On Fri, Apr 17, 2020 at 09:02:30AM -0700, akuster808 wrote:
>
>
> On 4/17/20 5:11 AM, Mikko Rapeli wrote:
> > RFC: currently unconditionally, but this may need to be changed because
> > so many patches are not passing this check. Even upstream backports
> > from Linus Torvalds are failing the check sometimes.
> >
> > This is test is important to keep all kernel patches in upstreamable state.
> > It would help keeping BSP and project specific kernel recipes in better
> > shape.
> > It will not help if BSP git trees contain 1000's of non-upstreamed
> > patches. Passing this script is mandatory for patches according to
> > https://www.kernel.org/doc/html/latest/process/submit-checklist.html
>
> I am not opposed to having more checks, I worry about the impact on build.
>
> Maybe have this in a separate class so folks can opt-in or out depending
> on final conclusion of this RFC.
Agreed. In theory this would belong to insane.bbclass, or would only be
added if some extra variable or similar is set in the recipes using
kernel.bbclass, or a separate class if you will. I would need some pointers
here.
I'm also playing with sparse so that I can easily do sparse checker builds of
kernels from different BSPs. Will hopefully use similar approach with that
too...
-Mikko
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#137276):
https://lists.openembedded.org/g/openembedded-core/message/137276
Mute This Topic: https://lists.openembedded.org/mt/73075458/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
