zhtengw Posted April 7, 2023 Author Share Posted April 7, 2023 Hi giganode, [ 43.663221] i915 0000:00:02.1: GuC interface version 0.1.0.0 [ 43.663869] i915 0000:00:02.1: IOV: Failed to get runtime info (-EPROTO) [ 43.669365] i915 0000:00:02.1: Device initialization failed (-71) [ 43.669368] i915: probe of 0000:00:02.1 failed with error -71 According these info from your dmesg, I can locate the error in running this function. It happens when not enough MMIO resources for SR-IOV. On 4/4/2023 at 4:19 PM, giganode said: Everything is set up.. except I can't find the SR-IOV Enable Setting. Hardware used is listed in my signature. The board is from ERYING - the one with the directly soldered 11800h - from AliExpress. Can you add "pci=realloc,assign-busses" to your boot kernel command line to have a try? By the way, I don't know how to see your signature to get your hardware info. Link to comment
giganode Posted April 7, 2023 Share Posted April 7, 2023 10 hours ago, zhtengw said: Can you add "pci=realloc,assign-busses" to your boot kernel command line to have a try? I tried. It's still exactly the same output. 10 hours ago, zhtengw said: By the way, I don't know how to see your signature to get your hardware info. Oh okay. Model:RM42-502 with Intel Core i7-11800H and 24TB with Double Parity M/B:INTEL HM570 Version Default string s/n Default string BIOS:American Megatrends International, LLC. Version THM570106 Dated 06/14/2022 CPU:11th Gen Intel® Core™ i7-11800H @ 2.30GHz HVM:Enabled IOMMU:Enabled Cache:L1 Cache: 384 KiB, L1 Cache: 256 KiB, L2 Cache: 10 MiB, L3 Cache: 24 MiB Memory:16 GiB DDR4 (max. installable capacity 128 GiB) Network:bond0: fault-tolerance (active-backup), mtu 1500 Kernel:Linux 6.1.21-Unraid x86_64 Link to comment
zhtengw Posted April 8, 2023 Author Share Posted April 8, 2023 11 hours ago, giganode said: I tried. It's still exactly the same output. Oh okay. Model:RM42-502 with Intel Core i7-11800H and 24TB with Double Parity M/B:INTEL HM570 Version Default string s/n Default string BIOS:American Megatrends International, LLC. Version THM570106 Dated 06/14/2022 CPU:11th Gen Intel® Core™ i7-11800H @ 2.30GHz HVM:Enabled IOMMU:Enabled Cache:L1 Cache: 384 KiB, L1 Cache: 256 KiB, L2 Cache: 10 MiB, L3 Cache: 24 MiB Memory:16 GiB DDR4 (max. installable capacity 128 GiB) Network:bond0: fault-tolerance (active-backup), mtu 1500 Kernel:Linux 6.1.21-Unraid x86_64 I borrowed a laptop with i5-11300H and intel HM570 chipset which in the same series as your platform. And I got the same "code 43" error in Windows Guest, even through I installed the latest intel graphics driver. It looks like a adaptation issue of the intel driver for the mobile platform. Link to comment
Daigurren69 Posted April 9, 2023 Share Posted April 9, 2023 Everytime I stop the VM the SRIOV disappears and all my VGA devices disappear from System Devices including /dev/dri.....Any idea why? Link to comment
zixuan Posted April 10, 2023 Share Posted April 10, 2023 Thanks for the contribution of the author to the SR-IOV in unRaid. I started to use this plugin from the first verison till now. It's good for me to passthrough the gpu UHD 730 (i3-12100) to a windows 10 guest. However, after I updated the latest version (2023.04.06), troubles came to me. Simply, after I start the win10 guest with SR-IOV gpu, the host crashed within 10 min and can never restart untill I re-pluged the RAM. Now I have to uninstall the plugin and look forward to the next version. Thanks anyway for this plug! Link to comment
zhtengw Posted April 10, 2023 Author Share Posted April 10, 2023 11 minutes ago, zixuan said: Thanks for the contribution of the author to the SR-IOV in unRaid. I started to use this plugin from the first verison till now. It's good for me to passthrough the gpu UHD 730 (i3-12100) to a windows 10 guest. However, after I updated the latest version (2023.04.06), troubles came to me. Simply, after I start the win10 guest with SR-IOV gpu, the host crashed within 10 min and can never restart untill I re-pluged the RAM. Now I have to uninstall the plugin and look forward to the next version. Thanks anyway for this plug! Thanks for your feedback. I am using i3-12100 too and do not meet this issue. And I can not figure out what happen to you now. The 2023.04.06 updates only the setting page. If it is convenient to you, would you please install the version 2023.04.01 with the URL below to have a test? https://raw.githubusercontent.com/zhtengw/unraid-i915-sriov/525f2d329c6a7bea471964cf8f9364b2b8980ac5/i915-sriov.plg Link to comment
Cheater Posted April 10, 2023 Share Posted April 10, 2023 When I use SR-IOV the Intel XE graphics card is correctly detected. I'm using Teamviewer to view the screen, but the resolution is just 640x480. I assume this is because no physical screen is attached to adapt the resolution to. Any ideas how to resolve this? Link to comment
yxlqqufo Posted April 12, 2023 Share Posted April 12, 2023 我是6.11.5版本,配置是13600+z690,虚拟机使用虚拟显卡,但是显示代码43,如附件,这是我的信息,麻烦看下是哪里配置错了?除了安装插件意外还需要在哪里设置吗? Link to comment
zhtengw Posted April 13, 2023 Author Share Posted April 13, 2023 7 hours ago, yxlqqufo said: 我是6.11.5版本,配置是13600+z690,虚拟机使用虚拟显卡,但是显示代码43,如附件,这是我的信息,麻烦看下是哪里配置错了?除了安装插件意外还需要在哪里设置吗? 你好,您的虚拟机类型是不是q35,改成i440fx看看行不行 Link to comment
BHWang Posted April 13, 2023 Share Posted April 13, 2023 8 hours ago, yxlqqufo said: 我是6.11.5版本,配置是13600+z690,虚拟机使用虚拟显卡,但是显示代码43,如附件,这是我的信息,麻烦看下是哪里配置错了?除了安装插件意外还需要在哪里设置吗? Windows內的 intel顯示驅動裝了? Link to comment
balvras Posted April 15, 2023 Share Posted April 15, 2023 Hi all, I just wanted to let you know that the latest Unraid version 6.12.0-rc3 breaks the installation of the plugin. plugin: installing: i915-sriov.plg Executing hook script: pre_plugin_checks plugin: downloading: i915-sriov.plg ... done plugin: downloading: unraid-i915-sriov-2023.04.06.txz ... done +============================================================================== | Skipping package unraid-i915-sriov-2023.04.06 (already installed) +============================================================================== patching file usr/local/emhttp/plugins/dynamix.vm.manager/include/libvirt.php Hunk #1 succeeded at 809 (offset 29 lines). +============================================================================== | WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING | | Don't close this window with the red 'X' in the top right corner | until the 'DONE' button is displayed! | | WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING +============================================================================== ----------Downloading i915-sriov module Package for kernel v6.1.23---------- ---------This could take some time, please don't close this window!----------- --------Can't download i915-sriov module Package for kernel v6.1.23----------- plugin: run failed: '/bin/bash' returned 1 Executing hook script: post_plugin_checks Link to comment
zhtengw Posted April 15, 2023 Author Share Posted April 15, 2023 40 minutes ago, balvras said: Hi all, I just wanted to let you know that the latest Unraid version 6.12.0-rc3 breaks the installation of the plugin. plugin: installing: i915-sriov.plg Executing hook script: pre_plugin_checks plugin: downloading: i915-sriov.plg ... done plugin: downloading: unraid-i915-sriov-2023.04.06.txz ... done +============================================================================== | Skipping package unraid-i915-sriov-2023.04.06 (already installed) +============================================================================== patching file usr/local/emhttp/plugins/dynamix.vm.manager/include/libvirt.php Hunk #1 succeeded at 809 (offset 29 lines). +============================================================================== | WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING | | Don't close this window with the red 'X' in the top right corner | until the 'DONE' button is displayed! | | WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING +============================================================================== ----------Downloading i915-sriov module Package for kernel v6.1.23---------- ---------This could take some time, please don't close this window!----------- --------Can't download i915-sriov module Package for kernel v6.1.23----------- plugin: run failed: '/bin/bash' returned 1 Executing hook script: post_plugin_checks Fixed may try again now. 1 Link to comment
Anoyou Posted April 15, 2023 Share Posted April 15, 2023 中文大佬你好,请问这个插件的直通方案是不是无法输出?我尝试用显卡诱骗器,但还是除了mstsc远程外,其他远程方法都是640x480,除此之外没有其他问题。谢谢 unraid 6.12 rc2 Link to comment
zhtengw Posted April 15, 2023 Author Share Posted April 15, 2023 1 hour ago, Anoyou said: 中文大佬你好,请问这个插件的直通方案是不是无法输出?我尝试用显卡诱骗器,但还是除了mstsc远程外,其他远程方法都是640x480,除此之外没有其他问题。谢谢 unraid 6.12 rc2 你好,SR-IOV的虚拟显卡目前是没法显示输出的,用Windows自带的RDP服务可以保证不错的远程桌面体验,用mstsc或者其他支持rdp协议的远程桌面客户端都差不多的。VNC、向日葵这些体验就差一些了。 Link to comment
giganode Posted April 16, 2023 Share Posted April 16, 2023 On 4/8/2023 at 3:42 AM, zhtengw said: I borrowed a laptop with i5-11300H and intel HM570 chipset which in the same series as your platform. And I got the same "code 43" error in Windows Guest, even through I installed the latest intel graphics driver. It looks like a adaptation issue of the intel driver for the mobile platform. Oh okay.. sad to hear. I guess because its an mobile chipset, this will not change in the future. Link to comment
Seq Posted April 17, 2023 Share Posted April 17, 2023 (edited) Intel GPU TOP Does not work. It's just blank: with the only choice being: Edited April 17, 2023 by Seq Link to comment
SimonF Posted April 17, 2023 Share Posted April 17, 2023 1 hour ago, Seq said: Intel GPU TOP Does not work. It's just blank: with the only choice being: This is gpu stat you should only use the Physical address not the logical ones but will only work correctly on 6.12 Link to comment
maxAtmin Posted April 21, 2023 Share Posted April 21, 2023 12400 + Z690 unraid 6.11.5 Tested the 1st version in plugin store. After updated to 2023.04.06, windows guest gpu got error code 43. Tried rollback to 2023.04.01, it's still code 43. Link to comment
maxAtmin Posted April 21, 2023 Share Posted April 21, 2023 15 minutes ago, maxAtmin said: 12400 + Z690 unraid 6.11.5 Tested the 1st version in plugin store. After updated to 2023.04.06, windows guest gpu got error code 43. Tried rollback to 2023.04.01, it's still code 43. Update: Reinstall 2023.04.06, then install latest intel GPU driver in Windows 10 guest. The installation failed. But the SR-IOV gpu works after reboot. Link to comment
joykingdom Posted April 22, 2023 Share Posted April 22, 2023 6.12 RC3 After reboot VM Autostart disabled due to vfio-bind error Link to comment
SimonF Posted April 22, 2023 Share Posted April 22, 2023 4 hours ago, joykingdom said: 6.12 RC3 After reboot VM Autostart disabled due to vfio-bind error You need to look at the VFIO log to see which devices didn't bind. This is to safe guard against hardware changes that may crash host if VM starts without correct hardware passthru. Link to comment
joykingdom Posted April 22, 2023 Share Posted April 22, 2023 1 hour ago, SimonF said: You need to look at the VFIO log to see which devices didn't bind. This is to safe guard against hardware changes that may crash host if VM starts without correct hardware passthru. VFIO log Loading config from /boot/config/vfio-pci.cfg BIND=0000:05:00.0|8086:1528 0000:05:00.1|8086:1528 0000:08:00.0|8086:125c 0000:09:00.0|8086:125c 0000:00:02.1|8086:4692 0000:00:02.2|8086:4692--- Processing 0000:05:00.0 8086:1528 Vendor:Device 8086:1528 found at 0000:05:00.0 IOMMU group members (sans bridges): /sys/bus/pci/devices/0000:05:00.0/iommu_group/devices/0000:05:00.0 Binding... success... Device 8086:1528 at 0000:05:00.0 bound to vfio-pci --- Processing 0000:05:00.1 8086:1528 Vendor:Device 8086:1528 found at 0000:05:00.1 IOMMU group members (sans bridges): /sys/bus/pci/devices/0000:05:00.1/iommu_group/devices/0000:05:00.1 Binding... success... Device 8086:1528 at 0000:05:00.1 bound to vfio-pci --- Processing 0000:08:00.0 8086:125c Vendor:Device 8086:125c found at 0000:08:00.0 IOMMU group members (sans bridges): /sys/bus/pci/devices/0000:08:00.0/iommu_group/devices/0000:08:00.0 Binding... success... Device 8086:125c at 0000:08:00.0 bound to vfio-pci --- Processing 0000:09:00.0 8086:125c Vendor:Device 8086:125c found at 0000:09:00.0 IOMMU group members (sans bridges): /sys/bus/pci/devices/0000:09:00.0/iommu_group/devices/0000:09:00.0 Binding... success... Device 8086:125c at 0000:09:00.0 bound to vfio-pci --- Processing 0000:00:02.1 8086:4692 --- Processing 0000:00:02.2 8086:4692 --- Devices listed in /sys/bus/pci/drivers/vfio-pci: lrwxrwxrwx 1 root root 0 Apr 22 09:29 0000:05:00.0 -> ../../../../devices/pci0000:00/0000:00:1c.4/0000:05:00.0 lrwxrwxrwx 1 root root 0 Apr 22 09:29 0000:05:00.1 -> ../../../../devices/pci0000:00/0000:00:1c.4/0000:05:00.1 lrwxrwxrwx 1 root root 0 Apr 22 09:29 0000:08:00.0 -> ../../../../devices/pci0000:00/0000:00:1d.0/0000:06:00.0/0000:07:03.0/0000:08:00.0 lrwxrwxrwx 1 root root 0 Apr 22 09:29 0000:09:00.0 -> ../../../../devices/pci0000:00/0000:00:1d.0/0000:06:00.0/0000:07:07.0/0000:09:00.0 vfio-pci binding complete tower-diagnostics-20230422-1733.zip Link to comment
SimonF Posted April 22, 2023 Share Posted April 22, 2023 1 hour ago, joykingdom said: VFIO log Loading config from /boot/config/vfio-pci.cfg BIND=0000:05:00.0|8086:1528 0000:05:00.1|8086:1528 0000:08:00.0|8086:125c 0000:09:00.0|8086:125c 0000:00:02.1|8086:4692 0000:00:02.2|8086:4692--- Processing 0000:05:00.0 8086:1528 Vendor:Device 8086:1528 found at 0000:05:00.0 IOMMU group members (sans bridges): /sys/bus/pci/devices/0000:05:00.0/iommu_group/devices/0000:05:00.0 Binding... success... Device 8086:1528 at 0000:05:00.0 bound to vfio-pci --- Processing 0000:05:00.1 8086:1528 Vendor:Device 8086:1528 found at 0000:05:00.1 IOMMU group members (sans bridges): /sys/bus/pci/devices/0000:05:00.1/iommu_group/devices/0000:05:00.1 Binding... success... Device 8086:1528 at 0000:05:00.1 bound to vfio-pci --- Processing 0000:08:00.0 8086:125c Vendor:Device 8086:125c found at 0000:08:00.0 IOMMU group members (sans bridges): /sys/bus/pci/devices/0000:08:00.0/iommu_group/devices/0000:08:00.0 Binding... success... Device 8086:125c at 0000:08:00.0 bound to vfio-pci --- Processing 0000:09:00.0 8086:125c Vendor:Device 8086:125c found at 0000:09:00.0 IOMMU group members (sans bridges): /sys/bus/pci/devices/0000:09:00.0/iommu_group/devices/0000:09:00.0 Binding... success... Device 8086:125c at 0000:09:00.0 bound to vfio-pci --- Processing 0000:00:02.1 8086:4692 --- Processing 0000:00:02.2 8086:4692 --- Devices listed in /sys/bus/pci/drivers/vfio-pci: lrwxrwxrwx 1 root root 0 Apr 22 09:29 0000:05:00.0 -> ../../../../devices/pci0000:00/0000:00:1c.4/0000:05:00.0 lrwxrwxrwx 1 root root 0 Apr 22 09:29 0000:05:00.1 -> ../../../../devices/pci0000:00/0000:00:1c.4/0000:05:00.1 lrwxrwxrwx 1 root root 0 Apr 22 09:29 0000:08:00.0 -> ../../../../devices/pci0000:00/0000:00:1d.0/0000:06:00.0/0000:07:03.0/0000:08:00.0 lrwxrwxrwx 1 root root 0 Apr 22 09:29 0000:09:00.0 -> ../../../../devices/pci0000:00/0000:00:1d.0/0000:06:00.0/0000:07:07.0/0000:09:00.0 vfio-pci binding complete tower-diagnostics-20230422-1733.zip 169.92 kB · 0 downloads Issue is VFIO check is completed before plugin is loaded so 2.1 and 2.2 don't exist. You don't need to bind them to VFIO. For now suggest remove bindings from system devices. Link to comment
joykingdom Posted April 22, 2023 Share Posted April 22, 2023 9 minutes ago, SimonF said: Issue is VFIO check is completed before plugin is loaded so 2.1 and 2.2 don't exist. You don't need to bind them to VFIO. For now suggest remove bindings from system devices. Before 6.12 RC3 is ok So I can not use IGPU SR-IOV again? Link to comment
SimonF Posted April 22, 2023 Share Posted April 22, 2023 26 minutes ago, joykingdom said: Before 6.12 RC3 is ok So I can not use IGPU SR-IOV again? Yes you can use SR-IOV but you dont need to bind the VFs to VFIO. Check for Hardware changes was added into rc3. Passthru of the vgpu will still work. Sorry ignore my previous statements, it does look like the plugin is loaded before vfio is process. I will review the logs. But removing binds should allow Auto start to work again. You don't need to bind gpus, but is recommended. Link to comment
Recommended Posts