| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | vc4: Add some dumping for STORE_TILE_BUFFER_GENERAL. | Eric Anholt | 2015-01-15 | 1 | -1/+79 |
| * | vc4: Add dumping for the TILE_RENDERING_MODE_CONFIG packet. | Eric Anholt | 2015-01-15 | 1 | -1/+70 |
| * | vc4: Fix CL dumping trying to dump too far. | Eric Anholt | 2015-01-15 | 1 | -2/+2 |
| * | vc4: Fix texture type masking. | Eric Anholt | 2015-01-15 | 1 | -1/+1 |
| * | vc4: Colormask should apply after all other fragment ops (like logic op). | Eric Anholt | 2015-01-15 | 1 | -9/+18 |
| * | vc4: No turning unpack arguments into small immediates. | Eric Anholt | 2015-01-15 | 1 | -0/+3 |
| * | vc4: Move the tests for src needing to be an A register to vc4_qir.c. | Eric Anholt | 2015-01-15 | 3 | -17/+28 |
| * | vc4: Don't swap the raddr on instructions doing unpacks. | Eric Anholt | 2015-01-15 | 1 | -0/+5 |
| * | vc4: Don't let pairing happen with badly mismatched unpack flags. | Eric Anholt | 2015-01-15 | 1 | -0/+39 |
| * | vc4: Don't let pairing happen with badly mismatched pack flags. | Eric Anholt | 2015-01-15 | 1 | -0/+39 |
| * | vc4: Fix early Z behavior on hardware. | Eric Anholt | 2015-01-15 | 1 | -2/+1 |
| * | Revert "radeonsi: only set BC_OPTIMIZE_DISABLE when necessary" | Michel Dänzer | 2015-01-15 | 2 | -15/+6 |
| * | st/clover: Adapt to TargetLibraryInfo.h move in LLVM SVN r226078 | Michel Dänzer | 2015-01-15 | 1 | -0/+4 |
| * | freedreno/ir3: handle "holes" in inputs | Rob Clark | 2015-01-13 | 1 | -1/+31 |
| * | r600g: fix build failure when building the driver without LLVM | Marek Olšák | 2015-01-12 | 1 | -0/+4 |
| * | vc4: Clamp the inputs to the blend equation to [0, 1]. | Eric Anholt | 2015-01-11 | 1 | -1/+10 |
| * | vc4: Add a little helper for clamping to [0,1]. | Eric Anholt | 2015-01-11 | 1 | -4/+10 |
| * | vc4: Fix up statechange management for uncompiled/compiled FS/VS. | Eric Anholt | 2015-01-11 | 2 | -11/+10 |
| * | vc4: Fix clear color setup for RGB565. | Eric Anholt | 2015-01-11 | 1 | -1/+4 |
| * | vc4: Avoid the save/restore of r3 for raddr conflicts, just use ra31. | Eric Anholt | 2015-01-11 | 2 | -38/+11 |
| * | vc4: Allow dead code elimination of VPM reads. | Eric Anholt | 2015-01-10 | 2 | -1/+44 |
| * | vc4: Cook up the draw-time VPM setup info during shader compile. | Eric Anholt | 2015-01-10 | 4 | -11/+28 |
| * | vc4: Split two notions of instructions having side effects. | Eric Anholt | 2015-01-10 | 5 | -4/+15 |
| * | vc4: Redo VPM reads as a read file. | Eric Anholt | 2015-01-10 | 5 | -16/+16 |
| * | vc4: Fix miscalculation of the VPM space. | Eric Anholt | 2015-01-10 | 1 | -1/+1 |
| * | vc4: Pack VPM attr contents according to just the size of the attribute. | Eric Anholt | 2015-01-10 | 3 | -11/+9 |
| * | vc4: Restructure color packing as a series of channel replacements. | Eric Anholt | 2015-01-10 | 4 | -49/+60 |
| * | vc4: Fix the no-copy-propagating-from-TLB_COLOR_READ check. | Eric Anholt | 2015-01-10 | 1 | -1/+1 |
| * | vc4: Move global seqno short-circuiting to vc4_wait_seqno(). | Eric Anholt | 2015-01-10 | 2 | -6/+3 |
| * | st/wgl: Ignore ulVersion in DrvValidateVersion. | José Fonseca | 2015-01-08 | 1 | -2/+10 |
| * | freedreno/ir3: fix pos_regid > max_reg | Rob Clark | 2015-01-07 | 4 | -41/+121 |
| * | freedreno/ir3: start on indirect gpr reads | Rob Clark | 2015-01-07 | 3 | -8/+146 |
| * | freedreno/ir3: make reg array dynamic | Rob Clark | 2015-01-07 | 4 | -13/+50 |
| * | freedreno/ir3: simplify RA | Rob Clark | 2015-01-07 | 8 | -777/+622 |
| * | freedreno/ir3: regmask support for relative addr | Rob Clark | 2015-01-07 | 2 | -17/+51 |
| * | freedreno/ir3: split up ssa_src | Rob Clark | 2015-01-07 | 1 | -23/+34 |
| * | freedreno/ir3: drop instr_clone() stuff | Rob Clark | 2015-01-07 | 2 | -49/+17 |
| * | freedreno/ir3: runtime enable RA debug for DEBUG builds | Rob Clark | 2015-01-07 | 1 | -1/+6 |
| * | freedreno/ir3: handle relative addr in ir3_dump | Rob Clark | 2015-01-07 | 1 | -1/+8 |
| * | freedreno/ir3: legalize vs unused sam dst components | Rob Clark | 2015-01-07 | 2 | -2/+9 |
| * | freedreno/ir3: hack for old compiler | Rob Clark | 2015-01-07 | 1 | -0/+23 |
| * | tgsi: track max array per file | Rob Clark | 2015-01-07 | 2 | -0/+4 |
| * | tgsi: keep track of read vs written indirects | Rob Clark | 2015-01-07 | 2 | -0/+8 |
| * | Revert "radeonsi: reduce the size of si_pm4_state" | Marek Olšák | 2015-01-08 | 2 | -3/+12 |
| * | radeonsi: Fix crash when destroying si_screen | Tom Stellard | 2015-01-07 | 1 | -2/+4 |
| * | radeonsi: enable LLVM optimizations that assume no NaNs for non-compute shaders | Marek Olšák | 2015-01-07 | 3 | -4/+12 |
| * | radeonsi: emit SURFACE_SYNC last | Marek Olšák | 2015-01-07 | 1 | -23/+35 |
| * | radeonsi: flush all CB/DB caches unconditionally when changing the framebuffer | Marek Olšák | 2015-01-07 | 1 | -11/+7 |
| * | radeonsi: change TC cache flushing strategy for textures | Marek Olšák | 2015-01-07 | 2 | -4/+6 |
| * | radeonsi: improve and fix streamout flushing | Marek Olšák | 2015-01-07 | 3 | -10/+40 |
