On Fri, Nov 7, 2025 at 4:06 PM Jinpu Wang <[email protected]> wrote:
>
> Hi Jason,
>
> On Fri, Nov 7, 2025 at 2:02 AM Jason Wang <[email protected]> wrote:
> >
> > On Thu, Nov 6, 2025 at 10:28 PM Jinpu Wang <[email protected]> wrote:
> > >
> > > Hi Peter,
> > > On Wed, Nov 5, 2025 at 11:17 PM Peter Xu <[email protected]> wrote:
> > > >
> > > > On Wed, Nov 05, 2025 at 10:27:59AM +0100, Jinpu Wang wrote:
> > > > > > > These are not present (or not supported) on QEMU 8.2.10, which
> > > > > > > causes
> > > > > > > the migration state load to fail.
> > > > > >
> > > > > > Interesting, we've already done the compat work:
> > > > > >
> > > > > > GlobalProperty hw_compat_8_1[] = {
> > > > > > { TYPE_PCI_BRIDGE, "x-pci-express-writeable-slt-bug", "true" },
> > > > > > { "ramfb", "x-migrate", "off" },
> > > > > > { "vfio-pci-nohotplug", "x-ramfb-migrate", "off" },
> > > > > > { "igb", "x-pcie-flr-init", "off" },
> > > > > > { TYPE_VIRTIO_NET, "host_uso", "off"},
> > > > > > { TYPE_VIRTIO_NET, "guest_uso4", "off"},
> > > > > > { TYPE_VIRTIO_NET, "guest_uso6", "off"},
> > > > > > };
> > > > > > const size_t hw_compat_8_1_len = G_N_ELEMENTS(hw_compat_8_1);
> > > > > Yeah, I noticed the same.
> > > >
> > > > AFAICT, this is a known issue..
> > > >
> > > > Thomas and I used to suggest we should not turn on USO* by default by
> > > > probing kernel, but only allow user choosing it explicitly in a VM
> > > > setup. IOW, dest qemu should stop booting at all when kernel is too old
> > > > (when user chose the feature).
> > > I feel this is the approach we should have picked.
> > > >
> > > > See:
> > > >
> > > > https://lore.kernel.org/all/ZqQNKZ9_OPhDq2AK@x1n/
> > > Is there any effort to allow migration from new OS support the USO
> > > features to old OS doesn't support it?
> >
> > You can teach your management to disable USO via the qemu command line.
>
> I added "host_uso=false,guest_uso4=false,guest_uso6=false" for -device
> virtio-net-pci
> But migration still fails with slightly different error:
> char device redirected to /dev/pts/1 (label charserial0)
> 2025-11-07T15:00:45.528098Z qemu-8.2: Features 0x10130afffa7
> unsupported. Allowed features: 0x179bfffe7
> 2025-11-07T15:00:45.528245Z qemu-8.2: Failed to load virtio-net:virtio
> 2025-11-07T15:00:45.528253Z qemu-8.2: error while loading state for
> instance 0x0 of device '0000:00:02.0:06.0/virtio-net'
>
> I suppose it is VIRTIO_F_RING_RESET?
> Any idea? I didn't find option to disable it
Found it, "host_uso=false,guest_uso4=false,guest_uso6=false,queue_reset=false"
is needed to allow migration from new OS to old OS
> >
> > > Any hint to make it work?
> >
> > Thanks
>
> Thx!
> >
> > > >
> > > > Thanks,
> > > >
> > > > --
> > > > Peter Xu
> > > >
> > > Thx for the help.
> > >
> >
