[oe] [meta-oe][PATCH] cpuburn-neon: Upgrade to 2014066 version to fix build break
Koen Kooi
koen at dominion.thruhere.net
Fri Jun 27 14:08:34 UTC 2014
Op 27 jun. 2014, om 15:59 heeft Otavio Salvador <otavio at ossystems.com.br> het volgende geschreven:
> Hello Koen,
>
> On Fri, Jun 27, 2014 at 10:40 AM, Koen Kooi <koen at dominion.thruhere.net> wrote:
>>
>> Op 27 jun. 2014, om 15:09 heeft Otavio Salvador <otavio at ossystems.com.br> het volgende geschreven:
>>
>>> Hello Koen,
>>>
>>> On Fri, Jun 27, 2014 at 6:29 AM, Koen Kooi <koen at dominion.thruhere.net> wrote:
>>>> Lauren Post schreef op 26-06-14 16:37:
>>>>> Replace SRC_URI with new location, md5sum and new name since previous
>>>>> download will not work. Without this fix cpuburn-neon has a broken
>>>>> build.
>>>>
>>>> Why not make the recipe clone the git repo?
>>>
>>> I pondered about this when looking at the issue with Lauren. The
>>> recipe was like this before and this build break is causing issues for
>>> several people using meta-fsl-demos as we include it in our test
>>> images.
>>>
>>> The long term solution should be to rework it (also because it has
>>> extra tests and those are not being included) but I think this fix is
>>> urgent and known to work for now.
>>
>> It's still the wrong way to fix it, rework it to use git://, this will break again when the repo gets updated.
>>
>>> This fix needs to be backported to old branches as well as current
>>> recipe fails to fetch.
>>
>> if you're going to backport it make sure it's the right fix.
>
> Instead we can use, for now, the below URL so it stays working even if
> it is updated.
>
> https://raw.githubusercontent.com/ssvb/cpuburn-arm/dd5c5ba58d2b0b23cfab4a286f9d3f5510000f20/cpuburn-a8.S
>
> Fine for this fix?
A lot less fragile :)
> I will update it to use the clone and also include
> the extra tests (for A8, A9 and A7 plus the JPEG ones) soon.
>
> --
> Otavio Salvador O.S. Systems
> http://www.ossystems.com.br http://code.ossystems.com.br
> Mobile: +55 (53) 9981-7854 Mobile: +1 (347) 903-9750
More information about the Openembedded-devel
mailing list