What to do about the poor bitbake Quality Control?

What to do about the poor bitbake Quality Control?

[AJ ONeal]

This is about the 4th time that I've `rm -rf`-d everything and started from
scratch and never yet got a working `bitbake omap3-console-image`

Every time I `git pull` it's a different set of problems. It's never the
same package twice, but they never all compile for the basic console image
anywhere from 4 to 20 hours in.

Fortunately, I do have a machine that did build omap3-console-image a few
weeks ago at work. But I've never been able to get it on my machine at home
(same os) to start playing around with.

What needs to happen in order for things to git checked that they work with
at least a few common meta-tasks before being pushed out?
And is there anything I can do to help? (Donate some machine time to run
tests, for example)


Obviously, we all have our day jobs and whatnot, but this is becoming really
frustrating for me and I'd like to see things improve.

AJ ONeal

Re: What to do about the poor bitbake Quality Control?

[J. L.]

I agree with you AJ, its become a serious pain to just get a built
image. I myself have only built once completely and have yet to again
since figuring out the changes I needed to make to that image. I would
like to help myself but someone like me whose not very competent in
this area finds the documentation a tad lacking to properly learn the
ways to help. I still have yet to figure out how to even get my own
program from the net to build with OE as just following the hello
world does not seem to be enough for me to grasp how to do it with an
actual problem. Which runs me into my next issue I can not properly
modify the way some programs build to get the features I want.

I have also noticed on the gumstix boards while most of the time it
seems its the same errors people get, every once in awhile you see
some people with one issue and others with another.

I am sure this wont be of much use as a post but figured I would chime
in from an inexperienced beginner trying to get a footing in here and
to actually be able to use things properly.



On Sat, Aug 14, 2010 at 8:46 AM, AJ ONeal <coolaj86@gmail.com> wrote:
> This is about the 4th time that I've `rm -rf`-d everything and started from
> scratch and never yet got a working `bitbake omap3-console-image`
>
> Every time I `git pull` it's a different set of problems. It's never the
> same package twice, but they never all compile for the basic console image
> anywhere from 4 to 20 hours in.
>
> Fortunately, I do have a machine that did build omap3-console-image a few
> weeks ago at work. But I've never been able to get it on my machine at home
> (same os) to start playing around with.
>
> What needs to happen in order for things to git checked that they work with
> at least a few common meta-tasks before being pushed out?
> And is there anything I can do to help? (Donate some machine time to run
> tests, for example)
>
>
> Obviously, we all have our day jobs and whatnot, but this is becoming really
> frustrating for me and I'd like to see things improve.
>
> AJ ONeal
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

Re: What to do about the poor bitbake Quality Control?

[AJ ONeal]

I've been compiling some common general issues
http://fastr.github.com/articles/Troubleshooting-bitbake.html

Here's some quick-n-dirty instructions akin to using devshell:


   1. cd ${OVEROTOP}/tmp/work/${ARCH}/${PACKAGE}_${VER}_r${REV}/${BUILD}
      - ${ARCH} is probably armv7a-angstrom-linux-gnueabi
      - ${BUILD} is a directory which is not NOT src or temp such as the
      package name or git or trunk or svn
      - ${VER} and ${REV} - duh.
   2. `cp ../temp/run.do_compile.${OLD_PID}
      - ${OLD_PID} is a number like 3597 or 4352
      - do_compile could be any task - do_install, etc
   3. vim run.do_compile.1234
      1. comment out do_compile() (or whatever)
      2. add bash --norc
   4. ./run.do_compile.1234 will put you in an environment with all
   variables set
   5. make (or whatever) to try to build, debug issues
   6. exit (when done to go back to shell without run.do_compile settings)

This is based on the tip on openembedded wiki, but it took me a while to
understand what theirs meant so I cleaned it up in a fashion which is
hopefully crystal clear.

AJ ONeal

On Sat, Aug 14, 2010 at 10:16 AM, J. L. <vwyodapink@gmail.com> wrote:

> I agree with you AJ, its become a serious pain to just get a built
> image. I myself have only built once completely and have yet to again
> since figuring out the changes I needed to make to that image. I would
> like to help myself but someone like me whose not very competent in
> this area finds the documentation a tad lacking to properly learn the
> ways to help. I still have yet to figure out how to even get my own
> program from the net to build with OE as just following the hello
> world does not seem to be enough for me to grasp how to do it with an
> actual problem. Which runs me into my next issue I can not properly
> modify the way some programs build to get the features I want.
>
> I have also noticed on the gumstix boards while most of the time it
> seems its the same errors people get, every once in awhile you see
> some people with one issue and others with another.
>
> I am sure this wont be of much use as a post but figured I would chime
> in from an inexperienced beginner trying to get a footing in here and
> to actually be able to use things properly.
>
>
>
> On Sat, Aug 14, 2010 at 8:46 AM, AJ ONeal <coolaj86@gmail.com> wrote:
> > This is about the 4th time that I've `rm -rf`-d everything and started
> from
> > scratch and never yet got a working `bitbake omap3-console-image`
> >
> > Every time I `git pull` it's a different set of problems. It's never the
> > same package twice, but they never all compile for the basic console
> image
> > anywhere from 4 to 20 hours in.
> >
> > Fortunately, I do have a machine that did build omap3-console-image a few
> > weeks ago at work. But I've never been able to get it on my machine at
> home
> > (same os) to start playing around with.
> >
> > What needs to happen in order for things to git checked that they work
> with
> > at least a few common meta-tasks before being pushed out?
> > And is there anything I can do to help? (Donate some machine time to run
> > tests, for example)
> >
> >
> > Obviously, we all have our day jobs and whatnot, but this is becoming
> really
> > frustrating for me and I'd like to see things improve.
> >
> > AJ ONeal
> > _______________________________________________
> > Openembedded-devel mailing list
> > Openembedded-devel@lists.openembedded.org
> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
> >
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

Re: What to do about the poor bitbake Quality Control?

[J. L.]

Right on thanks that does seem to answer a couple questions I have
come across so far especially the devshell. I will give making my
recipe another shot again today and see if I can get further with it
or even crazier get it to build correctly. Thanks again hopefully we
will see some other input to your original statement.



On Sat, Aug 14, 2010 at 10:06 AM, AJ ONeal <coolaj86@gmail.com> wrote:
> I've been compiling some common general issues
> http://fastr.github.com/articles/Troubleshooting-bitbake.html
>
> Here's some quick-n-dirty instructions akin to using devshell:
>
>
>   1. cd ${OVEROTOP}/tmp/work/${ARCH}/${PACKAGE}_${VER}_r${REV}/${BUILD}
>      - ${ARCH} is probably armv7a-angstrom-linux-gnueabi
>      - ${BUILD} is a directory which is not NOT src or temp such as the
>      package name or git or trunk or svn
>      - ${VER} and ${REV} - duh.
>   2. `cp ../temp/run.do_compile.${OLD_PID}
>      - ${OLD_PID} is a number like 3597 or 4352
>      - do_compile could be any task - do_install, etc
>   3. vim run.do_compile.1234
>      1. comment out do_compile() (or whatever)
>      2. add bash --norc
>   4. ./run.do_compile.1234 will put you in an environment with all
>   variables set
>   5. make (or whatever) to try to build, debug issues
>   6. exit (when done to go back to shell without run.do_compile settings)
>
> This is based on the tip on openembedded wiki, but it took me a while to
> understand what theirs meant so I cleaned it up in a fashion which is
> hopefully crystal clear.
>
> AJ ONeal
>
> On Sat, Aug 14, 2010 at 10:16 AM, J. L. <vwyodapink@gmail.com> wrote:
>
>> I agree with you AJ, its become a serious pain to just get a built
>> image. I myself have only built once completely and have yet to again
>> since figuring out the changes I needed to make to that image. I would
>> like to help myself but someone like me whose not very competent in
>> this area finds the documentation a tad lacking to properly learn the
>> ways to help. I still have yet to figure out how to even get my own
>> program from the net to build with OE as just following the hello
>> world does not seem to be enough for me to grasp how to do it with an
>> actual problem. Which runs me into my next issue I can not properly
>> modify the way some programs build to get the features I want.
>>
>> I have also noticed on the gumstix boards while most of the time it
>> seems its the same errors people get, every once in awhile you see
>> some people with one issue and others with another.
>>
>> I am sure this wont be of much use as a post but figured I would chime
>> in from an inexperienced beginner trying to get a footing in here and
>> to actually be able to use things properly.
>>
>>
>>
>> On Sat, Aug 14, 2010 at 8:46 AM, AJ ONeal <coolaj86@gmail.com> wrote:
>> > This is about the 4th time that I've `rm -rf`-d everything and started
>> from
>> > scratch and never yet got a working `bitbake omap3-console-image`
>> >
>> > Every time I `git pull` it's a different set of problems. It's never the
>> > same package twice, but they never all compile for the basic console
>> image
>> > anywhere from 4 to 20 hours in.
>> >
>> > Fortunately, I do have a machine that did build omap3-console-image a few
>> > weeks ago at work. But I've never been able to get it on my machine at
>> home
>> > (same os) to start playing around with.
>> >
>> > What needs to happen in order for things to git checked that they work
>> with
>> > at least a few common meta-tasks before being pushed out?
>> > And is there anything I can do to help? (Donate some machine time to run
>> > tests, for example)
>> >
>> >
>> > Obviously, we all have our day jobs and whatnot, but this is becoming
>> really
>> > frustrating for me and I'd like to see things improve.
>> >
>> > AJ ONeal
>> > _______________________________________________
>> > Openembedded-devel mailing list
>> > Openembedded-devel@lists.openembedded.org
>> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>> >
>>
>> _______________________________________________
>> Openembedded-devel mailing list
>> Openembedded-devel@lists.openembedded.org
>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

Re: What to do about the poor bitbake Quality Control?

[Frans Meulenbroeks]

2010/8/14 AJ ONeal <coolaj86@gmail.com>:
> This is about the 4th time that I've `rm -rf`-d everything and started from
> scratch and never yet got a working `bitbake omap3-console-image`
>
> Every time I `git pull` it's a different set of problems. It's never the
> same package twice, but they never all compile for the basic console image
> anywhere from 4 to 20 hours in.
>
> Fortunately, I do have a machine that did build omap3-console-image a few
> weeks ago at work. But I've never been able to get it on my machine at home
> (same os) to start playing around with.
>
> What needs to happen in order for things to git checked that they work with
> at least a few common meta-tasks before being pushed out?
> And is there anything I can do to help? (Donate some machine time to run
> tests, for example)
>
>
> Obviously, we all have our day jobs and whatnot, but this is becoming really
> frustrating for me and I'd like to see things improve.
>

If you are using dev head, you're living on the edge. If you can't
stand that better use the stable branch.

As far as there are issues:
Either report them, or better fix them.
(or as a workaround you could find the commit wthat broke it and
report that and perhaps temporariy locally revert that commit if you
cannot fix it).

Having said that:
I don't really see many issues. Maybe I'm building a different set of
packages or for a different set of hw or for a different distro.

Frans

Re: What to do about the poor bitbake Quality Control?

[Frans Meulenbroeks]

2010/8/14 Frans Meulenbroeks <fransmeulenbroeks@gmail.com>:

>
> If you are using dev head, you're living on the edge. If you can't
> stand that better use the stable branch.
>
> As far as there are issues:
> Either report them, or better fix them.
> (or as a workaround you could find the commit wthat broke it and
> report that and perhaps temporariy locally revert that commit if you
> cannot fix it).
>
> Having said that:
> I don't really see many issues. Maybe I'm building a different set of
> packages or for a different set of hw or for a different distro.
>
> Frans
>
Addition: I noticed when pressing send that you also mailed to
gumstix-users, so apparently you do have a gumstix.
I haven't so I have never build for it and cannot really help you with
it as I am not planning on buying one (so even though I could build I
cannot test).

What you also can do is enable logging to tinderbox then at least the
info of the faulty builds becomes available for others.

Frans

Re: What to do about the poor bitbake Quality Control?

[Philip Balister]

On 08/14/2010 12:01 PM, Frans Meulenbroeks wrote:
> 2010/8/14 AJ ONeal<coolaj86@gmail.com>:
>> This is about the 4th time that I've `rm -rf`-d everything and started from
>> scratch and never yet got a working `bitbake omap3-console-image`
>>
>> Every time I `git pull` it's a different set of problems. It's never the
>> same package twice, but they never all compile for the basic console image
>> anywhere from 4 to 20 hours in.
>>
>> Fortunately, I do have a machine that did build omap3-console-image a few
>> weeks ago at work. But I've never been able to get it on my machine at home
>> (same os) to start playing around with.
>>
>> What needs to happen in order for things to git checked that they work with
>> at least a few common meta-tasks before being pushed out?
>> And is there anything I can do to help? (Donate some machine time to run
>> tests, for example)
>>
>>
>> Obviously, we all have our day jobs and whatnot, but this is becoming really
>> frustrating for me and I'd like to see things improve.
>>
>
> If you are using dev head, you're living on the edge. If you can't
> stand that better use the stable branch.
>
> As far as there are issues:
> Either report them, or better fix them.
> (or as a workaround you could find the commit wthat broke it and
> report that and perhaps temporariy locally revert that commit if you
> cannot fix it).
>
> Having said that:
> I don't really see many issues. Maybe I'm building a different set of
> packages or for a different set of hw or for a different distro.

I think they are building from the gumstix overo repo. We are interested 
in failures (at least failures for recipes that are in .dev), but we 
can't guarantee when our fixes will hit the gumstix repository.

Philip

Philip

Re: What to do about the poor bitbake Quality Control?

[Elvis Dowson]

Hi,

On Aug 14, 2010, at 11:14 PM, Philip Balister wrote:

> On 08/14/2010 12:01 PM, Frans Meulenbroeks wrote:
>> 2010/8/14 AJ ONeal<coolaj86@gmail.com>:
>>> This is about the 4th time that I've `rm -rf`-d everything and started from
>>> scratch and never yet got a working `bitbake omap3-console-image`
>>> 
>>> Every time I `git pull` it's a different set of problems. It's never the
>>> same package twice, but they never all compile for the basic console image
>>> anywhere from 4 to 20 hours in.
>>> 

> I think they are building from the gumstix overo repo. We are interested in failures (at least failures for recipes that are in .dev), but we can't guarantee when our fixes will hit the gumstix repository.

The overo-oe repository setup instructions pull from a gitorious branch, that is based on the oe development branch. I too have faced these issues everytime. It really is a pain, one can never be sure if the overo-oe repo will compile reliably after a git pull. Its a total waste of time. 

Elvis Dowson

Re: What to do about the poor bitbake Quality Control?

[AJ ONeal]

How do I enable logging to tenderbox? I'd be happy to do that.
I googled it to little avail:
http://www.google.com/search?hl=en&q=tinderbox+openembedded


For reference, here's the process I've been following:


rm ~/overo-oe -rf                 # *clean it all out*

mkdir -p ~/overo-oe
cd ~/overo-oe

git clone git://gitorious.org/gumstix-oe/mainline.git org.openembedded.dev
      # *get a fresh copy of the repo*

cd org.openembedded.dev
git checkout --track -b overo origin/overo
cd ~/overo-oe
git clone git://git.openembedded.net/bitbake bitbake
cd bitbake
git checkout 1.8.18
cd ~/overo-oe
cp -r org.openembedded.dev/contrib/gumstix/build .
cp ~/.bashrc ~/bashrc.bak
cat ~/overo-oe/build/profile >> ~/.bashrc
source ~/overo-oe/build/profile

bitbake omap3-console-image                       # *make the basic image
from a fresh start*


AJ ONeal

On Sat, Aug 14, 2010 at 1:04 PM, Frans Meulenbroeks <
fransmeulenbroeks@gmail.com> wrote:

> 2010/8/14 Frans Meulenbroeks <fransmeulenbroeks@gmail.com>:
>
> >
> > If you are using dev head, you're living on the edge. If you can't
> > stand that better use the stable branch.
> >
> > As far as there are issues:
> > Either report them, or better fix them.
> > (or as a workaround you could find the commit wthat broke it and
> > report that and perhaps temporariy locally revert that commit if you
> > cannot fix it).
> >
> > Having said that:
> > I don't really see many issues. Maybe I'm building a different set of
> > packages or for a different set of hw or for a different distro.
> >
> > Frans
> >
> Addition: I noticed when pressing send that you also mailed to
> gumstix-users, so apparently you do have a gumstix.
> I haven't so I have never build for it and cannot really help you with
> it as I am not planning on buying one (so even though I could build I
> cannot test).
>
> What you also can do is enable logging to tinderbox then at least the
> info of the faulty builds becomes available for others.
>
> Frans
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

Re: What to do about the poor bitbake Quality Control?

[Frans Meulenbroeks]

2010/8/14 AJ ONeal <coolaj86@gmail.com>:
> How do I enable logging to tenderbox? I'd be happy to do that.
> I googled it to little avail:
> http://www.google.com/search?hl=en&q=tinderbox+openembedded

Add this to your local.conf:

INHERIT += "oestats-client"
OESTATS_SERVER = "tinderbox.openembedded.net"
OESTATS_BUILDER ="YOURNAME GOES HERE!"

>
>
> For reference, here's the process I've been following:
>
>
> rm ~/overo-oe -rf                 # *clean it all out*
>
> mkdir -p ~/overo-oe
> cd ~/overo-oe
>
> git clone git://gitorious.org/gumstix-oe/mainline.git org.openembedded.dev
>      # *get a fresh copy of the repo*

So you are cloning fro a tree that might be a copy of the openembedded
tree. If you want to be sure to have the real openembedded tree clone
from git.openembedded.org/openembedded. we have
repo.or.cz/r/openembedded.git as ro mirror

>
> cd org.openembedded.dev
> git checkout --track -b overo origin/overo

and this pulls the overo branch within that tree.

As it stands I have no idea what is in that tree and branch
And, unless someone donates me an overo, I guess it'll remain like that.
Guess this holds for most of the dev's on this list.

If there is an issue with a recipe that is reproducible in the
org.openembedded.dev branch of the official repository we'll be happy
to look into it, but until then I think you should discuss this with
the gumstix/overo people.

Best regards, Frans

> cd ~/overo-oe
> git clone git://git.openembedded.net/bitbake bitbake
> cd bitbake
> git checkout 1.8.18
> cd ~/overo-oe
> cp -r org.openembedded.dev/contrib/gumstix/build .
> cp ~/.bashrc ~/bashrc.bak
> cat ~/overo-oe/build/profile >> ~/.bashrc
> source ~/overo-oe/build/profile
>
> bitbake omap3-console-image                       # *make the basic image
> from a fresh start*
>
>
> AJ ONeal
>

Re: What to do about the poor bitbake Quality Control?

[Koen Kooi]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 14-08-10 17:46, AJ ONeal wrote:
> This is about the 4th time that I've `rm -rf`-d everything and started from
> scratch and never yet got a working `bitbake omap3-console-image`

That recipe isn't in OE, so it seems you're building from what we call
"some random tree", which is unsupported. Complain to the people in
charge of that tree and tell them to fix things.

If people don't care enough to get there stuff upstream, upstream won't
care about problems in that tree.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFMaFqHMkyGM64RGpERAnigAJ9sMzq5aph9ONHXjOc016N26xNvFwCgs2XJ
GeX7r5WwkbKeoGi3rSShMdk=
=miVm
-----END PGP SIGNATURE-----

Re: What to do about the poor bitbake Quality Control?

[Gary Thomas]

On 08/14/2010 01:01 PM, Frans Meulenbroeks wrote:
> 2010/8/14 AJ ONeal<coolaj86@gmail.com>:
>> This is about the 4th time that I've `rm -rf`-d everything and started from
>> scratch and never yet got a working `bitbake omap3-console-image`
>>
>> Every time I `git pull` it's a different set of problems. It's never the
>> same package twice, but they never all compile for the basic console image
>> anywhere from 4 to 20 hours in.
>>
>> Fortunately, I do have a machine that did build omap3-console-image a few
>> weeks ago at work. But I've never been able to get it on my machine at home
>> (same os) to start playing around with.
>>
>> What needs to happen in order for things to git checked that they work with
>> at least a few common meta-tasks before being pushed out?
>> And is there anything I can do to help? (Donate some machine time to run
>> tests, for example)
>>
>>
>> Obviously, we all have our day jobs and whatnot, but this is becoming really
>> frustrating for me and I'd like to see things improve.
>>
>
> If you are using dev head, you're living on the edge. If you can't
> stand that better use the stable branch.

The biggest problem (as discussed at great length already) is that the
distance from 'dev' to 'stable' can be measured in years :-(  'stable'
just isn't useful at all for "current" work...

> As far as there are issues:
> Either report them, or better fix them.
> (or as a workaround you could find the commit wthat broke it and
> report that and perhaps temporariy locally revert that commit if you
> cannot fix it).
>
> Having said that:
> I don't really see many issues. Maybe I'm building a different set of
> packages or for a different set of hw or for a different distro.

Turning a blind eye, I would say.  I (and others) have nearly 0%
success building any Angstrom image on an x86_64 based host.  I've tried
the latest Ubuntu and Fedora - neither works, both yield different failures.
At least I have i686 build hosts to fall back on, but there are issues there
as well.

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

Re: What to do about the poor bitbake Quality Control?

[Gary Thomas]

On 08/15/2010 03:22 PM, Koen Kooi wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 14-08-10 17:46, AJ ONeal wrote:
>> This is about the 4th time that I've `rm -rf`-d everything and started from
>> scratch and never yet got a working `bitbake omap3-console-image`
>
> That recipe isn't in OE, so it seems you're building from what we call
> "some random tree", which is unsupported. Complain to the people in
> charge of that tree and tell them to fix things.
>
> If people don't care enough to get there stuff upstream, upstream won't
> care about problems in that tree.

I suggest that _you_ practice what you preach.  There have been tons
of messages on this list complaining about recipes that you are responsible
for (or at least have the last checkin on) and nothing ever seems to get
done about it.

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

Re: What to do about the poor bitbake Quality Control?

[Cliff Brake]

On Sun, Aug 15, 2010 at 11:38 PM, Gary Thomas <gary@mlbassoc.com> wrote:

> The biggest problem (as discussed at great length already) is that the
> distance from 'dev' to 'stable' can be measured in years :-(  'stable'
> just isn't useful at all for "current" work...

I must admit that some percentage of the time I also experience build
errors, but I've simply adapted my work-flow to deal with it.  For my
project work, I simply build over the course of a week or so until I
get something stable, and then lock down my OE version for my
development.

I think it would be very useful to have a "stable" branch that is only
synchronised with dev when X number of targets build from a clean
build.  It seems like this would be high value, with little effort.
Of course there will be corner things that break, but at least a new
beagleboard user can check out something and have reasonable
confidence that it will build images.

Does anyone have suggestions for the branch name and a reasonable
subset of machines and build targets?  Perhaps someone is already
running these clean builds?  At one point we had a machine at OSUOSL
dedicated to this purpose, but no one ever set it up.

Thanks,
Cliff

-- 
=================
http://bec-systems.com

Re: What to do about the poor bitbake Quality Control?

[Martyn Welch]

On 17/08/10 16:02, Cliff Brake wrote:
> I think it would be very useful to have a "stable" branch that is only
> synchronised with dev when X number of targets build from a clean
> build.  It seems like this would be high value, with little effort.
> Of course there will be corner things that break, but at least a new
> beagleboard user can check out something and have reasonable
> confidence that it will build images.
>
> Does anyone have suggestions for the branch name and a reasonable
> subset of machines and build targets?  Perhaps someone is already
> running these clean builds?  At one point we had a machine at OSUOSL
> dedicated to this purpose, but no one ever set it up.
>

So something like Debian's stable, testing and unstable[1]?

Martyn

[1] http://www.debian.org/releases/


-- 
Martyn Welch (Principal Software Engineer)   |   Registered in England and
GE Intelligent Platforms                     |   Wales (3828642) at 100
T +44(0)127322748                            |   Barbirolli Square, 
Manchester,
E martyn.welch@ge.com                        |   M2 3AB  VAT:GB 927559189