Commit graph

25525 commits

Author SHA1 Message Date
José Fonseca
36249348ed llvmpipe: More intrinsic helpers. 2009-08-29 09:21:29 +01:00
José Fonseca
50d77141e8 llvmpipe: Fix typo. 2009-08-29 09:21:29 +01:00
José Fonseca
81644f6315 util: Fix SCons build. 2009-08-29 09:21:28 +01:00
José Fonseca
19b31d0ac8 llvmpipe: Document the typo in llvm-c/Core.h. 2009-08-29 09:21:28 +01:00
José Fonseca
fb56786f60 util: Auto-generate u_format_table.c from SConscript. 2009-08-29 09:21:28 +01:00
José Fonseca
9285f159e0 llvmpipe: Simple description/instructions. 2009-08-29 09:21:28 +01:00
José Fonseca
8edcc8abe3 llvmpipe: Add missing include. 2009-08-29 09:21:27 +01:00
José Fonseca
2101c2941c llvmpipe: Tiles in rgba8 format. 2009-08-29 09:21:27 +01:00
José Fonseca
3ce1abf950 llvmpipe: Store tile color in SoA. 2009-08-29 09:21:27 +01:00
José Fonseca
29d94a9cbb llvmpipe: Code generate color masking. 2009-08-29 09:21:27 +01:00
José Fonseca
2d6b39f05e llvmpipe: Use the generated SoA blending code. 2009-08-29 09:21:27 +01:00
José Fonseca
e7fc21434d llvmpipe: Handle disabled blending too. 2009-08-29 09:21:27 +01:00
José Fonseca
62c91998ef xlib: Ensure one screen instance. 2009-08-29 09:21:27 +01:00
José Fonseca
0318f3e53e llvmpipe: Split the texture cache from the color/depth/stencil cache. 2009-08-29 09:21:27 +01:00
José Fonseca
b836b2593c xlib: Complete llvmpipe integration. 2009-08-29 09:21:27 +01:00
José Fonseca
2529ed5616 llvmpipe: SoA blending.
Throughput seems to be 4x higher.
2009-08-29 09:21:27 +01:00
José Fonseca
20f50b845b util: Fix blend factor dumping. 2009-08-29 09:21:26 +01:00
José Fonseca
3130b9921c llvmpipe: Update blending test. 2009-08-29 09:21:26 +01:00
José Fonseca
60584affcf llvmpipe: Handle more conversions.
Not all successfully yet though..
2009-08-29 09:21:26 +01:00
José Fonseca
145de19548 llvmpipe: Use full set of sse4 min/max intrinsics. 2009-08-29 09:21:26 +01:00
José Fonseca
a9771d2b75 llvmpipe: Fix one const generation for some signed integers. 2009-08-29 09:21:26 +01:00
José Fonseca
8244d6e5ad llvmpipe: Allow floating types without sign.
With the meaning that all values are assumed to be positive.
2009-08-29 09:21:26 +01:00
José Fonseca
9493260fdc llvmpipe: Centralize lp_build_context initialization. 2009-08-29 09:21:25 +01:00
José Fonseca
b07d19a885 llvmpipe: Normalize the cycles with the number of channel.
So that we have a comparable number for different formats.
2009-08-29 09:21:25 +01:00
José Fonseca
b874a7b807 llvmpipe: Handle ubyte -> float conversion too. 2009-08-29 09:21:25 +01:00
José Fonseca
b441a0b658 llvmpipe: Make lp_const_offset visible. 2009-08-29 09:21:24 +01:00
José Fonseca
684a47f0f6 llvmpipe: Only get the preprocessor flags from llvm-config.
Otherwise we get -O2 optimization flag, preventing proper debugging.
2009-08-29 09:21:24 +01:00
José Fonseca
8988424ee8 llvmpipe: Bootstrap type conversions. 2009-08-29 09:21:24 +01:00
José Fonseca
b19cb0080c llvmpipe: Use same type for reference vectors. 2009-08-29 09:21:24 +01:00
José Fonseca
d52dce0ffb llvmpipe: Fix floating point const scale factor. 2009-08-29 09:21:24 +01:00
José Fonseca
e6ebebc485 llvmpipe: Factor out shared test code into a separate module. 2009-08-29 09:21:24 +01:00
José Fonseca
138428bade llvmpipe: Cleanup constant helpers. 2009-08-29 09:21:23 +01:00
José Fonseca
627d6a6b04 llvmpipe: Move intrinsic helpers to a separate module. 2009-08-29 09:21:23 +01:00
José Fonseca
1165c30dc2 llvmpipe: Collect richer blend data. 2009-08-29 09:21:23 +01:00
José Fonseca
28e46458bc llvmpipe: Some notes about PSHUF. 2009-08-29 09:21:23 +01:00
José Fonseca
1dd7bb17c7 llvmpipe: Optimize blend swizzles by using bitmasks instead of shuffles for ubytes. 2009-08-29 09:21:23 +01:00
José Fonseca
a77084ea4b llvmpipe: Write data to a tsv file for posterior analysis. 2009-08-29 09:21:23 +01:00
José Fonseca
f081bacc25 llvmpipe: Move p_build_context to lp_bld_type.h
As it will be shared with more modules.
2009-08-29 09:21:23 +01:00
José Fonseca
696f7f2be5 util: New file dedicated to dump state in human/machine readable format. 2009-08-29 09:21:23 +01:00
José Fonseca
5940ba26f7 llvmpipe: Don't use llvm.readcyclecounter.
LLVM during optimization reorders the rdtsc instructions next to each
other, which makes it pretty useless.
2009-08-29 09:21:22 +01:00
José Fonseca
64611e086d llvmpipe: Separate constant building. 2009-08-29 09:21:22 +01:00
José Fonseca
c87fab0008 llvmpipe: Move type support functions into a separate file. 2009-08-29 09:21:22 +01:00
José Fonseca
b94e22e4be llvmpipe: Measure the number of cycles taken for blending. 2009-08-29 09:21:22 +01:00
José Fonseca
5778970f15 llvmpipe: Minor comments / debug mnemonics. 2009-08-29 09:21:22 +01:00
José Fonseca
02eaa32d83 llvmpipe: Optimize away min/max with equal operands. 2009-08-29 09:21:22 +01:00
José Fonseca
9e92244440 llvmpipe: Improve generated 8bit blending code. 2009-08-29 09:21:22 +01:00
José Fonseca
ede73258a7 llvmpipe: Get blending of normalized 8bit unsigned integers working. 2009-08-29 09:21:22 +01:00
José Fonseca
a6622e6c54 llvmpipe: Specialize arithmetic operations. 2009-08-29 09:21:22 +01:00
José Fonseca
272dadbe4e llvmpipe: Introduce a custom typing system.
Straightforward representation of floating-point/fixed-point/integer,
normalized/scaled, signed/unsigned SIMD vector types.
2009-08-29 09:21:22 +01:00
José Fonseca
f478b6fe76 llvmpipe: Avoid C99-isms. 2009-08-29 09:21:21 +01:00