Ben Skeggs
7b6fb34e9d
nouveau: use usage, not uninitialised flags value...
2009-01-12 16:59:33 +10:00
Brian Paul
516dd9b361
cell: optimize unpack_colors() function, saving 12 cycles
2009-01-11 15:28:38 -07:00
Brian Paul
6324c77e01
cell: move color unpacking code into separate function
2009-01-11 15:18:28 -07:00
Brian Paul
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
Brian Paul
c4a782041b
cell: datatype clean-ups in SPE rtasm
2009-01-11 14:22:00 -07:00
Brian Paul
c73b8c4131
cell: simplify the 'optional register' code
2009-01-11 14:06:39 -07:00
Brian Paul
91fac69537
cell: asst datatype clean-ups
2009-01-11 13:52:58 -07:00
Brian Paul
097da27f55
cell: move depth/stencil code into separate function
2009-01-11 13:40:28 -07:00
Brian Paul
33ba88a0df
cell: clean-up, re-indent, comments
2009-01-11 13:23:44 -07:00
Brian Paul
83a525af95
cell: use tgsi_dump_instruction() instead of spe_comment()
2009-01-10 14:58:45 -07:00
Brian Paul
2acf07983f
gallium: use tgsi_dump_instruction() instead of ppc_comment()
2009-01-10 14:58:45 -07:00
Brian Paul
d4394bb768
gallium: remove unused struct type
2009-01-10 14:58:45 -07:00
Brian Paul
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
Brian Paul
1922ea965a
gallium: emit comments in TGSI->PPC codegen
2009-01-10 14:58:44 -07:00
Brian Paul
2ebd969f0f
gallium: code to dump/debug PPC code (disabled)
2009-01-10 14:58:44 -07:00
Brian Paul
7acaeb8775
gallium: added comment/annotation support to PPC rtasm
2009-01-10 14:58:44 -07:00
Brian Paul
fba6dac380
cell: added rule to produce .s assembly files
2009-01-10 14:58:44 -07:00
Brian Paul
2b26a92cd3
gallium: s/false/FALSE/
2009-01-10 14:58:44 -07:00
Younes Manton
6cee4b8d7e
mesa: Update .gitignore
2009-01-10 14:11:30 -05:00
Younes Manton
0b03cd4ea7
nouveau: Update nv30 swizzling.
2009-01-10 14:09:59 -05:00
Younes Manton
8ee238be75
nouveau: Factor out common winsys bits into libnouveaudrm.a
2009-01-10 13:52:07 -05:00
Younes Manton
734b3cb182
g3dvl: Use Gallium thread wrappers.
2009-01-10 13:52:07 -05:00
Younes Manton
1e9c3efcc7
g3dvl: Use Gallium MALLOC wrappers.
2009-01-10 13:52:07 -05:00
Younes Manton
db1021a37c
g3dvl: Get rid of old unbuffered motion compensation code.
2009-01-10 13:52:06 -05:00
Younes Manton
3c1b790c31
g3dvl: Map vertex bufs once per frame/flush.
2009-01-10 13:52:06 -05:00
Younes Manton
9beb004885
nouveau: Catch some more leaks.
2009-01-10 13:52:06 -05:00
Younes Manton
c10db52fcd
g3dvl: Fix some memory leaks.
2009-01-10 13:52:06 -05:00
Younes Manton
87e39466dc
g3dvl: Allocate one set of bufs, let winsys rename them as necessary.
2009-01-10 13:52:06 -05:00
Younes Manton
af2a856caa
nouveau: Add busy() query, determines if BOs can be mapped immediately.
2009-01-10 13:52:05 -05:00
Younes Manton
7289c388f4
nouveau: Use PIPE_BUFFER_USAGE_CPU_* instead of custom.
2009-01-10 13:52:05 -05:00
Younes Manton
0e1301ec8f
g3dvl: Expand YCbCr to full RGB range by default.
2009-01-10 13:52:05 -05:00
Younes Manton
9cbcdd86d0
g3dvl: Work around mplayer weirdness in XvMCPutSurface().
2009-01-10 13:52:05 -05:00
Younes Manton
ebd38dd0d6
g3dvl: Flag textures holding incoming data as dynamic.
2009-01-10 13:52:05 -05:00
Younes Manton
34c0281879
nouveau: Put dynamic textures in GART for CPU access and don't swizzle.
...
Also flag shadows as dynamic since they're for CPU access as well.
2009-01-10 13:52:05 -05:00
Younes Manton
fbf418d9dc
gallium: Define PIPE_TEXTURE_USAGE_DYNAMIC.
...
Knowing how the client intends to use the texture will give the
driver the opportunity to optimize for such cases.
2009-01-10 13:52:04 -05:00
Younes Manton
72aa42e594
nouveau: Swizzle textures larger than nv04 SIFM limit in parts.
...
Limit of SIFM on nv40 is 1024x1024, not sure about others.
2009-01-10 13:52:04 -05:00
José Fonseca
7844b4e730
draw: Add missing include.
2009-01-09 21:35:09 +00:00
Alan Hourihane
834c93687f
st: build fix
2009-01-09 11:23:47 +00:00
Brian Paul
ffe7bda603
mesa: 7.3-rc-1 Makefile changes
2009-01-09 11:16:44 +00:00
Brian Paul
4441e0a7b2
mesa: latest glxext.h header, no version change
2009-01-09 11:16:44 +00:00
Brian Paul
258c55db11
mesa: import glext.h version 44
2009-01-09 11:16:43 +00:00
Brian Paul
bd03d9bdbb
docs: dri2proto, libdrm tweaks
2009-01-09 11:16:42 +00:00
Brian Paul
233a58ff56
glsl: fix typo in the vec2 += operator function
2009-01-09 11:16:42 +00:00
Brian Paul
f75910e9b7
mesa: set version string to 7.3-rc1
2009-01-09 11:16:42 +00:00
Brian Paul
da37940973
docs: more 7.3 doc updates
2009-01-09 11:16:41 +00:00
Brian Paul
178f1ff486
docs: import 7.2 relnotes, start on 7.3 relnotes
2009-01-09 11:16:41 +00:00
Brian Paul
ebccb250f7
glsl: fix broken +=, -=, *=, /= operators
...
These functions need to return the final computed value.
Now expressions such as a = (b += c) work properly.
Also, no need to use __asm intrinsics in these functions. The resulting
code is the same when using ordinary arithmetic operators and is more legible.
2009-01-09 11:16:40 +00:00
Brian Paul
5273a5f0d7
mesa: fix off-by-one bug in _mesa_delete_instructions()
2009-01-09 11:16:40 +00:00
Eric Anholt
28c503d4bb
i965: Fix GLSL FS DPH to return the right value instead of src0.w * src1.w.
2009-01-09 11:16:40 +00:00
Eric Anholt
739d38c8e9
i965: Remove worrisome comment about _NEW_PROGRAM signaling fp change.
...
Everything now depends on either BRW_NEW_FRAGMENT_PROGRAM or
BRW_NEW_VERTEX_PROGRAM.
2009-01-09 11:16:39 +00:00