X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=drivers%2Fnet%2Fhns3%2Fhns3_ethdev_vf.c;h=8d9b7979c806d1687b61d5ba496ef31d418b4dad;hb=70d2f42110576c670a67abd35a4bb6a2d0b00980;hp=b5f3e9a867847d8ad07bb3e6d3f6614d09bc0ab6;hpb=28f2e6357662e14643ad156e14d0e35b1d4cea7c;p=dpdk.git diff --git a/drivers/net/hns3/hns3_ethdev_vf.c b/drivers/net/hns3/hns3_ethdev_vf.c index b5f3e9a867..8d9b7979c8 100644 --- a/drivers/net/hns3/hns3_ethdev_vf.c +++ b/drivers/net/hns3/hns3_ethdev_vf.c @@ -1,30 +1,11 @@ /* SPDX-License-Identifier: BSD-3-Clause - * Copyright(c) 2018-2019 Hisilicon Limited. + * Copyright(c) 2018-2021 HiSilicon Limited. */ -#include -#include -#include -#include -#include -#include -#include #include - #include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include +#include #include -#include #include #include @@ -59,13 +40,26 @@ static enum hns3_reset_level hns3vf_get_reset_level(struct hns3_hw *hw, static int hns3vf_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu); static int hns3vf_dev_configure_vlan(struct rte_eth_dev *dev); +static int hns3vf_add_mc_mac_addr(struct hns3_hw *hw, + struct rte_ether_addr *mac_addr); +static int hns3vf_remove_mc_mac_addr(struct hns3_hw *hw, + struct rte_ether_addr *mac_addr); +static int hns3vf_dev_link_update(struct rte_eth_dev *eth_dev, + __rte_unused int wait_to_complete); + /* set PCI bus mastering */ -static void +static int hns3vf_set_bus_master(const struct rte_pci_device *device, bool op) { uint16_t reg; + int ret; - rte_pci_read_config(device, ®, sizeof(reg), PCI_COMMAND); + ret = rte_pci_read_config(device, ®, sizeof(reg), PCI_COMMAND); + if (ret < 0) { + PMD_INIT_LOG(ERR, "Failed to read PCI offset 0x%x", + PCI_COMMAND); + return ret; + } if (op) /* set the master bit */ @@ -73,7 +67,7 @@ hns3vf_set_bus_master(const struct rte_pci_device *device, bool op) else reg &= ~(PCI_COMMAND_MASTER); - rte_pci_write_config(device, ®, sizeof(reg), PCI_COMMAND); + return rte_pci_write_config(device, ®, sizeof(reg), PCI_COMMAND); } /** @@ -90,16 +84,34 @@ hns3vf_find_pci_capability(const struct rte_pci_device *device, int cap) uint8_t pos; uint8_t id; int ttl; + int ret; + + ret = rte_pci_read_config(device, &status, sizeof(status), PCI_STATUS); + if (ret < 0) { + PMD_INIT_LOG(ERR, "Failed to read PCI offset 0x%x", PCI_STATUS); + return 0; + } - rte_pci_read_config(device, &status, sizeof(status), PCI_STATUS); if (!(status & PCI_STATUS_CAP_LIST)) return 0; ttl = MAX_PCIE_CAPABILITY; - rte_pci_read_config(device, &pos, sizeof(pos), PCI_CAPABILITY_LIST); + ret = rte_pci_read_config(device, &pos, sizeof(pos), + PCI_CAPABILITY_LIST); + if (ret < 0) { + PMD_INIT_LOG(ERR, "Failed to read PCI offset 0x%x", + PCI_CAPABILITY_LIST); + return 0; + } + while (ttl-- && pos >= PCI_STD_HEADER_SIZEOF) { - rte_pci_read_config(device, &id, sizeof(id), - (pos + PCI_CAP_LIST_ID)); + ret = rte_pci_read_config(device, &id, sizeof(id), + (pos + PCI_CAP_LIST_ID)); + if (ret < 0) { + PMD_INIT_LOG(ERR, "Failed to read PCI offset 0x%x", + (pos + PCI_CAP_LIST_ID)); + break; + } if (id == 0xFF) break; @@ -107,8 +119,13 @@ hns3vf_find_pci_capability(const struct rte_pci_device *device, int cap) if (id == cap) return (int)pos; - rte_pci_read_config(device, &pos, sizeof(pos), - (pos + PCI_CAP_LIST_NEXT)); + ret = rte_pci_read_config(device, &pos, sizeof(pos), + (pos + PCI_CAP_LIST_NEXT)); + if (ret < 0) { + PMD_INIT_LOG(ERR, "Failed to read PCI offset 0x%x", + (pos + PCI_CAP_LIST_NEXT)); + break; + } } return 0; } @@ -118,40 +135,136 @@ hns3vf_enable_msix(const struct rte_pci_device *device, bool op) { uint16_t control; int pos; + int ret; pos = hns3vf_find_pci_capability(device, PCI_CAP_ID_MSIX); if (pos) { - rte_pci_read_config(device, &control, sizeof(control), + ret = rte_pci_read_config(device, &control, sizeof(control), (pos + PCI_MSIX_FLAGS)); + if (ret < 0) { + PMD_INIT_LOG(ERR, "Failed to read PCI offset 0x%x", + (pos + PCI_MSIX_FLAGS)); + return -ENXIO; + } + if (op) control |= PCI_MSIX_FLAGS_ENABLE; else control &= ~PCI_MSIX_FLAGS_ENABLE; - rte_pci_write_config(device, &control, sizeof(control), - (pos + PCI_MSIX_FLAGS)); + ret = rte_pci_write_config(device, &control, sizeof(control), + (pos + PCI_MSIX_FLAGS)); + if (ret < 0) { + PMD_INIT_LOG(ERR, "failed to write PCI offset 0x%x", + (pos + PCI_MSIX_FLAGS)); + return -ENXIO; + } + return 0; } + return -ENXIO; } static int -hns3vf_add_mac_addr(struct rte_eth_dev *dev, struct rte_ether_addr *mac_addr, - __attribute__ ((unused)) uint32_t idx, - __attribute__ ((unused)) uint32_t pool) +hns3vf_add_uc_mac_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr) { - struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); + /* mac address was checked by upper level interface */ char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; int ret; - rte_spinlock_lock(&hw->lock); ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_UNICAST, HNS3_MBX_MAC_VLAN_UC_ADD, mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN, false, NULL, 0); + if (ret) { + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + mac_addr); + hns3_err(hw, "failed to add uc mac addr(%s), ret = %d", + mac_str, ret); + } + return ret; +} + +static int +hns3vf_remove_uc_mac_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr) +{ + /* mac address was checked by upper level interface */ + char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; + int ret; + + ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_UNICAST, + HNS3_MBX_MAC_VLAN_UC_REMOVE, + mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN, + false, NULL, 0); + if (ret) { + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + mac_addr); + hns3_err(hw, "failed to add uc mac addr(%s), ret = %d", + mac_str, ret); + } + return ret; +} + +static int +hns3vf_add_mc_addr_common(struct hns3_hw *hw, struct rte_ether_addr *mac_addr) +{ + char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; + struct rte_ether_addr *addr; + int ret; + int i; + + for (i = 0; i < hw->mc_addrs_num; i++) { + addr = &hw->mc_addrs[i]; + /* Check if there are duplicate addresses */ + if (rte_is_same_ether_addr(addr, mac_addr)) { + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + addr); + hns3_err(hw, "failed to add mc mac addr, same addrs" + "(%s) is added by the set_mc_mac_addr_list " + "API", mac_str); + return -EINVAL; + } + } + + ret = hns3vf_add_mc_mac_addr(hw, mac_addr); + if (ret) { + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + mac_addr); + hns3_err(hw, "failed to add mc mac addr(%s), ret = %d", + mac_str, ret); + } + return ret; +} + +static int +hns3vf_add_mac_addr(struct rte_eth_dev *dev, struct rte_ether_addr *mac_addr, + __rte_unused uint32_t idx, + __rte_unused uint32_t pool) +{ + struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); + char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; + int ret; + + rte_spinlock_lock(&hw->lock); + + /* + * In hns3 network engine adding UC and MC mac address with different + * commands with firmware. We need to determine whether the input + * address is a UC or a MC address to call different commands. + * By the way, it is recommended calling the API function named + * rte_eth_dev_set_mc_addr_list to set the MC mac address, because + * using the rte_eth_dev_mac_addr_add API function to set MC mac address + * may affect the specifications of UC mac addresses. + */ + if (rte_is_multicast_ether_addr(mac_addr)) + ret = hns3vf_add_mc_addr_common(hw, mac_addr); + else + ret = hns3vf_add_uc_mac_addr(hw, mac_addr); + rte_spinlock_unlock(&hw->lock); if (ret) { - rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, mac_addr); - hns3_err(hw, "Failed to add mac addr(%s) for vf: %d", mac_str, + hns3_err(hw, "failed to add mac addr(%s), ret = %d", mac_str, ret); } @@ -168,15 +281,17 @@ hns3vf_remove_mac_addr(struct rte_eth_dev *dev, uint32_t idx) int ret; rte_spinlock_lock(&hw->lock); - ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_UNICAST, - HNS3_MBX_MAC_VLAN_UC_REMOVE, - mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN, false, - NULL, 0); + + if (rte_is_multicast_ether_addr(mac_addr)) + ret = hns3vf_remove_mc_mac_addr(hw, mac_addr); + else + ret = hns3vf_remove_uc_mac_addr(hw, mac_addr); + rte_spinlock_unlock(&hw->lock); if (ret) { - rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, mac_addr); - hns3_err(hw, "Failed to remove mac addr(%s) for vf: %d", + hns3_err(hw, "failed to remove mac addr(%s), ret = %d", mac_str, ret); } } @@ -192,14 +307,10 @@ hns3vf_set_default_mac_addr(struct rte_eth_dev *dev, char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; int ret; - if (!rte_is_valid_assigned_ether_addr(mac_addr)) { - rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, - mac_addr); - hns3_err(hw, "Failed to set mac addr, addr(%s) invalid.", - mac_str); - return -EINVAL; - } - + /* + * It has been guaranteed that input parameter named mac_addr is valid + * address in the rte layer of DPDK framework. + */ old_addr = (struct rte_ether_addr *)hw->mac.mac_addr; rte_spinlock_lock(&hw->lock); memcpy(addr_bytes, mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN); @@ -208,12 +319,27 @@ hns3vf_set_default_mac_addr(struct rte_eth_dev *dev, ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_UNICAST, HNS3_MBX_MAC_VLAN_UC_MODIFY, addr_bytes, - HNS3_TWO_ETHER_ADDR_LEN, false, NULL, 0); + HNS3_TWO_ETHER_ADDR_LEN, true, NULL, 0); if (ret) { - rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, - mac_addr); - hns3_err(hw, "Failed to set mac addr(%s) for vf: %d", mac_str, - ret); + /* + * The hns3 VF PMD driver depends on the hns3 PF kernel ethdev + * driver. When user has configured a MAC address for VF device + * by "ip link set ..." command based on the PF device, the hns3 + * PF kernel ethdev driver does not allow VF driver to request + * reconfiguring a different default MAC address, and return + * -EPREM to VF driver through mailbox. + */ + if (ret == -EPERM) { + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + old_addr); + hns3_warn(hw, "Has permanet mac addr(%s) for vf", + mac_str); + } else { + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + mac_addr); + hns3_err(hw, "Failed to set mac addr(%s) for vf: %d", + mac_str, ret); + } } rte_ether_addr_copy(mac_addr, @@ -228,39 +354,39 @@ hns3vf_configure_mac_addr(struct hns3_adapter *hns, bool del) { struct hns3_hw *hw = &hns->hw; struct rte_ether_addr *addr; - enum hns3_mbx_mac_vlan_subcode opcode; char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; - int ret = 0; + int err = 0; + int ret; int i; - if (del) - opcode = HNS3_MBX_MAC_VLAN_UC_REMOVE; - else - opcode = HNS3_MBX_MAC_VLAN_UC_ADD; for (i = 0; i < HNS3_VF_UC_MACADDR_NUM; i++) { addr = &hw->data->mac_addrs[i]; - if (!rte_is_valid_assigned_ether_addr(addr)) + if (rte_is_zero_ether_addr(addr)) continue; - rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, addr); - hns3_dbg(hw, "rm mac addr: %s", mac_str); - ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_UNICAST, opcode, - addr->addr_bytes, RTE_ETHER_ADDR_LEN, - false, NULL, 0); + if (rte_is_multicast_ether_addr(addr)) + ret = del ? hns3vf_remove_mc_mac_addr(hw, addr) : + hns3vf_add_mc_mac_addr(hw, addr); + else + ret = del ? hns3vf_remove_uc_mac_addr(hw, addr) : + hns3vf_add_uc_mac_addr(hw, addr); + if (ret) { - hns3_err(hw, "Failed to remove mac addr for vf: %d", - ret); - break; + err = ret; + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + addr); + hns3_err(hw, "failed to %s mac addr(%s) index:%d " + "ret = %d.", del ? "remove" : "restore", + mac_str, i, ret); } } - return ret; + return err; } static int -hns3vf_add_mc_mac_addr(struct hns3_adapter *hns, +hns3vf_add_mc_mac_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr) { char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; - struct hns3_hw *hw = &hns->hw; int ret; ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_MULTICAST, @@ -268,22 +394,20 @@ hns3vf_add_mc_mac_addr(struct hns3_adapter *hns, mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN, false, NULL, 0); if (ret) { - rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, mac_addr); hns3_err(hw, "Failed to add mc mac addr(%s) for vf: %d", mac_str, ret); - return ret; } - return 0; + return ret; } static int -hns3vf_remove_mc_mac_addr(struct hns3_adapter *hns, +hns3vf_remove_mc_mac_addr(struct hns3_hw *hw, struct rte_ether_addr *mac_addr) { char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; - struct hns3_hw *hw = &hns->hw; int ret; ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_MULTICAST, @@ -291,56 +415,102 @@ hns3vf_remove_mc_mac_addr(struct hns3_adapter *hns, mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN, false, NULL, 0); if (ret) { - rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, mac_addr); hns3_err(hw, "Failed to remove mc mac addr(%s) for vf: %d", mac_str, ret); - return ret; } - return 0; + return ret; } static int -hns3vf_set_mc_mac_addr_list(struct rte_eth_dev *dev, - struct rte_ether_addr *mc_addr_set, - uint32_t nb_mc_addr) +hns3vf_set_mc_addr_chk_param(struct hns3_hw *hw, + struct rte_ether_addr *mc_addr_set, + uint32_t nb_mc_addr) { - struct hns3_adapter *hns = dev->data->dev_private; - struct hns3_hw *hw = &hns->hw; - struct rte_ether_addr *addr; char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; - int cur_addr_num; - int set_addr_num; - int num; - int ret; - int i; + struct rte_ether_addr *addr; + uint32_t i; + uint32_t j; if (nb_mc_addr > HNS3_MC_MACADDR_NUM) { - hns3_err(hw, "Failed to set mc mac addr, nb_mc_addr(%d) " + hns3_err(hw, "failed to set mc mac addr, nb_mc_addr(%u) " "invalid. valid range: 0~%d", nb_mc_addr, HNS3_MC_MACADDR_NUM); return -EINVAL; } - set_addr_num = (int)nb_mc_addr; - for (i = 0; i < set_addr_num; i++) { + /* Check if input mac addresses are valid */ + for (i = 0; i < nb_mc_addr; i++) { addr = &mc_addr_set[i]; if (!rte_is_multicast_ether_addr(addr)) { - rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, addr); hns3_err(hw, - "Failed to set mc mac addr, addr(%s) invalid.", + "failed to set mc mac addr, addr(%s) invalid.", mac_str); return -EINVAL; } + + /* Check if there are duplicate addresses */ + for (j = i + 1; j < nb_mc_addr; j++) { + if (rte_is_same_ether_addr(addr, &mc_addr_set[j])) { + hns3_ether_format_addr(mac_str, + RTE_ETHER_ADDR_FMT_SIZE, + addr); + hns3_err(hw, "failed to set mc mac addr, " + "addrs invalid. two same addrs(%s).", + mac_str); + return -EINVAL; + } + } + + /* + * Check if there are duplicate addresses between mac_addrs + * and mc_addr_set + */ + for (j = 0; j < HNS3_VF_UC_MACADDR_NUM; j++) { + if (rte_is_same_ether_addr(addr, + &hw->data->mac_addrs[j])) { + hns3_ether_format_addr(mac_str, + RTE_ETHER_ADDR_FMT_SIZE, + addr); + hns3_err(hw, "failed to set mc mac addr, " + "addrs invalid. addrs(%s) has already " + "configured in mac_addr add API", + mac_str); + return -EINVAL; + } + } } + + return 0; +} + +static int +hns3vf_set_mc_mac_addr_list(struct rte_eth_dev *dev, + struct rte_ether_addr *mc_addr_set, + uint32_t nb_mc_addr) +{ + struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); + struct rte_ether_addr *addr; + int cur_addr_num; + int set_addr_num; + int num; + int ret; + int i; + + ret = hns3vf_set_mc_addr_chk_param(hw, mc_addr_set, nb_mc_addr); + if (ret) + return ret; + rte_spinlock_lock(&hw->lock); cur_addr_num = hw->mc_addrs_num; for (i = 0; i < cur_addr_num; i++) { num = cur_addr_num - i - 1; addr = &hw->mc_addrs[num]; - ret = hns3vf_remove_mc_mac_addr(hns, addr); + ret = hns3vf_remove_mc_mac_addr(hw, addr); if (ret) { rte_spinlock_unlock(&hw->lock); return ret; @@ -349,9 +519,10 @@ hns3vf_set_mc_mac_addr_list(struct rte_eth_dev *dev, hw->mc_addrs_num--; } + set_addr_num = (int)nb_mc_addr; for (i = 0; i < set_addr_num; i++) { addr = &mc_addr_set[i]; - ret = hns3vf_add_mc_mac_addr(hns, addr); + ret = hns3vf_add_mc_mac_addr(hw, addr); if (ret) { rte_spinlock_unlock(&hw->lock); return ret; @@ -380,12 +551,12 @@ hns3vf_configure_all_mc_mac_addr(struct hns3_adapter *hns, bool del) if (!rte_is_multicast_ether_addr(addr)) continue; if (del) - ret = hns3vf_remove_mc_mac_addr(hns, addr); + ret = hns3vf_remove_mc_mac_addr(hw, addr); else - ret = hns3vf_add_mc_mac_addr(hns, addr); + ret = hns3vf_add_mc_mac_addr(hw, addr); if (ret) { err = ret; - rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, addr); hns3_err(hw, "Failed to %s mc mac addr: %s for vf: %d", del ? "Remove" : "Restore", mac_str, ret); @@ -395,7 +566,8 @@ hns3vf_configure_all_mc_mac_addr(struct hns3_adapter *hns, bool del) } static int -hns3vf_set_promisc_mode(struct hns3_hw *hw, bool en_bc_pmc) +hns3vf_set_promisc_mode(struct hns3_hw *hw, bool en_bc_pmc, + bool en_uc_pmc, bool en_mc_pmc) { struct hns3_mbx_vf_to_pf_cmd *req; struct hns3_cmd_desc desc; @@ -403,57 +575,251 @@ hns3vf_set_promisc_mode(struct hns3_hw *hw, bool en_bc_pmc) req = (struct hns3_mbx_vf_to_pf_cmd *)desc.data; + /* + * The hns3 VF PMD driver depends on the hns3 PF kernel ethdev driver, + * so there are some features for promiscuous/allmulticast mode in hns3 + * VF PMD driver as below: + * 1. The promiscuous/allmulticast mode can be configured successfully + * only based on the trusted VF device. If based on the non trusted + * VF device, configuring promiscuous/allmulticast mode will fail. + * The hns3 VF device can be confiruged as trusted device by hns3 PF + * kernel ethdev driver on the host by the following command: + * "ip link set vf turst on" + * 2. After the promiscuous mode is configured successfully, hns3 VF PMD + * driver can receive the ingress and outgoing traffic. In the words, + * all the ingress packets, all the packets sent from the PF and + * other VFs on the same physical port. + * 3. Note: Because of the hardware constraints, By default vlan filter + * is enabled and couldn't be turned off based on VF device, so vlan + * filter is still effective even in promiscuous mode. If upper + * applications don't call rte_eth_dev_vlan_filter API function to + * set vlan based on VF device, hns3 VF PMD driver will can't receive + * the packets with vlan tag in promiscuoue mode. + */ hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_MBX_VF_TO_PF, false); req->msg[0] = HNS3_MBX_SET_PROMISC_MODE; req->msg[1] = en_bc_pmc ? 1 : 0; + req->msg[2] = en_uc_pmc ? 1 : 0; + req->msg[3] = en_mc_pmc ? 1 : 0; + req->msg[4] = hw->promisc_mode == HNS3_LIMIT_PROMISC_MODE ? 1 : 0; ret = hns3_cmd_send(hw, &desc, 1); if (ret) - hns3_err(hw, "Set promisc mode fail, status is %d", ret); + hns3_err(hw, "Set promisc mode fail, ret = %d", ret); + + return ret; +} + +static int +hns3vf_dev_promiscuous_enable(struct rte_eth_dev *dev) +{ + struct hns3_adapter *hns = dev->data->dev_private; + struct hns3_hw *hw = &hns->hw; + int ret; + + ret = hns3vf_set_promisc_mode(hw, true, true, true); + if (ret) + hns3_err(hw, "Failed to enable promiscuous mode, ret = %d", + ret); + return ret; +} + +static int +hns3vf_dev_promiscuous_disable(struct rte_eth_dev *dev) +{ + bool allmulti = dev->data->all_multicast ? true : false; + struct hns3_adapter *hns = dev->data->dev_private; + struct hns3_hw *hw = &hns->hw; + int ret; + + ret = hns3vf_set_promisc_mode(hw, true, false, allmulti); + if (ret) + hns3_err(hw, "Failed to disable promiscuous mode, ret = %d", + ret); + return ret; +} + +static int +hns3vf_dev_allmulticast_enable(struct rte_eth_dev *dev) +{ + struct hns3_adapter *hns = dev->data->dev_private; + struct hns3_hw *hw = &hns->hw; + int ret; + + if (dev->data->promiscuous) + return 0; + + ret = hns3vf_set_promisc_mode(hw, true, false, true); + if (ret) + hns3_err(hw, "Failed to enable allmulticast mode, ret = %d", + ret); + return ret; +} + +static int +hns3vf_dev_allmulticast_disable(struct rte_eth_dev *dev) +{ + struct hns3_adapter *hns = dev->data->dev_private; + struct hns3_hw *hw = &hns->hw; + int ret; + + if (dev->data->promiscuous) + return 0; + + ret = hns3vf_set_promisc_mode(hw, true, false, false); + if (ret) + hns3_err(hw, "Failed to disable allmulticast mode, ret = %d", + ret); + return ret; +} + +static int +hns3vf_restore_promisc(struct hns3_adapter *hns) +{ + struct hns3_hw *hw = &hns->hw; + bool allmulti = hw->data->all_multicast ? true : false; + + if (hw->data->promiscuous) + return hns3vf_set_promisc_mode(hw, true, true, true); + + return hns3vf_set_promisc_mode(hw, true, false, allmulti); +} + +static int +hns3vf_bind_ring_with_vector(struct hns3_hw *hw, uint8_t vector_id, + bool mmap, enum hns3_ring_type queue_type, + uint16_t queue_id) +{ + struct hns3_vf_bind_vector_msg bind_msg; + const char *op_str; + uint16_t code; + int ret; + + memset(&bind_msg, 0, sizeof(bind_msg)); + code = mmap ? HNS3_MBX_MAP_RING_TO_VECTOR : + HNS3_MBX_UNMAP_RING_TO_VECTOR; + bind_msg.vector_id = vector_id; + + if (queue_type == HNS3_RING_TYPE_RX) + bind_msg.param[0].int_gl_index = HNS3_RING_GL_RX; + else + bind_msg.param[0].int_gl_index = HNS3_RING_GL_TX; + + bind_msg.param[0].ring_type = queue_type; + bind_msg.ring_num = 1; + bind_msg.param[0].tqp_index = queue_id; + op_str = mmap ? "Map" : "Unmap"; + ret = hns3_send_mbx_msg(hw, code, 0, (uint8_t *)&bind_msg, + sizeof(bind_msg), false, NULL, 0); + if (ret) + hns3_err(hw, "%s TQP %u fail, vector_id is %u, ret is %d.", + op_str, queue_id, bind_msg.vector_id, ret); return ret; } +static int +hns3vf_init_ring_with_vector(struct hns3_hw *hw) +{ + uint16_t vec; + int ret; + int i; + + /* + * In hns3 network engine, vector 0 is always the misc interrupt of this + * function, vector 1~N can be used respectively for the queues of the + * function. Tx and Rx queues with the same number share the interrupt + * vector. In the initialization clearing the all hardware mapping + * relationship configurations between queues and interrupt vectors is + * needed, so some error caused by the residual configurations, such as + * the unexpected Tx interrupt, can be avoid. + */ + vec = hw->num_msi - 1; /* vector 0 for misc interrupt, not for queue */ + if (hw->intr.mapping_mode == HNS3_INTR_MAPPING_VEC_RSV_ONE) + vec = vec - 1; /* the last interrupt is reserved */ + hw->intr_tqps_num = RTE_MIN(vec, hw->tqps_num); + for (i = 0; i < hw->intr_tqps_num; i++) { + /* + * Set gap limiter/rate limiter/quanity limiter algorithm + * configuration for interrupt coalesce of queue's interrupt. + */ + hns3_set_queue_intr_gl(hw, i, HNS3_RING_GL_RX, + HNS3_TQP_INTR_GL_DEFAULT); + hns3_set_queue_intr_gl(hw, i, HNS3_RING_GL_TX, + HNS3_TQP_INTR_GL_DEFAULT); + hns3_set_queue_intr_rl(hw, i, HNS3_TQP_INTR_RL_DEFAULT); + /* + * QL(quantity limiter) is not used currently, just set 0 to + * close it. + */ + hns3_set_queue_intr_ql(hw, i, HNS3_TQP_INTR_QL_DEFAULT); + + ret = hns3vf_bind_ring_with_vector(hw, vec, false, + HNS3_RING_TYPE_TX, i); + if (ret) { + PMD_INIT_LOG(ERR, "VF fail to unbind TX ring(%d) with " + "vector: %u, ret=%d", i, vec, ret); + return ret; + } + + ret = hns3vf_bind_ring_with_vector(hw, vec, false, + HNS3_RING_TYPE_RX, i); + if (ret) { + PMD_INIT_LOG(ERR, "VF fail to unbind RX ring(%d) with " + "vector: %u, ret=%d", i, vec, ret); + return ret; + } + } + + return 0; +} + static int hns3vf_dev_configure(struct rte_eth_dev *dev) { - struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); - struct hns3_rss_conf *rss_cfg = &hw->rss_info; + struct hns3_adapter *hns = dev->data->dev_private; + struct hns3_hw *hw = &hns->hw; struct rte_eth_conf *conf = &dev->data->dev_conf; enum rte_eth_rx_mq_mode mq_mode = conf->rxmode.mq_mode; uint16_t nb_rx_q = dev->data->nb_rx_queues; uint16_t nb_tx_q = dev->data->nb_tx_queues; struct rte_eth_rss_conf rss_conf; + uint32_t max_rx_pkt_len; uint16_t mtu; + bool gro_en; int ret; + hw->cfg_max_queues = RTE_MAX(nb_rx_q, nb_tx_q); + /* - * Hardware does not support where the number of rx and tx queues is - * not equal in hip08. + * Some versions of hardware network engine does not support + * individually enable/disable/reset the Tx or Rx queue. These devices + * must enable/disable/reset Tx and Rx queues at the same time. When the + * numbers of Tx queues allocated by upper applications are not equal to + * the numbers of Rx queues, driver needs to setup fake Tx or Rx queues + * to adjust numbers of Tx/Rx queues. otherwise, network engine can not + * work as usual. But these fake queues are imperceptible, and can not + * be used by upper applications. */ - if (nb_rx_q != nb_tx_q) { - hns3_err(hw, - "nb_rx_queues(%u) not equal with nb_tx_queues(%u)! " - "Hardware does not support this configuration!", - nb_rx_q, nb_tx_q); - return -EINVAL; + ret = hns3_set_fake_rx_or_tx_queues(dev, nb_rx_q, nb_tx_q); + if (ret) { + hns3_err(hw, "fail to set Rx/Tx fake queues, ret = %d.", ret); + hw->cfg_max_queues = 0; + return ret; } + hw->adapter_state = HNS3_NIC_CONFIGURING; if (conf->link_speeds & ETH_LINK_SPEED_FIXED) { hns3_err(hw, "setting link speed/duplex not supported"); - return -EINVAL; + ret = -EINVAL; + goto cfg_err; } - hw->adapter_state = HNS3_NIC_CONFIGURING; - /* When RSS is not configured, redirect the packet queue 0 */ if ((uint32_t)mq_mode & ETH_MQ_RX_RSS_FLAG) { + conf->rxmode.offloads |= DEV_RX_OFFLOAD_RSS_HASH; + hw->rss_dis_flag = false; rss_conf = conf->rx_adv_conf.rss_conf; - if (rss_conf.rss_key == NULL) { - rss_conf.rss_key = rss_cfg->key; - rss_conf.rss_key_len = HNS3_RSS_KEY_SIZE; - } - ret = hns3_dev_rss_hash_update(dev, &rss_conf); if (ret) goto cfg_err; @@ -464,12 +830,18 @@ hns3vf_dev_configure(struct rte_eth_dev *dev) * according to the maximum RX packet length. */ if (conf->rxmode.offloads & DEV_RX_OFFLOAD_JUMBO_FRAME) { - /* - * Security of max_rx_pkt_len is guaranteed in dpdk frame. - * Maximum value of max_rx_pkt_len is HNS3_MAX_FRAME_LEN, so it - * can safely assign to "uint16_t" type variable. - */ - mtu = (uint16_t)HNS3_PKTLEN_TO_MTU(conf->rxmode.max_rx_pkt_len); + max_rx_pkt_len = conf->rxmode.max_rx_pkt_len; + if (max_rx_pkt_len > HNS3_MAX_FRAME_LEN || + max_rx_pkt_len <= HNS3_DEFAULT_FRAME_LEN) { + hns3_err(hw, "maximum Rx packet length must be greater " + "than %u and less than %u when jumbo frame enabled.", + (uint16_t)HNS3_DEFAULT_FRAME_LEN, + (uint16_t)HNS3_MAX_FRAME_LEN); + ret = -EINVAL; + goto cfg_err; + } + + mtu = (uint16_t)HNS3_PKTLEN_TO_MTU(max_rx_pkt_len); ret = hns3vf_dev_mtu_set(dev, mtu); if (ret) goto cfg_err; @@ -480,11 +852,22 @@ hns3vf_dev_configure(struct rte_eth_dev *dev) if (ret) goto cfg_err; + /* config hardware GRO */ + gro_en = conf->rxmode.offloads & DEV_RX_OFFLOAD_TCP_LRO ? true : false; + ret = hns3_config_gro(hw, gro_en); + if (ret) + goto cfg_err; + + hns3_init_rx_ptype_tble(dev); + hw->adapter_state = HNS3_NIC_CONFIGURED; return 0; cfg_err: + hw->cfg_max_queues = 0; + (void)hns3_set_fake_rx_or_tx_queues(dev, 0, 0); hw->adapter_state = HNS3_NIC_INITIALIZED; + return ret; } @@ -508,24 +891,45 @@ hns3vf_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu) uint32_t frame_size = mtu + HNS3_ETH_OVERHEAD; int ret; - if (dev->data->dev_started) { - hns3_err(hw, "Failed to set mtu, port %u must be stopped " - "before configuration", dev->data->port_id); - return -EBUSY; - } - - if (rte_atomic16_read(&hw->reset.resetting)) { + /* + * The hns3 PF/VF devices on the same port share the hardware MTU + * configuration. Currently, we send mailbox to inform hns3 PF kernel + * ethdev driver to finish hardware MTU configuration in hns3 VF PMD + * driver, there is no need to stop the port for hns3 VF device, and the + * MTU value issued by hns3 VF PMD driver must be less than or equal to + * PF's MTU. + */ + if (__atomic_load_n(&hw->reset.resetting, __ATOMIC_RELAXED)) { hns3_err(hw, "Failed to set mtu during resetting"); return -EIO; } + /* + * when Rx of scattered packets is off, we have some possibility of + * using vector Rx process function or simple Rx functions in hns3 PMD + * driver. If the input MTU is increased and the maximum length of + * received packets is greater than the length of a buffer for Rx + * packet, the hardware network engine needs to use multiple BDs and + * buffers to store these packets. This will cause problems when still + * using vector Rx process function or simple Rx function to receiving + * packets. So, when Rx of scattered packets is off and device is + * started, it is not permitted to increase MTU so that the maximum + * length of Rx packets is greater than Rx buffer length. + */ + if (dev->data->dev_started && !dev->data->scattered_rx && + frame_size > hw->rx_buf_len) { + hns3_err(hw, "failed to set mtu because current is " + "not scattered rx mode"); + return -EOPNOTSUPP; + } + rte_spinlock_lock(&hw->lock); ret = hns3vf_config_mtu(hw, mtu); if (ret) { rte_spinlock_unlock(&hw->lock); return ret; } - if (frame_size > RTE_ETHER_MAX_LEN) + if (mtu > RTE_ETHER_MTU) dev->data->dev_conf.rxmode.offloads |= DEV_RX_OFFLOAD_JUMBO_FRAME; else @@ -542,13 +946,22 @@ hns3vf_dev_infos_get(struct rte_eth_dev *eth_dev, struct rte_eth_dev_info *info) { struct hns3_adapter *hns = eth_dev->data->dev_private; struct hns3_hw *hw = &hns->hw; + uint16_t q_num = hw->tqps_num; + + /* + * In interrupt mode, 'max_rx_queues' is set based on the number of + * MSI-X interrupt resources of the hardware. + */ + if (hw->data->dev_conf.intr_conf.rxq == 1) + q_num = hw->intr_tqps_num; - info->max_rx_queues = hw->tqps_num; + info->max_rx_queues = q_num; info->max_tx_queues = hw->tqps_num; info->max_rx_pktlen = HNS3_MAX_FRAME_LEN; /* CRC included */ - info->min_rx_bufsize = hw->rx_buf_len; + info->min_rx_bufsize = HNS3_MIN_BD_BUF_SIZE; info->max_mac_addrs = HNS3_VF_UC_MACADDR_NUM; info->max_mtu = info->max_rx_pktlen - HNS3_ETH_OVERHEAD; + info->max_lro_pkt_size = HNS3_MAX_LRO_SIZE; info->rx_offload_capa = (DEV_RX_OFFLOAD_IPV4_CKSUM | DEV_RX_OFFLOAD_UDP_CKSUM | @@ -556,22 +969,31 @@ hns3vf_dev_infos_get(struct rte_eth_dev *eth_dev, struct rte_eth_dev_info *info) DEV_RX_OFFLOAD_SCTP_CKSUM | DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM | DEV_RX_OFFLOAD_OUTER_UDP_CKSUM | - DEV_RX_OFFLOAD_KEEP_CRC | DEV_RX_OFFLOAD_SCATTER | DEV_RX_OFFLOAD_VLAN_STRIP | - DEV_RX_OFFLOAD_QINQ_STRIP | DEV_RX_OFFLOAD_VLAN_FILTER | - DEV_RX_OFFLOAD_JUMBO_FRAME); - info->tx_queue_offload_capa = DEV_TX_OFFLOAD_MBUF_FAST_FREE; + DEV_RX_OFFLOAD_JUMBO_FRAME | + DEV_RX_OFFLOAD_RSS_HASH | + DEV_RX_OFFLOAD_TCP_LRO); info->tx_offload_capa = (DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM | DEV_TX_OFFLOAD_IPV4_CKSUM | DEV_TX_OFFLOAD_TCP_CKSUM | DEV_TX_OFFLOAD_UDP_CKSUM | DEV_TX_OFFLOAD_SCTP_CKSUM | - DEV_TX_OFFLOAD_VLAN_INSERT | - DEV_TX_OFFLOAD_QINQ_INSERT | DEV_TX_OFFLOAD_MULTI_SEGS | - info->tx_queue_offload_capa); + DEV_TX_OFFLOAD_TCP_TSO | + DEV_TX_OFFLOAD_VXLAN_TNL_TSO | + DEV_TX_OFFLOAD_GRE_TNL_TSO | + DEV_TX_OFFLOAD_GENEVE_TNL_TSO | + DEV_TX_OFFLOAD_MBUF_FAST_FREE | + hns3_txvlan_cap_get(hw)); + + if (hns3_dev_outer_udp_cksum_supported(hw)) + info->tx_offload_capa |= DEV_TX_OFFLOAD_OUTER_UDP_CKSUM; + + if (hns3_dev_indep_txrx_supported(hw)) + info->dev_capa = RTE_ETH_DEV_CAPA_RUNTIME_RX_QUEUE_SETUP | + RTE_ETH_DEV_CAPA_RUNTIME_TX_QUEUE_SETUP; info->rx_desc_lim = (struct rte_eth_desc_lim) { .nb_max = HNS3_MAX_RING_DESC, @@ -583,13 +1005,33 @@ hns3vf_dev_infos_get(struct rte_eth_dev *eth_dev, struct rte_eth_dev_info *info) .nb_max = HNS3_MAX_RING_DESC, .nb_min = HNS3_MIN_RING_DESC, .nb_align = HNS3_ALIGN_RING_DESC, + .nb_seg_max = HNS3_MAX_TSO_BD_PER_PKT, + .nb_mtu_seg_max = hw->max_non_tso_bd_num, }; - info->vmdq_queue_num = 0; + info->default_rxconf = (struct rte_eth_rxconf) { + .rx_free_thresh = HNS3_DEFAULT_RX_FREE_THRESH, + /* + * If there are no available Rx buffer descriptors, incoming + * packets are always dropped by hardware based on hns3 network + * engine. + */ + .rx_drop_en = 1, + .offloads = 0, + }; + info->default_txconf = (struct rte_eth_txconf) { + .tx_rs_thresh = HNS3_DEFAULT_TX_RS_THRESH, + .offloads = 0, + }; - info->reta_size = HNS3_RSS_IND_TBL_SIZE; + info->reta_size = hw->rss_ind_tbl_size; info->hash_key_size = HNS3_RSS_KEY_SIZE; info->flow_type_rss_offloads = HNS3_ETH_RSS_SUPPORT; + + info->default_rxportconf.burst_size = HNS3_DEFAULT_PORT_CONF_BURST_SIZE; + info->default_txportconf.burst_size = HNS3_DEFAULT_PORT_CONF_BURST_SIZE; + info->default_rxportconf.nb_queues = HNS3_DEFAULT_PORT_CONF_QUEUES_NUM; + info->default_txportconf.nb_queues = HNS3_DEFAULT_PORT_CONF_QUEUES_NUM; info->default_rxportconf.ring_size = HNS3_DEFAULT_RING_DESC; info->default_txportconf.ring_size = HNS3_DEFAULT_RING_DESC; @@ -625,12 +1067,11 @@ hns3vf_check_event_cause(struct hns3_adapter *hns, uint32_t *clearval) /* Fetch the events from their corresponding regs */ cmdq_stat_reg = hns3_read_dev(hw, HNS3_VECTOR0_CMDQ_STAT_REG); - if (BIT(HNS3_VECTOR0_RST_INT_B) & cmdq_stat_reg) { rst_ing_reg = hns3_read_dev(hw, HNS3_FUN_RST_ING); hns3_warn(hw, "resetting reg: 0x%x", rst_ing_reg); hns3_atomic_set_bit(HNS3_VF_RESET, &hw->reset.pending); - rte_atomic16_set(&hw->reset.disable_cmd, 1); + __atomic_store_n(&hw->reset.disable_cmd, 1, __ATOMIC_RELAXED); val = hns3_read_dev(hw, HNS3_VF_RST_ING); hns3_write_dev(hw, HNS3_VF_RST_ING, val | HNS3_VF_RST_ING_BIT); val = cmdq_stat_reg & ~BIT(HNS3_VECTOR0_RST_INT_B); @@ -670,14 +1111,13 @@ hns3vf_interrupt_handler(void *param) enum hns3vf_evt_cause event_cause; uint32_t clearval; - if (hw->irq_thread_id == 0) - hw->irq_thread_id = pthread_self(); - /* Disable interrupt */ hns3vf_disable_irq0(hw); /* Read out interrupt causes */ event_cause = hns3vf_check_event_cause(hns, &clearval); + /* Clear interrupt causes */ + hns3vf_clear_event_cause(hw, clearval); switch (event_cause) { case HNS3VF_VECTOR0_EVENT_RST: @@ -690,33 +1130,242 @@ hns3vf_interrupt_handler(void *param) break; } - /* Clear interrupt causes */ - hns3vf_clear_event_cause(hw, clearval); - /* Enable interrupt */ hns3vf_enable_irq0(hw); } +static void +hns3vf_set_default_dev_specifications(struct hns3_hw *hw) +{ + hw->max_non_tso_bd_num = HNS3_MAX_NON_TSO_BD_PER_PKT; + hw->rss_ind_tbl_size = HNS3_RSS_IND_TBL_SIZE; + hw->rss_key_size = HNS3_RSS_KEY_SIZE; + hw->intr.int_ql_max = HNS3_INTR_QL_NONE; +} + +static void +hns3vf_parse_dev_specifications(struct hns3_hw *hw, struct hns3_cmd_desc *desc) +{ + struct hns3_dev_specs_0_cmd *req0; + + req0 = (struct hns3_dev_specs_0_cmd *)desc[0].data; + + hw->max_non_tso_bd_num = req0->max_non_tso_bd_num; + hw->rss_ind_tbl_size = rte_le_to_cpu_16(req0->rss_ind_tbl_size); + hw->rss_key_size = rte_le_to_cpu_16(req0->rss_key_size); + hw->intr.int_ql_max = rte_le_to_cpu_16(req0->intr_ql_max); +} + +static int +hns3vf_check_dev_specifications(struct hns3_hw *hw) +{ + if (hw->rss_ind_tbl_size == 0 || + hw->rss_ind_tbl_size > HNS3_RSS_IND_TBL_SIZE_MAX) { + hns3_warn(hw, "the size of hash lookup table configured (%u)" + " exceeds the maximum(%u)", hw->rss_ind_tbl_size, + HNS3_RSS_IND_TBL_SIZE_MAX); + return -EINVAL; + } + + return 0; +} + +static int +hns3vf_query_dev_specifications(struct hns3_hw *hw) +{ + struct hns3_cmd_desc desc[HNS3_QUERY_DEV_SPECS_BD_NUM]; + int ret; + int i; + + for (i = 0; i < HNS3_QUERY_DEV_SPECS_BD_NUM - 1; i++) { + hns3_cmd_setup_basic_desc(&desc[i], HNS3_OPC_QUERY_DEV_SPECS, + true); + desc[i].flag |= rte_cpu_to_le_16(HNS3_CMD_FLAG_NEXT); + } + hns3_cmd_setup_basic_desc(&desc[i], HNS3_OPC_QUERY_DEV_SPECS, true); + + ret = hns3_cmd_send(hw, desc, HNS3_QUERY_DEV_SPECS_BD_NUM); + if (ret) + return ret; + + hns3vf_parse_dev_specifications(hw, desc); + + return hns3vf_check_dev_specifications(hw); +} + +void +hns3vf_update_push_lsc_cap(struct hns3_hw *hw, bool supported) +{ + uint16_t val = supported ? HNS3_PF_PUSH_LSC_CAP_SUPPORTED : + HNS3_PF_PUSH_LSC_CAP_NOT_SUPPORTED; + uint16_t exp = HNS3_PF_PUSH_LSC_CAP_UNKNOWN; + struct hns3_vf *vf = HNS3_DEV_HW_TO_VF(hw); + + if (vf->pf_push_lsc_cap == HNS3_PF_PUSH_LSC_CAP_UNKNOWN) + __atomic_compare_exchange(&vf->pf_push_lsc_cap, &exp, &val, 0, + __ATOMIC_ACQUIRE, __ATOMIC_ACQUIRE); +} + +static void +hns3vf_get_push_lsc_cap(struct hns3_hw *hw) +{ +#define HNS3_CHECK_PUSH_LSC_CAP_TIMEOUT_MS 500 + + struct rte_eth_dev *dev = &rte_eth_devices[hw->data->port_id]; + int32_t remain_ms = HNS3_CHECK_PUSH_LSC_CAP_TIMEOUT_MS; + uint16_t val = HNS3_PF_PUSH_LSC_CAP_NOT_SUPPORTED; + uint16_t exp = HNS3_PF_PUSH_LSC_CAP_UNKNOWN; + struct hns3_vf *vf = HNS3_DEV_HW_TO_VF(hw); + + __atomic_store_n(&vf->pf_push_lsc_cap, HNS3_PF_PUSH_LSC_CAP_UNKNOWN, + __ATOMIC_RELEASE); + + (void)hns3_send_mbx_msg(hw, HNS3_MBX_GET_LINK_STATUS, 0, NULL, 0, false, + NULL, 0); + + while (remain_ms > 0) { + rte_delay_ms(HNS3_POLL_RESPONE_MS); + if (__atomic_load_n(&vf->pf_push_lsc_cap, __ATOMIC_ACQUIRE) != + HNS3_PF_PUSH_LSC_CAP_UNKNOWN) + break; + remain_ms--; + } + + /* + * When exit above loop, the pf_push_lsc_cap could be one of the three + * state: unknown (means pf not ack), not_supported, supported. + * Here config it as 'not_supported' when it's 'unknown' state. + */ + __atomic_compare_exchange(&vf->pf_push_lsc_cap, &exp, &val, 0, + __ATOMIC_ACQUIRE, __ATOMIC_ACQUIRE); + + if (__atomic_load_n(&vf->pf_push_lsc_cap, __ATOMIC_ACQUIRE) == + HNS3_PF_PUSH_LSC_CAP_SUPPORTED) { + hns3_info(hw, "detect PF support push link status change!"); + } else { + /* + * Framework already set RTE_ETH_DEV_INTR_LSC bit because driver + * declared RTE_PCI_DRV_INTR_LSC in drv_flags. So here cleared + * the RTE_ETH_DEV_INTR_LSC capability. + */ + dev->data->dev_flags &= ~RTE_ETH_DEV_INTR_LSC; + } +} + +static int +hns3vf_get_capability(struct hns3_hw *hw) +{ + struct rte_pci_device *pci_dev; + struct rte_eth_dev *eth_dev; + uint8_t revision; + int ret; + + eth_dev = &rte_eth_devices[hw->data->port_id]; + pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev); + + /* Get PCI revision id */ + ret = rte_pci_read_config(pci_dev, &revision, HNS3_PCI_REVISION_ID_LEN, + HNS3_PCI_REVISION_ID); + if (ret != HNS3_PCI_REVISION_ID_LEN) { + PMD_INIT_LOG(ERR, "failed to read pci revision id, ret = %d", + ret); + return -EIO; + } + hw->revision = revision; + + if (revision < PCI_REVISION_ID_HIP09_A) { + hns3vf_set_default_dev_specifications(hw); + hw->intr.mapping_mode = HNS3_INTR_MAPPING_VEC_RSV_ONE; + hw->intr.gl_unit = HNS3_INTR_COALESCE_GL_UINT_2US; + hw->tso_mode = HNS3_TSO_SW_CAL_PSEUDO_H_CSUM; + hw->drop_stats_mode = HNS3_PKTS_DROP_STATS_MODE1; + hw->min_tx_pkt_len = HNS3_HIP08_MIN_TX_PKT_LEN; + hw->rss_info.ipv6_sctp_offload_supported = false; + hw->promisc_mode = HNS3_UNLIMIT_PROMISC_MODE; + return 0; + } + + ret = hns3vf_query_dev_specifications(hw); + if (ret) { + PMD_INIT_LOG(ERR, + "failed to query dev specifications, ret = %d", + ret); + return ret; + } + + hw->intr.mapping_mode = HNS3_INTR_MAPPING_VEC_ALL; + hw->intr.gl_unit = HNS3_INTR_COALESCE_GL_UINT_1US; + hw->tso_mode = HNS3_TSO_HW_CAL_PSEUDO_H_CSUM; + hw->drop_stats_mode = HNS3_PKTS_DROP_STATS_MODE2; + hw->min_tx_pkt_len = HNS3_HIP09_MIN_TX_PKT_LEN; + hw->rss_info.ipv6_sctp_offload_supported = true; + hw->promisc_mode = HNS3_LIMIT_PROMISC_MODE; + + return 0; +} + static int hns3vf_check_tqp_info(struct hns3_hw *hw) { - uint16_t tqps_num; + if (hw->tqps_num == 0) { + PMD_INIT_LOG(ERR, "Get invalid tqps_num(0) from PF."); + return -EINVAL; + } - tqps_num = hw->tqps_num; - if (tqps_num > HNS3_MAX_TQP_NUM_PER_FUNC || tqps_num == 0) { - PMD_INIT_LOG(ERR, "Get invalid tqps_num(%u) from PF. valid " - "range: 1~%d", - tqps_num, HNS3_MAX_TQP_NUM_PER_FUNC); + if (hw->rss_size_max == 0) { + PMD_INIT_LOG(ERR, "Get invalid rss_size_max(0) from PF."); return -EINVAL; } - if (hw->rx_buf_len == 0) - hw->rx_buf_len = HNS3_DEFAULT_RX_BUF_LEN; - hw->alloc_rss_size = RTE_MIN(hw->rss_size_max, hw->tqps_num); + hw->tqps_num = RTE_MIN(hw->rss_size_max, hw->tqps_num); return 0; } +static int +hns3vf_get_port_base_vlan_filter_state(struct hns3_hw *hw) +{ + uint8_t resp_msg; + int ret; + + ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_VLAN, + HNS3_MBX_GET_PORT_BASE_VLAN_STATE, NULL, 0, + true, &resp_msg, sizeof(resp_msg)); + if (ret) { + if (ret == -ETIME) { + /* + * Getting current port based VLAN state from PF driver + * will not affect VF driver's basic function. Because + * the VF driver relies on hns3 PF kernel ether driver, + * to avoid introducing compatibility issues with older + * version of PF driver, no failure will be returned + * when the return value is ETIME. This return value has + * the following scenarios: + * 1) Firmware didn't return the results in time + * 2) the result return by firmware is timeout + * 3) the older version of kernel side PF driver does + * not support this mailbox message. + * For scenarios 1 and 2, it is most likely that a + * hardware error has occurred, or a hardware reset has + * occurred. In this case, these errors will be caught + * by other functions. + */ + PMD_INIT_LOG(WARNING, + "failed to get PVID state for timeout, maybe " + "kernel side PF driver doesn't support this " + "mailbox message, or firmware didn't respond."); + resp_msg = HNS3_PORT_BASE_VLAN_DISABLE; + } else { + PMD_INIT_LOG(ERR, "failed to get port based VLAN state," + " ret = %d", ret); + return ret; + } + } + hw->port_base_vlan_cfg.state = resp_msg ? + HNS3_PORT_BASE_VLAN_ENABLE : HNS3_PORT_BASE_VLAN_DISABLE; + return 0; +} + static int hns3vf_get_queue_info(struct hns3_hw *hw) { @@ -733,7 +1382,6 @@ hns3vf_get_queue_info(struct hns3_hw *hw) memcpy(&hw->tqps_num, &resp_msg[0], sizeof(uint16_t)); memcpy(&hw->rss_size_max, &resp_msg[2], sizeof(uint16_t)); - memcpy(&hw->rx_buf_len, &resp_msg[4], sizeof(uint16_t)); return hns3vf_check_tqp_info(hw); } @@ -759,21 +1407,65 @@ hns3vf_get_queue_depth(struct hns3_hw *hw) return 0; } +static void +hns3vf_update_caps(struct hns3_hw *hw, uint32_t caps) +{ + if (hns3_get_bit(caps, HNS3VF_CAPS_VLAN_FLT_MOD_B)) + hns3_set_bit(hw->capability, + HNS3_DEV_SUPPORT_VF_VLAN_FLT_MOD_B, 1); +} + static int -hns3vf_get_tc_info(struct hns3_hw *hw) +hns3vf_get_num_tc(struct hns3_hw *hw) { - uint8_t resp_msg; + uint8_t num_tc = 0; + uint32_t i; + + for (i = 0; i < HNS3_MAX_TC_NUM; i++) { + if (hw->hw_tc_map & BIT(i)) + num_tc++; + } + return num_tc; +} + +static int +hns3vf_get_basic_info(struct hns3_hw *hw) +{ + uint8_t resp_msg[HNS3_MBX_MAX_RESP_DATA_SIZE]; + struct hns3_basic_info *basic_info; int ret; - ret = hns3_send_mbx_msg(hw, HNS3_MBX_GET_TCINFO, 0, NULL, 0, - true, &resp_msg, sizeof(resp_msg)); + ret = hns3_send_mbx_msg(hw, HNS3_MBX_GET_BASIC_INFO, 0, NULL, 0, + true, resp_msg, sizeof(resp_msg)); if (ret) { - hns3_err(hw, "VF request to get TC info from PF failed %d", - ret); + hns3_err(hw, "failed to get basic info from PF, ret = %d.", + ret); + return ret; + } + + basic_info = (struct hns3_basic_info *)resp_msg; + hw->hw_tc_map = basic_info->hw_tc_map; + hw->num_tc = hns3vf_get_num_tc(hw); + hw->pf_vf_if_version = basic_info->pf_vf_if_version; + hns3vf_update_caps(hw, basic_info->caps); + + return 0; +} + +static int +hns3vf_get_host_mac_addr(struct hns3_hw *hw) +{ + uint8_t host_mac[RTE_ETHER_ADDR_LEN]; + int ret; + + ret = hns3_send_mbx_msg(hw, HNS3_MBX_GET_MAC_ADDR, 0, NULL, 0, + true, host_mac, RTE_ETHER_ADDR_LEN); + if (ret) { + hns3_err(hw, "Failed to get mac addr from PF: %d", ret); return ret; } - hw->hw_tc_map = resp_msg; + memcpy(hw->mac.mac_addr, host_mac, RTE_ETHER_ADDR_LEN); return 0; } @@ -784,6 +1476,21 @@ hns3vf_get_configuration(struct hns3_hw *hw) int ret; hw->mac.media_type = HNS3_MEDIA_TYPE_NONE; + hw->rss_dis_flag = false; + + /* Get device capability */ + ret = hns3vf_get_capability(hw); + if (ret) { + PMD_INIT_LOG(ERR, "failed to get device capability: %d.", ret); + return ret; + } + + hns3vf_get_push_lsc_cap(hw); + + /* Get basic info from PF */ + ret = hns3vf_get_basic_info(hw); + if (ret) + return ret; /* Get queue configuration from PF */ ret = hns3vf_get_queue_info(hw); @@ -795,42 +1502,78 @@ hns3vf_get_configuration(struct hns3_hw *hw) if (ret) return ret; - /* Get tc configuration from PF */ - return hns3vf_get_tc_info(hw); + /* Get user defined VF MAC addr from PF */ + ret = hns3vf_get_host_mac_addr(hw); + if (ret) + return ret; + + return hns3vf_get_port_base_vlan_filter_state(hw); } -static void -hns3vf_set_tc_info(struct hns3_adapter *hns) +static int +hns3vf_set_tc_queue_mapping(struct hns3_adapter *hns, uint16_t nb_rx_q, + uint16_t nb_tx_q) { struct hns3_hw *hw = &hns->hw; - uint16_t nb_rx_q = hw->data->nb_rx_queues; - uint16_t new_tqps; - uint8_t i; - - hw->num_tc = 0; - for (i = 0; i < HNS3_MAX_TC_NUM; i++) - if (hw->hw_tc_map & BIT(i)) - hw->num_tc++; - - new_tqps = RTE_MIN(hw->tqps_num, nb_rx_q); - hw->alloc_rss_size = RTE_MIN(hw->rss_size_max, new_tqps / hw->num_tc); - hw->alloc_tqps = hw->alloc_rss_size * hw->num_tc; - hns3_tc_queue_mapping_cfg(hw); + return hns3_queue_to_tc_mapping(hw, nb_rx_q, nb_tx_q); } static void hns3vf_request_link_info(struct hns3_hw *hw) { - uint8_t resp_msg; + struct hns3_vf *vf = HNS3_DEV_HW_TO_VF(hw); + bool send_req; int ret; - if (rte_atomic16_read(&hw->reset.resetting)) + if (__atomic_load_n(&hw->reset.resetting, __ATOMIC_RELAXED)) + return; + + send_req = vf->pf_push_lsc_cap == HNS3_PF_PUSH_LSC_CAP_NOT_SUPPORTED || + vf->req_link_info_cnt > 0; + if (!send_req) return; + ret = hns3_send_mbx_msg(hw, HNS3_MBX_GET_LINK_STATUS, 0, NULL, 0, false, - &resp_msg, sizeof(resp_msg)); - if (ret) - hns3_err(hw, "Failed to fetch link status from PF: %d", ret); + NULL, 0); + if (ret) { + hns3_err(hw, "failed to fetch link status, ret = %d", ret); + return; + } + + if (vf->req_link_info_cnt > 0) + vf->req_link_info_cnt--; +} + +void +hns3vf_update_link_status(struct hns3_hw *hw, uint8_t link_status, + uint32_t link_speed, uint8_t link_duplex) +{ + struct rte_eth_dev *dev = &rte_eth_devices[hw->data->port_id]; + struct hns3_vf *vf = HNS3_DEV_HW_TO_VF(hw); + struct hns3_mac *mac = &hw->mac; + int ret; + + /* + * PF kernel driver may push link status when VF driver is in resetting, + * driver will stop polling job in this case, after resetting done + * driver will start polling job again. + * When polling job started, driver will get initial link status by + * sending request to PF kernel driver, then could update link status by + * process PF kernel driver's link status mailbox message. + */ + if (!__atomic_load_n(&vf->poll_job_started, __ATOMIC_RELAXED)) + return; + + if (hw->adapter_state != HNS3_NIC_STARTED) + return; + + mac->link_status = link_status; + mac->link_speed = link_speed; + mac->link_duplex = link_duplex; + ret = hns3vf_dev_link_update(dev, 0); + if (ret == 0 && dev->data->dev_conf.intr_conf.lsc != 0) + hns3_start_report_lse(dev); } static int @@ -858,7 +1601,7 @@ hns3vf_vlan_filter_set(struct rte_eth_dev *dev, uint16_t vlan_id, int on) struct hns3_hw *hw = &hns->hw; int ret; - if (rte_atomic16_read(&hw->reset.resetting)) { + if (__atomic_load_n(&hw->reset.resetting, __ATOMIC_RELAXED)) { hns3_err(hw, "vf set vlan id failed during resetting, vlan_id =%u", vlan_id); @@ -874,6 +1617,26 @@ hns3vf_vlan_filter_set(struct rte_eth_dev *dev, uint16_t vlan_id, int on) return ret; } +static int +hns3vf_en_vlan_filter(struct hns3_hw *hw, bool enable) +{ + uint8_t msg_data; + int ret; + + if (!hns3_dev_vf_vlan_flt_supported(hw)) + return 0; + + msg_data = enable ? 1 : 0; + ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_VLAN, + HNS3_MBX_ENABLE_VLAN_FILTER, &msg_data, + sizeof(msg_data), true, NULL, 0); + if (ret) + hns3_err(hw, "%s vlan filter failed, ret = %d.", + enable ? "enable" : "disable", ret); + + return ret; +} + static int hns3vf_en_hw_strip_rxvtag(struct hns3_hw *hw, bool enable) { @@ -884,7 +1647,8 @@ hns3vf_en_hw_strip_rxvtag(struct hns3_hw *hw, bool enable) ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_VLAN, HNS3_MBX_VLAN_RX_OFF_CFG, &msg_data, sizeof(msg_data), false, NULL, 0); if (ret) - hns3_err(hw, "vf enable strip failed, ret =%d", ret); + hns3_err(hw, "vf %s strip failed, ret = %d.", + enable ? "enable" : "disable", ret); return ret; } @@ -895,20 +1659,40 @@ hns3vf_vlan_offload_set(struct rte_eth_dev *dev, int mask) struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); struct rte_eth_conf *dev_conf = &dev->data->dev_conf; unsigned int tmp_mask; + int ret = 0; + + if (__atomic_load_n(&hw->reset.resetting, __ATOMIC_RELAXED)) { + hns3_err(hw, "vf set vlan offload failed during resetting, " + "mask = 0x%x", mask); + return -EIO; + } tmp_mask = (unsigned int)mask; + + if (tmp_mask & ETH_VLAN_FILTER_MASK) { + rte_spinlock_lock(&hw->lock); + /* Enable or disable VLAN filter */ + if (dev_conf->rxmode.offloads & DEV_RX_OFFLOAD_VLAN_FILTER) + ret = hns3vf_en_vlan_filter(hw, true); + else + ret = hns3vf_en_vlan_filter(hw, false); + rte_spinlock_unlock(&hw->lock); + if (ret) + return ret; + } + /* Vlan stripping setting */ if (tmp_mask & ETH_VLAN_STRIP_MASK) { rte_spinlock_lock(&hw->lock); /* Enable or disable VLAN stripping */ if (dev_conf->rxmode.offloads & DEV_RX_OFFLOAD_VLAN_STRIP) - hns3vf_en_hw_strip_rxvtag(hw, true); + ret = hns3vf_en_hw_strip_rxvtag(hw, true); else - hns3vf_en_hw_strip_rxvtag(hw, false); + ret = hns3vf_en_hw_strip_rxvtag(hw, false); rte_spinlock_unlock(&hw->lock); } - return 0; + return ret; } static int @@ -994,9 +1778,10 @@ hns3vf_dev_configure_vlan(struct rte_eth_dev *dev) } /* Apply vlan offload setting */ - ret = hns3vf_vlan_offload_set(dev, ETH_VLAN_STRIP_MASK); + ret = hns3vf_vlan_offload_set(dev, ETH_VLAN_STRIP_MASK | + ETH_VLAN_FILTER_MASK); if (ret) - hns3_err(hw, "dev config vlan offload failed, ret =%d", ret); + hns3_err(hw, "dev config vlan offload failed, ret = %d.", ret); return ret; } @@ -1017,11 +1802,10 @@ hns3vf_keep_alive_handler(void *param) struct rte_eth_dev *eth_dev = (struct rte_eth_dev *)param; struct hns3_adapter *hns = eth_dev->data->dev_private; struct hns3_hw *hw = &hns->hw; - uint8_t respmsg; int ret; ret = hns3_send_mbx_msg(hw, HNS3_MBX_KEEP_ALIVE, 0, NULL, 0, - false, &respmsg, sizeof(uint8_t)); + false, NULL, 0); if (ret) hns3_err(hw, "VF sends keeping alive cmd failed(=%d)", ret); @@ -1039,8 +1823,8 @@ hns3vf_service_handler(void *param) /* * The query link status and reset processing are executed in the - * interrupt thread.When the IMP reset occurs, IMP will not respond, - * and the query operation will time out after 30ms. In the case of + * interrupt thread. When the IMP reset occurs, IMP will not respond, + * and the query operation will timeout after 30ms. In the case of * multiple PF/VFs, each query failure timeout causes the IMP reset * interrupt to fail to respond within 100ms. * Before querying the link status, check whether there is a reset @@ -1055,6 +1839,60 @@ hns3vf_service_handler(void *param) eth_dev); } +static void +hns3vf_start_poll_job(struct rte_eth_dev *dev) +{ +#define HNS3_REQUEST_LINK_INFO_REMAINS_CNT 3 + + struct hns3_vf *vf = HNS3_DEV_PRIVATE_TO_VF(dev->data->dev_private); + + if (vf->pf_push_lsc_cap == HNS3_PF_PUSH_LSC_CAP_SUPPORTED) + vf->req_link_info_cnt = HNS3_REQUEST_LINK_INFO_REMAINS_CNT; + + __atomic_store_n(&vf->poll_job_started, 1, __ATOMIC_RELAXED); + + hns3vf_service_handler(dev); +} + +static void +hns3vf_stop_poll_job(struct rte_eth_dev *dev) +{ + struct hns3_vf *vf = HNS3_DEV_PRIVATE_TO_VF(dev->data->dev_private); + + rte_eal_alarm_cancel(hns3vf_service_handler, dev); + + __atomic_store_n(&vf->poll_job_started, 0, __ATOMIC_RELAXED); +} + +static int +hns3_query_vf_resource(struct hns3_hw *hw) +{ + struct hns3_vf_res_cmd *req; + struct hns3_cmd_desc desc; + uint16_t num_msi; + int ret; + + hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_QUERY_VF_RSRC, true); + ret = hns3_cmd_send(hw, &desc, 1); + if (ret) { + hns3_err(hw, "query vf resource failed, ret = %d", ret); + return ret; + } + + req = (struct hns3_vf_res_cmd *)desc.data; + num_msi = hns3_get_field(rte_le_to_cpu_16(req->vf_intr_vector_number), + HNS3_VF_VEC_NUM_M, HNS3_VF_VEC_NUM_S); + if (num_msi < HNS3_MIN_VECTOR_NUM) { + hns3_err(hw, "Just %u msi resources, not enough for vf(min:%d)", + num_msi, HNS3_MIN_VECTOR_NUM); + return -EINVAL; + } + + hw->num_msi = num_msi; + + return 0; +} + static int hns3vf_init_hardware(struct hns3_adapter *hns) { @@ -1062,7 +1900,7 @@ hns3vf_init_hardware(struct hns3_adapter *hns) uint16_t mtu = hw->data->mtu; int ret; - ret = hns3vf_set_promisc_mode(hw, true); + ret = hns3vf_set_promisc_mode(hw, true, false, false); if (ret) return ret; @@ -1082,17 +1920,22 @@ hns3vf_init_hardware(struct hns3_adapter *hns) goto err_init_hardware; } - ret = hns3vf_set_alive(hw, true); + /* + * In the initialization clearing the all hardware mapping relationship + * configurations between queues and interrupt vectors is needed, so + * some error caused by the residual configurations, such as the + * unexpected interrupt, can be avoid. + */ + ret = hns3vf_init_ring_with_vector(hw); if (ret) { - PMD_INIT_LOG(ERR, "Failed to VF send alive to PF: %d", ret); + PMD_INIT_LOG(ERR, "Failed to init ring intr vector: %d", ret); goto err_init_hardware; } - hns3vf_request_link_info(hw); return 0; err_init_hardware: - (void)hns3vf_set_promisc_mode(hw, false); + (void)hns3vf_set_promisc_mode(hw, false, false, false); return ret; } @@ -1131,6 +1974,13 @@ hns3vf_init_vf(struct rte_eth_dev *eth_dev) goto err_cmd_init; } + hns3_tx_push_init(eth_dev); + + /* Get VF resource */ + ret = hns3_query_vf_resource(hw); + if (ret) + goto err_cmd_init; + rte_spinlock_init(&hw->mbx_resp.lock); hns3vf_clear_event_cause(hw, 0); @@ -1153,34 +2003,55 @@ hns3vf_init_vf(struct rte_eth_dev *eth_dev) goto err_get_config; } - rte_eth_random_addr(hw->mac.mac_addr); /* Generate a random mac addr */ + ret = hns3_tqp_stats_init(hw); + if (ret) + goto err_get_config; + + /* Hardware statistics of imissed registers cleared. */ + ret = hns3_update_imissed_stats(hw, true); + if (ret) { + hns3_err(hw, "clear imissed stats failed, ret = %d", ret); + goto err_set_tc_queue; + } + + ret = hns3vf_set_tc_queue_mapping(hns, hw->tqps_num, hw->tqps_num); + if (ret) { + PMD_INIT_LOG(ERR, "failed to set tc info, ret = %d.", ret); + goto err_set_tc_queue; + } ret = hns3vf_clear_vport_list(hw); if (ret) { PMD_INIT_LOG(ERR, "Failed to clear tbl list: %d", ret); - goto err_get_config; + goto err_set_tc_queue; } ret = hns3vf_init_hardware(hns); if (ret) - goto err_get_config; + goto err_set_tc_queue; + + hns3_rss_set_default_args(hw); - hns3_set_default_rss_args(hw); + ret = hns3vf_set_alive(hw, true); + if (ret) { + PMD_INIT_LOG(ERR, "Failed to VF send alive to PF: %d", ret); + goto err_set_tc_queue; + } - (void)hns3_stats_reset(eth_dev); return 0; +err_set_tc_queue: + hns3_tqp_stats_uninit(hw); + err_get_config: hns3vf_disable_irq0(hw); rte_intr_disable(&pci_dev->intr_handle); hns3_intr_unregister(&pci_dev->intr_handle, hns3vf_interrupt_handler, eth_dev); err_intr_callback_register: - hns3_cmd_uninit(hw); - err_cmd_init: + hns3_cmd_uninit(hw); hns3_cmd_destroy_queue(hw); - err_cmd_init_queue: hw->io_base = NULL; @@ -1197,8 +2068,11 @@ hns3vf_uninit_vf(struct rte_eth_dev *eth_dev) PMD_INIT_FUNC_TRACE(); hns3_rss_uninit(hns); + (void)hns3_config_gro(hw, false); (void)hns3vf_set_alive(hw, false); - (void)hns3vf_set_promisc_mode(hw, false); + (void)hns3vf_set_promisc_mode(hw, false, false, false); + hns3_flow_uninit(eth_dev); + hns3_tqp_stats_uninit(hw); hns3vf_disable_irq0(hw); rte_intr_disable(&pci_dev->intr_handle); hns3_intr_unregister(&pci_dev->intr_handle, hns3vf_interrupt_handler, @@ -1208,59 +2082,45 @@ hns3vf_uninit_vf(struct rte_eth_dev *eth_dev) hw->io_base = NULL; } -static int -hns3vf_bind_ring_with_vector(struct rte_eth_dev *dev, uint8_t vector_id, - bool mmap, uint16_t queue_id) - -{ - struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); - struct hns3_vf_bind_vector_msg bind_msg; - uint16_t code; - int ret; - - memset(&bind_msg, 0, sizeof(bind_msg)); - code = mmap ? HNS3_MBX_MAP_RING_TO_VECTOR : - HNS3_MBX_UNMAP_RING_TO_VECTOR; - bind_msg.vector_id = vector_id; - bind_msg.ring_num = 1; - bind_msg.param[0].ring_type = HNS3_RING_TYPE_RX; - bind_msg.param[0].tqp_index = queue_id; - bind_msg.param[0].int_gl_index = HNS3_RING_GL_RX; - - ret = hns3_send_mbx_msg(hw, code, 0, (uint8_t *)&bind_msg, - sizeof(bind_msg), false, NULL, 0); - if (ret) { - hns3_err(hw, "Map TQP %d fail, vector_id is %d, ret is %d.", - queue_id, vector_id, ret); - return ret; - } - - return 0; -} - static int hns3vf_do_stop(struct hns3_adapter *hns) { struct hns3_hw *hw = &hns->hw; - bool reset_queue; + int ret; hw->mac.link_status = ETH_LINK_DOWN; - if (rte_atomic16_read(&hw->reset.disable_cmd) == 0) { + /* + * The "hns3vf_do_stop" function will also be called by .stop_service to + * prepare reset. At the time of global or IMP reset, the command cannot + * be sent to stop the tx/rx queues. The mbuf in Tx/Rx queues may be + * accessed during the reset process. So the mbuf can not be released + * during reset and is required to be released after the reset is + * completed. + */ + if (__atomic_load_n(&hw->reset.resetting, __ATOMIC_RELAXED) == 0) + hns3_dev_release_mbufs(hns); + + if (__atomic_load_n(&hw->reset.disable_cmd, __ATOMIC_RELAXED) == 0) { hns3vf_configure_mac_addr(hns, true); - reset_queue = true; - } else - reset_queue = false; - return hns3_stop_queues(hns, reset_queue); + ret = hns3_reset_all_tqps(hns); + if (ret) { + hns3_err(hw, "failed to reset all queues ret = %d", + ret); + return ret; + } + } + return 0; } static void hns3vf_unmap_rx_interrupt(struct rte_eth_dev *dev) { + struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev); struct rte_intr_handle *intr_handle = &pci_dev->intr_handle; - uint8_t base = 0; - uint8_t vec = 0; + uint8_t base = RTE_INTR_VEC_ZERO_OFFSET; + uint8_t vec = RTE_INTR_VEC_ZERO_OFFSET; uint16_t q_id; if (dev->data->dev_conf.intr_conf.rxq == 0) @@ -1272,8 +2132,9 @@ hns3vf_unmap_rx_interrupt(struct rte_eth_dev *dev) base = RTE_INTR_VEC_RXTX_OFFSET; } if (rte_intr_dp_is_en(intr_handle)) { - for (q_id = 0; q_id < dev->data->nb_rx_queues; q_id++) { - (void)hns3vf_bind_ring_with_vector(dev, vec, false, + for (q_id = 0; q_id < hw->used_rx_queues; q_id++) { + (void)hns3vf_bind_ring_with_vector(hw, vec, false, + HNS3_RING_TYPE_RX, q_id); if (vec < base + intr_handle->nb_efd - 1) vec++; @@ -1287,13 +2148,14 @@ hns3vf_unmap_rx_interrupt(struct rte_eth_dev *dev) } } -static void +static int hns3vf_dev_stop(struct rte_eth_dev *dev) { struct hns3_adapter *hns = dev->data->dev_private; struct hns3_hw *hw = &hns->hw; PMD_INIT_FUNC_TRACE(); + dev->data->dev_started = 0; hw->adapter_state = HNS3_NIC_STOPPING; hns3_set_rxtx_function(dev); @@ -1301,31 +2163,35 @@ hns3vf_dev_stop(struct rte_eth_dev *dev) /* Disable datapath on secondary process. */ hns3_mp_req_stop_rxtx(dev); /* Prevent crashes when queues are still in use. */ - rte_delay_ms(hw->tqps_num); + rte_delay_ms(hw->cfg_max_queues); rte_spinlock_lock(&hw->lock); - if (rte_atomic16_read(&hw->reset.resetting) == 0) { + if (__atomic_load_n(&hw->reset.resetting, __ATOMIC_RELAXED) == 0) { + hns3_stop_tqps(hw); hns3vf_do_stop(hns); - hns3_dev_release_mbufs(hns); + hns3vf_unmap_rx_interrupt(dev); hw->adapter_state = HNS3_NIC_CONFIGURED; } - rte_eal_alarm_cancel(hns3vf_service_handler, dev); + hns3_rx_scattered_reset(dev); + hns3vf_stop_poll_job(dev); + hns3_stop_report_lse(dev); rte_spinlock_unlock(&hw->lock); - hns3vf_unmap_rx_interrupt(dev); + return 0; } -static void +static int hns3vf_dev_close(struct rte_eth_dev *eth_dev) { struct hns3_adapter *hns = eth_dev->data->dev_private; struct hns3_hw *hw = &hns->hw; + int ret = 0; if (rte_eal_process_type() != RTE_PROC_PRIMARY) - return; + return 0; if (hw->adapter_state == HNS3_NIC_STARTED) - hns3vf_dev_stop(eth_dev); + ret = hns3vf_dev_stop(eth_dev); hw->adapter_state = HNS3_NIC_CLOSING; hns3_reset_abort(hns); @@ -1336,10 +2202,38 @@ hns3vf_dev_close(struct rte_eth_dev *eth_dev) hns3vf_uninit_vf(eth_dev); hns3_free_all_queues(eth_dev); rte_free(hw->reset.wait_data); - rte_free(eth_dev->process_private); - eth_dev->process_private = NULL; hns3_mp_uninit_primary(); - hns3_warn(hw, "Close port %d finished", hw->data->port_id); + hns3_warn(hw, "Close port %u finished", hw->data->port_id); + + return ret; +} + +static int +hns3vf_fw_version_get(struct rte_eth_dev *eth_dev, char *fw_version, + size_t fw_size) +{ + struct hns3_adapter *hns = eth_dev->data->dev_private; + struct hns3_hw *hw = &hns->hw; + uint32_t version = hw->fw_version; + int ret; + + ret = snprintf(fw_version, fw_size, "%lu.%lu.%lu.%lu", + hns3_get_field(version, HNS3_FW_VERSION_BYTE3_M, + HNS3_FW_VERSION_BYTE3_S), + hns3_get_field(version, HNS3_FW_VERSION_BYTE2_M, + HNS3_FW_VERSION_BYTE2_S), + hns3_get_field(version, HNS3_FW_VERSION_BYTE1_M, + HNS3_FW_VERSION_BYTE1_S), + hns3_get_field(version, HNS3_FW_VERSION_BYTE0_M, + HNS3_FW_VERSION_BYTE0_S)); + if (ret < 0) + return -EINVAL; + + ret += 1; /* add the size of '\0' */ + if (fw_size < (size_t)ret) + return ret; + else + return 0; } static int @@ -1361,13 +2255,19 @@ hns3vf_dev_link_update(struct rte_eth_dev *eth_dev, case ETH_SPEED_NUM_40G: case ETH_SPEED_NUM_50G: case ETH_SPEED_NUM_100G: - new_link.link_speed = mac->link_speed; + case ETH_SPEED_NUM_200G: + if (mac->link_status) + new_link.link_speed = mac->link_speed; break; default: - new_link.link_speed = ETH_SPEED_NUM_100M; + if (mac->link_status) + new_link.link_speed = ETH_SPEED_NUM_UNKNOWN; break; } + if (!mac->link_status) + new_link.link_speed = ETH_SPEED_NUM_NONE; + new_link.link_duplex = mac->link_duplex; new_link.link_status = mac->link_status ? ETH_LINK_UP : ETH_LINK_DOWN; new_link.link_autoneg = @@ -1380,17 +2280,21 @@ static int hns3vf_do_start(struct hns3_adapter *hns, bool reset_queue) { struct hns3_hw *hw = &hns->hw; + uint16_t nb_rx_q = hw->data->nb_rx_queues; + uint16_t nb_tx_q = hw->data->nb_tx_queues; int ret; - hns3vf_set_tc_info(hns); - - ret = hns3_start_queues(hns, reset_queue); - if (ret) { - hns3_err(hw, "Failed to start queues: %d", ret); + ret = hns3vf_set_tc_queue_mapping(hns, nb_rx_q, nb_tx_q); + if (ret) return ret; - } - return 0; + hns3_enable_rxd_adv_layout(hw); + + ret = hns3_init_queues(hns, reset_queue); + if (ret) + hns3_err(hw, "failed to init queues, ret = %d.", ret); + + return ret; } static int @@ -1399,33 +2303,34 @@ hns3vf_map_rx_interrupt(struct rte_eth_dev *dev) struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev); struct rte_intr_handle *intr_handle = &pci_dev->intr_handle; struct hns3_hw *hw = HNS3_DEV_PRIVATE_TO_HW(dev->data->dev_private); + uint8_t base = RTE_INTR_VEC_ZERO_OFFSET; + uint8_t vec = RTE_INTR_VEC_ZERO_OFFSET; uint32_t intr_vector; - uint8_t base = 0; - uint8_t vec = 0; uint16_t q_id; int ret; - if (dev->data->dev_conf.intr_conf.rxq == 0) + /* + * hns3 needs a separate interrupt to be used as event interrupt which + * could not be shared with task queue pair, so KERNEL drivers need + * support multiple interrupt vectors. + */ + if (dev->data->dev_conf.intr_conf.rxq == 0 || + !rte_intr_cap_multiple(intr_handle)) return 0; - /* disable uio/vfio intr/eventfd mapping */ rte_intr_disable(intr_handle); + intr_vector = hw->used_rx_queues; + /* It creates event fd for each intr vector when MSIX is used */ + if (rte_intr_efd_enable(intr_handle, intr_vector)) + return -EINVAL; - /* check and configure queue intr-vector mapping */ - if (rte_intr_cap_multiple(intr_handle) || - !RTE_ETH_DEV_SRIOV(dev).active) { - intr_vector = dev->data->nb_rx_queues; - /* It creates event fd for each intr vector when MSIX is used */ - if (rte_intr_efd_enable(intr_handle, intr_vector)) - return -EINVAL; - } - if (rte_intr_dp_is_en(intr_handle) && !intr_handle->intr_vec) { + if (intr_handle->intr_vec == NULL) { intr_handle->intr_vec = rte_zmalloc("intr_vec", - dev->data->nb_rx_queues * sizeof(int), 0); + hw->used_rx_queues * sizeof(int), 0); if (intr_handle->intr_vec == NULL) { - hns3_err(hw, "Failed to allocate %d rx_queues" - " intr_vec", dev->data->nb_rx_queues); + hns3_err(hw, "Failed to allocate %u rx_queues" + " intr_vec", hw->used_rx_queues); ret = -ENOMEM; goto vf_alloc_intr_vec_error; } @@ -1435,41 +2340,71 @@ hns3vf_map_rx_interrupt(struct rte_eth_dev *dev) vec = RTE_INTR_VEC_RXTX_OFFSET; base = RTE_INTR_VEC_RXTX_OFFSET; } - if (rte_intr_dp_is_en(intr_handle)) { - for (q_id = 0; q_id < dev->data->nb_rx_queues; q_id++) { - ret = hns3vf_bind_ring_with_vector(dev, vec, true, - q_id); - if (ret) - goto vf_bind_vector_error; - intr_handle->intr_vec[q_id] = vec; - if (vec < base + intr_handle->nb_efd - 1) - vec++; - } + + for (q_id = 0; q_id < hw->used_rx_queues; q_id++) { + ret = hns3vf_bind_ring_with_vector(hw, vec, true, + HNS3_RING_TYPE_RX, q_id); + if (ret) + goto vf_bind_vector_error; + intr_handle->intr_vec[q_id] = vec; + /* + * If there are not enough efds (e.g. not enough interrupt), + * remaining queues will be bond to the last interrupt. + */ + if (vec < base + intr_handle->nb_efd - 1) + vec++; } rte_intr_enable(intr_handle); return 0; vf_bind_vector_error: - rte_intr_efd_disable(intr_handle); - if (intr_handle->intr_vec) { - free(intr_handle->intr_vec); - intr_handle->intr_vec = NULL; - } - return ret; + free(intr_handle->intr_vec); + intr_handle->intr_vec = NULL; vf_alloc_intr_vec_error: rte_intr_efd_disable(intr_handle); return ret; } +static int +hns3vf_restore_rx_interrupt(struct hns3_hw *hw) +{ + struct rte_eth_dev *dev = &rte_eth_devices[hw->data->port_id]; + struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev); + struct rte_intr_handle *intr_handle = &pci_dev->intr_handle; + uint16_t q_id; + int ret; + + if (dev->data->dev_conf.intr_conf.rxq == 0) + return 0; + + if (rte_intr_dp_is_en(intr_handle)) { + for (q_id = 0; q_id < hw->used_rx_queues; q_id++) { + ret = hns3vf_bind_ring_with_vector(hw, + intr_handle->intr_vec[q_id], true, + HNS3_RING_TYPE_RX, q_id); + if (ret) + return ret; + } + } + + return 0; +} + +static void +hns3vf_restore_filter(struct rte_eth_dev *dev) +{ + hns3_restore_rss_filter(dev); +} + static int hns3vf_dev_start(struct rte_eth_dev *dev) { struct hns3_adapter *hns = dev->data->dev_private; struct hns3_hw *hw = &hns->hw; - int ret = 0; + int ret; PMD_INIT_FUNC_TRACE(); - if (rte_atomic16_read(&hw->reset.resetting)) + if (__atomic_load_n(&hw->reset.resetting, __ATOMIC_RELAXED)) return -EBUSY; rte_spinlock_lock(&hw->lock); @@ -1480,15 +2415,55 @@ hns3vf_dev_start(struct rte_eth_dev *dev) rte_spinlock_unlock(&hw->lock); return ret; } + ret = hns3vf_map_rx_interrupt(dev); + if (ret) + goto map_rx_inter_err; + + /* + * There are three register used to control the status of a TQP + * (contains a pair of Tx queue and Rx queue) in the new version network + * engine. One is used to control the enabling of Tx queue, the other is + * used to control the enabling of Rx queue, and the last is the master + * switch used to control the enabling of the tqp. The Tx register and + * TQP register must be enabled at the same time to enable a Tx queue. + * The same applies to the Rx queue. For the older network enginem, this + * function only refresh the enabled flag, and it is used to update the + * status of queue in the dpdk framework. + */ + ret = hns3_start_all_txqs(dev); + if (ret) + goto map_rx_inter_err; + + ret = hns3_start_all_rxqs(dev); + if (ret) + goto start_all_rxqs_fail; + hw->adapter_state = HNS3_NIC_STARTED; rte_spinlock_unlock(&hw->lock); - ret = hns3vf_map_rx_interrupt(dev); - if (ret) - return ret; + hns3_rx_scattered_calc(dev); hns3_set_rxtx_function(dev); hns3_mp_req_start_rxtx(dev); - rte_eal_alarm_set(HNS3VF_SERVICE_INTERVAL, hns3vf_service_handler, dev); + + hns3vf_restore_filter(dev); + + /* Enable interrupt of all rx queues before enabling queues */ + hns3_dev_all_rx_queue_intr_enable(hw, true); + hns3_start_tqps(hw); + + if (dev->data->dev_conf.intr_conf.lsc != 0) + hns3vf_dev_link_update(dev, 0); + hns3vf_start_poll_job(dev); + + return ret; + +start_all_rxqs_fail: + hns3_stop_all_txqs(dev); +map_rx_inter_err: + (void)hns3vf_do_stop(hns); + hw->adapter_state = HNS3_NIC_CONFIGURED; + rte_spinlock_unlock(&hw->lock); + return ret; } @@ -1521,9 +2496,25 @@ hns3vf_is_reset_pending(struct hns3_adapter *hns) struct hns3_hw *hw = &hns->hw; enum hns3_reset_level reset; + /* + * According to the protocol of PCIe, FLR to a PF device resets the PF + * state as well as the SR-IOV extended capability including VF Enable + * which means that VFs no longer exist. + * + * HNS3_VF_FULL_RESET means PF device is in FLR reset. when PF device + * is in FLR stage, the register state of VF device is not reliable, + * so register states detection can not be carried out. In this case, + * we just ignore the register states and return false to indicate that + * there are no other reset states that need to be processed by driver. + */ + if (hw->reset.level == HNS3_VF_FULL_RESET) + return false; + + /* Check the registers to confirm whether there is reset pending */ hns3vf_check_event_cause(hns, NULL); reset = hns3vf_get_reset_level(hw, &hw->reset.pending); - if (hw->reset.level != HNS3_NONE_RESET && hw->reset.level < reset) { + if (hw->reset.level != HNS3_NONE_RESET && reset != HNS3_NONE_RESET && + hw->reset.level < reset) { hns3_warn(hw, "High level reset %d is pending", reset); return true; } @@ -1561,7 +2552,7 @@ hns3vf_wait_hardware_ready(struct hns3_adapter *hns) hns3_warn(hw, "hardware is ready, delay 1 sec for PF reset complete"); return -EAGAIN; } else if (wait_data->result == HNS3_WAIT_TIMEOUT) { - gettimeofday(&tv, NULL); + hns3_clock_gettime(&tv); hns3_warn(hw, "Reset step4 hardware not ready after reset time=%ld.%.6ld", tv.tv_sec, tv.tv_usec); return -ETIME; @@ -1571,7 +2562,7 @@ hns3vf_wait_hardware_ready(struct hns3_adapter *hns) wait_data->hns = hns; wait_data->check_completion = is_vf_reset_done; wait_data->end_ms = (uint64_t)HNS3VF_RESET_WAIT_CNT * - HNS3VF_RESET_WAIT_MS + get_timeofday_ms(); + HNS3VF_RESET_WAIT_MS + hns3_clock_gettime_ms(); wait_data->interval = HNS3VF_RESET_WAIT_MS * USEC_PER_MSEC; wait_data->count = HNS3VF_RESET_WAIT_CNT; wait_data->result = HNS3_WAIT_REQUEST; @@ -1583,15 +2574,17 @@ static int hns3vf_prepare_reset(struct hns3_adapter *hns) { struct hns3_hw *hw = &hns->hw; - int ret = 0; + int ret; if (hw->reset.level == HNS3_VF_FUNC_RESET) { ret = hns3_send_mbx_msg(hw, HNS3_MBX_RESET, 0, NULL, 0, true, NULL, 0); + if (ret) + return ret; } - rte_atomic16_set(&hw->reset.disable_cmd, 1); + __atomic_store_n(&hw->reset.disable_cmd, 1, __ATOMIC_RELAXED); - return ret; + return 0; } static int @@ -1601,18 +2594,27 @@ hns3vf_stop_service(struct hns3_adapter *hns) struct rte_eth_dev *eth_dev; eth_dev = &rte_eth_devices[hw->data->port_id]; - rte_eal_alarm_cancel(hns3vf_service_handler, eth_dev); + if (hw->adapter_state == HNS3_NIC_STARTED) { + /* + * Make sure call update link status before hns3vf_stop_poll_job + * because update link status depend on polling job exist. + */ + hns3vf_update_link_status(hw, ETH_LINK_DOWN, hw->mac.link_speed, + hw->mac.link_duplex); + hns3vf_stop_poll_job(eth_dev); + } hw->mac.link_status = ETH_LINK_DOWN; hns3_set_rxtx_function(eth_dev); rte_wmb(); /* Disable datapath on secondary process. */ hns3_mp_req_stop_rxtx(eth_dev); - rte_delay_ms(hw->tqps_num); + rte_delay_ms(hw->cfg_max_queues); rte_spinlock_lock(&hw->lock); if (hw->adapter_state == HNS3_NIC_STARTED || hw->adapter_state == HNS3_NIC_STOPPING) { + hns3_enable_all_queues(hw, false); hns3vf_do_stop(hns); hw->reset.mbuf_deferred_free = true; } else @@ -1623,7 +2625,7 @@ hns3vf_stop_service(struct hns3_adapter *hns) * from table space. Hence, for function reset software intervention is * required to delete the entries. */ - if (rte_atomic16_read(&hw->reset.disable_cmd) == 0) + if (__atomic_load_n(&hw->reset.disable_cmd, __ATOMIC_RELAXED) == 0) hns3vf_configure_all_mc_mac_addr(hns, true); rte_spinlock_unlock(&hw->lock); @@ -1639,8 +2641,64 @@ hns3vf_start_service(struct hns3_adapter *hns) eth_dev = &rte_eth_devices[hw->data->port_id]; hns3_set_rxtx_function(eth_dev); hns3_mp_req_start_rxtx(eth_dev); + if (hw->adapter_state == HNS3_NIC_STARTED) { + hns3vf_start_poll_job(eth_dev); + + /* Enable interrupt of all rx queues before enabling queues */ + hns3_dev_all_rx_queue_intr_enable(hw, true); + /* + * Enable state of each rxq and txq will be recovered after + * reset, so we need to restore them before enable all tqps; + */ + hns3_restore_tqp_enable_state(hw); + /* + * When finished the initialization, enable queues to receive + * and transmit packets. + */ + hns3_enable_all_queues(hw, true); + } + + return 0; +} + +static int +hns3vf_check_default_mac_change(struct hns3_hw *hw) +{ + char mac_str[RTE_ETHER_ADDR_FMT_SIZE]; + struct rte_ether_addr *hw_mac; + int ret; + + /* + * The hns3 PF ethdev driver in kernel support setting VF MAC address + * on the host by "ip link set ..." command. If the hns3 PF kernel + * ethdev driver sets the MAC address for VF device after the + * initialization of the related VF device, the PF driver will notify + * VF driver to reset VF device to make the new MAC address effective + * immediately. The hns3 VF PMD driver should check whether the MAC + * address has been changed by the PF kernel ethdev driver, if changed + * VF driver should configure hardware using the new MAC address in the + * recovering hardware configuration stage of the reset process. + */ + ret = hns3vf_get_host_mac_addr(hw); + if (ret) + return ret; + + hw_mac = (struct rte_ether_addr *)hw->mac.mac_addr; + ret = rte_is_zero_ether_addr(hw_mac); + if (ret) { + rte_ether_addr_copy(&hw->data->mac_addrs[0], hw_mac); + } else { + ret = rte_is_same_ether_addr(&hw->data->mac_addrs[0], hw_mac); + if (!ret) { + rte_ether_addr_copy(hw_mac, &hw->data->mac_addrs[0]); + hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, + &hw->data->mac_addrs[0]); + hns3_warn(hw, "Default MAC address has been changed to:" + " %s by the host PF kernel ethdev driver", + mac_str); + } + } - hns3vf_service_handler(eth_dev); return 0; } @@ -1650,6 +2708,10 @@ hns3vf_restore_conf(struct hns3_adapter *hns) struct hns3_hw *hw = &hns->hw; int ret; + ret = hns3vf_check_default_mac_change(hw); + if (ret) + return ret; + ret = hns3vf_configure_mac_addr(hns, false); if (ret) return ret; @@ -1658,10 +2720,26 @@ hns3vf_restore_conf(struct hns3_adapter *hns) if (ret) goto err_mc_mac; + ret = hns3vf_restore_promisc(hns); + if (ret) + goto err_vlan_table; + ret = hns3vf_restore_vlan_conf(hns); if (ret) goto err_vlan_table; + ret = hns3vf_get_port_base_vlan_filter_state(hw); + if (ret) + goto err_vlan_table; + + ret = hns3vf_restore_rx_interrupt(hw); + if (ret) + goto err_vlan_table; + + ret = hns3_restore_gro_conf(hw); + if (ret) + goto err_vlan_table; + if (hw->adapter_state == HNS3_NIC_STARTED) { ret = hns3vf_do_start(hns, false); if (ret) @@ -1669,6 +2747,13 @@ hns3vf_restore_conf(struct hns3_adapter *hns) hns3_info(hw, "hns3vf dev restart successful!"); } else if (hw->adapter_state == HNS3_NIC_STOPPING) hw->adapter_state = HNS3_NIC_CONFIGURED; + + ret = hns3vf_set_alive(hw, true); + if (ret) { + hns3_err(hw, "failed to VF send alive to PF: %d", ret); + goto err_vlan_table; + } + return 0; err_vlan_table: @@ -1719,8 +2804,10 @@ hns3vf_reset_service(void *param) * The interrupt may have been lost. It is necessary to handle * the interrupt to recover from the error. */ - if (rte_atomic16_read(&hns->hw.reset.schedule) == SCHEDULE_DEFERRED) { - rte_atomic16_set(&hns->hw.reset.schedule, SCHEDULE_REQUESTED); + if (__atomic_load_n(&hw->reset.schedule, __ATOMIC_RELAXED) == + SCHEDULE_DEFERRED) { + __atomic_store_n(&hw->reset.schedule, SCHEDULE_REQUESTED, + __ATOMIC_RELAXED); hns3_err(hw, "Handling interrupts in delayed tasks"); hns3vf_interrupt_handler(&rte_eth_devices[hw->data->port_id]); reset_level = hns3vf_get_reset_level(hw, &hw->reset.pending); @@ -1729,7 +2816,7 @@ hns3vf_reset_service(void *param) hns3_atomic_set_bit(HNS3_VF_RESET, &hw->reset.pending); } } - rte_atomic16_set(&hns->hw.reset.schedule, SCHEDULE_NONE); + __atomic_store_n(&hw->reset.schedule, SCHEDULE_NONE, __ATOMIC_RELAXED); /* * Hardware reset has been notified, we now have to poll & check if @@ -1737,14 +2824,13 @@ hns3vf_reset_service(void *param) */ reset_level = hns3vf_get_reset_level(hw, &hw->reset.pending); if (reset_level != HNS3_NONE_RESET) { - gettimeofday(&tv_start, NULL); + hns3_clock_gettime(&tv_start); hns3_reset_process(hns, reset_level); - gettimeofday(&tv, NULL); + hns3_clock_gettime(&tv); timersub(&tv, &tv_start, &tv_delta); - msec = tv_delta.tv_sec * MSEC_PER_SEC + - tv_delta.tv_usec / USEC_PER_MSEC; + msec = hns3_clock_calctime_ms(&tv_delta); if (msec > HNS3_RESET_PROCESS_MS) - hns3_err(hw, "%d handle long time delta %" PRIx64 + hns3_err(hw, "%d handle long time delta %" PRIu64 " ms time=%ld.%.6ld", hw->reset.level, msec, tv.tv_sec, tv.tv_usec); } @@ -1760,14 +2846,18 @@ hns3vf_reinit_dev(struct hns3_adapter *hns) if (hw->reset.level == HNS3_VF_FULL_RESET) { rte_intr_disable(&pci_dev->intr_handle); - hns3vf_set_bus_master(pci_dev, true); + ret = hns3vf_set_bus_master(pci_dev, true); + if (ret < 0) { + hns3_err(hw, "failed to set pci bus, ret = %d", ret); + return ret; + } } /* Firmware command initialize */ ret = hns3_cmd_init(hw); if (ret) { hns3_err(hw, "Failed to init cmd: %d", ret); - goto err_cmd_init; + return ret; } if (hw->reset.level == HNS3_VF_FULL_RESET) { @@ -1775,8 +2865,8 @@ hns3vf_reinit_dev(struct hns3_adapter *hns) * UIO enables msix by writing the pcie configuration space * vfio_pci enables msix in rte_intr_enable. */ - if (pci_dev->kdrv == RTE_KDRV_IGB_UIO || - pci_dev->kdrv == RTE_KDRV_UIO_GENERIC) { + if (pci_dev->kdrv == RTE_PCI_KDRV_IGB_UIO || + pci_dev->kdrv == RTE_PCI_KDRV_UIO_GENERIC) { if (hns3vf_enable_msix(pci_dev, true)) hns3_err(hw, "Failed to enable msix"); } @@ -1784,32 +2874,31 @@ hns3vf_reinit_dev(struct hns3_adapter *hns) rte_intr_enable(&pci_dev->intr_handle); } - ret = hns3_reset_all_queues(hns); + ret = hns3_reset_all_tqps(hns); if (ret) { hns3_err(hw, "Failed to reset all queues: %d", ret); - goto err_init; + return ret; } ret = hns3vf_init_hardware(hns); if (ret) { hns3_err(hw, "Failed to init hardware: %d", ret); - goto err_init; + return ret; } return 0; - -err_cmd_init: - hns3vf_set_bus_master(pci_dev, false); -err_init: - hns3_cmd_uninit(hw); - return ret; } static const struct eth_dev_ops hns3vf_eth_dev_ops = { + .dev_configure = hns3vf_dev_configure, .dev_start = hns3vf_dev_start, .dev_stop = hns3vf_dev_stop, .dev_close = hns3vf_dev_close, .mtu_set = hns3vf_dev_mtu_set, + .promiscuous_enable = hns3vf_dev_promiscuous_enable, + .promiscuous_disable = hns3vf_dev_promiscuous_disable, + .allmulticast_enable = hns3vf_dev_allmulticast_enable, + .allmulticast_disable = hns3vf_dev_allmulticast_disable, .stats_get = hns3_stats_get, .stats_reset = hns3_stats_reset, .xstats_get = hns3_dev_xstats_get, @@ -1818,13 +2907,21 @@ static const struct eth_dev_ops hns3vf_eth_dev_ops = { .xstats_get_by_id = hns3_dev_xstats_get_by_id, .xstats_get_names_by_id = hns3_dev_xstats_get_names_by_id, .dev_infos_get = hns3vf_dev_infos_get, + .fw_version_get = hns3vf_fw_version_get, .rx_queue_setup = hns3_rx_queue_setup, .tx_queue_setup = hns3_tx_queue_setup, .rx_queue_release = hns3_dev_rx_queue_release, .tx_queue_release = hns3_dev_tx_queue_release, + .rx_queue_start = hns3_dev_rx_queue_start, + .rx_queue_stop = hns3_dev_rx_queue_stop, + .tx_queue_start = hns3_dev_tx_queue_start, + .tx_queue_stop = hns3_dev_tx_queue_stop, .rx_queue_intr_enable = hns3_dev_rx_queue_intr_enable, .rx_queue_intr_disable = hns3_dev_rx_queue_intr_disable, - .dev_configure = hns3vf_dev_configure, + .rxq_info_get = hns3_rxq_info_get, + .txq_info_get = hns3_txq_info_get, + .rx_burst_mode_get = hns3_rx_burst_mode_get, + .tx_burst_mode_get = hns3_tx_burst_mode_get, .mac_addr_add = hns3vf_add_mac_addr, .mac_addr_remove = hns3vf_remove_mac_addr, .mac_addr_set = hns3vf_set_default_mac_addr, @@ -1834,11 +2931,12 @@ static const struct eth_dev_ops hns3vf_eth_dev_ops = { .rss_hash_conf_get = hns3_dev_rss_hash_conf_get, .reta_update = hns3_dev_rss_reta_update, .reta_query = hns3_dev_rss_reta_query, - .filter_ctrl = hns3_dev_filter_ctrl, + .flow_ops_get = hns3_dev_flow_ops_get, .vlan_filter_set = hns3vf_vlan_filter_set, .vlan_offload_set = hns3vf_vlan_offload_set, .get_reg = hns3_get_regs, .dev_supported_ptypes_get = hns3_dev_supported_ptypes_get, + .tx_done_cleanup = hns3_tx_done_cleanup, }; static const struct hns3_reset_ops hns3vf_reset_ops = { @@ -1860,31 +2958,35 @@ hns3vf_dev_init(struct rte_eth_dev *eth_dev) PMD_INIT_FUNC_TRACE(); - eth_dev->process_private = (struct hns3_process_private *) - rte_zmalloc_socket("hns3_filter_list", - sizeof(struct hns3_process_private), - RTE_CACHE_LINE_SIZE, eth_dev->device->numa_node); - if (eth_dev->process_private == NULL) { - PMD_INIT_LOG(ERR, "Failed to alloc memory for process private"); - return -ENOMEM; - } - - /* initialize flow filter lists */ - hns3_filterlist_init(eth_dev); + hns3_flow_init(eth_dev); hns3_set_rxtx_function(eth_dev); eth_dev->dev_ops = &hns3vf_eth_dev_ops; + eth_dev->rx_queue_count = hns3_rx_queue_count; if (rte_eal_process_type() != RTE_PROC_PRIMARY) { - hns3_mp_init_secondary(); + ret = hns3_mp_init_secondary(); + if (ret) { + PMD_INIT_LOG(ERR, "Failed to init for secondary " + "process, ret = %d", ret); + goto err_mp_init_secondary; + } hw->secondary_cnt++; + hns3_tx_push_init(eth_dev); return 0; } - hns3_mp_init_primary(); + ret = hns3_mp_init_primary(); + if (ret) { + PMD_INIT_LOG(ERR, + "Failed to init for primary process, ret = %d", + ret); + goto err_mp_init_primary; + } hw->adapter_state = HNS3_NIC_UNINITIALIZED; hns->is_vf = true; hw->data = eth_dev->data; + hns3_parse_devargs(eth_dev); ret = hns3_reset_init(hw); if (ret) @@ -1910,16 +3012,26 @@ hns3vf_dev_init(struct rte_eth_dev *eth_dev) goto err_rte_zmalloc; } + /* + * The hns3 PF ethdev driver in kernel support setting VF MAC address + * on the host by "ip link set ..." command. To avoid some incorrect + * scenes, for example, hns3 VF PMD driver fails to receive and send + * packets after user configure the MAC address by using the + * "ip link set ..." command, hns3 VF PMD driver keep the same MAC + * address strategy as the hns3 kernel ethdev driver in the + * initialization. If user configure a MAC address by the ip command + * for VF device, then hns3 VF PMD driver will start with it, otherwise + * start with a random MAC address in the initialization. + */ + if (rte_is_zero_ether_addr((struct rte_ether_addr *)hw->mac.mac_addr)) + rte_eth_random_addr(hw->mac.mac_addr); rte_ether_addr_copy((struct rte_ether_addr *)hw->mac.mac_addr, ð_dev->data->mac_addrs[0]); + hw->adapter_state = HNS3_NIC_INITIALIZED; - /* - * Pass the information to the rte_eth_dev_close() that it should also - * release the private port resources. - */ - eth_dev->data->dev_flags |= RTE_ETH_DEV_CLOSE_REMOVE; - if (rte_atomic16_read(&hns->hw.reset.schedule) == SCHEDULE_PENDING) { + if (__atomic_load_n(&hw->reset.schedule, __ATOMIC_RELAXED) == + SCHEDULE_PENDING) { hns3_err(hw, "Reschedule reset service after dev_init"); hns3_schedule_reset(hns); } else { @@ -1937,12 +3049,16 @@ err_init_vf: rte_free(hw->reset.wait_data); err_init_reset: + hns3_mp_uninit_primary(); + +err_mp_init_primary: +err_mp_init_secondary: eth_dev->dev_ops = NULL; eth_dev->rx_pkt_burst = NULL; + eth_dev->rx_descriptor_status = NULL; eth_dev->tx_pkt_burst = NULL; eth_dev->tx_pkt_prepare = NULL; - rte_free(eth_dev->process_private); - eth_dev->process_private = NULL; + eth_dev->tx_descriptor_status = NULL; return ret; } @@ -1956,12 +3072,7 @@ hns3vf_dev_uninit(struct rte_eth_dev *eth_dev) PMD_INIT_FUNC_TRACE(); if (rte_eal_process_type() != RTE_PROC_PRIMARY) - return -EPERM; - - eth_dev->dev_ops = NULL; - eth_dev->rx_pkt_burst = NULL; - eth_dev->tx_pkt_burst = NULL; - eth_dev->tx_pkt_prepare = NULL; + return 0; if (hw->adapter_state < HNS3_NIC_CLOSING) hns3vf_dev_close(eth_dev); @@ -1988,12 +3099,12 @@ eth_hns3vf_pci_remove(struct rte_pci_device *pci_dev) static const struct rte_pci_id pci_id_hns3vf_map[] = { { RTE_PCI_DEVICE(PCI_VENDOR_ID_HUAWEI, HNS3_DEV_ID_100G_VF) }, { RTE_PCI_DEVICE(PCI_VENDOR_ID_HUAWEI, HNS3_DEV_ID_100G_RDMA_PFC_VF) }, - { .vendor_id = 0, /* sentinel */ }, + { .vendor_id = 0, }, /* sentinel */ }; static struct rte_pci_driver rte_hns3vf_pmd = { .id_table = pci_id_hns3vf_map, - .drv_flags = RTE_PCI_DRV_NEED_MAPPING, + .drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC, .probe = eth_hns3vf_pci_probe, .remove = eth_hns3vf_pci_remove, }; @@ -2001,3 +3112,7 @@ static struct rte_pci_driver rte_hns3vf_pmd = { RTE_PMD_REGISTER_PCI(net_hns3_vf, rte_hns3vf_pmd); RTE_PMD_REGISTER_PCI_TABLE(net_hns3_vf, pci_id_hns3vf_map); RTE_PMD_REGISTER_KMOD_DEP(net_hns3_vf, "* igb_uio | vfio-pci"); +RTE_PMD_REGISTER_PARAM_STRING(net_hns3_vf, + HNS3_DEVARG_RX_FUNC_HINT "=vec|sve|simple|common " + HNS3_DEVARG_TX_FUNC_HINT "=vec|sve|simple|common " + HNS3_DEVARG_DEV_CAPS_MASK "=<1-65535> ");