| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | tgsi: remove redundant CND0 opcode | Keith Whitwell | 2009-09-01 | 1 | -4/+0 |
| | | | | | Can be implemented with CMP src2, src1, src0 | ||||
| * | scons: Tool for LLVM. Gracefully disable llvmpipe if LLVM not found. | José Fonseca | 2009-09-01 | 1 | -3/+5 |
| | | |||||
| * | llvmpipe: Fallback to element-wise comparisons when no comparison intrinsic ↵ | José Fonseca | 2009-08-31 | 1 | -2/+45 |
| | | | | | | | | | | | | | is available. Although selection of vector elements is valid LLVM IR, no machine target supports it yet. This is a last-resort option, but it allows llvmpipe to be used on any target supported by LLVM without modifications. Obviously better performance is attainable by emitting SIMD intrinsics where otherwise LLVM doesn't. | ||||
| * | llvmpipe: In the abcense of a min/max intrinsic, use the comparison intrinsics. | José Fonseca | 2009-08-31 | 3 | -11/+7 |
| | | |||||
| * | llvmpipe: Actually commit lp_buffer.[ch]. | José Fonseca | 2009-08-31 | 2 | -0/+205 |
| | | | | | Thanks Michel. | ||||
| * | llvmpipe: Texture cache in 4 ubytes instead of 4 floats. | José Fonseca | 2009-08-30 | 3 | -19/+19 |
| | | | | | | | | | | This is more a short term experiment than a long term commitment, as we'll need to support higher precision textures too, as this will all be be replaced by runtime generated code. With this change most Mesa demos fps increased around 10%. Not a huge improvement, but not a negligible one either. | ||||
| * | llvmpipe: Use util_format_read_4f to read texture tiles. | José Fonseca | 2009-08-30 | 1 | -6/+17 |
| | | |||||
| * | llvmpipe: Texture cache is read-only. No need to flush. | José Fonseca | 2009-08-30 | 3 | -48/+0 |
| | | |||||
| * | llvmpipe: Eliminate internal pipe_buffer_map/unmap usage. | José Fonseca | 2009-08-30 | 1 | -9/+3 |
| | | |||||
| * | llvmpipe: List lp_buffer.c in the Makefile. | José Fonseca | 2009-08-29 | 1 | -0/+1 |
| | | |||||
| * | llvmpipe: Define an winsys for LLVM. Drop pipe_winsys | José Fonseca | 2009-08-29 | 7 | -79/+196 |
| | | | | | | | | | lp_winsys will eventually be unified with softpipe's eventually, but we are free to move quicker since we don't have the myriad of users yet. Will provide a pipe_winsys adaptor from Keith's softpipe-private-winsys soon. | ||||
| * | llvmpipe: Update for changes in master. | José Fonseca | 2009-08-29 | 3 | -12/+10 |
| | | |||||
| * | llvmpipe: Isolate sampling from TGSI translation. | José Fonseca | 2009-08-29 | 5 | -94/+145 |
| | | |||||
| * | llvmpipe: Better structure dereference naming. | José Fonseca | 2009-08-29 | 2 | -6/+11 |
| | | |||||
| * | llvmpipe: Fix shader variant key construction. | José Fonseca | 2009-08-29 | 1 | -2/+2 |
| | | | | | Fixes the blank screen on non-64bit mode. | ||||
| * | llvmpipe: Fix build. | José Fonseca | 2009-08-29 | 2 | -2/+2 |
| | | |||||
| * | llvmpipe: Merge all pixel format related files. | José Fonseca | 2009-08-29 | 6 | -255/+153 |
| | | |||||
| * | llvmpipe: Factor out and optimize the input interpolation. | José Fonseca | 2009-08-29 | 7 | -169/+521 |
| | | | | | | | | Special attention is given to the interpolation of side by side quads. Multiplications are made only for the first quad. Interpolation of inputs for posterior quads are done exclusively with additions, and perspective divide if necessary. | ||||
| * | llvmpipe: Pre-declare fetch_texel. | José Fonseca | 2009-08-29 | 5 | -28/+66 |
| | | |||||
| * | llvmpipe: Pass the alpha ref value and blend color in the jit context. | José Fonseca | 2009-08-29 | 8 | -50/+86 |
| | | |||||
| * | llvmpipe: Pass fragment context to generated function in a single structure. | José Fonseca | 2009-08-29 | 8 | -33/+108 |
| | | |||||
| * | llvmpipe: Structure manipulation helpers. | José Fonseca | 2009-08-29 | 4 | -0/+119 |
| | | |||||
| * | llvmpipe: Centralize the C <-> JIT interfaces in one place. | José Fonseca | 2009-08-29 | 7 | -43/+153 |
| | | |||||
| * | llvmpipe: Drop blend derived state. | José Fonseca | 2009-08-29 | 4 | -161/+6 |
| | | | | | Already included in the fragment shader. | ||||
| * | llvmpipe: keep texture stride 16byte aligned | Keith Whitwell | 2009-08-29 | 1 | -1/+1 |
| | | | | | Fixes crash when resizing gears window. | ||||
| * | llvmpipe: Squash the quad pipeline. | José Fonseca | 2009-08-29 | 10 | -351/+120 |
| | | | | | It had been reduced to one fixed stage. | ||||
| * | llvmpipe: s/uni/scalar/. | José Fonseca | 2009-08-29 | 9 | -47/+47 |
| | | | | | More obvious name. | ||||
| * | llvmpipe: Add a bunch of comments. | José Fonseca | 2009-08-29 | 17 | -70/+346 |
| | | | | | Description/rationale/to-do items, while I still remember them... | ||||
| * | llvmpipe: Split off vs stuff from lp_state_fs.c. | José Fonseca | 2009-08-29 | 4 | -60/+98 |
| | | | | | lp_state_fs.c is already too big without it. | ||||
| * | llvmpipe: Avoid duplicate clamping when converting between types. | José Fonseca | 2009-08-29 | 1 | -8/+15 |
| | | |||||
| * | llvmpipe: Generate the fragment pipeline into a single function. | José Fonseca | 2009-08-29 | 13 | -339/+327 |
| | | | | | Still hackish. Will document and optimize later. | ||||
| * | llvmpipe: Get conversions working correctly for all integer types. | José Fonseca | 2009-08-29 | 5 | -83/+389 |
| | | |||||
| * | llvmpipe: Delete the quad polygon stipple stage. | José Fonseca | 2009-08-29 | 3 | -87/+0 |
| | | | | | Not used now -- stipple done by the draw module. May code generate later. | ||||
| * | llvmpipe: Early depth testing. | José Fonseca | 2009-08-29 | 11 | -46/+166 |
| | | |||||
| * | llvmpipe: Disassemble beyond referred code beyond the RET instruction. | José Fonseca | 2009-08-29 | 1 | -2/+32 |
| | | |||||
| * | llvmpipe: Split control flow function declarations and notes. | José Fonseca | 2009-08-29 | 11 | -38/+91 |
| | | |||||
| * | llvmpipe: Back port recent softpipe-opt improvements from Keith. | José Fonseca | 2009-08-29 | 6 | -142/+466 |
| | | |||||
| * | llvmpipe: Don't limit the number of tiles in the cache. | José Fonseca | 2009-08-29 | 3 | -239/+96 |
| | | |||||
| * | llvmpipe: Clean dead tile cache code. | José Fonseca | 2009-08-29 | 2 | -28/+0 |
| | | |||||
| * | llvmpipe: Simplify tile clears. Use pipe_fill_rect instead of scratch tile. | José Fonseca | 2009-08-29 | 3 | -66/+39 |
| | | |||||
| * | llvmpipe: Start hiding llvmpipe_cached_tile. | José Fonseca | 2009-08-29 | 3 | -8/+8 |
| | | |||||
| * | llvmpipe: Drop depth/stencil support from tile cache. | José Fonseca | 2009-08-29 | 2 | -83/+18 |
| | | |||||
| * | llvmpipe: Code generate the depth test, and include in the shader. | José Fonseca | 2009-08-29 | 14 | -869/+172 |
| | | | | | | Only 32bit depth/stencil surfaces supported for now. Stencil ops not implemented yet. | ||||
| * | llvmpipe: Update texture block from format. | José Fonseca | 2009-08-29 | 1 | -0/+3 |
| | | |||||
| * | llvmpipe: Depth test to LLVM IR translation. | José Fonseca | 2009-08-29 | 4 | -2/+248 |
| | | |||||
| * | llvmpipe: Centralize mask update logic. | José Fonseca | 2009-08-29 | 7 | -34/+43 |
| | | |||||
| * | llvmpipe: Clamped float to unsigned norm via mantissa manipulation. | José Fonseca | 2009-08-29 | 3 | -17/+113 |
| | | |||||
| * | llvmpipe: Several fixes of the conversion test. | José Fonseca | 2009-08-29 | 4 | -34/+55 |
| | | |||||
| * | llvmpipe: Generate instrinsics for integer comparisons. | José Fonseca | 2009-08-29 | 1 | -0/+71 |
| | | | | | | It is a little messy, given the available instrucions form SIMD integer comparisons is rather limited. | ||||
| * | llvmpipe: Mention the ability to get microbenchmarks from the unit tests. | José Fonseca | 2009-08-29 | 1 | -0/+4 |
| | | | | | Even better would be to have a -h/--help option to the tests... | ||||
 |
index : mesa.git | |
| Unnamed repository; edit this file 'description' to name the repository. |
| summaryrefslogtreecommitdiffstats |