Gitiles
Code Review Sign In
gerrit-public.fairphone.software / platform / external / XNNPACK / refs/tags/rel/13/fp3/6.A.018.0 / src / qc8-dwconv / gen / up16x25-minmax-fp32-neon-mla8-ld128.c
  1. 7be427a Disable MSan and TSan in most microkernels with Out-of-Bounds reads by Marat Dukhan · 2 years, 6 months ago
  2. 03efa0f Reoptimize FP32 requantization in NEON QS8/QC8/QU8 GEMM/IGEMM/DWCONV by Marat Dukhan · 2 years, 7 months ago
  3. 5a31dc6 Optimize FP32 requantization in NEON QS8/QC8/QU8 GEMM/IGEMM/DWCONV by Marat Dukhan · 2 years, 7 months ago
  4. 7988a18 Refactoring xnn_qs8_minmax_params for NEON/NEONv8 by Marat Dukhan · 2 years, 7 months ago
  5. 5f7cf55 Avoid using gcc-specific intrinsics in NEON microkernels by Marat Dukhan · 2 years, 7 months ago
  6. 5f2939f QS8/QC8 DWCONV NEON MUL8/MLA8 microkernels using 128-bit loads by Marat Dukhan · 3 years ago[Copied (65%) from src/qc8-dwconv/gen/up16x25-minmax-fp32-neon-mla8.c]
  7. caccd8e Accumulate in 16 bits once in NEON QS8/QC8 DWCONV before extending to 32 bits by Marat Dukhan · 3 years ago