VM wont start with NVMe device passed through

Software & Operating Systems Linux
1

i am attempting to pass through my NVMe device to my VM, lspci -nnk shows that vfio-pci is loaded into the device, but attempting to start the VM yields this error:

error: internal error: process exited while connecting to monitor: 2019-01-11T23:19:26.245272Z qemu-system-x86_64: -device vfio-pci,host=41:00.0,id=hostdev2,bootindex=
1,bus=pci.0,addr=0x8,rombar=0: vfio 0000:41:00.0: failed to add PCI capability 0x11[0x50]@0xb0: table & pba overlap, or they don’t fit in BARs, or don’t align

ive never seen this error before and have no idea what could be causing it.
any help is apreciated.

2

More detail would be nice, (hardware used, software etc)

3

my motherboard is the asus prime x399-a and the NVMe device in question is the crucial CT500P1SSD8

4

Ive seen this before where NVMe devices are using the SMI SM2262 controller. Detailed here:
https://bugzilla.kernel.org/show_bug.cgi?id=202055.
I think there’s some things detailed in the later posts which might help patch the issue.

Samsung NVMe devices dont have this issue.

2 Likes
5

Looks like im also having this issue.

Crucial, CT500P1SSD8

Error: failed to add PCI capability 0x11[0x50]@0xb0: table & pba overlap, or they don’t fit in BARs, or don’t align

Maybe I need to refund my Amazon M.2 Drive before March 3rd lol.

6

Rats… I also have the same issue as well (I own a ADATA SX8200NP)…

I have not tried the patch yet.

Edit: I am also going to assume that the patch has not been upstreamed, right?

Update: Just edit your xml (taken from this comment):

Here is an example that Alex Williamson provided. Make sure that your apply it to the right PCI device.

<domain type='kvm' xmlns:qemu='http://libvirt.org/schemas/domain/qemu/1.0'>
...
    <hostdev mode='subsystem' type='pci' managed='yes'>
      <source>
        <address domain='0x0000' bus='0x01' slot='0x00' function='0x0'/>
      </source>
      <alias name='ua-sm2262'/>
      <address type='pci' domain='0x0000' bus='0x02' slot='0x00' function='0x0'/>
    </hostdev>
...
  <qemu:commandline>
    <qemu:arg value='-set'/>
    <qemu:arg value='device.ua-sm2262.x-msix-relocation=bar2'/>
  </qemu:commandline>

Make sure to read the comment though, there are limitations with this workaround, but I can confirm that this works wonders for my Hackintosh VM.

1 Like
7

For Intel SSD 660p series, the latest firmware is required for pass-through to work without issues. The latest one should be from mid 2020 at the time of writing this.

Not even any workarounds or special settings for qemu/libvirt are needed, just a simple pass-through setup as with any other PCIe device.

8

Which tool do you use to update the Intel SSD 660p firmware (preferably under linux)? I’m getting this with a tool found on Intel’s website:

root@debian:~# intelmas load -intelssd 1
WARNING! You have selected to update the drives firmware! 
Proceed with the update? (Y|N): Y
Checking for firmware update...
Error: This drive is no longer supported.
root@debian:~#

EDIT: A tool from Solidigm website seems to have worked (sorry can’t post link from my account)