lkml.org 
[lkml]   [2017]   [Nov]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH v2 25/35] nds32: Build infrastructure
On Wed, Nov 29, 2017 at 10:10 AM, Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
> Hi Arnd,
>
> On Wed, Nov 29, 2017 at 9:58 AM, Arnd Bergmann <arnd@arndb.de> wrote:
>> On Wed, Nov 29, 2017 at 9:39 AM, Greentime Hu <green.hu@gmail.com> wrote:
>>> 2017-11-27 22:21 GMT+08:00 Arnd Bergmann <arnd@arndb.de>:
>>>> On Mon, Nov 27, 2017 at 1:28 PM, Greentime Hu <green.hu@gmail.com> wrote:
>>>>> diff --git a/arch/nds32/Kconfig.cpu b/arch/nds32/Kconfig.cpu
>>>>> +config CPU_CACHE_NONALIASING
>>>>> + bool "Non-aliasing cache"
>>>>> + help
>>>>> + If this CPU is using VIPT data cache and its cache way size is larger
>>>>> + than page size, say N. If it is using PIPT data cache, say Y.
>>>>> +
>>>>> + If unsure, say Y.
>>>>
>>>> Can you determine this from the CPU type?
>>>
>>> There is no cpu register to determine it. It also depeneds on page
>>> size and way size however page size is configurable by software.
>>> These codes are determined at compile time will be benefit to code
>>> size and performance.
>>> IMHO, I think it would be better to be determined here.
>>
>> I meant determining it at compile time from other Kconfig symbols,
>> if that's possible. Do the CPU cores each have a fixed way-size?
>> If they do, it could be done like
>>
>> menu "CPU selection"
>>
>> config CPU_N15
>> bool "AndesCore N15"
>> select CPU_CACHE_NONALIASING
>>
>> config CPU_N13
>> bool "AndesCore N15"
>> select CPU_CACHE_NONALIASING if PAGE_SIZE_16K
>>
>> ...
>>
>> endmenu
>>
>> and then you can use the same CPU_... symbols to make other decisions
>> as well, e.g. CPU specific compiler optimizations.
>
> Do you want to support multiple CPU types in a single kernel image
> (I see no "choice" statement above)?
> If yes, you may have a mix of aliasing and non-aliasing caches, so
> you may want to invert the logic, and select CPU_CACHE_ALIASING
> instead.

Right, my mistake.

Arnd

\
 
 \ /
  Last update: 2017-11-29 10:26    [W:0.104 / U:1.664 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site