summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* util: Add util_next_power_of_two() for rounding a uint up to a POT.Younes Manton2009-09-271-0/+20
|
* r300g: fix erroneous conditionMaciej Cencora2009-09-271-1/+1
|
* r300g: don't force vertex position for HW TCL pathMaciej Cencora2009-09-271-1/+3
| | | | It could be generated by vertex shader.
* r300g: add some assertionsMaciej Cencora2009-09-271-0/+3
| | | | Not sure why we are getting a shader with two inputs with position semantic, but we don't know how to handle it correctly so it's better to stop the app than lock the machine.
* r300g: plug memory leakMaciej Cencora2009-09-271-1/+1
|
* r300g: add some debugging infoMaciej Cencora2009-09-271-10/+20
|
* r300g: fix r300 rs pathDave Airlie2009-09-281-2/+2
|
* r300g: rewrite RS state setup.Dave Airlie2009-09-281-24/+13
| | | | | Not 100% sure this is correct, but its more correct than what was here previous however it may require changes in the input routing for the frag shader.
* r300g: silence compiler warningDave Airlie2009-09-281-1/+1
|
* i915g: Submit direct vertex buffersJakob Bornecrantz2009-09-271-2/+31
|
* i915g: Fix warningJakob Bornecrantz2009-09-271-1/+1
|
* i915g: Tile shared buffers as wellJakob Bornecrantz2009-09-271-0/+47
|
* i915g: Make sure to map tiled buffers via the gttJakob Bornecrantz2009-09-272-1/+12
|
* i915g: Use booleanJakob Bornecrantz2009-09-271-1/+1
|
* i915g: Enable reuse of buffersJakob Bornecrantz2009-09-271-0/+1
|
* st/egl: Remove buffer from screenJakob Bornecrantz2009-09-272-5/+3
| | | | | | | It is no longer needed since the new drm api code, and it incorrectly checked if the buffer where there for testing completeness when it should have checked the texture instead.
* r600 : Enable draw_prim.Richard Li2009-09-271-2/+0
|
* softpipe: Grab a ref when the fb is set.Younes Manton2009-09-272-4/+9
| | | | | | | Nasty bug when the surface is freed and another is allocated right on top of it. The next time we set the fb state SP thinks it's the same surface and doesn't flush, and when the flush eventually happens the surface belongs to a completely different texture.
* progs/rbug: Link to mathJakob Bornecrantz2009-09-261-1/+1
|
* r300g: add tx depth support in register.Dave Airlie2009-09-262-0/+8
| | | | also enable cube/3d bits in txformat reg
* r300g: only pass complete texture state to hw setup functionDave Airlie2009-09-261-15/+12
| | | | | No point passing things twice here, also allows more state to be setup properly.
* r300g: add missing break in OQ emitDave Airlie2009-09-261-0/+1
|
* llvmpipe: Allow building with LLVM 2.6 too.José Fonseca2009-09-263-0/+15
|
* r300g: report GL1.5, enable cap bits for OQ and shadow.Dave Airlie2009-09-261-4/+2
| | | | | | | Its not like it works well on 1.3 so may as well reach for greater heights. Signed-off-by: Dave Airlie <airlied@redhat.com>
* r300g: add z16 unorm texture formatDave Airlie2009-09-261-0/+2
|
* r300g: fix texture pitch to correct value.Dave Airlie2009-09-261-1/+1
| | | | pitch is pixels - 1, not bytes.
* Merge branch 'asm-shader-rework-2'Ian Romanick2009-09-2510-2124/+3025
|\ | | | | | | | | Conflicts: src/mesa/shader/program_parse.tab.c
| * NV fp lexer: Add UP4B and UP4UB instructions that were previously missingIan Romanick2009-09-242-667/+686
| |
| * Merge branch 'master' into asm-shader-rework-2Ian Romanick2009-09-10306-9924/+18165
| |\ | | | | | | | | | | | | | | | | | | Conflicts: src/mesa/shader/lex.yy.c src/mesa/shader/program_parse.tab.c src/mesa/shader/program_parse.tab.h
| * | NV fp parser: Add support for condition codesIan Romanick2009-09-104-881/+1183
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conditional write masks and the condition-code based KIL instruction are all supported. The specific behavior of KIL in the following shader may or may not match the behavior of other implementations: !!ARBfp1.0 TEMP GT; MOVC GT, fragment.texcoord[0]; KIL GT.x; END Should be it interpreted as 'KIL srcReg' or as 'KIL ccTest'? The current parser will interpret it as 'KIL srcReg'.
| * | ARB prog lexer: Add missing #include to silence compile warningIan Romanick2009-09-102-173/+175
| | |
| * | ARB prog parser: Differentiate between used and unused names in the lexerIan Romanick2009-09-105-834/+851
| | | | | | | | | | | | | | | The lexer will return IDENTIFIER only when the name does not have an associated symbol. Otherwise USED_IDENTIFIER is returned.
| * | NV fp parser: Support instruction and TEMP / OUTPUT sizesIan Romanick2009-09-046-1506/+1463
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adds support for declaring TEMP and OUTPUT variables as 'LONG' or 'SHORT' precision. The precision specifiers are parsed, but they are currently ignored. Some support for this may be added in the future, but neither Intel hardware nor, as far as I'm aware, Radeon hardware support multiple precisions. Also adds support for instruction precision ('X', 'H', and 'R') suffixes and instruction condition code output ('C') suffix. This results in a fairly major change to the lexer. Instructions are matched with all the possible suffix strings. The suffix string are then carved off by a context (i.e., which program mode and options are set) aware parser that converts the suffixes to bits in prog_instruction. This could have been handled in the same way _SAT was originally handled in the lexer, but it would have resulted in a very large lexer with lots of opportunity for cut-and-paste errors.
| * | ARB prog parser: Add new constructor for asm_instructionIan Romanick2009-09-043-334/+400
| | | | | | | | | | | | | | | The new constructor copies fields from the prog_instruction that the parser expects the lexer to set.
| * | NV fp parser: Add support for absolute value operator on instruction operandsIan Romanick2009-09-032-840/+942
| | |
| * | Enable GL_NV_fragment_program_option for software renderingIan Romanick2009-09-032-0/+4
| | | | | | | | | | | | At this point the extension is not fully implemented.
| * | NV fp parser: Support new scalar constant behaviorIan Romanick2009-09-032-729/+770
| | | | | | | | | | | | | | | | | | ARBfp requires scalar constants have a '.x' suffix, but NVfp_option does not. This shows up with instructions that require a scalar parameter (e.g., COS).
| * | NV fp parser: Parse TXD instructionIan Romanick2009-09-012-842/+941
| | |
| * | NV fp lexer: Add new opcodesIan Romanick2009-08-315-1191/+1455
| | |
| * | NV fp: Parse 'OPTION NV_fragment_program' in ARB assembly shadersIan Romanick2009-08-312-0/+12
| | |
| * | NV fp: Add tracking for NV_fragment_program_optionIan Romanick2009-08-312-0/+2
| | |
* | | st/xorg: Re-enable accelerated fills and copies.Michel Dänzer2009-09-251-1/+1
| | | | | | | | | | | | | | | | | | These seem to work well enough now with the new code. Composite acceleration isn't quite there yet and thus remains disabled in xorg_composite_bind_state() for now.
* | | st/xorg: Flush render cache if but only if a source has pending write ↵Michel Dänzer2009-09-251-1/+9
| | | | | | | | | | | | operations.
* | | st/xorg: Reject Composite acceleration for some cases not working yet.Michel Dänzer2009-09-251-0/+10
| | |
* | | st/xorg: Make sure struct is fully initialized.Michel Dänzer2009-09-251-1/+1
| | | | | | | | | | | | gcc complained about a missing initializer.
* | | st/xorg: Use generic semantic for Composite mask coordinates.Michel Dänzer2009-09-251-2/+2
| | |
* | | st/xorg: Flesh out EXA PrepareComposite hook a little.Michel Dänzer2009-09-251-2/+28
| | | | | | | | | | | | | | | Check that the formats are supported, and don't crash with source-only pictures.
* | | st/xorg: Use I8 format instead of A8 for depth 8 pixmaps.Michel Dänzer2009-09-251-1/+1
| | | | | | | | | | | | Seems to work better for Composite acceleration.
* | | st/xorg: Bind rasterizer state for copies.Michel Dänzer2009-09-251-0/+2
| | |
* | | st/xorg: Source-only pictures always have format PICT_a8r8g8b8.Michel Dänzer2009-09-251-21/+3
| | | | | | | | | | | | | | | See xserver/render/picture.c:createSourcePicture(). This both simplifies the code and avoids a crash because pFormat is NULL.