By default, windows.h exposes a large number of problematic define statements
which are UpperCamelCase, such as a define from `CreateWindow` to
`CreateWindow{A,W}`.
As many of these names are generic (e.g. CreateFile, CreateWindow), they can
mess up Gecko code that may legitimately have its own methods with the same
names.
The header also defines some traditional SCREAMING_SNAKE_CASE defines which
can mess up our code by conflicting with local values.
This patch adds a simple code generator which generates wrappers for these
defines, and uses them to wrap the windows.h wrapper using the `stl_wrappers`
mechanism, allowing us to use windows.h in more places.
Differential Revision: https://phabricator.services.mozilla.com/D10932
Converted NS_STYLE_BORDER_STYLE_* consts to enum class. Updated corresponding values to enum class. reduced BCCornerInfo struct values to fit StyleBorderStyle values inside struct. Added defaults to switches that do not fully cover all instances of StyleBorderStyle.
It was pointed out in a review by jgilbert that glMapBuffer only supports
writing to the mapped range on an OpenGL ES profile and using it to read is
undefined behaviour. We now use glMapBufferRange when available, which does
support reading on both OpenGL and OpenGL ES profiles, and allows capturing
screenshots on Android. When it is not available, we fall back to glMapBuffer
(e.g., for macOS).
Differential Revision: https://phabricator.services.mozilla.com/D12341
--HG--
extra : moz-landing-system : lando
(Unless there were other profiler actions, as I'm not sure yet whether it would
be safe to skip them when the profiler is paused; another bug should
investigate that.)
Differential Revision: https://phabricator.services.mozilla.com/D11308
--HG--
extra : moz-landing-system : lando
In order to reduce the cost of running marionette tests on a virtual machine
with a GPU, add a marionette-gpu job, and run the WebRender rollout test added
in the previous patch in this new job.
Depends on D10528
Differential Revision: https://phabricator.services.mozilla.com/D12241
--HG--
extra : moz-landing-system : lando
Add test that when we restart the browser with a default value set on
gfx.webrender.all.qualified, Firefox saves that value and checks respects
the saved value when initializing WebRender.
Depends on D10527
Differential Revision: https://phabricator.services.mozilla.com/D10528
--HG--
extra : moz-landing-system : lando
Normandy's Preference Rollout code sets default values on prefs, not user
values (see uses of PrefUtils.setPref() in PreferenceRolloutAction.jsm).
Default prefs are not persistent; unlike user prefs, changes to default pref
values are not stored on disk. Changes to default values are only made on the
in-memory copy of the pref's value, and thus don't survive a browser restart.
Normandy changes the rolled out prefs early on in the startup of the browser,
but not before gfxPlatform::Init() runs. So that means gfx can't use Normandy
pref rollout to gradually rollout WebRender to release, as
gfxPlatform::InitWebRenderConfig() won't see the rolled out version of the
pref in time to turn on WebRender.
So to work around this, add a profile-before-change shutdown observer that
saves the default value of the gfx.webrender.all.qualified pref to a new user
pref, gfx.webrender.all.qualified.default. We check that on startup and
emulate the behavior that the pref system would have if that pref default
value had already been set by Normandy.
Differential Revision: https://phabricator.services.mozilla.com/D10527
--HG--
extra : moz-landing-system : lando
In order to reduce the cost of running marionette tests on a virtual machine
with a GPU, add a marionette-gpu job, and run the WebRender rollout test added
in the previous patch in this new job.
Depends on D10528
Differential Revision: https://phabricator.services.mozilla.com/D12241
--HG--
extra : moz-landing-system : lando
Add test that when we restart the browser with a default value set on
gfx.webrender.all.qualified, Firefox saves that value and checks respects
the saved value when initializing WebRender.
Depends on D10527
Differential Revision: https://phabricator.services.mozilla.com/D10528
--HG--
extra : moz-landing-system : lando
Normandy's Preference Rollout code sets default values on prefs, not user
values (see uses of PrefUtils.setPref() in PreferenceRolloutAction.jsm).
Default prefs are not persistent; unlike user prefs, changes to default pref
values are not stored on disk. Changes to default values are only made on the
in-memory copy of the pref's value, and thus don't survive a browser restart.
Normandy changes the rolled out prefs early on in the startup of the browser,
but not before gfxPlatform::Init() runs. So that means gfx can't use Normandy
pref rollout to gradually rollout WebRender to release, as
gfxPlatform::InitWebRenderConfig() won't see the rolled out version of the
pref in time to turn on WebRender.
So to work around this, add a profile-before-change shutdown observer that
saves the default value of the gfx.webrender.all.qualified pref to a new user
pref, gfx.webrender.all.qualified.default. We check that on startup and
emulate the behavior that the pref system would have if that pref default
value had already been set by Normandy.
Differential Revision: https://phabricator.services.mozilla.com/D10527
--HG--
extra : moz-landing-system : lando
Previously, WebRender ignored clip_node_id on the clip/scroll stack
when pushing clips or reference frames. This got fixed to be more consistent in:
https://github.com/servo/webrender/pull/3315
Now Gecko can use the clip chains generated for display items naturally,
instead of smuggling the last clip through the scroll_node_id, which is what
was happening in this hacky code branch being removed.
Differential Revision: https://phabricator.services.mozilla.com/D12216
--HG--
extra : moz-landing-system : lando
Here we make use of the parts added in parts 1 and 2 to hold onto
recycled surfaces for as long as necessary to prevent the animated image
decoder from reusing them until WebRender is done with them.
Differential Revision: https://phabricator.services.mozilla.com/D10902
Originally it made sense to share the code, but now the latter has
become too specialized to reuse it. Fork it off here and update it later
parts in this series.
Differential Revision: https://phabricator.services.mozilla.com/D10901
SourceSurfaceSharedData can be wrapped by RecyclingDataSurface in order
to facilitate its lifetime management. As long as the latter is alive,
the former cannot be reused by the animated image decoder for a future
frame's contents. SharedSurfacesAnimation will need to hold onto the
RecyclingDataSurface as long as WebRender is using the surface.
Differential Revision: https://phabricator.services.mozilla.com/D10899
When we replace the external image ID an image key points to, the
previous external image ID may still be used by WebRender. We currently
wait until the frame has been rendered to release our hold on said
surface. With this patch, we will communicate the image key and previous
external image ID pairing to the owning process when releasing to let it
know that it can reuse it (e.g. for an animated image). Additionally we
now use the new textures updated checkpoint which should happen sooner
than the frame rendered checkpoint, but guarantee that WebRender is no
longer using the old external image ID.
Differential Revision: https://phabricator.services.mozilla.com/D10897
Speculative fix for Bug 1507068. Checks if mParent is not null before trying to get a pointer to the toolbar animator
Differential Revision: https://phabricator.services.mozilla.com/D11957
--HG--
extra : moz-landing-system : lando
This required some changes to the page structure to exercise the codepath
that bug 1478335 fixed.
Differential Revision: https://phabricator.services.mozilla.com/D11437
--HG--
extra : moz-landing-system : lando
adjust fuzzy-if statements for win10 tests that are fialing on new windows 10 ami image
Differential Revision: https://phabricator.services.mozilla.com/D11914
--HG--
extra : moz-landing-system : lando
Due to changes in where things end up in the unified builds.
Depends on D8482
Differential Revision: https://phabricator.services.mozilla.com/D8483
--HG--
extra : moz-landing-system : lando
Previously, the ASR overrides contained Maybe<ClipId>, where Nothing() corresponded to taking the top of the clip/scroll stack instead of overriding. This change removes the associated complexity by ensuring that we always provide the ClipId.
Differential Revision: https://phabricator.services.mozilla.com/D11813
--HG--
extra : moz-landing-system : lando
Due to changes in where things end up in the unified builds.
Depends on D8482
Differential Revision: https://phabricator.services.mozilla.com/D8483
--HG--
extra : moz-landing-system : lando