Add Stockfish namespace.

[stockfish] / src / nnue / layers / affine_transform.h

diff --git a/src/nnue/layers/affine_transform.h b/src/nnue/layers/affine_transform.h
index ab2beab7168ebe2e92d6e6be560903eccab1a055..d2713c5aaf6078a556cb0b953fa819f37968f87d 100644 (file)
--- a/src/nnue/layers/affine_transform.h
+++ b/src/nnue/layers/affine_transform.h
@@ -1,6 +1,6 @@
 /*
   Stockfish, a UCI chess playing engine derived from Glaurung 2.1
-  Copyright (C) 2004-2020 The Stockfish developers (see AUTHORS file)
+  Copyright (C) 2004-2021 The Stockfish developers (see AUTHORS file)
 
   Stockfish is free software: you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
@@ -24,7 +24,7 @@
 #include <iostream>
 #include "../nnue_common.h"
 
-namespace Eval::NNUE::Layers {
+namespace Stockfish::Eval::NNUE::Layers {
 
   // Affine transformation layer
   template <typename PreviousLayer, IndexType OutputDimensions>
@@ -301,20 +301,40 @@ namespace Eval::NNUE::Layers {
       }
       else if constexpr (kOutputDimensions == 1)
       {
-          constexpr IndexType kNumChunks = kPaddedInputDimensions / kSimdWidth;
-
-          vec_t sum0 = vec_setzero();
-
-          const auto row0 = reinterpret_cast<const vec_t*>(&weights_[0]);
-
-          for (int j = 0; j < (int)kNumChunks; ++j)
+#if defined (USE_AVX512)
+          if constexpr (kPaddedInputDimensions % (kSimdWidth * 2) != 0)
           {
-              const vec_t in = input_vector[j];
-
-              vec_add_dpbusd_32(sum0, in, row0[j]);
+              constexpr IndexType kNumChunks = kPaddedInputDimensions / kSimdWidth;
+              const auto input_vector256 = reinterpret_cast<const __m256i*>(input);
+
+              __m256i sum0 = _mm256_setzero_si256();
+              const auto row0 = reinterpret_cast<const __m256i*>(&weights_[0]);
+
+              for (int j = 0; j < (int)kNumChunks; ++j)
+              {
+                  const __m256i in = input_vector256[j];
+                  m256_add_dpbusd_epi32(sum0, in, row0[j]);
+              }
+              output[0] = m256_hadd(sum0, biases_[0]);
+          }
+          else
+#endif
+          {
+#if defined (USE_AVX512)
+              constexpr IndexType kNumChunks = kPaddedInputDimensions / (kSimdWidth * 2);
+#else
+              constexpr IndexType kNumChunks = kPaddedInputDimensions / kSimdWidth;
+#endif
+              vec_t sum0 = vec_setzero();
+              const auto row0 = reinterpret_cast<const vec_t*>(&weights_[0]);
+
+              for (int j = 0; j < (int)kNumChunks; ++j)
+              {
+                  const vec_t in = input_vector[j];
+                  vec_add_dpbusd_32(sum0, in, row0[j]);
+              }
+              output[0] = vec_hadd(sum0, biases_[0]);
           }
-
-          output[0] = vec_hadd(sum0, biases_[0]);
       }
 
 #else
@@ -439,6 +459,6 @@ namespace Eval::NNUE::Layers {
 #endif
   };
 
-}  // namespace Eval::NNUE::Layers
+}  // namespace Stockfish::Eval::NNUE::Layers
 
 #endif // #ifndef NNUE_LAYERS_AFFINE_TRANSFORM_H_INCLUDED