summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* llvmpipe: Optimize away min/max with equal operands.José Fonseca2009-08-291-0/+14
|
* llvmpipe: Improve generated 8bit blending code.José Fonseca2009-08-291-0/+29
|
* llvmpipe: Get blending of normalized 8bit unsigned integers working.José Fonseca2009-08-292-115/+347
|
* llvmpipe: Specialize arithmetic operations.José Fonseca2009-08-294-244/+420
|
* llvmpipe: Introduce a custom typing system.José Fonseca2009-08-295-42/+228
| | | | | Straightforward representation of floating-point/fixed-point/integer, normalized/scaled, signed/unsigned SIMD vector types.
* llvmpipe: Avoid C99-isms.José Fonseca2009-08-292-23/+21
|
* llvmpipe: Factor out intrisic calling code into a reusable function.José Fonseca2009-08-291-40/+31
|
* llvmpipe: Cleanup test programs.José Fonseca2009-08-293-63/+83
|
* llvmpipe: Blending.José Fonseca2009-08-295-2/+965
| | | | The code
* llvmpipe: Arithmetic helpers.José Fonseca2009-08-294-0/+420
|
* llvmpipe: fastpath for interpolated z16 less depthtestingKeith Whitwell2009-08-291-4/+139
| | | | | | | Because this is interpolated (ie. early) depth, we can build in an assumption about the quads emitted by triangle setup, ie that they are actually linear spans. Interpolate z over those spans in z16 format to save on math & conversion.
* llvmpipe: cope with nr_cbufs == 0Keith Whitwell2009-08-291-3/+13
| | | | Disable blend code when no color buffer
* llvmpipe: Translate logicops.José Fonseca2009-08-294-0/+115
|
* llvmpipe: Separate pixel packing/unpacking from loading/storing.José Fonseca2009-08-298-44/+185
|
* xlib: Integrate with llvmpipe.José Fonseca2009-08-291-0/+513
| | | | Actually, the makefile/sconscript changes were already commit by mistake.
* llvmpipe: Adjust for pixel layout changes.José Fonseca2009-08-293-2/+4
|
* util: Make pixel format layout more meaningful.José Fonseca2009-08-293-109/+114
| | | | | Namelly, explicitly distinguish formats coded by arithmetic manipulation, from formats layed out in array.
* util: Don't commit u_format_table.cJosé Fonseca2009-08-292-1925/+1
| | | | At least not until stabilizes, to reduce noise.
* llvmpipe: Fix typo in copyright.José Fonseca2009-08-295-5/+5
|
* llvmpipe: Add test cases for pixels.José Fonseca2009-08-291-42/+95
| | | | Actually some of this test cases need more testing...
* util: Fix r5g6b5 description.José Fonseca2009-08-293-10/+13
|
* llvmpipe: Update SConscript.José Fonseca2009-08-291-5/+0
|
* llvmpipe: Simple shell script to aid port softpipe changes to llvmpipe.José Fonseca2009-08-291-0/+28
|
* llvmpipe: move all depth/stencil/alpha pixel processing into one stageKeith Whitwell2009-08-2913-957/+580
|
* llvmpipe: fix error in scissor state dependenciesKeith Whitwell2009-08-291-2/+10
|
* llvmpipe: cleanup framebuffer state routine slightlyKeith Whitwell2009-08-291-40/+10
|
* llvmpipe: move all color-combine code into lp_quad_blend.cKeith Whitwell2009-08-2910-972/+727
| | | | | Consolidate the read-modify-write color combining code from the blend, colormask and output stages.
* llvmpipe: fix typoKeith Whitwell2009-08-291-1/+1
|
* llvmpipe: example fastpaths in blendingKeith Whitwell2009-08-291-11/+121
|
* llvmpipe: actually pass >1 quad from triangle routineKeith Whitwell2009-08-298-139/+135
| | | | First attempt
* llvmpipe: expand quad pipeline to process >1 quad at a timeKeith Whitwell2009-08-2914-633/+745
| | | | | | This is part one -- we still only pass a single quad down, but the code can now cope with more. The quads must all be from the same tile.
* llvmpipe: rip out old mulithread supportKeith Whitwell2009-08-294-305/+57
|
* llvmpipe: avoid flushing depth buffer cache on swapbuffersKeith Whitwell2009-08-292-9/+16
| | | | | | | | | There's no need to push out depth buffer contents on swapbuffers. Note that this change doesn't throw away depth buffer changes, it simply holds them in the cache over calls to swapbuffers. The hope is that swapbuffers will be followed by a clear() which means in that case we won't have to write the changes out.
* llvmpipe: also shortcircuit non-texture tile lookupsKeith Whitwell2009-08-292-18/+31
|
* llvmpipe: shortcircuit repeated lookups of the same tileKeith Whitwell2009-08-293-97/+156
| | | | | | | | | The lp_tile_cache is often called repeatedly to look up the same tile. Add a cache (to the cache) of the single tile most recently retreived and make a quick inline check to see if this matches the subsequent request. Add a tile_address bitfield struct to make this check easier.
* llvmpipe: remove unused vars in lp_setup.cKeith Whitwell2009-08-291-5/+0
|
* llvmpipe: use bitwise logic to setup quad masks in lp_setupKeith Whitwell2009-08-291-25/+40
|
* llvmpipe: simplify flush_spansKeith Whitwell2009-08-291-53/+19
| | | | No loss of performance, but simpler code.
* llvmpipe: make some small steps to flush texture cache less frequentlyKeith Whitwell2009-08-296-35/+70
| | | | No performance gain yet, but the code is a bit cleaner.
* llvmpipe: remove backwards dependency from tilecache to llvmpipeKeith Whitwell2009-08-2914-65/+62
| | | | | | | | The tile cache is a utility, it shouldn't know anything about the entity which is making use of it (ie llvmpipe). Remove llvmpipe parameter to all the tilecache function calls, and also remove the need to keep a llvmpipe pointer in the sampler structs.
* llvmpipe: Rename preprocessor symbols too.José Fonseca2009-08-2930-113/+113
|
* llvmpipe: Pixel packing/unpacking and loop code generators.José Fonseca2009-08-297-1/+689
| | | | | Just a small proof of concept plus a standalone test app. Not integrated with the rest of the driver yet.
* util: Pixel format database.José Fonseca2009-08-297-0/+2397
| | | | | | | There are some inconsistencies in pipe_format, but above all, there simply aren't enough bits in an enum to conveniently store all information about a pixel format we need to be able to dynamically generate pixel packing/unpacking code.
* llvmpipe: Fork softpipe for experimentation with llvm.José Fonseca2009-08-2956-0/+11080
|
* Put 'extern' first on the line to silence GCC warnings.Ian Romanick2009-08-281-2/+2
|
* i965: Use VBOs in the VBO module on 965, now that we have ARB_map_buffer_range.Eric Anholt2009-08-281-0/+2
| | | | This looks like it's a small win on blender.
* vbo: Fix array pointer calculation for MapBufferRange-mapped vertex data.Eric Anholt2009-08-281-1/+3
| | | | | We would end up with the offset from the start of the mapping rather than the offset from the start of the buffer.
* intel: Add support for GL_ARB_map_buffer_range.Eric Anholt2009-08-285-35/+189
| | | | | Passes glean's bufferObject test, and should provide good performance in the cases applications are expected to use.
* i965: Implement ARB_oq CheckQuery in the intended way.Eric Anholt2009-08-281-6/+1
| | | | | Previously we blocked because I hadn't added the libdrm function. Now it's there, so update your libdrm.
* radeon: Fix swtcl emit pediction.Pauli Nieminen2009-08-294-45/+70
| | | | | | Problem was to find the correct place to run prediction. Only place that is called for every primitive is ALLOC_VERTS so we have to do prediction there before allocation.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 22:30:57 +0000