Re: [RFC PATCH 1/2] efi: Make 'efi_enabled' a function to query EFI facilities

[Tim Gardner <rtg.canonical@gmail.com>]

On 01/04/2013 09:15 AM, Matt Fleming wrote:
> On Fri, 2013-01-04 at 08:08 -0700, Tim Gardner wrote:
>> On 01/03/2013 06:18 AM, Matt Fleming wrote:
>>> From: Matt Fleming <matt.fleming@intel.com>
>>>
>>
>> snip
>>
>>>   /*
>>> - * We play games with efi_enabled so that the compiler will, if possible, remove
>>> - * EFI-related code altogether.
>>> + * We play games with efi_enabled so that the compiler will, if
>>> + * possible, remove EFI-related code altogether.
>>>    */
>>> +#define EFI_BOOT		0x00000001 /* Were we booted from EFI? */
>>> +#define EFI_SYSTEM_TABLES	0x00000002 /* Can we use EFI system tables? */
>>> +#define EFI_CONFIG_TABLES	0x00000004 /* Can we use EFI config tables? */
>>> +#define EFI_RUNTIME_SERVICES	0x00000004 /* Can we use runtime services? */
>>> +#define EFI_MEMMAP		0x00000008 /* Can we use EFI memory map? */
>>> +#define EFI_64BIT		0x00000010 /* Is the firmware 64-bit? */
>>> +
>>
>> Your use of test_bit() and set_bit() imply that these macros should be
>> bit numbers, not bit masks. It'll work until you define a mask with an
>> integer value greater then 31.
>
> They're not intended to be bitmasks in the sense that no two bits are
> set in each constant (and I am aware of the upper limit).
>
> I have no problem changing the above values to bit numbers if that would
> be less confusing.
>

When you do change them to bit numbers you should also note that 
EFI_CONFIG_TABLES and EFI_RUNTIME_SERVICES have the same value, which I 
believe is in error.

rtg
-- 
Tim Gardner tim.gardner@canonical.com