LFS: Unstable and Testing
jeremy at jutley.org
Fri Jul 30 17:14:36 PDT 2004
> Hi all,
> Can someone fill me in on what the purpose of LFS unstable and testing
> is? I seem to have a misconception.
> I can't help but notice that every commit is done to both books. I was
> under the impression that unstable was for trying out new stuff. After
> it was determined that it was good and desirable for testing, it was
> moved there.
> I thought testing was the future next release of LFS. But the way I've
> seen it going is there is no difference between testing and unstable.
> Brand new stuff is going into both books at the same time.
> Why have two branches, it just seems like extra work if they're both
> parallel branches? Please, someone, clarify this for me. I am under
> the impression that what goes into testing has been tested, is good,
> and is what the community wants to see in the next release of LFS as
> it has been tested in unstable and discussed, *then* moved to testing.
Most of what we're doing right now are simple package updates, so they
apply equally well to both branches. Simple stuff like that doesn't
really require a whole lot of testing, other than to make sure they build
properly on the systems in question - with my commits today. The only
real experimental thing in unstable right now is hotplug, but soon I hope
to get working with newer glibc's, testing the code for GCC 3.4.2 (and
maybe at some point even GCC 3.5), and other experimental stuff. You'll
note that while unstable was upgraded to the latest HJL binutils, the
testing branch was actually downgraded to FSF 2.15 release - another
instance of making sure the testing book is as stable as possible.
So they really are separate branches, it's just that the updates applied
today were valid for both branches.
Hopefully this clears things up for you, if not, let me know what you're
unclear about and I'll do my best to clear it up for you.
More information about the lfs-dev