3ef419f4a5
egl: fix makecurrent with null drawable/context
2009-01-14 11:58:06 +00:00
d96c89e579
gallium: Disable memory debugging for Windows OGL.
...
Unfortunately both Mesa and Gallium use the same defines for memory
allocation (MALLOC, FREE, etc), and worse, some times memory is allocated
with one set and freed with the other set, causing the homegrown memory
debugger to trip on itself.
In the future mesa and gallium should use different names, but for now,
memory debugging on Windows will have to be carried with different tools..
2009-01-14 11:40:32 +00:00
eb1f01a9d9
progs: List tri-clear.
2009-01-14 11:40:31 +00:00
83155aa11f
scons: Use .a suffix for MinGW.
...
This allows to build MinGW and MSVC within the same dir.
2009-01-14 11:40:31 +00:00
529f86fb11
intel: Add a none working GEM backend for intel
2009-01-14 12:28:31 +01:00
f6d09531ff
cell: Specify constant as float for CEILF().
...
Without the f, the constant is treated as a double, resulting in
slower arithmetic and libgcc conversion calls each time CEILF()
is used.
2009-01-14 12:49:28 +11:00
67c7f94a21
glsl: fix a comment typo
2009-01-13 23:59:18 +00:00
8373347c05
glsl: support sampler arrays.
2009-01-13 23:54:46 +00:00
f883c14560
nv50: fix progs/tests/manytex
...
Previously all squares were textured with the same texture.. not quite what
the demo was supposed to look like!
2009-01-13 13:57:22 +10:00
adee4b9021
nv50: get glxgears showing all 3 gears instead of 1!!
...
This fixes a lot of other things where not all the geometry got drawn
also.
2009-01-13 13:57:22 +10:00
8337c78d91
nv50: change some magic reg, makes more things work
...
No real idea what this does.. but a lot of things that misrendered and
made the GPU throw a DATA_ERROR now work.. I'm wondering what side-effects
we'll see from this :)
2009-01-13 13:57:22 +10:00
e8b0088692
nv50: add DXTn formats
2009-01-13 13:57:22 +10:00
68bb26b62d
nv50: shadow mapping
2009-01-13 13:57:22 +10:00
f7c2010525
nv50: aniso
2009-01-13 13:57:22 +10:00
918fc55e5f
nv50: occlusion queries
...
Not quite working, but the general idea is right I think.
2009-01-13 13:57:22 +10:00
d2442016af
nv50: implement KIL enough for progs/fp/kil to work
2009-01-13 13:57:22 +10:00
068107b5ad
cell: Add missing suffix to SHUFFLE macro
2009-01-13 14:02:18 +11:00
402e6752b5
cell: allocate batch buffers w/ 16-byte alignment
...
Replace cell_batch{align,alloc)*() with cell_batch_alloc16(), allocating
multiples of 16 bytes that are 16 byte aligned.
Opcodes are stored in preferred slot of SPU machine word.
Various structures are explicitly padded to 16 byte multiples.
Added STATIC_ASSERT().
2009-01-12 16:25:28 -07:00
45604ffac7
gallivm: Make sure the bitcode buffer is followed by a 0 byte.
...
May fail to parse otherwise.
2009-01-12 15:05:05 +01:00
f43e621e22
gallivm: Print error message from ParseBitcodeFile() in case it fails.
2009-01-12 12:39:31 +01:00
359bbe7432
gallivm: Explicitly specify the LLVM components we need.
2009-01-12 12:37:13 +01:00
f586c31fa6
gallivm: Adapt to header file move in LLVM 2.4.
2009-01-12 12:34:27 +01:00
7a90ace9c8
nv50: make rtt work again
2009-01-12 16:59:35 +10:00
ac6516101b
nv50: fix handling of depth textures
2009-01-12 16:59:35 +10:00
39bcc39717
nv50: another typo..
2009-01-12 16:59:35 +10:00
ecb2eb4c99
nouveau: fix warning
2009-01-12 16:59:34 +10:00
73f1857aee
nv50: fix assertion failure
2009-01-12 16:59:34 +10:00
f935f35287
nv50: remove previous hack to manage tiled surfaces
2009-01-12 16:59:34 +10:00
515c3d9bc1
nv50: fix a typo and a thinko
2009-01-12 16:59:34 +10:00
ed8f0b753b
nv50: enable npot textures
2009-01-12 16:59:34 +10:00
b01d0077af
nv50: disable shader debug
2009-01-12 16:59:34 +10:00
08b6534bc8
nv50: any cpu access to a texture is done on its backing images
...
Still a little dodgy:
- RTT will hit an assertion (hopefully!) and fail
- 3D textures with depth >= 32 will cause bad things to happen
2009-01-12 16:59:33 +10:00
103020f264
nv50: create buffers for each image that makes up a texture
2009-01-12 16:59:33 +10:00
df266471b1
nouveau: return buffer map to something sane.
...
Sorry, but no, we're not doing this.. Correctness always takes precedence
over speed. Implement this higher up where you know it's safe to do so,
and doesn't break other things in the process.
2009-01-12 16:59:33 +10:00
7b6fb34e9d
nouveau: use usage, not uninitialised flags value...
2009-01-12 16:59:33 +10:00
516dd9b361
cell: optimize unpack_colors() function, saving 12 cycles
2009-01-11 15:28:38 -07:00
6324c77e01
cell: move color unpacking code into separate function
2009-01-11 15:18:28 -07:00
b27eb7cb4f
cell: re-order the z/stencil fetch/extract/convert instructions for better perf
...
The new instruction order is 10 cycles faster.
2009-01-11 15:11:00 -07:00
c4a782041b
cell: datatype clean-ups in SPE rtasm
2009-01-11 14:22:00 -07:00
c73b8c4131
cell: simplify the 'optional register' code
2009-01-11 14:06:39 -07:00
91fac69537
cell: asst datatype clean-ups
2009-01-11 13:52:58 -07:00
097da27f55
cell: move depth/stencil code into separate function
2009-01-11 13:40:28 -07:00
33ba88a0df
cell: clean-up, re-indent, comments
2009-01-11 13:23:44 -07:00
83a525af95
cell: use tgsi_dump_instruction() instead of spe_comment()
2009-01-10 14:58:45 -07:00
2acf07983f
gallium: use tgsi_dump_instruction() instead of ppc_comment()
2009-01-10 14:58:45 -07:00
d4394bb768
gallium: remove unused struct type
2009-01-10 14:58:45 -07:00
0c71313970
gallium: fix register clobber bug in TGSI->PPC codegen
...
When negating a src vector that's stored in a altivec register, need to put
negated value into a new register so we don't upset the original value.
This solves the dark colors in the mandelbrot GLSL demo.
Also, use new predicate functions to check if a TGSI temp is stored in
an altivec register.
2009-01-10 14:58:44 -07:00
1922ea965a
gallium: emit comments in TGSI->PPC codegen
2009-01-10 14:58:44 -07:00
2ebd969f0f
gallium: code to dump/debug PPC code (disabled)
2009-01-10 14:58:44 -07:00
7acaeb8775
gallium: added comment/annotation support to PPC rtasm
2009-01-10 14:58:44 -07:00
Alan Hourihane