[Feature Request] A test for ACPI _REV

[Alex Hung]

Perhaps we can develop an OS detection test - it will look up _OSI and
_REV in AML and report accordingly.

If we found out other new creative approaches, we could add to this test, too.

On Wed, Mar 18, 2015 at 5:11 PM, Colin Ian King
<colin.king at canonical.com> wrote:
> On 18/03/15 09:04, Alex Hung wrote:
>> It starts with a recent patch by Matthew Garrett:
>> http://comments.gmane.org/gmane.linux.acpi.devel/74165
>>
>> I have seen a lot _REV uses for OS detect in many major OEM systems,
>> including but not limiting to Dell and HP in Matthew's comments. It is
>> probably a know-how among BIOS community. However, Matthew's patch
>> will certainly change the game after it gets into kernel.
>>
>> Today I have receive a feature request as below:
>> =========================================================
>> ... make sure that we are aware of this when there is a test for _REV
>> included in our BIOSes.  I'd like to make sure that if _REV is found
>> in the BIOS to be tested that it at least starts a discussion to
>> understand what it's doing and if it's the right thing for the system.
>> =========================================================
>>
>> I am looking for some discussion and feedback from FWTS community
>> whether this should be included in FWTS.
>
> I'd prefer it if fwts could check for this directly rather than
> indirectly trying to scan for kernel messages.
>
>>
>> Alternatively, it may be possible to modify kernel to generate a
>> message like _OSI:
>>     [Firmware Bug]: ACPI: BIOS _OSI(Linux) query ignored
>> and fwts can try to catch it in klog test, assuming kernel can include
>> this patch.
>>
>



-- 
Cheers,
Alex Hung