| Commit message (Expand) | Author | Age | Files | Lines |
* | radeonsi: remove flatshade from the shader key | Marek Olšák | 2015-01-07 | 3 | -7/+7 |
* | radeonsi: remove special handling of TGSI_INTERPOLATE_COLOR in shader codegen | Marek Olšák | 2015-01-07 | 1 | -6/+10 |
* | radeonsi: implement VERTEXID_NOBASE and BASEVERTEX system values | Marek Olšák | 2015-01-07 | 1 | -0/+10 |
* | radeonsi: fix VertexID for OpenGL | Marek Olšák | 2015-01-07 | 1 | -2/+5 |
* | radeonsi: clarify a hw bug in shader exports | Marek Olšák | 2015-01-07 | 1 | -5/+10 |
* | radeonsi: use ordered compares for SSG and face selection | Marek Olšák | 2015-01-07 | 2 | -3/+3 |
* | radeonsi: remove unused and not useful variables | Marek Olšák | 2015-01-07 | 3 | -6/+1 |
* | radeonsi: remove init config from states | Marek Olšák | 2015-01-07 | 6 | -5/+4 |
* | radeonsi: reduce the size of si_pm4_state | Marek Olšák | 2015-01-07 | 2 | -12/+3 |
* | vc4: Fix scaling W projection of the Z coordinate when there's a Z offset. | Eric Anholt | 2015-01-06 | 1 | -3/+3 |
* | vc4: Fix deletion from the program cache. | Eric Anholt | 2015-01-06 | 1 | -1/+1 |
* | vc4: Skip storing the Z/S contents when it's invalidated. | Eric Anholt | 2015-01-06 | 1 | -0/+11 |
* | radeon/llvm: Use amdgcn triple for SI+ on LLVM >= 3.6 | Tom Stellard | 2015-01-06 | 4 | -16/+27 |
* | radeonsi: Cache LLVMTargetMachine object in si_screen | Tom Stellard | 2015-01-06 | 6 | -26/+51 |
* | nvc0: add name to magic number | Ilia Mirkin | 2015-01-05 | 1 | -2/+2 |
* | nvc0: regenerate rnndb headers | Ilia Mirkin | 2015-01-05 | 17 | -837/+1157 |
* | nv50: regenerate rnndb headers | Ilia Mirkin | 2015-01-05 | 11 | -358/+451 |
* | nv50: enable texture compression | Tobias Klausmann | 2015-01-05 | 2 | -3/+26 |
* | nv50/ir: enable sat modifier for OP_SUB | Ilia Mirkin | 2015-01-05 | 1 | -1/+1 |
* | nv50/ir: Add sat modifier for mul | Roy Spliet | 2015-01-05 | 2 | -1/+7 |
* | nv50,nvc0: avoid doing work inside of an assert | Ilia Mirkin | 2015-01-05 | 2 | -2/+4 |
* | nv50/ir: fix texture offsets in release builds | Ilia Mirkin | 2015-01-05 | 2 | -2/+4 |
* | r300g: handle vertex format PIPE_FORMAT_NONE | Marek Olšák | 2015-01-04 | 1 | -2/+11 |
* | nv50/ir: Fold sat into mad | Roy Spliet | 2015-01-01 | 1 | -1/+1 |
* | nv50/ir: fold MAD when one of the multiplicands is const | Ilia Mirkin | 2015-01-01 | 1 | -0/+23 |
* | radeonsi: fix warnings | Marek Olšák | 2015-01-01 | 2 | -1/+3 |
* | vc4: Fix memory leak as of 0404e7fe0ac2a6234a11290b4b1596e8bc127a4b. | Eric Anholt | 2014-12-31 | 1 | -5/+5 |
* | nv50,nvc0: set vertex id base to index_bias | Ilia Mirkin | 2014-12-30 | 5 | -7/+35 |
* | nv50,nvc0: implement half_pixel_center | Tiziano Bacocco | 2014-12-30 | 8 | -14/+11 |
* | vc4: Only render tiles where the scissor ever intersected them. | Eric Anholt | 2014-12-30 | 4 | -10/+52 |
* | vc4: Move draw call reset handling to a helper function. | Eric Anholt | 2014-12-30 | 1 | -23/+31 |
* | vc4: Drop the content of vc4_flush_resource(). | Eric Anholt | 2014-12-30 | 1 | -4/+4 |
* | vc4: Handle unaligned accesses in CL emits. | Eric Anholt | 2014-12-25 | 2 | -26/+78 |
* | vc4: Don't bother zero-initializing the shader reloc indices. | Eric Anholt | 2014-12-25 | 1 | -2/+2 |
* | vc4: Fix the argument type for cl_u16(). | Eric Anholt | 2014-12-25 | 1 | -1/+1 |
* | radeonsi: Don't modify PA_SC_RASTER_CONFIG register value if rb_mask == 0 | Michel Dänzer | 2014-12-25 | 1 | -2/+4 |
* | vc4: Optimize CL emits by doing size checks up front. | Eric Anholt | 2014-12-24 | 5 | -16/+66 |
* | vc4: Avoid repeated hindex lookups in the loop over tiles. | Eric Anholt | 2014-12-24 | 2 | -15/+24 |
* | freedreno/ir3: split out legalize pass | Rob Clark | 2014-12-23 | 5 | -154/+214 |
* | freedreno/ir3: ra debug | Rob Clark | 2014-12-23 | 3 | -17/+61 |
* | radeonsi: force NaNs to 0 | Marek Olšák | 2014-12-21 | 1 | -4/+8 |
* | r300g: implement ARR opcode | David Heidelberg | 2014-12-21 | 4 | -4/+16 |
* | freedreno/a4xx: blend-color | Rob Clark | 2014-12-20 | 1 | -0/+13 |
* | freedreno/a4xx: alpha-test | Rob Clark | 2014-12-20 | 1 | -0/+2 |
* | freedreno: update generated headers | Rob Clark | 2014-12-20 | 6 | -61/+151 |
* | freedreno/ir3: trans_kill cleanup | Rob Clark | 2014-12-20 | 1 | -12/+7 |
* | freedreno/ir3: hack for standalone compiler | Rob Clark | 2014-12-20 | 1 | -1/+5 |
* | vc4: Coalesce MOVs into VPM with the instructions generating the values. | Eric Anholt | 2014-12-18 | 4 | -15/+143 |
* | vc4: Redefine VPM writes as a (destination) QIR register file. | Eric Anholt | 2014-12-17 | 3 | -7/+19 |
* | vc4: Add support for turning constant uniforms into small immediates. | Eric Anholt | 2014-12-17 | 13 | -46/+283 |