external/XNNPACK.git - [no description]

Age	Commit message (Expand)	Author
2022-08-03	Remove stale TODO	Zhi An Ng
2022-08-03	Rename table loglut to vlog	Frank Barchard
2022-08-03	Remove xnnpack.h dependency from microkernel test and benchmarks	Zhi An Ng
2022-08-03	Return xnn_status instead of hard coded integers in JIT generators	Zhi An Ng
2022-08-03	Add vld2_r to AArch32 assembler, this will be used later for fused operations.	Zhi An Ng
2022-08-02	apply sort to gemm headers	Frank Barchard
2022-08-02	U32 VLOG microkernel to compute natural log	Frank Barchard
2022-08-02	Change JIT generator to return uint8_t instead of xnn_status to remove depend...	Zhi An Ng
2022-08-02	Add some AArch32 assembler instructions	Zhi An Ng
2022-08-02	Add extended multiplication math functions	Marat Dukhan
2022-08-01	Convert AArch32 and AArch64 F32 GEMM and IGEMM microkernels used for default ...	Zhi An Ng
2022-08-01	Do not redefine GNU_SOURCE if it's already defined by something already in th...	XNNPACK Team
2022-08-01	Fix CMake build	Marat Dukhan
2022-07-29	S16 Front End Microkernel improve naming consistency	Frank Barchard
2022-07-28	CS16 squareabs microkernel for NEON	Frank Barchard
2022-07-28	Remove FP16 includes from all microkernels	Marat Dukhan
2022-07-28	CS16 squareabs microkernel	Frank Barchard
2022-07-27	Rename xnn_q[su]8_addsub_minmax_params to xnn_q[su]8_add_minmax_params	Marat Dukhan
2022-07-27	Refactor declarations of microkernel parameters	Marat Dukhan
2022-07-27	S16 vlshift microkernel	Frank Barchard
2022-07-27	S16 window use post increment	Frank Barchard
2022-07-27	Window microkernel tester add shift value to LOG	Frank Barchard
2022-07-27	Replace FP16 for float<->uint32 bitcasts with internal functions	Marat Dukhan
2022-07-27	Add benchmark::utils::CheckNEON to NEON benchmarks	Frank Barchard
2022-07-27	Fix CMake builds	Zhi An Ng
2022-07-26	Fix Bazel build by adding a new :riscv64 target so that :riscv select will no...	Zhi An Ng
2022-07-26	Rename absmax to maxabs	Frank Barchard
2022-07-26	Use remap to support growing code and weights buffer	Zhi An Ng
2022-07-26	Exclude clang-cl from MSVC-specific implementation of math functions	Marat Dukhan
2022-07-26	Introduce math_clz_nonzero_u32 function	Marat Dukhan
2022-07-26	U32 SQRT evaluation stub for the Hashemian algorithm	Marat Dukhan
2022-07-25	s16 rabsmax C and NEON microkernels	Frank Barchard
2022-07-25	Remove no longer used output_zero_point, output_min, and output_max fields fr...	Zhi An Ng
2022-07-25	Remove no longer used pad_before_channels and pad_after_channels field from x...	Zhi An Ng
2022-07-25	Actually fix Bazel build by removing the entire rule	Zhi An Ng
2022-07-25	Fix Bazel builds	Zhi An Ng
2022-07-25	Add an option to create weights cache specifying the capacity of the underlyi...	Zhi An Ng
2022-07-25	Rename asr_s32/asr_s64 to math_asr_s32/math_asr_s64	Marat Dukhan
2022-07-25	Add debug logging to help debug high memory bandwidth issues	Zhi An Ng
2022-07-25	U32 SQRT evaluation stub for multiplication-/division-free algorithm	Marat Dukhan
2022-07-24	Additional scalar SQRT U32 evaluation stubs	Marat Dukhan
2022-07-22	S16 WINDOW NEON microkernels	Frank Barchard
2022-07-22	Evaluation stubs for U32 SQRT operation in scalar TFLM implementation	Marat Dukhan
2022-07-21	Fix missing includes, stop using gmock due to CXX17 issues	Zhi An Ng
2022-07-21	Add some AArch64 assemblers methods	Zhi An Ng
2022-07-21	Add ld3r to AArch64 assembler	Zhi An Ng
2022-07-20	xnn_s16_window_minmax_ukernel__scalar microkernel	Frank Barchard
2022-07-20	Include <numeric> for std::accumulate	Frank Barchard
2022-07-20	Fix enabling/disabling options in CMake build	Marat Dukhan
2022-07-20	Consistently use op->weights cache instead of passing caches around operators	Zhi An Ng