Christian König
de2eec6a3d
[g3dvl] move scaling to macroblocksize into vertex shader
2010-11-05 00:20:33 +01:00
Christian König
325233fdef
[g3dvl] move vertex normalisation into vertex shader
2010-11-04 23:23:42 +01:00
Christian König
b4c5c6f51a
[g3dvl] rework shader a bit
2010-11-04 21:41:39 +01:00
Christian König
2e4a7b7306
Fix zero block handling for field based mc
2010-11-03 00:38:07 +01:00
Christian König
0b75203c59
First try of field based mc
2010-10-30 01:42:16 +02:00
Christian König
41ed47d6b8
Merge branch 'master' of ssh://git.freedesktop.org/git/mesa/mesa into pipe-video
...
Conflicts:
src/gallium/include/pipe/p_format.h
2010-10-28 20:24:56 +02:00
José Fonseca
5479fa34d9
gallium: Avoid using __doc__ in python scripts.
2010-10-28 17:38:18 +01:00
Vinson Lee
a54ab4960b
st/mesa: Silence uninitialized variable warning.
...
Fixes this GCC warning.
state_tracker/st_program.c: In function 'st_print_shaders':
state_tracker/st_program.c:735: warning: 'sh' may be used uninitialized in this function
2010-10-28 06:08:19 -07:00
Tom Stellard
aa43176ebd
r300/compiler: Use rc_get_readers_normal() for presubtract optimizations
2010-10-27 22:49:50 -07:00
Kenneth Graunke
cbc966b57b
i965: Add bit operation support to the fragment shader backend.
2010-10-27 13:55:30 -07:00
Eric Anholt
9e3641bd0d
i965: Make FS uniforms be the actual type of the uniform at upload time.
...
This fixes some insanity that would otherwise be required for GLSL
1.30 bit ops or gen6 integer uniform operations in general, at the
cost of upload-time pain. Given that we only have that pain because
mesa's mangling our integer uniforms to be floats, this something that
should be fixed outside of the shader codegen.
2010-10-27 13:54:35 -07:00
Ian Romanick
502943049a
docs: add GL_EXT_separate_shader_objects to release notes
2010-10-27 13:45:29 -07:00
Ian Romanick
817ed68710
intel: Enable GL_EXT_separate_shader_objects in Intel drivers
2010-10-27 13:35:53 -07:00
Ian Romanick
f48915ec52
swrast: Enable GL_EXT_separate_shader_objects in software paths
2010-10-27 13:35:53 -07:00
Ian Romanick
84eba3ef71
Track separate programs for each stage
...
The assumption is that all stages are the same program or that
varyings are passed between stages using built-in varyings.
2010-10-27 13:35:53 -07:00
Ian Romanick
75c6f47288
mesa: Track an ActiveProgram distinct from CurrentProgram
...
ActiveProgram is the GL_EXT_separate_shader_objects state variable
used for glUniform calls. glUseProgram also sets this.
2010-10-27 13:35:53 -07:00
Ian Romanick
01abcf3b79
mesa: Add display list support for GL_EXT_separate_shader_objects functions
2010-10-27 13:35:53 -07:00
Ian Romanick
c72aa7fa58
mesa: Skeletal support for GL_EXT_separate_shader_objects
...
Really just filling in the entry points. None of them do anything
other than validate their inputs.
2010-10-27 13:35:53 -07:00
Ian Romanick
b97794c041
mesa: Add infrastructure to track GL_EXT_separate_shader_objects
2010-10-27 13:35:53 -07:00
Ian Romanick
44f6e17ebb
glapi: Commit files changed by previous commit
2010-10-27 13:35:53 -07:00
Ian Romanick
206e904f3c
glapi: Add GL_EXT_separate_shader_objects
2010-10-27 13:35:52 -07:00
Kenneth Graunke
3acc826520
Fix build on systems where "python" is python 3.
...
First, it changes autoconf to use a "python2" binary when available,
rather than plain "python" (which is ambiguous). Secondly, it changes
the Makefiles to use $(PYTHON) $(PYTHON_FLAGS) rather than calling
python directly.
Signed-off-by: Xavier Chantry <chantry.xavier@gmail.com>
Signed-off-by: Matthew William Cox <matt@mattcox.ca>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
2010-10-27 12:49:53 -07:00
Marek Olšák
676c3f08bd
r300g: add a default channel ordering of texture border for unhandled formats
...
It should fix the texture border for compressed textures.
Broken since 8449a4772a .
2010-10-27 21:21:23 +02:00
Alex Deucher
8ff7885e8f
r600c: add missing radeon_prepare_render() call on evergreen
2010-10-27 14:30:50 -04:00
Alex Deucher
b194b9b238
r100: revalidate after radeon_update_renderbuffers
...
This is a port of 603741a86d
to r100.
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
2010-10-27 13:53:29 -04:00
Vinson Lee
80adc8ac3b
swrast: Print out format on unexpected failure in _swrast_ReadPixels.
2010-10-27 10:16:18 -07:00
Vinson Lee
1b92eb1a4b
egl: Remove unnecessary headers.
2010-10-27 09:51:11 -07:00
Vinson Lee
e7343cd704
mesa: Remove unnecessary header.
2010-10-27 09:38:33 -07:00
Vinson Lee
21ce44374a
st/mesa: Remove unnecessary header.
2010-10-27 09:33:13 -07:00
Vinson Lee
d674ee2a4d
r600g: Silence uninitialized variable warnings.
2010-10-27 09:26:27 -07:00
Vinson Lee
d4cdd2fab0
mesa: Remove unnecessary headers.
2010-10-27 09:09:47 -07:00
Vinson Lee
3c8106402f
r300g: Silence uninitialized variable warning.
...
Fixes this GCC warning.
r300_state_derived.c: In function 'r300_update_derived_state':
r300_state_derived.c:593: warning: 'r' may be used uninitialized in this function
r300_state_derived.c:593: note: 'r' was declared here
2010-10-27 09:02:00 -07:00
Tilman Sauerbeck
8ad9d83fdf
r600g: Destroy the blitter.
...
This fix got lost in the state rework merge.
Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
2010-10-27 08:44:47 +02:00
Tilman Sauerbeck
c6b10cd986
r600g: In radeon_bo(), call LIST_INITHEAD early.
...
radeon_bo_destroy() will want to read the list field. Without this patch,
we'd end up evaluating the list pointers before they have been properly
set up when we destroyed the newly created bo if it cannot be mapped.
Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
2010-10-27 08:44:35 +02:00
Brian Paul
412b960883
mesa: rename function to _mesa_is_format_integer_color()
...
Be a bit more clear about its operation.
2010-10-26 20:30:42 -06:00
Brian Paul
ab50148fda
mesa: fix bug in _mesa_is_format_integer()
...
We only want to return true if it's an integer _color_ format, not a
depth and/or stencil format.
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=31143
2010-10-26 20:25:23 -06:00
Brian Paul
9b3c4d3e67
mesa: remove the unused _mesa_is_fragment_shader_active() function
...
This reverts commit 013d5ffeec .
2010-10-26 18:05:37 -06:00
Brian Paul
ccef2110ed
mesa: call _mesa_valid_to_render() in glDrawPixels, glCopyPixels, glBitmap
...
This lets us simplify and consolidate some state checking code.
This implements the GL_INVALID_OPERATION check for all drawing commands
required by GL_EXT_texture_integer.
2010-10-26 18:05:37 -06:00
Brian Paul
705978e283
mesa: do integer FB / shader validation check in _mesa_valid_to_render()
2010-10-26 18:05:37 -06:00
Eric Anholt
bb4f12f538
i965: Disable register spilling on gen6 until it's fixed.
...
Avoids GPU hang on glsl-fs-convolution-1.
2010-10-26 15:07:25 -07:00
Eric Anholt
00bfdac5b8
i965: Fix VS URB entry sizing.
...
I'm trying to clamp to a minimum of 1 URB row, not a maximum of 1.
Fixes:
glsl-kwin-blur
glsl-max-varying
glsl-routing
2010-10-26 15:07:10 -07:00
Eric Anholt
88087ba1bf
i965: Drop the eot argument to read messages, which can never be set.
2010-10-26 13:46:09 -07:00
Eric Anholt
3ee5d68075
i965: Add support for constant buffer loads on gen6.
...
Fixes glsl-fs-uniform-array-5.
2010-10-26 13:17:54 -07:00
Eric Anholt
519835de04
i965: Set up the constant buffer on gen6 when it's needed.
...
This was slightly confused because gen6_wm_constants does the push
constant buffer, while brw_wm_constants does pull constants.
2010-10-26 13:15:01 -07:00
Eric Anholt
6488cf46f5
i965: Fix typo in comment about state flags.
2010-10-26 12:19:46 -07:00
Eric Anholt
33c4b2370f
i965: Handle new ir_unop_round_even in channel expression splitting.
2010-10-26 11:23:27 -07:00
Eric Anholt
62452e7d94
i965: Add support for discard instructions on gen6.
...
It's a little more painful than before because we don't have the handy
mask register any more, and have to make do with cooking up a value
out of the flag register.
2010-10-26 11:21:44 -07:00
Eric Anholt
9b1d26f78f
i965: Add disasm for the flag register.
2010-10-26 11:21:44 -07:00
Eric Anholt
0e8c834ffa
i965: Clear some undefined fields of g0 when using them for gen6 FB writes.
...
This doesn't appear to help any testcases I'm looking at, but it looks
like it's required.
2010-10-26 10:34:14 -07:00
Eric Anholt
1732a8bc72
i965: Use SENDC on the first render target write on gen6.
...
This is apparently required, as the thread will be initiated while it
still has dependencies, and this is what waits for those to be
resolved before writing color.
2010-10-26 10:34:10 -07:00