[meta-freescale] [PATCH] imx-gpu-viv: Avoid duplicate values in PROVIDES

[Gary Thomas]

On 2015-06-17 08:31, Otavio Salvador wrote:
> On Wed, Jun 17, 2015 at 11:25 AM, Gary Thomas <gary at mlbassoc.com> wrote:
>> On 2015-06-17 08:06, Otavio Salvador wrote:
>>>
>>> On Tue, Jun 16, 2015 at 10:18 AM, Gary Thomas <gary at mlbassoc.com> wrote:
>>>>
>>>> Additional packages are added to the PROVIDES variable, depending on
>>>> the type of processor being supported, e.g. i.MX6Q, i.MX6SX, etc.
>>>> The old way this was being set could add multiple versions of the
>>>> same package to the PROVIDES variable which in turn can cause confusion
>>>> with sstate, e.g. a number of i.MX6 machines use :mx6q:mx6dl in their
>>>> overrides whereas others might only have :mx6q.  With the previous
>>>> setting of PROVIDES, this package would not be sharable via sstate
>>>> between such machines.
>>>>
>>>> This patch changes the process so that these extra packages are only
>>>> added once, improving the [re]usability of sstate.
>>>>
>>>> Signed-off-by: Gary Thomas <gary at mlbassoc.com>
>>>
>>>
>>> The patch looks very good but it fails to apply on master; please
>>> rebase and send a v2. While on that, please pay attention to the
>>> README file to proper send the subject-prefix for the respective
>>> layer.
>>>
>>
>> The patch was based on meta-fsl-arm master(f137cc67f8) which seems
>> to be the latest version.
>>
>> What am I missing?
>
> Applying: imx-gpu-viv: Avoid duplicate values in PROVIDES
> Using index info to reconstruct a base tree...
> error: patch failed: recipes-graphics/imx-gpu-viv/imx-gpu-viv.inc:11
> error: recipes-graphics/imx-gpu-viv/imx-gpu-viv.inc: patch does not apply
> Did you hand edit your patch?
> It does not apply to blobs recorded in its index.
> Cannot fall back to three-way merge.
> Patch failed at 0001 imx-gpu-viv: Avoid duplicate values in PROVIDES
> The copy of the patch that failed is found in:
>     /home/otavio/src/yocto/sources/meta-fsl-arm/.git/rebase-apply/patch
> When you have resolved this problem, run "git am --continue".
> If you prefer to skip this patch, run "git am --skip" instead.
> To restore the original branch and stop patching, run "git am --abort".
>
> it seems to not been based on a publish hash. You might have a local
> commit not merged yet. Please try to rebase and resend.
>

I'm still confused - it applies just fine for me:

gthomas at Zeus:/tmp$ git clone git://git.yoctoproject.org/meta-fsl-arm
Cloning into 'meta-fsl-arm'...
remote: Counting objects: 10025, done.
remote: Compressing objects: 100% (3426/3426), done.
remote: Total 10025 (delta 6192), reused 10014 (delta 6181)
Receiving objects: 100% (10025/10025), 37.81 MiB | 1.39 MiB/s, done.
Resolving deltas: 100% (6192/6192), done.
Checking connectivity... done.
gthomas at Zeus:/tmp$ cd meta-fsl-arm
gthomas at Zeus:/tmp/meta-fsl-arm$ git am </tmp/0001-imx-gpu-viv-Avoid-duplicate-values-in-PROVIDES.patch
Applying: imx-gpu-viv: Avoid duplicate values in PROVIDES

What am I doing differently from you?

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------