Re: Build error while following Appendix A. Yocto Project Development manual

[Tom Zanussi <tom.zanussi@intel.com>]

On Tue, 2011-11-01 at 15:47 -0700, Robert P. J. Day wrote:
> some questions/observations as i'm getting close to the end of this
> appendix (some of which you've already seen and addressed), all done
> on my 64-bit ubuntu 11.10 system:
> 
>   * section A.3: rather than force the reader to follow a link or go
> to an earlier section, why not just supply the command to check out
> meta-intel/?
> 
>   $ git clone http://git.yoctoproject.org/cgit/cgit.cgi/meta-intel
> 

Makes sense to me.  Also, the same comment would apply to the main poky
repo as well.

>   * section A.5.1: is it actually *necessary* to delete the
> crownbay.conf file, and to change the comments in the mymachine.conf
> file?  admittedly, it's nice to keep things consistent, but if the
> text claims that something is "needed", then it really should be
> required.

No, not strictly necessary, but assuming you're using these instructions
to create a new BSP, definitely desirable.

> 
>   * section A.5.2.1: as we've noted, that first command can be tossed.
> in addition, should one remove the "crownbay" directory from that
> location?  since "crownbay" content is being removed from other
> locations, it would be nice to be consistent.
> 
>   * section A.5.2.2: delete first command
> 

I think that's right.

>   * section A.5.2.4: the "SRCREV_meta" values listed in the appendix
> don't match what's in the actual file.  so what are the correct
> values?
> 

Well, this example seems to assume you want to base it on the atom-pc
branch, so I guess we should use the value for that branch in meta-yocto
(meta-yocto/recipes-kernel/linux/linux-yocto_3.0.bbappend).  The meta
branch SRCREV should be inherited from the main recipe
(meta/recipes-kernel/linux/linux-yocto_3.0.bb) which looks like it
matches the current value and doesn't need to be changed.

Or at least that's what I'm building now.  I'll post the full transcript
assuming I get a good build out of it...

Tom

>   ... back to work ...
> 
> rday
>