commit e2c4a2a9f8 fixed one thing but broke many others things :-/
hopfully this fixes that.
It turned out that the journal was mistakenly setting the OVERRIDE_LAYER0
flush option for all entries, but some other logic errors were also
uncovered in _cogl_material_equal.
To help us handle sliced textures; When flushing materials there is an
override option that can be given to replace the texture name for layer0
so we may iterate the slices without needing to modify the material
in use.
Since improving the journal's ability to batch state changes we added a
_cogl_material_equals function that is used by the journal to compare
materials and identify when a state change is required, but this wasn't
correctly considering the layer0 override resulting in false positives that
meant the journal wouldn't update the GL state and the first texture name
was used for all slices.
The cost of glGetFloatv with Mesa is still representing a majority of our
time in OpenGL for some applications, and the last thing left using this is
the current-matrix API when getting the projection matrix.
This adds a matrix stack for the projection matrix, so all getting, setting
and modification of the projection matrix is now managed by Cogl and it's only
when we come to draw that we flush changes to the matrix to OpenGL.
This also brings us closer to being able to drop internal use of the
deprecated OpenGL matrix functions, re: commit 54159f5a1d
Fixes and adds a unit test for creating and drawing using materials with
COGL_INVALID_HANDLE texture layers.
This may be valid if for example the user has set a texture combine string
that only references a constant color.
_cogl_material_flush_layers_gl_state will bind the fallback texture for any
COGL_INVALID_HANDLE layer, later though we could explicitly check when the
current blend mode does't actually reference a texture source in which case
binding the fallback texture is redundant.
This tests drawing using cogl_rectangle, cogl_polygon and
cogl_vertex_buffer_draw.
This simplifies the vertex data uploading in the journal, and could improve
performance. Modifying a VBO mid-scene could reqire synchronizing with the
GPU or some form of shadowing/copying to avoid modifying data that the GPU
is currently processing; the buffer was also being marked as GL_STATIC_DRAW
which could have made things worse.
Now we simply create a GL_STATIC_DRAW VBO for each flush and and delete it
when we are finished.
Since most Clutter actors aren't much more than textured quads; flushing the
journal typically involves lots of 'change modelview; draw quad' sequences.
The amount of overhead involved in uploading a new modelview and queuing
that primitive is huge in comparison to simply transforming 4 vertices by
the current modelview when logging quads. (Note if your GPU supports HW
vertex transform, then it still does the projective and viewport transforms)
At the same time a --cogl-debug=disable-software-transform option has been
added for comparison and debugging.
This change allows typical pick scenes to be batched into a single draw call
and I'm seeing test-pick run over 200% faster with this. (i965 + Mesa
7.6-devel)
Enabling this option makes Cogl trace how the journal is managing to batch
your rectangles. The journal staggers how it emmits state to the GL driver
and the batches will normally get smaller for each stage, but ideally you
don't want to be in a situation where Cogl is only able to draw one quad per
modelview change and draw call.
E.g. this is a fairly ideal example:
BATCHING: journal len = 101
BATCHING: vbo offset batch len = 101
BATCHING: material batch len = 101
BATCHING: modelview batch len = 101
This isn't:
BATCHING: journal len = 1
BATCHING: vbo offset batch len = 1
BATCHING: material batch len = 1
BATCHING: modelview batch len = 1
BATCHING: journal len = 1
BATCHING: vbo offset batch len = 1
BATCHING: material batch len = 1
BATCHING: modelview batch len = 1
<repeat>
When this option is used Cogl will print a trace of all quads that get
logged into the journal, and a trace of quads as they get flushed.
If you are seeing a bug with the geometry being drawn by Cogl this may give
some clues by letting you sanity check the numbers being logged vs the
numbers being emitted.
We now put the color of materials into the vertex array used by the journal
instead of calling glColor() but the number of requests for the material
color were quite expensive so we have changed the material color to
internally be byte components instead of floats to avoid repeat conversions
and added _cogl_material_get_colorubv as a fast-path for the journal to
copy data into the vertex array.
The number of material layers enabled when logging a quad in the journal
determines the stride of the corresponding vertex data (since we need a set
of texture coordinates for each layer.) By padding data in the case where we
have only one layer we can avoid a change in stride if we are mixing single
and double layer primitives in a scene (e.g. relevent for a composite
manager that may use 2 layers for all shaped windows) Avoiding stride
changes means we can minimize calls to gl{Vertex,Color}Pointer when flushing
the journal.
Since we need to update the texcoord pointers when the actual number of
layers changes, this adds another batch_and_call() stage to deal with
glTexCoordPointer and enabling/disabling the client arrays.
Previously the journal was always flushed at the end of
_cogl_rectangles_with_multitexture_coords, (i.e. the end of any
cogl_rectangle* calls) but now we have broadened the potential for batching
geometry. In ideal circumstances we will only flush once per scene.
In summary the journal works like this:
When you use any of the cogl_rectangle* APIs then nothing is emitted to the
GPU at this point, we just log one or more quads into the journal. A
journal entry consists of the quad coordinates, an associated material
reference, and a modelview matrix. Ideally the journal only gets flushed
once at the end of a scene, but in fact there are things to consider that
may cause unwanted flushing, including:
- modifying materials mid-scene
This is because each quad in the journal has an associated material
reference (i.e. not copy), so if you try and modify a material that is
already referenced in the journal we force a flush first)
NOTE: For now this means you should avoid using cogl_set_source_color()
since that currently uses a single shared material. Later we
should change it to use a pool of materials that is recycled
when the journal is flushed.
- modifying any state that isn't currently logged, such as depth, fog and
backface culling enables.
The first thing that happens when flushing, is to upload all the vertex data
associated with the journal into a single VBO.
We then go through a process of splitting up the journal into batches that
have compatible state so they can be emitted to the GPU together. This is
currently broken up into 3 levels so we can stagger the state changes:
1) we break the journal up according to changes in the number of material layers
associated with logged quads. The number of layers in a material determines
the stride of the associated vertices, so we have to update our vertex
array offsets at this level. (i.e. calling gl{Vertex,Color},Pointer etc)
2) we further split batches up according to material compatability. (e.g.
materials with different textures) We flush material state at this level.
3) Finally we split batches up according to modelview changes. At this level
we update the modelview matrix and actually emit the actual draw command.
This commit is largely about putting the initial design in-place; this will be
followed by other changes that take advantage of the extended batching.
We were calculating our vertex stride and allocating our vertex array
differently depending on whether the user passed TRUE for use_color or not.
The problem was that we were always writting color data to the array
regardless of use_color.
There was also a bug with _cogl_texture_sliced_polygon in that it was
writing byte color components but we were expecting float components. We
now use byte components in _cogl_multitexture_unsliced_polygon too and pass
GL_UNSIGNED_BYTE to glColorPointer.
The texture filters are now a property of the material layer rather
than the texture object. Whenever a texture is painted with a material
it sets the filters on all of the GL textures in the Cogl texture. The
filter is cached so that it won't be changed unnecessarily.
The automatic mipmap generation has changed so that the mipmaps are
only generated when the texture is painted instead of every time the
data changes. Changing the texture sets a flag to mark that the
mipmaps are dirty. This works better if the FBO extension is available
because we can use glGenerateMipmap. If the extension is not available
it will temporarily enable automatic mipmap generation and reupload
the first pixel of each slice. This requires tracking the data for the
first pixel.
The COGL_TEXTURE_AUTO_MIPMAP flag has been replaced with
COGL_TEXTURE_NO_AUTO_MIPMAP so that it will default to
auto-mipmapping. The mipmap generation is now effectively free if you
are not using a mipmap filter mode so you would only want to disable
it if you had some special reason to generate your own mipmaps.
ClutterTexture no longer has to store its own copy of the filter
mode. Instead it stores it in the material and the property is
directly set and read from that. This fixes problems with the filters
getting out of sync when a cogl handle is set on the texture
directly. It also avoids the mess of having to rerealize the texture
if the filter quality changes to HIGH because Cogl will take of
generating the mipmaps if needed.
Instead of using GL_TRIANGLES and uploading the indices every time, it
now uses GL_QUADS instead on OpenGL. Under GLES it still uses indices
but it uses the new cogl_vertex_buffer_indices_get_for_quads function
to avoid uploading the vertices every time.
This requires the _cogl_vertex_buffer_indices_pointer_from_handle
function to be exposed privately to the rest of Cogl.
The static_indices array has been removed from the Cogl context.
At the moment Cogl doesn't do much batching of quads so most of the time we
are flushing a single quad at a time. This patch simplifies how we submit
those quads to OpenGL by using glDrawArrays with GL_TRIANGLE_FAN mode
instead of sending indexed vertices using GL_TRIANGLES mode.
Note: I hope to follow up soon with changes that improve our batching and
also move the indices into a VBO so they don't need to be re-validated every
time we call glDrawElements.
There were a number of functions intended to support creating of new
primitives using materials, but at this point they aren't used outside of
Cogl so until someone has a usecase and we can get feedback on this
API, it's being removed before we release Clutter 1.0.
COGL has a debug message system like Clutter's own. In parallel,
it also uses a coupld of #defines. Spread around there are also
calls to printf() instead to the more correct g_log* wrappers.
This commit tries to unify and clean up the macros and the
debug message handling inside COGL to be more consistent.
Adds missing notices, and ensures all the notices are consistent. The Cogl
blurb also now reads:
* Cogl
*
* An object oriented GL/GLES Abstraction/Utility Layer
This makes the #if 0'd debug code that was in _cogl_journal_flush_quad_batch
- which we have repeatedly found usefull for debugging various geometry
issues in Clutter apps - a runtime debug option.
The outline colors rotate in order from red to green to blue which can also
help confirm the order that your geometry really drawn.
The outlines are not affected by the current material state, so if you e.g.
have a blending bug where geometry mysteriously disappears this can confirm
if the underlying rectangles are actually being emitted but blending is
causing them to be invisible.
None of this code directly related to implementing CoglTextures, and the
code was needlessly duplicated between the GL and GLES backends. This moves
the cogl_rectangle* and cogl_polygon* code into common/cogl-primitives.c
makes which makes lot of sense since the two copies keep needlessly
diverging introducing or fixing bugs in one but not the other. For instance
I came accross one such bug regarding the enabling of texture units when
unifying the code.
This makes it consistent with cogl_rectangle_with_{multi,}texture_coords.
Notably the reason cogl_rectangle_with_{multi,}texture_coords wasn't changed
instead is that the former approach lets you describe back facing rectangles.
(though technically you could pass negative width/height values to achieve
this; it doesn't seem as neat.)
Conflicts:
clutter/cogl/gl/cogl-texture.c
clutter/cogl/gles/cogl-primitives.c
* cogl-material:
clutter-{clone-,}texture weren't updating their material opacity.
Updates GLES1 support for CoglMaterial
Normalizes gl vs gles code in preperation for synching material changes
Removes cogl_blend_func and cogl_alpha_func
Fully integrates CoglMaterial throughout the rest of Cogl
[cogl-material] Restore the GL_TEXTURE_ENV_MODE after material_rectangle
[cogl-material] Make the user_tex_coords parameter of _rectangle const
[test-cogl-material] Remove return value from material_rectangle_paint
Add cogl-material.h and cogl-matrix.h to libclutterinclude_HEADERS
[cogl-material] improvements for cogl_material_rectangle
[cogl-material] Adds a cogl_material_set_color function
[cogl-material] Some improvements for how we sync CoglMaterial state with OpenGL
[cogl-material] Converts clutter-texture/clutter-clone-texture to the material API
[doc] Hooks up cogl-material reference documentation
Updates previous GLES multi-texturing code to use CoglMaterial
Adds a CoglMaterial abstraction, which includes support for multi-texturing
[doc] Hooks up cogl-matrix reference documentation
Adds CoglMatrix utility code
[tests] Adds an interactive unit test for multi-texturing
[multi-texturing] This adds a new cogl_multi_texture API for GL,GLES1 + GLES2
This glues CoglMaterial in as the fundamental way that Cogl describes how to
fill in geometry.
It adds cogl_set_source (), which is used to set the material which will be
used by all subsequent drawing functions
It adds cogl_set_source_texture as a convenience for setting up a default
material with a single texture layer, and cogl_set_source_color is now also
a convenience for setting up a material with a solid fill.
"drawing functions" include, cogl_rectangle, cogl_texture_rectangle,
cogl_texture_multiple_rectangles, cogl_texture_polygon (though the
cogl_texture_* funcs have been renamed; see below for details),
cogl_path_fill/stroke and cogl_vertex_buffer_draw*.
cogl_texture_rectangle, cogl_texture_multiple_rectangles and
cogl_texture_polygon no longer take a texture handle; instead the current
source material is referenced. The functions have also been renamed to:
cogl_rectangle_with_texture_coords, cogl_rectangles_with_texture_coords
and cogl_polygon respectivly.
Most code that previously did:
cogl_texture_rectangle (tex_handle, x, y,...);
needs to be changed to now do:
cogl_set_source_texture (tex_handle);
cogl_rectangle_with_texture_coords (x, y,....);
In the less likely case where you were blending your source texture with a color
like:
cogl_set_source_color4ub (r,g,b,a); /* where r,g,b,a isn't just white */
cogl_texture_rectangle (tex_handle, x, y,...);
you will need your own material to do that:
mat = cogl_material_new ();
cogl_material_set_color4ub (r,g,b,a);
cogl_material_set_layer (mat, 0, tex_handle));
cogl_set_source_material (mat);
Code that uses the texture coordinates, 0, 0, 1, 1 don't need to use
cog_rectangle_with_texure_coords since these are the coordinates that
cogl_rectangle will use.
For cogl_texture_polygon; as well as dropping the texture handle, the
n_vertices and vertices arguments were transposed for consistency. So
code previously written as:
cogl_texture_polygon (tex_handle, 3, verts, TRUE);
need to be written as:
cogl_set_source_texture (tex_handle);
cogl_polygon (verts, 3, TRUE);
All of the unit tests have been updated to now use the material API and
test-cogl-material has been renamed to test-cogl-multitexture since any
textured quad is now technically a test of CoglMaterial but this test
specifically creates a material with multiple texture layers.
Note: The GLES backend has not been updated yet; that will be done in a
following commit.
To deal with all the corner cases that couldn't be scripted a number of patches
were written for the remaining 10% of the effort.
Note: again no API changes were made in Clutter, only in Cogl.
This is the result of running a number of sed and perl scripts over the code to
do 90% of the work in converting from 16.16 fixed to single precision floating
point.
Note: A pristine cogl-fixed.c has been maintained as a standalone utility API
so that applications may still take advantage of fixed point if they
desire for certain optimisations where lower precision may be acceptable.
Note: no API changes were made in Clutter, only in Cogl.
Overview of changes:
- Within clutter/* all usage of the COGL_FIXED_ macros have been changed to use
the CLUTTER_FIXED_ macros.
- Within cogl/* all usage of the COGL_FIXED_ macros have been completly stripped
and expanded into code that works with single precision floats instead.
- Uses of cogl_fixed_* have been replaced with single precision math.h
alternatives.
- Uses of COGL_ANGLE_* and cogl_angle_* have been replaced so we use a float for
angles and math.h replacements.
* clutter/cogl/cogl-path.h:
* clutter/cogl/common/cogl-primitives.c:
* clutter/cogl/common/cogl-primitives.h:
* clutter/cogl/gl/cogl-primitives.c:
* clutter/cogl/gles/cogl-primitives.c: Changed the semantics of
cogl_path_move_to. Previously this always started a new path but
now it instead starts a new disjoint sub path. The path isn't
cleared until you call either cogl_path_stroke, cogl_path_fill or
cogl_path_new. There are also cogl_path_stroke_preserve and
cogl_path_fill_preserve functions.
* clutter/cogl/gl/cogl-context.c:
* clutter/cogl/gl/cogl-context.h:
* clutter/cogl/gles/cogl-context.c:
* clutter/cogl/gles/cogl-context.h: Convert the path nodes array
to a GArray.
* clutter/cogl/gl/cogl-texture.c:
* clutter/cogl/gles/cogl-texture.c: Call cogl_clip_ensure
* clutter/cogl/common/cogl-clip-stack.c:
* clutter/cogl/common/cogl-clip-stack.h: Simplified the clip
stack code quite a bit to make it more maintainable. Previously
whenever you added a new clip it would go through a separate route
to immediately intersect with the current clip and when you
removed it again it would immediately rebuild the entire clip. Now
when you add or remove a clip it doesn't do anything immediately
but just sets a dirty flag instead.
* clutter/cogl/gl/cogl.c:
* clutter/cogl/gles/cogl.c: Taken away the code to intersect
stencil clips when there is exactly one stencil bit. It won't work
with path clips and I don't know of any platform that doesn't have
eight or zero stencil bits. It needs at least three bits to
intersect a path with an existing clip. cogl_features_init now
just decides you don't have a stencil buffer at all if you have
less than three bits.
* clutter/cogl/cogl.h.in: New functions and documentation.
* tests/interactive/test-clip.c: Replaced with a different test
that lets you add and remove clips. The three different mouse
buttons add clips in different shapes. This makes it easier to
test multiple levels of clipping.
* tests/interactive/test-cogl-primitives.c: Use
cogl_path_stroke_preserve when using the same path again.
* doc/reference/cogl/cogl-sections.txt: Document the new
functions.
Bug 1209 - Move fixed point API in COGL
* clutter/cogl/cogl-fixed.h:
* clutter/cogl/cogl.h.in:
* clutter/cogl/common/Makefile.am:
* clutter/cogl/common/cogl-fixed.c: Add fixed point API, modelled
after the ClutterFixed. The CoglFixed API supercedes the ClutterFixed
one and avoids the dependency of COGL on Clutter's own API.
* clutter/cogl/common/cogl-clip-stack.c:
* clutter/cogl/common/cogl-primitives.c:
* clutter/cogl/common/cogl-primitives.h: Update internal usage of
ClutterFixed to CoglFixed.
* clutter/cogl/gl/Makefile.am:
* clutter/cogl/gl/cogl-primitives.c:
* clutter/cogl/gl/cogl-texture.c:
* clutter/cogl/gl/cogl.c: Ditto, in the GL implementation of the
COGL API.
* clutter/cogl/gles/Makefile.am:
* clutter/cogl/gles/cogl-fbo.c:
* clutter/cogl/gles/cogl-gles2-wrapper.c:
* clutter/cogl/gles/cogl-primitives.c:
* clutter/cogl/gles/cogl-texture.c:
* clutter/cogl/gles/cogl.c: Ditto, in the GLES implementation of
the COGL API.
* clutter/pango/pangoclutter-glyph-cache.c:
* clutter/pango/pangoclutter-glyph-cache.h: Ditto, in the Pango
renderer glyphs cache.
* clutter/clutter-fixed.c:
* clutter/clutter-fixed.h: ClutterFixed and related API becomes
a simple transition API for bindings and public Clutter API.
* clutter/clutter-actor.c:
* clutter/clutter-alpha.c:
* clutter/clutter-backend.c:
* clutter/clutter-behaviour-depth.c:
* clutter/clutter-behaviour-ellipse.c:
* clutter/clutter-behaviour-path.c:
* clutter/clutter-behaviour-rotate.c:
* clutter/clutter-behaviour-scale.c:
* clutter/clutter-clone-texture.c:
* clutter/clutter-color.c:
* clutter/clutter-entry.c:
* clutter/clutter-stage.c:
* clutter/clutter-texture.c:
* clutter/clutter-timeline.c:
* clutter/clutter-units.h: Move from the internal usage of
ClutterFixed to CoglFixed.
* doc/reference/clutter/clutter-sections.txt:
* doc/reference/cogl/cogl-docs.sgml:
* doc/reference/cogl/cogl-sections.txt: Update the documentation.
* tests/test-cogl-tex-tile.c:
* tests/test-project.c: Fix tests after the API change
* README: Add release notes.
* clutter/cogl/gles/cogl-primitives.c:
* clutter/cogl/common/cogl-primitives.c: moved declaration of
gegl_rectangle and gegl_rectanglex here to satisfy linking
requirements when building the fruity backend.
* clutter/cogl/common/cogl-primitives.c: api review touch ups.
* clutter/cogl/gl/cogl-primitives.c: (cogl_path_fill),
(cogl_path_stroke): indentation.
* clutter/cogl/gles/cogl-primitives.c:
(_cogl_path_fill_nodes): free allocated resources.
* tests/test-cogl-primitives.c: updated to new API, added rotation to
test to show that cogl renders paths correct under perspective
distortion.
