Remove handcrafted MMX code

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

diff --git a/src/nnue/layers/clipped_relu.h b/src/nnue/layers/clipped_relu.h
index 65455df4944324a12870ca29d68c9ff5e0b379b1..48cd6c6934547a06852a248b6ad936733f8a0faa 100644 (file)
--- a/src/nnue/layers/clipped_relu.h
+++ b/src/nnue/layers/clipped_relu.h
@@ -1,6 +1,6 @@
 /*
   Stockfish, a UCI chess playing engine derived from Glaurung 2.1
-  Copyright (C) 2004-2021 The Stockfish developers (see AUTHORS file)
+  Copyright (C) 2004-2023 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
@@ -21,55 +21,50 @@
 #ifndef NNUE_LAYERS_CLIPPED_RELU_H_INCLUDED
 #define NNUE_LAYERS_CLIPPED_RELU_H_INCLUDED
 
+#include <algorithm>
+#include <cstdint>
+#include <iosfwd>
+
 #include "../nnue_common.h"
 
 namespace Stockfish::Eval::NNUE::Layers {
 
   // Clipped ReLU
-  template <typename PreviousLayer>
+  template <IndexType InDims>
   class ClippedReLU {
    public:
     // Input/output type
-    using InputType = typename PreviousLayer::OutputType;
+    using InputType = std::int32_t;
     using OutputType = std::uint8_t;
-    static_assert(std::is_same<InputType, std::int32_t>::value, "");
 
     // Number of input/output dimensions
-    static constexpr IndexType InputDimensions =
-        PreviousLayer::OutputDimensions;
+    static constexpr IndexType InputDimensions = InDims;
     static constexpr IndexType OutputDimensions = InputDimensions;
+    static constexpr IndexType PaddedOutputDimensions =
+        ceil_to_multiple<IndexType>(OutputDimensions, 32);
 
-    // Size of forward propagation buffer used in this layer
-    static constexpr std::size_t SelfBufferSize =
-        ceil_to_multiple(OutputDimensions * sizeof(OutputType), CacheLineSize);
-
-    // Size of the forward propagation buffer used from the input layer to this layer
-    static constexpr std::size_t BufferSize =
-        PreviousLayer::BufferSize + SelfBufferSize;
+    using OutputBuffer = OutputType[PaddedOutputDimensions];
 
     // Hash value embedded in the evaluation file
-    static constexpr std::uint32_t get_hash_value() {
+    static constexpr std::uint32_t get_hash_value(std::uint32_t prevHash) {
       std::uint32_t hashValue = 0x538D24C7u;
-      hashValue += PreviousLayer::get_hash_value();
+      hashValue += prevHash;
       return hashValue;
     }
 
     // Read network parameters
-    bool read_parameters(std::istream& stream) {
-      return previousLayer.read_parameters(stream);
+    bool read_parameters(std::istream&) {
+      return true;
     }
 
     // Write network parameters
-    bool write_parameters(std::ostream& stream) const {
-      return previousLayer.write_parameters(stream);
+    bool write_parameters(std::ostream&) const {
+      return true;
     }
 
     // Forward propagation
-    const OutputType* propagate(
-        const TransformedFeatureType* transformedFeatures, char* buffer) const {
-      const auto input = previousLayer.propagate(
-          transformedFeatures, buffer + SelfBufferSize);
-      const auto output = reinterpret_cast<OutputType*>(buffer);
+    void propagate(
+        const InputType* input, OutputType* output) const {
 
   #if defined(USE_AVX2)
       if constexpr (InputDimensions % SimdWidth == 0) {
@@ -140,24 +135,6 @@ namespace Stockfish::Eval::NNUE::Layers {
       }
       constexpr IndexType Start = NumChunks * SimdWidth;
 
-  #elif defined(USE_MMX)
-      constexpr IndexType NumChunks = InputDimensions / SimdWidth;
-      const __m64 k0x80s = _mm_set1_pi8(-128);
-      const auto in = reinterpret_cast<const __m64*>(input);
-      const auto out = reinterpret_cast<__m64*>(output);
-      for (IndexType i = 0; i < NumChunks; ++i) {
-        const __m64 words0 = _mm_srai_pi16(
-            _mm_packs_pi32(in[i * 4 + 0], in[i * 4 + 1]),
-            WeightScaleBits);
-        const __m64 words1 = _mm_srai_pi16(
-            _mm_packs_pi32(in[i * 4 + 2], in[i * 4 + 3]),
-            WeightScaleBits);
-        const __m64 packedbytes = _mm_packs_pi16(words0, words1);
-        out[i] = _mm_subs_pi8(_mm_adds_pi8(packedbytes, k0x80s), k0x80s);
-      }
-      _mm_empty();
-      constexpr IndexType Start = NumChunks * SimdWidth;
-
   #elif defined(USE_NEON)
       constexpr IndexType NumChunks = InputDimensions / (SimdWidth / 2);
       const int8x8_t Zero = {0};
@@ -177,13 +154,9 @@ namespace Stockfish::Eval::NNUE::Layers {
 
       for (IndexType i = Start; i < InputDimensions; ++i) {
         output[i] = static_cast<OutputType>(
-            std::max(0, std::min(127, input[i] >> WeightScaleBits)));
+            std::clamp(input[i] >> WeightScaleBits, 0, 127));
       }
-      return output;
     }
-
-   private:
-    PreviousLayer previousLayer;
   };
 
 }  // namespace Stockfish::Eval::NNUE::Layers