Got Error Message when starting VM PCI_CONFIG_SPACE_SIZE' failed


Recommended Posts

Hello Everybody, im new to this community and wants to start working with unraid but i can´t passthrough my Intel GPU to the VM. Every Time im trying zu start the vm this Error Message comes out.

 

internal error: qemu unexpectedly closed the monitor: qemu-system-x86_64: hw/pci/pcie.c:686: pcie_add_capability: Assertion `prev >= PCI_CONFIG_SPACE_SIZE' failed.

 

I have passthrough the cpu and display / sound from mainboard to the vm.

Choosen Q35-2.7 as Maschine and selected SeaBios.  (tried with i440fx-2.7 lets start the vm but no picture on the hdmi Signal, only unraid shell )

So i have read, that Q35-2.7 has to be choosen to get the screen on hdmi, but here i got the above error message :(

 

I have the ASROCK j4205-ITX with Apollo Lake CPU. The Maschine would be used mainly for Kodi and by the way for smb ftp and other stuff....

My Bios is 1.10b ( got it from support for vt-d support enabled ) because the 1.10 don´t have this option.

 

Here are my System Informations:

 

Model: Custom

M/B: ASRock - J4205-ITX

CPU: Intel® Pentium® CPU J4205 @ 1.50GHz

HVM: Enabled

IOMMU: Enabled

Cache: 224 kB, 2048 kB

Memory: 16 GB (max. installable capacity 16 GB)

Network: bond0: fault-tolerance (active-backup), mtu 1500

eth0: 1000 Mb/s, full duplex, mtu 1500

Kernel: Linux 4.8.12-unRAID x86_64

OpenSSL: 1.0.2j

UNRAID VERSION: 6.3.0-rc6

 

i can´t load the LOG PAGE, it loads all the time and shows me only a blank screen.

 

So i heard about unraid and wants to start with it. I hope anyone can help me to get working on this board and software. I like the feeling like VMware ESX :) and the possibility for a lot of things.

And excuse my englisch, please. Not my Main Language at all :)

 

Best Regards

 

Link to comment
  • 4 weeks later...

I can confirm it happens to me as well, same board and bios 1.20, shame as it is an awesome board for the rest, I added a PCI-e x1 Marvell card for another 4 SATA ports, total of 8 SATA3, it's new hardware and doesn't boot on any kernel < 4.5 (incredibly slow and impossible to take boot on unRAID 6.2 as kernel is 4.4), I'm sure support will get better.

 

Note: bios 1.20 has the option for VT-D so no need to use a special rom.

 

 

Link to comment

Does anyone from Limitech know if there are known issues with Apollo Lake SoC?

 

This was originally open as a defect and I think it should have stayed as that.

 

libvirtd.log

2017-01-27 15:52:53.170+0000: 21493: error : qemuMonitorIORead:586 : Unable to read from monitor: Connection reset by peer
2017-01-27 15:52:53.171+0000: 21493: error : qemuProcessReportLogError:1809 : internal error: qemu unexpectedly closed the monitor: qemu-system-x86_64: hw/pci/pcie.c:686: pcie_add_capability: Assertion `prev >= PCI_CONFIG_SPACE_SIZE' failed.
2017-01-27 15:52:53.372+0000: 21498: error : qemuProcessReportLogError:1809 : internal error: process exited while connecting to monitor: qemu-system-x86_64: hw/pci/pcie.c:686: pcie_add_capability: Assertion `prev >= PCI_CONFIG_SPACE_SIZE' failed.

 

qemy LibreELEC.log

2017-01-27 15:52:51.612+0000: starting up libvirt version: 2.4.0, qemu version: 2.7.0, hostname: Mammuth
LC_ALL=C PATH=/bin:/sbin:/usr/bin:/usr/sbin HOME=/ QEMU_AUDIO_DRV=none /usr/local/sbin/qemu -name guest=LibreELEC,debug-threads=on -S -object secret,id=masterKey0,format=raw,file=/var/lib/libvirt/qemu/domain-2-LibreELEC/master-key.aes -machine pc-q35-2.7,accel=kvm,usb=off,dump-guest-core=off,mem-merge=off -drive file=/usr/share/qemu/ovmf-x64/OVMF_CODE-pure-efi.fd,if=pflash,format=raw,unit=0,readonly=on -drive file=/etc/libvirt/qemu/nvram/27238c84-c8d3-39c1-a1c3-68f6677077ce_VARS-pure-efi.fd,if=pflash,format=raw,unit=1 -m 1024 -realtime mlock=off -smp 2,sockets=1,cores=2,threads=1 -uuid 27238c84-c8d3-39c1-a1c3-68f6677077ce -display none -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/domain-2-LibreELEC/monitor.sock,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc,driftfix=slew -global kvm-pit.lost_tick_policy=discard -no-hpet -no-shutdown -boot strict=on -device i82801b11-bridge,id=pci.1,bus=pcie.0,addr=0x1e -device pci-bridge,chassis_nr=2,id=pci.2,bus=pci.1,addr=0x0 -device ich9-usb-ehci1,id=usb,bus=pcie.0,addr=0x7.0x7 -device ich9-usb-uhci1,masterbus=usb.0,firstport=0,bus=pcie.0,multifunction=on,addr=0x7 -device ich9-usb-uhci2,masterbus=usb.0,firstport=2,bus=pcie.0,addr=0x7.0x1 -device ich9-usb-uhci3,masterbus=usb.0,firstport=4,bus=pcie.0,addr=0x7.0x2 -device virtio-serial-pci,id=virtio-serial0,bus=pci.2,addr=0x3 -drive file=/mnt/user/domains/LibreELEC/LibreELEC-unRAID.x86_64-7.0.1_1.img,format=raw,if=none,id=drive-virtio-disk2,readonly=on,cache=writeback -device virtio-blk-pci,scsi=off,bus=pci.2,addr=0x4,drive=drive-virtio-disk2,id=virtio-disk2,bootindex=1 -fsdev local,security_model=passthrough,id=fsdev-fs0,path=/mnt/user/appdata/LibreELEC/ -device virtio-9p-pci,id=fs0,fsdev=fsdev-fs0,mount_tag=appconfig,bus=pci.2,addr=0x1 -netdev tap,fd=24,id=hostnet0,vhost=on,vhostfd=27 -device virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:ee:0d:53,bus=pci.2,addr=0x2 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -chardev socket,id=charchannel0,path=/var/lib/libvirt/qemu/channel/target/domain-2-LibreELEC/org.qemu.guest_agent.0,server,nowait -device virtserialport,bus=virtio-serial0.0,nr=1,chardev=charchannel0,id=channel0,name=org.qemu.guest_agent.0 -device vfio-pci,host=00:02.0,id=hostdev0,bus=pcie.0,addr=0x2 -device vfio-pci,host=00:0e.0,id=hostdev1,bus=pci.2,addr=0x5 -device virtio-balloon-pci,id=balloon0,bus=pci.2,addr=0x6 -msg timestamp=on
Domain id=2 is tainted: high-privileges
char device redirected to /dev/pts/1 (label charserial0)
qemu-system-x86_64: hw/pci/pcie.c:686: pcie_add_capability: Assertion `prev >= PCI_CONFIG_SPACE_SIZE' failed.
2017-01-27 15:52:53.372+0000: shutting down, reason=failed

Link to comment
  • 3 weeks later...

did someone tried the new 6.3.1 version?

 

Version 6.3.1 2017-02-08

------------------------

 

Base distro:

 

- apcupsd: version 3.14.14 (rev2 -- added modbus usb support)

- gd: version 2.2.4 (CVE-2016-9317, CVE-2016-6912)

- libusb-compat: version 0.1.5

- qemu: version 2.7.1

- samba: version 4.5.5

 

Linux kernel:

 

- version 4.9.8

Link to comment
  • 3 weeks later...

I did get it to boot. Seems it was whenever I choose the integrated sound card ("Intel Sunrise Point-H HD Audio") it wouldn't boot. Choosing no sound card lets it boot fine. Maybe the integrated sound card and video are in the separate groups?

Edited by drzaph
Link to comment
  • 2 weeks later...

my system:  ASRock - J4205-ITX + 16GB Ram

unRAID 6.3.2

 

Tried OpenElec 6.0.3 but soon as i pick integrated GK (Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series Integrated Graphics Controller (rev 0b)) instead of VNC - it freezes the whole NAS/OS and i have to reboot the whole system manually.

 

Tried to:

Create OpenElec 7.0.1 VM with GK - freezes right away (hard reset)

switch " Enable PCIe ACS Override: " yes/no no difference

---

LibreElec test was the same, soon aj i put IGK intead of VNC, it freezes and need to do hard reset :(

Tried Machine: i440fx and Q35 2.7 as well

BIOS: OVFM, SeaBios

 

After the reboot it also disable whole VMs menu - so i need to enable it again

 

When its gonna be fixed? I was planing to purchase it, but the 2nd main function - Media Center is not working :(

Its been ongoing issue since november/december and not fixed yet! Aru you guys in Limetech doing something about it?

 

Thanks

 

 

 

Link to comment
21 hours ago, killeriq said:

update:

Now seems that im able to run

LibreElec - but never goes into Kodi it self, just showing loadig , starting in text mode and then it loops in controller screen information each 30s

OpenElec - shows just black screen

unraidtower-diagnostics-20170315-2313.zip

 

In reviewing your diagnostics, there are a few events which jump out at me:

Mar 15 23:10:40 unRAIDTower kernel: vfio_cap_init: 0000:00:0e.0 pci config conflict @0x80, was cap 0x9 now cap 0x10
Mar 15 23:10:40 unRAIDTower kernel: vfio_cap_init: 0000:00:0e.0 pci config conflict @0x81, was cap 0x9 now cap 0x10
Mar 15 23:10:40 unRAIDTower kernel: vfio_cap_init: 0000:00:0e.0 pci config conflict @0x82, was cap 0x9 now cap 0x10
Mar 15 23:10:40 unRAIDTower kernel: vfio_cap_init: 0000:00:0e.0 pci config conflict @0x83, was cap 0x9 now cap 0x10
Mar 15 23:11:18 unRAIDTower kernel: DMAR: DRHD: handling fault status reg 2
Mar 15 23:11:18 unRAIDTower kernel: DMAR: [DMA Write] Request device [00:02.0] fault addr 0 [fault reason 02] Present bit in context entry is clear

These messages show up after starting your VM.  I don't see these messages on my test system, though we don't have the same hardware as you.

 

I think the issue here is that something is funky with that hardware platform that doesn't make it work correctly with IGD assignment.  I'm curious if regular assignment of other PCI devices (such as a discrete GPU, USB controller, or otherwise) have the same issue on your system or if its only the IGD device exhibiting this abnormal behavior.  I'm going to send the diagnostics you provided to a contact on the VFIO project (Alex Williamson) but chances are this either will require a BIOS update from the vendor or a code change in the software / kernel.  This won't be a quick fix.

Link to comment

To be perfectly honest, we have no idea.  The issue is that there are very few users of that particular motherboard in combination with the features you want to leverage.  I would doubt even ASRock had ever tested the use of IGD assignment with that motherboard.  You could definitely contact ASRock and make mention of what you are trying to do and the problems you are facing, but I have a feeling they will push back and just blame the software (not unRAID OS specifically, but KVM/QEMU).

 

The only thing I needed from you was your diagnostics which contains all the log events necessary.

Link to comment

hmm...great so i just got a new ITX useless piece of HW - which cant be used in the way i wanted.

It can be only as NAS and VMs and Kodi only as VNC so no HDMI output or sound.

What are my options?

 

1. wait for fix 

2. get some additional graphic card - please suggest which one

3. get different ITX board - please suggest which one

4. get Qnap which just work - they use some Celeron j3060

Link to comment
20 hours ago, killeriq said:

hmm...great so i just got a new ITX useless piece of HW - which cant be used in the way i wanted.

It can be only as NAS and VMs and Kodi only as VNC so no HDMI output or sound.

What are my options?

 

1. wait for fix 

2. get some additional graphic card - please suggest which one

3. get different ITX board - please suggest which one

4. get Qnap which just work - they use some Celeron j3060

 

We have a list of tested hardware here:  http://lime-technology.com/wiki/index.php/UnRAID_6/Overview#Lime_Technology_Tested_Components

 

You can also search the forums for other posts from folks using virtualization and GPU pass through for recommendations, but I think it'd be bad for us to make recommendations for hardware we haven't personally tested ourselves.  There is also this community maintained spreadsheet of tested configurations that applies to QEMU/KVM/VFIO in general:  https://docs.google.com/spreadsheets/d/1LnGpTrXalwGVNy0PWJDURhyxa3sgqkGXmvNCIvIMenk/edit#gid=0

 

In general, building a custom unRAID system can go one of two ways:  you can go with recommendations from us and our community which are very likely to work or you can go off of hardware spec sheets and DIY the entire project.  If you go with the latter, you always run the risk of being the first to jump in the water to test a new combination of hardware and software features, and thus be left out in the wind for a bit while issues get stamped out.  In those scenarios, you have to do exactly what you did:  create a post in the forum and include your diagnostics so we can get more information.  From here, it's the waiting game.  We rely on upstream projects like KVM/QEMU/VFIO to resolve these types of issues because the issues originate from their projects.  If you were having a storage issue with your drives or controllers, we'd be able to do a lot more directly because we maintain that ourselves.

  • Upvote 1
Link to comment
  • 6 months later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.