Re: "can't evaluate _CRS: 1" messages since b355cee [ACPI / resources: ignore invalid ACPI device resources]

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

On Tuesday, March 11, 2014 01:11:19 PM Zhang, Rui wrote:
> 
> > -----Original Message-----
> > From: Rafael J. Wysocki [mailto:rjw@rjwysocki.net]
> > Sent: Tuesday, March 11, 2014 9:23 PM
> > To: Zhang, Rui
> > Cc: Markus Trippelsdorf; linux-acpi@vger.kernel.org; Wysocki, Rafael J;
> > linux-kernel@vger.kernel.org
> > Subject: Re: "can't evaluate _CRS: 1" messages since b355cee [ACPI /
> > resources: ignore invalid ACPI device resources]
> > Importance: High
> > 
> > On Tuesday, March 11, 2014 12:59:47 PM Zhang, Rui wrote:
> > > I think the reason problem is that:
> > > 1. there are zero length IO/Memory resources for these machines, and
> > for those resources
> > >    a) w/o commit b355cee88e3b1a193f0e9a81db810f6f83ad728b,
> > acpi_dev_resource_memory/io returns true
> > >    b) w commit b355cee88e3b1a193f0e9a81db810f6f83ad728b,
> > > acpi_dev_resource_memory/io returns false 2. in
> > pnpacpi_allocated_resource(), when acpi_dev_resource_memory/io()
> > returns false, it translates into "the resource is not IO/Memory
> > resource".
> > >
> > > IMO, the root cause is that the BOOL return value of
> > > acpi_dev_resource_memory/io() can not represent the following three
> > > situations
> > > 1) a valid IO/Memory resource
> > > 2) an invalid IO/Memory resource
> > > 3) not an IO/Memory resource
> > 
> > Well, can we make acpi_dev_resource_memory/io return an int, then, so
> > that it can cover all of the cases it needs to cover correctly and so
> > that we can get rid of those confusing messages?
> > 
> Sure, I will submit a patch soon.

Thanks!

-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.