[PATCH v4 4/5] arm: Add [U]EFI runtime services support

Leif Lindholm leif.lindholm at linaro.org
Tue Jan 14 10:25:23 EST 2014


On 14 January 2014 13:26, Arnd Bergmann <arnd at arndb.de> wrote:
>> > > Are you suggesting that I should #ifndef ARM in common code, or that I
>> > > should neglect to document what the common code will do with data it is
>> > > given by UEFI?
>> >
>> > It would probably be good to document the fact that it won't work,
>> > possibly even having a BUG_ON statement in the code for this case.
>>
>> Why?
>>
>> You'll only touch that pointer if you enable CONFIG_ACPI, and if you
>> do you probably want that address. Sounds a bit hostile to throw a BUG
>> in the face of someone who's (for example) just succeeded to get Linux
>> running on a Windows RT device.
>
> But we know that it can't work unless a lot of other things get changed
> in the kernel.

We know using ACPI cannot work without updates to the kernel.
That doesn't mean we need to throw a BUG just because the
firmware tells us a table exists.

>> Although no other architectures supported by UEFI support big-endian,
>> so to be honest, I don't want to have to be the first one to validate
>> that in order to get the basic support into the kernel.
>
> I think there was a project to run UEFI on PowerPC on some stage, though
> I can't find any code now.

That does sound familiar, but there is nothing in the specification.

/
    Leif



More information about the linux-arm-kernel mailing list