From: Mike Larkin <mlarkin@nested.page>
Subject: Re: [0/5] vmd: generate and expose ACPI tables to the guest via fw_cfg
To: Miguel Landaeta <miguel@miguel.cc>
Cc: tech@openbsd.org
Date: Wed, 10 Jun 2026 10:31:47 -0700

On Tue, Jun 09, 2026 at 02:03:06PM +0000, Miguel Landaeta wrote:
> I tested these changes on a guest running -current, please see below the
> relevant diff from dmesg output.
>
> I also tested a recent Linux kernel (6.18.x) and it still boots
> successfully.
>
>
> 15,16c15,16
> < bios0 at mainbus0: SMBIOS rev. 2.4 @ 0xf2220 (10 entries)
> < bios0: vendor SeaBIOS version "1.16.3p1-OpenBSD-vmm" date 01/01/2011
> ---
> > bios0 at mainbus0: SMBIOS rev. 2.4 @ 0xf1d40 (10 entries)
> > bios0: vendor SeaBIOS version "1.16.3p2-OpenBSD-vmm" date 01/01/2011
> 18,19c18,23
> < acpi at bios0 not configured
> < cpu0 at mainbus0: (uniprocessor)
> ---
> > acpi0 at bios0: ACPI 3.0
> > acpi0: sleep states
> > acpi0: tables DSDT FACP APIC
> > acpi0: wakeup devices
> > acpimadt0 at acpi0 addr 0xfee00000: PC-AT compat
> > cpu0 at mainbus0: apid 0 (boot processor)
> > cpu0: apic clock running at 0MHz
>

Thanks for the diffs. I already have a couple different implementations that
are further along than this however, so I'll probably continue merging those into
one workable version. Those include working lapic and ioapic emulation as well
as acpi table generation, etc...