Mike,

I think Robin preferred to have kernel support application L1 stack by
default in former discussion.


Sonic
 

>-----Original Message-----
>From: [email protected] 
>[mailto:[email protected]] On 
>Behalf Of Mike Frysinger
>Sent: Wednesday, June 30, 2010 3:24 AM
>To: [email protected]
>Cc: [email protected]
>Subject: Re: [Linux-kernel-commits] [8953] 
>trunk/arch/blackfin: blackfin:always permit l1 app stack if l1 
>exception stack is not used
>
>On Tue, Jun 29, 2010 at 05:39,  <[email protected]> wrote:
>> Revision 8953 Author bhsong Date 2010-06-29 05:39:11 -0400 (Tue, 29 
>> Jun
>> 2010)
>>
>> Log Message
>>
>> blackfin: always permit l1 app stack if l1 exception stack 
>is not used
>>
>> -config APP_STACK_L1
>> -    bool "Support locating application stack in L1 Scratch Memory"
>> -    default y
>> -    help
>> -      If enabled the application stack can be located in L1
>> -      scratch memory (less latency).
>> -
>> -      Currently only works with FLAT binaries.
>> -
>>  config EXCEPTION_L1_SCRATCH
>>      bool "Locate exception stack in L1 Scratch Memory"
>>      default n
>> -    depends on !APP_STACK_L1
>>      help
>>        Whenever an exception occurs, use the L1 Scratch memory for
>>        stack storage.  You cannot place the stacks of FLAT binaries
>
>the resulting ifdefs are confusing.  the old way is clear:
>#ifdef CONFIG_APP_STACK_L1
>    /* do stuff related to having app stack in L1 scratch */ #endif
>
>the new way is unclear and it complicates adding new options 
>related to L1 scratchpad usage
>
>if you dont want the Kconfig option to be shown, then make it hidden:
>config APP_STACK_L1
>       bool
>       default y
>       depends !EXCEPTION_L1_SCRATCH && !CONFIG_SMP
>
>that said, i dont see what's wrong with having the Kconfig be 
>explicit in the first place -mike 
>_______________________________________________
>Linux-kernel-commits mailing list
>[email protected]
>https://blackfin.uclinux.org/mailman/listinfo/linux-kernel-commits
>
_______________________________________________
Linux-kernel-commits mailing list
[email protected]
https://blackfin.uclinux.org/mailman/listinfo/linux-kernel-commits

Reply via email to