Re: [RFC PATCH 6/6] hw/arm/cubieboard: Disable unsupported M-USB in device tree blob

[Philippe Mathieu-Daudé]

On 1/6/20 5:07 PM, Peter Maydell wrote:
> On Mon, 30 Dec 2019 at 11:10, Philippe Mathieu-Daudé <address@hidden> wrote:
> > We do not model the Sunxi Multipoint USB.
> > The Linux kernel OOPS when booting:
>
> > This is not critical but confusing. To avoid the Linux kernel to
> > probe this device, mark it disabled in the device tree blob.
> >
> > Signed-off-by: Philippe Mathieu-Daudé <address@hidden>
> > ---
> > I'm not sure if this is a QEMU anti-pattern or bad practice.
> > I know we prefer to be as close to the hardware as possible,
> > but here the hardware is not changed, the dtb is. However
> > this makes the guest behave differently. At least we don't
> > have to manually edit the dts. If this is only annoying for
> > acceptance testing, we might consider manually editing the
> > dts in the tests setup().
>
> In general I'd definitely prefer it if we avoided editing
> the DTB for things like this (we have very few boards that
> edit the DTB and none of them do it as workarounds for
> buggy/missing device emulation).

Yes I understand (I was expecting this answer, but better to have it 
explicit).

> Is it possible to have
> at least a dummy implementation of the controller that's
> enough to make the probe function fail more cleanly?

Niek gave some pointers. Anyway The kernel is not crashing, it is just 
oopsing, so this is not fatal (the VM is still usable).

Thanks!

Phil.