Brian Paul
61b7da074e
llvmpipe: make min/max lod and lod bias dynamic state
...
Before, changing any of these sampler values triggered generation
of new JIT code. Added a new flag for the special case of
min_lod == max_lod which is hit during auto mipmap generation.
2010-09-24 09:47:37 -06:00
Jerome Glisse
7967b46e65
r600g: fix compilation after change to evergreend.h
...
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24 10:43:57 -04:00
Jerome Glisse
eff1af65af
r600g: evergreen fix for new design
...
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24 10:41:01 -04:00
Jerome Glisse
cb3aed80db
r600g: move use_mem_constants flags for new designs structure alignment
...
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24 10:41:01 -04:00
Jerome Glisse
3672bc14af
r600g: fix typo in evergreen define (resource are in [0x30000;0x34000] range)
...
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24 10:41:01 -04:00
Brian Paul
f5c810c42f
st/mesa: use the wrapped renderbuffer in CopyPixels()
...
Fixes assertion failures when copying stencil pixels.
NOTE: this is a candidate for the 7.9 branch.
2010-09-24 08:27:06 -06:00
Brian Paul
10dcc989ab
st/mesa: add missing MESA_FORMAT_S8 case in st_mesa_format_to_pipe_format()
...
NOTE: this is a candidate for the 7.9 branch.
2010-09-24 08:24:43 -06:00
Brian Paul
9f7c8053e0
mesa: fix assertions to handle srgb formats
...
http://bugs.freedesktop.org/show_bug.cgi?id=30333
NOTE: This is a candidate for the 7.9 branch.
2010-09-24 07:55:49 -06:00
Luca Barbieri
1154765429
d3d1x: CRLF -> LF in progs
2010-09-24 15:12:20 +02:00
Luca Barbieri
7e81c67c8b
d3d1x: stop using GLX in demos, just use the default visual
2010-09-24 15:12:19 +02:00
Luca Barbieri
db1fbb1efc
d3d1x: assert if X visual is not among enumerated visuals
2010-09-24 15:12:19 +02:00
Luca Barbieri
f1063cfee2
d3d1x: don't crash on drivers not supporting vertex or geometry sampling
2010-09-24 15:12:19 +02:00
Luca Barbieri
b632d9fce3
nvfx: add RGB framebuffer format support in addition to BGR
2010-09-24 15:12:19 +02:00
Luca Barbieri
d0ee833dee
nvfx: allow setting NULL constant buffers
2010-09-24 15:12:19 +02:00
Andre Maasikas
8b63ed4e6c
r600g: break alu clause earlier
...
we still have constants to add and next int may need also 6 slots
2010-09-24 13:26:19 +03:00
Luca Barbieri
c7a064b4d5
d3d1x: fix linking of dxbc2tgsi
2010-09-24 09:51:15 +02:00
Luca Barbieri
54ee7721a1
d3d1x: draw to the correct buffer
2010-09-24 09:15:49 +02:00
Luca Barbieri
0f4ec3f72c
d3d1x: fix CheckMultisampleQualityLevels
2010-09-24 09:15:49 +02:00
Luca Barbieri
0e40b41cee
d3d1x: don't assert on unsupported resource types
2010-09-24 09:15:49 +02:00
Luca Barbieri
4babdc7844
d3d1x: add untested support for geometry shader translation
2010-09-24 09:15:49 +02:00
Luca Barbieri
f71f8c7d18
d3d1x: add shader dumping
2010-09-24 09:15:49 +02:00
Dave Airlie
11cd1612a1
r600g: fix polygon mode
...
this fixes glean'pointSprite test.
2010-09-24 18:58:16 +10:00
Dave Airlie
efa111a6cb
r600g: fixup sprite coord enable.
...
this fixes piglit glsl-fs-pointcoord
2010-09-24 16:36:54 +10:00
Dave Airlie
428b101af9
r600g: fix typo in r700 alu emit
2010-09-24 16:12:02 +10:00
Dave Airlie
59276b8541
r600g: fixup VP->FP output->input routing.
...
We need to map the TGSI semantics to each other using the hw semantic ids.
this fixes glsl-kwin-blur and glsl-routing.
2010-09-24 14:59:19 +10:00
Dave Airlie
e74d26d82a
r600g: fixup tex wrapping.
...
the clamp edge/clamp cases were reversed.
2010-09-24 13:51:54 +10:00
Dave Airlie
4e27e935ca
r600g: drop index_offset parameter to index buffer translate.
...
r600 doesn't need this as we always have working index bias
2010-09-24 12:38:14 +10:00
Dave Airlie
cf0162be13
r600g: fix draw-elements and draw-elements-base-vertex
2010-09-24 12:34:43 +10:00
Dave Airlie
95e04c3d74
r600g: some more vertex formats
2010-09-24 12:34:43 +10:00
Dave Airlie
b7ab9ee84e
r600g: add some more vertex format support.
...
adds the sscaled formats, this passes some more of the draw-vertices tests.
2010-09-24 12:34:43 +10:00
Dave Airlie
4388087f19
r600g: add vert support for 16/16 and 16/16/16 floats.
...
makes draw-vertices-half-float pass
2010-09-24 12:34:43 +10:00
Marek Olšák
85a45dcd5d
Build r300g by default
...
NOTE: This will go to 7.9 as well.
2010-09-24 02:58:50 +02:00
Marek Olšák
9f35dcd24c
r300g: fix the border color for every format other than PIPE_FORMAT_B8G8R8A8
...
TX_BORDER_COLOR should be formatted according to the texture format.
Also the interaction with ARB_texture_swizzle should be fixed too.
NOTE: This is a candidate for the 7.9 branch.
2010-09-24 02:57:36 +02:00
Marek Olšák
7d28ec8500
r300g: fix a copy-paste typo for logging
2010-09-24 02:33:34 +02:00
Marek Olšák
a333485386
r300g: make accessing map_list and buffer_handles thread-safe
...
NOTE: This is a candidate for the 7.9 branch.
2010-09-24 02:29:05 +02:00
Marek Olšák
206d92912c
r300g: fixup long-lived BO maps being incorrectly unmapped when flushing
...
Based on commit 3ddc714b20 by Dave Airlie.
NOTE: This is a candidate for the 7.9 branch.
2010-09-24 02:29:04 +02:00
Marek Olšák
68afbe89c7
util: make calling remove_from_list multiple times in a row safe
...
This commit fixes an infinite loop in foreach_s if remove_from_list is used
more than once on the same item with other list operations in between.
NOTE: This is a candidate for the 7.9 branch because the commit
"r300g: fixup long-lived BO maps being incorrectly unmapped when flushing"
depends on it.
2010-09-24 02:29:04 +02:00
Eric Anholt
f46523e0bc
i915: Remove a dead if (0) block.
2010-09-23 16:34:10 -07:00
Eric Anholt
64ff468d6f
intel: Remove dead intelIsTextureResident().
...
It always returned 1 (GL_TRUE), which is the same thing that happens when
the driver hook isn't present.
2010-09-23 16:30:58 -07:00
Eric Anholt
f9e6f401e1
unichrome: Mostly revert my convolution removal changes.
...
For this driver, the minimum pitch alignment stuff does appear to be
necessary, so leave the separate munged width/height variable in
place.
2010-09-23 16:20:33 -07:00
Eric Anholt
1c0646a826
radeon: Remove copied minimum pitch alignment code.
...
This is already covered by radeon_mipmap_tree.c, and my convolution
cleanups broke in the presence of this code. Thanks to Marek Olšák
for tracking down the relevant miptree code for me.
2010-09-23 16:20:25 -07:00
Eric Anholt
fae1855946
intel: Replace my intel_texture_bitmap code with _mesa_meta_Bitmap.
...
The meta code is more general than mine, and appears to pass the same
sets of tests (piglit + some oglconform).
2010-09-23 16:04:55 -07:00
Eric Anholt
2337f364b1
intel: Remove unnecessary minimum pitch alignment to 32 bytes.
...
This broke with the cleanup I did in convolution removal. It's
unnecessary anyway since region_alloc_tiled adjusts pitches for us (64
byte alignment)
2010-09-23 16:04:55 -07:00
Tom Stellard
92762842a0
r300g: Always try to build libr300compiler.a
...
Make libr300compiler.a a PHONY target so that this library will always be
built. This fixes the problem of libr300compiler.a not being updated
when r300g is being built and r300c is not.
This is a candidate for the Mesa 7.9 branch.
2010-09-23 15:04:35 -07:00
Eric Anholt
d26211e499
intel: Remove disabled stencil drawpixels acceleration.
...
We still retain the fallback override for GL_STENCIL_INDEX, because
the metaops version fails at oglconform.
2010-09-23 14:58:37 -07:00
Dave Airlie
c0c0c4b96b
r300g: fix point sprite coord.
...
handled elsewhere now.
thanks to Droste on irc for pointing out the fix
2010-09-24 07:46:59 +10:00
Jerome Glisse
b360c050b6
r600g: initial evergreen support in new path
...
This doesn't work yet.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-23 17:10:28 -04:00
Tilman Sauerbeck
ce8c71817b
r600g: Destroy the blitter.
...
Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
2010-09-23 22:36:00 +02:00
Eric Anholt
a62efdf82c
mesa: Remove EXT_convolution.
...
More optional code.
2010-09-23 13:25:45 -07:00
Eric Anholt
73578ba9c4
mesa: Remove SGI_color_matrix.
...
Another optional ARB_imaging subset extension.
2010-09-23 13:25:45 -07:00