Re: [PATCH 1/2] i386/acpi: fix inconsistent QEMU/OVMF device paths

[vit9696]

I am afraid I may not be available for this till the next weekend, but if the time permits then, I can have a look. Actually have another patch to upstream.

On Mon, Mar 1, 2021 at 00:37, Michael S. Tsirkin <mst@redhat.com> wrote:
>   On Sun, Feb 28, 2021 at 09:28:26PM +0000, vit9696 wrote:
> > Thomas, macOS is not really "special" here, it is rather that you will not
> > frequently use boot options in a VM. One of the most popular uses for boot
> > options is to switch between the operating systems, but for virtual machines
> > it is rarely the case. However, macOS does indeed use boot options for itself.
> > One example is to install updates. As long as the created boot option is not
> > valid an automated reboot during the update installation may result in the
> > wrong bootloader being chosen or in a stall within the firmware UI awaiting
> > manual boot option selection.
> >
> > Michael, does your suggestion mean that the default approach will be to keep
> > the new behaviour, but if you manually specify an older q35 machine version it
> > will provide the original behaviour. If so, it seems fair to me.
> >
> > Best regards,
> > Vitaly
>
> Exactly. Vitaly, could you cook up a patch like this?
>
> --
> MST

signature.asc
Description: OpenPGP digital signature