK; if (status) { - hclge_log_error(dev, "QCN_ECC_RINT", - &hclge_qcn_ecc_rint[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + reset_level = hclge_log_error(dev, "QCN_ECC_RINT", + &hclge_qcn_ecc_rint[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } /* log NCSI errors */ desc_data = (__le32 *)&desc[9]; status = le32_to_cpu(*desc_data) & HCLGE_NCSI_ECC_INT_MASK; if (status) { - hclge_log_error(dev, "NCSI_ECC_INT_RPT", - &hclge_ncsi_err_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_CORE_RESET); + reset_level = hclge_log_error(dev, "NCSI_ECC_INT_RPT", + &hclge_ncsi_err_int[0], status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } /* clear all main PF RAS errors */ @@ -1000,6 +1260,7 @@ static int hclge_handle_pf_ras_error(struct hclge_dev *hdev, { struct hnae3_ae_dev *ae_dev = hdev->ae_dev; struct device *dev = &hdev->pdev->dev; + enum hnae3_reset_type reset_level; __le32 *desc_data; u32 status; int ret; @@ -1018,38 +1279,47 @@ static int hclge_handle_pf_ras_error(struct hclge_dev *hdev, /* log SSU(Storage Switch Unit) errors */ status = le32_to_cpu(desc[0].data[0]); if (status) { - hclge_log_error(dev, "SSU_PORT_BASED_ERR_INT", - &hclge_ssu_port_based_err_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_GLOBAL_RESET); + reset_level = hclge_log_error(dev, "SSU_PORT_BASED_ERR_INT", + &hclge_ssu_port_based_err_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } status = le32_to_cpu(desc[0].data[1]); if (status) { - hclge_log_error(dev, "SSU_FIFO_OVERFLOW_INT", - &hclge_ssu_fifo_overflow_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_GLOBAL_RESET); + reset_level = hclge_log_error(dev, "SSU_FIFO_OVERFLOW_INT", + &hclge_ssu_fifo_overflow_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } status = le32_to_cpu(desc[0].data[2]); if (status) { - hclge_log_error(dev, "SSU_ETS_TCG_INT", - &hclge_ssu_ets_tcg_int[0], status); - HCLGE_SET_DEFAULT_RESET_REQUEST(HNAE3_GLOBAL_RESET); + reset_level = hclge_log_error(dev, "SSU_ETS_TCG_INT", + &hclge_ssu_ets_tcg_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); } /* log IGU(Ingress Unit) EGU(Egress Unit) TNL errors */ desc_data = (__le32 *)&desc[1]; status = le32_to_cpu(*desc_data) & HCLGE_IGU_EGU_TNL_INT_MASK; - if (status) - hclge_log_error(dev, "IGU_EGU_TNL_INT_STS", - &hclge_igu_egu_tnl_int[0], status); + if (status) { + reset_level = hclge_log_error(dev, "IGU_EGU_TNL_INT_STS", + &hclge_igu_egu_tnl_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); + } /* log PPU(RCB) errors */ desc_data = (__le32 *)&desc[3]; status = le32_to_cpu(*desc_data) & HCLGE_PPU_PF_INT_RAS_MASK; - if (status) - hclge_log_error(dev, "PPU_PF_ABNORMAL_INT_ST0", - &hclge_ppu_pf_abnormal_int[0], status); + if (status) { + reset_level = hclge_log_error(dev, "PPU_PF_ABNORMAL_INT_ST0", + &hclge_ppu_pf_abnormal_int[0], + status); + HCLGE_SET_DEFAULT_RESET_REQUEST(reset_level); + } /* clear all PF RAS errors */ hclge_cmd_reuse_desc(&desc[0], false); @@ -1343,14 +1613,12 @@ int hclge_handle_hw_msix_error(struct hclge_dev *hdev, { struct device *dev = &hdev->pdev->dev; u32 mpf_bd_num, pf_bd_num, bd_num; + enum hnae3_reset_type reset_level; struct hclge_desc desc_bd; struct hclge_desc *desc; __le32 *desc_data; - int ret = 0; u32 status; - - /* set default handling */ - set_bit(HNAE3_FUNC_RESET, reset_requests); + int ret; /* query the number of bds for the MSIx int status */ hclge_cmd_setup_basic_desc(&desc_bd, HCLGE_QUERY_MSIX_INT_STS_BD_NUM, @@ -1390,9 +1658,10 @@ int hclge_handle_hw_msix_error(struct hclge_dev *hdev, desc_data = (__le32 *)&desc[1]; status = le32_to_cpu(*desc_data); if (status) { - hclge_log_error(dev, "MAC_AFIFO_TNL_INT_R", - &hclge_mac_afifo_tnl_int[0], status); - set_bit(HNAE3_GLOBAL_RESET, reset_requests); + reset_level = hclge_log_error(dev, "MAC_AFIFO_TNL_INT_R", + &hclge_mac_afifo_tnl_int[0], + status); + set_bit(reset_level, reset_requests); } /* log PPU(RCB) MPF errors */ @@ -1400,9 +1669,11 @@ int hclge_handle_hw_msix_error(struct hclge_dev *hdev, status = le32_to_cpu(*(desc_data + 2)) & HCLGE_PPU_MPF_INT_ST2_MSIX_MASK; if (status) { - hclge_log_error(dev, "PPU_MPF_ABNORMAL_INT_ST2", - &hclge_ppu_mpf_abnormal_int_st2[0], status); - set_bit(HNAE3_CORE_RESET, reset_requests); + reset_level = + hclge_log_error(dev, "PPU_MPF_ABNORMAL_INT_ST2", + &hclge_ppu_mpf_abnormal_int_st2[0], + status); + set_bit(reset_level, reset_requests); } /* clear all main PF MSIx errors */ @@ -1436,24 +1707,31 @@ int hclge_handle_hw_msix_error(struct hclge_dev *hdev, /* log SSU PF errors */ status = le32_to_cpu(desc[0].data[0]) & HCLGE_SSU_PORT_INT_MSIX_MASK; if (status) { - hclge_log_error(dev, "SSU_PORT_BASED_ERR_INT", - &hclge_ssu_port_based_pf_int[0], status); - set_bit(HNAE3_GLOBAL_RESET, reset_requests); + reset_level = hclge_log_error(dev, "SSU_PORT_BASED_ERR_INT", + &hclge_ssu_port_based_pf_int[0], + status); + set_bit(reset_level, reset_requests); } /* read and log PPP PF errors */ desc_data = (__le32 *)&desc[2]; status = le32_to_cpu(*desc_data); - if (status) - hclge_log_error(dev, "PPP_PF_ABNORMAL_INT_ST0", - &hclge_ppp_pf_abnormal_int[0], status); + if (status) { + reset_level = hclge_log_error(dev, "PPP_PF_ABNORMAL_INT_ST0", + &hclge_ppp_pf_abnormal_int[0], + status); + set_bit(reset_level, reset_requests); + } /* log PPU(RCB) PF errors */ desc_data = (__le32 *)&desc[3]; status = le32_to_cpu(*desc_data) & HCLGE_PPU_PF_INT_MSIX_MASK; - if (status) - hclge_log_error(dev, "PPU_PF_ABNORMAL_INT_ST", - &hclge_ppu_pf_abnormal_int[0], status); + if (status) { + reset_level = hclge_log_error(dev, "PPU_PF_ABNORMAL_INT_ST", + &hclge_ppu_pf_abnormal_int[0], + status); + set_bit(reset_level, reset_requests); + } /* clear all PF MSIx errors */ hclge_cmd_reuse_desc(&desc[0], false); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h index fc06828..4a2e82f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h @@ -112,6 +112,7 @@ struct hclge_hw_blk { struct hclge_hw_error { u32 int_msk; const char *msg; + enum hnae3_reset_type reset_level; }; int hclge_hw_error_set_state(struct hclge_dev *hdev, bool state); -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 10/12] net: hns3: do not initialize MDIO bus when PHY is inexistent 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (8 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 09/12] net: hns3: set dividual reset level for all RAS and MSI-X errors Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 11/12] net: hns3: free the pending skb when clean RX ring Huazhong Tan ` (2 subsequent siblings) 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Jian Shen, Peng Li, Huazhong Tan From: Jian Shen <shenjian15@huawei.com> For some cases, PHY may not be connected to MDIO bus, then the driver will initialize fail since MDIO bus initialization fails. This patch fixes it by skipping the MDIO bus initialization when PHY is inexistent. Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Peng Li <lipeng321@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c index 48eda2c..12be4e2 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c @@ -121,12 +121,18 @@ static int hclge_mdio_read(struct mii_bus *bus, int phyid, int regnum) int hclge_mac_mdio_config(struct hclge_dev *hdev) { +#define PHY_INEXISTENT 255 + struct hclge_mac *mac = &hdev->hw.mac; struct phy_device *phydev; struct mii_bus *mdio_bus; int ret; - if (hdev->hw.mac.phy_addr >= PHY_MAX_ADDR) { + if (hdev->hw.mac.phy_addr == PHY_INEXISTENT) { + dev_info(&hdev->pdev->dev, + "no phy device is connected to mdio bus\n"); + return 0; + } else if (hdev->hw.mac.phy_addr >= PHY_MAX_ADDR) { dev_err(&hdev->pdev->dev, "phy_addr(%d) is too large.\n", hdev->hw.mac.phy_addr); return -EINVAL; -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 11/12] net: hns3: free the pending skb when clean RX ring 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (9 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 10/12] net: hns3: do not initialize MDIO bus when PHY is inexistent Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 12/12] net: hns3: code optimization for command queue' spin lock Huazhong Tan 2019-04-14 20:47 ` [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver David Miller 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Peng Li, Huazhong Tan From: Peng Li <lipeng321@huawei.com> If there is pending skb in RX flow when close the port, and the pending buffer is not cleaned, the new packet will be added to the pending skb when the port opens again, and the first new packet has error data. This patch cleans the pending skb when clean RX ring. Signed-off-by: Peng Li <lipeng321@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index a67dfdd..9233438 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -3960,6 +3960,13 @@ static int hns3_clear_rx_ring(struct hns3_enet_ring *ring) ring_ptr_move_fw(ring, next_to_use); } + /* Free the pending skb in rx ring */ + if (ring->skb) { + dev_kfree_skb_any(ring->skb); + ring->skb = NULL; + ring->pending_buf = 0; + } + return 0; } -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH V2 net-next 12/12] net: hns3: code optimization for command queue' spin lock 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (10 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 11/12] net: hns3: free the pending skb when clean RX ring Huazhong Tan @ 2019-04-14 1:47 ` Huazhong Tan 2019-04-14 20:47 ` [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver David Miller 12 siblings, 0 replies; 14+ messages in thread From: Huazhong Tan @ 2019-04-14 1:47 UTC (permalink / raw) To: davem Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm, Peng Li, Huazhong Tan From: Peng Li <lipeng321@huawei.com> This patch removes some redundant BH disable when initializing and uninitializing command queue. Signed-off-by: Peng Li <lipeng321@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c | 4 ++-- drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c index cf7ff4a..fbd904e 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c @@ -355,7 +355,7 @@ int hclge_cmd_init(struct hclge_dev *hdev) int ret; spin_lock_bh(&hdev->hw.cmq.csq.lock); - spin_lock_bh(&hdev->hw.cmq.crq.lock); + spin_lock(&hdev->hw.cmq.crq.lock); hdev->hw.cmq.csq.next_to_clean = 0; hdev->hw.cmq.csq.next_to_use = 0; @@ -364,7 +364,7 @@ int hclge_cmd_init(struct hclge_dev *hdev) hclge_cmd_init_regs(&hdev->hw); - spin_unlock_bh(&hdev->hw.cmq.crq.lock); + spin_unlock(&hdev->hw.cmq.crq.lock); spin_unlock_bh(&hdev->hw.cmq.csq.lock); clear_bit(HCLGE_STATE_CMD_DISABLE, &hdev->state); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c index 054556e..1b428d4 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c @@ -334,7 +334,7 @@ int hclgevf_cmd_init(struct hclgevf_dev *hdev) int ret; spin_lock_bh(&hdev->hw.cmq.csq.lock); - spin_lock_bh(&hdev->hw.cmq.crq.lock); + spin_lock(&hdev->hw.cmq.crq.lock); /* initialize the pointers of async rx queue of mailbox */ hdev->arq.hdev = hdev; @@ -348,7 +348,7 @@ int hclgevf_cmd_init(struct hclgevf_dev *hdev) hclgevf_cmd_init_regs(&hdev->hw); - spin_unlock_bh(&hdev->hw.cmq.crq.lock); + spin_unlock(&hdev->hw.cmq.crq.lock); spin_unlock_bh(&hdev->hw.cmq.csq.lock); clear_bit(HCLGEVF_STATE_CMD_DISABLE, &hdev->state); -- 2.7.4 ^ permalink raw reply related [flat|nested] 14+ messages in thread
* Re: [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan ` (11 preceding siblings ...) 2019-04-14 1:47 ` [PATCH V2 net-next 12/12] net: hns3: code optimization for command queue' spin lock Huazhong Tan @ 2019-04-14 20:47 ` David Miller 12 siblings, 0 replies; 14+ messages in thread From: David Miller @ 2019-04-14 20:47 UTC (permalink / raw) To: tanhuazhong; +Cc: netdev, linux-kernel, salil.mehta, yisen.zhuang, linuxarm From: Huazhong Tan <tanhuazhong@huawei.com> Date: Sun, 14 Apr 2019 09:47:34 +0800 > This patch-set includes code optimizations and bugfixes for the HNS3 > ethernet controller driver. > > [patch 1/12 - 4/12] optimizes the VLAN freature and adds support for port > based VLAN, fixes some related bugs about the current implementation. > > [patch 5/12 - 12/12] includes some other code optimizations for the HNS3 > ethernet controller driver. > > Change log: > V1->V2: modifies some patches' commint log and code. These commit log messages are a lot better, thanks! Series applied. ^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2019-04-14 20:48 UTC | newest] Thread overview: 14+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2019-04-14 1:47 [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 01/12] net: hns3: modify VLAN initialization to be compatible with port based VLAN Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 02/12] net: hns3: fix VLAN offload handle for VLAN inserted by port Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 03/12] net: hns3: fix set port based VLAN for PF Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 04/12] net: hns3: fix set port based VLAN issue for VF Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 05/12] net: hns3: minor refactor for hns3_rx_checksum Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 06/12] net: hns3: add hns3_gro_complete for HW GRO process Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 07/12] net: hns3: always assume no drop TC for performance reason Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 08/12] net: hns3: divide shared buffer between TC Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 09/12] net: hns3: set dividual reset level for all RAS and MSI-X errors Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 10/12] net: hns3: do not initialize MDIO bus when PHY is inexistent Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 11/12] net: hns3: free the pending skb when clean RX ring Huazhong Tan 2019-04-14 1:47 ` [PATCH V2 net-next 12/12] net: hns3: code optimization for command queue' spin lock Huazhong Tan 2019-04-14 20:47 ` [PATCH V2 net-next 00/12] code optimizations & bugfixes for HNS3 driver David Miller
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).