| Commit message (Expand) | Author | Age | Files | Lines |
* | radeon/llvm: Add support for v4f32 stores on R600 | Tom Stellard | 2012-09-21 | 3 | -9/+27 |
* | radeon/llvm: Add support for i8 reads on R600 | Tom Stellard | 2012-09-21 | 3 | -0/+25 |
* | radeon/llvm: Expand vector fadd and fmul on R600 | Tom Stellard | 2012-09-21 | 1 | -0/+3 |
* | radeon/llvm: Add optimization for FP_ROUND | Tom Stellard | 2012-09-21 | 2 | -0/+27 |
* | radeon/llvm: Replace AMDGPU pow intrinsic with the llvm version | Tom Stellard | 2012-09-21 | 4 | -7/+26 |
* | llvmpipe: fix overflow bug in total texture size computation | Brian Paul | 2012-09-20 | 1 | -2/+16 |
* | r600g/llvm: rs780/rs880 are r600 asics | Alex Deucher | 2012-09-20 | 1 | -2/+2 |
* | r300/compiler: Use precomputed q values in the register allocator | Tom Stellard | 2012-09-19 | 1 | -1/+69 |
* | r300g: Init regalloc state during context creation | Tom Stellard | 2012-09-19 | 8 | -155/+204 |
* | r300/compiler: Don't create register classes for inputs | Tom Stellard | 2012-09-19 | 1 | -14/+1 |
* | ra: Add q_values parameter to ra_set_finalize() | Tom Stellard | 2012-09-19 | 1 | -1/+1 |
* | r600g: Invalidate texture cache when creating vertex buffers for compute v2 | Tom Stellard | 2012-09-19 | 1 | -1/+3 |
* | r600g: Use LOOP_START_DX10 for loops | Tom Stellard | 2012-09-19 | 3 | -2/+11 |
* | r600g: Set the correct value of COLOR*_DIM for RATs | Tom Stellard | 2012-09-19 | 1 | -2/+2 |
* | r600g: Make sure to initialize DB_DEPTH_CONTROL register for compute | Tom Stellard | 2012-09-19 | 1 | -1/+3 |
* | r600g: Add some comments and debug printfs to compute code | Tom Stellard | 2012-09-19 | 2 | -5/+53 |
* | r600g: Add missing break to case statement | Tom Stellard | 2012-09-19 | 1 | -0/+1 |
* | radeon/llvm: Emit ISA for ALU instructions in the R600 code emitter | Michal Sciubidlo | 2012-09-19 | 10 | -167/+359 |
* | radeon/llvm: Only support 512 constant registers on R600 | Tom Stellard | 2012-09-19 | 1 | -1/+1 |
* | radeon/llvm: Add a fdiv pattern. | Vincent Lejeune | 2012-09-18 | 1 | -3/+10 |
* | radeon/llvm: reserve also corresponding 128bits reg | Vincent Lejeune | 2012-09-18 | 1 | -0/+1 |
* | softpipe: implement the new can_create_resource() function | Brian Paul | 2012-09-17 | 2 | -5/+29 |
* | llvmpipe: implement the new can_create_resource() function | Brian Paul | 2012-09-17 | 1 | -5/+23 |
* | llvmpipe: set max cube texture size to 4K x 4K | Brian Paul | 2012-09-17 | 2 | -1/+2 |
* | radeon/llvm: Inital flow control support for SI | Tom Stellard | 2012-09-17 | 7 | -2/+168 |
* | r600g: Close a memory leak of llvm byte streams | Xinya Zhang | 2012-09-17 | 1 | -0/+1 |
* | radeon/llvm: Fix unused variable warning | Tom Stellard | 2012-09-17 | 1 | -1/+0 |
* | radeon/llvm: Move kernel arg lowering into R600TargetLowering class | Tom Stellard | 2012-09-17 | 6 | -470/+35 |
* | radeon/llvm: Match integer add/sub for SI. | Michel Dänzer | 2012-09-17 | 1 | -2/+8 |
* | radeon/llvm: Complete integer comparison patterns for SI. | Michel Dänzer | 2012-09-17 | 1 | -4/+12 |
* | radeon/llvm: Match AMDGPUfract on SI. | Michel Dänzer | 2012-09-17 | 1 | -1/+3 |
* | radeon/llvm: Match int_AMDGPU_floor for SI. | Michel Dänzer | 2012-09-17 | 1 | -1/+3 |
* | radeon/llvm: Match vector logical operations on SI. | Michel Dänzer | 2012-09-17 | 1 | -3/+9 |
* | softpipe: update SP_MAX_TEXTURE_3D_LEVELS comment | Brian Paul | 2012-09-16 | 1 | -1/+1 |
* | radeon/llvm: Support frint on SI | Christian König | 2012-09-14 | 1 | -1/+3 |
* | r600g: consolidate decompression code for the blitter | Marek Olšák | 2012-09-14 | 1 | -29/+37 |
* | r600g: do not require MSAA renderbuffer support if not asked for | Marek Olšák | 2012-09-14 | 2 | -16/+4 |
* | r300g: fix colormask with non-BGRA formats | Marek Olšák | 2012-09-14 | 4 | -31/+160 |
* | r600g: reduce quant mode on evergreen+ | Alex Deucher | 2012-09-13 | 1 | -1/+1 |
* | radeonsi: don't use a staging resource for large transfers | Marek Olšák | 2012-09-13 | 1 | -10/+0 |
* | r600g: don't use a staging resource for large transfers | Marek Olšák | 2012-09-13 | 1 | -10/+0 |
* | r600g: convert the remnants of VGT state into immediate register writes/atoms v4 | Marek Olšák | 2012-09-13 | 8 | -57/+65 |
* | r600g: emit the primitive type and associated regs only if the type is changed | Marek Olšák | 2012-09-13 | 5 | -48/+38 |
* | r600g: add clip_misc_state for clip registers emitted in draw_vbo | Marek Olšák | 2012-09-13 | 8 | -22/+44 |
* | r600g: fix computing how much space is needed for a draw command | Marek Olšák | 2012-09-13 | 2 | -6/+12 |
* | r600g: fix the number of CS dwords of cb_misc_state | Marek Olšák | 2012-09-13 | 2 | -2/+2 |
* | r600g: atomize clip state | Marek Olšák | 2012-09-13 | 6 | -148/+38 |
* | r600g: atomize blend color | Marek Olšák | 2012-09-13 | 6 | -27/+25 |
* | r600g: atomize viewport state | Marek Olšák | 2012-09-13 | 7 | -40/+28 |
* | r600g: atomize stencil ref state | Marek Olšák | 2012-09-13 | 7 | -51/+56 |