[Openembedded-architecture] layer.conf and layerindex change proposal

[Paul Eggleton]

On Thu, 09 Jun 2016 23:44:54 Peter Kjellerstedt wrote:
> > -----Original Message-----
> > From: openembedded-architecture-bounces at lists.openembedded.org
> > [mailto:openembedded-architecture-bounces at lists.openembedded.org] On
> > Behalf Of Paul Eggleton
> > Sent: den 9 juni 2016 05:48
> > To: Mark Hatle
> > Cc: openembedded-architecture at lists.openembedded.org
> > Subject: Re: [Openembedded-architecture] layer.conf and layerindex
> > change proposal
> > 
> > On Wed, 08 Jun 2016 18:01:44 Mark Hatle wrote:
> > > On 6/8/16 5:24 PM, Richard Purdie wrote:
> > > > On Wed, 2016-06-08 at 16:15 -0500, Mark Hatle wrote:
> > > >> In addition, many of the layers I work with have a clear set of
> > > >> required layers
> > > >> (LAYERDEPENDS), but also have 'recommended' layers...  So I'd love
> > 
> > to
> > 
> > > >> see a
> > > >> LAYERRECOMMENDS as well.
> > > >> 
> > > >> For this one, in bitbake, we could add the LAYERRECOMMENDS to the
> > > >> priority
> > > >> sorting code -- but if the layer is not included it's not an error
> > 
> > or
> > 
> > > >> warning.
> > > > 
> > > > I don't understand this. Why would the fact a layer recommends
> > 
> > another
> > 
> > > > layer change priority sorting code? I don't think the dependency
> > > > variable does that?
> > > 
> > > In bitbake/lib/bb/cooker.py:
> > > 
> > > 
> > > look around line 1148 and 1178.
> > > 
> > > 
> > > The existing LAYERDEPENDS does influence the priority in some way.
> > 
> > (First
> > 
> > > chunk verifies dependencies, second is a function that actually
> > 
> > calculates
> > 
> > > layer priority based on results of the validation.)  I was actually
> > > surprised to see this code, I didn't realize the system did this in
> > 
> > any
> > 
> > > way...
> > 
> > Heh, it's one of those times when you think "what? who wrote that
> > code?" and then you find out it was yourself 5 years ago. I vaguely
> > remember writing this but not much about the intent, however if you
> > look at the commit message it's supposed to set the priority
> > automatically *only* if the priority isn't specified, and a cursory
> > look at calc_layer_priority() seems to verify that.
> > Here's the commit message:
> > 
> > http://git.openembedded.org/bitbake/commit/?id=115b89fa279b64e79da0f72caf7
> > b30965a83fab1> 
> > There's also a reference to the following bugzilla entry:
> >   https://bugzilla.yoctoproject.org/show_bug.cgi?id=790
> > 
> > Due to copying other layer.conf boilerplate most layers do specify a
> > priority and therefore I suspect this code is rarely exercised in
> > practice.
> 
> Don't bet on it. We rely on this feature for all our layers...
> 
> It is somewhat documented in the manual in the section about
> BBFILE_PRIORITY. However, it just says that if BBFILE_PRIORITY is
> not specified, then the layer's priority is based on layer
> dependencies, but it does not state how. There is a reference to
> LAYERDEPENDS for more information, but its documentation does not
> touch on the subject of priorities.

OK, that's good feedback - I guess what that says is we really should ensure 
that it is properly documented, so I've filed this bug:

  https://bugzilla.yoctoproject.org/show_bug.cgi?id=9753 

We should probably ensure it is covered by the automated tests as well.

Cheers,
Paul

-- 

Paul Eggleton
Intel Open Source Technology Centre