2017-11-17 21:44:22

by Greentime Hu

[permalink] [raw]
Subject: Re: [PATCH 26/31] nds32: Build infrastructure

2017-11-10 16:26 GMT+08:00 Greentime Hu <[email protected]>:
> 2017-11-09 18:33 GMT+08:00 Arnd Bergmann <[email protected]>:
>> On Thu, Nov 9, 2017 at 10:02 AM, Greentime Hu <[email protected]> wrote:
>>> 2017-11-08 18:16 GMT+08:00 Arnd Bergmann <[email protected]>:
>>>> On Wed, Nov 8, 2017 at 6:55 AM, Greentime Hu <[email protected]> wrote:
>>
>>>>> +config GENERIC_CALIBRATE_DELAY
>>>>> + def_bool y
>>>>
>>>> It's better to avoid the delay loop completely and skip the calibration,
>>>> if your hardware allows.
>>>
>>> Thanks.
>>> Do you mean that this config should be def_bool n?
>>> why? Almost all arch enable it.
>>
>> It depends on what your hardware can do. If you have a way to see how much
>> time has passed that is guaranteed to be reliable on all machines, then
>> use that instead.
>>
>> On a lot of architectures, it's not possible, so they have to fall back to using
>> the delay loop.
>
> I get it. I will discuss it with our HW colleagues.
> We may get these informations in some registers.

Hi, Arnd:

I think I can't set it to default n because it will be called in start_kernel.

start_kernel() -> calibrate_delay()

If I don't enable this config, it will link error because it didn't
build init/calibrate.c

From 1584341045984523442@xxx Fri Nov 17 19:02:43 +0000 2017
X-GM-THRID: 1583483358934427114
X-Gmail-Labels: Inbox,Category Forums,HistoricalUnread


2017-11-17 18:38:07

by Arnd Bergmann

[permalink] [raw]
Subject: Re: [PATCH 26/31] nds32: Build infrastructure

On Fri, Nov 17, 2017 at 1:39 PM, Greentime Hu <[email protected]> wrote:
> 2017-11-10 16:26 GMT+08:00 Greentime Hu <[email protected]>:
>> 2017-11-09 18:33 GMT+08:00 Arnd Bergmann <[email protected]>:
>>> On Thu, Nov 9, 2017 at 10:02 AM, Greentime Hu <[email protected]> wrote:
>>>> 2017-11-08 18:16 GMT+08:00 Arnd Bergmann <[email protected]>:
>>>>> On Wed, Nov 8, 2017 at 6:55 AM, Greentime Hu <[email protected]> wrote:
>>>
>>>>>> +config GENERIC_CALIBRATE_DELAY
>>>>>> + def_bool y
>>>>>
>>>>> It's better to avoid the delay loop completely and skip the calibration,
>>>>> if your hardware allows.
>>>>
>>>> Thanks.
>>>> Do you mean that this config should be def_bool n?
>>>> why? Almost all arch enable it.
>>>
>>> It depends on what your hardware can do. If you have a way to see how much
>>> time has passed that is guaranteed to be reliable on all machines, then
>>> use that instead.
>>>
>>> On a lot of architectures, it's not possible, so they have to fall back to using
>>> the delay loop.
>>
>> I get it. I will discuss it with our HW colleagues.
>> We may get these informations in some registers.
>
> Hi, Arnd:
>
> I think I can't set it to default n because it will be called in start_kernel.
>
> start_kernel() -> calibrate_delay()
>
> If I don't enable this config, it will link error because it didn't
> build init/calibrate.c

You will have to provide an architecture-specific implementation of
this function, please ahve a look at what others are doing, e.g. tile,
openrisc or
h8300.

Arnd

From 1584229833824536820@xxx Thu Nov 16 13:35:03 +0000 2017
X-GM-THRID: 1583483358934427114
X-Gmail-Labels: Inbox,Category Forums,HistoricalUnread