| Age | Commit message (Collapse) | Author |
|---|
|
|
|
Before, we were sometimes rendering into a stale texture because
st_finalize_texture() would discard the old texture and create a new one.
Moved st_update_framebuffer atom after texture validation so that we
can create a new renderbuffer surface if the texture changes.
Also, split texture validation into two parts: finalize_textures and
update_textures. Do finalize_textures first to avoid getting into the
situtation where we're doing a pipe->surface_copy() mid-way through
state validation.
Some debug code still in place, but disabled...
|
|
|
|
|
|
This fixes the case where the app calls SwapBuffers then calls glReadPixels
to read the front color buffer. We now keep track of when the front buffer
is a _logically_ copy of the back buffer (after SwapBuffers) and read from
the back color buffer instead of the front.
|
|
|
|
elsewhere
|
|
(blit, gen-mipmaps, quad-clear, etc)
Also, additional cso_set_*() functions for viewport, framebuffer, blend color,
etc. state.
|
|
now.
|
|
Previously, we set _NEW_PIXEL and _NEW_COLOR in these functions, respectively.
That correponds to the GL attribute groups, but doesn't make much sense
otherwise. This could improve validation efficiency in a few places too.
It looks like all the drivers are already checking for _NEW_BUFFERS in the
right places (since that's the bit for FBO state) so we can trim out
_NEW_PIXEL and _NEW_COLOR at any time.
|
|
|
|
|
|
The memcmp is insufficient for eliminating redundant framebuffer state changes.
|
|
|
|
Remove surface ptr from gl_renderbuffer.
Use st_renderbuffer in most places.
More clean-up.
|
|
|
|
|
|
Z testing now works with i915 driver.
Add gl_renderbuffer::surface pointer (and reverse pointer).
Remove intel_surface and xmesa_surface types - no longer used.
|
|
|
|
|
|
We should be able to render to any depth/format of X window now.
|
|
|
