Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | softpipe: Update SConscript. | José Fonseca | 2009-09-24 | 1 | -6/+0 |
| | |||||
* | r600 : disable draw_prim for now. | Richard Li | 2009-09-24 | 1 | -1/+4 |
| | |||||
* | gdi: Update for WGL state tracker interface changes. | José Fonseca | 2009-09-24 | 2 | -8/+16 |
| | |||||
* | wgl: DWM integration. | José Fonseca | 2009-09-24 | 10 | -67/+380 |
| | |||||
* | mesa: Fix missing finite symbol error on Windows. | José Fonseca | 2009-09-24 | 1 | -2/+1 |
| | | | | | Caused by some weird logic regarding the __WIN32__ define which made the finite definition dependent on the header include order. | ||||
* | r600: fix up ordering of functions in draw prims path | Alex Deucher | 2009-09-23 | 1 | -16/+15 |
| | | | | | Shaders and IB need to be updated and allocated before calling validatebuffers. | ||||
* | r600: fix r700PredictRenderSize for draw prims path | Alex Deucher | 2009-09-23 | 1 | -7/+12 |
| | |||||
* | r600 : add hw index buffer draw support. | Richard Li | 2009-09-23 | 1 | -12/+61 |
| | |||||
* | Finish removing glcore | Adam Jackson | 2009-09-23 | 1 | -84/+0 |
| | |||||
* | r600: fix copy/paste typo | Alex Deucher | 2009-09-23 | 1 | -1/+1 |
| | |||||
* | r600: fix some warnings | Alex Deucher | 2009-09-23 | 2 | -3/+4 |
| | |||||
* | r600: add support for CUBE textures, also TXP | Andre Maasikas | 2009-09-23 | 2 | -47/+263 |
| | | | | seems to work here ... | ||||
* | llvmpipe: move tile cache datatypes into .c file since they're private | Brian Paul | 2009-09-23 | 2 | -36/+38 |
| | |||||
* | softpipe: replace macros with inline functions | Brian Paul | 2009-09-23 | 1 | -21/+32 |
| | | | | And update comments. | ||||
* | softpipe: white-space/formatting fixes and updated comments | Brian Paul | 2009-09-23 | 1 | -86/+87 |
| | |||||
* | llvmpipe: added max texture/surface size sanity check | Brian Paul | 2009-09-23 | 1 | -0/+6 |
| | | | | Carried over from softpipe driver. | ||||
* | llvmpipe: increase MAX_WIDTH/HEIGHT to match max texture size | Brian Paul | 2009-09-23 | 1 | -2/+2 |
| | |||||
* | Merge branch 'mesa_7_6_branch' | Brian Paul | 2009-09-23 | 3 | -8/+8 |
|\ | | | | | | | | | | | Conflicts: src/gallium/drivers/softpipe/sp_tile_cache.c | ||||
| * | softpipe: added max texture/surface size sanity check | Brian Paul | 2009-09-23 | 1 | -0/+6 |
| | | |||||
| * | softpipe: increase MAX_WIDTH/HEIGTH 4096 to match max texture size | Brian Paul | 2009-09-23 | 1 | -2/+2 |
| | | |||||
* | | Merge branch 'softpipe-opt' | Keith Whitwell | 2009-09-23 | 45 | -4314/+4577 |
|\ \ | | | | | | | | | | | | | | | | | | | Conflicts: progs/demos/cubemap.c src/gallium/drivers/softpipe/sp_tex_sample.c src/gallium/drivers/softpipe/sp_texture.c | ||||
| * | | softpipe: fix bugs in POT texture sampling when texture is not square | root | 2009-09-22 | 1 | -6/+28 |
| | | | | | | | | | | | | | | | Before, if level was greater than the logbase2(base size) we were doing a negative bit shift and winding up with garbage values. | ||||
| * | | softpipe: additional assertions | Brian Paul | 2009-09-22 | 1 | -0/+4 |
| | | | |||||
| * | | softpipe: disable a _debug_printf() | Brian Paul | 2009-09-22 | 1 | -0/+2 |
| | | | |||||
| * | | softpipe: need to write depth/stencil values even when stencil fails | Keith Whitwell | 2009-09-22 | 1 | -12/+10 |
| | | | |||||
| * | | softpipe: set quad->facing value | Keith Whitwell | 2009-09-22 | 1 | -0/+1 |
| | | | |||||
| * | | softpipe: fix polygon stipple | Keith Whitwell | 2009-09-22 | 1 | -2/+0 |
| | | | |||||
| * | | softpipe: fix occlusion counting | Keith Whitwell | 2009-09-22 | 1 | -2/+17 |
| | | | |||||
| * | | mesa/swrast: use one fewer divide in swrast's choose_cube_face also | Keith Whitwell | 2009-08-24 | 1 | -2/+6 |
| | | | | | | | | | | | | Same change as for softpipe | ||||
| * | | softpipe: use one fewer divide in sample_cube | Keith Whitwell | 2009-08-24 | 1 | -3/+6 |
| | | | | | | | | | | | | | | | GCC won't do this for us. Makes a bigger difference to cubemap fps than previous set of compilcated rearrangements. | ||||
| * | | softpipe: separate out 2d and cube img filter functions | Keith Whitwell | 2009-08-24 | 1 | -4/+88 |
| | | | |||||
| * | | softpipe: make the various get_texel routines more similar | Keith Whitwell | 2009-08-24 | 1 | -92/+176 |
| | | | | | | | | | | | | | | | Remove arguments, return const float * by default. Add specialized 3d versions and remove 3d texture support from the others. | ||||
| * | | softpipe: lift tex_address construction up to img_filter | Keith Whitwell | 2009-08-23 | 1 | -25/+35 |
| | | | | | | | | | | | | For fastpaths at least, can avoid recalculating this sometimes. | ||||
| * | | softpipe: remove old prim_setup draw stage | Keith Whitwell | 2009-08-23 | 10 | -386/+59 |
| | | | | | | | | | | | | | | | Everything now goes through the draw_vbuf handler, the same as regular drivers. | ||||
| * | | softpipe: add missing header | Keith Whitwell | 2009-08-23 | 1 | -0/+1 |
| | | | |||||
| * | | Merge branch 'tex-tilecache' into softpipe-opt | Keith Whitwell | 2009-08-23 | 13 | -228/+464 |
| |\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conflicts: src/gallium/drivers/softpipe/sp_state_derived.c src/gallium/drivers/softpipe/sp_state_sampler.c src/gallium/drivers/softpipe/sp_tex_sample.c src/gallium/drivers/softpipe/sp_tex_sample.h src/gallium/drivers/softpipe/sp_tile_cache.c | ||||
| | * | | softpipe: fix typo in clear_tile | Keith Whitwell | 2009-08-18 | 1 | -1/+1 |
| | | | | |||||
| | * | | softpipe: split texture and surface tile caches | Keith Whitwell | 2009-08-18 | 13 | -224/+460 |
| | | | | | | | | | | | | | | | | | | | | These do similar jobs but with largely disjoint code. Will want to evolve them separately going forward. | ||||
| * | | | softpipe: remove duplicate #include, move another | Brian Paul | 2009-08-21 | 1 | -2/+1 |
| | | | | |||||
| * | | | softpipe: remove tex sample dependencies on softpipe | Brian Paul | 2009-08-21 | 1 | -6/+3 |
| | | | | | | | | | | | | | | | | | | | | The texture sampling code doesn't really have any dependencies on the rest of softpipe, just the tile cache. | ||||
| * | | | softpipe: minor code refactoring to remove softpipe/tile cache dependencies | Brian Paul | 2009-08-21 | 3 | -16/+24 |
| | | | | | | | | | | | | | | | | The tile cache code now has no hard dependencies on softpipe. | ||||
| * | | | softpipe: remove unused #includes, white-space clean-up | Brian Paul | 2009-08-21 | 1 | -5/+2 |
| | | | | |||||
| * | | | softpipe: remove unneeded const qualifier | Brian Paul | 2009-08-21 | 1 | -1/+1 |
| | | | | |||||
| * | | | softpipe: per-unit sampler varients | Brian Paul | 2009-08-21 | 3 | -5/+25 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Can't share sampler varients across multiple tex units because the texture pointer is in the sampler varient. That prevents different textures per unit. Fixes progs/demos/multiarb, progs/glsl/samplers, etc. | ||||
| * | | | softpipe: add missing PIPE_TEXTURE_CUBE case in get_lambda_func() | Brian Paul | 2009-08-21 | 1 | -0/+1 |
| | | | | | | | | | | | | | | | | Fixes progs/demos/cubemap | ||||
| * | | | softpipe: add missing sp_sampler_varient_destroy | Keith Whitwell | 2009-08-21 | 1 | -0/+8 |
| | | | | |||||
| * | | | softpipe: remove redundant comparison, make test easier to understand | Brian Paul | 2009-08-21 | 1 | -2/+1 |
| | | | | |||||
| * | | | softpipe: fix min/mag filter typo | Brian Paul | 2009-08-21 | 1 | -1/+1 |
| | | | | |||||
| * | | | softpipe: fix s/t/p typos | Brian Paul | 2009-08-21 | 1 | -3/+3 |
| | | | | |||||
| * | | | softpipe: rework texture sampling code | Keith Whitwell | 2009-08-21 | 11 | -1100/+1350 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Split into component pieces, stitch together at runtime using function pointers. Make it possible to utilize the existing fastpaths as image-level filters for generic mip-filtering routines. Remove special case for rectangle filtering, as it can now be handled by the 2d path. As most of the mesa demo texturing was already covered by fast paths, its harder to find examples of speedups, but tunnel gets a boost as mip-nearest filtering is now able to access the img_2d_linear_wrap_POT functions for sampling within a mipmap level. |