should layers define BB_DANGLINGAPPENDS_WARNONLY ?= "1"?

should layers define BB_DANGLINGAPPENDS_WARNONLY ?= "1"?

[Robert P. J. Day]

  perusing the meta-musl layer, and noticed in layer.conf the line:

BB_DANGLINGAPPENDS_WARNONLY ?= "1"

i thought this was a bit odd since (at least to me) it seems like the
sort of thing a *developer* might set in their local.conf to
temporarily deal with a dangling .bbappend file -- it never occurred
to me that it would be appropriate for a layer to define that value
itself as an inherent part of its own ... layerness.

  a quick grep of the couple dozen layers i have checked out revealed
two layers that do that:

./meta-musl/conf/layer.conf:BB_DANGLINGAPPENDS_WARNONLY ?= "1"
./meta-yocto-kernel-extras/meta-kernel-dev/conf/layer.conf:BB_DANGLINGAPPENDS_WARNONLY
?= "true"

  i'm curious what others think of this approach.

rday

-- 

========================================================================
Robert P. J. Day                                 Ottawa, Ontario, CANADA
                        http://crashcourse.ca

Twitter:                                       http://twitter.com/rpjday
LinkedIn:                               http://ca.linkedin.com/in/rpjday
========================================================================

Re: should layers define BB_DANGLINGAPPENDS_WARNONLY ?= "1"?

[Paul Eggleton]

Hi Robert,

On Monday 02 November 2015 12:54:33 Robert P. J. Day wrote:
>   perusing the meta-musl layer, and noticed in layer.conf the line:
> 
> BB_DANGLINGAPPENDS_WARNONLY ?= "1"
> 
> i thought this was a bit odd since (at least to me) it seems like the
> sort of thing a *developer* might set in their local.conf to
> temporarily deal with a dangling .bbappend file -- it never occurred
> to me that it would be appropriate for a layer to define that value
> itself as an inherent part of its own ... layerness.
> 
>   a quick grep of the couple dozen layers i have checked out revealed
> two layers that do that:
> 
> ./meta-musl/conf/layer.conf:BB_DANGLINGAPPENDS_WARNONLY ?= "1"
> ./meta-yocto-kernel-extras/meta-kernel-dev/conf/layer.conf:BB_DANGLINGAPPEND
> S_WARNONLY ?= "true"
> 
>   i'm curious what others think of this approach.

I'd say no, they shouldn't really. As a user of the layer you might not 
realise that it's making this change to your configuration.

Cheers,
Paul

-- 

Paul Eggleton
Intel Open Source Technology Centre

Re: should layers define BB_DANGLINGAPPENDS_WARNONLY ?= "1"?

[Gary Thomas]

On 2015-11-02 10:54, Robert P. J. Day wrote:
>
>    perusing the meta-musl layer, and noticed in layer.conf the line:
>
> BB_DANGLINGAPPENDS_WARNONLY ?= "1"
>
> i thought this was a bit odd since (at least to me) it seems like the
> sort of thing a *developer* might set in their local.conf to
> temporarily deal with a dangling .bbappend file -- it never occurred
> to me that it would be appropriate for a layer to define that value
> itself as an inherent part of its own ... layerness.
>
>    a quick grep of the couple dozen layers i have checked out revealed
> two layers that do that:
>
> ./meta-musl/conf/layer.conf:BB_DANGLINGAPPENDS_WARNONLY ?= "1"
> ./meta-yocto-kernel-extras/meta-kernel-dev/conf/layer.conf:BB_DANGLINGAPPENDS_WARNONLY
> ?= "true"
>
>    i'm curious what others think of this approach.

Sounds dangerous, bordering on incorrect (or irresponsible) to me.
This could easily mask a problem and is not layer specific, i.e.
the warning would then apply to all layers, not just meta-musl.

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

Re: should layers define BB_DANGLINGAPPENDS_WARNONLY ?= "1"?

[Khem Raj]

On Mon, Nov 2, 2015 at 9:54 AM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
>
>   perusing the meta-musl layer, and noticed in layer.conf the line:
>
> BB_DANGLINGAPPENDS_WARNONLY ?= "1"

should be avoided although meta-musl is a bit core in nature.
this is so that it can interface with several different compilers
coming from different layers and also different oe-core releases can
work with same branch of meta-musl.
 one may not have same layers attached to it. Ideally, those patches
should now become part
of the original layers as they have been accepted in gcc upstream and
so on. Hopefully as I will take
this into oe-core in coming months this should disappear.

>
> i thought this was a bit odd since (at least to me) it seems like the
> sort of thing a *developer* might set in their local.conf to
> temporarily deal with a dangling .bbappend file -- it never occurred
> to me that it would be appropriate for a layer to define that value
> itself as an inherent part of its own ... layerness.
>
>   a quick grep of the couple dozen layers i have checked out revealed
> two layers that do that:
>
> ./meta-musl/conf/layer.conf:BB_DANGLINGAPPENDS_WARNONLY ?= "1"
> ./meta-yocto-kernel-extras/meta-kernel-dev/conf/layer.conf:BB_DANGLINGAPPENDS_WARNONLY
> ?= "true"
>
>   i'm curious what others think of this approach.
>
> rday
>
> --
>
> ========================================================================
> Robert P. J. Day                                 Ottawa, Ontario, CANADA
>                         http://crashcourse.ca
>
> Twitter:                                       http://twitter.com/rpjday
> LinkedIn:                               http://ca.linkedin.com/in/rpjday
> ========================================================================
>
> --
> _______________________________________________
> yocto mailing list
> yocto@yoctoproject.org
> https://lists.yoctoproject.org/listinfo/yocto

Re: should layers define BB_DANGLINGAPPENDS_WARNONLY ?= "1"?

[Robert P. J. Day]

On Mon, 2 Nov 2015, Khem Raj wrote:

> On Mon, Nov 2, 2015 at 9:54 AM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
> >
> >   perusing the meta-musl layer, and noticed in layer.conf the line:
> >
> > BB_DANGLINGAPPENDS_WARNONLY ?= "1"
>
> should be avoided although meta-musl is a bit core in nature.
> this is so that it can interface with several different compilers
> coming from different layers and also different oe-core releases can
> work with same branch of meta-musl.
>  one may not have same layers attached to it. Ideally, those patches
> should now become part
> of the original layers as they have been accepted in gcc upstream and
> so on. Hopefully as I will take
> this into oe-core in coming months this should disappear.

  ok, that's fair ... i didn't consider this a major issue, i just
thought it was a bit odd.

rday

-- 

========================================================================
Robert P. J. Day                                 Ottawa, Ontario, CANADA
                        http://crashcourse.ca

Twitter:                                       http://twitter.com/rpjday
LinkedIn:                               http://ca.linkedin.com/in/rpjday
========================================================================