32a95cb70c
mesa: Fix GetVertexAttrib* inside display lists.
...
GetVertexAttrib*{,ARB} is no longer aliased to the NV calls.
This fixes tracing yofrankie with apitrace, given it requires accurate
results from GetVertexAttribiv*.
NOTE: This is a candidate for the stable branches.
2011-05-12 01:09:55 +01:00
95c93651b4
r300/compiler: Fix bug in rc_get_variables()
...
Variables that share readers were not always being linked together.
https://bugs.freedesktop.org/show_bug.cgi?id=36939
2011-05-11 16:16:29 -07:00
4612554dce
r300/compiler: Limit instructions to 3 source selects
...
Some presubtract conversions were generating more than 3 source
selects.
https://bugs.freedesktop.org/show_bug.cgi?id=36527
Note: This is a candidate for the 7.10 branch.
2011-05-11 16:16:29 -07:00
d1e8195c07
r300/compiler: Add simple unit test framework
...
Plus three tests for rc_inst_can_use_presub()
2011-05-11 16:16:29 -07:00
b9f2750900
r600g: fix flushes on rs780/rs880
...
They need the same hack as rv670.
Fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=35312
Signed-off-by: Alex Deucher <alexdeucher@gmail.com >
2011-05-11 12:10:01 -04:00
7a271151a5
mesa: Don't append fog code for programs that don't output color.
...
Fixes fdo 36919.
NOTE: This is a candidate for the stable branches.
It should be cherry-picked to the sames branches that
3aa21f93dc
2011-05-11 14:04:08 +01:00
2812b60a84
svga/drm: Implement svga_winsys_screen::get_hw_version.
2011-05-10 16:36:35 +01:00
a67ecfeebc
st/wgl: Remove buggy assertion.
...
The assertion is wrong, now that state tracker can cope with a window with
zero width or height.
2011-05-10 16:13:05 +01:00
d6b248652a
nv50: fix mistake in pipe caps for pre-NVA0 cards
...
Introduced by 531b12af35
2011-05-10 10:16:54 +02:00
68074387a4
ir_to_mesa: Emit TXD instruction.
...
Mesa already supports this because of NV_fragment_program.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org >
Tested-by: Marek Olšák <maraeo@gmail.com >
2011-05-09 11:23:54 -07:00
5a3a242a8f
glsl: Add compiler support for ARB_shader_texture_lod.
...
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org >
Tested-by: Marek Olšák <maraeo@gmail.com >
2011-05-09 11:23:54 -07:00
1d4eb8d5d8
mesa: Add ARB_shader_texture_lod to the extension list; off by default.
...
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org >
2011-05-09 11:23:54 -07:00
4e1b1cbd0d
nv50/nvc0: make transfers aware of PIPE_TRANSFER_MAP_DIRECTLY
...
If state tracker asked us to map resource directly and we can't
do it (because of tiling), return NULL instead of doing full transfer
- state tracker should handle it and fallback to some other method
or repeat transfer without PIPE_TRANSFER_MAP_DIRECTLY.
It greatly improves performance of xorg state tracker on nv50+,
because its fallback (DFS/UTS) is much faster than full transfer.
2011-05-09 20:19:19 +02:00
86852236a3
r300/compiler: align memory allocations to 8-bytes
...
Eliminates unaligned accesses on strict architectures. Spotted by Jay
Estabrook.
Signed-off-by: Matt Turner <mattst88@gmail.com >
NOTE: This is a candidate for the 7.10 branch.
2011-05-09 20:01:58 +02:00
7297f2ac3c
r300g: unreference a zbuffer surface after decompression
2011-05-09 20:01:58 +02:00
56b39fcd8b
r300g: remove redundant state variable hyperz_locked
2011-05-09 20:01:58 +02:00
801fbdf286
mesa: document instructions ir_to_mesa emits
...
GLSL stopped using:
BRA, EXP, LOG, LRP, NRM3, NRM4, XPD.
GLSL started using:
KIL, SCS, SSG, SWZ.
(omg why SWZ? isn't proc_src_register flexible enough?)
GLSL doesn't use these opcodes some Radeons do support:
ARR, DP2A, DST, LRP, XPD.
These opcodes are now unused:
AND, NOT, NRM3, NRM4, OR, XOR.
(plus maybe the NV extensions which are unused by Gallium)
In addition to that, we don't use two-dimensional indirect addressing,
which the Mesa IR can do.
2011-05-09 20:01:58 +02:00
39be542f73
r300c: Fix up for register allocator rewrite.
...
Was broken by commit fe622bac0c
2011-05-09 18:26:50 +02:00
b9e8ea6a27
r600g: anisotropic filtering support for evergreen hw
...
Signed-off-by: Jerome Glisse <jglisse@redhat.com >
2011-05-09 12:18:41 -04:00
ad904cdf98
r600g: add support for anisotropic filtering
2011-05-09 12:09:16 -04:00
9d792d013c
gallium: Factor out unknown endianness detection.
2011-05-09 12:29:47 +02:00
991c1f0249
Don't allow compilation if endianness isn't known
...
PIPE_ARCH_UNKNOWN_ENDIAN is used no where else. All #else branches of
ifdef PIPE_ARCH_LITTLE assume big-endian. Not #error'ing out here
only serves to allow bad things to happen.
Signed-off-by: Matt Turner <mattst88@gmail.com >
2011-05-09 12:23:21 +02:00
5e945863ca
p_config.h: Rely on glibc endianness definitions when available
...
Suggested by Julien Cristau.
Signed-off-by: Matt Turner <mattst88@gmail.com >
2011-05-09 12:22:59 +02:00
bd3ce30ee7
st/xorg: add some debugging messages to xorg_exa.c
2011-05-09 12:22:50 +02:00
6e0263fdd4
st/xorg: remove DEBUG_PRINT macro and add exa_debug_printf
...
Localizes preprocessor usage to one place.
2011-05-09 12:22:45 +02:00
7c5eaf5a2d
st/xorg: fix compilation of xorg_exa.c with DEBUG_PRINT set to 1
2011-05-09 12:22:24 +02:00
ce31aaec02
[g3dvl] move dct_type detection out of get_macroblock_modes
2011-05-08 22:28:33 +02:00
7074801ed7
cell: fix unbalanced parenthesis
...
Signed-off-by: Nicolas Kaiser <nikai@nikai.net >
Signed-off-by: Brian Paul <brianp@vmware.com >
2011-05-08 07:41:26 -06:00
3b773d06d2
[g3dvl] fix field selection of mb without mc
2011-05-08 13:39:56 +02:00
626352648a
[g3dvl] memcpy the DCT buffer instead of writing it directly
...
It looks like texture buffers are not cached so this seems to be alot faster
2011-05-07 18:09:31 +02:00
e06a09ac37
vdpau: use multiple buffers to speed up rendering
2011-05-07 15:28:04 +02:00
cd13ec253a
vdpau: implement PresentationQueueSetBackgroundColor
2011-05-07 15:00:52 +02:00
e0cc970a54
vdpau: as long as we don't have a background picture we don't want an alpha channel
2011-05-07 14:51:17 +02:00
6a5d2d7967
[g3dvl] rgba surfaces are clearing if the alpha channel is one
2011-05-07 14:49:58 +02:00
a67a0bb60c
[g3dvl] using reciprocals for size calc is sometimes not precise enough
2011-05-07 14:39:35 +02:00
213b9004a6
Merge remote-tracking branch 'origin/master' into pipe-video
...
Conflicts:
src/gallium/drivers/r600/r600_state.c
2011-05-07 14:11:40 +02:00
03615c02d8
r300/compiler: implement TXD and TXL opcodes
2011-05-07 02:51:38 +02:00
d8e222d887
r300g: handle seamless_cube_map caps
2011-05-07 02:51:38 +02:00
6bb230970c
egl/wayland: Follow wl_resource_destroy changes
2011-05-06 16:50:45 -04:00
531b12af35
nv50,nvc0: activate seamless cube map filtering
2011-05-06 21:11:03 +02:00
27d3e0b25c
docs: update relnotes-7.11
2011-05-06 20:14:28 +02:00
483d2ea5e3
docs: update GL3 status
2011-05-06 20:14:28 +02:00
d931b0d8b3
r600g: implement seamless_cube_map for evergreen
...
The r600/r700 support will follow soon.
2011-05-06 20:13:05 +02:00
a5f0a11477
gallium: implement seamless cubemap extensions
...
Reviewed-by: Brian Paul <brianp@vmware.com >
2011-05-06 20:06:30 +02:00
bf65c9ea41
mesa: handle TEXTURE_CUBE_MAP_SEAMLESS in SamplerParameter
...
Reviewed-by: Brian Paul <brianp@vmware.com >
2011-05-06 20:06:30 +02:00
a19c42ffc6
mesa: implement AMD_seamless_cubemap_per_texture
2011-05-06 20:06:30 +02:00
3b20a89941
r600g: avoid recursion in rv670 flush workaround
...
Should fix:
https://bugs.freedesktop.org/show_bug.cgi?id=36914
Signed-off-by: Alex Deucher <alexdeucher@gmail.com >
2011-05-06 13:19:21 -04:00
3b2bc4ac5b
gallium/xorg st/nv50: add PIPE_BIND_CURSOR
...
We need to distinguish surfaces for mouse cursors from scanouts,
because nv50 hardware display engine ignores tiling flags.
2011-05-06 19:09:23 +02:00
8c3226be90
Add pci id for FirePro 2270
...
Signed-off-by: Kostas Georgiou <georgiou@opengamma.com >
2011-05-06 13:03:40 -04:00
2c3232da7a
egl_dri2: add new radeon pci ids
...
There seriously needs to be a better way to do
this.
Signed-off-by: Alex Deucher <alexdeucher@gmail.com >
2011-05-06 13:02:24 -04:00
José Fonseca