| Commit message (Expand) | Author | Age | Files | Lines |
* | gallium: add PIPE_CAP_TGSI_VOTE for when the VOTE ops are allowed | Ilia Mirkin | 2016-06-06 | 17 | -0/+17 |
* | gallium: add VOTE_* opcodes to implement GL_ARB_shader_group_vote | Ilia Mirkin | 2016-06-06 | 3 | -1/+26 |
* | mesa: hook up core bits of GL_ARB_shader_group_vote | Ilia Mirkin | 2016-06-06 | 12 | -1/+67 |
* | glsl: Make opt_copy_propagation_elements actually propagate into loops. | Kenneth Graunke | 2016-06-06 | 1 | -2/+17 |
* | glsl: Make opt_copy_propagation actually propagate into loops. | Kenneth Graunke | 2016-06-06 | 1 | -6/+24 |
* | nv50/ir: use round toward 0 when converting doubles to integers | Samuel Pitoiset | 2016-06-06 | 1 | -1/+3 |
* | gallium/radeon: don't re-set BO metadata after CMASK deallocation | Marek Olšák | 2016-06-06 | 1 | -1/+0 |
* | st/mesa: change SQRT lowering to fix the game Risen | Marek Olšák | 2016-06-06 | 1 | -7/+9 |
* | radeonsi: add a performance tweak for 4 SE parts | Marek Olšák | 2016-06-06 | 1 | -0/+11 |
* | radeonsi: simplify PRIMGROUP_SIZE computation for tessellation | Marek Olšák | 2016-06-06 | 1 | -9/+1 |
* | r600g: use hw MSAA resolve for non-trivial resolves | Marek Olšák | 2016-06-06 | 1 | -9/+53 |
* | radeonsi: use hw MSAA resolve for non-trivial resolves | Marek Olšák | 2016-06-06 | 1 | -10/+54 |
* | mesa/program_resource: return -1 for index if no location. | Dave Airlie | 2016-06-07 | 1 | -3/+16 |
* | radeonsi: set descriptor dirty mask on shader buffer unbind | Nicolai Hähnle | 2016-06-06 | 1 | -0/+1 |
* | st/mesa: fix resource leak in try_pbo_readpixels | Nicolai Hähnle | 2016-06-06 | 1 | -1/+1 |
* | tgsi: fix mixed data type comparison in tgsi_point_sprite.c | Charmaine Lee | 2016-06-06 | 1 | -3/+3 |
* | svga: print shader linkage info when tgsi debug bit is on | Charmaine Lee | 2016-06-06 | 1 | -2/+5 |
* | st/mesa: check shader image format support before using PBO download | Ilia Mirkin | 2016-06-06 | 1 | -0/+6 |
* | tgsi: use truncf in micro_trunc | Lars Hamre | 2016-06-06 | 1 | -4/+4 |
* | i965/gs/scalar: Fix load input for doubles | Samuel Iglesias Gonsálvez | 2016-06-06 | 1 | -18/+54 |
* | i965/fs: fix offset when loading double vector input varyings | Samuel Iglesias Gonsálvez | 2016-06-06 | 1 | -1/+21 |
* | i965/fs: fix FS_OPCODE_CINTERP for unpacked double input varyings | Samuel Iglesias Gonsálvez | 2016-06-06 | 1 | -1/+12 |
* | glsl: geom shader max_vertices layout must match. | Dave Airlie | 2016-06-06 | 3 | -9/+13 |
* | anv/pipeline: Add support for caching the push constant map | Jason Ekstrand | 2016-06-06 | 1 | -4/+29 |
* | glsl: use enum glsl_interface_packing in more places. (v2) | Dave Airlie | 2016-06-06 | 11 | -36/+46 |
* | i965: don't use NumLayers for 3D textures. | Dave Airlie | 2016-06-06 | 1 | -1/+1 |
* | glsl: for anonymous struct matching use without_array() (v3) | Dave Airlie | 2016-06-06 | 1 | -3/+20 |
* | glsl/ast: don't crash when func_name is NULL | Dave Airlie | 2016-06-06 | 1 | -0/+4 |
* | glsl: handle ast_aggregate in has_sequence_subexpression. (v2) | Dave Airlie | 2016-06-06 | 1 | -1/+1 |
* | mesa: Try to unbreak the MSVC build. | Kenneth Graunke | 2016-06-05 | 2 | -0/+8 |
* | mesa: Add MESA_SHADER_CAPTURE_PATH for writing .shader_test files. | Kenneth Graunke | 2016-06-05 | 3 | -0/+77 |
* | nv50,nvc0: fix BGR10_A2UI vertex format | Ilia Mirkin | 2016-06-05 | 1 | -1/+1 |
* | nvc0: do not clear surfaces bins in the validate function | Samuel Pitoiset | 2016-06-05 | 2 | -5/+2 |
* | nvc0: re-validate images after launching a grid on Fermi | Samuel Pitoiset | 2016-06-05 | 1 | -0/+3 |
* | radeonsi: fix images with level > 0 | Marek Olšák | 2016-06-05 | 1 | -1/+1 |
* | nvc0: reduce overhead from always marking images dirty | Ilia Mirkin | 2016-06-04 | 1 | -9/+36 |
* | nvc0: reduce overhead from always marking buffers dirty | Ilia Mirkin | 2016-06-04 | 1 | -6/+20 |
* | nvc0: fix memory barrier flag handling | Ilia Mirkin | 2016-06-04 | 1 | -9/+16 |
* | nvc0: mark bound buffer range valid | Ilia Mirkin | 2016-06-04 | 3 | -0/+9 |
* | anv/entrypoints: don't go using wayland/xcb unless they are configured | Dave Airlie | 2016-06-05 | 1 | -6/+9 |
* | gallium/radeon: don't use the DMA ring for pipelined buffer uploads | Marek Olšák | 2016-06-04 | 1 | -5/+4 |
* | r600g: don't flush caches when binding shader resources | Marek Olšák | 2016-06-04 | 4 | -31/+26 |
* | r600g: only do necessary cache flushes in cp_dma_copy_buffer | Marek Olšák | 2016-06-04 | 1 | -14/+1 |
* | r600g: only do necessary cache flushes in cp_dma_clear_buffer | Marek Olšák | 2016-06-04 | 2 | -14/+18 |
* | r600g: remove a CP DMA workaround that's not needed anymore | Marek Olšák | 2016-06-04 | 1 | -6/+0 |
* | r600g: fix CP DMA hazard with index buffer fetches (v3) | Marek Olšák | 2016-06-04 | 7 | -7/+93 |
* | r600g: properly sync CP with CP DMA on R6xx | Marek Olšák | 2016-06-04 | 1 | -1/+8 |
* | r600g: write WAIT_UNTIL in the correct place | Marek Olšák | 2016-06-04 | 1 | -8/+11 |
* | gallium/radeon: rename allocator_so_filled_size -> allocator_zeroed_memory | Marek Olšák | 2016-06-04 | 3 | -6/+6 |
* | gallium/u_suballoc: allow different alignment for each allocation | Marek Olšák | 2016-06-04 | 8 | -21/+20 |