gtk4.git
6 years agotestsuite: Turn all CSS test ui files into regular windows
Benjamin Otte [Fri, 14 Feb 2020 18:27:17 +0000 (19:27 +0100)]
testsuite: Turn all CSS test ui files into regular windows

Use undecorated windows instead of popups.

6 years agotestsuite: Make a11y tests use regular windows
Benjamin Otte [Fri, 14 Feb 2020 18:25:23 +0000 (19:25 +0100)]
testsuite: Make a11y tests use regular windows

6 years agoMerge branch 'wip/otte/for-master' into 'master'
Benjamin Otte [Fri, 14 Feb 2020 17:12:17 +0000 (17:12 +0000)]
Merge branch 'wip/otte/for-master' into 'master'

reftests: Change popup windows to non-decorated windows

See merge request GNOME/gtk!1435

6 years agoreftests: Change popup windows to non-decorated windows
Benjamin Otte [Fri, 14 Feb 2020 16:54:38 +0000 (17:54 +0100)]
reftests: Change popup windows to non-decorated windows

This allows getting rid of popup windows.

6 years agoMerge branch 'zhaoqiang/gtk-To_avoid_compiler_warning'
Matthias Clasen [Fri, 14 Feb 2020 13:47:36 +0000 (08:47 -0500)]
Merge branch 'zhaoqiang/gtk-To_avoid_compiler_warning'

6 years agoMerge branch 'master-subtract-base-size' into 'master'
Matthias Clasen [Fri, 14 Feb 2020 13:19:42 +0000 (13:19 +0000)]
Merge branch 'master-subtract-base-size' into 'master'

gdk: Subtract base size when checking aspect ratio

See merge request GNOME/gtk!1331

6 years agoMerge branch 'wip/otte/for-master' into 'master'
Mohammed Sadiq [Fri, 14 Feb 2020 02:39:21 +0000 (02:39 +0000)]
Merge branch 'wip/otte/for-master' into 'master'

slicelistmodel: Fix two wrong computations

See merge request GNOME/gtk!1433

6 years agoslicelistmodel: Fix two wrong computations
Benjamin Otte [Fri, 14 Feb 2020 02:13:49 +0000 (03:13 +0100)]
slicelistmodel: Fix two wrong computations

Both of those are thinkos during math.
Found by Mohammed Sadiq.

Testcases triggering both have been added.

6 years agoForce-create style contexts in realize
Matthias Clasen [Fri, 14 Feb 2020 02:14:01 +0000 (21:14 -0500)]
Force-create style contexts in realize

This is necessary as long as style contexts are
required for proper css change tracking.

Fixes #2435

6 years agoMerge branch 'master' into 'master'
Emmanuele Bassi [Thu, 13 Feb 2020 19:53:33 +0000 (19:53 +0000)]
Merge branch 'master' into 'master'

Fixed OpenGL extension detection for extensions promoted to OpenGL core.

Closes #2428

See merge request GNOME/gtk!1424

6 years agoFix detection of OpenGL 3.3 core GL_ARB_timer_query.
David Hogan [Wed, 12 Feb 2020 20:55:59 +0000 (07:55 +1100)]
Fix detection of OpenGL 3.3 core GL_ARB_timer_query.

Some systems (notably macOS) will not allow enumeration of an extension that has been promoted to core OpenGL for context in use. This change assumes that GL_ARB_timer_query is available on OpenGL 3.3+.

I could not find definitive information on whether GL_ARB_debug_output or GL_KHR_debug have been added to core. Other extensions in use were addressed by https://gitlab.gnome.org/GNOME/gtk/merge_requests/1422 .

6 years agoFixed OpenGL extension detection for extensions promoted to OpenGL core.
David Hogan [Wed, 12 Feb 2020 19:52:24 +0000 (06:52 +1100)]
Fixed OpenGL extension detection for extensions promoted to OpenGL core.

For a given OpenGL context, macOS in particular does not support enumeration / detection of OpenGL features that have been promoted to core OpenGL functionality. It is possible other drivers are the same. This change assumes support for GL_ARB_texture_non_power_of_two with OpenGL 2.0+, GL_ARB_texture_rectangle with OpenGL 3.1+ and GL_EXT_framebuffer_blit with OpenGL 3.0+. I failed to find definitive information on whether GL_GREMEDY_frame_terminator has been promoted to OpenGL core, or whether GL_ANGLE_framebuffer_blit or GL_EXT_unpack_subimage have been promoted to core in OpenGL ES. This change results in a significant GtkGLArea performance boost on macOS.

Closes #2428

6 years agoMerge branch 'ci-flatpak' into 'master'
Emmanuele Bassi [Thu, 13 Feb 2020 18:10:29 +0000 (18:10 +0000)]
Merge branch 'ci-flatpak' into 'master'

CI pipeline changes for Flatpak jobs

See merge request GNOME/gtk!1429

6 years agoci: Remove G_MESSAGES_DEBUG
Emmanuele Bassi [Thu, 13 Feb 2020 17:10:19 +0000 (17:10 +0000)]
ci: Remove G_MESSAGES_DEBUG

We don't need all the debugging messages.

6 years agoci: Allow flatpak jobs to fail
Emmanuele Bassi [Thu, 13 Feb 2020 17:09:35 +0000 (17:09 +0000)]
ci: Allow flatpak jobs to fail

We have more accurate CI jobs, now; the Flatpak jobs are nice to have.

6 years agoMerge branch 'gi-fixes' into 'master'
Matthias Clasen [Thu, 13 Feb 2020 16:38:41 +0000 (16:38 +0000)]
Merge branch 'gi-fixes' into 'master'

Various fixes for documentation and introspectable API

See merge request GNOME/gtk!1427

6 years agoMake release builds mandatory for ci
Matthias Clasen [Thu, 13 Feb 2020 16:21:15 +0000 (11:21 -0500)]
Make release builds mandatory for ci

The tests are now passing in release builds, lets keep
it that way.

6 years agotestsuite: Handle icontheme test better
Matthias Clasen [Thu, 13 Feb 2020 15:56:26 +0000 (10:56 -0500)]
testsuite: Handle icontheme test better

Skip just the failing tests in release builds.

6 years agogsk: Fix the compiler warning differently
Matthias Clasen [Thu, 13 Feb 2020 15:42:21 +0000 (10:42 -0500)]
gsk: Fix the compiler warning differently

6 years agoMerge branch 'report-fixes' into 'master'
Emmanuele Bassi [Thu, 13 Feb 2020 15:04:04 +0000 (15:04 +0000)]
Merge branch 'report-fixes' into 'master'

Fixes for the CI test reports

See merge request GNOME/gtk!1426

6 years agoci: Use per-suite anchors
Emmanuele Bassi [Thu, 13 Feb 2020 14:45:08 +0000 (14:45 +0000)]
ci: Use per-suite anchors

Anchors are global, so they need to be namespaced.

6 years agoci: Rearrange the results in the HTML report
Emmanuele Bassi [Thu, 13 Feb 2020 14:26:40 +0000 (14:26 +0000)]
ci: Rearrange the results in the HTML report

Move the failures up top, so they stand out.

6 years agoci: Fix the HTML anchor in the report
Emmanuele Bassi [Thu, 13 Feb 2020 14:23:59 +0000 (14:23 +0000)]
ci: Fix the HTML anchor in the report

6 years agoci: Use result instead of exit code in the JUnit report
Emmanuele Bassi [Thu, 13 Feb 2020 13:56:33 +0000 (13:56 +0000)]
ci: Use result instead of exit code in the JUnit report

We have a result code coming from Meson which is more accurate than just
looking at the exit code of the unit.

6 years agoci: Unexpected passes are failures
Emmanuele Bassi [Thu, 13 Feb 2020 13:54:22 +0000 (13:54 +0000)]
ci: Unexpected passes are failures

List them as such in our reports.

6 years agoci: Put the branch name in the HTML header
Emmanuele Bassi [Thu, 13 Feb 2020 13:54:02 +0000 (13:54 +0000)]
ci: Put the branch name in the HTML header

6 years agoci: Move HTML report meta from header to article
Emmanuele Bassi [Thu, 13 Feb 2020 13:52:57 +0000 (13:52 +0000)]
ci: Move HTML report meta from header to article

Makes the bar smaller.

6 years agoci: Re-enable ccache
Emmanuele Bassi [Thu, 13 Feb 2020 14:04:41 +0000 (14:04 +0000)]
ci: Re-enable ccache

6 years agoRemove gtk_style_context_get_parent()
Emmanuele Bassi [Tue, 11 Feb 2020 17:34:09 +0000 (17:34 +0000)]
Remove gtk_style_context_get_parent()

The function was declared and mentioned in the documentation, but it has
been removed in commit 416182a20d062d5865916f86909354ec85d00dd1.

6 years agogi: Skip gtk_custom_layout_new()
Emmanuele Bassi [Tue, 11 Feb 2020 17:29:56 +0000 (17:29 +0000)]
gi: Skip gtk_custom_layout_new()

The whole GtkCustomLayout class is mostly a C convenience API, and
there's no reason why high level languages should use it.

6 years agoFix the function name in the gtk-doc stanza
Emmanuele Bassi [Tue, 11 Feb 2020 17:28:18 +0000 (17:28 +0000)]
Fix the function name in the gtk-doc stanza

The documentation is still referring to the old function's name.

6 years agoMatch argument name between declaration and definition
Emmanuele Bassi [Tue, 11 Feb 2020 17:26:51 +0000 (17:26 +0000)]
Match argument name between declaration and definition

Otherwise the documentation and the introspection data will complain.

6 years agodocs: Match the argument name with its declaration
Emmanuele Bassi [Tue, 11 Feb 2020 17:25:44 +0000 (17:25 +0000)]
docs: Match the argument name with its declaration

6 years agoAdd a notify function to GdkContentProvider
Emmanuele Bassi [Tue, 11 Feb 2020 17:24:01 +0000 (17:24 +0000)]
Add a notify function to GdkContentProvider

The callback-based content providers need a GDestroyNotify function to
free the data passed to them on construction, otherwise they are going
to leak.

6 years agoRemove border-half-pixel reftest from xfail
Matthias Clasen [Thu, 13 Feb 2020 14:39:27 +0000 (09:39 -0500)]
Remove border-half-pixel reftest from xfail

It passes both locally and on ci now.

6 years agoMerge branch 'wip/otte/no-clip-on-draw' into 'master'
Matthias Clasen [Thu, 13 Feb 2020 13:43:59 +0000 (13:43 +0000)]
Merge branch 'wip/otte/no-clip-on-draw' into 'master'

Various fixes to rendernode drawing with Cairo

See merge request GNOME/gtk!1425

6 years agogdk: Remove gdk_cairo_get_clip_rectangle()
Benjamin Otte [Wed, 12 Feb 2020 22:45:51 +0000 (23:45 +0100)]
gdk: Remove gdk_cairo_get_clip_rectangle()

The function is fundamentally broken for unbounded surfaces.
If a surface is unbounded, we cannot represent this as a
cairo_rectangle_int_t, and using the return value doesn't work because
it's already used for something else.

In GTK3, unbounded surfaces aren't a problem, but GTK4 uses recording
surfaces.

So better remove that function before we keep using it and using it
wrong.

6 years agocairoblur: Get rid of gdk_cairo_get_clip_rectangle() call
Benjamin Otte [Thu, 13 Feb 2020 01:48:19 +0000 (02:48 +0100)]
cairoblur: Get rid of gdk_cairo_get_clip_rectangle() call

It's about to be removed.

6 years agogsk: Clip shadow node before push_group()
Benjamin Otte [Thu, 13 Feb 2020 00:53:12 +0000 (01:53 +0100)]
gsk: Clip shadow node before push_group()

We want to be sure to push a group that's as small as possible, so we
clip to the child's bounds, because that's the smallest we can go.

6 years agogsk: Add a utility function for rectangles
Benjamin Otte [Thu, 13 Feb 2020 00:48:03 +0000 (01:48 +0100)]
gsk: Add a utility function for rectangles

... and use it.

6 years agocairoblur: Move the check for early exit
Benjamin Otte [Wed, 12 Feb 2020 23:14:29 +0000 (00:14 +0100)]
cairoblur: Move the check for early exit

That means we only have one place where we check all kinds of early
exits.

6 years agorendernode: Avoid rounding errors
Benjamin Otte [Wed, 12 Feb 2020 17:44:11 +0000 (18:44 +0100)]
rendernode: Avoid rounding errors

Compute the pattern matrix directly instead of transforming the cairo_t.

This ensures that when node_size / texture_size is some obscure floating
point value, we don't get rounding issues from scaling by it once we
draw the texture_size rectangle.

I have no actual failure where this comes in handy, but I had written
the code anyway, so decided to keep it.

6 years agotestsuite: Add tests for rounded rect code
Benjamin Otte [Thu, 13 Feb 2020 04:23:31 +0000 (05:23 +0100)]
testsuite: Add tests for rounded rect code

... including the bug that was fixed in the parent commit.

6 years agoroundedrect: Fix inlining of graphene functions
Benjamin Otte [Thu, 13 Feb 2020 06:33:18 +0000 (07:33 +0100)]
roundedrect: Fix inlining of graphene functions

graphene treats equality for contains() operations as always matching,
so do the same thing.

This is because unlike integer math, floating point cannot do the "as
close as possible to the point, but not reaching it" operation that
integer does by just subtracting 1.

6 years agoroundedrect: Fix gsk_rounded_rect_intersects_rect()
Benjamin Otte [Thu, 13 Feb 2020 04:21:49 +0000 (05:21 +0100)]
roundedrect: Fix gsk_rounded_rect_intersects_rect()

The previous code would return FALSE for
  gsk_rounded_rect_intersects_rect (&rounded, &rounded.bounds);
if rounded was indeed rounded.

6 years agotestsuite: Don't use gdk_cairo_get_clip_rectangle()
Benjamin Otte [Wed, 12 Feb 2020 22:44:09 +0000 (23:44 +0100)]
testsuite: Don't use gdk_cairo_get_clip_rectangle()

I want to remove it.

6 years agotestsuite: Don't destroy surfaces that are still used
Benjamin Otte [Thu, 13 Feb 2020 02:23:51 +0000 (03:23 +0100)]
testsuite: Don't destroy surfaces that are still used

Coercing the surfaces must not unref the old surface, because the old
surface is going to be saved to a file.

6 years agotestsuite: Remove accidentally checked in file
Benjamin Otte [Thu, 13 Feb 2020 05:14:05 +0000 (06:14 +0100)]
testsuite: Remove accidentally checked in file

This file should go in the listview branch I guess.

6 years agoFix flickery hover
Matthias Clasen [Thu, 13 Feb 2020 03:45:24 +0000 (22:45 -0500)]
Fix flickery hover

Commit 47c44644b109e0 was a bit overzealous in fixing
compiler warnings. We still need to call collect_textures,
even if we don't need the number that it returns.

6 years agoMerge branch 'wip/baedert/cairo-paint' into 'master'
Matthias Clasen [Wed, 12 Feb 2020 23:44:38 +0000 (23:44 +0000)]
Merge branch 'wip/baedert/cairo-paint' into 'master'

Remove cairo_clip() when drawing render nodes

Closes #2431

See merge request GNOME/gtk!1418

6 years agoRemove unused signal enum values
Matthias Clasen [Wed, 12 Feb 2020 23:24:59 +0000 (18:24 -0500)]
Remove unused signal enum values

GtkWidget lost even more signals, so we don't need
these enum values anymore.

6 years agopopovermenu: Avoid an uninitialized variable
Matthias Clasen [Wed, 12 Feb 2020 22:58:41 +0000 (17:58 -0500)]
popovermenu: Avoid an uninitialized variable

Pointed out by Timm/clang.

Fixes https://gitlab.gnome.org/GNOME/gtk/issues/2434

6 years agoMerge branch 'ci-pages' into 'master'
Matthias Clasen [Wed, 12 Feb 2020 22:35:58 +0000 (22:35 +0000)]
Merge branch 'ci-pages' into 'master'

Rename the SCSS files for our themes

Closes #2423

See merge request GNOME/gtk!1423

6 years agoRename the SCSS files for our themes
Emmanuele Bassi [Wed, 12 Feb 2020 20:18:27 +0000 (20:18 +0000)]
Rename the SCSS files for our themes

It seems that Meson's gnome.compile_resources() cannot deal with two
files with the same name under different directories, which breaks the
build parallelism because the GResource file ends up not depending on
either the Adwaita or the HighContrast gtk-contained.css file.

This commit only changes the on-disk names of the Adwaita and
HighContrast SCSS files, and the corresponding generated CSS files; the
files in the GResource are going to be aliased to the old file names, to
minimise the breakage. We might want to change the theme entry points at
some later date, if we decide to commit to this naming scheme.

Fixes: #2423
See Meson bug: https://github.com/mesonbuild/meson/issues/6615

6 years agoFix the Emoji chooser finalize
Matthias Clasen [Wed, 12 Feb 2020 19:23:01 +0000 (14:23 -0500)]
Fix the Emoji chooser finalize

If the Emoji chooser is finalized before the idle
has run, it would spew criticals, breaking some tests.
Avoid that.

6 years agoMerge branch 'gbsneto/fix-emojichooser-header' into 'master'
Georges Basile Stavracas Neto [Wed, 12 Feb 2020 18:56:19 +0000 (18:56 +0000)]
Merge branch 'gbsneto/fix-emojichooser-header' into 'master'

build: Install gtkemojichooser.h

See merge request GNOME/gtk!1420

6 years agobuild: Install gtkemojichooser.h
Georges Basile Stavracas Neto [Wed, 12 Feb 2020 18:41:49 +0000 (15:41 -0300)]
build: Install gtkemojichooser.h

It was made public at ef7172dc1ee86a, but the header file
wasn't added to the public headers list.

Add the header file to the list.

6 years agoMerge branch 'ci-pages' into 'master'
Emmanuele Bassi [Wed, 12 Feb 2020 18:38:39 +0000 (18:38 +0000)]
Merge branch 'ci-pages' into 'master'

Small CI pipeline fixes

See merge request GNOME/gtk!1419

6 years agoci: Update flatpak job
Emmanuele Bassi [Wed, 12 Feb 2020 18:21:00 +0000 (18:21 +0000)]
ci: Update flatpak job

Use `flatpak build` and point at the nightly repo when generating the
bundle.

6 years agoci: Deploy the API ref before flatpak
Emmanuele Bassi [Wed, 12 Feb 2020 18:10:49 +0000 (18:10 +0000)]
ci: Deploy the API ref before flatpak

6 years agoci: Fix linking of the CI assets in the report
Emmanuele Bassi [Wed, 12 Feb 2020 18:03:48 +0000 (18:03 +0000)]
ci: Fix linking of the CI assets in the report

The report is relative to the build directory, and so are the generated
assets.

6 years agotestsuite: Warn if a necessary env var is missing
Matthias Clasen [Tue, 11 Feb 2020 19:09:24 +0000 (14:09 -0500)]
testsuite: Warn if a necessary env var is missing

Better to warn if we know why the tests are
going to fail. Saves headscratching later.

6 years agoMove performance tests to the right testsuite
Matthias Clasen [Tue, 11 Feb 2020 19:09:00 +0000 (14:09 -0500)]
Move performance tests to the right testsuite

This was a copy-paste error.

6 years agorendernode: Don't clip when _draw()ing
Timm Bäder [Wed, 12 Feb 2020 15:22:44 +0000 (16:22 +0100)]
rendernode: Don't clip when _draw()ing

6 years agorendernode: draw() only in node bounds
Timm Bäder [Wed, 12 Feb 2020 15:21:38 +0000 (16:21 +0100)]
rendernode: draw() only in node bounds

We currently disable when draw()ing nodes using the cairo fallback path,
which means we can't just use cairo_paint(). Use a proper rectangle
instead.

Fixes #2431

6 years agoMerge branch 'cleanup-profiler-marks' into 'master'
Matthias Clasen [Wed, 12 Feb 2020 14:31:40 +0000 (14:31 +0000)]
Merge branch 'cleanup-profiler-marks' into 'master'

Cleanup profiler marks

See merge request GNOME/gtk!1417

6 years agoprofile: Use separate names for frameclock marks instead of using details
Alexander Larsson [Wed, 12 Feb 2020 13:26:16 +0000 (14:26 +0100)]
profile: Use separate names for frameclock marks instead of using details

The marks are averaged based on the name, so this makes more sense.

Also rename the map/unmap marks to have the same capitalization as
everything else.

6 years agoprofiler: Add G_GNUC_PRINTF markers to silence warnings
Alexander Larsson [Wed, 12 Feb 2020 12:35:42 +0000 (13:35 +0100)]
profiler: Add G_GNUC_PRINTF markers to silence warnings

I was getting CI failures like:
 ../gdk/gdkprofiler.c: In function ‘add_markvf’:
../gdk/gdkprofiler.c:111:3: error: function ‘add_markvf’ might be a candidate for ‘gnu_printf’ format attribute [-Werror=suggest-attribute=format]

6 years agoprofiler: Clean up profiler marks for frameclock
Alexander Larsson [Wed, 12 Feb 2020 12:20:32 +0000 (13:20 +0100)]
profiler: Clean up profiler marks for frameclock

This drops the marks for before/after-paint as they are internal
things that very rarely use any time, and also flush/resume-events
as any events reported here will get separate marks so will be easy
to see anyway.

Also, we rename the entire frameclock cycle to "frameclock cycle"
rather than "paint_idle" which is rather cryptic.

6 years agoprofiler: Add profiler marks for when surfaces are mapped and unmapped
Alexander Larsson [Wed, 12 Feb 2020 11:44:43 +0000 (12:44 +0100)]
profiler: Add profiler marks for when surfaces are mapped and unmapped

This is done on the X11 side so we can tell when the map/unmap takes
effect and how long it took since the map request.

6 years agoicon theme: Don't add profiler marks for short async icon loads
Alexander Larsson [Wed, 12 Feb 2020 11:43:27 +0000 (12:43 +0100)]
icon theme: Don't add profiler marks for short async icon loads

We do a bunch of preloads, and most of these are not really interesting
anyway.

6 years agobuilder: Don't add profiler marks for short parses
Alexander Larsson [Wed, 12 Feb 2020 11:42:29 +0000 (12:42 +0100)]
builder: Don't add profiler marks for short parses

There are a lot of thes (since we're recursing), so don't spew the
output with uninteresting ones.

6 years agoprofiler: Add _end_mark() version of _add_mark()
Alexander Larsson [Wed, 12 Feb 2020 10:25:34 +0000 (11:25 +0100)]
profiler: Add _end_mark() version of _add_mark()

These don't take a duration, instead they call g_get_monotonic_time() to
and subtract the start time for it.

Almost all our calls are like this, and this makes the callsites clearer
and avoids inlining the clock call into the call site.

6 years agoprofiler: Make profiler-is-running a macro
Alexander Larsson [Wed, 12 Feb 2020 10:05:01 +0000 (11:05 +0100)]
profiler: Make profiler-is-running a macro

When we use if (GDK_PROFILER_IS_RUNNING) this means we get an
inlined if (FALSE) when the compiler support is not compiled in, which
gets rid of all the related code completely.

We also expand to  G_UNLIKELY(gdk_profiler_is_running ()) in the supported
case which might cause somewhat better code generation.

6 years agoConvert all profiler times from nsec to usec
Alexander Larsson [Wed, 12 Feb 2020 09:44:17 +0000 (10:44 +0100)]
Convert all profiler times from nsec to usec

usec is the scale of the monotonic timer which is where we get almost
all the times from. The only actual source of nsec is the opengl
GPU time (but who knows what the actual resulution of that is).

Changing this to usec allows us to get rid of " * 1000" in a *lot* of
places all over the codebase, which are ugly and confusing.

6 years agoUse markf in one more place
Alexander Larsson [Wed, 12 Feb 2020 09:40:26 +0000 (10:40 +0100)]
Use markf in one more place

6 years agofixup! Add gdk_profiler_add_markf() to do printf formating
Alexander Larsson [Wed, 12 Feb 2020 09:11:53 +0000 (10:11 +0100)]
fixup! Add gdk_profiler_add_markf() to do printf formating

6 years agoUse the new gdk_profiler_add_markf() function
Alexander Larsson [Wed, 12 Feb 2020 08:59:03 +0000 (09:59 +0100)]
Use the new gdk_profiler_add_markf() function

6 years agoAdd gdk_profiler_add_markf() to do printf formating
Alexander Larsson [Wed, 12 Feb 2020 08:56:10 +0000 (09:56 +0100)]
Add gdk_profiler_add_markf() to do printf formating

This allows us to avoid hand-rolling g_strdup_printf calls,
but also moves the printf into the called function where
it doesn't bloat the code of the calling function if the profiler
is not running.

6 years agoUpdate Esperanto translation
Carmen Bianca BAKKER [Wed, 12 Feb 2020 07:43:58 +0000 (07:43 +0000)]
Update Esperanto translation

6 years agoMerge branch 'mark-events-with-kind' into 'master'
Matthias Clasen [Tue, 11 Feb 2020 19:01:10 +0000 (19:01 +0000)]
Merge branch 'mark-events-with-kind' into 'master'

profiler: Always mark events with the kind

See merge request GNOME/gtk!1414

6 years agoprofiler: Always mark events with the kind
Alexander Larsson [Tue, 11 Feb 2020 18:51:49 +0000 (19:51 +0100)]
profiler: Always mark events with the kind

6 years agoMerge branch 'wip/baedert/outline-radius' into 'master'
Matthias Clasen [Tue, 11 Feb 2020 16:59:11 +0000 (16:59 +0000)]
Merge branch 'wip/baedert/outline-radius' into 'master'

css boxes: Compute box from border box

Closes #2425

See merge request GNOME/gtk!1408

6 years agoMerge branch 'share-gl-programs' into 'master'
Emmanuele Bassi [Tue, 11 Feb 2020 16:30:16 +0000 (16:30 +0000)]
Merge branch 'share-gl-programs' into 'master'

GskGLRenderer: Share programs between different renderers in same display

See merge request GNOME/gtk!1413

6 years agoMerge branch 'handle-max-texture-size' into 'master'
Matthias Clasen [Tue, 11 Feb 2020 16:23:02 +0000 (16:23 +0000)]
Merge branch 'handle-max-texture-size' into 'master'

gl: Don't require too large texture for offsreen ops

See merge request GNOME/gtk!1411

6 years agoMerge branch 'fix-angle-normalization' into 'master'
Matthias Clasen [Tue, 11 Feb 2020 16:17:14 +0000 (16:17 +0000)]
Merge branch 'fix-angle-normalization' into 'master'

gsk: Fix angle normalization

See merge request GNOME/gtk!1409

6 years agoGskGLRenderer: Share programs between different renderers in same display
Alexander Larsson [Tue, 11 Feb 2020 15:56:24 +0000 (16:56 +0100)]
GskGLRenderer: Share programs between different renderers in same display

This is similar to how we share texture atlases. Some added complexity
in that the program state also needed to be shared, so it had to move to
the shared Programs object.

With this change realization of additional GskRenderers when opening
popups went from ~60msec to ~35 msec on average.

6 years agoMerge branch 'ci-jobs' into 'master'
Emmanuele Bassi [Tue, 11 Feb 2020 15:25:48 +0000 (15:25 +0000)]
Merge branch 'ci-jobs' into 'master'

Update CI jobs

See merge request GNOME/gtk!1412

6 years agoci: Allow release job to fail
Emmanuele Bassi [Tue, 11 Feb 2020 15:12:14 +0000 (15:12 +0000)]
ci: Allow release job to fail

We need to investigate release-specific failures, and possibly fix our
test suite to deal with debug-specific poking through the internal
state.

For the time being, allow the release job to fail, but this is strictly
a temporary measure.

6 years agotestsuite/gtk: Add xfail tests
Emmanuele Bassi [Tue, 11 Feb 2020 14:53:57 +0000 (14:53 +0000)]
testsuite/gtk: Add xfail tests

Some tests are expected to fail.

Some tests are expected to fail in non-debugging builds.

The icontheme test is the latter.

6 years agovulkan: Handle VK_ERROR_UNKNOWN
Emmanuele Bassi [Tue, 11 Feb 2020 14:07:43 +0000 (14:07 +0000)]
vulkan: Handle VK_ERROR_UNKNOWN

A newly added Vulkan 1.2 error.

6 years agoci: Add a style check pass
Emmanuele Bassi [Tue, 11 Feb 2020 13:28:16 +0000 (13:28 +0000)]
ci: Add a style check pass

We should at least try to get the style check done by a machine, as
opposed to have humans go through merge requests with coding style
nitpicks.

6 years agoci: Generate the report for the release builds
Emmanuele Bassi [Wed, 29 Jan 2020 13:36:23 +0000 (14:36 +0100)]
ci: Generate the report for the release builds

We are going to need to hide the report generation into the test runner
script, as we want the job to produce the reports even in case of
failure, instead of bailing out immediately.

6 years agoDeclare global counters only in debug builds
Emmanuele Bassi [Wed, 29 Jan 2020 13:03:56 +0000 (14:03 +0100)]
Declare global counters only in debug builds

6 years agoSilence compiler warnings in non-debug builds
Emmanuele Bassi [Wed, 29 Jan 2020 13:02:41 +0000 (14:02 +0100)]
Silence compiler warnings in non-debug builds

6 years agoDeclare global counter only in debug builds
Emmanuele Bassi [Wed, 29 Jan 2020 13:02:16 +0000 (14:02 +0100)]
Declare global counter only in debug builds

6 years agoMerge branch 'fix-revealer-scale' into 'master'
Matthias Clasen [Tue, 11 Feb 2020 14:06:36 +0000 (14:06 +0000)]
Merge branch 'fix-revealer-scale' into 'master'

revealer: Fix child size allocation at small scales

See merge request GNOME/gtk!1410

6 years agoci: Add missing newline escape
Emmanuele Bassi [Wed, 29 Jan 2020 11:31:00 +0000 (12:31 +0100)]
ci: Add missing newline escape

Otherwise we're going to run the accessibility test suite, and it's
going to fail.

6 years agoci: Remove static build job
Emmanuele Bassi [Wed, 29 Jan 2020 11:23:06 +0000 (12:23 +0100)]
ci: Remove static build job

Now that we figured out why the build was failing on Fedora when the
profiling code was enabled, we can go back to building both shared and
static libraries by default.

6 years agoci: Use the new Fedora image
Emmanuele Bassi [Wed, 29 Jan 2020 11:18:09 +0000 (12:18 +0100)]
ci: Use the new Fedora image