[GIT PULL v2][for 3.17] pull request for hisilicon terminal soc

Olof Johansson olof at lixom.net
Wed Jul 30 23:14:18 PDT 2014


On Wed, Jul 30, 2014 at 11:08 PM, xuwei <xuwei5 at hisilicon.com> wrote:
>
>
> On 2014/7/31 13:32, Olof Johansson wrote:
>> Hi,
>>
>> There were a few things that still needed fixing here -- I took care
>> of most of them. See below.
>>
>>
>> On Tue, Jul 29, 2014 at 12:44 AM, xuwei <xuwei5 at hisilicon.com> wrote:
>>> Hi Olof, Hi Kevin, Hi Arnd,
>>>
>>> Please help to merge pull request for hisilicon terminal soc.
>>>
>>> Best Regards,
>>> Wei
>>>
>>>
>>> The following changes since commit 7171511eaec5bf23fb06078f59784a3a0626b38f:
>>>
>>>   Linux 3.16-rc1 (2014-06-15 17:45:28 -1000)
>>>
>>> are available in the git repository at:
>>>
>>>   git://github.com/hisilicon/linux-hisi.git tags/hix5hd2-for-3.17-v2
>>>
>>> for you to fetch changes up to eeea1ab9b53f311f634e87d9208daa564a97dbba:
>>>
>>>   ARM: hisi: remove smp from machine descriptor (2014-07-29 14:54:11 +0800)
>>>
>>> ----------------------------------------------------------------
>>> ARM: mach-hisi: Hisilicon teminal soc udaptes for 3.17
>>>
>>> - Rename Hi3716 to HiX5HD2
>>> HiX5HD2 is a SoC with dual Cortex A9 cores for STB market.
>>> And original Hi3xxx SoC is for mobile market.
>>> To avoid confusing the two types of SoCs and also because of
>>> different implementation(SMP, IPs integarted and earlycon configure),
>>> rename Hi3716 to HIX5HD2 ann add ARCH_HIX5HD2.
>>>
>>> - Add HiX5HD2 STB SoC and development kit board support
>>> - Change to use power reset driver to reboot for HiX5HD2
>>> - Add ARCH_HISI for the common configure of hisilicon SoCs
>>> - Change to use CPU_METHOD_OF_DECLARE for SMP on HiX5HD2 and hi3xxx
>>> - Enable L2 aux value for Hi3620
>>>
>>> ----------------------------------------------------------------
>>> Haifeng Yan (3):
>>>       ARM: debug: Rename Hi3716 to HIX5HD2
>>>       ARM: hisi: enable hix5hd2 SoC
>>>       ARM: dts: Add hix5hd2-dkb dts file.
>
> Hi Olof,
>
>> It is confusing -- is the SoC named HIX5HD2 or Hisilicon X5HD2? Seems
>> like there are references to both.
>
> The full name is Hisilicon X5HD2.
> But we also call it HIX5HD2.

Ok. I think for consistency, it's better to call it X5HD2 everywhere
in the kernel where it's already prefixed by Hisilicon, such as
compatible strings and help texts on config options.

>> What I did was that I renamed the dts and dtsi to hisi-x5hd2  to have
>> the vendor/family as prefix. Given the few boards per SoC family you
>> have had so far, it makes more sense to prefix it with hisi.
>>
>
> It is OK for us.
> And we will follow the rule in the future.
>
>>> Haojian Zhuang (5):
>>>       ARM: hisi: add ARCH_HISI
>>>       power: reset: move hisilicon reboot code
>>>       ARM: config: enable ARCH_HIX5HD2
>>>       ARM: hisi: set l2 aux for pl310
>>
>> This patch makes no sense to still carry. If things still need to be
>> resolved, then it should just be dropped.
>>
>>>       ARM: hisi: remove smp from machine descriptor
>>
>>
>> Also, in general: Wei, you seem to have added your Acked-by instead of
>> Signed-off-by on the commits. When you apply them, you need to do so
>> with your sign off.
>
> Do you mean I need add both "Acked-by" and "Signed-off-by"?or just add "Signed-off-by"?

Just S-o-b. See Documentation/SubmittingPatches.txt for more
explanation on what the different tags mean.

>> What I have done is that I have applied most of the above patches
>> directly instead of merging the branch, and I have done the touchups
>> described above (dropped the L2 patch and renamed the dts/dtsi).
>>
>> If for some reason, the name really should be hisi-hix5hd2, then let
>> me know quickly so I can revisit before it goes in. But I don't thik
>> so -- I find more hits with "x5hd2" out there than with "hix5hd2". If
>> anything, the remaining traces of hix5hd2 should probably be renamed
>> too (machine descriptor, etc).
>
> It is OK with "x5hd2".
>
>> (I also fixed a few spelling errors in the patch descriptions here and
>> there, etc).
>
> Thanks!
>
> Best Regards,
> Wei


-Olof



More information about the linux-arm-kernel mailing list