Making our multi-monitor, dock and projector support better.
The consensus of this session was that the remaining plumbing bits that are not working are plain bugs. The rest was design work presented by Stewart Wilson of the design team.
seb128, 2012-05-23: dropping the q target, we didn't sign to fix all the bugs listed there in q and don't want to track them this way, we will still watch the list and fix what we can
bryce, 2011-11-15: Copied in work-items from the multimedia-
raof, 2011-11-16: Dropping WIs superseded by libxrandr-utils and design work.
bryce, 2011-11-18: I've linked a bunch of bugs Stewart Wilson flagged as multi-monitor issues. Most are not 'blockers' but I've opted to include them all for completeness, and to make it easier to track overall progress. The blockers are set to Critical importance (Red bug with '!' on their back), and those are the ones we're primarily focused on in getting resolved for Precise.
I've also linked a slew of X bug reports that looked vaguely relevant. Some may turn out to not really be relevant (or not really be X bugs) once we analyze them closer. A few are long-standing X issues we probably won't be able to fix in this time frame but should give thought to. A lot are probably also highly HW-specific, but still ought to be investigated. There's probably also a bunch I've missed, but this set is going to give us more than plenty to do.
I've also crafted a few Arsenal reports listing bugs for desktop teams that have been tagged 'dual-head':
At this time we are showing one bug task Fix Released, one Fix Committed, a couple In Progress and four Invalid.
bryce, 2011-12-21: I've gone through all of the linked bug reports and marked ones that can still be reproduced in precise, or requested information where it was unclear how to reproduce. A large bunch I verified are fixed now. A few are obsolete, a few were dupes of other bugs, a few request changes that contradict the spec, and some were ambiguous but the reporter never followed up and they've expired. I dropped a few linked bugs that actually aren't multi-monitor issues.
Of the 102 bugs currently linked:
* 13 are fix released
* 1 fix committed
* 0 in progress
* 13 are invalid or expired
* 3 are wontfix or opinion
Thus we're about 30% of the way through the bug reports. Many of these bugs are deep-seated thorny problems we're unlikely to get fixed in the next few months, so I don't think we'll ever get to 100%. If we hit 60% by Precise release I'd think we did well.
bryce, 2012-02-13: Completed re-verification and re-triaging of all linked bugs. 94 bugs are linked at this point. (Some are unlinked due to not being multi-monitor issues after all, and a few new ones are added due to recent changes.)
Of the 94 currently linked:
* 27 are fix released
* 0 fix committed
* 1 in progress
* 21 invalid or expired
* 5 wontfix or opinion
So we're 56% through the bug reports, well on target to hit the 60% closure objective.
Except for a couple nvidia bugs and one intel bug, the remaining X.org bugs are Wishlist and not easy things to address this cycle.
There are still a couple handfuls of control center bugs but they are mostly low severity corner cases and wishlist items; the most serious issues from this list have been addressed.
bryce, 2012-05-14: At this point post-release the stats are:
Of the 94 currently linked:
* 32 are fix released
* 0 are fix committed
* 0 are in progress
* 22 are invalid or expired
* 8 are wontfix or opinion
That puts us at 66% closure.
[gema] QA to review the use cases whenever they are fully drafted and give feedback: TODO
[bryce] Associate blocking Unity Launchpad bugs with blueprint: DONE
[bryce] Associate blocking Gnome Launchpad bugs with blueprint: DONE
[bryce] Associate blocking X.org Launchpad bugs with blueprint: DONE
[repete] Associate blocking bugs with blueprint: DONE
[swilson] Publish design work: DONE
[repete] Collect existing use-cases for multiple monitor support: TODO
[raof] Investigate how to distinguish projectors / TVs / monitors.: TODO
[raof] Investigate VGA hotplug detection - is it enabled everywhere it could be? (yes): DONE
[raof] Add support to X for keeping the KMS-set mode on startup: TODO
* Blueprints in grey have been implemented.