Unable to get Arc Pro B50 to function

T

Tightrope2718

New Member
May 10, 2026
2
0
1
Hello,

I have been struggling with this issue for awhile, but I am unable to proceed further. I cannot seem to find anyone else who is having a similar issue to me.

In my server I have an Arc A380 that I wanted to upgrade to an Arc Pro B50. My Arc A380 works perfectly. I have it passed through to a VM without any issues. My configuration files are listed below.

/etc/modules
Code: 
vfio
vfio_iommu_type1
vfio_pci
vfio_virqfd

/etc/modprobe.d/pve-blacklist.conf
Code: 
blacklist nouveau
blacklist nvidiafb
blacklist radeon
blacklist i915
blacklist xe
blacklist snd_hda_intel

/etc/modprobe.d/vfio.conf
Code: 
options vfio-pci ids=0e:00.0,0f:00.0

/etc/default/grub
Code: 
GRUB_DEFAULT=0
GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR=`( . /etc/os-release && echo ${NAME} )`
GRUB_CMDLINE_LINUX_DEFAULT="quiet amd_iommu=on iommu=pt"
GRUB_CMDLINE_LINUX=""

lspci -vvv
Code: 
0e:00.0 VGA compatible controller: Intel Corporation DG2 [Arc A380] (rev 05) (prog-if 00 [VGA controller])
        Subsystem: Device 172f:3943
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 64 bytes
        Interrupt: pin ? routed to IRQ 289
        IOMMU group: 33
        Region 0: Memory at fa000000 (64-bit, non-prefetchable) [size=16M]
        Region 2: Memory at 7c00000000 (64-bit, prefetchable) [size=8G]
        Expansion ROM at fb000000 [disabled] [size=2M]
        Capabilities: [40] Vendor Specific Information: Len=0c <?>
        Capabilities: [70] Express (v2) Endpoint, IntMsgNum 0
                DevCap: MaxPayload 128 bytes, PhantFunc 0, Latency L0s <64ns, L1 <1us
                        ExtTag+ AttnBtn- AttnInd- PwrInd- RBE+ FLReset+ SlotPowerLimit 0W TEE-IO-
                DevCtl: CorrErr- NonFatalErr- FatalErr- UnsupReq-
                        RlxdOrd+ ExtTag+ PhantFunc- AuxPwr- NoSnoop+ FLReset-
                        MaxPayload 128 bytes, MaxReadReq 128 bytes
                DevSta: CorrErr- NonFatalErr- FatalErr- UnsupReq- AuxPwr- TransPend-
                LnkCap: Port #0, Speed 2.5GT/s, Width x1, ASPM L0s L1, Exit Latency L0s <64ns, L1 <1us
                        ClockPM- Surprise- LLActRep- BwNot- ASPMOptComp+
                LnkCtl: ASPM L1 Enabled; RCB 64 bytes, LnkDisable- CommClk-
                        ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
                LnkSta: Speed 2.5GT/s, Width x1
                        TrErr- Train- SlotClk- DLActive- BWMgmt- ABWMgmt-
                DevCap2: Completion Timeout: Range B, TimeoutDis+ NROPrPrP- LTR+
                         10BitTagComp+ 10BitTagReq+ OBFF Not Supported, ExtFmt+ EETLPPrefix-
                         EmergencyPowerReduction Not Supported, EmergencyPowerReductionInit-
                         FRS- TPHComp- ExtTPHComp-
                         AtomicOpsCap: 32bit- 64bit- 128bitCAS-
                DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis-
                         AtomicOpsCtl: ReqEn-
                         IDOReq- IDOCompl- LTR+ EmergencyPowerReductionReq-
                         10BitTagReq- OBFF Disabled, EETLPPrefixBlk-
                LnkCap2: Supported Link Speeds: 2.5GT/s, Crosslink- Retimer- 2Retimers- DRS-
                LnkCtl2: Target Link Speed: 2.5GT/s, EnterCompliance- SpeedDis-
                         Transmit Margin: Normal Operating Range, EnterModifiedCompliance- ComplianceSOS-
                         Compliance Preset/De-emphasis: -6dB de-emphasis, 0dB preshoot
                LnkSta2: Current De-emphasis Level: -6dB, EqualizationComplete- EqualizationPhase1-
                         EqualizationPhase2- EqualizationPhase3- LinkEqualizationRequest-
                         Retimer- 2Retimers- CrosslinkRes: unsupported
        Capabilities: [ac] MSI: Enable+ Count=1/1 Maskable+ 64bit+
                Address: 00000000fee00000  Data: 0000
                Masking: 00000000  Pending: 00000000
        Capabilities: [d0] Power Management version 3
                Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0+,D1-,D2-,D3hot+,D3cold-)
                Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
        Capabilities: [100 v1] Alternative Routing-ID Interpretation (ARI)
                ARICap: MFVC- ACS-, Next Function: 0
                ARICtl: MFVC- ACS-, Function Group: 0
        Capabilities: [420 v1] Physical Resizable BAR
                BAR 2: current size: 8GB, supported: 256MB 512MB 1GB 2GB 4GB 8GB
        Capabilities: [400 v1] Latency Tolerance Reporting
                Max snoop latency: 1048576ns
                Max no snoop latency: 1048576ns
        Kernel driver in use: vfio-pci
        Kernel modules: i915, xe

As you can see, my BIOS settings are correct for resizable bar and it is enabled on the A380. My motherboard is a Gigabyte X570 Aorus Ultra. I have made sure that (CSM = Disabled, Above 4G Decode = Enabled, Resizeable Bar = Auto, IOMMU = enabled) and has the latest BIOS installed.

First thing I did was installed the B50 into a Windows computer and performed all firmware updates (done on a computer with a Gigabyte B550M Gaming X WIFI). When I install the B50 into my Proxmox server things get very weird. I have tried this on all three kernels I have available on my machine:
  • 6.17.13-7-pve
  • 7.0.0-3-pve
  • 7.0.2-2-pve
With my current /etc/modprobe.d/pve-blacklist.conf, /etc/modprobe.d/vfio.conf and /etc/default/grub, I observe the following behavior:

First, display output freezes immediately after Starting pvestatd.service is shown on screen.

lspci -vvv
Code: 
0e:00.0 VGA compatible controller: Intel Corporation Battlemage G21 [Intel Graphics] (prog-if 00 [VGA controller])
Subsystem: Intel Corporation Device 1114
!!! Unknown header type 7f
IOMMU group: 33
Region 0: Memory at d0000000 (64-bit, prefetchable) [size=16M]
Region 2: Memory at c0000000 (64-bit, prefetchable) [size=256M]
Expansion ROM at fbe00000 [disabled] [size=2M]
Kernel modules: xe

0f:00.0 Audio device: Intel Corporation Device e2f7
Subsystem: Intel Corporation Device 1114
!!! Unknown header type 7f
IOMMU group: 34
Region 0: Memory at fc000000 (64-bit, non-prefetchable) [size=16K]
Kernel modules: snd_hda_intel

dmesg | grep 0e:00.0 shows no output

journalctl | grep 0e:00.0
Code: 
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: [8086:e212] type 00 class 0x030000 PCIe Endpoint
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 0 [mem 0xe0000000-0xe0ffffff 64bit pref]
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 2 [mem 0xd0000000-0xdfffffff 64bit pref]
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: ROM [mem 0xfbe00000-0xfbffffff pref]
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: PME# supported from D0 D3hot
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 0 [mem 0x00000000-0x00ffffff 64bit pref]
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 0 [mem 0x00000000-0x01ffffff 64bit pref]: contains BAR 0 for 2 VFs
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 2 [mem 0x00000000-0x1ffffffff 64bit pref]
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 2 [mem 0x00000000-0x3ffffffff 64bit pref]: contains BAR 2 for 2 VFs
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: vgaarb: setting as boot VGA device
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: vgaarb: bridge control possible
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: vgaarb: VGA device added: decodes=io+mem,owns=none,locks=none
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 2 [mem size 0x400000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 2 [mem size 0x400000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 0 [mem size 0x02000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 0 [mem size 0x02000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 2 [mem size 0x400000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 2 [mem size 0x400000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 2 [mem size 0x10000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 2 [mem size 0x10000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 0 [mem size 0x01000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 0 [mem size 0x01000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 0 [mem size 0x02000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 0 [mem size 0x02000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 2 [mem size 0x10000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 2 [mem size 0x10000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 0 [mem size 0x01000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 0 [mem size 0x01000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 2 [mem size 0x400000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: VF BAR 2 [mem size 0x400000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 2 [mem size 0x10000000 64bit pref]: can't assign; no space
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 2 [mem size 0x10000000 64bit pref]: failed to assign
May 09 21:51:48 pve kernel: pci 0000:0e:00.0: BAR 0 [mem size 0x01000000 64bit pref]: can't assign; no space

I then modified my config files below. I thought I would do this to try going the SR-IOV route instead of pass through and encountered different errors.

/etc/modprobe.d/pve-blacklist.conf
Code: 
blacklist nouveau
blacklist nvidiafb
blacklist radeon

/etc/modprobe.d/vfio.conf is now blank

Again, display output freezes immediately after Starting pvestatd.service is shown on screen.

lspci -vvv is unchanged

Code: 
0e:00.0 VGA compatible controller: Intel Corporation Battlemage G21 [Intel Graphics] (prog-if 00 [VGA controller])
        Subsystem: Intel Corporation Device 1114
        !!! Unknown header type 7f
        IOMMU group: 33
        Region 0: Memory at d0000000 (64-bit, prefetchable) [size=16M]
        Region 2: Memory at c0000000 (64-bit, prefetchable) [size=256M]
        Expansion ROM at fbe00000 [disabled] [size=2M]
        Kernel modules: xe

0f:00.0 Audio device: Intel Corporation Device e2f7
        Subsystem: Intel Corporation Device 1114
        !!! Unknown header type 7f
        IOMMU group: 34
        Region 0: Memory at fc000000 (64-bit, non-prefetchable) [size=16K]
        Kernel modules: snd_hda_intel

dmesg | grep 0e:00.0
Code: 
[ 12.166759] xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
[ 12.191572] xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
[ 12.192714] xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
[ 13.254961] xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
[ 13.275326] xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
[ 13.276219] xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
[ 14.343133] xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
[ 14.364154] xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
[ 14.364992] xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
[ 15.430547] xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
[ 15.430781] xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
[ 15.430784] xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
[ 15.430899] xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
[ 15.430905] xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
[ 15.431899] xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
[ 15.431902] xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
[ 15.432644] xe 0000:0e:00.0: [drm] *ERROR* [CRTC:150:pipe A] DSB 0 is busy
[ 25.542264] xe 0000:0e:00.0: [drm] *ERROR* [CRTC:150:pipe A] flip_done timed out
[ 25.543305] xe 0000:0e:00.0: [drm] *ERROR* [CRTC:150:pipe A] DSB 0 timed out waiting for idle (current head=0xfedaffff, head=0xfedaffff, tail=0xfedaffff)

journalctl | grep 0e:00.0
Code: 
May 09 22:08:46 pve kernel: xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
May 09 22:08:47 pve kernel: xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
May 09 22:08:47 pve kernel: xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
May 09 22:08:47 pve kernel: xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
May 09 22:08:48 pve kernel: xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
May 09 22:08:48 pve kernel: xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
May 09 22:08:48 pve kernel: xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
May 09 22:08:49 pve kernel: xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
May 09 22:08:49 pve kernel: xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
May 09 22:08:49 pve kernel: xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
May 09 22:08:50 pve kernel: xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
May 09 22:08:50 pve kernel: xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
May 09 22:08:50 pve kernel: xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
May 09 22:08:51 pve kernel: xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
May 09 22:08:51 pve kernel: xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
May 09 22:08:51 pve kernel: xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request
May 09 22:08:52 pve kernel: xe 0000:0e:00.0: [drm] vblank wait timed out on crtc 0
May 09 22:08:52 pve kernel: xe 0000:0e:00.0: [drm] *ERROR* Tile0: GT0: Force wake domain 0: wake. MMIO unreliable (forcewake register returns 0xFFFFFFFF)!
May 09 22:08:52 pve kernel: xe 0000:0e:00.0: [drm] Tile0: GT0: Forcewake domain 0x1 failed to acknowledge awake request


I am unsure what all these errors mean. Trying to search for them has been unfruitful. I have not found anyone else who has also had any issues swapping from a working Arc A-series GPU to an Arc B-series GPU either. I am not sure if this is a hardware compatibility issue, configuration issue or software/driver issue. Any help would be greatly appreciated.
 
Last edited:
I don’t own an Arc Pro B50, so I can’t say for sure, but I use an Arc B580 and have never made any configuration changes under /etc/modprobe.d/.
It will work properly simply by adding the necessary vfio-related entries to `modules` and specifying the following in the `<VMID>.conf` file.

I don’t know if that error would occur if I were to configure those settings.
Wouldn't it be better to just delete any settings whose purpose is unclear?

// The only setting I have configured

Code: 
cat /etc/pve/qemu-server/101.conf | grep hostpci
hostpci0: 0000:07:00;0000:08:00,pcie=1

cat /etc/modules
vfio
vfio_iommu_type1
vfio_pci

*The only boot option specified is `iommu=pt`.

// Proof that nothing has been changed

Code: 
cat /etc/modprobe.d/*
# The microcode module attempts to apply a microcode update when
# it autoloads.  This is not always safe, so we block it by default.
blacklist microcode
options kvm ignore_msrs=1 report_ignored_msrs=0
options kvm-intel nested=Y
options nct6683 force=1
# This file contains a list of modules which are not supported by Proxmox VE

# nvidiafb see bugreport https://bugzilla.proxmox.com/show_bug.cgi?id=701
blacklist nvidiafb
options zfs zfs_arc_max=17179869184

*I assume you're trying to configure dGPU passthrough. That's why I'm commenting. If you're trying to use it as a vGPU, your settings are probably all wrong.
 
Last edited:
can you please post the complete journal output from the last boot (journalctl -b) ? maybe we can see more what might be an issue?

i have an arc pro b50 here in my workstation and after updating the firmware it works fine
 
uzumo said:
I don’t own an Arc Pro B50, so I can’t say for sure, but I use an Arc B580 and have never made any configuration changes under /etc/modprobe.d/.
It will work properly simply by adding the necessary vfio-related entries to `modules` and specifying the following in the `<VMID>.conf` file.

I don’t know if that error would occur if I were to configure those settings.
Wouldn't it be better to just delete any settings whose purpose is unclear?

// The only setting I have configured

Code: 
cat /etc/pve/qemu-server/101.conf | grep hostpci
hostpci0: 0000:07:00;0000:08:00,pcie=1

cat /etc/modules
vfio
vfio_iommu_type1
vfio_pci

*The only boot option specified is `iommu=pt`.

// Proof that nothing has been changed

Code: 
cat /etc/modprobe.d/*
# The microcode module attempts to apply a microcode update when
# it autoloads.  This is not always safe, so we block it by default.
blacklist microcode
options kvm ignore_msrs=1 report_ignored_msrs=0
options kvm-intel nested=Y
options nct6683 force=1
# This file contains a list of modules which are not supported by Proxmox VE

# nvidiafb see bugreport https://bugzilla.proxmox.com/show_bug.cgi?id=701
blacklist nvidiafb
options zfs zfs_arc_max=17179869184

*I assume you're trying to configure dGPU passthrough. That's why I'm commenting. If you're trying to use it as a vGPU, your settings are probably all wrong.
Click to expand...

Thanks for your suggestions. I forgot to add that the failure of the B50 isn't occurring when the VM starts up. I have had the VM completely disabled. My goal was to do passthrough (only one VM needs the GPU), but since vfio wasn't working, I tried enabling the xe and going the sr-iov route which also didn't work. For reference, the line in my config file for passing my A380 is similar to yours hostpci0: 0000:0e:00,pcie=1,x-vga=1,romfile=dg2_gfx_fwupdate_SOC2.bin

I tried editing my /etc/modules to be identical to yours. I also added tried adding options kvm ignore_msrs=1 report_ignored_msrs=0, options kvm-intel nested=Y and options nct6683 force=1 and didn't change any behavior, unfortunately.

I also have ASPM disabled in BIOS, but also added pcie_aspm=off it to my GRUB_CMDLINE_LINUX_DEFAULT so it now looks like GRUB_CMDLINE_LINUX_DEFAULT="quiet amd_iommu=on iommu=pt pcie_aspm=off", and this also didn't change any behavior.
dcsapak said:
can you please post the complete journal output from the last boot (journalctl -b) ? maybe we can see more what might be an issue?

i have an arc pro b50 here in my workstation and after updating the firmware it works fine
Click to expand...
I have attached 3 logs for you. One is with the A380 which works fine, one is with the B50 and xe driver blacklisted and vfio enabled, and the other is with the vfio disabled and the xe driver not blacklisted.
 

Attachments

You must log in or register to reply here.
Top Bottom
Back