| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2009-11-03 | llvmpipe: Sanitise shader semantic and interpolator handling. | Michal Krol | |
| Handle the remaining semantic names and indices. Respect color interpolator when not flatshading. Based on Michal's softpipe commit eb699d64ec7057032139baccedcb0694ca41d706. | |||
| 2009-11-03 | llvmpipe: remove old prim_setup draw stage | Keith Whitwell | |
| Everything now goes through the draw_vbuf handler, the same as regular drivers. Based on Keith's commit 4fe0fc3eba1f79beda890a5016359d549bab6ad4. | |||
| 2009-11-03 | llvmpipe: Respect gl_rasterization_rules in primitive setup. | José Fonseca | |
| Based on Michal's identical commit for softpipe (ca9c413647bf9efb5ed770e3a655bc758075aec7). | |||
| 2009-10-28 | llvmpipe: Make sure the JIT engine and X86 target are linked on MSVC build. | José Fonseca | |
| Basically mimic the llvm 2.6 way of linking execution engines and targets. | |||
| 2009-10-28 | llvmpipe: Implement round() for MSVC. | José Fonseca | |
| 2009-10-25 | llvmpipe: Human friendlier sampler state dump. | José Fonseca | |
| 2009-10-25 | llvmpipe: Dump the sampler state of the shader key. | José Fonseca | |
| 2009-10-25 | llvmpipe: Immediate multiplication. | José Fonseca | |
| 2009-10-25 | llvmpipe: Fast path for sampling rgba8 textures with linear filtering. | José Fonseca | |
| Implement Keith's suggestion of doing most of the sampling with 16x8 and 8x16 AoS, and only doing the conversion to floating point SoA at the very last step. Improves gloss performance by 10%. | |||
| 2009-10-25 | llvmpipe: Make lerping work for 8.8 fixed point values. | José Fonseca | |
| 2009-10-25 | llvmpipe: New module to help make assertions about formats. | José Fonseca | |
| 2009-10-25 | llvmpipe: New function to unpack rgba8 formats into 4 x u8n AoS. | José Fonseca | |
| 2009-10-25 | llvmpipe: Split the format swizzling step from the unpacking. | José Fonseca | |
| 2009-10-25 | llvmpipe: Allow different signs when unpacking. | José Fonseca | |
| 2009-10-25 | llvmpipe: Add inlines to quickly generate types matching the native SIMD ↵ | José Fonseca | |
| register bitwidth. | |||
| 2009-10-25 | llvmpipe: Merge lp_build_load_rgba_soa into lp_build_sample_texel. | José Fonseca | |
| 2009-10-25 | llvmpipe: Factor our pixel offset computation. | José Fonseca | |
| 2009-10-25 | llvmpipe: Share testing infrastructure with lp_test_format. | José Fonseca | |
| 2009-10-25 | llvmpipe: Eliminate lp_build_store_rgba_aos. | José Fonseca | |
| 2009-10-25 | llvmpipe: Eliminate lp_build_load_rgba_aos. | José Fonseca | |
| 2009-10-25 | llvmpipe: Move a few format/sampling functions into better space. | José Fonseca | |
| 2009-10-23 | gallium: remove the swizzling parts of ExtSwizzle | Keith Whitwell | |
| These haven't been used by the mesa state tracker since the conversion to tgsi_ureg, and it seems that none of the other state trackers are using it either. This helps simplify one of the biggest suprises when starting off with TGSI shaders. | |||
| 2009-10-23 | gallium: remove noise opcodes | Keith Whitwell | |
| Provide a dummy implementation in the GL state tracker (move 0.5 to the destination regs). At some point, a motivated person could add a better implementation of noise. Currently not even the nvidia binary drivers do anything more than this. In any case, the place to do this is in the GL state tracker, not the poor driver. | |||
| 2009-10-22 | llvmpipe: Avoid yet another variable size array. | José Fonseca | |
| 2009-10-22 | llvmpipe: Define rdtsc for MSVC. | José Fonseca | |
| 2009-10-22 | llvmpipe: Avoid variable size arrays. | José Fonseca | |
| Not really variable size, but MSVC still doesn't like them. | |||
| 2009-10-22 | llvmpipe: Use the pack/unpack functions for 8bit unsigned norm multiplication. | José Fonseca | |
| 2009-10-22 | llvmpipe: Utility function to double the bit width of a type. | José Fonseca | |
| 2009-10-22 | llvmpipe: Factor vector packing/unpacking to a separate source file. | José Fonseca | |
| These functions will be needed to implement many of the 8bit operations, and they are quite complex on its own. | |||
| 2009-10-22 | llvmpipe: Call util_cpu_detect() from the unit tests. | José Fonseca | |
| 2009-10-21 | llvmpipe: Remove extraneous name in lp_type pre-declaration. | José Fonseca | |
| 2009-10-18 | llvmpipe: Allocate texture storage for whole quads. | José Fonseca | |
| 2009-10-14 | llvmpipe: Use ALIGN_STACK. | José Fonseca | |
| 2009-10-09 | llvmpipe: Eliminate constant mapping/unmapping. | José Fonseca | |
| 2009-10-04 | llvmpipe: Ensure tile cache transfers are mapped before flushing it. | José Fonseca | |
| 2009-10-04 | llvmpipe: Autogenerate lp_tile_soa.c from u_format.csv. | José Fonseca | |
| This is just a temporary change until we code generate the tile read/write functions in runtime. The new code avoids an extra memcpy that exists in u_tile.c functions, from which lp_tile_soa.c was originally based. This achieves up to 5% improvement, particularly in frames with little geometry overlap. | |||
| 2009-10-04 | llvmpipe: Remove loop testing from format testing. | José Fonseca | |
| Loop building will be rewritten. | |||
| 2009-10-04 | llvmpipe: Match header's protection macro with filename. | José Fonseca | |
| 2009-10-04 | llvmpipe: Adjust format assertion. | José Fonseca | |
| We support array layout too -- if it has a single channel. | |||
| 2009-10-03 | llvmpipe: Fetch tile only if a color buffer is bound. | José Fonseca | |
| 2009-10-02 | gallium: Preparations for adding more PIPE_TRANSFER_* usage flags. | Michel Dänzer | |
| Always test for PIPE_TRANSFER_READ/WRITE using the bit-wise and operator, and add a pipe_transfer_buffer_flags() helper for getting the buffer usage flags corresponding to them. | |||
| 2009-10-01 | gallium: remove depth.occlusion_count flag | Keith Whitwell | |
| This was redundant as drivers can just keep track of whether they are inside a begin/end query pair. We want to add more query types later and also support nested queries, none of which map well onto a flag like this. No driver appeared to be using the flag. | |||
| 2009-09-29 | llvmpipe: Workaround for bug in llvm 2.5. | José Fonseca | |
| The combination of fptosi and sitofp (necessary for trunc/floor/ceil/round implementation) somehow becomes invalid code. Skip the instruction combining pass when SSE4.1 is not available. | |||
| 2009-09-29 | llvmpipe: First verify LLVM IR, only then run optimizing passes. | José Fonseca | |
| 2009-09-29 | llvmpipe: Runtime cpu checks for lp_build_min_simple too. | José Fonseca | |
| 2009-09-29 | llvmpipe: Implement non SSE4.1 versions of floor and round. | José Fonseca | |
| 2009-09-29 | llvmpipe: Emit SSE intrinsics based on runtime cpu capability check. | José Fonseca | |
| Note that llvmpipe still doesn't run on any processor yet: if you don't have a recent processor with SSE4.1 you will still likely end up hitting a code path for which a generic non-sse4 version is not implemented yet. | |||
| 2009-09-28 | llvmpipe: Grab a ref when the fb is set. | Younes Manton | |
| Based on softpipe commit a77226071f6814a53358a5d6caff685889d0e4ec. | |||
| 2009-09-26 | llvmpipe: Allow building with LLVM 2.6 too. | José Fonseca | |
| 2009-09-24 | llvmpipe: add missing __FUNCTION__ parameter to debug_printf() calls | Brian Paul | |
 |
index : android-x86-mesa.git | |
| Androïd/x86 port of Mesa drivers | Hugues Hiegel |
| summaryrefslogtreecommitdiff |