| Commit message (Expand) | Author | Age | Files | Lines |
* | radeonsi: do per-pixel clipping based on viewport states | Marek Olšák | 2016-04-08 | 2 | -11/+85 |
* | nv50/ir: do not try to attach JOIN ops to ATOM | Samuel Pitoiset | 2016-04-07 | 1 | -1/+1 |
* | radeonsi: raise number of samplers per shader to 32 | Nicolai Hähnle | 2016-04-07 | 1 | -3/+3 |
* | radeonsi: expand the compressed color and depth texture masks to 64 bits | Nicolai Hähnle | 2016-04-07 | 3 | -18/+18 |
* | radeonsi: replace magic 16 by SI_NUM_USER_SAMPLERS | Nicolai Hähnle | 2016-04-07 | 1 | -1/+1 |
* | gallium: raise PIPE_MAX_SAMPLERS to 32 | Nicolai Hähnle | 2016-04-07 | 1 | -1/+1 |
* | tgsi/scan: add an assert for the size of the samplers_declared bitfield | Nicolai Hähnle | 2016-04-07 | 1 | -1/+2 |
* | draw/aaline: stronger guard against no free samplers (v2) | Nicolai Hähnle | 2016-04-07 | 1 | -2/+4 |
* | util/pstipple: stronger guard against no free samplers (v2) | Nicolai Hähnle | 2016-04-07 | 1 | -2/+4 |
* | svga: new SVGA_MSAA env var to disable/enable MSAA pixel formats | Brian Paul | 2016-04-07 | 1 | -2/+4 |
* | svga: add some trivial null pointer checks | Brian Paul | 2016-04-07 | 3 | -0/+9 |
* | trace: add missing set_shader_images() | Samuel Pitoiset | 2016-04-07 | 3 | -0/+81 |
* | radeonsi: disable perfect ZPASS counts for PIPE_QUERY_OCCLUSION_PREDICATE | Marek Olšák | 2016-04-07 | 3 | -5/+16 |
* | radeonsi: don't use the real barrier instruction in tess ctrl shaders | Marek Olšák | 2016-04-07 | 1 | -0/+8 |
* | Revert "clover: Fix build against clang SVN >= r265359" | Michel Dänzer | 2016-04-07 | 1 | -3/+0 |
* | r600: use radeon_emit in a few more places in evergreen_compute | Dave Airlie | 2016-04-07 | 1 | -4/+4 |
* | r600: make compute global buffer functions static. | Dave Airlie | 2016-04-07 | 2 | -98/+86 |
* | r600: make two compute functions static. | Dave Airlie | 2016-04-07 | 2 | -5/+3 |
* | r600: using pipe_grid_info more in evergreen_compute. | Dave Airlie | 2016-04-07 | 2 | -26/+21 |
* | r600: in evergreen_compute use ctx consistently instead of ctx_ | Dave Airlie | 2016-04-07 | 1 | -25/+25 |
* | r600: use rctx consistently in evergreen_compute.c | Dave Airlie | 2016-04-07 | 1 | -74/+74 |
* | r600: cleanup whitespace in evergreen_compute.c | Dave Airlie | 2016-04-07 | 1 | -87/+75 |
* | r600g: Enable ARB_framebuffer_no_attachments | Edward O'Callaghan | 2016-04-07 | 1 | -1/+1 |
* | radeonsi: Enable ARB_framebuffer_no_attachments | Edward O'Callaghan | 2016-04-07 | 1 | -1/+1 |
* | radeonsi: Improve assert info out of si_set_framebuffer_state() | Edward O'Callaghan | 2016-04-07 | 1 | -0/+2 |
* | radeonsi: Allow 16 samples MSAA mode for PIPE_FORMAT_NONE | Edward O'Callaghan | 2016-04-07 | 1 | -0/+5 |
* | softpipe: Set samples and layers in set_framebuffer_state() cb | Edward O'Callaghan | 2016-04-07 | 1 | -0/+2 |
* | gallium/trace: Dump no.of samples and layers in fb state | Edward O'Callaghan | 2016-04-07 | 1 | -0/+2 |
* | gallium: Put no.of {samples,layers} into pipe_framebuffer_state | Edward O'Callaghan | 2016-04-07 | 3 | -0/+40 |
* | gallium: Add PIPE_CAP_FRAMEBUFFER_NO_ATTACHMENT | Edward O'Callaghan | 2016-04-07 | 16 | -0/+23 |
* | radeonsi: set shader calling conventions | Bas Nieuwenhuizen | 2016-04-06 | 1 | -1/+16 |
* | freedreno/ir3: insert extra move into phi | Rob Clark | 2016-04-05 | 1 | -0/+10 |
* | freedreno/ir3: eliminate unnecessary absneg's | Rob Clark | 2016-04-05 | 2 | -3/+26 |
* | clover: Fix build against clang SVN >= r265359 | Michel Dänzer | 2016-04-05 | 1 | -0/+3 |
* | radeonsi: use bounded indexing for samplers | Bas Nieuwenhuizen | 2016-04-05 | 1 | -1/+4 |
* | radeonsi: use bounded indexing for constant buffers | Bas Nieuwenhuizen | 2016-04-05 | 1 | -2/+3 |
* | gallium/radeon: allow multiple exports of the same texture with different usage | Marek Olšák | 2016-04-05 | 1 | -21/+33 |
* | freedreno/ir3: deal with duplicate phi sources | Rob Clark | 2016-04-04 | 1 | -5/+20 |
* | freedreno/ir3: fix silly brain-fart in RA | Rob Clark | 2016-04-04 | 1 | -2/+1 |
* | freedreno/ir3: don't cp into phi's | Rob Clark | 2016-04-04 | 1 | -0/+6 |
* | freedreno/ir3: we can't store immediate values | Rob Clark | 2016-04-04 | 1 | -0/+13 |
* | freedreno/ir3: add dumping for use/def/live-in/live-out | Rob Clark | 2016-04-04 | 3 | -10/+42 |
* | freedreno/ir3: drop unused instr category arg | Rob Clark | 2016-04-04 | 5 | -114/+108 |
* | freedreno/ir3: remove ir3_instruction::category | Rob Clark | 2016-04-04 | 10 | -93/+84 |
* | freedreno/ir3: encode instruction category in opc_t | Rob Clark | 2016-04-04 | 5 | -192/+201 |
* | nvc0: add hardware ETC2 and ASTC support on GK20A and GM107+ | Ilia Mirkin | 2016-04-04 | 3 | -2/+64 |
* | gallivm: Introduce lp_format_intrinsic. | Jose Fonseca | 2016-04-04 | 3 | -14/+54 |
* | gallivm: Use llvm.fabs. | Jose Fonseca | 2016-04-03 | 1 | -8/+3 |
* | gallivm: Prefer backend agnostic intrinsic for rounding. | Jose Fonseca | 2016-04-03 | 1 | -7/+39 |
* | gallivm: Add debug option to force SSE2. | Jose Fonseca | 2016-04-03 | 1 | -11/+14 |