i40e AMD-Vi IO_PAGE_FAULT

i40e AMD-Vi IO_PAGE_FAULT

我们的一台服务器经常崩溃(Supermicro,Ubuntu 22.04),显然是在网络负载过大时。日志文件中充满了以下消息:

kernel: [514257.305733] i40e 0000:02:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0020 address=0x79ea8113f60 flags=0x0000]

这是网卡:

i40e: Intel(R) Ethernet Connection XL710 Network Driver
i40e: Copyright (c) 2013 - 2019 Intel Corporation.
i40e 0000:02:00.0: fw 8.5.67516 api 1.15 nvm 8.50 0x8000be1e 1.3295.0 [8086:15ff] [15d9:1c76]
i40e 0000:02:00.0: MAC address: 7c:c2:55:9d:d2:78
i40e 0000:02:00.0: FW LLDP is enabled
i40e 0000:02:00.0 eth0: NIC Link is Up, 10 Gbps Full Duplex, Flow Control: None
i40e 0000:02:00.0: PCI-Express: Speed 8.0GT/s Width x4
i40e 0000:02:00.0: PCI-Express bandwidth available for this device may be insufficient for optimal performance.
i40e 0000:02:00.0: Please move the device to a different PCI-e link with more lanes and/or higher transfer rate.
i40e 0000:02:00.0: Features: PF-id[0] VFs: 64 VSIs: 66 QP: 119 RSS FD_ATR FD_SB NTUPLE DCB VxLAN Geneve PTP VEPA
i40e 0000:02:00.1: fw 8.5.67516 api 1.15 nvm 8.50 0x8000be1e 1.3295.0 [8086:15ff] [15d9:1c76]
i40e 0000:02:00.1: MAC address: 7c:c2:55:9d:d2:79
i40e 0000:02:00.1: FW LLDP is enabled
i40e 0000:02:00.1: PCI-Express: Speed 8.0GT/s Width x4
i40e 0000:02:00.1: PCI-Express bandwidth available for this device may be insufficient for optimal performance.
i40e 0000:02:00.1: Please move the device to a different PCI-e link with more lanes and/or higher transfer rate.
i40e 0000:02:00.1: Features: PF-id[1] VFs: 64 VSIs: 66 QP: 119 RSS FD_ATR FD_SB NTUPLE DCB VxLAN Geneve PTP VEPA
i40e 0000:02:00.0 enp2s0f0: renamed from eth0
i40e 0000:02:00.1 enp2s0f1: renamed from eth1
i40e 0000:02:00.0: entering allmulti mode.

驱动程序(“modinfo i40e”):

filename: /lib/modules/5.15.0-87-generic/kernel/drivers/net/ethernet/intel/i40e/i40e.ko
license: GPL v2 description: Intel(R) Ethernet Connection XL710 Network Driver
author: Intel Corporation, <[email protected]>
srcversion: 263AAE07A824C74401C1729 alias: pci:v00008086d0000158Bsv*sd*bc*sc*i*
alias: pci:v00008086d0000158Asv*sd*bc*sc*i*

有人知道吗?在与 IO_PAGE_FAULT 类似的情况下,有时建议使用 GRUB_CMDLINE_LINUX_DEFAULT="iommu=soft"。也许我应该将速度从 10 Gbit/s 降低到 1 Gbit/s 作为测试?

谢谢

斯蒂芬

相关内容