Re: [PATCH] PNPACPI: do ACPI binding directly

["Rafael J. Wysocki" <rjw@rjwysocki.net>]

On Tuesday, July 08, 2014 09:17:09 AM Zhang Rui wrote:
> On Mon, 2014-07-07 at 22:03 +0200, Rafael J. Wysocki wrote:
> > On Monday, July 07, 2014 10:27:26 PM Zhang Rui wrote:
> > > On Mon, 2014-07-07 at 14:53 +0200, Rafael J. Wysocki wrote:
> > > > On Wednesday, July 02, 2014 08:54:49 PM Zhang Rui wrote:
> > > > > From 7daac7fc7cd7b605ccd84f10fc206cedf6170e89 Mon Sep 17 00:00:00 2001
> > > > > From: Zhang Rui <rui.zhang@intel.com>
> > > > > Date: Fri, 20 Jun 2014 10:14:07 +0800
> > > > > Subject: [PATCH] PNPACPI: do ACPI binding directly
> > > > > 
> > > > > PNPACPI uses acpi_bus_type to do ACPI binding for the PNPACPI devices.
> > > > > 
> > > > > This is overkill because PNPACPI code already knows which ACPI
> > > > > device object to bind during PNPACPI device enumeration.
> > > > > 
> > > > > This patch removes acpi_pnp_bus and does the binding by invoking
> > > > > acpi_bind_one() directly after device enumerated.
> > > > > 
> > > > > This also fixes a bug in the previous code that some PNPACPI devices failed
> > > > > to be bound because
> > > > > 1. the ACPI device _HID is not pnpid, e.g. "MSFT0001", but its _CID is,
> > > > >    e.g. "PNP0303", thus ACPI _CID is used as the pnp device device id.
> > > > > 2. device is bound only if the pnp device id matches the ACPI device _HID.
> > > > > 
> > > > > Tested-by: Prigent Christophe <christophe.prigent@intel.com>
> > > > > Signed-off-by: Zhang Rui <rui.zhang@intel.com>
> > > > 
> > > > Does this fix a regression?
> > > 
> > > Hmmm, no, the problem exists in all previous kernel versions IMO.
> > 
> > OK, so it is applicable to -stable too?  Or would it need to be backported?
> > 
> I don't think we need it in stable kernel, because this is not a
> critical issue.
> I've seen only two devices with this problem, and all of them don't have
> ACPI _PSx/_PRx method. This means that, although pnpacpi_resume() fails
> because of this bug, nothing really bad happens because pnpacpi_resume()
> is actually a no-op for such devices.

I thought so.

I'll queue it up for 3.17, then, because it is a framework-level change
rather than an isolated fix.

Rafael