6.9 Progress And Plans For The Future
Hi all,
I'm happy to announce that for the first time I'm running 6.9rc0
packages. I finally got the tree to build this afternoon on i386, and they
seem to be running smoothly so far (aside from that libICE bug I just fixed
a little while ago). They are, however, in no way ready to go in to
experimental for the following reasons:
1) They require a newer render and xrender than we currently are shipping.
I have the packages on my hard drive, but I need to commit the code to
svn. I'm going to check on doing that later this evening. Another issue
is that I'm not listed as an uploader for these packages. I'm going to
add myself to that field unless someone objects, so I can keep this
moving ahead. These will, of course, go in to experimental until Xorg
6.9 goes in to unstable.
2) xterm was officially split out from the upstream tree. I've heard it's
going back in for rc1 (or the final release, whichever is first) but
we may as well push forward with transitioning to Thomas Dickey as our
canonical upstream. If Ender doesn't get back about this tonight, I'll
start setting up the repo and moving the packaging material over to
split it out officially.
3) It will undoubtedly FTBFS on any non-i386 arch because I haven't done
the requisite manifest work at all for these arches. Once the above are
done, I'm going to get working on these other arches. This is not an
actual blocker for uploading to experimental, but I'd like to get this
cleared up as soon as possible so that we don't totally break unstable
for five rounds of uploads due to the manifest stuff again.
Porters reading this list, or just anyone who owns a minority arch, if
you want to send build logs so that I can fix up the manifest stuff for
your arch, please check out the 6.9 branch once I get render and
xrender updated and uploaded to experimental. The build log will have a
diff at the end for the manifest for your arch, and this will be
critical in getting things to fully build for you.
Once 6.9 actually is in experimental, I'm going to actively contact
porter lists and get them to help fixing the builds on their arches.
So that's where we're at with 6.9. We're getting to the stage where any
work on the 6.8 tree is polish, although there's still plenty to be done
there, and fortunately it will transfer over easily to the 6.9 tree.
However, in order to keep momentum, I plan to set a freeze on the 6.8 tree
at the point where 6.8 reaches testing. During this time, all changes to
the 6.8 tree must be approved by me. At the same time, the 6.9 tree and 7.0
tree (if it exists) will be fully open for development.
I'm also planning to set a freeze on the 6.9 packages at some point,
where any changes to those packages must be approved by me. During this
freeze the 7.0 tree will be fully open for development. I will, of course,
let you know when this is, but it definitely will not happen until 6.9 is
built on all arches and uploaded to unstable. Unfortunately, I can't give
as much of a milestone as "when it enters testing" like I can for 6.8, but
I'll keep you all posted.
Finally, the plan for 7.0 for those who are interested. We will, yet
again, steal the packages wholesale from Daniel and Ubuntu one more time.
This will hopefully be the last time we have to do this, as I'd like to
stay more synchronized with them in the future, by stealing patches and
staying up to date with upstream. We'll have to do another audit in full,
just like for 6.8, and there's a whole new build system to learn as well.
The benefits to the modular tree should be apparent, and we stand to gain a
lot by moving towards 7.0 quickly. I haven't yet decided when I'm going to
start work on 7.0 in full, although anyone who is interested is welcome to
do so ahead of me.
That's it. Questions? Comments? Concerns? Free alcohol for your XSFRM?
- David Nusinow
Reply to: