I have a Dell R750 server with ESXI 7.0 installed, and use A30 passthrough to Guest OS Ubuntu 20.04. However, when I open the NVIDIA control panel, there is no display, and an error is reported in the log as shown in the figure below
Jul 8 10:48:17 natx-virtual-machine kernel:[903.827918]NVRM: This PCI I/O region assigned to your nVIDIA device is invalid:
Jul 8 10:48:17 natx-virtual-machine kernel: [903.827918]NVRM:BAR1 iS OM @ Ox0(PCI:0000:0b:00.0)
Jul 8 10:48:17 natx-virtual-machine kernel:[ 903.829185] nvidia: probe of 0000:0b:00.0 failed witherror-1
Jul 8 10:48:17 natx-virtual-machine kernel: [ 903.829196]NVRM: The NVIDIA pr obe routine failed for 1device(s).
Jul8 10:48:17 natx-virtual-machine kernel:[903.829197] NVRM:None of the N VIDIA devices were initialized.
Jul 8 10:48:17 natx-virtual-machine kernel:[903.829349] nvidia-nvlink: Unre gistered the Nvlink Core, major device number 235
Jul 8 10:48:17 natx-virtual-machine systemd-udevd[587]:nvidia: Process/sbin/modprobe nvidia-drm’failed with exit code 1.
Jul 8 10:48:17 natx-virtual-machine kernel:[ 903.930254] nvidia-nvlink: Nvli nk Core is being initialized, major device number 235
Jul 8 10:48:17 natx-virtual-machine kernel:[ 903.930261]NVRM: This PCI I/0 region assigned to your NVIDIA device is invalid:
Jul 8 10:48:17 natx-virtual-machine kernel:[ 903.930261]NVRM: BAR1 is OM @0x0(PCI:0000:0b:00.0)
Jul 8 10:48:17 natx-virtual-machine kernel:[ 903.931538] nvidia: probe of 0o
00:0b:00.0 failed with error-1
Jul 8 10:48:17 natx-virtual-machine kernel:[ 903.931548] NVRM: The NVIDIA pr
obe routine failed for 1device(s).
Jul 8 10:48:17 natx-virtual-machine kernel:[ 903.931549]NVRM:None of the N
VIDIA devices were initialized.
Jul 8 10:48:17 natx-virtual-machine kernel:[ 903.931717] nvidia-nvlink: Unre gistered the Nvlink Core, major device number 2351htvanaty uirtua1 marhino.ivariions.
I’ve set up the BIOS, test with 56T 12T 512G ,same .
Memory Mapped I/O above 4 GB enable
memory mapped i/o base 56T/12T/512GB
@generix hello, i did everything you’ve suggested in this thread and many more like this one, i wanna boot virtual ubuntu 20.04 in a host with 8 3090 with esxi, and nvidia-smi shows only 6.
output of lspci -vvv -d 10de: for unknown device:
1c:00.0 VGA compatible controller: NVIDIA Corporation GA102 [GeForce RTX 3090] (rev a1) (prog-if 00 [VGA controller])
DeviceName: pciPassthru6 Subsystem: Gigabyte Technology Co., Ltd Device 403b Physical Slot: 257
Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- SERR- <PERR- INTx-
Interrupt: pin A routed to IRQ 17
Region 0: Memory at f6000000 (32-bit, non-prefetchable) [size=16M]
Region 1: Memory at (64-bit, prefetchable)
Region 3: Memory at 1e000000 (64-bit, prefetchable) [size=32M] Region 5: I/O ports at b000 [size=128]
Capabilities: Kernel modules: nvidiafb, nouveau, nvidia_drm, nvidia
the working one have assied memory at Region 1
1b:00.0 VGA compatible controller: NVIDIA Corporation GA102 [GeForce RTX 3090] (rev a1) (prog-if 00 [VGA controller]) DeviceName: pciPassthru2 Subsystem: Gigabyte Technology Co., Ltd Device 403b Physical Slot: 256 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+ Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- SERR- <PERR- INTx- Latency: 64, Cache Line Size: 64 bytes Interrupt: pin A routed to IRQ 73 Region 0: Memory at fa000000 (32-bit, non-prefetchable) [size=16M] Region 1: Memory at 80000000 (64-bit, prefetchable) [size=256M] Region 3: Memory at 9e000000 (64-bit, prefetchable) [size=32M] Region 5: I/O ports at 7000 [size=128] Capabilities: Kernel driver in use: nvidia Kernel modules: nvidiafb, nouveau, nvidia_drm, nvidia
i have tried windows, everything is fine and 8 gpus working.
tried pci=realloc=off
this is dmesg | grep NVRM
[ 3.917540] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid:
NVRM: BAR1 is 0M @ 0x0 (PCI:0000:05:00.0)
[ 4.151071] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid:
NVRM: BAR1 is 0M @ 0x0 (PCI:0000:1c:00.0)
[ 4.151091] NVRM: The NVIDIA probe routine failed for 2 device(s).
[ 4.151096] NVRM: loading NVIDIA UNIX Open Kernel Module for x86_64 515.76
Release Build (dvs-builder@U16-T02-3-3) Mon Sep 12 19:20:55 UTC 2022
[ 8.626153] NVRM objClInitPcieChipset: *** Chipset Setup Function Error!
thanks for ur help
Please rather try pci=realloc
Please run nvidia-bug-report.sh as root and attach the resulting nvidia-bug-report.log.gz file to your post.