XCP-ng
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Groups
    • Register
    • Login

    PCIe Pass-through lanes and lane performance

    Scheduled Pinned Locked Moved Compute
    36 Posts 11 Posters 3.5k Views 9 Watching
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • poddingueP Offline
      poddingue Vates 🪐 @JamesG
      last edited by

      Before you go further down the bridge tree, redakula called this one early in the thread and it's worth rereading.

      Intel Arc cards put an internal PCIe bridge between the slot and the GPU, and lspci on Linux only reports the top of that chain. His own A310 shows PCIe x8 4.0 at x4 4.0 in GPU-Z under Windows while Linux insists it is x1. That is the same bridge topology you're now staring at on 81, 82 and 83.

      TeddyAstie said much the same when he called it mostly a display issue, and he later checked that resizable BAR does work on UEFI guests since the OVMF build supports it. So the Gen1x1 reading may just be lspci describing the bridge rather than the card. I could easily be wrong, and the people already in this thread know this hardware far better than I do. 🤷

      J 1 Reply Last reply Reply Quote 0
      • J Online
        JamesG @poddingue
        last edited by

        @poddingue

        I'm in agreement with all of this. I can clearly see the Intel card starting at slot 81 and ending at 84. Unfortunately, in XCP-ng 8.3, only 83 (the "gpu" itself) is visible as an assignable PCI pass-through device. When I pass through just slot 83, I get a dysfunctional result.

        Again...I'm a little out of my expertise here, but my assumption here is that in order to actually get the full GPU to work, I need to get slots 81-84 passed over to the VM.

        Problem...In XCP-ng 8.2, you could mask off PCI devices with the pciback.hide function:

        /opt/xensource/libexec/xen-cmdline --set-dom0 "xen-pciback.hide=(0000:04:01.0)(0000:00:19.0)"

Or in my case:

/opt/xensource/libexec/xen-cmdline --set-dom0 "xen-pciback.hide=(0000:81:00.0)(0000:82.01.0)(0000:82:02.0)(0000:83:00.0)(0000:84:00.0)"

        Then you would assign those same PCI devices to the guest VM.

        In 8.3 you do this with an xe command:

        xe pci-list

and

xe pci-disable-dom0-access uuid=<pci uuid>

        The bridge devices for the Intel card aren't being included in the pci-list of XCP-ng and are therefore not assignable to guests. For example, running "xe pci-list" on a host shows me this single entry relevant to the Intel card:

        uuid ( RO)           : f3e6842b-ad1b-9e93-dbea-214dde28618a
    vendor-name ( RO): Intel Corporation
    device-name ( RO): Device e212
         pci-id ( RO): 0000:83:00.0

        With the new 8.3 method of assigning PCI resources to guest VM's, I don't see how I can fully pass-through the Intel GPU to a guest. I won't necessarily call this a "bug" but it is an unexpected and perhaps edge case behavior that Vates didn't expect.

        A 2 Replies Last reply Reply Quote 0
        • olivierlambertO Offline
          olivierlambert Vates 🪐 Co-Founder CEO
          last edited by

          Also a question for @Team-XAPI-Network

          1 Reply Last reply Reply Quote 1
          • A Offline
            andriy.sultanov Vates 🪐 XAPI & Network Team @JamesG
            last edited by

            @JamesG You can still manually use xen-cmdline to pass through 81-84 to the VM on 8.3 (this is in fact what xe pci-disable-dom0-access does behind the scenes).

            I'll try to see why other devices are omitted and how this can be fixed.

            1 Reply Last reply Reply Quote 0
            • A Offline
              andriy.sultanov Vates 🪐 XAPI & Network Team @JamesG
              last edited by

              @JamesG Could you pass through 81-84 manually and confirm that it actually fixes the issues for you?

              J 1 Reply Last reply Reply Quote 0
              • J Online
                JamesG @andriy.sultanov
                last edited by

                @andriy.sultanov Absolutely. I wasn't sure if the old way was incompatible with the new way or not.

                So the old 8.2 method will still work without conflict in 8.3??

                If so, I'll have an answer for you shortly.

                A 1 Reply Last reply Reply Quote 0
                • A Offline
                  andriy.sultanov Vates 🪐 XAPI & Network Team @JamesG
                  last edited by

                  @JamesG it should be compatible, yes.

                  you can check the current command-line value with /opt/xensource/libexec/xen-cmdline --get-dom0 xen-pciback.hide and add all 81-84 devices with --set-dom0

                  J 1 Reply Last reply Reply Quote 0
                  • J Online
                    JamesG @andriy.sultanov
                    last edited by

                    @andriy.sultanov First problem...

                    /opt/xensource/libexec/xen-cmdline --get-dom0 xen-pciback.hide
xen-pciback.hide=(0000:81:00.0)(0000:82.01.0)(0000:82:02.0)(0000:83:00.0)(0000:84:00.0)

                    That looks good...

                    xl pci-assignable-list
libxl: error: libxl_pci.c:444:libxl_device_pci_assignable_list: Looks like pciback driver not loaded

                    Not so much...

                    I did go ahead and assign those PCI devices to the guest, but of course it won't start, complaining that the pciback driver isn't loaded.

                    J 1 Reply Last reply Reply Quote 0
                    • J Online
                      JamesG @JamesG
                      last edited by

                      Nevermind...Typo...Rebooting now to see...

                      J 1 Reply Last reply Reply Quote 0
                      • J Online
                        JamesG @JamesG
                        last edited by

                        Argh...

                        Okay...First up:

                        lspci:

                        81:00.0 PCI bridge: Intel Corporation Device e2ff (rev 01)
                        82:01.0 PCI bridge: Intel Corporation Device e2f0
                        82:02.0 PCI bridge: Intel Corporation Device e2f1
                        83:00.0 VGA compatible controller: Intel Corporation Device e212
                        84:00.0 Audio device: Intel Corporation Device e2f7

                        Execute the hide:

                        /opt/xensource/libexec/xen-cmdline --set-dom0 "xen-pciback.hide=(0000:81:00.0)(0000:82:01.0)(0000:82:02.0)(0000:83:00.0)(0000:84:00.0)"

                        Reboot and verify the hide:

                        xl pci-assignable-list
                        0000:84:00.0
                        0000:82:01.0
                        0000:83:00.0
                        0000:81:00.0
                        0000:82:02.0

                        Looks good....make the assignment:

                        xe vm-param-set other-config:pci=0/0000:81:00.0,0/0000:82:01.0,0/0000:82:02.0,0/0000:83:00.0,0/0000:84:00.0 uuid=9d76f849-16bf-0414-9072-6904176bbb01

                        Verify:

                        xe vm-param-list uuid=9d76f849-16bf-0414-9072-6904176bbb01

                        other-config (MRW): pci: 0/0000:81:00.0,0/0000:82:01.0,0/0000:82:02.0,0/0000:83:00.0,0/0000:84:00.0; import_task: OpaqueRef:9e15d1aa-5e79-79d9-af81-4dc299725217; xo:9d76f849: {"creation":{"date":"2025-12-16T19:19:14.724Z","template":"1f8728f9-6354-4a68-adc3-222c34f2bb94","user":"ae9cf9c9-bbcc-4dc9-b1bb-d9fb1b657c60"}}; base_template_name: Windows 11; mac_seed: 5891089b-e9ec-3050-30e5-564db3e6e170; install-methods: cdrom

                        Starting up the VM results in this:

                          "params": [
    "xenopsd internal error: Cannot_add(0000:82:02.0, Device_common.QMP_Error(1, \"{\\\"error\\\":{\\\"class\\\":\\\"GenericError\\\",\\\"desc\\\":\\\"Failed to initialize 11/15, type = 0x1, rc: -1\\\",\\\"data\\\":{}},\\\"id\\\":\\\"qmp-000006-1\\\"}\"))"
  ],

                        I'm gonna strip it all back, reboot and start over.

                        J 1 Reply Last reply Reply Quote 0
                        • J Online
                          JamesG @JamesG
                          last edited by

                          Yep...

                          Stripped it all out....Rebooted...

                          Set the pcihideback for dom0...Rebooted.

                          Verified all devices were assignable.

                          Assigned to guest VM, booted...No go.

                          Initially the log said it couldn't assign 82:02.0
                          So I removed it.
                          Then it couldn't assign 82:01.0
                          So I removed it.
                          Then it couldn't assign 81:00.0
                          So I removed it.

                          Now the VM works...But we're effectively back to just passing through the GPU itself, and I guess the audio device.

                          No functional change.

                          TeddyAstieT 1 Reply Last reply Reply Quote 0
                          • TeddyAstieT Offline
                            TeddyAstie Vates 🪐 XCP-ng Team Xen Guru @JamesG
                            last edited by

                            @JamesG

                            You can't PCI Passthrough a PCI bridge (and even if you could, it wouldn't be very useful), as it's a part of PCI internals and not really something the guest requires to use the device.

                            As I asked previously, is there any issue (aside display like PCIe 1.0 1x) with the device, is it behavior correctly or having the expected performance ?

                            J 1 Reply Last reply Reply Quote 0
                            • J Online
                              JamesG @TeddyAstie
                              last edited by

                              @TeddyAstie From what I've tried...It won't work at all in Linux.

                              It kind of works in Windows, and while I do see some load on the GPU when running something, there's as much if not more load on the CPU. It's not what I expected at least.

                              I've never really gamed on PC, and definitely never Linux. My only uses for GPU's have been in media composition which was always on dedicated systems and Nvidia GPUs. More recently I've been doing some AI tasks, again with passed-through NVidia GPUs.

                              Side tangent...
                              I ran into another workload last night where the app wanted OpenGL...But OpenGL doesn't work with the "normal" VGA adapter in XCP-ng, and I don't have a GPU on that particular pool to try a pass-through. That lead me to VirGL, but it doesn't look like that's gone anywhere.

                              1 Reply Last reply Reply Quote 0

                              Hello! It looks like you're interested in this conversation, but you don't have an account yet.

                              Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.

                              With your input, this post could be even better 💗

                              Register Login
                              • First post
                                Last post