1.17 status and schedule
keithp at keithp.com
Sun Jan 4 17:35:31 PST 2015
Ok, it's (almost) the first working day of the year, and we're supposed
to be shipping 1.17. I think we're quite close, but I want to make sure
everyone is ready for the release before it happens.
From my list of pending patches, we've got:
some optimizations for lighter-weight hardware. Very useful,
don't seem like material for 1.17.
There has been work on getting page flipping going here; given
that modesetting is all new functionality for 1.17, we can't
have any regressions, so if that code looks solid and is useful,
I'd be tempted to include it.
Aaron posted a pair of patches dealing with GPU screen handling;
no review yet, but it seems like a useful fix if it is obvious,
correct and not scary. Review needed.
Some warning fixes related to dtrace and DMX. These should wait
until after 1.17.
I posted a patch series that fixes Present interactions with
compositing managers. It seems 'obviously' correct to me, but it
hasn't seen any review. Unless someone gets really excited about
this, it'll have to wait until after 1.17.
Tomasz posted a patch to move some shm file descriptors out of
the low range used for X clients; I've seen no discussion of
this, but it looks like it'd be nice to have
Daniel posted a patch series to clear keys that were pressed
when switching back to X, and Peter and I agreed that the
results were useful, but that the implementation could be done a
bit differently with better results. I posted an
Xwayland-specific patch that papered over the worst problems.
Any of the more complete solutions seem a bit too invasive to me
for 1.17, but the simple hack seems like a good interim solution
if someone wants to push it.
Tiling properties. This is (alas) an ABI change, so it will need
to wait until after 1.17. I think we'll want to merge this along
with the other randr changes to add 'monitors' so that we can
make gnome work properly with most 4k monitors.
Beyond 1.17, if we can get a bunch of the pending glamor, modesetting
and randr changes merged soon after the release, I'd like to consider
doing a 1.18 release on a shorter schedule than 6 months. We've finally
reached the point where a pile of new feature code is waiting to land
once the merge window opens. We may be able to have a useful release
with a very short merge window.
As always, this is just a suggestion. Failing a fairly strong rough
consensus, we should plan on retaining our existing 6 month schedule.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 810 bytes
Desc: not available
More information about the xorg-devel