Merge branch 'official-stockfish:master' into Main

Joachim26 · Jan 17, 2024 · 869f2ba · 869f2ba
2 parents b0910da + c8bc2ce
commit 869f2ba
Show file tree

Hide file tree

Showing 13 changed files with 189 additions and 225 deletions.
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
@@ -19,7 +19,7 @@ making contributions to Stockfish.
 In case you do not have a C++ compiler installed, you can follow the
 instructions from our wiki.
 
-- [Linux][linux-compiling-link]
+- [Ubuntu][ubuntu-compiling-link]
 - [Windows][windows-compiling-link]
 - [macOS][macos-compiling-link]
 
@@ -92,6 +92,6 @@ Thank you for contributing to Stockfish and helping us make it even better!
 [discussions-link]:       https://github.com/official-stockfish/Stockfish/discussions/new
 [creating-my-first-test]: https://github.com/official-stockfish/fishtest/wiki/Creating-my-first-test#create-your-test
 [issue-tracker-link]:     https://github.com/official-stockfish/Stockfish/issues
-[linux-compiling-link]:   https://github.com/official-stockfish/Stockfish/wiki/Compiling-from-source#linux
-[windows-compiling-link]: https://github.com/official-stockfish/Stockfish/wiki/Compiling-from-source#windows
-[macos-compiling-link]:   https://github.com/official-stockfish/Stockfish/wiki/Compiling-from-source#macos
+[ubuntu-compiling-link]:  https://github.com/official-stockfish/Stockfish/wiki/Developers#user-content-installing-a-compiler-1
+[windows-compiling-link]: https://github.com/official-stockfish/Stockfish/wiki/Developers#user-content-installing-a-compiler
+[macos-compiling-link]:   https://github.com/official-stockfish/Stockfish/wiki/Developers#user-content-installing-a-compiler-2
diff --git a/scripts/get_native_properties.sh b/scripts/get_native_properties.sh
@@ -79,7 +79,7 @@ case $uname_s in
       'aarch64')
         file_os='android'
         true_arch='armv8'
-        if check_flags 'dotprod'; then
+        if check_flags 'asimddp'; then
           true_arch="$true_arch-dotprod"
         fi
         ;;

diff --git a/src/bitboard.cpp b/src/bitboard.cpp
@@ -44,15 +44,13 @@ Bitboard BishopTable[0x1480];  // To store bishop attacks
 
 void init_magics(PieceType pt, Bitboard table[], Magic magics[]);
 
-}
-
 // Returns the bitboard of target square for the given step
 // from the given square. If the step is off the board, returns empty bitboard.
-inline Bitboard safe_destination(Square s, int step) {
+Bitboard safe_destination(Square s, int step) {
     Square to = Square(s + step);
     return is_ok(to) && distance(s, to) <= 2 ? square_bb(to) : Bitboard(0);
 }
-
+}
 
 // Returns an ASCII representation of a bitboard suitable
 // to be printed to standard output. Useful for debugging.

diff --git a/src/misc.h b/src/misc.h
@@ -19,12 +19,14 @@
 #ifndef MISC_H_INCLUDED
 #define MISC_H_INCLUDED
 
+#include <algorithm>
 #include <cassert>
 #include <chrono>
 #include <cstddef>
 #include <cstdint>
 #include <iosfwd>
 #include <string>
+#include <vector>
 
 #define stringify2(x) #x
 #define stringify(x) stringify2(x)
@@ -188,6 +190,19 @@ struct CommandLine {
     std::string workingDirectory;  // path of the working directory
 };
 
+namespace Utility {
+
+template<typename T, typename Predicate>
+void move_to_front(std::vector<T>& vec, Predicate pred) {
+    auto it = std::find_if(vec.begin(), vec.end(), pred);
+
+    if (it != vec.end())
+    {
+        std::rotate(vec.begin(), it, it + 1);
+    }
+}
+}
+
 }  // namespace Stockfish
 
 #endif  // #ifndef MISC_H_INCLUDED
diff --git a/src/movepick.cpp b/src/movepick.cpp
@@ -198,15 +198,13 @@ void MovePicker::score() {
                                                : 0;
 
             // malus for putting piece en prise
-            m.value -=
-              !(threatenedPieces & from)
-                ? (pt == QUEEN
-                     ? bool(to & threatenedByRook) * 50000 + bool(to & threatenedByMinor) * 10000
-                   : pt == ROOK
-                     ? bool(to & threatenedByMinor) * 25000 + bool(to & threatenedByPawn) * 10000
-                   : pt != PAWN ? bool(to & threatenedByPawn) * 15000
-                                : 0)
-                : 0;
+            m.value -= !(threatenedPieces & from)
+                       ? (pt == QUEEN ? bool(to & threatenedByRook) * 50000
+                                          + bool(to & threatenedByMinor) * 10000
+                          : pt == ROOK ? bool(to & threatenedByMinor) * 25000
+                          : pt != PAWN ? bool(to & threatenedByPawn) * 15000
+                                       : 0)
+                       : 0;
         }
 
         else  // Type == EVASIONS

diff --git a/src/nnue/layers/affine_transform.h b/src/nnue/layers/affine_transform.h
@@ -200,21 +200,18 @@ class AffineTransform {
         #define vec_setzero _mm512_setzero_si512
         #define vec_set_32 _mm512_set1_epi32
         #define vec_add_dpbusd_32 Simd::m512_add_dpbusd_epi32
-        #define vec_add_dpbusd_32x2 Simd::m512_add_dpbusd_epi32x2
         #define vec_hadd Simd::m512_hadd
     #elif defined(USE_AVX2)
             using vec_t = __m256i;
         #define vec_setzero _mm256_setzero_si256
         #define vec_set_32 _mm256_set1_epi32
         #define vec_add_dpbusd_32 Simd::m256_add_dpbusd_epi32
-        #define vec_add_dpbusd_32x2 Simd::m256_add_dpbusd_epi32x2
         #define vec_hadd Simd::m256_hadd
     #elif defined(USE_SSSE3)
             using vec_t = __m128i;
         #define vec_setzero _mm_setzero_si128
         #define vec_set_32 _mm_set1_epi32
         #define vec_add_dpbusd_32 Simd::m128_add_dpbusd_epi32
-        #define vec_add_dpbusd_32x2 Simd::m128_add_dpbusd_epi32x2
         #define vec_hadd Simd::m128_hadd
     #endif
 
@@ -231,16 +228,14 @@ class AffineTransform {
             for (IndexType k = 0; k < NumRegs; ++k)
                 acc[k] = biasvec[k];
 
-            for (IndexType i = 0; i < NumChunks; i += 2)
+            for (IndexType i = 0; i < NumChunks; ++i)
             {
-                const vec_t in0 = vec_set_32(input32[i + 0]);
-                const vec_t in1 = vec_set_32(input32[i + 1]);
+                const vec_t in0 = vec_set_32(input32[i]);
                 const auto  col0 =
-                  reinterpret_cast<const vec_t*>(&weights[(i + 0) * OutputDimensions * 4]);
-                const auto col1 =
-                  reinterpret_cast<const vec_t*>(&weights[(i + 1) * OutputDimensions * 4]);
+                  reinterpret_cast<const vec_t*>(&weights[i * OutputDimensions * 4]);
+
                 for (IndexType k = 0; k < NumRegs; ++k)
-                    vec_add_dpbusd_32x2(acc[k], in0, col0[k], in1, col1[k]);
+                    vec_add_dpbusd_32(acc[k], in0, col0[k]);
             }
 
             vec_t* outptr = reinterpret_cast<vec_t*>(output);
@@ -250,7 +245,6 @@ class AffineTransform {
     #undef vec_setzero
     #undef vec_set_32
     #undef vec_add_dpbusd_32
-    #undef vec_add_dpbusd_32x2
     #undef vec_hadd
         }
         else if constexpr (OutputDimensions == 1)
@@ -263,14 +257,12 @@ class AffineTransform {
         #define vec_setzero _mm256_setzero_si256
         #define vec_set_32 _mm256_set1_epi32
         #define vec_add_dpbusd_32 Simd::m256_add_dpbusd_epi32
-        #define vec_add_dpbusd_32x2 Simd::m256_add_dpbusd_epi32x2
         #define vec_hadd Simd::m256_hadd
     #elif defined(USE_SSSE3)
             using vec_t = __m128i;
         #define vec_setzero _mm_setzero_si128
         #define vec_set_32 _mm_set1_epi32
         #define vec_add_dpbusd_32 Simd::m128_add_dpbusd_epi32
-        #define vec_add_dpbusd_32x2 Simd::m128_add_dpbusd_epi32x2
         #define vec_hadd Simd::m128_hadd
     #endif
 
@@ -294,7 +286,6 @@ class AffineTransform {
     #undef vec_setzero
     #undef vec_set_32
     #undef vec_add_dpbusd_32
-    #undef vec_add_dpbusd_32x2
     #undef vec_hadd
         }
 #else

diff --git a/src/nnue/layers/simd.h b/src/nnue/layers/simd.h
@@ -87,21 +87,6 @@ m512_hadd128x16_interleave(__m512i sum0, __m512i sum1, __m512i sum2, __m512i sum
     #endif
 }
 
-[[maybe_unused]] static void
-m512_add_dpbusd_epi32x2(__m512i& acc, __m512i a0, __m512i b0, __m512i a1, __m512i b1) {
-
-    #if defined(USE_VNNI)
-    acc = _mm512_dpbusd_epi32(acc, a0, b0);
-    acc = _mm512_dpbusd_epi32(acc, a1, b1);
-    #else
-    __m512i product0 = _mm512_maddubs_epi16(a0, b0);
-    __m512i product1 = _mm512_maddubs_epi16(a1, b1);
-    product0         = _mm512_madd_epi16(product0, _mm512_set1_epi16(1));
-    product1         = _mm512_madd_epi16(product1, _mm512_set1_epi16(1));
-    acc              = _mm512_add_epi32(acc, _mm512_add_epi32(product0, product1));
-    #endif
-}
-
 #endif
 
 #if defined(USE_AVX2)
@@ -124,21 +109,6 @@ m512_add_dpbusd_epi32x2(__m512i& acc, __m512i a0, __m512i b0, __m512i a1, __m512
     #endif
 }
 
-[[maybe_unused]] static void
-m256_add_dpbusd_epi32x2(__m256i& acc, __m256i a0, __m256i b0, __m256i a1, __m256i b1) {
-
-    #if defined(USE_VNNI)
-    acc = _mm256_dpbusd_epi32(acc, a0, b0);
-    acc = _mm256_dpbusd_epi32(acc, a1, b1);
-    #else
-    __m256i product0 = _mm256_maddubs_epi16(a0, b0);
-    __m256i product1 = _mm256_maddubs_epi16(a1, b1);
-    product0         = _mm256_madd_epi16(product0, _mm256_set1_epi16(1));
-    product1         = _mm256_madd_epi16(product1, _mm256_set1_epi16(1));
-    acc              = _mm256_add_epi32(acc, _mm256_add_epi32(product0, product1));
-    #endif
-}
-
 #endif
 
 #if defined(USE_SSSE3)
@@ -156,27 +126,10 @@ m256_add_dpbusd_epi32x2(__m256i& acc, __m256i a0, __m256i b0, __m256i a1, __m256
     acc              = _mm_add_epi32(acc, product0);
 }
 
-[[maybe_unused]] static void
-m128_add_dpbusd_epi32x2(__m128i& acc, __m128i a0, __m128i b0, __m128i a1, __m128i b1) {
-
-    __m128i product0 = _mm_maddubs_epi16(a0, b0);
-    __m128i product1 = _mm_maddubs_epi16(a1, b1);
-    product0         = _mm_madd_epi16(product0, _mm_set1_epi16(1));
-    product1         = _mm_madd_epi16(product1, _mm_set1_epi16(1));
-    acc              = _mm_add_epi32(acc, _mm_add_epi32(product0, product1));
-}
-
 #endif
 
 #if defined(USE_NEON_DOTPROD)
 
-[[maybe_unused]] static void dotprod_m128_add_dpbusd_epi32x2(
-  int32x4_t& acc, int8x16_t a0, int8x16_t b0, int8x16_t a1, int8x16_t b1) {
-
-    acc = vdotq_s32(acc, a0, b0);
-    acc = vdotq_s32(acc, a1, b1);
-}
-
 [[maybe_unused]] static void
 dotprod_m128_add_dpbusd_epi32(int32x4_t& acc, int8x16_t a, int8x16_t b) {
 
@@ -198,13 +151,6 @@ dotprod_m128_add_dpbusd_epi32(int32x4_t& acc, int8x16_t a, int8x16_t b) {
     return neon_m128_reduce_add_epi32(sum) + bias;
 }
 
-[[maybe_unused]] static void
-neon_m128_add_dpbusd_epi32x2(int32x4_t& acc, int8x8_t a0, int8x8_t b0, int8x8_t a1, int8x8_t b1) {
-
-    int16x8_t product = vmull_s8(a0, b0);
-    product           = vmlal_s8(product, a1, b1);
-    acc               = vpadalq_s16(acc, product);
-}
 #endif
 
 #if USE_NEON >= 8