1
0
Fork 0
mutter-performance-source/gl/cogl-primitives.c
Robert Bragg 2b9478a665 [cogl] Improving Cogl journal to minimize driver overheads + GPU state changes
Previously the journal was always flushed at the end of
_cogl_rectangles_with_multitexture_coords, (i.e.  the end of any
cogl_rectangle* calls) but now we have broadened the potential for batching
geometry.  In ideal circumstances we will only flush once per scene.

In summary the journal works like this:

When you use any of the cogl_rectangle* APIs then nothing is emitted to the
GPU at this point, we just log one or more quads into the journal.  A
journal entry consists of the quad coordinates, an associated material
reference, and a modelview matrix.  Ideally the journal only gets flushed
once at the end of a scene, but in fact there are things to consider that
may cause unwanted flushing, including:

- modifying materials mid-scene
    This is because each quad in the journal has an associated material
    reference (i.e. not copy), so if you try and modify a material that is
    already referenced in the journal we force a flush first)

    NOTE: For now this means you should avoid using cogl_set_source_color()
	      since that currently uses a single shared material. Later we
	  should change it to use a pool of materials that is recycled
	  when the journal is flushed.

- modifying any state that isn't currently logged, such as depth, fog and
  backface culling enables.

The first thing that happens when flushing, is to upload all the vertex data
associated with the journal into a single VBO.

We then go through a process of splitting up the journal into batches that
have compatible state so they can be emitted to the GPU together.  This is
currently broken up into 3 levels so we can stagger the state changes:

1) we break the journal up according to changes in the number of material layers
   associated with logged quads. The number of layers in a material determines
   the stride of the associated vertices, so we have to update our vertex
   array offsets at this level. (i.e. calling gl{Vertex,Color},Pointer etc)
2) we further split batches up according to material compatability. (e.g.
   materials with different textures) We flush material state at this level.
3) Finally we split batches up according to modelview changes. At this level
   we update the modelview matrix and actually emit the actual draw command.

This commit is largely about putting the initial design in-place; this will be
followed by other changes that take advantage of the extended batching.
2009-06-30 17:13:34 +01:00

274 lines
8.1 KiB
C

/*
* Cogl
*
* An object oriented GL/GLES Abstraction/Utility Layer
*
* Copyright (C) 2007,2008,2009 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, write to the
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
* Boston, MA 02111-1307, USA.
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "cogl.h"
#include "cogl-internal.h"
#include "cogl-context.h"
#include "cogl-clip-stack.h"
#include "cogl-material-private.h"
#include <string.h>
#include <gmodule.h>
#include <math.h>
#define _COGL_MAX_BEZ_RECURSE_DEPTH 16
void
_cogl_path_add_node (gboolean new_sub_path,
float x,
float y)
{
CoglPathNode new_node;
_COGL_GET_CONTEXT (ctx, NO_RETVAL);
new_node.x = x;
new_node.y = y;
new_node.path_size = 0;
if (new_sub_path || ctx->path_nodes->len == 0)
ctx->last_path = ctx->path_nodes->len;
g_array_append_val (ctx->path_nodes, new_node);
g_array_index (ctx->path_nodes, CoglPathNode, ctx->last_path).path_size++;
if (ctx->path_nodes->len == 1)
{
ctx->path_nodes_min.x = ctx->path_nodes_max.x = x;
ctx->path_nodes_min.y = ctx->path_nodes_max.y = y;
}
else
{
if (x < ctx->path_nodes_min.x) ctx->path_nodes_min.x = x;
if (x > ctx->path_nodes_max.x) ctx->path_nodes_max.x = x;
if (y < ctx->path_nodes_min.y) ctx->path_nodes_min.y = y;
if (y > ctx->path_nodes_max.y) ctx->path_nodes_max.y = y;
}
}
void
_cogl_path_stroke_nodes ()
{
guint path_start = 0;
gulong enable_flags = COGL_ENABLE_VERTEX_ARRAY;
CoglMaterialFlushOptions options;
_COGL_GET_CONTEXT (ctx, NO_RETVAL);
enable_flags |= _cogl_material_get_cogl_enable_flags (ctx->source_material);
cogl_enable (enable_flags);
options.flags = COGL_MATERIAL_FLUSH_DISABLE_MASK;
/* disable all texture layers */
options.disable_layers = (guint32)~0;
_cogl_material_flush_gl_state (ctx->source_material, &options);
_cogl_current_matrix_state_flush ();
while (path_start < ctx->path_nodes->len)
{
CoglPathNode *path = &g_array_index (ctx->path_nodes, CoglPathNode,
path_start);
GE( glVertexPointer (2, GL_FLOAT, sizeof (CoglPathNode),
(guchar *) path
+ G_STRUCT_OFFSET (CoglPathNode, x)) );
GE( glDrawArrays (GL_LINE_STRIP, 0, path->path_size) );
path_start += path->path_size;
}
}
static void
_cogl_path_get_bounds (floatVec2 nodes_min,
floatVec2 nodes_max,
float *bounds_x,
float *bounds_y,
float *bounds_w,
float *bounds_h)
{
*bounds_x = nodes_min.x;
*bounds_y = nodes_min.y;
*bounds_w = nodes_max.x - *bounds_x;
*bounds_h = nodes_max.y - *bounds_y;
}
void
_cogl_add_path_to_stencil_buffer (floatVec2 nodes_min,
floatVec2 nodes_max,
guint path_size,
CoglPathNode *path,
gboolean merge)
{
guint path_start = 0;
guint sub_path_num = 0;
float bounds_x;
float bounds_y;
float bounds_w;
float bounds_h;
gulong enable_flags = COGL_ENABLE_VERTEX_ARRAY;
CoglHandle prev_source;
_COGL_GET_CONTEXT (ctx, NO_RETVAL);
_cogl_journal_flush ();
/* Just setup a simple material that doesn't use texturing... */
prev_source = cogl_handle_ref (ctx->source_material);
cogl_set_source (ctx->stencil_material);
_cogl_material_flush_gl_state (ctx->source_material, NULL);
enable_flags |=
_cogl_material_get_cogl_enable_flags (ctx->source_material);
cogl_enable (enable_flags);
_cogl_path_get_bounds (nodes_min, nodes_max,
&bounds_x, &bounds_y, &bounds_w, &bounds_h);
if (merge)
{
GE( glStencilMask (2) );
GE( glStencilFunc (GL_LEQUAL, 0x2, 0x6) );
}
else
{
GE( glClear (GL_STENCIL_BUFFER_BIT) );
GE( glStencilMask (1) );
GE( glStencilFunc (GL_LEQUAL, 0x1, 0x3) );
}
GE( glEnable (GL_STENCIL_TEST) );
GE( glStencilOp (GL_INVERT, GL_INVERT, GL_INVERT) );
GE( glColorMask (FALSE, FALSE, FALSE, FALSE) );
GE( glDepthMask (FALSE) );
while (path_start < path_size)
{
/* NB: after calling _cogl_journal_flush the current matrix
* state is undefined */
_cogl_current_matrix_state_flush ();
GE( glVertexPointer (2, GL_FLOAT, sizeof (CoglPathNode),
(guchar *) path
+ G_STRUCT_OFFSET (CoglPathNode, x)) );
GE( glDrawArrays (GL_TRIANGLE_FAN, 0, path->path_size) );
if (sub_path_num > 0)
{
/* Union the two stencil buffers bits into the least
significant bit */
GE( glStencilMask (merge ? 6 : 3) );
GE( glStencilOp (GL_ZERO, GL_REPLACE, GL_REPLACE) );
cogl_rectangle (bounds_x, bounds_y,
bounds_x + bounds_w, bounds_y + bounds_h);
_cogl_journal_flush ();
GE( glStencilOp (GL_INVERT, GL_INVERT, GL_INVERT) );
}
GE( glStencilMask (merge ? 4 : 2) );
path_start += path->path_size;
path += path->path_size;
sub_path_num++;
}
if (merge)
{
/* Now we have the new stencil buffer in bit 1 and the old
stencil buffer in bit 0 so we need to intersect them */
GE( glStencilMask (3) );
GE( glStencilFunc (GL_NEVER, 0x2, 0x3) );
GE( glStencilOp (GL_DECR, GL_DECR, GL_DECR) );
/* Decrement all of the bits twice so that only pixels where the
value is 3 will remain */
_cogl_set_current_matrix (COGL_MATRIX_PROJECTION);
_cogl_current_matrix_push ();
_cogl_current_matrix_identity ();
/* Cogl generally assumes the modelview matrix is current, so since
* cogl_rectangle will be flushing GL state and emitting geometry
* to OpenGL it will be confused if we leave the projection matrix
* active... */
_cogl_set_current_matrix (COGL_MATRIX_MODELVIEW);
_cogl_current_matrix_push ();
_cogl_current_matrix_identity ();
cogl_rectangle (-1.0, -1.0, 1.0, 1.0);
cogl_rectangle (-1.0, -1.0, 1.0, 1.0);
_cogl_journal_flush ();
_cogl_current_matrix_pop ();
_cogl_set_current_matrix (COGL_MATRIX_PROJECTION);
_cogl_current_matrix_pop ();
_cogl_set_current_matrix (COGL_MATRIX_MODELVIEW);
}
GE( glStencilMask (~(GLuint) 0) );
GE( glDepthMask (TRUE) );
GE( glColorMask (TRUE, TRUE, TRUE, TRUE) );
GE( glStencilFunc (GL_EQUAL, 0x1, 0x1) );
GE( glStencilOp (GL_KEEP, GL_KEEP, GL_KEEP) );
/* restore the original material */
cogl_set_source (prev_source);
cogl_handle_unref (prev_source);
}
void
_cogl_path_fill_nodes ()
{
float bounds_x;
float bounds_y;
float bounds_w;
float bounds_h;
_COGL_GET_CONTEXT (ctx, NO_RETVAL);
_cogl_path_get_bounds (ctx->path_nodes_min, ctx->path_nodes_max,
&bounds_x, &bounds_y, &bounds_w, &bounds_h);
_cogl_add_path_to_stencil_buffer (ctx->path_nodes_min,
ctx->path_nodes_max,
ctx->path_nodes->len,
&g_array_index (ctx->path_nodes,
CoglPathNode, 0),
ctx->clip.stencil_used);
cogl_rectangle (bounds_x, bounds_y,
bounds_x + bounds_w, bounds_y + bounds_h);
/* The stencil buffer now contains garbage so the clip area needs to
be rebuilt */
ctx->clip.stack_dirty = TRUE;
}