pull request for atomic

yangwei weiyang wei.a.yang at gmail.com
Thu Aug 16 15:56:13 UTC 2012


Hi all

I have submitted a new pull request for PowerPC atomic support and
seven new atomic test cases for UP architecture. The address is
https://github.com/RTEMS/rtems/pull/5. Any comments are welcome!

2012/8/1 yangwei weiyang <wei.a.yang at gmail.com>:
> 2012/8/1 Gedare Bloom <gedare at rtems.org>:
>> I'm confused. I thought I read in the documentation that if an
>> operation is unsupported by an architecture then the operation does
>> nothing. Instead are you saying you would like to check at
> Right, If an operation is not supported by an architecture it will do
> nothing. But if an atomic operation implementation differs according
> to its architecture configure, such as for powerpc and powerpc-64 its
> 64-bit atomic implementation will differ.
>> compile-time whether 64-bit atomics are supported somehow?
>>
> I do not mean check whether 64-bit atomic operation are supported,
> instead of compiling different atomic implementation.
>
>> Off the top of my head I do not know if that capability exists.
>>
>> -Gedare
>>
>> On Tue, Jul 31, 2012 at 9:16 PM, yangwei weiyang <wei.a.yang at gmail.com> wrote:
>>> Hi Gedare,
>>>
>>> First really thank you for your detailed comments. I have modified all
>>> the error you pointed. And there still a question about the 64-bit
>>> support for atomic operations. In my opinion for a architecture it
>>> maybe have 64-bit support or not. So there should be a macro to
>>> indicate whether it supports 64-bit. Is there a macro in RTEMS?
>>>
>>> 2012/7/31 Gedare Bloom <gedare at rtems.org>:
>>>> I have made some comments on your new version. If you have further
>>>> questions or answers to any questions I put please send an email I am
>>>> more likely to see it.
>>>>
>>>> -Gedare
>>>>
>>>> On Wed, Jul 25, 2012 at 2:00 AM,  <wei.a.yang at gmail.com> wrote:
>>>>>
>>>>> 在 2012-7-25,2:42,Gedare Bloom <gedare at rtems.org> 写道:
>>>>>
>>>>>> On Tue, Jul 24, 2012 at 12:29 PM, yangwei weiyang <wei.a.yang at gmail.com> wrote:
>>>>>>> 2012/7/23 Gedare Bloom <gedare at rtems.org>:
>>>>>>>> On Mon, Jul 23, 2012 at 4:45 AM, yangwei weiyang <wei.a.yang at gmail.com> wrote:
>>>>>>>>> 2012/7/23 Sebastian Huber <sebastian.huber at embedded-brains.de>:
>>>>>>>>>> Hi,
>>>>>>>>>>
>>>>>>>>>> is it possible to squash the commits together?  This makes it easier to
>>>>>>>>>> review.
>>>>>>>>> Hi Sebastian, firstly thank you for your comments. The commits are a
>>>>>>>>> little trivial because all commits are submitted step by step. And i
>>>>>>>>> do not know how to make a pull request on github with a single patch
>>>>>>>>> from first commit to last commit. Or i can generate a patch manually
>>>>>>>>> and send it to mail list.
>>>>>>>>
>>>>>>>> git rebase -i
>>>>>>>>
>>>>>>>> should let you specify revision range and squash commits together.
>>>>>>>> then you can push a new branch and make new request.
>>>>>>> Hi Gedare, Could you tell me more details about this command.
>>>>>>> I use git rebase master atomic but all commits do not squash together.
>>>>>>
>>>>>> http://git-scm.com/book/en/Git-Tools-Rewriting-History#Squashing-Commits
>>>>> Hi Gedare. Thank you for your material, it is very useful for me. And I have squash all the commits and create a new branch atomic-review to pull request. Any comments are welcome.
>>>>>>
>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Since you wrote the files probably on your own they should have your
>>>>>>>>>> copyright.
>>>>>>>>>>
>>>>>>>>> OK, i will add the copyright to the appropriate files.
>>>>>>>>>> The file "cpukit/score/cpu/i386/rtems/score/atomic_cpu.h" should include a
>>>>>>>>>> generic file e.g. "cpukit/score/include/rtems/score/genericcpuatomic.h" with
>>>>>>>>>> the type definitions.  I think we can use this generic file on every CPU
>>>>>>>>>> port.
>>>>>>>>>>
>>>>>>>>> Yeah, agree. And RTEMS support so many architectures so i will think
>>>>>>>>> about what types should be generic to all cpu architecture.
>>>>>>>>>> I suggest to use the following type names:
>>>>>>>>>>
>>>>>>>>>> Atomic_Int
>>>>>>>>>> Atomic_Long
>>>>>>>>>> Atomic_Int32
>>>>>>>>>> Atomic_Int64
>>>>>>>>>> Atomic_Pointer
>>>>>>>>>>
>>>>>>>>> OK. and which file follow this rule that i can study the name rule on RTEMS?
>>>>>>>>>> I suggest to rename "atomic_cpu.h" into "cpuatomic.h".  Currently there is
>>>>>>>>>> no header file in the score with a "_".
>>>>>>>>>>
>>>>>>>>> OK.
>>>>>>>>>>
>>>>>>>>>> On 07/22/2012 06:08 PM, yangwei weiyang wrote:
>>>>>>>>>>>
>>>>>>>>>>> Hi all,
>>>>>>>>>>>
>>>>>>>>>>> I have made a pull request for review the source code of atomic, and
>>>>>>>>>>> could you give me any comments? thank you!
>>>>>>>>>>>
>>>>>>>>>>> pull request link is blow:
>>>>>>>>>>> https://github.com/RTEMS/rtems/pull/1
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> Sebastian Huber, embedded brains GmbH
>>>>>>>>>>
>>>>>>>>>> Address : Obere Lagerstr. 30, D-82178 Puchheim, Germany
>>>>>>>>>> Phone   : +49 89 18 90 80 79-6
>>>>>>>>>> Fax     : +49 89 18 90 80 79-9
>>>>>>>>>> E-Mail  : sebastian.huber at embedded-brains.de
>>>>>>>>>> PGP     : Public key available on request.
>>>>>>>>>>
>>>>>>>>>> Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> rtems-devel mailing list
>>>>>>>>>> rtems-devel at rtems.org
>>>>>>>>>> http://www.rtems.org/mailman/listinfo/rtems-devel
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Wei Yang
>>>>>>>>> Best Regards
>>>>>>>>>
>>>>>>>>> wei.a.yang at gmail.com
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> rtems-devel mailing list
>>>>>>>>> rtems-devel at rtems.org
>>>>>>>>> http://www.rtems.org/mailman/listinfo/rtems-devel
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Wei Yang
>>>>>>> Best Regards
>>>>>>>
>>>>>>> wei.a.yang at gmail.com
>>>
>>>
>>>
>>> --
>>> Wei Yang
>>> Best Regards
>>>
>>> wei.a.yang at gmail.com
>
>
>
> --
> Wei Yang
> Best Regards
>
> wei.a.yang at gmail.com



-- 
Wei Yang
Best Regards

wei.a.yang at gmail.com




More information about the devel mailing list