lp:~raof/mir/notify-shell-of-display-configuration

Branch merges

Propose for merging

No branches dependent on this one.

Rejected for merging into lp:mir

Alan Griffiths: Disapprove on 2016-09-16

Cemil Azizoglu (community): Approve on 2016-09-14

Daniel van Vugt: Abstain on 2016-09-12

Mir CI Bot: Approve (continuous-integration) on 2016-09-08

Diff: 245 lines (+179/-1)

4 files modified

include/server/mir/shell/display_configuration_controller.h (+17/-0)
src/server/scene/mediating_display_changer.cpp (+16/-1)
src/server/scene/mediating_display_changer.h (+5/-0)
tests/unit-tests/scene/test_mediating_display_changer.cpp (+141/-0)

api

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

3695. By Chris Halse Rogers on 2016-09-08

...and moke set_configured_callback threadsafe.

3694. By Chris Halse Rogers on 2016-09-08

More complete and correct set_configured_callback()

3693. By Chris Halse Rogers on 2016-09-08

Add (start of) a configured-notification to msh::DisplayConfigurationController().

There are all sorts of times that it'd be useful in our test code to know when a configuration
change had occurred, and I'm sure that shells will find it equally useful.

3692. By Brandon Schaefer on 2016-09-08

If you were to confine a surface that was unfocused, mir was more then happy to confine to that region. Not what is wanted! So check we are the focused surface before updating the seat.

Approved by mir-ci-bot, Kevin DuBois.

3691. By Brandon Schaefer on 2016-09-07

Gamma support for KMS. Other platforms unsupported.

Approved by Cemil Azizoglu, mir-ci-bot, Kevin DuBois.

3690. By Alan Griffiths on 2016-09-07

Fix most of the diagnostic messages doxygen gives on "make doc".

Approved by Cemil Azizoglu, mir-ci-bot.

3689. By Chris Halse Rogers on 2016-09-07

Bump timeout for ANR acceptance tests.

These require the server to process three separate alarm notifications across
3 seconds, process them, and wait for the client process to handle them.

Bump the timeouts to values that can't possibly be exceeded, even on our
poor oversubscribed CI hardware.

Plausibly fixes: https://bugs.launchpad.net/mir/+bug/1570327. Fixes: https://bugs.launchpad.net/bugs/1570327.

Approved by mir-ci-bot, Cemil Azizoglu, Daniel van Vugt.

3688. By Daniel van Vugt on 2016-09-07

Add basic support for DisplayReport to the x11 platform.

I have added as much as makes sense for this graphics platform, but
really only needed report_vsync to get some log messages for other
work I'm doing. The relevant class for that X::DisplayBuffer isn't
under test yet, but I don't feel this is important enough to
introduce a test suite for it here.

Approved by Cemil Azizoglu, mir-ci-bot, Chris Halse Rogers.

3687. By Kevin DuBois on 2016-09-07

nested: change the construction of mgn::DisplayBuffer so that it creates a MirBufferStream, and then uses that to create the backing MirSurface. Nested passthrough will toggle between that MirBufferStream and a chain (or chains) depending on what a particular composition calls for.

Approved by mir-ci-bot, Alexandros Frantzis.

3686. By Cemil Azizoglu on 2016-09-06

These tests pass when they should fail. This MP fixes that so they pass for the right reasons.
.

Approved by mir-ci-bot, Brandon Schaefer, Daniel van Vugt.