AuroraMiddleware/gtk2
1
0
Fork 0
forked from AuroraMiddleware/gtk
Code Issues Pull Requests Projects Releases Wiki Activity

gsk: Ensure that non-drawing nodes are supported

Browse Source 
Non-drawing nodes should skip the texturing phase.
This commit is contained in:
Emmanuele Bassi 2016-08-09 16:30:13 +01:00
parent b2f5a69742
commit c9514b45ff
1 changed files with 48 additions and 51 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

99
gsk/gskglrenderer.c
View File

@ -413,19 +413,22 @@ render_item (GskGLRenderer *self,
glUseProgram (item->render_data.program_id);
/* Use texture unit 0 for the source */
glUniform1i (item->render_data.source_location, 0);
gsk_gl_driver_bind_source_texture (self->gl_driver, item->render_data.texture_id);
if (item->parent_data != NULL)
if (item->render_data.texture_id != 0)
{
glUniform1i (item->render_data.blendMode_location, item->blend_mode);
/* Use texture unit 0 for the source */
glUniform1i (item->render_data.source_location, 0);
gsk_gl_driver_bind_source_texture (self->gl_driver, item->render_data.texture_id);
/* Use texture unit 1 for the mask */
if (item->parent_data->texture_id != 0)
if (item->parent_data != NULL)
{
glUniform1i (item->render_data.mask_location, 1);
gsk_gl_driver_bind_mask_texture (self->gl_driver, item->parent_data->texture_id);
glUniform1i (item->render_data.blendMode_location, item->blend_mode);
/* Use texture unit 1 for the mask */
if (item->parent_data->texture_id != 0)
{
glUniform1i (item->render_data.mask_location, 1);
gsk_gl_driver_bind_mask_texture (self->gl_driver, item->parent_data->texture_id);
}
}
}
@ -659,22 +662,6 @@ gsk_gl_renderer_add_render_item (GskGLRenderer *self,
else
item.parent_data = NULL;
if (render_node_needs_render_target (node))
{
item.render_data.render_target_id =
gsk_gl_driver_create_texture (self->gl_driver, item.size.width, item.size.height);
gsk_gl_driver_init_texture_empty (self->gl_driver, item.render_data.render_target_id);
gsk_gl_driver_create_render_target (self->gl_driver, item.render_data.render_target_id, TRUE, TRUE);
item.children = g_array_sized_new (FALSE, FALSE, sizeof (RenderItem),
gsk_render_node_get_n_children (node));
}
else
{
item.render_data.render_target_id = self->texture_id;
item.children = NULL;
}
/* Select the program to use */
if (parent != NULL)
program_id = self->blend_program_id;
@ -700,6 +687,39 @@ gsk_gl_renderer_add_render_item (GskGLRenderer *self,
item.render_data.uv_location =
gsk_shader_builder_get_attribute_location (self->shader_builder, program_id, self->attributes[UV]);
if (render_node_needs_render_target (node))
{
item.render_data.render_target_id =
gsk_gl_driver_create_texture (self->gl_driver, item.size.width, item.size.height);
gsk_gl_driver_init_texture_empty (self->gl_driver, item.render_data.render_target_id);
gsk_gl_driver_create_render_target (self->gl_driver, item.render_data.render_target_id, TRUE, TRUE);
item.children = g_array_sized_new (FALSE, FALSE, sizeof (RenderItem),
gsk_render_node_get_n_children (node));
}
else
{
item.render_data.render_target_id = self->texture_id;
item.children = NULL;
}
surface = gsk_render_node_get_surface (node);
/* If the node does not draw anything, we skip it */
if (surface == NULL && item.render_data.render_target_id == self->texture_id)
goto out;
/* Upload the Cairo surface to a GL texture */
item.render_data.texture_id = gsk_gl_driver_create_texture (self->gl_driver,
item.size.width,
item.size.height);
gsk_gl_driver_bind_source_texture (self->gl_driver, item.render_data.texture_id);
gsk_gl_driver_init_texture_with_surface (self->gl_driver,
item.render_data.texture_id,
surface,
self->gl_min_filter,
self->gl_mag_filter);
/* Create the vertex buffers holding the geometry of the quad */
{
GskQuadVertex vertex_data[N_VERTICES] = {
@ -723,40 +743,17 @@ gsk_gl_renderer_add_render_item (GskGLRenderer *self,
gsk_renderer_get_projection (GSK_RENDERER (self), &projection);
item.z = project_item (&projection, &mv);
/* TODO: This should really be an asset atlas, to avoid uploading a ton
* of textures. Ideally we could use a single Cairo surface to get around
* the GL texture limits and reorder the texture data on the CPU side and
* do a single upload; alternatively, we could use a separate FBO and
* render each texture into it
*/
surface = gsk_render_node_get_surface (node);
if (surface != NULL)
{
/* Upload the Cairo surface to a GL texture */
item.render_data.texture_id = gsk_gl_driver_create_texture (self->gl_driver,
item.size.width,
item.size.height);
gsk_gl_driver_bind_source_texture (self->gl_driver, item.render_data.texture_id);
gsk_gl_driver_init_texture_with_surface (self->gl_driver,
item.render_data.texture_id,
surface,
self->gl_min_filter,
self->gl_mag_filter);
}
GSK_NOTE (OPENGL, g_print ("%*sAdding node <%s>[%p] to render items\n",
2 * node_depth (node), "",
node->name != NULL ? node->name : "unnamed",
node));
g_array_append_val (render_items, item);
ritem = &g_array_index (render_items,
RenderItem,
render_items->len - 1);
ritem = &g_array_index (render_items, RenderItem, render_items->len - 1);
if (item.children != NULL)
render_items = item.children;
out:
gsk_render_node_iter_init (&iter, node);
while (gsk_render_node_iter_next (&iter, &child))
gsk_gl_renderer_add_render_item (self, render_items, child, ritem);