Ben Skeggs [Mon, 5 Jan 2009 03:06:43 +0000 (14:06 +1100)]
nv50: add TXP to TEX case.. not correct, but anyway..
Ben Skeggs [Mon, 5 Jan 2009 02:41:12 +0000 (13:41 +1100)]
nv50: indentation for TEX is a little overenthusiastic
Ben Skeggs [Mon, 5 Jan 2009 02:38:47 +0000 (13:38 +1100)]
nv50: ensure we actually get contiguous regs for TEX insn.
Still many more horrible things to fix here...
Jonathan Adamczewski [Mon, 5 Jan 2009 14:52:17 +0000 (07:52 -0700)]
cell: SIMDize sorting in setup_sort_vertices()
Put setup.v{min,mid,max,provoke} into a union with qword vertex_headers.
Rewrite vertex sorting to more efficiently handle the packed data items.
Reduces spu_tri.o by ~128 bytes.
Jonathan Adamczewski [Mon, 5 Jan 2009 14:49:48 +0000 (07:49 -0700)]
cell: SIMDize some subtractions
Put edge.{dx,dy} into a union with a vector and perform subtractions in
setup_sort_vertices() on vectors.
Reduces spu_tri.o by ~300 bytes.
Jakob Bornecrantz [Mon, 5 Jan 2009 10:50:57 +0000 (11:50 +0100)]
progs/glsl: Add ignores
Jakob Bornecrantz [Mon, 5 Jan 2009 10:49:45 +0000 (11:49 +0100)]
progs/glsl: Remove double target def for vert-or-frag-only
Jonathan Adamczewski [Mon, 5 Jan 2009 01:48:46 +0000 (18:48 -0700)]
cell: improvements to spu_tri.c
Replace int setup.span{left,right}[2] with vec_uint4 setup.span.quad
SIMDize calculate_mask() and inline into into flush_spans()
Set setup.span.quad members using spu_shuffle() or spu_sel().
Reduces spu_tri.o by ~116 bytes.
Jonathan Adamczewski [Mon, 5 Jan 2009 01:46:44 +0000 (18:46 -0700)]
cell: new spu_shuffle.h header
Facilitates creation of shuffle patterns for use with spu_shuffle()
and si_shufb() intrinsics.
To be used by subsequent patches.
Brian Paul [Mon, 5 Jan 2009 01:40:48 +0000 (18:40 -0700)]
cell: initial codegen support for fragment shader loops
Basic for/while loops work now.
Only one level of loop nesting is supported at this time (same for if/else).
The progs/glsl/mandelbrot demo works, but the colors are too dim.
Brian Paul [Mon, 5 Jan 2009 01:38:42 +0000 (18:38 -0700)]
cell: add support for PIPE_CAP_TEXTURE_MIRROR_REPEAT query
We don't really support this texwrap mode yet, but this enables GL 2.1
Brian Paul [Sun, 4 Jan 2009 17:14:19 +0000 (10:14 -0700)]
cell: clean-up, improve SPU code generation
Start on ARL and address-relative indexing too.
Brian Paul [Sun, 4 Jan 2009 17:04:44 +0000 (10:04 -0700)]
gallium: fix MaxTextureCoordUnits init
Fixes Cell regression.
Brian [Sat, 3 Jan 2009 21:44:34 +0000 (14:44 -0700)]
cell: fix typo (s/10/0)
Brian [Tue, 23 Dec 2008 21:46:25 +0000 (15:46 -0600)]
demos: remove redundant Makefile lines from prev merge
Brian Paul [Fri, 2 Jan 2009 23:55:33 +0000 (16:55 -0700)]
softpipe: increase number of texture samplers/units to PIPE_MAX_SAMPLERS (16)
Brian Paul [Fri, 2 Jan 2009 23:54:11 +0000 (16:54 -0700)]
gallium: assorted state tracker fixes for > 8 texture samplers
Brian Paul [Fri, 2 Jan 2009 23:52:00 +0000 (16:52 -0700)]
mesa: replace CC with APP_CC in progs/glsl/Makefile
Brian Paul [Fri, 2 Jan 2009 23:32:26 +0000 (16:32 -0700)]
Merge commit 'origin/master' into gallium-0.2
Conflicts:
src/mesa/main/ffvertex_prog.c
src/mesa/main/texenvprogram.c
Brian Paul [Fri, 2 Jan 2009 23:16:16 +0000 (16:16 -0700)]
gallium: fix texcoord loop for rasterpos attributes
Brian Paul [Fri, 2 Jan 2009 23:15:43 +0000 (16:15 -0700)]
gallium: clamp MaxVertexTextureImageUnits against Mesa limit
Brian Paul [Fri, 2 Jan 2009 23:15:16 +0000 (16:15 -0700)]
mesa: remove redudant lines for libglapi.a
Brian Paul [Fri, 2 Jan 2009 19:26:15 +0000 (12:26 -0700)]
mesa: fix another "out of samplers" problem
Now only the samplers that are actually used by texture() functions are
saved in the uniform variable list. Before, we could run out of samplers
if too many were declared while only some of them were actually used.
Brian Paul [Thu, 1 Jan 2009 21:04:57 +0000 (14:04 -0700)]
i965: increase number of texture samplers to 16
This lets GLSL shaders use up to 16 samplers.
Fixed function is still limited to 8 textures.
Tested with progs/glsl/samplers.c
Brian Paul [Thu, 1 Jan 2009 19:03:07 +0000 (12:03 -0700)]
i965: comments, clean-ups, re-order some functions
Brian Paul [Thu, 1 Jan 2009 00:16:44 +0000 (17:16 -0700)]
i965: added OPCODE_NRM3/4
Brian Paul [Thu, 1 Jan 2009 00:16:27 +0000 (17:16 -0700)]
i965: fix comment
Brian Paul [Thu, 1 Jan 2009 00:01:44 +0000 (17:01 -0700)]
i965: indentation and formatting fixes
Brian Paul [Wed, 31 Dec 2008 23:49:58 +0000 (16:49 -0700)]
i965: implement OPCODE_NRM3/NRM4
Brian Paul [Wed, 31 Dec 2008 23:43:57 +0000 (16:43 -0700)]
i965: whitespace, comment changes
Brian Paul [Wed, 31 Dec 2008 23:38:10 +0000 (16:38 -0700)]
mesa: comments, whitespace changes
Brian Paul [Thu, 1 Jan 2009 21:02:17 +0000 (14:02 -0700)]
demos: added progs/glsl/samplers.c to test all available texture samplers
Brian Paul [Thu, 1 Jan 2009 15:07:54 +0000 (08:07 -0700)]
mesa: updated comments about GLSL constants
Brian Paul [Wed, 31 Dec 2008 18:54:02 +0000 (11:54 -0700)]
mesa: increase max texture image units and GLSL samplers to 16
The max texture coord units is still 8. All the fixed-function paths are
still limited to 8 too. But GLSL shaders can use more samplers now.
Note that some texcoord-related data structures are declared to be 16
elements in size rather than 8. This just simplifies the code in a few
places; the extra elements aren't accessible to the user.
These changes haven't been extensively tested yet, but sanity checking has
been done.
It should be possible to increase the max image units/samplers to 32 without
doing anything special. Beyond that we'll need longer bitfields in a few
places.
José Fonseca [Wed, 31 Dec 2008 15:03:35 +0000 (15:03 +0000)]
python: Pass a zero offset to util_draw_vertex_buffer.
José Fonseca [Wed, 31 Dec 2008 13:29:09 +0000 (13:29 +0000)]
util: List new file in SConscript.
José Fonseca [Wed, 31 Dec 2008 13:05:39 +0000 (13:05 +0000)]
Merge commit 'origin/gallium-0.1' into gallium-0.2
José Fonseca [Tue, 30 Dec 2008 20:12:28 +0000 (20:12 +0000)]
i915: Avoid ptr->int conversion.
José Fonseca [Tue, 30 Dec 2008 20:10:24 +0000 (20:10 +0000)]
draw: Avoid integer overflow converting pointers on 64bit archs.
Not really an error, as we only care for the lower 4 bits.
Eric Anholt [Wed, 31 Dec 2008 08:29:49 +0000 (00:29 -0800)]
intel: Share passthrough transform setup between glBitmap and glDrawPixels.
The DrawPixels path was missing glViewport care, so blender's toolbar icons
would go to the wrong places.
Bug #19118.
Eric Anholt [Wed, 31 Dec 2008 08:02:43 +0000 (00:02 -0800)]
intel: Add support for glBitmap as metaops using GL calls.
This lets us avoid software fallbacks when clients forget to turn some state
off (engine demo) or just do crazy things to test conformance (OGLC).
This should probably be brought into mesa generic code so other drivers can
make use of it.
Bug #19016.
Brian Paul [Wed, 31 Dec 2008 01:01:15 +0000 (18:01 -0700)]
Merge commit 'origin/master' into gallium-0.2
Conflicts:
src/mesa/main/config.h
Brian Paul [Wed, 31 Dec 2008 00:50:39 +0000 (17:50 -0700)]
mesa: fix bug in evaluation of structure fields
Fixes incorrect size information. See bug 19273.
Brian Paul [Wed, 31 Dec 2008 00:11:32 +0000 (17:11 -0700)]
mesa: allow variable indexing into the predefined uniform variable arrays
This allows code such as "vec4 a = gl_LightSource[i].ambient;" to work.
When a built-in uniform array is indexed with a variable index we need to
"unroll" the whole array into the parameter list (aka constant buffer) because
we don't know which elements may be accessed at compile-time. In the case of
the gl_LightSource array of size [8], we emit 64 state references into the
parameter array (8 elements times 8 vec4s per gl_LightSourceParameters
struct).
Previously, we only allowed constant-indexed references to uniform arrays
(such as gl_LightSource[2].position) which resulted in a single state reference
being added to the parameter array, not 64. We still optimize this case.
Users should be aware that using "gl_LightSource[i].ambient" in their shaders
is a bit expensive since state validation will involve updating all 64
light source entries in the parameter list.
Brian Paul [Wed, 31 Dec 2008 00:03:09 +0000 (17:03 -0700)]
mesa: better error message when running out of GLSL samplers
Brian Paul [Wed, 31 Dec 2008 00:00:36 +0000 (17:00 -0700)]
mesa: comments for some state vars
Brian Paul [Wed, 31 Dec 2008 00:00:06 +0000 (17:00 -0700)]
mesa: increase max constants/uniforms to 256 (vec4 vectors)
José Fonseca [Tue, 30 Dec 2008 19:21:30 +0000 (19:21 +0000)]
gallium: Initialize var before use.
José Fonseca [Tue, 30 Dec 2008 17:22:24 +0000 (17:22 +0000)]
glut: List new source file in sconscript.
José Fonseca [Tue, 30 Dec 2008 17:21:25 +0000 (17:21 +0000)]
mesa: Do not specify types in bitfields.
As advised by gcc -pedantic.
José Fonseca [Tue, 30 Dec 2008 17:21:15 +0000 (17:21 +0000)]
draw: Do not specify types in bitfields.
As advised by gcc -pedantic.
José Fonseca [Tue, 30 Dec 2008 17:15:34 +0000 (17:15 +0000)]
gallium: Remove unused variables.
José Fonseca [Tue, 30 Dec 2008 17:14:48 +0000 (17:14 +0000)]
gallium: Don't redefine INLINE.
INLINE is commonly used in third-party headers.
José Fonseca [Tue, 30 Dec 2008 17:13:38 +0000 (17:13 +0000)]
scons: Specify C99 throughout all the tree.
MSVC may not support full C99, but supports more than plain C90. And
-pedantic without -std=c99 generates too many spurious warnings
(specially C++ style comments) to be of any use.
Note that using certain C99 features in the cross-platform parts of Gallium
is still not possible; namely mid-of-scope variable declarations and named
structure initializers will break MSVC builds.
José Fonseca [Tue, 30 Dec 2008 17:06:51 +0000 (17:06 +0000)]
rtasm: Remove spurious semi-colons after function bodies.
José Fonseca [Tue, 30 Dec 2008 16:35:46 +0000 (16:35 +0000)]
gdi: Remove accidental keypresses.
Brian Paul [Tue, 30 Dec 2008 14:57:16 +0000 (07:57 -0700)]
mesa: updated compilation documentation
Brian Paul [Thu, 18 Dec 2008 23:08:42 +0000 (16:08 -0700)]
demos: minor fixes to twoside.c glsl demo
Xiang, Haihao [Tue, 30 Dec 2008 03:25:45 +0000 (11:25 +0800)]
intel: disable ATI_texture_env_combine3 for i830( and related device).
Thanks to Eric for pointing it out.
Eric Anholt [Wed, 24 Dec 2008 00:08:40 +0000 (16:08 -0800)]
dri: Fix driWaitForMSC32 when divisor >= 2 and msc < 0.
We'd come up with a negative remainder, while we were looking for the positive
version of it in the loop conditional. And, since the "did we hit our target"
break was disabled for the target_msc == 0 ("Just make the divisor/remainder
work") path, we'd never exit.
Simplify the code by just using int64_t all over instead of trying to do it
in a u32 space.
Alex Deucher [Mon, 29 Dec 2008 17:28:12 +0000 (12:28 -0500)]
R300: missing semicolon
Xiang, Haihao [Mon, 29 Dec 2008 01:30:41 +0000 (09:30 +0800)]
intel: enable ATI_texture_env_combine3. Fixes #17707
Dave Airlie [Sun, 28 Dec 2008 06:48:36 +0000 (16:48 +1000)]
r300: remove the unknowns from the indx_buffer code
Xiang, Haihao [Wed, 24 Dec 2008 01:26:46 +0000 (09:26 +0800)]
i915: separate the fog term from the specular color term.
Previously fog parameter and specular color are packed into the
same dword. Note specular color should be packed in BGRA for device,
so if fog parameter and specular color all are present, fog parameter
will dirty the alpha term of specular color. This fixes rendering
issue when playing 'Yo Frankie' on 915/945.
Eric Anholt [Tue, 23 Dec 2008 22:45:37 +0000 (14:45 -0800)]
intel: Fix glBitmap clipping for DRI1.
Dave Airlie [Sat, 20 Dec 2008 01:38:23 +0000 (17:38 -0800)]
Remove third buffer support from Mesa.
This is part of the deprecated pageflipping infrastructure.
Alan Hourihane [Tue, 23 Dec 2008 15:33:46 +0000 (15:33 +0000)]
egl: fix egl closure
Alan Hourihane [Tue, 23 Dec 2008 14:18:36 +0000 (14:18 +0000)]
egl: fix startup query version
Alan Hourihane [Tue, 23 Dec 2008 13:20:18 +0000 (13:20 +0000)]
egl: support GLXFBConfigs, pbuffers and pixmaps.
José Fonseca [Mon, 22 Dec 2008 20:23:59 +0000 (20:23 +0000)]
softpipe: Don't fill surfaces's winsys fields.
This is sometimes checked to distinguish between texture views and
(deprecated) standalone surfaces.
José Fonseca [Mon, 22 Dec 2008 20:20:58 +0000 (20:20 +0000)]
softpipe: Call surface_alloc_storage to get the pipebuffer for display targets.
Otherwise blitting from display target surfaces to front screen fails in
several platforms.
José Fonseca [Mon, 22 Dec 2008 19:31:52 +0000 (19:31 +0000)]
Merge commit 'origin/gallium-0.1' into gallium-0.2
Conflicts:
src/gallium/auxiliary/util/Makefile
José Fonseca [Thu, 11 Dec 2008 05:42:25 +0000 (14:42 +0900)]
gdi: Cleanup sconsfile.
Jerome Glisse [Thu, 18 Dec 2008 18:56:45 +0000 (19:56 +0100)]
softpipe: initialize refcount and winsys
Jerome Glisse [Thu, 18 Dec 2008 12:34:27 +0000 (13:34 +0100)]
softpipe: convert to use texture instead of surface
José Fonseca [Mon, 22 Dec 2008 16:55:27 +0000 (16:55 +0000)]
gallium: const correctness.
José Fonseca [Sun, 21 Dec 2008 13:48:43 +0000 (13:48 +0000)]
Ignore new tests executables.
José Fonseca [Sat, 20 Dec 2008 12:59:51 +0000 (12:59 +0000)]
gallium: Fix typo in define name.
Brian Paul [Fri, 19 Dec 2008 23:56:56 +0000 (16:56 -0700)]
Merge commit 'origin/master' into gallium-0.2
Brian Paul [Fri, 19 Dec 2008 23:37:06 +0000 (16:37 -0700)]
gallium: begin adapting Ian's 3D mipmap gen code to gallium utility lib
Unfinished, a big no-op for now.
Brian Paul [Fri, 19 Dec 2008 23:14:48 +0000 (16:14 -0700)]
gallium: Fix typeo in mipmap filter for GL_UNSIGNED_SHORT_1_5_5_5_REV
This is copied from Ian's commit
a330933bb75c38148668637cd22b90d75d39506f
Ian Romanick [Fri, 19 Dec 2008 21:03:51 +0000 (13:03 -0800)]
Add do_row_3d for mipmapping 3D textures
Previously 3D textures were mipmapped using multiple passed through
the 2D mipmap generation code. This had 3 disadvantages. First, the
extra passes were slow. Second, this required the allocation of a
temporary buffer to hold intermediate data. Third, and most
important, the extra passes caused loss of additional bits due to
integer division / bit-shifting.
With this change, our mipmapgen conformance test passes for
non-compressed texture formats.
Ian Romanick [Fri, 19 Dec 2008 20:58:01 +0000 (12:58 -0800)]
Fix typeo in mipmap filter for GL_UNSIGNED_SHORT_1_5_5_5_REV
Ian Romanick [Thu, 18 Dec 2008 22:13:26 +0000 (14:13 -0800)]
965 / GLSL: Use full precision for EXP instruction
The partial precision mode doesn't have quite enough bits of precision
to pass conformance tests.
Ian Romanick [Thu, 18 Dec 2008 22:11:06 +0000 (14:11 -0800)]
GLSL: The LOG2 macro doesn't have enough precision
It looks like the LOG2 macro only has 8 or 9 bits of precission, but
the ARB_vertex_program spec says "accurate to at least 10 bits".
Eric Anholt [Fri, 19 Dec 2008 00:00:09 +0000 (16:00 -0800)]
intel: Fix glBitmap top/bottom clipping.
Bug #19139.
Eric Anholt [Thu, 18 Dec 2008 06:10:57 +0000 (22:10 -0800)]
intel: Don't forget the source bitmap size when clipping the size we draw.
Eric Anholt [Thu, 18 Dec 2008 05:18:20 +0000 (21:18 -0800)]
intel: Update mesa state in blit operations that want post-scissor draw bounds.
Eric Anholt [Thu, 18 Dec 2008 05:18:00 +0000 (21:18 -0800)]
intel: don't clip to scissor-clipped read framebuffer bounds in copypixels.
José Fonseca [Fri, 19 Dec 2008 20:06:11 +0000 (20:06 +0000)]
gallium: Simple and efficient cache.
Fixed size hash table. Collisions are handled by simply destroying the
previous entry.
It hasn't received much testing yet.
Brian Paul [Fri, 19 Dec 2008 15:01:03 +0000 (08:01 -0700)]
cell: fix build breakage
Alan Hourihane [Fri, 19 Dec 2008 14:59:43 +0000 (14:59 +0000)]
gallium: set ST_NEW_FRAMEBUFFER when setting new surfaces
Brian Paul [Fri, 19 Dec 2008 14:57:09 +0000 (07:57 -0700)]
mesa: s/CC/APP_CC/ in progs/glsl/Makefile
Brian Paul [Fri, 19 Dec 2008 14:33:17 +0000 (07:33 -0700)]
gallium: replace #elif with #else
Brian Paul [Fri, 19 Dec 2008 14:32:52 +0000 (07:32 -0700)]
gallium: replace #elif with #else
Eric Anholt [Fri, 19 Dec 2008 02:42:06 +0000 (18:42 -0800)]
intel: Move copyteximage source clipping out of copytexsubimage.
glCopyTexSubImage already gets the (correct) clipping for us, so it doesn't
need the path. While moving the clipping out, replace the code with the mesa
path to do the same job.
Eric Anholt [Fri, 19 Dec 2008 02:31:25 +0000 (18:31 -0800)]
mesa: Clip copytexsubimage to read framebuffer bounds, not scissor region.
Eric Anholt [Fri, 19 Dec 2008 02:23:51 +0000 (18:23 -0800)]
mesa: Correct _mesa_clip_to_region() off-by-one.
Note how if:
x + width == xmax + 0: width -= 0
x + width == xmax + 1: width -= 0
x + width == xmax + 2: width -= 1
So, the function was clipping to [xmin, xmax+1), not [xmin, xmax) like it was
supposed to. Same for ymax.
Brian Paul [Fri, 19 Dec 2008 01:06:38 +0000 (18:06 -0700)]
tgsi: scan for additional info: uses_fogcoord, uses_frontfacing
Brian Paul [Fri, 19 Dec 2008 01:06:38 +0000 (18:06 -0700)]
tgsi: scan for additional info: uses_fogcoord, uses_frontfacing