summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* llvmpipe: make vertex-related functions static, clean-up initializationsBrian Paul2010-04-283-21/+19
|
* llvmpipe: make blend-related functions static, clean-up initializationsBrian Paul2010-04-283-42/+38
|
* llvmpipe: make sampler-related functions static, clean-up initializationsBrian Paul2010-04-283-46/+27
|
* llvmpipe: remove some unneeded shader structure fieldsBrian Paul2010-04-282-9/+2
|
* llvmpipe: dump shaders if LP_DEBUG=tgsiBrian Paul2010-04-282-0/+12
|
* draw: use a cast wrapperBrian Paul2010-04-281-4/+11
|
* draw: put 'create' in the vs varient create function namesBrian Paul2010-04-285-12/+16
|
* radeong: fix scons buildJoakim Sindholt2010-04-281-0/+1
|
* draw: add missing function parameterBrian Paul2010-04-281-1/+2
|
* softpipe: consolidate tex image offset codeBrian Paul2010-04-281-29/+32
|
* llvmpipe: remove unused write_depthBrian Paul2010-04-283-15/+6
|
* softpipe: use consistant names for local varsBrian Paul2010-04-281-36/+33
|
* gallium: remove trailing commaBrian Paul2010-04-281-1/+1
|
* gallium/util: convert //-style commentsBrian Paul2010-04-281-2/+2
|
* gallium/util: convert //-style commentsBrian Paul2010-04-281-2/+2
|
* llvmpipe: Remove unnecessary header.Vinson Lee2010-04-281-1/+0
|
* r300g: move ZTOP to separate fileMarek Olšák2010-04-286-73/+142
| | | | Ideally, HiZ should go in r300_hyperz.c next to ZTOP.
* gallium: Drop BGNFOR, ENDFOR, REP, and ENDREP opcodes.José Fonseca2010-04-271-4/+2
| | | | | Arcane DX9 opcodes that never were emitted by Mesa and have no DX10 counterpart. Slated for removal for quite some time now.
* gallium/docs: Drop BGNFOR, ENDFOR, REP, and ENDREP opcodes.José Fonseca2010-04-271-51/+0
|
* tgsi: Drop BGNFOR, ENDFOR, REP, and ENDREP opcodes.José Fonseca2010-04-277-161/+4
|
* r300: Drop BGNFOR, ENDFOR, REP, and ENDREP opcodes.José Fonseca2010-04-271-2/+0
|
* cell: Drop BGNFOR, ENDFOR, REP, and ENDREP opcodes.José Fonseca2010-04-271-12/+0
|
* svga: Drop BGNFOR, ENDFOR, REP, and ENDREP opcodes.José Fonseca2010-04-271-3/+0
|
* gallivm: Drop BGNFOR, ENDFOR, REP, and ENDREP opcodes.José Fonseca2010-04-271-16/+0
|
* llvmpipe: remove lp_scene_map_buffers(), lp_scene_unmap_buffers()Brian Paul2010-04-271-58/+4
|
* llvmpipe: implement max scene sizeBrian Paul2010-04-274-2/+61
| | | | | | | | When the size of the scene (binned data plus referenced resources/textures) exceeds LP_MAX_SCENE_SIZE flush/render the scene. This could be improved in various ways but is a good start. Fixes piglit streaming-texture-leak test.
* llvmpipe: added llvmpipe_resource_size()Brian Paul2010-04-272-0/+25
|
* llvmpipe: comment-out unused field for nowBrian Paul2010-04-271-1/+1
|
* llvmipe: update commentsBrian Paul2010-04-271-3/+3
|
* util: fix assorted problems in the image packing functionsBrian Paul2010-04-271-84/+70
|
* gallivm: Disable llvm.cos.v4f32 and llvm.sin.v4f32 instrinsics on Windows.José Fonseca2010-04-271-0/+18
| | | | | | | | Runtime linking doesn't quite work. Just comment then out for now to prevent crashes. These will go away in the future because calling 4 times CRT's cosf()/sinf() is over-precise and under-performing.
* gallivm: Ensure all allocas are in the first block.José Fonseca2010-04-276-41/+97
| | | | Refactor the code to make this easier.
* i965g: Fix scons build of winsysJakob Bornecrantz2010-04-271-0/+2
|
* st/dri: Don't check for null when user ensures non-nullJakob Bornecrantz2010-04-271-30/+22
|
* st/dri: Use flush function of old context not newJakob Bornecrantz2010-04-271-1/+1
|
* r300g: fix mipmapped texture3D size calculationMarek Olšák2010-04-271-1/+1
| | | | | | I accidentally inverted the condition. :( Broken since the commit 2579fe4044012a552c8be222ffae76b48de13592. FDO bug #27851.
* r300g: compensate for non-atomized emit_query_end in the CSMarek Olšák2010-04-272-4/+11
| | | | And reserve a little more space just in case.
* nvfx: Move src/gallium/drivers to beginning of SCons include path.Vinson Lee2010-04-271-1/+1
| | | | | | | | | | | | libdrm-2.4.20 and earlier include the nouveau/nouveau_class.h header. A later version of libdrm will not ship this header. Mesa also has this header at src/gallium/drivers. The symbol NV34TCL_VTXFMT_TYPE_HALF is needed by nvfx_vbo.c. This symbol is not in the libdrm copy of the header but is in the Mesa copy of the header. This patch moves src/gallium/drivers to the beginning of the include paths such that when building on hosts with libdrm-2.4.20 or ealier the build uses the copy in Mesa.
* r300g: reserve CS space for index offsetMarek Olšák2010-04-271-1/+1
| | | | | | | This has caused CS overflow since the commit 671f1e1229877d459cfd76e2cf5ea89c8d881036. TA3D works here again.
* llvmpipe: Remove unused variable.Vinson Lee2010-04-261-2/+0
|
* r300g: add support for more unaligned vertex formatsMarek Olšák2010-04-271-1/+33
| | | | | | An aligned stride is still needed for it to work. I am slowly approaching the Sauerbraten milestone!
* r300g: setup PSC first, then align vertex formatsMarek Olšák2010-04-271-3/+5
| | | | This allows us to put correct swizzles for undefined components in PSC.
* r300g: set undefined vertex swizzles to (0,0,0,1) respectivelyMarek Olšák2010-04-271-1/+8
|
* draw/llvm: implement fetch elts pathsZack Rusin2010-04-263-31/+160
| | | | | | | we were only running the llvm paths when the input elts were linear, now we can handle abritrary fetch elts arrays. we do this by generating two paths - linear and fetch_elts one and just selecting the right one at run time.
* r300g: disable point sprites (again)Marek Olšák2010-04-261-1/+4
| | | | They cause lockups. The users doesn't seem to like them.
* r300g: fix warnings when printing uint64_tMarek Olšák2010-04-261-2/+4
| | | | Though not nice, this is the correct way.
* r300g: fix warnings by using the const qualifierMarek Olšák2010-04-262-2/+2
| | | | See also the libdrm commit af98ccf4dd5dcb1b904ec32b9bd1521e6bf7dda5.
* r300g: use CS_OUT_TABLE betterMatt Turner2010-04-261-13/+3
| | | | | | Reviewed-by: Corbin Simpson <[email protected]> Signed-off-by: Matt Turner <[email protected]> Signed-off-by: Marek Olšák <[email protected]>
* gallivm: BGNFOR/ENDFOR fallthrough to BGNLOOP/ENDLOOPAlan Hourihane2010-04-261-10/+2
|
* draw: Always use the llvm middle end when available & enabled.José Fonseca2010-04-262-17/+23
|