This updates the inner loops of the cull function so now the vertices of
the polygon being culled are iterated in the inner loop instead of the
clip planes and we count how many vertices are outside the current
plane so we can bail out immediately if all the vertices are outside of
any plane and so we can correctly track partial intersections with the
clip region.
The previous approach could catch some partial intersections but for
example a rectangle that was larger than the clip region centred over
the clip region with all corners outside would be reported as outside,
not as a partial intersection.
In 047227fb cogl_atlas_new was changed so that it can take a flags
parameter to specify whether to clear the new atlases and whether to
copy images to the new atlas after reorganisation. This was done so
that the atlas code could be shared with the glyph cache. At some
point during the development of this patch the flag was just a single
boolean instead and this is accidentally how it is used from the glyph
cache. The glyph cache therefore passes 'TRUE' as the set of flags
which means it will only get the 'clear' flag and not the
'disable-migration' flag. When the glyph cache gets full it will
therefore try to copy the texture to the new atlas as well as
redrawing them with cairo. This causes problems because the glyph
cache needs to work in situations where there is no FBO support.
In _cogl_pipeline_prune_empty_layer_difference if the layer's parent
has no owner then it just takes ownership of it. However this could
theoretically end up taking ownership of the root layer because
according to the comment above in the same function that should never
have an owner. This patch just adds an extra check to ensure that the
unowned layer has a parent.
http://bugzilla.clutter-project.org/show_bug.cgi?id=2588
In _cogl_pipeline_prune_empty_layer_difference if we are reverting to
the immediate parent of an empty/redundant layer then it is not enough
to simply add a reference to the pipeline's ->layer_differences list
without also updating parent_layer->owner to point back to its new
owner.
This oversight was leading us to break the invariable that all layers
referenced in layer_differences have an owner and was also causing us to
break another invariable whereby after calling
_cogl_pipeline_layer_pre_change_notify the returned layer must always be
owned by the given 'required_owner'.
http://bugzilla.clutter-project.org/show_bug.cgi?id=2588
When we come to presenting the result of a clipped redraw to the front
buffer with a blit we need to ensure that all the rendering is done,
otherwise redraw operations that are slower than the framerate can queue
up in the pipeline during a heavy animation, causing a larger and larger
backlog of rendering visible as lag to the user.
Note: Since calling glFinish() and sycnrhonizing the CPU with the GPU is
far from ideal, we hope that this is only a short term solution.
One idea is to using sync objects to track render completion so we can
throttle the backlog (ideally with an additional extension that lets us
get notifications in our mainloop instead of having to busy wait for the
completion.)
Another option is to support clipped redraws by reusing the contents of
old back buffers such that we can flip instead of using a blit and then
we can use GLX_INTEL_swap_events to throttle. For this though we would
still probably want an additional extension so we can report the limited
region of the window damage to X/compositors.
Thanks to Owen Taylor and Alexander Larsson for reporting the problem.
clutter_clone_get_paint_volume was being exported from the shared
library because the function wasn't declared static. This function
shouldn't be exposed because it should be accessed through
clutter_actor_get_paint_volume.
The texture containing the image for the redirected actor will always
be painted at a 1:1 texel:pixel ratio so there's no need to use linear
filtering. This should also counteract some of the effects of rounding
errors when calculating the geometry for the quad.
if cross compiling clutter using mingw using an out of tree build
directory then a pre-requisite for creating the resources.o file
containing the transparent cursor is for the win32 directory itself to
be created at $(top_builddir)/clutter/win32.
glib already has a data type to manage a list of callbacks called a
GHookList so we might as well use it instead of maintaining Cogl's own
type. The glib version may be slightly more efficient because it
avoids using a GList and instead encodes the prev and next pointers
directly in the GHook structure. It also has more features than
CoglCallbackList.
Previously we were applying the culling optimization to any actor
painted without considering that we may be painting to an offscreen
framebuffer where the stage clip isn't applicable.
For now we simply expose a getter for the current draw framebuffer
and we can assume that a return value of NULL corresponds to the
stage.
Note: This will need to be updated as stages start to be backed by real
CoglFramebuffer objects and so we won't get NULL in those cases.
To give quick visibility to the things going on relating to clipping and
culling this adds some more CLIPPING debug notes to clutter-actor.c and
clutter-stage.c
As documented in cogl-pipeline-private.h, there is a precedence to the
ClutterPaintVolume bitfields that should be considered whenever we
implement code that manipulates PaintVolumes...
Firstly if ->is_empty == TRUE then the values for ->is_complete and
->is_2d are undefined, so we should typically check ->is_empty as the
first priority.
This fixes a bug in _clutter_paint_volume_cull() whereby we were
checking pv->is_complete before checking pv->is_empty which was
resulting in assertions for actors with no size.
Drawing and clipping to paths is generally quite expensive because the
geometry has to be tessellated into triangles in a single VBO which
breaks up the journal batching. If we can detect when the path
contains just a single rectangle then we can instead divert to calling
cogl_rectangle which will take advantage of the journal, or by pushing
a rectangle clip which usually ends up just using the scissor.
This patch adds a boolean to each path to mark when it is a
rectangle. It gets cleared whenever a node is added or gets set to
TRUE whenever cogl2_path_rectangle is called. This doesn't try to
catch cases where a rectangle is composed by cogl_path_line_to and
cogl_path_move_to commands.
ClutterDragAction should be able to use the newly added ClutterSettings
property exposing the system's drag threshold.
Currently, the x-drag-threshold and the y-drag-threshold properties (and
relative accessors) use an unsigned integer for their values; we should
be able to safely expand the range to include -1 as the minimum value,
and use this new value to tell the ClutterDragAction that it should query
the ClutterSettings object for the drag threshold.
The storage of the properties has been changed, albeit in a compatible
way, as GObject installs a uint ↔ int transformation function for GValue
automatically.
The setter for the drag thresholds has been changes to use a signed
integer, but the getter has been updated to always Do The Right Thing™:
it never returns -1 but, instead, will return the valid drag threshold,
either from the value set or from the Settings singleton.
This change is ABI compatible.
http://bugzilla.clutter-project.org/show_bug.cgi?id=2583
ClutterState is missing some documentation on how to define transitions
using ClutterScript definitions; it is also missing some example code
for both the C API and the ClutterScript syntax.
The allocation of the ClutterBox is not enough to be used as the paint
volume, because children might decide to paint outside that area.
Instead, we should use the allocation if the Box has a background color
and then do what Group does, and union all the paint volumes of the
children.
http://bugzilla.clutter-project.org/show_bug.cgi?id=2600
In 9ff04e8a99 the builtin uniforms were moved to the common shader
boilerplate. However the common boilerplate is positioned before the
default precision specifier on GLES2 so it would fail to compile
because the uniforms end up with no precision in the fragment
shader. This patch just moves the precision specifier to above the
common boilerplate.
The CLUTTER_ENTER and CLUTTER_LEAVE event types were mistakenly ignored
by clutter_event_get_device(), when returning the device from a
non-allocated ClutterEvent.
There's an optimisation in clutter-actor.c to avoid calculating the
last known paint volume whenever culling and clipped redraws are both
disabled. However there was a small thinko in the logic so that it
would also avoid calculating the paint volume whenever only one of the
debug flags is enabled. This fixes it to explicitly check that the two
flags are not both enabled before skipping the paint volume
calculation.
Instead of unconditionally combining the modelview and projection
matrices and then iterating each of the vertices to call
cogl_matrix_transform_point for each one in turn we now only combine the
matrices if there are more than 4 vertices (with less than 4 vertices
its less work to transform them separately) and we use the new
cogl_vertex_{transform,project}_points APIs which can hopefully
vectorize the transformations.
Finally the perspective divide and viewport scale is done in a separate
loop at the end and we don't do the spurious perspective divide and
viewport scale for the z component.
Previously each time we needed to retrieve the model transform for a
given actor we would call the apply_transform vfunc which would build up
a transformation matrix based on the actor's current anchor point, its
scale, its allocation and rotation. The apply_transform implementation
would repeatedly call API like cogl_matrix_rotate, cogl_matrix_translate
and cogl_matrix_scale.
All this micro matrix manipulation APIs were starting to show up in the
profiles of dynamic applications so this adds priv->transform matrix
cache which maintains the combined result of the actors scale, rotation
and anchor point etc. Whenever something like the rotation changes then
then the matrix is marked as dirty, but so long as the matrix isn't
dirty then the apply_transform vfunc now just calls cogl_matrix_multiply
with the cached transform matrix.
This implements a variation of frustum culling whereby we convert screen
space clip rectangles into eye space mini-frustums so that we don't have
to repeatedly transform actor paint-volumes all the way into screen
coordinates to perform culling, we just have to apply the modelview
transform and then determine each points distance from the planes that
make up the clip frustum.
By avoiding the projective transform, perspective divide and viewport
scale for each point culled this makes culling much cheaper.
This simplifies the implementation of the ClutterStage apply_transform
vfunc by using the new cogl_matrix_view_2d_in_perspective utility API
which can setup up a view transform for a given perspective so that a
cross section of the view frustum at an arbitrary depth can be mapped
directly to 2D stage coordinates with (0,0) at the top left.
This adds two new experimental functions to cogl-matrix.c:
cogl_matrix_view_2d_in_perspective and cogl_matrix_view_2d_in_frustum
which can be used to setup a view transform that maps a 2D coordinate
system (0,0) top left and (width,height) bottom right to the current
viewport.
Toolkits such as Clutter that want to mix 2D and 3D drawing can use
these APIs to position a 2D coordinate system at an arbitrary depth
inside a 3D perspective projected viewing frustum.
Firstly Clutter shouldn't be using OpenGL directly so this needed
changing but also conceptually it doesn't make sense for
clutter_stage_read_pixels to validate the requested area to read against
the viewport it would make more sense to compare against the window
size. Finally checking that the width of the area is less than the
viewport or window width without considering the x isn't enough to know
if the area extends outside the windows bounds. (same for the height)
This patch removes the validation of the read area from
clutter_stage_read_pixels and instead we now simply rely on the
semantics of cogl_read_pixels for reading areas outside the window
bounds.
OpenGL < 4.0 only supports integer based viewports and internally we
have a mixture of code using floats and integers for viewports. This
patch switches all viewports throughout clutter and cogl to be
represented using floats considering that in the future we may want to
take advantage of floating point viewports with modern hardware/drivers.
This makes a change to the original point_in_poly algorithm from:
http://www.ecse.rpi.edu/Homepages/wrf/Research/Short_Notes/pnpoly.html
The aim was to tune the test so that tests against screen aligned
rectangles are more resilient to some in-precision in how we transformed
that rectangle into screen coordinates. In particular gnome-shell was
finding that for some stage sizes then row 0 of the stage would become a
dead zone when going through the software picking fast-path and this was
because the y position of screen aligned rectangles could end up as
something like 0.00024 and the way the algorithm works it doesn't have
any epsilon/fuz factor to consider that in-precision.
We've avoided introducing an epsilon factor to the comparisons since we
feel there's a risk of changing some semantics in ways that might not be
desirable. One of those is that if you transform two polygons which
share an edge and test a point close to that edge then this algorithm
will currently give a positive result for only one polygon.
Another concern is the way this algorithm resolves the corner case where
the horizontal ray being cast to count edge crossings may cross directly
through a vertex. The solution is based on the "idea of Simulation of
Simplicity" and "pretends to shift the ray infinitesimally down so that
it either clearly intersects, or clearly doesn't touch". I'm not
familiar with the idea myself so I expect a misplaced epsilon is likely
to break that aspect of the algorithm.
The simple solution this patch applies is to pixel align the polygon
vertices which should eradicate most noise due to in-precision.
https://bugzilla.gnome.org/show_bug.cgi?id=641197
Anything that is not CLUTTER_INIT_SUCCESS is to be considered an error.
This fixes the Clutter initialization sequence to actually error out
on pre-conditions and backend initialization failures.
clutter_offscreen_effect_pre_paint was using the unitialized value of
the ‘box’ variable whenever the actor doesn't have a paint
volume. This patch makes it just set the offset to 0,0 instead.
When removing the opacity override in the post_paint implementation,
ClutterOffscreenEffect would always set the override back to -1. This
ends up cancelling out the effect of any overrides from outer effects
which means that if any actor has multiple effects attached then it
would apply the opacity multiple times.
To fix this, the effect now preserves the old value of the opacity
override and restores that instead of setting -1.
http://bugzilla.clutter-project.org/show_bug.cgi?id=2541
This is needed if an effect wants to temporarily override the paint
opacity. It needs to be able to restore the old opacity override in
the post_paint handler otherwise it would replace the effect of the
opacity override from any outer effects.
http://bugzilla.clutter-project.org/show_bug.cgi?id=2541
