Commit graph

35332 commits

Author SHA1 Message Date
Ian Romanick
3832706f81 glsl2: Initialize ast_declarator_list::invariant in constructor 2010-07-01 17:10:11 -07:00
Ian Romanick
12873fa4e3 glsl2: Don't bounds check unsize array redeclarations
This along with several previous commits fix test CorrectUnsizedArray.frag.
2010-07-01 14:10:19 -07:00
Ian Romanick
127308b4be glsl2: Add gl_MaxTextureCoords 2010-07-01 13:30:50 -07:00
Ian Romanick
cd00d5b88c glsl2: Default delcaration of gl_TexCoord is unsized 2010-07-01 13:17:54 -07:00
Dan Nicholson
442c37e2ef Use GLW_CFLAGS when building libGLw
We check for libX11 and libXt, so we might as well use the CFLAGS
pkg-config tells us about.
2010-07-01 13:00:02 -07:00
Ian Romanick
5466b63968 glsl2: Change order of semaintic checks on variable declarations
This will make it easier to support more (valid) kinds of redeclarations.
2010-07-01 12:56:49 -07:00
Fernando Carrijo
d2fe97a209 mesa: Purge macros NEED_EVENTS and NEED_REPLIES
Signed-off-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
Signed-off-by: Brian Paul <brianp@vmware.com>
2010-07-01 13:07:07 -06:00
Brian Paul
1d298a3764 gallium/cso: unbind sampler views in cso_release_all() 2010-07-01 13:07:07 -06:00
Brian Paul
5b90f83aee mesa: free xform feedback hash table 2010-07-01 13:07:07 -06:00
Brian Paul
30f46e7b4c st/mesa: fix comment 2010-07-01 13:07:07 -06:00
Eric Anholt
411fb36b7c ir_to_mesa: Fill in remaining ops, remove default case for expression types.
We should now have support for all the expression types we need for
GLSL 1.20.
2010-07-01 11:24:38 -07:00
Eric Anholt
8761fcc2bf ir_to_mesa: Add support for ir_unop_rcp.
This isn't used at the moment, but will be soon.
2010-07-01 11:23:02 -07:00
Eric Anholt
d1b07167b9 glsl2: Update README for what I've been thinking about with expr types work. 2010-07-01 11:07:58 -07:00
Eric Anholt
8a1f186cc5 glsl2: Add a pass to convert mod(a, b) to b * fract(a/b).
This is used by the Mesa IR backend to implement mod, fixing glsl-fs-mod.
2010-07-01 11:07:23 -07:00
Eric Anholt
9acf618f24 glsl2: Remove dead member from dead code visitor. 2010-07-01 11:07:22 -07:00
Eric Anholt
d925c91730 glsl2: Add ir_unop_fract as an expression type.
Most backends will prefer seeing this to seeing (a - floor(a)), so
represent it explicitly.
2010-07-01 11:07:22 -07:00
Brian Paul
17c176eb73 llvmpipe: silence pointer type warnings 2010-07-01 09:15:31 -06:00
José Fonseca
8d93f360c5 gallivm: Support 4 x unorm8 in lp_build_fetch_rgba_aos().
Uses code and ideas from Brian Paul.
2010-07-01 15:02:17 +01:00
José Fonseca
b919bb7f61 gallivm: Allow to conversions to/from registers of different sizes.
Allow for example to convert from 4 x float32 to 4 x unorm8 and vice versa.

Uses code and ideas from Brian Paul.
2010-07-01 15:02:15 +01:00
José Fonseca
a70ec096aa gallivm: Support extended swizzles in lp_build_swizzle1_aos().
And rename to lp_build_swizzle_aos().
2010-07-01 15:02:12 +01:00
José Fonseca
e277d5c1f6 gallivm: Setup a global optimization pass.
Modules are still free to setup their own optimization passes, but for the
normal case it should not be necessary.
2010-07-01 15:02:10 +01:00
José Fonseca
83ced5a918 llvmpipe: Remove lp_build_swizzle2_aos().
Unnecessary special case.
2010-07-01 15:02:04 +01:00
Eric Anholt
5e4dd061d1 ir_to_mesa: Add support for discard instructions.
Fixes glsl-fs-discard-01.
2010-06-30 17:31:06 -07:00
Eric Anholt
f5b3b2a01a glsl2: Don't break sign() down by vector components. 2010-06-30 17:29:12 -07:00
Eric Anholt
3acd92a91f ir_to_mesa: Add support for ir_unop_sign.
Fixes glsl-fs-sign, glsl-vs-sign.
2010-06-30 17:29:12 -07:00
Eric Anholt
e558786a3e i965: Add support for OPCODE_SSG.
The old compiler didn't use SSG, and instead emitted SGT/SGT/SUB.  We
can do a little better for SSG than we do for the SGT series.
2010-06-30 17:29:09 -07:00
Eric Anholt
d6ebe9b16b ir_to_mesa: Add missing no-op type conversions.
Fixes glsl-fs-step.
2010-06-30 16:23:32 -07:00
Eric Anholt
02d615306e glsl2: Fix reversed value of step().
It's 0.0 if x < edge, not 1.0.  Partial fix for glsl-fs-step.
2010-06-30 16:23:15 -07:00
Eric Anholt
e64a4aaacb ir_to_mesa: Note which of our expr ops are unsupported 1.30 features. 2010-06-30 16:18:06 -07:00
Eric Anholt
4e16a7b526 glsl2: Fix up the implementation of fract() for vector types.
There's no need to split each vector component out, just do vector ops.
2010-06-30 15:59:07 -07:00
Eric Anholt
285ff93819 ir_to_mesa: Initialize the (we never use it) abs field of Mesa src regs. 2010-06-30 15:49:04 -07:00
Eric Anholt
ea6b34cce4 ir_to_mesa: Send the negate field on to Mesa IR.
Fixes glsl-fs-neg.
2010-06-30 15:47:11 -07:00
Eric Anholt
698b844443 ir_to_mesa: When generating a swizzle, respect the reg's current swizzle.
Fixes depth-tex-modes-glsl.
2010-06-30 15:30:00 -07:00
Kenneth Graunke
77049a702a glsl2: Implement AST->HIR support for the "discard" instruction. 2010-06-30 14:54:58 -07:00
Kenneth Graunke
16efab1c4d glsl2: Define new ir_discard instruction. 2010-06-30 14:54:58 -07:00
Eric Anholt
88c20c46b8 ir_to_mesa: Support gl_FragDepth.
Fixes glsl-bug-22603.
2010-06-30 14:55:45 -07:00
Eric Anholt
97eba76b8c glsl2: Allow a fragment shader to not write a color.
I can't find any text justifying this check, and it caused a
reasonable-looking shader in glsl-bug-22603 (which writes only
gl_FragDepth) to fail.
2010-06-30 14:51:50 -07:00
Eric Anholt
3e2127b9de mesa: Don't look in unallocatd param slots for parameter values.
glsl-derivs would add 40.0, 0.0, and 1.0 in order.  When we went
looking for 0.0, we'd find it in the second slot of the param, and use
it, but param->Size would still be 1.  When we went to add 1.0 and
didn't find it, we'd put allocate it to that second slot and the 0.0
would actualy end up being 1.0.

Fixes glsl-derivs, glsl-deriv-varyings.
2010-06-30 14:46:06 -07:00
Eric Anholt
c64da87611 ir_to_mesa: Fix the indexing of attributes in the program's Attributes.
This fixes GetAttribLocation returning VERT_ATTRIB_GENERIC1 instead of 1,
caught by glsl-dlist-getattriblocation.
2010-06-30 14:44:14 -07:00
Kenneth Graunke
c7f4ff193a glsl2: Fix storing of dead memory in the symbol table.
decl->identifier is part of the AST, so it doesn't live very long.
Instead, add var->name which is owned by var.
2010-06-30 13:52:24 -07:00
Kenneth Graunke
629198b96a glsl2: Preprocessed source doesn't need to live past compile time. 2010-06-30 13:52:24 -07:00
Kenneth Graunke
c6099a65f8 glsl2: Create new talloc contexts the "right" way. 2010-06-30 13:52:24 -07:00
Kenneth Graunke
953ff1283d glsl2: Use _mesa_glsl_parse_state as the talloc parent, not glsl_shader.
_mesa_glsl_parse_state should be the parent for all temporary allocation
done while compiling a shader.  glsl_shader should only be used as the
parent for the shader's final IR---the _result_ of compilation.

Since many IR instructions may be added or discarded during optimization
passes, IR should not ever be allocated to glsl_shader directly.

Done via sed -i s/talloc_parent(state)/state/g and s/talloc_parent(st)/st/g.

This also removes a ton of talloc_parent calls, which may help performance.
2010-06-30 13:52:24 -07:00
Kenneth Graunke
116f1d4f95 glsl2: Steal the live IR and free the rest of the junk. 2010-06-30 13:52:24 -07:00
Brian Paul
fb06b543a5 llvmpipe: another null pointer check 2010-06-30 14:05:45 -06:00
Brian Paul
2b3e1ad731 llvmpipe: use dummy tile when out of memory 2010-06-30 14:05:45 -06:00
Ian Romanick
ef5f194831 linker: Don't dynamically allocate slots for linked shaders
The can be at most one shader per stage.  There are currently only two
stages.  There is zero reason to dynamically size this array.
2010-06-30 12:20:44 -07:00
Eric Anholt
edcb9c2b06 ir_to_mesa: Tell Mesa about our choices for vertex attribute locations. 2010-06-30 12:02:31 -07:00
Ian Romanick
982e3798d8 linker: Don't automatically allocate VERT_ATTRIB_GENERIC0 2010-06-30 12:02:11 -07:00
Eric Anholt
849e18153c glsl2: Use Mesa's gl_shader_program instead of our own struct glsl_program.
This avoids more allocation and shuffling of data around.
2010-06-30 12:01:42 -07:00