[PATCH v9 18/21] ARM64 / ACPI: Select ACPI_REDUCED_HARDWARE_ONLY if ACPI is enabled on ARM64

From: lorenzo.pieralisi@arm.com (Lorenzo Pieralisi)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v9 18/21] ARM64 / ACPI: Select ACPI_REDUCED_HARDWARE_ONLY if ACPI is enabled on ARM64
Date: Tue, 10 Mar 2015 14:16:46 +0000	[thread overview]
Message-ID: <20150310141645.GA25123@red-moon> (raw)
In-Reply-To: <54FEE227.4000808@linaro.org>

On Tue, Mar 10, 2015 at 12:23:03PM +0000, Hanjun Guo wrote:
> On 2015?03?07? 01:47, Lorenzo Pieralisi wrote:
> > On Wed, Feb 25, 2015 at 08:39:58AM +0000, Hanjun Guo wrote:
> >> From: Al Stone <al.stone@linaro.org>
> >>
> >> ACPI reduced hardware mode is disabled by default, but ARM64
> >> can only run properly in ACPI hardware reduced mode, so select
> >> ACPI_REDUCED_HARDWARE_ONLY if ACPI is enabled on ARM64.
> >
> > Agreed. This will remove unneeded acpica code. I think though,
> > you should also check the FADT flag HW_REDUCED_ACPI at boot to
> > make sure that the tables are HW reduced compliant. I am flagging
> 
> Agreed, I will add that check on top of this patch set. We explicitly
> stated that we are using hardware reduced mode on ARM64 in the doc, so, 
> how about print some warning message and go on with the boot if we
> got the wrong flag HW_REDUCED_ACPI at boot?

Why would be boot if the ACPI tables are not HW_REDUCED_ACPI compliant ?

To look for trouble :) ? arm64 does not support any other ACPI model,
the missing flag should disable ACPI IMO on arm64.

You could add a check in this patch same place where you check
the FADT version and be done with this.

Thank you !
Lorenzo

> > this up because I noticed code in core code (not guarded by
> > ifdef) that checks:
> >
> > acpi_gbl_reduced_hardware
> >
> > and if it is not set it goes on instantiating fixed HW devices.
> >
> > drivers/acpi/scan.c:2567
> >
> > It is just a heads up, I think it is harmless, but one thing is
> > removing acpica code on ACPI HW reduced only, the other is to make
> > sure the kernel does not try to use ACPI HW features that acpica
> > can't support.
> 
> I agree, need some cleanups for the ACPICA code I think, and this
> was raised about 1 and half years ago and Intel folks need more
> consideration on that [1].
> 
> [1]: http://comments.gmane.org/gmane.linux.acpi.devel/63844
> 
> Thanks
> Hanjun
> 