Re: [edk2-devel] [PATCH 5/5] ArmPkg/ArmMmuLib: drop unused TT_ATTR_INDX_INVALID CPP symbol

["Ard Biesheuvel" <ard.biesheuvel@arm.com>]

On 4/2/20 12:23 PM, Leif Lindholm via groups.io wrote:
> On Sat, Mar 28, 2020 at 11:43:21 +0100, Ard Biesheuvel wrote:
>> TT_ATTR_INDX_INVALID is #define'd but never used so drop it.
>>
>> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
>> ---
>>   ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c | 3 ---
>>   1 file changed, 3 deletions(-)
>>
>> diff --git a/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c b/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c
>> index a82596d290f1..222ff817956f 100644
>> --- a/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c
>> +++ b/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c
>> @@ -19,9 +19,6 @@
>>   #include <Library/BaseLib.h>
>>   #include <Library/DebugLib.h>
>>   
>> -// We use this index definition to define an invalid block entry
>> -#define TT_ATTR_INDX_INVALID    ((UINT32)~0)
>> -
> 
> Since this is separately defined also in
> ArmPkg/Drivers/CpuDxe/AArch64/Mmu.c (£"$%^£"$?!?!?)
> could this patch be tweaked to instead "consolidate" the definitions
> to a central location, like ArmPkg/Include/Chipset/AArch64Mmu.h where
> all of the other TT_ATTR_INDX_ definitions live?
> 

That would imply that this value is somehow architected, which it is not.

ArmPkg/Drivers/CpuDxe/AArch64/Mmu.c is the only remaining user of this 
constant, and it only has meaning within the context of the routines 
therein.