lp:~cemil-azizoglu/mir/mir-render-surface

Branch merges

Propose for merging

No branches dependent on this one.

On hold for merging into lp:mir

Mir development team: Pending requested 2016-07-14

Diff: 1709 lines (+1267/-6)

27 files modified

CMakeLists.txt (+8/-0)
playground/CMakeLists.txt (+19/-0)
playground/eglflash_render_surface.c (+152/-0)
playground/render_surface.cpp (+253/-0)
src/client/CMakeLists.txt (+2/-0)
src/client/buffer_stream.cpp (+9/-0)
src/client/buffer_stream.h (+4/-0)
src/client/client_buffer_stream.h (+2/-0)
src/client/connection_surface_map.h (+6/-1)
src/client/error_stream.cpp (+16/-0)
src/client/error_stream.h (+8/-0)
src/client/mir_buffer_stream_api.cpp (+6/-1)
src/client/mir_connection.cpp (+29/-2)
src/client/mir_connection.h (+2/-0)
src/client/mir_render_surface.h (+58/-0)
src/client/mir_render_surface_api.cpp (+140/-0)
src/client/mir_surface.cpp (+2/-0)
src/client/mir_surface.h (+3/-1)
src/client/mir_surface_api.cpp (+40/-1)
src/client/render_surface.cpp (+278/-0)
src/client/render_surface.h (+128/-0)
src/client/screencast_stream.cpp (+5/-0)
src/client/screencast_stream.h (+1/-0)
src/client/surface_map.cpp (+14/-0)
src/client/symbols.map (+8/-0)
src/include/client/mir_toolkit/client_types_nbs.h (+1/-0)
src/include/client/mir_toolkit/mir_render_surface.h (+73/-0)

api

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

3647. By Cemil Azizoglu on 2016-07-14

MirRenderSurface works - no tests

3646. By Chris Halse Rogers on 2016-08-12

Provide subpixel arrangement info from the mesa-kms platform. Fixes: https://bugs.launchpad.net/bugs/1393578.

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

3645. By Andreas Pokorny on 2016-08-12

Use the pixel format of the individual buffer stream to infer shaped()

It is a valid scenario to have an opaque buffer (or buffer stream) for the first layer in a surface and have further transparent or also opaque layers on top. This change removes the constant "non_rectangular" parameter of BasicSurface, and instead looks at the actual streams that are attached to surface.

With mir_presentation_queue it might be worth thinking about going one step further and actually look at the mg::Buffer that would be drawn next. An attempt to do so showed that it would require a few further refactoring steps in our integration test setup.
.

Approved by Kevin DuBois, mir-ci-bot, Alexandros Frantzis.

3644. By Kevin DuBois on 2016-08-12

Have the MirSurface report the shell-designated size like it used to, instead of the spec-requested size.

fixes: LP: #1612256. Fixes: https://bugs.launchpad.net/bugs/1612256.

Approved by mir-ci-bot, Alan Griffiths.

3643. By Chris Halse Rogers on 2016-08-12

Improvements to mtd::FakeDisplay.

Throw an exception when writing to/reading from the configuration change event trigger fails rather than silently continuing, blithely unaware that the call has failed.

Switch from a pipe to an eventfd; eventfd is the solution tailored for this use-case.

Presented in the forlorn hope that it makes a test that appears to be flaky on vivid+clang, but does not appear that it's possible for it to be flaky, stops being flaky. Either by consistently passing, or consistently failing!.

Approved by Alexandros Frantzis, Kevin DuBois, mir-ci-bot.

3642. By Chris Halse Rogers on 2016-08-11

Add subpixel arrangement info to the display configuration API.

Current platforms only provide mir_subpixel_arrangement_unknown, and need to add support for setting this to platform-provided values before this will be useful.

Fixes (part one): https://bugs.launchpad.net/mir/+bug/1393578. Fixes: https://bugs.launchpad.net/bugs/1393578.

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

3641. By Alexandros Frantzis on 2016-08-11

Fix clang build on yakkety. Fixes: https://bugs.launchpad.net/bugs/1609612.

Approved by mir-ci-bot, Chris Halse Rogers, Daniel van Vugt, Alan Griffiths, Cemil Azizoglu.

3640. By Kevin DuBois on 2016-08-10

Increase timeout for MultiThreadedCompositor.cleans_up_after_throw_in_start test,
which can fail when CI is very slow

fixes: LP: #1595940. Fixes: https://bugs.launchpad.net/bugs/1595940.

Approved by mir-ci-bot, Alan Griffiths, Cemil Azizoglu.

3639. By Andreas Pokorny on 2016-08-10

Fix server hang ups and allocation failures when client recreate buffer streams

When a mir client destroys a buffer stream the destruction result callback will trigger a buffer release of the associated buffers. That buffer release request inevitable fails in session mediator because it fails to resolve the BufferStreamId. This change handles exceptions during the attempt to disassociate buffers from already deleted buffer streams to ensure that the actual buffer release can happen.

This error is not really a leak - the buffers were deleted when the client disconnects and the server would be able to serve new clients.

Approved by Kevin DuBois, mir-ci-bot.

3638. By Daniel van Vugt on 2016-08-10

On those devices that have KMS but no hardware cursor support,
fall back automatically to the software cursor (LP: #1610054). Fixes: https://bugs.launchpad.net/bugs/1610054.

Approved by Cemil Azizoglu, mir-ci-bot.