isa/mutter
1
0
Fork 0
mirror of https://github.com/brl/mutter.git synced 2024-12-27 05:12:15 +00:00
Code Issues Packages Projects Releases Wiki Activity
01e744455d
mutter/cogl/cogl-pipeline-state.c

2003 lines
64 KiB
C
Raw Normal View History

pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
/*
* Cogl
*
* An object oriented GL/GLES Abstraction/Utility Layer
*
* Copyright (C) 2008,2009,2010 Intel Corporation.
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library. If not, see
* <http://www.gnu.org/licenses/>.
*
*
*
* Authors:
* Robert Bragg <robert@linux.intel.com>
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "cogl-context-private.h"
#include "cogl-color-private.h"
#include "cogl-blend-string.h"
#include "cogl-util.h"
#include "cogl-depth-state-private.h"
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
#include "cogl-pipeline-state-private.h"
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
#include "cogl-snippet-private.h"
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
#include "string.h"
#ifndef GL_FUNC_ADD
#define GL_FUNC_ADD 0x8006
#endif
CoglPipeline *
_cogl_pipeline_get_user_program (CoglPipeline *pipeline)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), NULL);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_USER_SHADER);
return authority->big_state->user_program;
}
gboolean
_cogl_pipeline_color_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
return cogl_color_equal (&authority0->color, &authority1->color);
}
gboolean
_cogl_pipeline_lighting_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
CoglPipelineLightingState *state0 = &authority0->big_state->lighting_state;
CoglPipelineLightingState *state1 = &authority1->big_state->lighting_state;
if (memcmp (state0->ambient, state1->ambient, sizeof (float) * 4) != 0)
return FALSE;
if (memcmp (state0->diffuse, state1->diffuse, sizeof (float) * 4) != 0)
return FALSE;
if (memcmp (state0->specular, state1->specular, sizeof (float) * 4) != 0)
return FALSE;
if (memcmp (state0->emission, state1->emission, sizeof (float) * 4) != 0)
return FALSE;
if (state0->shininess != state1->shininess)
return FALSE;
return TRUE;
}
gboolean
_cogl_pipeline_alpha_func_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
CoglPipelineAlphaFuncState *alpha_state0 =
&authority0->big_state->alpha_state;
CoglPipelineAlphaFuncState *alpha_state1 =
&authority1->big_state->alpha_state;
return alpha_state0->alpha_func == alpha_state1->alpha_func;
}
gboolean
_cogl_pipeline_alpha_func_reference_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
CoglPipelineAlphaFuncState *alpha_state0 =
&authority0->big_state->alpha_state;
CoglPipelineAlphaFuncState *alpha_state1 =
&authority1->big_state->alpha_state;
return (alpha_state0->alpha_func_reference ==
alpha_state1->alpha_func_reference);
}
gboolean
_cogl_pipeline_blend_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
CoglPipelineBlendState *blend_state0 = &authority0->big_state->blend_state;
CoglPipelineBlendState *blend_state1 = &authority1->big_state->blend_state;
_COGL_GET_CONTEXT (ctx, FALSE);
#if defined(HAVE_COGL_GLES2) || defined(HAVE_COGL_GL)
if (ctx->driver != COGL_DRIVER_GLES1)
{
if (blend_state0->blend_equation_rgb != blend_state1->blend_equation_rgb)
return FALSE;
if (blend_state0->blend_equation_alpha !=
blend_state1->blend_equation_alpha)
return FALSE;
if (blend_state0->blend_src_factor_alpha !=
blend_state1->blend_src_factor_alpha)
return FALSE;
if (blend_state0->blend_dst_factor_alpha !=
blend_state1->blend_dst_factor_alpha)
return FALSE;
}
#endif
if (blend_state0->blend_src_factor_rgb !=
blend_state1->blend_src_factor_rgb)
return FALSE;
if (blend_state0->blend_dst_factor_rgb !=
blend_state1->blend_dst_factor_rgb)
return FALSE;
#if defined(HAVE_COGL_GLES2) || defined(HAVE_COGL_GL)
if (ctx->driver != COGL_DRIVER_GLES1 &&
(blend_state0->blend_src_factor_rgb == GL_ONE_MINUS_CONSTANT_COLOR ||
blend_state0->blend_src_factor_rgb == GL_CONSTANT_COLOR ||
blend_state0->blend_dst_factor_rgb == GL_ONE_MINUS_CONSTANT_COLOR ||
blend_state0->blend_dst_factor_rgb == GL_CONSTANT_COLOR))
{
if (!cogl_color_equal (&blend_state0->blend_constant,
&blend_state1->blend_constant))
return FALSE;
}
#endif
return TRUE;
}
gboolean
_cogl_pipeline_depth_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
if (authority0->big_state->depth_state.test_enabled == FALSE &&
authority1->big_state->depth_state.test_enabled == FALSE)
return TRUE;
else
{
CoglDepthState *s0 = &authority0->big_state->depth_state;
CoglDepthState *s1 = &authority1->big_state->depth_state;
return s0->test_enabled == s1->test_enabled &&
s0->test_function == s1->test_function &&
s0->write_enabled == s1->write_enabled &&
s0->range_near == s1->range_near &&
s0->range_far == s1->range_far;
}
}
gboolean
_cogl_pipeline_fog_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
CoglPipelineFogState *fog_state0 = &authority0->big_state->fog_state;
CoglPipelineFogState *fog_state1 = &authority1->big_state->fog_state;
if (fog_state0->enabled == fog_state1->enabled &&
cogl_color_equal (&fog_state0->color, &fog_state1->color) &&
fog_state0->mode == fog_state1->mode &&
fog_state0->density == fog_state1->density &&
fog_state0->z_near == fog_state1->z_near &&
fog_state0->z_far == fog_state1->z_far)
return TRUE;
else
return FALSE;
}
gboolean
_cogl_pipeline_point_size_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
return authority0->big_state->point_size == authority1->big_state->point_size;
}
gboolean
_cogl_pipeline_logic_ops_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
CoglPipelineLogicOpsState *logic_ops_state0 = &authority0->big_state->logic_ops_state;
CoglPipelineLogicOpsState *logic_ops_state1 = &authority1->big_state->logic_ops_state;
return logic_ops_state0->color_mask == logic_ops_state1->color_mask;
}
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
gboolean
_cogl_pipeline_cull_face_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
CoglPipelineCullFaceState *cull_face_state0
= &authority0->big_state->cull_face_state;
CoglPipelineCullFaceState *cull_face_state1
= &authority1->big_state->cull_face_state;
/* The cull face state is considered equal if two pipelines are both
set to no culling. If the front winding property is ever used for
anything else or the comparison is used not just for drawing then
this would have to change */
if (cull_face_state0->mode == COGL_PIPELINE_CULL_FACE_MODE_NONE)
return cull_face_state1->mode == COGL_PIPELINE_CULL_FACE_MODE_NONE;
return (cull_face_state0->mode == cull_face_state1->mode &&
cull_face_state0->front_winding == cull_face_state1->front_winding);
}
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
gboolean
_cogl_pipeline_user_shader_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
return (authority0->big_state->user_program ==
authority1->big_state->user_program);
}
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
typedef struct
{
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
const CoglBoxedValue **dst_values;
const CoglBoxedValue *src_values;
int override_count;
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
} GetUniformsClosure;
static gboolean
get_uniforms_cb (int uniform_num, void *user_data)
{
GetUniformsClosure *data = user_data;
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
if (data->dst_values[uniform_num] == NULL)
data->dst_values[uniform_num] = data->src_values + data->override_count;
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
data->override_count++;
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
return TRUE;
}
static void
_cogl_pipeline_get_all_uniform_values (CoglPipeline *pipeline,
const CoglBoxedValue **values)
{
GetUniformsClosure data;
_COGL_GET_CONTEXT (ctx, NO_RETVAL);
memset (values, 0,
sizeof (const CoglBoxedValue *) * ctx->n_uniform_names);
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
data.dst_values = values;
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
do
{
if ((pipeline->differences & COGL_PIPELINE_STATE_UNIFORMS))
pipeline-state: Fix crash in the get_all_uniform_values function The get_all_uniform_values function tries to walk the parent hierarchy of pipelines to find pipelines overriding the uniforms state and then grabs the values from the override. However it was accessing data inside the ‘big state’ even if the pipeline didn't override the uniforms state so it would crash if it encountered a parent pipeline with no big state. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2012-02-13 18:48:17 +00:00
{
const CoglPipelineUniformsState *uniforms_state =
&pipeline->big_state->uniforms_state;
data.override_count = 0;
data.src_values = uniforms_state->override_values;
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
pipeline-state: Fix crash in the get_all_uniform_values function The get_all_uniform_values function tries to walk the parent hierarchy of pipelines to find pipelines overriding the uniforms state and then grabs the values from the override. However it was accessing data inside the ‘big state’ even if the pipeline didn't override the uniforms state so it would crash if it encountered a parent pipeline with no big state. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2012-02-13 18:48:17 +00:00
_cogl_bitmask_foreach (&uniforms_state->override_mask,
get_uniforms_cb,
&data);
}
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
pipeline = _cogl_pipeline_get_parent (pipeline);
}
while (pipeline);
}
gboolean
_cogl_pipeline_uniforms_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
unsigned long *differences;
const CoglBoxedValue **values0, **values1;
int n_longs;
int i;
_COGL_GET_CONTEXT (ctx, FALSE);
if (authority0 == authority1)
return TRUE;
values0 = g_alloca (sizeof (const CoglBoxedValue *) * ctx->n_uniform_names);
values1 = g_alloca (sizeof (const CoglBoxedValue *) * ctx->n_uniform_names);
n_longs = COGL_FLAGS_N_LONGS_FOR_SIZE (ctx->n_uniform_names);
differences = g_alloca (n_longs * sizeof (unsigned long));
memset (differences, 0, sizeof (unsigned long) * n_longs);
_cogl_pipeline_compare_uniform_differences (differences,
authority0,
authority1);
_cogl_pipeline_get_all_uniform_values (authority0, values0);
_cogl_pipeline_get_all_uniform_values (authority1, values1);
COGL_FLAGS_FOREACH_START (differences, n_longs, i)
{
const CoglBoxedValue *value0 = values0[i];
const CoglBoxedValue *value1 = values1[i];
Fix comparing the uniforms state When comparing uniform values, it was not correctly handling the case where pipeline0 has the value set but pipeline1 does not (only the other way around) so it would crash. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:35:23 +00:00
if (value0 == NULL)
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
{
if (value1 != NULL && value1->type != COGL_BOXED_NONE)
return FALSE;
}
Fix comparing the uniforms state When comparing uniform values, it was not correctly handling the case where pipeline0 has the value set but pipeline1 does not (only the other way around) so it would crash. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:35:23 +00:00
else if (value1 == NULL)
{
if (value0 != NULL && value0->type != COGL_BOXED_NONE)
return FALSE;
}
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
else if (!_cogl_boxed_value_equal (value0, value1))
return FALSE;
}
COGL_FLAGS_FOREACH_END;
return TRUE;
}
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
gboolean
_cogl_pipeline_vertex_snippets_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
return _cogl_pipeline_snippet_list_equal (&authority0->big_state->
vertex_snippets,
&authority1->big_state->
vertex_snippets);
}
gboolean
_cogl_pipeline_fragment_snippets_state_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
return _cogl_pipeline_snippet_list_equal (&authority0->big_state->
fragment_snippets,
&authority1->big_state->
fragment_snippets);
}
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
void
cogl_pipeline_get_color (CoglPipeline *pipeline,
CoglColor *color)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_COLOR);
*color = authority->color;
}
/* This is used heavily by the cogl journal when logging quads */
void
_cogl_pipeline_get_colorubv (CoglPipeline *pipeline,
guint8 *color)
{
CoglPipeline *authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_COLOR);
_cogl_color_get_rgba_4ubv (&authority->color, color);
}
void
cogl_pipeline_set_color (CoglPipeline *pipeline,
const CoglColor *color)
{
CoglPipelineState state = COGL_PIPELINE_STATE_COLOR;
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
if (cogl_color_equal (color, &authority->color))
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, color, FALSE);
pipeline->color = *color;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_color_equal);
_cogl_pipeline_update_blend_enable (pipeline, state);
}
void
cogl_pipeline_set_color4ub (CoglPipeline *pipeline,
guint8 red,
guint8 green,
guint8 blue,
guint8 alpha)
{
CoglColor color;
cogl_color_init_from_4ub (&color, red, green, blue, alpha);
cogl_pipeline_set_color (pipeline, &color);
}
void
cogl_pipeline_set_color4f (CoglPipeline *pipeline,
float red,
float green,
float blue,
float alpha)
{
CoglColor color;
cogl_color_init_from_4f (&color, red, green, blue, alpha);
cogl_pipeline_set_color (pipeline, &color);
}
CoglPipelineBlendEnable
_cogl_pipeline_get_blend_enabled (CoglPipeline *pipeline)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), FALSE);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_BLEND_ENABLE);
return authority->blend_enable;
}
Add -Wmissing-declarations to maintainer flags and fix problems This option to GCC makes it give a warning whenever a global function is defined without a declaration. This should catch cases were we've defined a function but forgot to put it in a header. In that case it is either only used within one file so we should make it static or we should declare it in a header. The following changes where made to fix problems: • Some functions were made static • cogl-path.h (the one containing the 1.0 API) was split into two files, one defining the functions and one defining the enums so that cogl-path.c can include the enum and function declarations from the 2.0 API as well as the function declarations from the 1.0 API. • cogl2-clip-state has been removed. This only had one experimental function called cogl_clip_push_from_path but as this is unstable we might as well remove it favour of the equivalent cogl_framebuffer_* API. • The GLX, SDL and WGL winsys's now have a private header to define their get_vtable function instead of directly declaring in the C file where it is called. • All places that were calling COGL_OBJECT_DEFINE need to have the cogl_is_whatever function declared so these have been added either as a public function or in a private header. • Some files that were not including the header containing their function declarations have been fixed to do so. • Any unused error quark functions have been removed. If we later want them we should add them back one by one and add a declaration for them in a header. • _cogl_is_framebuffer has been renamed to cogl_is_framebuffer and made a public function with a declaration in cogl-framebuffer.h • Similarly for CoglOnscreen. • cogl_vdraw_indexed_attributes is called cogl_framebuffer_vdraw_indexed_attributes in the header. The definition has been changed to match the header. • cogl_index_buffer_allocate has been removed. This had no declaration and I'm not sure what it's supposed to do. • CoglJournal has been changed to use the internal CoglObject macro so that it won't define an exported cogl_is_journal symbol. • The _cogl_blah_pointer_from_handle functions have been removed. CoglHandle isn't used much anymore anyway and in the few places where it is used I think it's safe to just use the implicit cast from void* to the right type. • The test-utils.h header for the conformance tests explicitly disables the -Wmissing-declaration option using a pragma because all of the tests declare their main function without a header. Any mistakes relating to missing declarations aren't really important for the tests. • cogl_quaternion_init_from_quaternion and init_from_matrix have been given declarations in cogl-quaternion.h Reviewed-by: Robert Bragg <robert@linux.intel.com>
2012-03-06 18:21:28 +00:00
static gboolean
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
_cogl_pipeline_blend_enable_equal (CoglPipeline *authority0,
CoglPipeline *authority1)
{
return authority0->blend_enable == authority1->blend_enable ? TRUE : FALSE;
}
void
_cogl_pipeline_set_blend_enabled (CoglPipeline *pipeline,
CoglPipelineBlendEnable enable)
{
CoglPipelineState state = COGL_PIPELINE_STATE_BLEND_ENABLE;
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
_COGL_RETURN_IF_FAIL (enable > 1 &&
"don't pass TRUE or FALSE to _set_blend_enabled!");
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
if (authority->blend_enable == enable)
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
pipeline->blend_enable = enable;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_blend_enable_equal);
_cogl_pipeline_update_blend_enable (pipeline, state);
}
void
cogl_pipeline_get_ambient (CoglPipeline *pipeline,
CoglColor *ambient)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_LIGHTING);
cogl_color_init_from_4fv (ambient,
authority->big_state->lighting_state.ambient);
}
void
cogl_pipeline_set_ambient (CoglPipeline *pipeline,
const CoglColor *ambient)
{
CoglPipelineState state = COGL_PIPELINE_STATE_LIGHTING;
CoglPipeline *authority;
CoglPipelineLightingState *lighting_state;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
lighting_state = &authority->big_state->lighting_state;
if (cogl_color_equal (ambient, &lighting_state->ambient))
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
lighting_state = &pipeline->big_state->lighting_state;
lighting_state->ambient[0] = cogl_color_get_red_float (ambient);
lighting_state->ambient[1] = cogl_color_get_green_float (ambient);
lighting_state->ambient[2] = cogl_color_get_blue_float (ambient);
lighting_state->ambient[3] = cogl_color_get_alpha_float (ambient);
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_lighting_state_equal);
_cogl_pipeline_update_blend_enable (pipeline, state);
}
void
cogl_pipeline_get_diffuse (CoglPipeline *pipeline,
CoglColor *diffuse)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_LIGHTING);
cogl_color_init_from_4fv (diffuse,
authority->big_state->lighting_state.diffuse);
}
void
cogl_pipeline_set_diffuse (CoglPipeline *pipeline,
const CoglColor *diffuse)
{
CoglPipelineState state = COGL_PIPELINE_STATE_LIGHTING;
CoglPipeline *authority;
CoglPipelineLightingState *lighting_state;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
lighting_state = &authority->big_state->lighting_state;
if (cogl_color_equal (diffuse, &lighting_state->diffuse))
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
lighting_state = &pipeline->big_state->lighting_state;
lighting_state->diffuse[0] = cogl_color_get_red_float (diffuse);
lighting_state->diffuse[1] = cogl_color_get_green_float (diffuse);
lighting_state->diffuse[2] = cogl_color_get_blue_float (diffuse);
lighting_state->diffuse[3] = cogl_color_get_alpha_float (diffuse);
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_lighting_state_equal);
_cogl_pipeline_update_blend_enable (pipeline, state);
}
void
cogl_pipeline_set_ambient_and_diffuse (CoglPipeline *pipeline,
const CoglColor *color)
{
cogl_pipeline_set_ambient (pipeline, color);
cogl_pipeline_set_diffuse (pipeline, color);
}
void
cogl_pipeline_get_specular (CoglPipeline *pipeline,
CoglColor *specular)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_LIGHTING);
cogl_color_init_from_4fv (specular,
authority->big_state->lighting_state.specular);
}
void
cogl_pipeline_set_specular (CoglPipeline *pipeline, const CoglColor *specular)
{
CoglPipeline *authority;
CoglPipelineState state = COGL_PIPELINE_STATE_LIGHTING;
CoglPipelineLightingState *lighting_state;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
lighting_state = &authority->big_state->lighting_state;
if (cogl_color_equal (specular, &lighting_state->specular))
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
lighting_state = &pipeline->big_state->lighting_state;
lighting_state->specular[0] = cogl_color_get_red_float (specular);
lighting_state->specular[1] = cogl_color_get_green_float (specular);
lighting_state->specular[2] = cogl_color_get_blue_float (specular);
lighting_state->specular[3] = cogl_color_get_alpha_float (specular);
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_lighting_state_equal);
_cogl_pipeline_update_blend_enable (pipeline, state);
}
float
cogl_pipeline_get_shininess (CoglPipeline *pipeline)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), 0);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_LIGHTING);
return authority->big_state->lighting_state.shininess;
}
void
cogl_pipeline_set_shininess (CoglPipeline *pipeline,
float shininess)
{
CoglPipeline *authority;
CoglPipelineState state = COGL_PIPELINE_STATE_LIGHTING;
CoglPipelineLightingState *lighting_state;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
if (shininess < 0.0)
{
g_warning ("Out of range shininess %f supplied for pipeline\n",
shininess);
return;
}
authority = _cogl_pipeline_get_authority (pipeline, state);
lighting_state = &authority->big_state->lighting_state;
if (lighting_state->shininess == shininess)
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
lighting_state = &pipeline->big_state->lighting_state;
lighting_state->shininess = shininess;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_lighting_state_equal);
}
void
cogl_pipeline_get_emission (CoglPipeline *pipeline,
CoglColor *emission)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_LIGHTING);
cogl_color_init_from_4fv (emission,
authority->big_state->lighting_state.emission);
}
void
cogl_pipeline_set_emission (CoglPipeline *pipeline, const CoglColor *emission)
{
CoglPipeline *authority;
CoglPipelineState state = COGL_PIPELINE_STATE_LIGHTING;
CoglPipelineLightingState *lighting_state;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
lighting_state = &authority->big_state->lighting_state;
if (cogl_color_equal (emission, &lighting_state->emission))
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
lighting_state = &pipeline->big_state->lighting_state;
lighting_state->emission[0] = cogl_color_get_red_float (emission);
lighting_state->emission[1] = cogl_color_get_green_float (emission);
lighting_state->emission[2] = cogl_color_get_blue_float (emission);
lighting_state->emission[3] = cogl_color_get_alpha_float (emission);
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_lighting_state_equal);
_cogl_pipeline_update_blend_enable (pipeline, state);
}
static void
_cogl_pipeline_set_alpha_test_function (CoglPipeline *pipeline,
CoglPipelineAlphaFunc alpha_func)
{
CoglPipelineState state = COGL_PIPELINE_STATE_ALPHA_FUNC;
CoglPipeline *authority;
CoglPipelineAlphaFuncState *alpha_state;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
alpha_state = &authority->big_state->alpha_state;
if (alpha_state->alpha_func == alpha_func)
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
alpha_state = &pipeline->big_state->alpha_state;
alpha_state->alpha_func = alpha_func;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_alpha_func_state_equal);
}
static void
_cogl_pipeline_set_alpha_test_function_reference (CoglPipeline *pipeline,
float alpha_reference)
{
CoglPipelineState state = COGL_PIPELINE_STATE_ALPHA_FUNC_REFERENCE;
CoglPipeline *authority;
CoglPipelineAlphaFuncState *alpha_state;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
alpha_state = &authority->big_state->alpha_state;
if (alpha_state->alpha_func_reference == alpha_reference)
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
alpha_state = &pipeline->big_state->alpha_state;
alpha_state->alpha_func_reference = alpha_reference;
_cogl_pipeline_update_authority
(pipeline, authority, state,
_cogl_pipeline_alpha_func_reference_state_equal);
}
void
cogl_pipeline_set_alpha_test_function (CoglPipeline *pipeline,
CoglPipelineAlphaFunc alpha_func,
float alpha_reference)
{
_cogl_pipeline_set_alpha_test_function (pipeline, alpha_func);
_cogl_pipeline_set_alpha_test_function_reference (pipeline, alpha_reference);
}
CoglPipelineAlphaFunc
cogl_pipeline_get_alpha_test_function (CoglPipeline *pipeline)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), 0);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_ALPHA_FUNC);
return authority->big_state->alpha_state.alpha_func;
}
float
cogl_pipeline_get_alpha_test_reference (CoglPipeline *pipeline)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), 0.0f);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline,
COGL_PIPELINE_STATE_ALPHA_FUNC_REFERENCE);
return authority->big_state->alpha_state.alpha_func_reference;
}
Add -Wmissing-declarations to maintainer flags and fix problems This option to GCC makes it give a warning whenever a global function is defined without a declaration. This should catch cases were we've defined a function but forgot to put it in a header. In that case it is either only used within one file so we should make it static or we should declare it in a header. The following changes where made to fix problems: • Some functions were made static • cogl-path.h (the one containing the 1.0 API) was split into two files, one defining the functions and one defining the enums so that cogl-path.c can include the enum and function declarations from the 2.0 API as well as the function declarations from the 1.0 API. • cogl2-clip-state has been removed. This only had one experimental function called cogl_clip_push_from_path but as this is unstable we might as well remove it favour of the equivalent cogl_framebuffer_* API. • The GLX, SDL and WGL winsys's now have a private header to define their get_vtable function instead of directly declaring in the C file where it is called. • All places that were calling COGL_OBJECT_DEFINE need to have the cogl_is_whatever function declared so these have been added either as a public function or in a private header. • Some files that were not including the header containing their function declarations have been fixed to do so. • Any unused error quark functions have been removed. If we later want them we should add them back one by one and add a declaration for them in a header. • _cogl_is_framebuffer has been renamed to cogl_is_framebuffer and made a public function with a declaration in cogl-framebuffer.h • Similarly for CoglOnscreen. • cogl_vdraw_indexed_attributes is called cogl_framebuffer_vdraw_indexed_attributes in the header. The definition has been changed to match the header. • cogl_index_buffer_allocate has been removed. This had no declaration and I'm not sure what it's supposed to do. • CoglJournal has been changed to use the internal CoglObject macro so that it won't define an exported cogl_is_journal symbol. • The _cogl_blah_pointer_from_handle functions have been removed. CoglHandle isn't used much anymore anyway and in the few places where it is used I think it's safe to just use the implicit cast from void* to the right type. • The test-utils.h header for the conformance tests explicitly disables the -Wmissing-declaration option using a pragma because all of the tests declare their main function without a header. Any mistakes relating to missing declarations aren't really important for the tests. • cogl_quaternion_init_from_quaternion and init_from_matrix have been given declarations in cogl-quaternion.h Reviewed-by: Robert Bragg <robert@linux.intel.com>
2012-03-06 18:21:28 +00:00
static GLenum
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
arg_to_gl_blend_factor (CoglBlendStringArgument *arg)
{
if (arg->source.is_zero)
return GL_ZERO;
if (arg->factor.is_one)
return GL_ONE;
else if (arg->factor.is_src_alpha_saturate)
return GL_SRC_ALPHA_SATURATE;
else if (arg->factor.source.info->type ==
COGL_BLEND_STRING_COLOR_SOURCE_SRC_COLOR)
{
if (arg->factor.source.mask != COGL_BLEND_STRING_CHANNEL_MASK_ALPHA)
{
if (arg->factor.source.one_minus)
return GL_ONE_MINUS_SRC_COLOR;
else
return GL_SRC_COLOR;
}
else
{
if (arg->factor.source.one_minus)
return GL_ONE_MINUS_SRC_ALPHA;
else
return GL_SRC_ALPHA;
}
}
else if (arg->factor.source.info->type ==
COGL_BLEND_STRING_COLOR_SOURCE_DST_COLOR)
{
if (arg->factor.source.mask != COGL_BLEND_STRING_CHANNEL_MASK_ALPHA)
{
if (arg->factor.source.one_minus)
return GL_ONE_MINUS_DST_COLOR;
else
return GL_DST_COLOR;
}
else
{
if (arg->factor.source.one_minus)
return GL_ONE_MINUS_DST_ALPHA;
else
return GL_DST_ALPHA;
}
}
#if defined(HAVE_COGL_GLES2) || defined(HAVE_COGL_GL)
else if (arg->factor.source.info->type ==
COGL_BLEND_STRING_COLOR_SOURCE_CONSTANT)
{
if (arg->factor.source.mask != COGL_BLEND_STRING_CHANNEL_MASK_ALPHA)
{
if (arg->factor.source.one_minus)
return GL_ONE_MINUS_CONSTANT_COLOR;
else
return GL_CONSTANT_COLOR;
}
else
{
if (arg->factor.source.one_minus)
return GL_ONE_MINUS_CONSTANT_ALPHA;
else
return GL_CONSTANT_ALPHA;
}
}
#endif
g_warning ("Unable to determine valid blend factor from blend string\n");
return GL_ONE;
}
Add -Wmissing-declarations to maintainer flags and fix problems This option to GCC makes it give a warning whenever a global function is defined without a declaration. This should catch cases were we've defined a function but forgot to put it in a header. In that case it is either only used within one file so we should make it static or we should declare it in a header. The following changes where made to fix problems: • Some functions were made static • cogl-path.h (the one containing the 1.0 API) was split into two files, one defining the functions and one defining the enums so that cogl-path.c can include the enum and function declarations from the 2.0 API as well as the function declarations from the 1.0 API. • cogl2-clip-state has been removed. This only had one experimental function called cogl_clip_push_from_path but as this is unstable we might as well remove it favour of the equivalent cogl_framebuffer_* API. • The GLX, SDL and WGL winsys's now have a private header to define their get_vtable function instead of directly declaring in the C file where it is called. • All places that were calling COGL_OBJECT_DEFINE need to have the cogl_is_whatever function declared so these have been added either as a public function or in a private header. • Some files that were not including the header containing their function declarations have been fixed to do so. • Any unused error quark functions have been removed. If we later want them we should add them back one by one and add a declaration for them in a header. • _cogl_is_framebuffer has been renamed to cogl_is_framebuffer and made a public function with a declaration in cogl-framebuffer.h • Similarly for CoglOnscreen. • cogl_vdraw_indexed_attributes is called cogl_framebuffer_vdraw_indexed_attributes in the header. The definition has been changed to match the header. • cogl_index_buffer_allocate has been removed. This had no declaration and I'm not sure what it's supposed to do. • CoglJournal has been changed to use the internal CoglObject macro so that it won't define an exported cogl_is_journal symbol. • The _cogl_blah_pointer_from_handle functions have been removed. CoglHandle isn't used much anymore anyway and in the few places where it is used I think it's safe to just use the implicit cast from void* to the right type. • The test-utils.h header for the conformance tests explicitly disables the -Wmissing-declaration option using a pragma because all of the tests declare their main function without a header. Any mistakes relating to missing declarations aren't really important for the tests. • cogl_quaternion_init_from_quaternion and init_from_matrix have been given declarations in cogl-quaternion.h Reviewed-by: Robert Bragg <robert@linux.intel.com>
2012-03-06 18:21:28 +00:00
static void
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
setup_blend_state (CoglBlendStringStatement *statement,
GLenum *blend_equation,
GLint *blend_src_factor,
GLint *blend_dst_factor)
{
switch (statement->function->type)
{
case COGL_BLEND_STRING_FUNCTION_ADD:
*blend_equation = GL_FUNC_ADD;
break;
/* TODO - add more */
default:
g_warning ("Unsupported blend function given");
*blend_equation = GL_FUNC_ADD;
}
*blend_src_factor = arg_to_gl_blend_factor (&statement->args[0]);
*blend_dst_factor = arg_to_gl_blend_factor (&statement->args[1]);
}
gboolean
cogl_pipeline_set_blend (CoglPipeline *pipeline,
const char *blend_description,
GError **error)
{
CoglPipelineState state = COGL_PIPELINE_STATE_BLEND;
CoglPipeline *authority;
CoglBlendStringStatement statements[2];
CoglBlendStringStatement *rgb;
CoglBlendStringStatement *a;
GError *internal_error = NULL;
int count;
CoglPipelineBlendState *blend_state;
_COGL_GET_CONTEXT (ctx, FALSE);
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), FALSE);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
count =
_cogl_blend_string_compile (blend_description,
COGL_BLEND_STRING_CONTEXT_BLENDING,
statements,
&internal_error);
if (!count)
{
if (error)
g_propagate_error (error, internal_error);
else
{
g_warning ("Cannot compile blend description: %s\n",
internal_error->message);
g_error_free (internal_error);
}
return FALSE;
}
if (count == 1)
rgb = a = statements;
else
{
rgb = &statements[0];
a = &statements[1];
}
authority =
_cogl_pipeline_get_authority (pipeline, state);
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
blend_state = &pipeline->big_state->blend_state;
#if defined (HAVE_COGL_GL) || defined (HAVE_COGL_GLES2)
if (ctx->driver != COGL_DRIVER_GLES1)
{
setup_blend_state (rgb,
&blend_state->blend_equation_rgb,
&blend_state->blend_src_factor_rgb,
&blend_state->blend_dst_factor_rgb);
setup_blend_state (a,
&blend_state->blend_equation_alpha,
&blend_state->blend_src_factor_alpha,
&blend_state->blend_dst_factor_alpha);
}
else
#endif
{
setup_blend_state (rgb,
NULL,
&blend_state->blend_src_factor_rgb,
&blend_state->blend_dst_factor_rgb);
}
/* If we are the current authority see if we can revert to one of our
* ancestors being the authority */
if (pipeline == authority &&
_cogl_pipeline_get_parent (authority) != NULL)
{
CoglPipeline *parent = _cogl_pipeline_get_parent (authority);
CoglPipeline *old_authority =
_cogl_pipeline_get_authority (parent, state);
if (_cogl_pipeline_blend_state_equal (authority, old_authority))
pipeline->differences &= ~state;
}
/* If we weren't previously the authority on this state then we need
* to extended our differences mask and so it's possible that some
* of our ancestry will now become redundant, so we aim to reparent
* ourselves if that's true... */
if (pipeline != authority)
{
pipeline->differences |= state;
_cogl_pipeline_prune_redundant_ancestry (pipeline);
}
_cogl_pipeline_update_blend_enable (pipeline, state);
return TRUE;
}
void
cogl_pipeline_set_blend_constant (CoglPipeline *pipeline,
const CoglColor *constant_color)
{
_COGL_GET_CONTEXT (ctx, NO_RETVAL);
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
if (ctx->driver == COGL_DRIVER_GLES1)
return;
#if defined(HAVE_COGL_GLES2) || defined(HAVE_COGL_GL)
{
CoglPipelineState state = COGL_PIPELINE_STATE_BLEND;
CoglPipeline *authority;
CoglPipelineBlendState *blend_state;
authority = _cogl_pipeline_get_authority (pipeline, state);
blend_state = &authority->big_state->blend_state;
if (cogl_color_equal (constant_color, &blend_state->blend_constant))
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
blend_state = &pipeline->big_state->blend_state;
blend_state->blend_constant = *constant_color;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_blend_state_equal);
_cogl_pipeline_update_blend_enable (pipeline, state);
}
#endif
}
CoglHandle
cogl_pipeline_get_user_program (CoglPipeline *pipeline)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), COGL_INVALID_HANDLE);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_USER_SHADER);
return authority->big_state->user_program;
}
/* XXX: for now we don't mind if the program has vertex shaders
* attached but if we ever make a similar API public we should only
* allow attaching of programs containing fragment shaders. Eventually
* we will have a CoglPipeline abstraction to also cover vertex
* processing.
*/
void
cogl_pipeline_set_user_program (CoglPipeline *pipeline,
CoglHandle program)
{
CoglPipelineState state = COGL_PIPELINE_STATE_USER_SHADER;
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
if (authority->big_state->user_program == program)
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
if (program != COGL_INVALID_HANDLE)
{
_cogl_pipeline_set_fragend (pipeline, COGL_PIPELINE_FRAGEND_DEFAULT);
_cogl_pipeline_set_vertend (pipeline, COGL_PIPELINE_VERTEND_DEFAULT);
}
/* If we are the current authority see if we can revert to one of our
* ancestors being the authority */
if (pipeline == authority &&
_cogl_pipeline_get_parent (authority) != NULL)
{
CoglPipeline *parent = _cogl_pipeline_get_parent (authority);
CoglPipeline *old_authority =
_cogl_pipeline_get_authority (parent, state);
if (old_authority->big_state->user_program == program)
pipeline->differences &= ~state;
}
else if (pipeline != authority)
{
/* If we weren't previously the authority on this state then we
* need to extended our differences mask and so it's possible
* that some of our ancestry will now become redundant, so we
* aim to reparent ourselves if that's true... */
pipeline->differences |= state;
_cogl_pipeline_prune_redundant_ancestry (pipeline);
}
if (program != COGL_INVALID_HANDLE)
cogl_handle_ref (program);
if (authority == pipeline &&
pipeline->big_state->user_program != COGL_INVALID_HANDLE)
cogl_handle_unref (pipeline->big_state->user_program);
pipeline->big_state->user_program = program;
_cogl_pipeline_update_blend_enable (pipeline, state);
}
gboolean
cogl_pipeline_set_depth_state (CoglPipeline *pipeline,
const CoglDepthState *depth_state,
GError **error)
{
CoglPipelineState state = COGL_PIPELINE_STATE_DEPTH;
CoglPipeline *authority;
CoglDepthState *orig_state;
_COGL_GET_CONTEXT (ctx, FALSE);
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), FALSE);
_COGL_RETURN_VAL_IF_FAIL (depth_state->magic == COGL_DEPTH_STATE_MAGIC, FALSE);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
orig_state = &authority->big_state->depth_state;
if (orig_state->test_enabled == depth_state->test_enabled &&
orig_state->write_enabled == depth_state->write_enabled &&
orig_state->test_function == depth_state->test_function &&
orig_state->range_near == depth_state->range_near &&
orig_state->range_far == depth_state->range_far)
return TRUE;
if (ctx->driver == COGL_DRIVER_GLES1 &&
(depth_state->range_near != 0 ||
depth_state->range_far != 1))
{
g_set_error (error,
COGL_ERROR,
COGL_ERROR_UNSUPPORTED,
"glDepthRange not available on GLES 1");
return FALSE;
}
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
pipeline->big_state->depth_state = *depth_state;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_depth_state_equal);
return TRUE;
}
void
cogl_pipeline_get_depth_state (CoglPipeline *pipeline,
CoglDepthState *state)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_DEPTH);
*state = authority->big_state->depth_state;
}
CoglColorMask
cogl_pipeline_get_color_mask (CoglPipeline *pipeline)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), 0);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_LOGIC_OPS);
return authority->big_state->logic_ops_state.color_mask;
}
void
cogl_pipeline_set_color_mask (CoglPipeline *pipeline,
CoglColorMask color_mask)
{
CoglPipelineState state = COGL_PIPELINE_STATE_LOGIC_OPS;
CoglPipeline *authority;
CoglPipelineLogicOpsState *logic_ops_state;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
logic_ops_state = &authority->big_state->logic_ops_state;
if (logic_ops_state->color_mask == color_mask)
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
logic_ops_state = &pipeline->big_state->logic_ops_state;
logic_ops_state->color_mask = color_mask;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_logic_ops_state_equal);
}
void
_cogl_pipeline_set_fog_state (CoglPipeline *pipeline,
const CoglPipelineFogState *fog_state)
{
CoglPipelineState state = COGL_PIPELINE_STATE_FOG;
CoglPipeline *authority;
CoglPipelineFogState *current_fog_state;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
current_fog_state = &authority->big_state->fog_state;
if (current_fog_state->enabled == fog_state->enabled &&
cogl_color_equal (&current_fog_state->color, &fog_state->color) &&
current_fog_state->mode == fog_state->mode &&
current_fog_state->density == fog_state->density &&
current_fog_state->z_near == fog_state->z_near &&
current_fog_state->z_far == fog_state->z_far)
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
pipeline->big_state->fog_state = *fog_state;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_fog_state_equal);
}
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
void
pipeline: Make the backface culling experimental public This adds two new experimental public functions to replace the old internal _cogl_pipeline_set_cull_face_state function: void cogl_pipeline_set_cull_face_mode (CoglPipeline *pipeline, CoglPipelineCullFaceMode cull_face_mode); void cogl_pipeline_set_front_face_winding (CoglPipeline *pipeline, CoglWinding front_winding); There are also the corresponding getters. https://bugzilla.gnome.org/show_bug.cgi?id=663628 Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-08 13:58:33 +00:00
cogl_pipeline_set_cull_face_mode (CoglPipeline *pipeline,
CoglPipelineCullFaceMode cull_face_mode)
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
{
CoglPipelineState state = COGL_PIPELINE_STATE_CULL_FACE;
CoglPipeline *authority;
pipeline: Make the backface culling experimental public This adds two new experimental public functions to replace the old internal _cogl_pipeline_set_cull_face_state function: void cogl_pipeline_set_cull_face_mode (CoglPipeline *pipeline, CoglPipelineCullFaceMode cull_face_mode); void cogl_pipeline_set_front_face_winding (CoglPipeline *pipeline, CoglWinding front_winding); There are also the corresponding getters. https://bugzilla.gnome.org/show_bug.cgi?id=663628 Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-08 13:58:33 +00:00
CoglPipelineCullFaceState *cull_face_state;
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
pipeline: Make the backface culling experimental public This adds two new experimental public functions to replace the old internal _cogl_pipeline_set_cull_face_state function: void cogl_pipeline_set_cull_face_mode (CoglPipeline *pipeline, CoglPipelineCullFaceMode cull_face_mode); void cogl_pipeline_set_front_face_winding (CoglPipeline *pipeline, CoglWinding front_winding); There are also the corresponding getters. https://bugzilla.gnome.org/show_bug.cgi?id=663628 Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-08 13:58:33 +00:00
cull_face_state = &authority->big_state->cull_face_state;
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
pipeline: Make the backface culling experimental public This adds two new experimental public functions to replace the old internal _cogl_pipeline_set_cull_face_state function: void cogl_pipeline_set_cull_face_mode (CoglPipeline *pipeline, CoglPipelineCullFaceMode cull_face_mode); void cogl_pipeline_set_front_face_winding (CoglPipeline *pipeline, CoglWinding front_winding); There are also the corresponding getters. https://bugzilla.gnome.org/show_bug.cgi?id=663628 Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-08 13:58:33 +00:00
if (cull_face_state->mode == cull_face_mode)
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
pipeline: Make the backface culling experimental public This adds two new experimental public functions to replace the old internal _cogl_pipeline_set_cull_face_state function: void cogl_pipeline_set_cull_face_mode (CoglPipeline *pipeline, CoglPipelineCullFaceMode cull_face_mode); void cogl_pipeline_set_front_face_winding (CoglPipeline *pipeline, CoglWinding front_winding); There are also the corresponding getters. https://bugzilla.gnome.org/show_bug.cgi?id=663628 Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-08 13:58:33 +00:00
pipeline->big_state->cull_face_state.mode = cull_face_mode;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_cull_face_state_equal);
}
void
cogl_pipeline_set_front_face_winding (CoglPipeline *pipeline,
CoglWinding front_winding)
{
CoglPipelineState state = COGL_PIPELINE_STATE_CULL_FACE;
CoglPipeline *authority;
CoglPipelineCullFaceState *cull_face_state;
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
authority = _cogl_pipeline_get_authority (pipeline, state);
cull_face_state = &authority->big_state->cull_face_state;
if (cull_face_state->front_winding == front_winding)
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
pipeline->big_state->cull_face_state.front_winding = front_winding;
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_cull_face_state_equal);
}
CoglPipelineCullFaceMode
pipeline: Make the backface culling experimental public This adds two new experimental public functions to replace the old internal _cogl_pipeline_set_cull_face_state function: void cogl_pipeline_set_cull_face_mode (CoglPipeline *pipeline, CoglPipelineCullFaceMode cull_face_mode); void cogl_pipeline_set_front_face_winding (CoglPipeline *pipeline, CoglWinding front_winding); There are also the corresponding getters. https://bugzilla.gnome.org/show_bug.cgi?id=663628 Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-08 13:58:33 +00:00
cogl_pipeline_get_cull_face_mode (CoglPipeline *pipeline)
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
{
CoglPipelineState state = COGL_PIPELINE_STATE_CULL_FACE;
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline),
COGL_PIPELINE_CULL_FACE_MODE_NONE);
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
return authority->big_state->cull_face_state.mode;
}
pipeline: Make the backface culling experimental public This adds two new experimental public functions to replace the old internal _cogl_pipeline_set_cull_face_state function: void cogl_pipeline_set_cull_face_mode (CoglPipeline *pipeline, CoglPipelineCullFaceMode cull_face_mode); void cogl_pipeline_set_front_face_winding (CoglPipeline *pipeline, CoglWinding front_winding); There are also the corresponding getters. https://bugzilla.gnome.org/show_bug.cgi?id=663628 Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-08 13:58:33 +00:00
CoglWinding
cogl_pipeline_get_front_face_winding (CoglPipeline *pipeline)
{
CoglPipelineState state = COGL_PIPELINE_STATE_CULL_FACE;
CoglPipeline *authority;
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline),
COGL_PIPELINE_CULL_FACE_MODE_NONE);
authority = _cogl_pipeline_get_authority (pipeline, state);
return authority->big_state->cull_face_state.front_winding;
}
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
float
cogl_pipeline_get_point_size (CoglPipeline *pipeline)
{
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_VAL_IF_FAIL (cogl_is_pipeline (pipeline), FALSE);
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority =
_cogl_pipeline_get_authority (pipeline, COGL_PIPELINE_STATE_POINT_SIZE);
return authority->big_state->point_size;
}
void
cogl_pipeline_set_point_size (CoglPipeline *pipeline,
float point_size)
{
CoglPipelineState state = COGL_PIPELINE_STATE_POINT_SIZE;
CoglPipeline *authority;
Start to reduce dependence on glib Since we've had several developers from admirable projects say they would like to use Cogl but would really prefer not to pull in gobject,gmodule and glib as extra dependencies we are investigating if we can get to the point where glib is only an optional dependency. Actually we feel like we only make minimal use of glib anyway, so it may well be quite straightforward to achieve this. This adds a --disable-glib configure option that can be used to disable features that depend on glib. Actually --disable-glib doesn't strictly disable glib at this point because it's more helpful if cogl continues to build as we make incremental progress towards this. The first use of glib that this patch tackles is the use of g_return_val_if_fail and g_return_if_fail which have been replaced with equivalent _COGL_RETURN_VAL_IF_FAIL and _COGL_RETURN_IF_FAIL macros. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-10-13 21:34:30 +00:00
_COGL_RETURN_IF_FAIL (cogl_is_pipeline (pipeline));
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
authority = _cogl_pipeline_get_authority (pipeline, state);
if (authority->big_state->point_size == point_size)
return;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
pipeline->big_state->point_size = point_size;
_cogl_pipeline_update_authority (pipeline, authority, state,
_cogl_pipeline_point_size_equal);
}
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
static CoglBoxedValue *
_cogl_pipeline_override_uniform (CoglPipeline *pipeline,
int location)
{
CoglPipelineState state = COGL_PIPELINE_STATE_UNIFORMS;
CoglPipelineUniformsState *uniforms_state;
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
int override_index;
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
_COGL_GET_CONTEXT (ctx, NULL);
g_return_val_if_fail (cogl_is_pipeline (pipeline), NULL);
g_return_val_if_fail (location >= 0, NULL);
g_return_val_if_fail (location < ctx->n_uniform_names, NULL);
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
uniforms_state = &pipeline->big_state->uniforms_state;
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
/* Count the number of bits that are set below this location. That
should give us the position where our new value should lie */
override_index = _cogl_bitmask_popcount_upto (&uniforms_state->override_mask,
location);
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
_cogl_bitmask_set (&uniforms_state->changed_mask, location, TRUE);
/* If this pipeline already has an override for this value then we
can just use it directly */
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
if (_cogl_bitmask_get (&uniforms_state->override_mask, location))
return uniforms_state->override_values + override_index;
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
/* We need to create a new override value in the right position
within the array. This is pretty inefficient but the hope is that
it will be much more common to modify an existing uniform rather
than modify a new one so it is more important to optimise the
former case. */
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
if (uniforms_state->override_values == NULL)
{
g_assert (override_index == 0);
uniforms_state->override_values = g_new (CoglBoxedValue, 1);
}
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
else
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
{
/* We need to grow the array and copy in the old values */
CoglBoxedValue *old_values = uniforms_state->override_values;
int old_size = _cogl_bitmask_popcount (&uniforms_state->override_mask);
uniforms_state->override_values = g_new (CoglBoxedValue, old_size + 1);
/* Copy in the old values leaving a gap for the new value */
memcpy (uniforms_state->override_values,
old_values,
sizeof (CoglBoxedValue) * override_index);
pipeline: fix realloc of uniform overrides array If we need to realloc the uniforms overrides array for a pipeline to insert a new override then we copy the old state into the new allocation for the entries surrounding the inserted entry. This patch fixes a mistake in how we copied the old entries that follow the inserted entry since we were actually copying to begining of the new allocation and potentially reading from beyond the extents of the old allocation. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-12-30 01:08:10 +00:00
memcpy (uniforms_state->override_values + override_index + 1,
old_values + override_index,
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
sizeof (CoglBoxedValue) * (old_size - override_index));
g_free (old_values);
}
_cogl_boxed_value_init (uniforms_state->override_values + override_index);
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
_cogl_bitmask_set (&uniforms_state->override_mask, location, TRUE);
cogl-pipeline: Store the uniform overrides in an array instead of list Previously the uniform overrides were stored in a linked list. Now they are stored in a g_malloc'd array. The values are still tightly packed so that there is only a value for each uniform that has a corresponding bit in override_mask. The allocated size of the array always exactly corresponds to the number of bits set in the override_mask. This means that when a new uniform value is set on a pipeline it will have to grow the array and copy the old values in. The assumption is that setting a value for a new uniform is much less frequent then setting a value for an existing uniform so it makes more sense to optimise the latter. The advantage of using an array is that we can quickly jump to right boxed value given a uniform location by doing a population count in the bitmask for the number of bits less than the given uniform location. This can be done in O(1) time whereas the old approach using a list would scale by the number of bits set. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-04 17:56:44 +00:00
return uniforms_state->override_values + override_index;
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
}
void
cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline,
int uniform_location,
float value)
{
CoglBoxedValue *boxed_value;
boxed_value = _cogl_pipeline_override_uniform (pipeline, uniform_location);
_cogl_boxed_value_set_1f (boxed_value, value);
}
void
cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline,
int uniform_location,
int value)
{
CoglBoxedValue *boxed_value;
boxed_value = _cogl_pipeline_override_uniform (pipeline, uniform_location);
_cogl_boxed_value_set_1i (boxed_value, value);
}
void
cogl_pipeline_set_uniform_float (CoglPipeline *pipeline,
int uniform_location,
int n_components,
int count,
const float *value)
{
CoglBoxedValue *boxed_value;
boxed_value = _cogl_pipeline_override_uniform (pipeline, uniform_location);
_cogl_boxed_value_set_float (boxed_value, n_components, count, value);
}
void
cogl_pipeline_set_uniform_int (CoglPipeline *pipeline,
int uniform_location,
int n_components,
int count,
const int *value)
{
CoglBoxedValue *boxed_value;
boxed_value = _cogl_pipeline_override_uniform (pipeline, uniform_location);
_cogl_boxed_value_set_int (boxed_value, n_components, count, value);
}
void
cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline,
int uniform_location,
int dimensions,
int count,
gboolean transpose,
const float *value)
{
CoglBoxedValue *boxed_value;
boxed_value = _cogl_pipeline_override_uniform (pipeline, uniform_location);
_cogl_boxed_value_set_matrix (boxed_value,
dimensions,
count,
transpose,
value);
}
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
static void
_cogl_pipeline_add_vertex_snippet (CoglPipeline *pipeline,
CoglSnippet *snippet)
{
CoglPipelineState state = COGL_PIPELINE_STATE_VERTEX_SNIPPETS;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
_cogl_pipeline_snippet_list_add (&pipeline->big_state->vertex_snippets,
snippet);
}
static void
_cogl_pipeline_add_fragment_snippet (CoglPipeline *pipeline,
CoglSnippet *snippet)
{
CoglPipelineState state = COGL_PIPELINE_STATE_FRAGMENT_SNIPPETS;
/* - Flush journal primitives referencing the current state.
* - Make sure the pipeline has no dependants so it may be modified.
* - If the pipeline isn't currently an authority for the state being
* changed, then initialize that state from the current authority.
*/
_cogl_pipeline_pre_change_notify (pipeline, state, NULL, FALSE);
_cogl_pipeline_snippet_list_add (&pipeline->big_state->fragment_snippets,
snippet);
}
void
snippet: Move the hook to be a property of the snippet Instead of specifying the hook point when adding to the pipeline using a separate function for each hook, the hook is now a property of the snippet. The hook is set on construction and is then read-only. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 20:54:14 +00:00
cogl_pipeline_add_snippet (CoglPipeline *pipeline,
CoglSnippet *snippet)
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
{
snippet: Move the hook to be a property of the snippet Instead of specifying the hook point when adding to the pipeline using a separate function for each hook, the hook is now a property of the snippet. The hook is set on construction and is then read-only. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 20:54:14 +00:00
g_return_if_fail (cogl_is_pipeline (pipeline));
g_return_if_fail (cogl_is_snippet (snippet));
g_return_if_fail (snippet->hook < COGL_SNIPPET_FIRST_LAYER_HOOK);
if (snippet->hook < COGL_SNIPPET_FIRST_PIPELINE_FRAGMENT_HOOK)
_cogl_pipeline_add_vertex_snippet (pipeline, snippet);
else
_cogl_pipeline_add_fragment_snippet (pipeline, snippet);
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
}
gboolean
pipeline: Assume blend enabled when there are snippets Whenever snippets are enabled we can't determine whether the final color will be fully opaque so we just have to assume blending should be enabled. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 17:59:52 +00:00
_cogl_pipeline_has_non_layer_vertex_snippets (CoglPipeline *pipeline)
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
{
CoglPipeline *authority =
_cogl_pipeline_get_authority (pipeline,
COGL_PIPELINE_STATE_VERTEX_SNIPPETS);
return !COGL_LIST_EMPTY (&authority->big_state->vertex_snippets);
}
snippet: Add a hook for the layer texture coordinate transformation This adds a hook called COGL_SNIPPET_HOOK_TEXTURE_COORD_TRANSFORM. This can be used to alter the application of the layer user matrix to a texture coordinate or it can bypass it altogether. This is the first per-layer hook that affects the vertex shader state so the patch includes the boilerplate needed to get that to work. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-28 19:58:15 +00:00
static gboolean
check_layer_has_vertex_snippet (CoglPipelineLayer *layer,
void *user_data)
{
unsigned long state = COGL_PIPELINE_LAYER_STATE_VERTEX_SNIPPETS;
CoglPipelineLayer *authority =
_cogl_pipeline_layer_get_authority (layer, state);
gboolean *found_vertex_snippet = user_data;
if (!COGL_LIST_EMPTY (&authority->big_state->vertex_snippets))
{
*found_vertex_snippet = TRUE;
return FALSE;
}
return TRUE;
}
pipeline: Assume blend enabled when there are snippets Whenever snippets are enabled we can't determine whether the final color will be fully opaque so we just have to assume blending should be enabled. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 17:59:52 +00:00
gboolean
_cogl_pipeline_has_vertex_snippets (CoglPipeline *pipeline)
{
snippet: Add a hook for the layer texture coordinate transformation This adds a hook called COGL_SNIPPET_HOOK_TEXTURE_COORD_TRANSFORM. This can be used to alter the application of the layer user matrix to a texture coordinate or it can bypass it altogether. This is the first per-layer hook that affects the vertex shader state so the patch includes the boilerplate needed to get that to work. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-28 19:58:15 +00:00
gboolean found_vertex_snippet = FALSE;
if (_cogl_pipeline_has_non_layer_vertex_snippets (pipeline))
return TRUE;
_cogl_pipeline_foreach_layer_internal (pipeline,
check_layer_has_vertex_snippet,
&found_vertex_snippet);
return found_vertex_snippet;
pipeline: Assume blend enabled when there are snippets Whenever snippets are enabled we can't determine whether the final color will be fully opaque so we just have to assume blending should be enabled. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 17:59:52 +00:00
}
gboolean
_cogl_pipeline_has_non_layer_fragment_snippets (CoglPipeline *pipeline)
{
CoglPipeline *authority =
_cogl_pipeline_get_authority (pipeline,
COGL_PIPELINE_STATE_FRAGMENT_SNIPPETS);
return !COGL_LIST_EMPTY (&authority->big_state->fragment_snippets);
}
pipeline: Add a snippet hook for the texture lookup This adds a per-layer snippet hook for the texure lookup. Here the snippet can modify the texture coordinates used for the lookup or modify the texel resulting from the lookup. This is the first per-layer hook so this also adds the COGL_PIPELINE_LAYER_STATE_FRAGMENT_SNIPPETS state and all of the boilerplate needed to make that work. Most of the functions used by the pipeline state to manage the snippet list has been moved into cogl-pipeline-snippet.c so that it can be shared with the layer state. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 17:36:03 +00:00
static gboolean
check_layer_has_fragment_snippet (CoglPipelineLayer *layer,
void *user_data)
{
unsigned long state = COGL_PIPELINE_LAYER_STATE_FRAGMENT_SNIPPETS;
CoglPipelineLayer *authority =
_cogl_pipeline_layer_get_authority (layer, state);
gboolean *found_fragment_snippet = user_data;
if (!COGL_LIST_EMPTY (&authority->big_state->fragment_snippets))
{
*found_fragment_snippet = TRUE;
return FALSE;
}
return TRUE;
}
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
gboolean
_cogl_pipeline_has_fragment_snippets (CoglPipeline *pipeline)
{
pipeline: Add a snippet hook for the texture lookup This adds a per-layer snippet hook for the texure lookup. Here the snippet can modify the texture coordinates used for the lookup or modify the texel resulting from the lookup. This is the first per-layer hook so this also adds the COGL_PIPELINE_LAYER_STATE_FRAGMENT_SNIPPETS state and all of the boilerplate needed to make that work. Most of the functions used by the pipeline state to manage the snippet list has been moved into cogl-pipeline-snippet.c so that it can be shared with the layer state. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 17:36:03 +00:00
gboolean found_fragment_snippet = FALSE;
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
pipeline: Assume blend enabled when there are snippets Whenever snippets are enabled we can't determine whether the final color will be fully opaque so we just have to assume blending should be enabled. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 17:59:52 +00:00
if (_cogl_pipeline_has_non_layer_fragment_snippets (pipeline))
pipeline: Add a snippet hook for the texture lookup This adds a per-layer snippet hook for the texure lookup. Here the snippet can modify the texture coordinates used for the lookup or modify the texel resulting from the lookup. This is the first per-layer hook so this also adds the COGL_PIPELINE_LAYER_STATE_FRAGMENT_SNIPPETS state and all of the boilerplate needed to make that work. Most of the functions used by the pipeline state to manage the snippet list has been moved into cogl-pipeline-snippet.c so that it can be shared with the layer state. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 17:36:03 +00:00
return TRUE;
_cogl_pipeline_foreach_layer_internal (pipeline,
check_layer_has_fragment_snippet,
&found_fragment_snippet);
return found_fragment_snippet;
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
}
pipeline: split out all core state apis Since cogl-pipeline.c has become very unwieldy this make a start at trying to shape this code back into a manageable state. This patche moves all the API relating to core pipeline state into cogl-pipeline-state.c. This doesn't move code relating to layer state out nor does it move any of the code supporting the core design of CoglPipeline itself. This change alone factors out 2k lines of code from cogl-pipeline.c which is obviously a good start. The next step will be to factor out the layer state and then probably look at breaking all of this state code down into state-groups. Reviewed-by: Neil Roberts <neil@linux.intel.com>
2011-09-07 22:44:37 +00:00
void
_cogl_pipeline_hash_color_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
state->hash = _cogl_util_one_at_a_time_hash (state->hash, &authority->color,
_COGL_COLOR_DATA_SIZE);
}
void
_cogl_pipeline_hash_blend_enable_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
guint8 blend_enable = authority->blend_enable;
state->hash = _cogl_util_one_at_a_time_hash (state->hash, &blend_enable, 1);
}
void
_cogl_pipeline_hash_lighting_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
CoglPipelineLightingState *lighting_state =
&authority->big_state->lighting_state;
state->hash =
_cogl_util_one_at_a_time_hash (state->hash, lighting_state,
sizeof (CoglPipelineLightingState));
}
void
_cogl_pipeline_hash_alpha_func_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
CoglPipelineAlphaFuncState *alpha_state = &authority->big_state->alpha_state;
state->hash =
_cogl_util_one_at_a_time_hash (state->hash, &alpha_state->alpha_func,
sizeof (alpha_state->alpha_func));
}
void
_cogl_pipeline_hash_alpha_func_reference_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
CoglPipelineAlphaFuncState *alpha_state = &authority->big_state->alpha_state;
float ref = alpha_state->alpha_func_reference;
state->hash =
_cogl_util_one_at_a_time_hash (state->hash, &ref, sizeof (float));
}
void
_cogl_pipeline_hash_blend_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
CoglPipelineBlendState *blend_state = &authority->big_state->blend_state;
unsigned int hash;
_COGL_GET_CONTEXT (ctx, NO_RETVAL);
if (!authority->real_blend_enable)
return;
hash = state->hash;
#if defined(HAVE_COGL_GLES2) || defined(HAVE_COGL_GL)
if (ctx->driver != COGL_DRIVER_GLES1)
{
hash =
_cogl_util_one_at_a_time_hash (hash, &blend_state->blend_equation_rgb,
sizeof (blend_state->blend_equation_rgb));
hash =
_cogl_util_one_at_a_time_hash (hash, &blend_state->blend_equation_alpha,
sizeof (blend_state->blend_equation_alpha));
hash =
_cogl_util_one_at_a_time_hash (hash, &blend_state->blend_src_factor_alpha,
sizeof (blend_state->blend_src_factor_alpha));
hash =
_cogl_util_one_at_a_time_hash (hash, &blend_state->blend_dst_factor_alpha,
sizeof (blend_state->blend_dst_factor_alpha));
if (blend_state->blend_src_factor_rgb == GL_ONE_MINUS_CONSTANT_COLOR ||
blend_state->blend_src_factor_rgb == GL_CONSTANT_COLOR ||
blend_state->blend_dst_factor_rgb == GL_ONE_MINUS_CONSTANT_COLOR ||
blend_state->blend_dst_factor_rgb == GL_CONSTANT_COLOR)
{
hash =
_cogl_util_one_at_a_time_hash (hash, &blend_state->blend_constant,
sizeof (blend_state->blend_constant));
}
}
#endif
hash =
_cogl_util_one_at_a_time_hash (hash, &blend_state->blend_src_factor_rgb,
sizeof (blend_state->blend_src_factor_rgb));
hash =
_cogl_util_one_at_a_time_hash (hash, &blend_state->blend_dst_factor_rgb,
sizeof (blend_state->blend_dst_factor_rgb));
state->hash = hash;
}
void
_cogl_pipeline_hash_user_shader_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
CoglHandle user_program = authority->big_state->user_program;
state->hash = _cogl_util_one_at_a_time_hash (state->hash, &user_program,
sizeof (user_program));
}
void
_cogl_pipeline_hash_depth_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
CoglDepthState *depth_state = &authority->big_state->depth_state;
unsigned int hash = state->hash;
if (depth_state->test_enabled)
{
guint8 enabled = depth_state->test_enabled;
CoglDepthTestFunction function = depth_state->test_function;
hash = _cogl_util_one_at_a_time_hash (hash, &enabled, sizeof (enabled));
hash = _cogl_util_one_at_a_time_hash (hash, &function, sizeof (function));
}
if (depth_state->write_enabled)
{
guint8 enabled = depth_state->write_enabled;
float near_val = depth_state->range_near;
float far_val = depth_state->range_far;
hash = _cogl_util_one_at_a_time_hash (hash, &enabled, sizeof (enabled));
hash = _cogl_util_one_at_a_time_hash (hash, &near_val, sizeof (near_val));
hash = _cogl_util_one_at_a_time_hash (hash, &far_val, sizeof (far_val));
}
state->hash = hash;
}
void
_cogl_pipeline_hash_fog_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
CoglPipelineFogState *fog_state = &authority->big_state->fog_state;
unsigned long hash = state->hash;
if (!fog_state->enabled)
hash = _cogl_util_one_at_a_time_hash (hash, &fog_state->enabled,
sizeof (fog_state->enabled));
else
hash = _cogl_util_one_at_a_time_hash (hash, &fog_state,
sizeof (CoglPipelineFogState));
state->hash = hash;
}
void
_cogl_pipeline_hash_point_size_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
float point_size = authority->big_state->point_size;
state->hash = _cogl_util_one_at_a_time_hash (state->hash, &point_size,
sizeof (point_size));
}
void
_cogl_pipeline_hash_logic_ops_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
CoglPipelineLogicOpsState *logic_ops_state = &authority->big_state->logic_ops_state;
state->hash = _cogl_util_one_at_a_time_hash (state->hash, &logic_ops_state->color_mask,
sizeof (CoglColorMask));
}
Make backface culling be part of the legacy state This adds an internal function to set the backface culling state on a pipeline. This includes properties to set the culling mode (front, back or both) and also to set which face is considered the front (COGL_WINDING_CLOCKWISE or COGL_WINDING_COUNTER_CLOCKWISE). The actual front face flushed to GL depends on whether we are rendering to an offscreen buffer or not. This means that when changing between on- and off- screen framebuffers it now checks whether the last flushed pipeline has backface culling enabled and forces a reflush of the cull face state if so. The backface culling is now set on a pipeline as part of the legacy state. This is important because some code in Cogl assumes it can flush a temporary pipeline to revert to a known state, but previously this wouldn't disable backface culling so things such as flushing the clip stack could get confused. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-09-15 10:25:39 +00:00
void
_cogl_pipeline_hash_cull_face_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
CoglPipelineCullFaceState *cull_face_state
= &authority->big_state->cull_face_state;
/* The cull face state is considered equal if two pipelines are both
set to no culling. If the front winding property is ever used for
anything else or the hashing is used not just for drawing then
this would have to change */
if (cull_face_state->mode == COGL_PIPELINE_CULL_FACE_MODE_NONE)
state->hash =
_cogl_util_one_at_a_time_hash (state->hash,
&cull_face_state->mode,
sizeof (CoglPipelineCullFaceMode));
else
state->hash =
_cogl_util_one_at_a_time_hash (state->hash,
cull_face_state,
sizeof (CoglPipelineCullFaceState));
}
cogl-pipeline: Add support for setting uniform values This adds the following new public experimental functions to set uniform values on a CoglPipeline: void cogl_pipeline_set_uniform_1f (CoglPipeline *pipeline, int uniform_location, float value); void cogl_pipeline_set_uniform_1i (CoglPipeline *pipeline, int uniform_location, int value); void cogl_pipeline_set_uniform_float (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const float *value); void cogl_pipeline_set_uniform_int (CoglPipeline *pipeline, int uniform_location, int n_components, int count, const int *value); void cogl_pipeline_set_uniform_matrix (CoglPipeline *pipeline, int uniform_location, int dimensions, int count, gboolean transpose, const float *value); These are similar to the old functions used to set uniforms on a CoglProgram. To get a value to pass in as the uniform_location there is also: int cogl_pipeline_get_uniform_location (CoglPipeline *pipeline, const char *uniform_name); Conceptually the uniform locations are tied to the pipeline so that whenever setting a value for a new pipeline the application is expected to call this function. However in practice the uniform locations are global to the CoglContext. The names are stored in a linked list where the position in the list is the uniform location. The global indices are used so that each pipeline can store a mask of which uniforms it overrides. That way it is quicker to detect which uniforms are different from the last pipeline that used the same CoglProgramState so it can avoid flushing uniforms that haven't changed. Currently the values are not actually compared which means that it will only avoid flushing a uniform if there is a common ancestor that sets the value (or if the same pipeline is being flushed again - in which case the pipeline and its common ancestor are the same thing). The uniform values are stored in the big state of the pipeline as a sparse linked list. A bitmask stores which values have been overridden and only overridden values are stored in the linked list. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-03 17:20:43 +00:00
void
_cogl_pipeline_hash_uniforms_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
/* This isn't used anywhere yet because the uniform state doesn't
affect program generation. It's quite a hassle to implement so
let's just leave it until something actually needs it */
g_warn_if_reached ();
}
void
_cogl_pipeline_compare_uniform_differences (unsigned long *differences,
CoglPipeline *pipeline0,
CoglPipeline *pipeline1)
{
GSList *head0 = NULL;
GSList *head1 = NULL;
CoglPipeline *node0;
CoglPipeline *node1;
int len0 = 0;
int len1 = 0;
int count;
GSList *common_ancestor0;
GSList *common_ancestor1;
/* This algorithm is copied from
_cogl_pipeline_compare_differences(). It might be nice to share
the code more */
for (node0 = pipeline0; node0; node0 = _cogl_pipeline_get_parent (node0))
{
GSList *link = alloca (sizeof (GSList));
link->next = head0;
link->data = node0;
head0 = link;
len0++;
}
for (node1 = pipeline1; node1; node1 = _cogl_pipeline_get_parent (node1))
{
GSList *link = alloca (sizeof (GSList));
link->next = head1;
link->data = node1;
head1 = link;
len1++;
}
/* NB: There's no point looking at the head entries since we know both
* pipelines must have the same default pipeline as their root node. */
common_ancestor0 = head0;
common_ancestor1 = head1;
head0 = head0->next;
head1 = head1->next;
count = MIN (len0, len1) - 1;
while (count--)
{
if (head0->data != head1->data)
break;
common_ancestor0 = head0;
common_ancestor1 = head1;
head0 = head0->next;
head1 = head1->next;
}
for (head0 = common_ancestor0->next; head0; head0 = head0->next)
{
node0 = head0->data;
if ((node0->differences & COGL_PIPELINE_STATE_UNIFORMS))
{
const CoglPipelineUniformsState *uniforms_state =
&node0->big_state->uniforms_state;
_cogl_bitmask_set_flags (&uniforms_state->override_mask,
differences);
}
}
for (head1 = common_ancestor1->next; head1; head1 = head1->next)
{
node1 = head1->data;
if ((node1->differences & COGL_PIPELINE_STATE_UNIFORMS))
{
const CoglPipelineUniformsState *uniforms_state =
&node1->big_state->uniforms_state;
_cogl_bitmask_set_flags (&uniforms_state->override_mask,
differences);
}
}
}
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
void
_cogl_pipeline_hash_vertex_snippets_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
_cogl_pipeline_snippet_list_hash (&authority->big_state->vertex_snippets,
pipeline: Add a snippet hook for the texture lookup This adds a per-layer snippet hook for the texure lookup. Here the snippet can modify the texture coordinates used for the lookup or modify the texel resulting from the lookup. This is the first per-layer hook so this also adds the COGL_PIPELINE_LAYER_STATE_FRAGMENT_SNIPPETS state and all of the boilerplate needed to make that work. Most of the functions used by the pipeline state to manage the snippet list has been moved into cogl-pipeline-snippet.c so that it can be shared with the layer state. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 17:36:03 +00:00
&state->hash);
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
}
void
_cogl_pipeline_hash_fragment_snippets_state (CoglPipeline *authority,
CoglPipelineHashState *state)
{
_cogl_pipeline_snippet_list_hash (&authority->big_state->fragment_snippets,
pipeline: Add a snippet hook for the texture lookup This adds a per-layer snippet hook for the texure lookup. Here the snippet can modify the texture coordinates used for the lookup or modify the texel resulting from the lookup. This is the first per-layer hook so this also adds the COGL_PIPELINE_LAYER_STATE_FRAGMENT_SNIPPETS state and all of the boilerplate needed to make that work. Most of the functions used by the pipeline state to manage the snippet list has been moved into cogl-pipeline-snippet.c so that it can be shared with the layer state. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-25 17:36:03 +00:00
&state->hash);
cogl-pipeline: Add two hook points for adding shader snippets This adds two new public experimental functions for attaching CoglSnippets to two hook points on a CoglPipeline: void cogl_pipeline_add_vertex_hook (CoglPipeline *, CoglSnippet *) void cogl_pipeline_add_fragment_hook (CoglPipeline *, CoglSnippet *) The hooks are intended to be around the entire vertex or fragment processing. That means the pre string in the snippet will be inserted at the very top of the main function and the post function will be inserted at the very end. The declarations get inserted in the global scope. The snippets are stored in two separate linked lists with a structure containing an enum representing the hook point and a pointer to the snippet. The lists are meant to be for hooks that affect the vertex shader and fragment shader respectively. Although there are currently only two hooks and the names match these two lists, the intention is *not* that each new hook will be in a separate list. The separation of the lists is just to make it easier to determine which shader needs to be regenerated when a new snippet is added. When a pipeline becomes the authority for either the vertex or fragment snipper state, it simply copies the entire list from the previous authority (although of course the shader snippet objects are referenced instead of copied so it doesn't duplicate the source strings). Each string is inserted into its own block in the shader. This means that each string has its own scope so it doesn't need to worry about name collisions with variables in other snippets. However it does mean that the pre and post strings can't share variables. It could be possible to wrap both parts in one block and then wrap the actual inner hook code in another block, however this would mean that any further snippets within the outer snippet would be able to see those variables. Perhaps something to consider would be to put each snippet into its own function which calls another function between the pre and post strings to do further processing. The pipeline cache for generated programs was previously shared with the fragment shader cache because the state that affects vertex shaders was a subset of the state that affects fragment shaders. This is no longer the case because there is a separate state mask for vertex snippets so the program cache now has its own hash table. Reviewed-by: Robert Bragg <robert@linux.intel.com>
2011-11-17 16:52:21 +00:00
}
Reference in New Issue Copy Permalink