mesa.git
14 years agoMerge branch 'gallium-embedded'
José Fonseca [Wed, 3 Feb 2010 18:54:13 +0000 (18:54 +0000)]
Merge branch 'gallium-embedded'

14 years agogallium/docs: Document the OS module.
José Fonseca [Wed, 3 Feb 2010 15:56:36 +0000 (15:56 +0000)]
gallium/docs: Document the OS module.

14 years agotgsi: added debugging code to catch divide by zero
Brian Paul [Wed, 3 Feb 2010 15:50:32 +0000 (08:50 -0700)]
tgsi: added debugging code to catch divide by zero

14 years agotgsi: convert CHECK_INF_OR_NAN to inline function
Brian Paul [Wed, 3 Feb 2010 15:47:07 +0000 (08:47 -0700)]
tgsi: convert CHECK_INF_OR_NAN to inline function

And disable with if (0).  Inf/NaN can occur normally during program
execution.  Only enable the check code when needed during debugging.

14 years agomesa: Factor out the fb initialization details from _mesa_new_framebuffer.
Francisco Jerez [Wed, 3 Feb 2010 11:21:04 +0000 (03:21 -0800)]
mesa: Factor out the fb initialization details from _mesa_new_framebuffer.

This should make things easier for drivers wanting to work with a
"subclass" of gl_framebuffer.

The complementary "_mesa_initialize_framebuffer" function is now
called "_mesa_initialize_window_framebuffer" for the sake of
symmetry.

Signed-off-by: Brian Paul <brianp@vmware.com>
14 years agoos: Also add the aligned memory prototypes to the embedded section.
José Fonseca [Wed, 3 Feb 2010 12:57:49 +0000 (12:57 +0000)]
os: Also add the aligned memory prototypes to the embedded section.

14 years agoutil: Reimplement all utility functions in terms of the new OS abstraction.
José Fonseca [Wed, 3 Feb 2010 12:13:06 +0000 (12:13 +0000)]
util: Reimplement all utility functions in terms of the new OS abstraction.

14 years agoos: New OS abstraction module.
José Fonseca [Wed, 3 Feb 2010 12:11:58 +0000 (12:11 +0000)]
os: New OS abstraction module.

14 years agovbo: fix void * arithmetic warning
Keith Whitwell [Wed, 3 Feb 2010 11:52:14 +0000 (11:52 +0000)]
vbo: fix void * arithmetic warning

14 years agovbo: Fix up in-place splitting for non-contiguous/indexed primitives.
Francisco Jerez [Wed, 3 Feb 2010 11:18:28 +0000 (03:18 -0800)]
vbo: Fix up in-place splitting for non-contiguous/indexed primitives.

The in-place splitting code wasn't dealing with index buffers at all
(and it was being called from vbo_split_prims for too big index
buffers, causing some occasional corruption).

Additionally, it wasn't taking into account primitives arrays with
non-contiguous indices (e.g. given prim[0].start = 0 and prim[1].start
= max_verts, it would happily call back the driver with (max_index -
min_index) still greater than max_verts, causing infinite recursion).

It still doesn't handle too large indexed vertex buffers: use
vbo_split_copy for that.

14 years agotrace: Include the appropriate header instead of defining prototypes.
José Fonseca [Wed, 3 Feb 2010 11:06:23 +0000 (11:06 +0000)]
trace: Include the appropriate header instead of defining prototypes.

Avoids header when the header was included before.

14 years agoprogs/fpglsl: couple more tests
Keith Whitwell [Tue, 2 Feb 2010 20:20:11 +0000 (20:20 +0000)]
progs/fpglsl: couple more tests

14 years agoprogs/fpglsl: fp-tris for glsl
Keith Whitwell [Tue, 2 Feb 2010 20:01:30 +0000 (20:01 +0000)]
progs/fpglsl: fp-tris for glsl

14 years agosvga: fix TXD and TXL opcode translation
Keith Whitwell [Tue, 2 Feb 2010 18:01:52 +0000 (18:01 +0000)]
svga: fix TXD and TXL opcode translation

14 years agosvga: deriv insns not valid in dynamic flow control either
Keith Whitwell [Tue, 2 Feb 2010 17:45:32 +0000 (17:45 +0000)]
svga: deriv insns not valid in dynamic flow control either

14 years agosvga: texture from lod zero inside dynamic branching
Keith Whitwell [Tue, 2 Feb 2010 15:34:01 +0000 (15:34 +0000)]
svga: texture from lod zero inside dynamic branching

Texture derivatives are potentially undefined inside dynamic branches,
so hardwire lod zero in this case.  Treating all if/endif and loop
constructs as dynamic branches.

14 years agosvga: better method for generating white fs color outputs
Keith Whitwell [Tue, 26 Jan 2010 15:52:46 +0000 (15:52 +0000)]
svga: better method for generating white fs color outputs

14 years agomesa/st: bump the gallium version number
Keith Whitwell [Tue, 12 Jan 2010 16:19:29 +0000 (16:19 +0000)]
mesa/st: bump the gallium version number

This is a very informal version number, but there have been enough
changes that a bump is appropriate at this time.

14 years agost/mesa: remove some floating point divides in viewport calculation
Keith Whitwell [Mon, 11 Jan 2010 16:15:37 +0000 (16:15 +0000)]
st/mesa: remove some floating point divides in viewport calculation

Compiler can't usually turn x/2.0f into x * 0.5f, though we're happy
with either.

14 years agogallium-docs: Fix typo.
José Fonseca [Wed, 3 Feb 2010 08:56:34 +0000 (08:56 +0000)]
gallium-docs: Fix typo.

14 years agor300: fix compiler bugs introduced with MRT changes.
Dave Airlie [Wed, 3 Feb 2010 08:56:40 +0000 (18:56 +1000)]
r300: fix compiler bugs introduced with MRT changes.

the first looks like a definite bug, the second I'm not so confident
of but it works.

Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agoegl: Rename _eglPreloadForEach callbacks.
Chia-I Wu [Wed, 3 Feb 2010 04:38:03 +0000 (12:38 +0800)]
egl: Rename _eglPreloadForEach callbacks.

Rename _eglPreloadFile to _eglLoaderFile, and _eglPreloadFile to
_eglLoaderFile.  There are several _eglPreload* functions.  The
callbacks do not belong to them and it should be better to give them a
different prefix.

14 years agoegl: Fix a potential segfault in driver suffix matching.
Chia-I Wu [Wed, 3 Feb 2010 04:35:57 +0000 (12:35 +0800)]
egl: Fix a potential segfault in driver suffix matching.

The driver suffix might be NULL on some platforms.  Perform the matching
only when there is a suffix.

14 years agoegl: Remove unnecessary #ifdef's.
Chia-I Wu [Tue, 2 Feb 2010 09:05:56 +0000 (17:05 +0800)]
egl: Remove unnecessary #ifdef's.

There were some refactorings with the addition of EGL_DRIVERS_PATH.
They made some platform #ifdef's unnecessary.

14 years agoegl: Ignore certain environment variables when setuid/setgid.
Chia-I Wu [Tue, 2 Feb 2010 08:47:53 +0000 (16:47 +0800)]
egl: Ignore certain environment variables when setuid/setgid.

Specifically, ignore EGL_DRIVERS_PATH when an application is
setuid/setgid.  And ignore EGL_DRIVER when there is a slash in it.

14 years agoegl: Add EGL_DRIVERS_PATH environment variable.
Chia-I Wu [Tue, 2 Feb 2010 07:34:55 +0000 (15:34 +0800)]
egl: Add EGL_DRIVERS_PATH environment variable.

EGL_DRIVERS_PATH gives a list of colon-separated directories.  The given
directories will be searched when preloading drivers.

This is based on Mike Stroyan's patch, which honors the variable in
_eglPreloadDisplayDrivers.  It is extended to honor the variable also in
_eglPreloadUserDriver and _eglPreloadDefaultDriver in this version.

14 years agodraw: remove old hard-coded shader length limits
Brian Paul [Wed, 3 Feb 2010 04:34:06 +0000 (21:34 -0700)]
draw: remove old hard-coded shader length limits

14 years agotgsi: added tgsi_alloc_tokens()
Brian Paul [Wed, 3 Feb 2010 04:23:23 +0000 (21:23 -0700)]
tgsi: added tgsi_alloc_tokens()

14 years agodraw: re-implement free_bit() with ffs()
Brian Paul [Wed, 3 Feb 2010 04:17:56 +0000 (21:17 -0700)]
draw: re-implement free_bit() with ffs()

14 years agodraw: comments and var renaming
Brian Paul [Wed, 3 Feb 2010 04:16:25 +0000 (21:16 -0700)]
draw: comments and var renaming

14 years agotgsi: fix comment
Brian Paul [Wed, 3 Feb 2010 01:28:25 +0000 (18:28 -0700)]
tgsi: fix comment

14 years agodraw: add const qualifiers, fix return types
Brian Paul [Wed, 3 Feb 2010 01:54:12 +0000 (18:54 -0700)]
draw: add const qualifiers, fix return types

14 years agosoftpipe: remove unnecessary #include
Brian Paul [Wed, 3 Feb 2010 01:53:53 +0000 (18:53 -0700)]
softpipe: remove unnecessary #include

14 years agodraw: new, updated comments
Brian Paul [Wed, 3 Feb 2010 01:50:19 +0000 (18:50 -0700)]
draw: new, updated comments

Try to avoid future confusion between different, but similar functions.

14 years agollvmpipe: fix (potentially) broken AA points, AA lines
Brian Paul [Wed, 3 Feb 2010 01:43:11 +0000 (18:43 -0700)]
llvmpipe: fix (potentially) broken AA points, AA lines

This patch duplicates the softpipe fix seen in the preceeding commit.

14 years agosoftpipe: fix broken AA points, AA lines
Brian Paul [Wed, 3 Feb 2010 01:42:22 +0000 (18:42 -0700)]
softpipe: fix broken AA points, AA lines

This fixes a regression when the geometry shaders branch was merged
to master with commit 89d8577fb3036547ef0b47498cc8dc5c77f886e0.

progs/demos/pointblast and progs/redbook/anti work again.

14 years agogallium/docs: Opcode refs.
Corbin Simpson [Wed, 3 Feb 2010 01:07:26 +0000 (17:07 -0800)]
gallium/docs: Opcode refs.

There is a very real possibility that I may be enjoying this too much.

14 years agogallium/docs: TGSI notes on replication to dst, and also cleanups.
Corbin Simpson [Wed, 3 Feb 2010 01:02:28 +0000 (17:02 -0800)]
gallium/docs: TGSI notes on replication to dst, and also cleanups.

Slowly, surely, I'm making this better.

14 years agogallium/docs: Start turning notes into docs.
Corbin Simpson [Wed, 3 Feb 2010 00:36:34 +0000 (16:36 -0800)]
gallium/docs: Start turning notes into docs.

Yeah, what's that? You can read it? It's got, hmm, *readability*? Nice,
ain't it. :3

14 years agogallium/docs: Add opcode formatting.
Corbin Simpson [Wed, 3 Feb 2010 00:20:12 +0000 (16:20 -0800)]
gallium/docs: Add opcode formatting.

14 years agogallium/docs: default Z texture tuple is (z,z,z,1) for OpenGL
Brian Paul [Tue, 2 Feb 2010 23:27:07 +0000 (16:27 -0700)]
gallium/docs: default Z texture tuple is (z,z,z,1) for OpenGL

14 years agogallium/docs: s/convensions/conventions/
Brian Paul [Tue, 2 Feb 2010 23:20:26 +0000 (16:20 -0700)]
gallium/docs: s/convensions/conventions/

14 years agor300g: Texture format cleanups, per docs.
Corbin Simpson [Tue, 2 Feb 2010 21:03:31 +0000 (13:03 -0800)]
r300g: Texture format cleanups, per docs.

texenv is perfect now.

14 years agogallium: Move some compiler portability stuff into p_compiler.h
José Fonseca [Tue, 2 Feb 2010 20:55:08 +0000 (20:55 +0000)]
gallium: Move some compiler portability stuff into p_compiler.h

14 years agogallium/docs: Did somebody call for a table?
Corbin Simpson [Tue, 2 Feb 2010 20:44:22 +0000 (12:44 -0800)]
gallium/docs: Did somebody call for a table?

14 years agoTrack frag shader changes introduced by commit 4769566500be1a53dd9b4cc1a613aef439a0e3d8
Scott Moreau [Tue, 2 Feb 2010 18:59:53 +0000 (11:59 -0700)]
Track frag shader changes introduced by commit 4769566500be1a53dd9b4cc1a613aef439a0e3d8

Signed-off-by: Corbin Simpson <MostAwesomeDude@gmail.com>
14 years agogallium: Move deprecated pipe/internal/p_winsys_screen.h inside util/u_simple_screen.h
José Fonseca [Tue, 2 Feb 2010 16:09:56 +0000 (16:09 +0000)]
gallium: Move deprecated pipe/internal/p_winsys_screen.h inside util/u_simple_screen.h

14 years agomesa: Add a BITSET_FFS function.
Francisco Jerez [Tue, 2 Feb 2010 11:07:08 +0000 (03:07 -0800)]
mesa: Add a BITSET_FFS function.

It will be useful for the nouveau DRI driver and IMHO there's no
reason to keep it private.

Signed-off-by: Brian Paul <brianp@vmware.com>
14 years agogallium: Move p_thread.h and p_atomic.h out of gallium interfaces.
José Fonseca [Tue, 2 Feb 2010 15:18:01 +0000 (15:18 +0000)]
gallium: Move p_thread.h and p_atomic.h out of gallium interfaces.

Into os/os_thread.h and util/u_atomic.h respectively.

14 years agogallium/docs: table of texture formats and (x,y,z,w) tuples returned by TEX
Brian Paul [Tue, 2 Feb 2010 15:14:40 +0000 (08:14 -0700)]
gallium/docs: table of texture formats and (x,y,z,w) tuples returned by TEX

Maybe someone can convert this into a real table instead of just
preformatted text.

14 years agogallium: Make pipe_atomic a regular int32_t.
José Fonseca [Tue, 2 Feb 2010 14:43:48 +0000 (14:43 +0000)]
gallium: Make pipe_atomic a regular int32_t.

14 years agogallium: pipe/p_inlines.h -> util/u_inlines.h
José Fonseca [Tue, 2 Feb 2010 14:42:17 +0000 (14:42 +0000)]
gallium: pipe/p_inlines.h -> util/u_inlines.h

14 years agor300compiler: Add MRT number to debugging output.
Corbin Simpson [Wed, 11 Nov 2009 11:08:08 +0000 (03:08 -0800)]
r300compiler: Add MRT number to debugging output.

14 years agor300compiler, r300 classic, r300g: Add support for MRTs in the frag shader.
Corbin Simpson [Wed, 11 Nov 2009 11:06:51 +0000 (03:06 -0800)]
r300compiler, r300 classic, r300g: Add support for MRTs in the frag shader.

This maybe breaks the vert compiler. Hopefully not.

14 years agotgsi: ureg_DECL_gs_input() accepts shader semantics.
Michal Krol [Tue, 2 Feb 2010 09:42:13 +0000 (10:42 +0100)]
tgsi: ureg_DECL_gs_input() accepts shader semantics.

14 years agovega: implement tgsi_ureg shaders cache
Igor Oliveira [Tue, 2 Feb 2010 02:02:15 +0000 (22:02 -0400)]
vega: implement tgsi_ureg shaders cache

14 years agovega: change tgsi asm by tgsi_ureg
Igor Oliveira [Tue, 2 Feb 2010 02:01:51 +0000 (22:01 -0400)]
vega: change tgsi asm by tgsi_ureg

14 years agor200: Fix EXT_fogcoord rendering.
Pauli Nieminen [Tue, 2 Feb 2010 03:39:28 +0000 (05:39 +0200)]
r200: Fix EXT_fogcoord rendering.

The fogcoord calue was not pushed to GPU because of implicit float to int conversion.
Fix is to use float pointer to buffer object so no conversion is done in assigment

14 years agor300g: Fix funky segfault.
Corbin Simpson [Tue, 2 Feb 2010 03:47:46 +0000 (19:47 -0800)]
r300g: Fix funky segfault.

14 years agost/egl: Wrong context might be used in eglBindTexImage.
Chia-I Wu [Tue, 2 Feb 2010 03:22:51 +0000 (11:22 +0800)]
st/egl: Wrong context might be used in eglBindTexImage.

If the pbuffer surface is current, gctx is set to the current context.
Make sure it is set to OpenGL ES 1 context again before binding.

14 years agodocs: Update for EGL environment variables.
Chia-I Wu [Tue, 2 Feb 2010 03:05:19 +0000 (11:05 +0800)]
docs: Update for EGL environment variables.

Mention that EGL_DRIVER should be a full path, and add a link from
envvars.html to egl.html.

14 years agoprogs/es2: Fix an assertion failure in tri.
Chia-I Wu [Tue, 2 Feb 2010 02:40:32 +0000 (10:40 +0800)]
progs/es2: Fix an assertion failure in tri.

EGL_SURFACE_TYPE is a config attribute, not a surface attribute.  Thanks
to Mike Stroyan for pointing out this error.

14 years agost/mesa: fix texture deallocation bug
Brian Paul [Tue, 2 Feb 2010 02:04:16 +0000 (19:04 -0700)]
st/mesa: fix texture deallocation bug

This fixes a bug reported by Christoph Bumiller on mesa3d-dev.

When a texture is first created as RGBA, then re-defined with
glTexImage(internalFormat=GL_DEPTH_COMPONENT) we failed to deallocate
the original texture.  When this texture was bound as a FBO surface,
the depth/Z surface format was RGBA instead of Z.  Depending on the
driver this led to a failed assertion or FBO validation failure.

This patch does three things:

1. Remove ancient code that mysteriously tested if we were replacing
the smallest mipmap level and tested if the texture was not a cube map
texture.  I can't see any reason for those tests.

2. Move the width=height=depth=0 test to after the code which frees
texture data.  Calling glTexImage with width=height=depth=0 and data=NULL
is a way to free a single mipmap level.

3. Update the code comments.

There are no apparent conform, glean or piglit regressions from this change.

(cherry picked from commit 43e4b584227534e30e487e7fb7e99d6501cbcd85)

14 years agoprogs/tests: add another FBO test to exercise a RGBA vs. Z texture bug
Brian Paul [Tue, 2 Feb 2010 02:02:52 +0000 (19:02 -0700)]
progs/tests: add another FBO test to exercise a RGBA vs. Z texture bug

See comments in file for more details.

14 years agomesa: change _mesa_find_free_register() to find multiple free regs
Brian Paul [Tue, 2 Feb 2010 01:00:12 +0000 (18:00 -0700)]
mesa: change _mesa_find_free_register() to find multiple free regs

Before, _mesa_find_free_register() would scan the given shader to
find a free/unused register of the given type.  But subsequent calls
would return the same register again.  This caused a failure in the
_mesa_remove_output_reads() function which sometimes needs several
free temps.

Now use a new function which build a vector of 'used' flags and another
function which searches that vector for an unused register starting at
a position that's incremented for each call.

Fixes fd.o bug 26317.  Note that a regression test for this has been
added to the glean/glsl1 test.

(cherry picked from commit e0d01c9d7f46ccd531f8dd1a04c5ac067200ef1e)

14 years agor300g: Inline immediates where possible.
Corbin Simpson [Mon, 1 Feb 2010 20:47:03 +0000 (12:47 -0800)]
r300g: Inline immediates where possible.

Oh look, more gears. Seems to work just fine though.

14 years agost/mesa: remove duplicate calculation of fp input mapping
Keith Whitwell [Mon, 1 Feb 2010 20:42:34 +0000 (20:42 +0000)]
st/mesa: remove duplicate calculation of fp input mapping

This was being calculated the same way in two different places.
Now just do it in st_translate_fragment_program().

14 years agost/mesa: remove dead stfp input_map array
Keith Whitwell [Mon, 1 Feb 2010 20:32:32 +0000 (20:32 +0000)]
st/mesa: remove dead stfp input_map array

Was being calculated and not used.  Also was probably incorrect...

14 years agomesa: added _mesa_print_vp/p_inputs() functions (debug aids)
Brian Paul [Mon, 1 Feb 2010 20:03:25 +0000 (13:03 -0700)]
mesa: added _mesa_print_vp/p_inputs() functions (debug aids)

14 years agoglslcompiler: add glapi_nop.o to OBJECTS
Brian Paul [Sat, 30 Jan 2010 20:10:08 +0000 (13:10 -0700)]
glslcompiler: add glapi_nop.o to OBJECTS

14 years agogallium: Use MSVC atomic intrinsics directly instead of the Windows header wrappers.
José Fonseca [Mon, 1 Feb 2010 16:12:44 +0000 (16:12 +0000)]
gallium: Use MSVC atomic intrinsics directly instead of the Windows header wrappers.

14 years agogallium: Remove unnecessary includes. Add others to compensate.
José Fonseca [Mon, 1 Feb 2010 16:11:46 +0000 (16:11 +0000)]
gallium: Remove unnecessary includes. Add others to compensate.

14 years agomesa: Add missing includes.
José Fonseca [Mon, 1 Feb 2010 15:12:20 +0000 (15:12 +0000)]
mesa: Add missing includes.

14 years agollvmpipe: Add missing includes.
José Fonseca [Mon, 1 Feb 2010 15:11:34 +0000 (15:11 +0000)]
llvmpipe: Add missing includes.

14 years agoi915: Add missing includes.
José Fonseca [Mon, 1 Feb 2010 15:11:18 +0000 (15:11 +0000)]
i915: Add missing includes.

14 years agoutil: Add missing includes.
José Fonseca [Mon, 1 Feb 2010 15:11:06 +0000 (15:11 +0000)]
util: Add missing includes.

14 years agodraw: Add missing includes.
José Fonseca [Mon, 1 Feb 2010 15:10:51 +0000 (15:10 +0000)]
draw: Add missing includes.

14 years agogallium: Drop mutex-based pipe_atomic implementation.
José Fonseca [Mon, 1 Feb 2010 14:23:36 +0000 (14:23 +0000)]
gallium: Drop mutex-based pipe_atomic implementation.

First step to severe the dependency in OS abstractions from the Gallium
interfaces.

14 years agomesa: Remove unnecessary headers.
Vinson Lee [Mon, 1 Feb 2010 08:43:33 +0000 (00:43 -0800)]
mesa: Remove unnecessary headers.

14 years agoglx: Remove unnecessary headers.
Vinson Lee [Mon, 1 Feb 2010 08:22:50 +0000 (00:22 -0800)]
glx: Remove unnecessary headers.

14 years agost/dri: Remove duplicate type qualifier.
Vinson Lee [Mon, 1 Feb 2010 07:52:48 +0000 (23:52 -0800)]
st/dri: Remove duplicate type qualifier.

14 years agoi965g: Add missing va_end.
Vinson Lee [Mon, 1 Feb 2010 07:47:07 +0000 (23:47 -0800)]
i965g: Add missing va_end.

14 years agor300g: Fix memory leak on memory allocation failure.
Vinson Lee [Mon, 1 Feb 2010 06:38:25 +0000 (22:38 -0800)]
r300g: Fix memory leak on memory allocation failure.

14 years agor600: Remove duplicate assignment.
Vinson Lee [Mon, 1 Feb 2010 06:33:01 +0000 (22:33 -0800)]
r600: Remove duplicate assignment.

14 years agoegl: Remove unnecessary headers.
Vinson Lee [Mon, 1 Feb 2010 05:56:53 +0000 (21:56 -0800)]
egl: Remove unnecessary headers.

14 years agonv04: Update to new nouveau_class.h
Marcin Kościelnicki [Sun, 31 Jan 2010 19:31:01 +0000 (19:31 +0000)]
nv04: Update to new nouveau_class.h

Signed-off-by: Francisco Jerez <currojerez@riseup.net>
14 years agonv50/nv40/nv30: fix small memory leak of nouveau_resources on screen_destroy
Marcin Slusarz [Tue, 12 Jan 2010 14:41:50 +0000 (15:41 +0100)]
nv50/nv40/nv30: fix small memory leak of nouveau_resources on screen_destroy

Signed-off-by: Francisco Jerez <currojerez@riseup.net>
14 years agonv10, nv20: Fix build.
Francisco Jerez [Sun, 31 Jan 2010 16:18:09 +0000 (17:18 +0100)]
nv10, nv20: Fix build.

14 years agoegl: Clean up surface attributes.
Chia-I Wu [Sun, 31 Jan 2010 09:56:30 +0000 (17:56 +0800)]
egl: Clean up surface attributes.

Add missing attributes and use correct types in _EGLSurface.  Remove
ifdef tests that serve no purpose.  Update _eglQuerySurface and
_eglSurfaceAttrib for missing queries and checks.

14 years agoegl: Add support for more EGLImage extensions to EGL core.
Chia-I Wu [Sun, 31 Jan 2010 09:34:02 +0000 (17:34 +0800)]
egl: Add support for more EGLImage extensions to EGL core.

Add support EGL_KHR_vg_parent_image and EGL_KHR_gl_*.  This is as simple
as adding some flags that can be enabled.  Individual drivers need to
implement the extensions before enbaling the flags.

14 years agost/mesa: Remove unnecessary headers.
Vinson Lee [Sun, 31 Jan 2010 10:34:51 +0000 (02:34 -0800)]
st/mesa: Remove unnecessary headers.

14 years agomesa: Remove unnecessary headers.
Vinson Lee [Sun, 31 Jan 2010 10:30:12 +0000 (02:30 -0800)]
mesa: Remove unnecessary headers.

14 years agoi965: Silence uninitialized variable warning.
Vinson Lee [Sun, 31 Jan 2010 08:44:57 +0000 (00:44 -0800)]
i965: Silence uninitialized variable warning.

14 years agor300g: Silence uninitialized variable warning.
Vinson Lee [Sun, 31 Jan 2010 06:50:35 +0000 (22:50 -0800)]
r300g: Silence uninitialized variable warning.

14 years agoegl: Surface type validaction should depend on extensions supported.
Chia-I Wu [Sun, 31 Jan 2010 06:39:06 +0000 (14:39 +0800)]
egl: Surface type validaction should depend on extensions supported.

EGL_SCREEN_BIT_MESA is a valid bit only when EGL_MESA_screen_surface is
supported.

14 years agoegl: Initialize display configs with the display.
Chia-I Wu [Sun, 31 Jan 2010 06:32:31 +0000 (14:32 +0800)]
egl: Initialize display configs with the display.

This changes _eglInitConfig to take the display as its argument.

14 years agoegl: Initialize display resources with their display.
Chia-I Wu [Sun, 31 Jan 2010 05:33:57 +0000 (13:33 +0800)]
egl: Initialize display resources with their display.

Change _eglInitSurface, _eglInitContext, and _eglInitImage to take an
_EGLDisplay instead of an _EGLDriver.  This is a more natural form, and
plus, the display encodes information such as the extensions supported
that might be required for attribute list parsing.

14 years agoegl: Refactor _eglInitImage.
Chia-I Wu [Sun, 31 Jan 2010 05:04:12 +0000 (13:04 +0800)]
egl: Refactor _eglInitImage.

Refactor attribute list parsing code to _eglParseImageAttribList.

14 years agoegl: Refactor _eglInitContext.
Chia-I Wu [Sun, 31 Jan 2010 04:57:53 +0000 (12:57 +0800)]
egl: Refactor _eglInitContext.

Refactor attribute list parsing code to _eglParseContextAttribList.  Add
a check to make sure the config supports the client API.

14 years agoegl: Refactor _eglInitSurface.
Chia-I Wu [Sun, 31 Jan 2010 04:13:59 +0000 (12:13 +0800)]
egl: Refactor _eglInitSurface.

Refactor attribute list parsing code to _eglParseSurfaceAttribList.