Add SIMD NEON Optimization for QT_FP16 in Scalar Quantizer (#3166)

Summary: ### Description We have SIMD support for x86 architecture using AVX2 optimization. But, we don't have a similar optimization for ARM architecture in Scalar Quantizer for the Quantization Type `QT_FP16`. This PR adds SIMD support for ARM using NEON optimization. ### Issues resolved #3014 Pull Request resolved: #3166 Reviewed By: algoriddle, pemazare Differential Revision: D52510486 Pulled By: mdouze fbshipit-source-id: 2bb360475a0d9e0816c8e05b44d7da7f2e6b28c5
facebookresearch · Jan 11, 2024 · c3aa526 · c3aa526
1 parent 32f0e8c
commit c3aa526
Show file tree

Hide file tree

Showing 4 changed files with 362 additions and 3 deletions.
diff --git a/faiss/CMakeLists.txt b/faiss/CMakeLists.txt
@@ -189,6 +189,7 @@ set(FAISS_HEADERS
   utils/extra_distances.h
   utils/fp16-fp16c.h
   utils/fp16-inl.h
+  utils/fp16-arm.h
   utils/fp16.h
   utils/hamming-inl.h
   utils/hamming.h