Commit Graph

27298 Commits

Author SHA1 Message Date
Georges Basile Stavracas Neto
bc41a88c92 clutter/stage: Replace _cogl_util_get_eye_planes_for_screen_poly()
_cogl_util_get_eye_planes_for_screen_poly() is quite a complicated beast. Ever
since Clutter became a compositor toolkit, and specially after we switched to
graphene_frustum_t on paint volumes, we can brutally simplify this function.

The new code assumes camera is at (0, 0, 0) at world coordinates (i.e. before
applying the projection). We also consider that the redraw clip are at stage
coordinates. That means that converting the clip rectangle to world rectangle
is simply a matter of projecting the corresponding vertices using the "view"
matrix. Furthermore, we only need to project the top-left, and bottom-right
vertices, since top-right and bottom-left can be derived from those two.

The frustum setup still uses triplets of vertices to setup the planes, except
now the first vertex is always the camera (hardcoded to 0, 0, 0), and the other
two vertices are the projected clip rectangle vertices.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:46:26 -03:00
Georges Basile Stavracas Neto
f7d0461768 clutter/stage: Simplify frustum setup function
We guarantee to never pass NULL clips anymore, so there's no need
to check for such case.

Remove the check for NULL clip, and remove all related variables.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:46:26 -03:00
Georges Basile Stavracas Neto
2ca20783e5 clutter/actor: Remove region culling
We can trust the clip frusta array to encode this information now.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:46:26 -03:00
Georges Basile Stavracas Neto
d1d3ac6b20 clutter/stage: Use multiple frusta depending on the redraw clip
The redraw clip region may contain multiple clip rectangles. We currently
only use the extents of this region, but having multiple frusta for each
rectangle is a better alternative, and will allow us to remove the extra
projection we currently do.

Make the clip frustum an array, with multiple frusta.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:46:23 -03:00
Georges Basile Stavracas Neto
63d05565f6 clutter/stage: Move pick cache invalidation out of setup
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
cf9d06164d clutter/stage: Remove outdated comment
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
feb8397c97 clutter/stage: Move clip frustum to ClutterPaintContext
The clip planes / frustum are contextual to painting. In the past, for
the lack of a better place, it was added to ClutterStage, but now we
have an appropriate home for such data: ClutterPaintContext.

Move the frustum to the paint context.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
b6fa26cf45 clutter/stage: Don't setup view on pick
While refactoring the clipping planes / frustum code, it became more and
more evident that we do not need to update them while picking. Picking
nowadays goes through a completely different code path, that does not
rely on paint volume culling.

While it might be interesting to eventually also cull out based on paint
volumes, it certainly won't go through the painting code anymore.

Remove setting up the view when picking, and rename functions appropriatedly.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
496aea51af clutter: Remove CLUTTER_CULL_RESULT_PARTIAL
Culling paint volumes don't give this level of detail anymore, and in
fact knowing whether it was partially or fully in was only being used
in a debug path. For the purposes of culling, it doesn't matter if a
given actor is partially or completely inside the frustum; either way,
it must be painted.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
80bd44cba0 clutter/actor: Use switch instead of if/else
To improve legibility of the code.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
a84dccfd64 clutter/stage: Use graphene_frustum_t for clipping
Instead of 4 planes, use a graphene_frustum_t to store the clipping
planes.

The cautious reviewer might noticed that we are now setting up 6
planes: the 4 planes we were doing before, plus 2 extra planes in
the Z axis. These extra planes simulate an "infinite" Z far, and
an "on-camera" Z near.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
338a9275b4 clutter: Remove ClutterPlane
It is unused now.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
964229acf4 clutter/actor: Trivial code shuffling
Retrieving the stage from the actor is almost free, but this is a
hot path anyway and we can bail out before that.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
01ccc54ce8 clutter/actor: Remove unnecessary check
The stage clip is *never* NULL - it is a structure field of ClutterStage
itself.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
793ca68d8c clutter/stage: Use graphene_plane_t for clipping planes
It allows us to remove quite a bunch of code, and not deal with part
of the mind-melting maths behind it.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
175851eef7 clutter/stage: Simplify view setup
ClutterStage defines the 8 vertices of a frustum:

  4 ----------------------------- 5
  | \                           / |
  |  \                         /  |
  |   0 --------------------- 1   |
  |   |                       |   |
  |   |                       |   |
  |   3 --------------------- 2   |
  |  /                         \  |
  | /                           \ |
  7 ----------------------------- 6

Then, it uses triplets of vertices to create each  clipping plane.
It only sets up 4 planes (it doesn't clip based on depth), defined
by the following vertices:

 * 0 - 4 - 5
 * 1 - 5 - 6
 * 2 - 6 - 7
 * 0 - 7 - 4

The first 3 triplets are selected using the for-loop. However, the
last triplet is different, and is done out of the loop. It could
have been made simpler by using the "3 - 7 - 4" triplet.

Simplify the current code by using the suggested triplet, calculated
inside the for-loop.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
1f99395d20 clutter: Use graphene_matrix_inverse()
Instead of our own implementation that upscales, then downscales back,
use graphene_matrix_inverse() directly. This is possible after switching
to a z-near value that doesn't have problems with float precision.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
78c648f947 cluter/stage: Use 1.0 for z-near
It doesn't actually matter, since we don't really have cases where we
cross this value, but it's enough to prevent catastrophic cancellation
due to very small float numbers.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
d08f724bc3 clutter/actor: Round to 256ths when projecting for picking
Picking is specially sensitive for float precision, and tests can
easily fail when something changes, even if ever so slightly. A
simple way to workaround this is by adjusting the projected points
using the same procedure described at 67cc60cbda.

Round projected points for picking to 256ths.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Georges Basile Stavracas Neto
8809ee9e26 clutter/util: Generalize ROUND_TO_256THS
It'll be reused in other bits of the Clutter codebase. Move it to
an inline function in clutter-private.h

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
2020-10-16 13:37:44 -03:00
Jonas Ådahl
5a58ccbece texture-tower: Use memory management helpers more
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
1cc82cf759 background: Use memory management helpers more
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
2aea404b14 cogl: Use memory management helpers more
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
29cdc35654 clutter/offscreen-effect: Use memory management helpers more
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
88c830b9ab cogl/framebuffer: Move allocation and y-flip-ness to non-abstract types
Now that CoglFramebuffer is a GObject, we can start moving things to the
proper places.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
eb14da3874 cogl: Turn CoglFramebuffer, CoglOffscreen and CoglOnscreen into GObjects
A first step towards abandoning the CoglObject type system: convert
CoglFramebuffer, CoglOffscreen and CoglOnscreen into GObjects.
CoglFramebuffer is turned into an abstract GObject, while the two others
are currently final. The "winsys" and "platform" are still sprinkled
'void *' in the the non-abstract type instances however.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
de4e59a39b clutter: Don't use CoglHandle to store framebuffers
Better just not lose type informatoin.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
ea2f4efaef cogl: Remove cogl_offscreen_new_to_texture()
It's no longer used, and has been deprecated for a long time.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
4b939439ac tests/clutter: Stop using cogl_offscreen_new_to_texture()
Use the non-deprecated variant.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
b947dced40 clutter: Stop using cogl_offscreen_new_to_texture()
It's deprecated, drops errors silently, so change to the proper
function.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
e7ac3e2715 cogl/framebuffer: Clean up cleanup function
Use cogl_clear_object() and g_clear_pointer() helpers.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
aa99e505ad cogl/framebuffer: Move public struct fields to private struct
This will limit accidental manipulation of the framebuffer struct
fields, while also paving the way to change to GObject type hierarchy.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
ffde1c0124 cogl/onscreen: Remove unused _cogl_onscreen_new()
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
209b78afd8 cogl/framebuffer/gl: Move OpenGL driver fields to private struct
It's driver specific, so it shouldn't be kept in the generic data
structure.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:38 +00:00
Jonas Ådahl
e720ef1ceb cogl/framebuffer: Fix argument naming to blit_framebuffer()
The first argument is the framebuffer operated on, so in order to stay
consistest, rename 'src' to 'framebuffer'. The second is the
destination. The destination is commonly referred to as 'dst' elsewhere,
so rename 'dest' to 'dst'.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:37 +00:00
Jonas Ådahl
4edcc21a5c cogl/framebuffer: Remove alignment of struct fields
They weren't all aligned anyway.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:37 +00:00
Jonas Ådahl
e7b923289a cogl/framebuffer: Fix enum declaration coding style
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
2020-10-16 16:17:37 +00:00
Ray Strode
79e5ec57d2 cally: fix state set leak
cally_actor_action_do_action leaks a state set object in the
case where the actor is defunct, insensitive, or hidden.

This commit plugs the leak.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1225
2020-10-15 08:58:37 +00:00
Niels De Graef
3bc15ade04 ci: Generate a cover report for the test suite
Based on @ebassi's work from https://gitlab.gnome.org/GNOME/glib/merge_requests/769

This MR allows us to see the succeeded/failed tests each time the CI
pipeline is run.

https://gitlab.gnome.org/GNOME/mutter/merge_requests/528
2020-10-15 10:18:44 +02:00
Daniel van Vugt
df76d3deec renderer-native: Get the GBM buffer format from EGL to match the framebuffer
This eliminates two locations where the format was hardcoded.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1491
2020-10-15 07:45:06 +00:00
Robert Mader
ae0d48b6fd wayland/subsurface: Only show subsurface if parent is mapped
The spec states:
```
A sub-surface becomes mapped, when a non-NULL wl_buffer is applied
and the parent surface is mapped. The order of which one happens
first is irrelevant. A sub-surface is hidden if the parent becomes hidden,
or if a NULL wl_buffer is applied. These rules apply recursively
through the tree of surfaces.
```

In the past we relied on Clutter actor behaviour to realize the recursive
part - which then broke in
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/664
when we changed the actor hierachy in regards to subsurfaces.

Explicitly encode the desired behaviour in `MetaWaylandSubsurface`, fixing
the issue and making it future proof.

Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1384
2020-10-14 18:55:21 +02:00
Daniel van Vugt
df65f05e27 cogl: Remove unused swap-related COGL_FEATURE_ID_ flags
What we actually use is `COGL_WINSYS_FEATURE_SWAP_BUFFERS_EVENT`, which
is also set in the same locations.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1297
2020-10-14 16:27:23 +00:00
Peter Hutterer
e078a007d6 backends: add support for scroll button locking
Where enabled, the first click of the scroll button sets the button logically
down, the second click sets the button logically up.

This is an accessibility feature, it doesn't require users to keep holding the
button down while scrolling which is hard or impossible for some users.

gsettings-desktop-schemas merge request:
https://gitlab.gnome.org/GNOME/gsettings-desktop-schemas/-/merge_requests/39

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1432
2020-10-14 16:02:29 +00:00
Robert Mader
1c1c8b25ed window-actor: Freeze subsurfaces as well
In order to prevent glitches in animations, make sure we freeze all
surfaces of a window actor.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1339
2020-10-14 15:53:53 +00:00
Florian Müllner
22902a5e2c window: "Hide" edge resistance behind modifier key
Aligning windows manually with other windows has become less important
since the advent of tiling. This decreases the usefulness of edge
resistance, which in fact many users perceive as lag nowadays.

Account for that by limiting resistance to screen and monitor edges by
default, and only include windows when the control key is pressed.

https://bugzilla.gnome.org/show_bug.cgi?id=679609
2020-10-14 15:45:24 +00:00
Florian Müllner
f9edb6bad3 edge-resistance: Replace booleans with flags parameter
Multiple boolean parameters don't make for great API, so before we
add another one, replace them with a single flag parameter.

https://bugzilla.gnome.org/show_bug.cgi?id=679609
2020-10-14 15:45:23 +00:00
Florian Müllner
64ced1632e window: Don't override tile monitor
Commit 033f0d11bf added a fallback in case the tile monitor wasn't
set before, but didn't actually check for a previously set value.
As a result, the "fallback" is not set unconditionally, which may
differ from the expected monitor: The tile monitor is determined
by the pointer position, while the window's monitor is the one
where the biggest part of the window resides on.

https://gitlab.gnome.org/GNOME/mutter/-/issues/1389
2020-10-14 15:30:05 +00:00
Jonas Ådahl
43c6f70605 util: Don't expand meta_*() debug log arguments if topic not enabled
It's pointless to call into functions that produce information that will
end up nowhere, so lets not. This will generate less angst when doing
more intense data gathering and string generation in debug log calls.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1467
2020-10-14 14:29:36 +00:00
Bastien Nocera
f79d40077e meson: Fail if initfd is enabled but Xwayland doesn't support it
Fedora's mutter is built with --auto-features=enabled which enables
-initfd support. But the current meson code doesn't check whether
-initfd is available in Xwayland, leading to the wayland session not
starting up.

Check for -initfd's availability before enabling it.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1103
2020-10-14 15:58:40 +02:00
Corentin Noël
0730ff5b9e clutter/frame-clock: Skip the ClutterFrameListenerIface from the introspection
This can't possibly be used by introspected languages as it is a raw struct.

https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1413
2020-10-13 10:56:50 +02:00