Update NNUE architecture to SFNNv5. Update network to nn-3c0aa92af1da.nnue.

[stockfish] / src / nnue / nnue_architecture.h

diff --git a/src/nnue/nnue_architecture.h b/src/nnue/nnue_architecture.h
index 725b40fb43d2e87d9ab484603528f9965a06f59d..cac8373075e368b37e3c72493490c7e23920afcc 100644 (file)
--- a/src/nnue/nnue_architecture.h
+++ b/src/nnue/nnue_architecture.h
@@ -21,12 +21,15 @@
 #ifndef NNUE_ARCHITECTURE_H_INCLUDED
 #define NNUE_ARCHITECTURE_H_INCLUDED
 
+#include <memory>
+
 #include "nnue_common.h"
 
 #include "features/half_ka_v2_hm.h"
 
 #include "layers/affine_transform.h"
 #include "layers/clipped_relu.h"
+#include "layers/sqr_clipped_relu.h"
 
 #include "../misc.h"
 
@@ -46,8 +49,9 @@ struct Network
   static constexpr int FC_1_OUTPUTS = 32;
 
   Layers::AffineTransform<TransformedFeatureDimensions, FC_0_OUTPUTS + 1> fc_0;
-  Layers::ClippedReLU<FC_0_OUTPUTS> ac_0;
-  Layers::AffineTransform<FC_0_OUTPUTS, FC_1_OUTPUTS> fc_1;
+  Layers::SqrClippedReLU<FC_0_OUTPUTS + 1> ac_sqr_0;
+  Layers::ClippedReLU<FC_0_OUTPUTS + 1> ac_0;
+  Layers::AffineTransform<FC_0_OUTPUTS * 2, FC_1_OUTPUTS> fc_1;
   Layers::ClippedReLU<FC_1_OUTPUTS> ac_1;
   Layers::AffineTransform<FC_1_OUTPUTS, 1> fc_2;
 
@@ -88,28 +92,35 @@ struct Network
 
   std::int32_t propagate(const TransformedFeatureType* transformedFeatures)
   {
-    constexpr uint64_t alignment = CacheLineSize;
-
-    struct Buffer
+    struct alignas(CacheLineSize) Buffer
     {
       alignas(CacheLineSize) decltype(fc_0)::OutputBuffer fc_0_out;
+      alignas(CacheLineSize) decltype(ac_sqr_0)::OutputType ac_sqr_0_out[ceil_to_multiple<IndexType>(FC_0_OUTPUTS * 2, 32)];
       alignas(CacheLineSize) decltype(ac_0)::OutputBuffer ac_0_out;
       alignas(CacheLineSize) decltype(fc_1)::OutputBuffer fc_1_out;
       alignas(CacheLineSize) decltype(ac_1)::OutputBuffer ac_1_out;
       alignas(CacheLineSize) decltype(fc_2)::OutputBuffer fc_2_out;
+
+      Buffer()
+      {
+          std::memset(this, 0, sizeof(*this));
+      }
     };
 
-#if defined(ALIGNAS_ON_STACK_VARIABLES_BROKEN)
-    char bufferRaw[sizeof(Buffer) + alignment];
-    char* bufferRawAligned = align_ptr_up<alignment>(&bufferRaw[0]);
-    Buffer& buffer = *(new (bufferRawAligned) Buffer);
+#if defined(__clang__) && (__APPLE__)
+    // workaround for a bug reported with xcode 12
+    static thread_local auto tlsBuffer = std::make_unique<Buffer>();
+    // Access TLS only once, cache result.
+    Buffer& buffer = *tlsBuffer;
 #else
-    alignas(alignment) Buffer buffer;
+    alignas(CacheLineSize) static thread_local Buffer buffer;
 #endif
 
     fc_0.propagate(transformedFeatures, buffer.fc_0_out);
+    ac_sqr_0.propagate(buffer.fc_0_out, buffer.ac_sqr_0_out);
     ac_0.propagate(buffer.fc_0_out, buffer.ac_0_out);
-    fc_1.propagate(buffer.ac_0_out, buffer.fc_1_out);
+    std::memcpy(buffer.ac_sqr_0_out + FC_0_OUTPUTS, buffer.ac_0_out, FC_0_OUTPUTS * sizeof(decltype(ac_0)::OutputType));
+    fc_1.propagate(buffer.ac_sqr_0_out, buffer.fc_1_out);
     ac_1.propagate(buffer.fc_1_out, buffer.ac_1_out);
     fc_2.propagate(buffer.ac_1_out, buffer.fc_2_out);
 
@@ -118,10 +129,6 @@ struct Network
     std::int32_t fwdOut = int(buffer.fc_0_out[FC_0_OUTPUTS]) * (600*OutputScale) / (127*(1<<WeightScaleBits));
     std::int32_t outputValue = buffer.fc_2_out[0] + fwdOut;
 
-#if defined(ALIGNAS_ON_STACK_VARIABLES_BROKEN)
-    buffer.~Buffer();
-#endif
-
     return outputValue;
   }
 };