[PATCH RFC v3 01/21] ACPI: Only enumerate enabled (or functional) devices

[Rafael J. Wysocki]

On Thu, Dec 14, 2023 at 7:16 PM Rafael J. Wysocki <rafael at kernel.org> wrote:
>
> On Thu, Dec 14, 2023 at 7:10 PM Russell King (Oracle)
> <linux at armlinux.org.uk> wrote:
> >
> > On Thu, Dec 14, 2023 at 06:47:00PM +0100, Rafael J. Wysocki wrote:
> > > On Thu, Dec 14, 2023 at 6:32 PM Jonathan Cameron
> > > <Jonathan.Cameron at huawei.com> wrote:
> > > >
> > > > On Wed, 13 Dec 2023 12:49:16 +0000
> > > > Russell King (Oracle) <rmk+kernel at armlinux.org.uk> wrote:
> > > >
> > > > > From: James Morse <james.morse at arm.com>
> > > > >
> > > > > Today the ACPI enumeration code 'visits' all devices that are present.
> > > > >
> > > > > This is a problem for arm64, where CPUs are always present, but not
> > > > > always enabled. When a device-check occurs because the firmware-policy
> > > > > has changed and a CPU is now enabled, the following error occurs:
> > > > > | acpi ACPI0007:48: Enumeration failure
> > > > >
> > > > > This is ultimately because acpi_dev_ready_for_enumeration() returns
> > > > > true for a device that is not enabled. The ACPI Processor driver
> > > > > will not register such CPUs as they are not 'decoding their resources'.
> > > > >
> > > > > Change acpi_dev_ready_for_enumeration() to also check the enabled bit.
> > > > > ACPI allows a device to be functional instead of maintaining the
> > > > > present and enabled bit. Make this behaviour an explicit check with
> > > > > a reference to the spec, and then check the present and enabled bits.
> > > > > This is needed to avoid enumerating present && functional devices that
> > > > > are not enabled.
> > > > >
> > > > > Signed-off-by: James Morse <james.morse at arm.com>
> > > > > Tested-by: Miguel Luis <miguel.luis at oracle.com>
> > > > > Tested-by: Vishnu Pajjuri <vishnu at os.amperecomputing.com>
> > > > > Tested-by: Jianyong Wu <jianyong.wu at arm.com>
> > > > > Signed-off-by: Russell King (Oracle) <rmk+kernel at armlinux.org.uk>
> > > > > ---
> > > > > If this change causes problems on deployed hardware, I suggest an
> > > > > arch opt-in: ACPI_IGNORE_STA_ENABLED, that causes
> > > > > acpi_dev_ready_for_enumeration() to only check the present bit.
> > > >
> > > > My gut feeling (having made ACPI 'fixes' in the past that ran into
> > > > horribly broken firmware and had to be reverted) is reduce the blast
> > > > radius preemptively from the start. I'd love to live in a world were
> > > > that wasn't necessary but I don't trust all the generators of ACPI tables.
> > > > I'll leave it to Rafael and other ACPI experts suggest how narrow we should
> > > > make it though - arch opt in might be narrow enough.
> > >
> > > A chicken bit wouldn't help much IMO, especially in the cases when
> > > working setups get broken.
> > >
> > > I would very much prefer to limit the scope of it, say to processors
> > > only, in the first place.
> >
> > Thanks for the feedback and the idea.
> >
> > I guess we need something like:
> >
> >         if (device->status.present)
> >                 return device->device_type != ACPI_BUS_TYPE_PROCESSOR ||
> >                        device->status.enabled;
> >         else
> >                 return device->status.functional;
> >
> > so we only check device->status.enabled for processor-type devices?
>
> Yes, something like this.

However, that is not sufficient, because there are
ACPI_BUS_TYPE_DEVICE devices representing processors.

I'm not sure about a clean way to do it ATM.