<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <meta name="Generator" content="Microsoft Word 15 (filtered medium)"> <style></style> </head> <body lang="EN-US" link="blue" vlink="purple" style="word-wrap:break-word"> <div class="WordSection1"> Hi, Ben<o:p></o:p> This patch can only take effect after firmware v8.6. It cannot be used on firmware v8.4 and v8.5. The reason is that the firmware team gave a clear reply that there are many related bugs in firmware v8.4 and v8.5, and they were fixed in firmware v8.6. The firmware team recommends using v8.6.<o:p></o:p> <o:p> </o:p> Regards<o:p></o:p> Kevin<o:p></o:p> <o:p> </o:p> <div style="border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt"> <div> <div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0cm 0cm 0cm"> From: Ben Magistro <koncept1@gmail.com> Sent: 2022年6月10日 22:27 To: Liu, KevinX <kevinx.liu@intel.com> Cc: dev@dpdk.org; Zhang, Yuying <yuying.zhang@intel.com>; Xing, Beilei <beilei.xing@intel.com>; Yang, SteveX <stevex.yang@intel.com>; Zhang, RobinX <robinx.zhang@intel.com> Subject: Re: [PATCH v7] net/i40e: add outer VLAN processing<o:p></o:p> </div> </div> <o:p> </o:p> <div> I'm trying to understand if this change is at all related to an issue we are experiencing with newer firmwares (<a href="https://mails.dpdk.org/archives/dev/2022-April/238621.html" target="_blank">https://mails.dpdk.org/archives/dev/2022-April/238621.html</a>) that happened to start with 8.4 and affected qinq offload processing. I can say loading this patch and running testpmd does not seem to have any effect on the issue we are experiencing.<o:p></o:p> <div> <o:p> </o:p> </div> <div> Side note, there is also a minor typo in the comment block "i40e_vlan_tpie_set" vs "i40e_vlan_tpid_set".<o:p></o:p> </div> <div> <o:p> </o:p> </div> <div> Thanks<o:p></o:p> </div> </div> <o:p> </o:p> <div> <div> On Fri, Jun 10, 2022 at 4:30 AM Kevin Liu <<a href="mailto:kevinx.liu@intel.com" target="_blank">kevinx.liu@intel.com</a>> wrote:<o:p></o:p> </div> <blockquote style="border:none;border-left:solid #CCCCCC 1.0pt;padding:0cm 0cm 0cm 6.0pt;margin-left:4.8pt;margin-right:0cm"> From: Robin Zhang <<a href="mailto:robinx.zhang@intel.com" target="_blank">robinx.zhang@intel.com</a>> Outer VLAN processing is supported after firmware v8.4, kernel driver also change the default behavior to support this feature. To align with kernel driver, add support for outer VLAN processing in DPDK. But it is forbidden for firmware to change the Inner/Outer VLAN configuration while there are MAC/VLAN filters in the switch table. Therefore, we need to clear the MAC table before setting config, and then restore the MAC table after setting. This will not impact on an old firmware. Signed-off-by: Robin Zhang <<a href="mailto:robinx.zhang@intel.com" target="_blank">robinx.zhang@intel.com</a>> Signed-off-by: Kevin Liu <<a href="mailto:kevinx.liu@intel.com" target="_blank">kevinx.liu@intel.com</a>> --- drivers/net/i40e/i40e_ethdev.c | 94 ++++++++++++++++++++++++++++++++-- drivers/net/i40e/i40e_ethdev.h | 3 ++ 2 files changed, 92 insertions(+), 5 deletions(-) diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c index 755786dc10..4cae163cb9 100644 --- a/drivers/net/i40e/i40e_ethdev.c +++ b/drivers/net/i40e/i40e_ethdev.c @@ -2575,6 +2575,7 @@ i40e_dev_close(struct rte_eth_dev *dev) struct i40e_hw *hw = I40E_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; + struct rte_eth_rxmode *rxmode = &dev->data->dev_conf.rxmode; struct i40e_filter_control_settings settings; struct rte_flow *p_flow; uint32_t reg; @@ -2587,6 +2588,18 @@ i40e_dev_close(struct rte_eth_dev *dev) if (rte_eal_process_type() != RTE_PROC_PRIMARY) return 0; + /* + * It is a workaround, if the double VLAN is disabled when + * the program exits, an abnormal error will occur on the + * NIC. Need to enable double VLAN when dev is closed. + */ + if (pf->fw8_3gt) { + if (!(rxmode->offloads & RTE_ETH_RX_OFFLOAD_VLAN_EXTEND)) { + rxmode->offloads |= RTE_ETH_RX_OFFLOAD_VLAN_EXTEND; + i40e_vlan_offload_set(dev, RTE_ETH_VLAN_EXTEND_MASK); + } + } + ret = rte_eth_switch_domain_free(pf->switch_domain_id); if (ret) PMD_INIT_LOG(WARNING, "failed to free switch domain: %d", ret); @@ -3909,6 +3922,7 @@ i40e_vlan_tpid_set(struct rte_eth_dev *dev, struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private); int qinq = dev->data->dev_conf.rxmode.offloads & RTE_ETH_RX_OFFLOAD_VLAN_EXTEND; + u16 sw_flags = 0, valid_flags = 0; int ret = 0; if ((vlan_type != RTE_ETH_VLAN_TYPE_INNER && @@ -3927,15 +3941,32 @@ i40e_vlan_tpid_set(struct rte_eth_dev *dev, /* 802.1ad frames ability is added in NVM API 1.7*/ if (hw->flags & I40E_HW_FLAG_802_1AD_CAPABLE) { if (qinq) { + if (pf->fw8_3gt) { + sw_flags = I40E_AQ_SET_SWITCH_CFG_OUTER_VLAN; + valid_flags = I40E_AQ_SET_SWITCH_CFG_OUTER_VLAN; + } if (vlan_type == RTE_ETH_VLAN_TYPE_OUTER) hw->first_tag = rte_cpu_to_le_16(tpid); else if (vlan_type == RTE_ETH_VLAN_TYPE_INNER) hw->second_tag = rte_cpu_to_le_16(tpid); } else { - if (vlan_type == RTE_ETH_VLAN_TYPE_OUTER) - hw->second_tag = rte_cpu_to_le_16(tpid); + /* + * If tpid is equal to 0x88A8, indicates that the + * disable double VLAN operation is in progress. + * Need set switch configuration back to default. + */ + if (pf->fw8_3gt && tpid == RTE_ETHER_TYPE_QINQ) { + sw_flags = 0; + valid_flags = I40E_AQ_SET_SWITCH_CFG_OUTER_VLAN; + if (vlan_type == RTE_ETH_VLAN_TYPE_OUTER) + hw->first_tag = rte_cpu_to_le_16(tpid); + } else { + if (vlan_type == RTE_ETH_VLAN_TYPE_OUTER) + hw->second_tag = rte_cpu_to_le_16(tpid); + } } - ret = i40e_aq_set_switch_config(hw, 0, 0, 0, NULL); + ret = i40e_aq_set_switch_config(hw, sw_flags, + valid_flags, 0, NULL); if (ret != I40E_SUCCESS) { PMD_DRV_LOG(ERR, "Set switch config failed aq_err: %d", @@ -3987,8 +4018,13 @@ static int i40e_vlan_offload_set(struct rte_eth_dev *dev, int mask) { struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private); + struct i40e_mac_filter_info *mac_filter; struct i40e_vsi *vsi = pf->main_vsi; struct rte_eth_rxmode *rxmode; + struct i40e_mac_filter *f; + int i, num; + void *temp; + int ret; rxmode = &dev->data->dev_conf.rxmode; if (mask & RTE_ETH_VLAN_FILTER_MASK) { @@ -4007,6 +4043,33 @@ i40e_vlan_offload_set(struct rte_eth_dev *dev, int mask) } if (mask & RTE_ETH_VLAN_EXTEND_MASK) { + i = 0; + num = vsi->mac_num; + mac_filter = rte_zmalloc("mac_filter_info_data", + num * sizeof(*mac_filter), 0); + if (mac_filter == NULL) { + PMD_DRV_LOG(ERR, "failed to allocate memory"); + return I40E_ERR_NO_MEMORY; + } + + /* + * Outer VLAN processing is supported after firmware v8.4, kernel driver + * also change the default behavior to support this feature. To align with + * kernel driver, set switch config in 'i40e_vlan_tpie_set' to support for + * outer VLAN processing. But it is forbidden for firmware to change the + * Inner/Outer VLAN configuration while there are MAC/VLAN filters in the + * switch table. Therefore, we need to clear the MAC table before setting + * config, and then restore the MAC table after setting. This feature is + * recommended to be used in firmware v8.6. + */ + /* Remove all existing mac */ + RTE_TAILQ_FOREACH_SAFE(f, &vsi->mac_list, next, temp) { + mac_filter[i] = f->mac_info; + ret = i40e_vsi_delete_mac(vsi, &f->mac_info.mac_addr); + if (ret) + PMD_DRV_LOG(ERR, "i40e vsi delete mac fail."); + i++; + } if (rxmode->offloads & RTE_ETH_RX_OFFLOAD_VLAN_EXTEND) { i40e_vsi_config_double_vlan(vsi, TRUE); /* Set global registers with default ethertype. */ @@ -4014,9 +4077,19 @@ i40e_vlan_offload_set(struct rte_eth_dev *dev, int mask) RTE_ETHER_TYPE_VLAN); i40e_vlan_tpid_set(dev, RTE_ETH_VLAN_TYPE_INNER, RTE_ETHER_TYPE_VLAN); - } - else + } else { + if (pf->fw8_3gt) + i40e_vlan_tpid_set(dev, RTE_ETH_VLAN_TYPE_OUTER, + RTE_ETHER_TYPE_QINQ); i40e_vsi_config_double_vlan(vsi, FALSE); + } + /* Restore all mac */ + for (i = 0; i < num; i++) { + ret = i40e_vsi_add_mac(vsi, &mac_filter[i]); + if (ret) + PMD_DRV_LOG(ERR, "i40e vsi add mac fail."); + } + rte_free(mac_filter); } if (mask & RTE_ETH_QINQ_STRIP_MASK) { @@ -4846,6 +4919,17 @@ i40e_pf_parameter_init(struct rte_eth_dev *dev) return -EINVAL; } + /** + * Enable outer VLAN processing if firmware version is greater + * than v8.3 + */ + if (hw->aq.fw_maj_ver > 8 || + (hw->aq.fw_maj_ver == 8 && hw->aq.fw_min_ver > 3)) { + pf->fw8_3gt = true; + } else { + pf->fw8_3gt = false; + } + return 0; } diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h index a1ebdc093c..fe943a45ff 100644 --- a/drivers/net/i40e/i40e_ethdev.h +++ b/drivers/net/i40e/i40e_ethdev.h @@ -1188,6 +1188,9 @@ struct i40e_pf { /* Switch Domain Id */ uint16_t switch_domain_id; + /* When firmware > 8.3, the enable flag for outer VLAN processing */ + bool fw8_3gt; + struct i40e_vf_msg_cfg vf_msg_cfg; uint64_t prev_rx_bytes; uint64_t prev_tx_bytes; -- 2.34.1<o:p></o:p> </blockquote> </div> </div> </div> </body> </html>