/*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1
- Copyright (C) 2004-2021 The Stockfish developers (see AUTHORS file)
+ Copyright (C) 2004-2022 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
namespace Stockfish::Eval::NNUE {
+ using BiasType = std::int16_t;
+ using WeightType = std::int16_t;
+ using PSQTWeightType = std::int32_t;
+
// If vector instructions are enabled, we update and refresh the
// accumulator tile by tile such that each tile fits in the CPU's
// vector registers.
#define VECTOR
- static_assert(PSQTBuckets == 8, "Assumed by the current choice of constants.");
+ static_assert(PSQTBuckets % 8 == 0,
+ "Per feature PSQT values cannot be processed at granularity lower than 8 at a time.");
#ifdef USE_AVX512
typedef __m512i vec_t;
#define vec_add_psqt_32(a,b) _mm256_add_epi32(a,b)
#define vec_sub_psqt_32(a,b) _mm256_sub_epi32(a,b)
#define vec_zero_psqt() _mm256_setzero_si256()
- static constexpr IndexType NumRegs = 8; // only 8 are needed
- static constexpr IndexType NumPsqtRegs = 1;
+ #define NumRegistersSIMD 32
#elif USE_AVX2
typedef __m256i vec_t;
#define vec_add_psqt_32(a,b) _mm256_add_epi32(a,b)
#define vec_sub_psqt_32(a,b) _mm256_sub_epi32(a,b)
#define vec_zero_psqt() _mm256_setzero_si256()
- static constexpr IndexType NumRegs = 16;
- static constexpr IndexType NumPsqtRegs = 1;
+ #define NumRegistersSIMD 16
#elif USE_SSE2
typedef __m128i vec_t;
#define vec_add_psqt_32(a,b) _mm_add_epi32(a,b)
#define vec_sub_psqt_32(a,b) _mm_sub_epi32(a,b)
#define vec_zero_psqt() _mm_setzero_si128()
- static constexpr IndexType NumRegs = Is64Bit ? 16 : 8;
- static constexpr IndexType NumPsqtRegs = 2;
+ #define NumRegistersSIMD (Is64Bit ? 16 : 8)
#elif USE_MMX
typedef __m64 vec_t;
#define vec_add_psqt_32(a,b) _mm_add_pi32(a,b)
#define vec_sub_psqt_32(a,b) _mm_sub_pi32(a,b)
#define vec_zero_psqt() _mm_setzero_si64()
- static constexpr IndexType NumRegs = 8;
- static constexpr IndexType NumPsqtRegs = 4;
+ #define NumRegistersSIMD 8
#elif USE_NEON
typedef int16x8_t vec_t;
#define vec_add_psqt_32(a,b) vaddq_s32(a,b)
#define vec_sub_psqt_32(a,b) vsubq_s32(a,b)
#define vec_zero_psqt() psqt_vec_t{0}
- static constexpr IndexType NumRegs = 16;
- static constexpr IndexType NumPsqtRegs = 2;
+ #define NumRegistersSIMD 16
#else
#undef VECTOR
#endif
+
+ #ifdef VECTOR
+
+ // Compute optimal SIMD register count for feature transformer accumulation.
+
+ // We use __m* types as template arguments, which causes GCC to emit warnings
+ // about losing some attribute information. This is irrelevant to us as we
+ // only take their size, so the following pragma are harmless.
+ #pragma GCC diagnostic push
+ #pragma GCC diagnostic ignored "-Wignored-attributes"
+
+ template <typename SIMDRegisterType,
+ typename LaneType,
+ int NumLanes,
+ int MaxRegisters>
+ static constexpr int BestRegisterCount()
+ {
+ #define RegisterSize sizeof(SIMDRegisterType)
+ #define LaneSize sizeof(LaneType)
+
+ static_assert(RegisterSize >= LaneSize);
+ static_assert(MaxRegisters <= NumRegistersSIMD);
+ static_assert(MaxRegisters > 0);
+ static_assert(NumRegistersSIMD > 0);
+ static_assert(RegisterSize % LaneSize == 0);
+ static_assert((NumLanes * LaneSize) % RegisterSize == 0);
+
+ const int ideal = (NumLanes * LaneSize) / RegisterSize;
+ if (ideal <= MaxRegisters)
+ return ideal;
+
+ // Look for the largest divisor of the ideal register count that is smaller than MaxRegisters
+ for (int divisor = MaxRegisters; divisor > 1; --divisor)
+ if (ideal % divisor == 0)
+ return divisor;
+
+ return 1;
+ }
+
+ static constexpr int NumRegs = BestRegisterCount<vec_t, WeightType, TransformedFeatureDimensions, NumRegistersSIMD>();
+ static constexpr int NumPsqtRegs = BestRegisterCount<psqt_vec_t, PSQTWeightType, PSQTBuckets, NumRegistersSIMD>();
+
+ #pragma GCC diagnostic pop
+
+ #endif
+
+
+
// Input feature converter
class FeatureTransformer {
// Number of input/output dimensions
static constexpr IndexType InputDimensions = FeatureSet::Dimensions;
- static constexpr IndexType OutputDimensions = HalfDimensions * 2;
+ static constexpr IndexType OutputDimensions = HalfDimensions;
// Size of forward propagation buffer
static constexpr std::size_t BufferSize =
// Hash value embedded in the evaluation file
static constexpr std::uint32_t get_hash_value() {
- return FeatureSet::HashValue ^ OutputDimensions;
+ return FeatureSet::HashValue ^ (OutputDimensions * 2);
}
// Read network parameters
) / 2;
- #if defined(USE_AVX512)
-
- constexpr IndexType NumChunks = HalfDimensions / (SimdWidth * 2);
- static_assert(HalfDimensions % (SimdWidth * 2) == 0);
- const __m512i Control = _mm512_setr_epi64(0, 2, 4, 6, 1, 3, 5, 7);
- const __m512i Zero = _mm512_setzero_si512();
-
for (IndexType p = 0; p < 2; ++p)
{
- const IndexType offset = HalfDimensions * p;
- auto out = reinterpret_cast<__m512i*>(&output[offset]);
- for (IndexType j = 0; j < NumChunks; ++j)
- {
- __m512i sum0 = _mm512_load_si512(&reinterpret_cast<const __m512i*>
- (accumulation[perspectives[p]])[j * 2 + 0]);
- __m512i sum1 = _mm512_load_si512(&reinterpret_cast<const __m512i*>
- (accumulation[perspectives[p]])[j * 2 + 1]);
+ const IndexType offset = (HalfDimensions / 2) * p;
- _mm512_store_si512(&out[j], _mm512_permutexvar_epi64(Control,
- _mm512_max_epi8(_mm512_packs_epi16(sum0, sum1), Zero)));
- }
- }
- return psqt;
+#if defined(USE_AVX512)
- #elif defined(USE_AVX2)
+ constexpr IndexType OutputChunkSize = 512 / 8;
+ static_assert((HalfDimensions / 2) % OutputChunkSize == 0);
+ constexpr IndexType NumOutputChunks = HalfDimensions / 2 / OutputChunkSize;
- constexpr IndexType NumChunks = HalfDimensions / SimdWidth;
- constexpr int Control = 0b11011000;
- const __m256i Zero = _mm256_setzero_si256();
+ const __m512i Zero = _mm512_setzero_si512();
+ const __m512i One = _mm512_set1_epi16(127);
+ const __m512i Control = _mm512_setr_epi64(0, 2, 4, 6, 1, 3, 5, 7);
- for (IndexType p = 0; p < 2; ++p)
- {
- const IndexType offset = HalfDimensions * p;
- auto out = reinterpret_cast<__m256i*>(&output[offset]);
- for (IndexType j = 0; j < NumChunks; ++j)
+ const __m512i* in0 = reinterpret_cast<const __m512i*>(&(accumulation[perspectives[p]][0]));
+ const __m512i* in1 = reinterpret_cast<const __m512i*>(&(accumulation[perspectives[p]][HalfDimensions / 2]));
+ __m512i* out = reinterpret_cast< __m512i*>(output + offset);
+
+ for (IndexType j = 0; j < NumOutputChunks; j += 1)
{
- __m256i sum0 = _mm256_load_si256(&reinterpret_cast<const __m256i*>
- (accumulation[perspectives[p]])[j * 2 + 0]);
- __m256i sum1 = _mm256_load_si256(&reinterpret_cast<const __m256i*>
- (accumulation[perspectives[p]])[j * 2 + 1]);
+ const __m512i sum0a = _mm512_max_epi16(_mm512_min_epi16(in0[j * 2 + 0], One), Zero);
+ const __m512i sum0b = _mm512_max_epi16(_mm512_min_epi16(in0[j * 2 + 1], One), Zero);
+ const __m512i sum1a = _mm512_max_epi16(_mm512_min_epi16(in1[j * 2 + 0], One), Zero);
+ const __m512i sum1b = _mm512_max_epi16(_mm512_min_epi16(in1[j * 2 + 1], One), Zero);
- _mm256_store_si256(&out[j], _mm256_permute4x64_epi64(
- _mm256_max_epi8(_mm256_packs_epi16(sum0, sum1), Zero), Control));
+ const __m512i pa = _mm512_srli_epi16(_mm512_mullo_epi16(sum0a, sum1a), 7);
+ const __m512i pb = _mm512_srli_epi16(_mm512_mullo_epi16(sum0b, sum1b), 7);
+
+ out[j] = _mm512_permutexvar_epi64(Control, _mm512_packs_epi16(pa, pb));
}
- }
- return psqt;
- #elif defined(USE_SSE2)
+#elif defined(USE_AVX2)
- #ifdef USE_SSE41
- constexpr IndexType NumChunks = HalfDimensions / SimdWidth;
- const __m128i Zero = _mm_setzero_si128();
- #else
- constexpr IndexType NumChunks = HalfDimensions / SimdWidth;
- const __m128i k0x80s = _mm_set1_epi8(-128);
- #endif
+ constexpr IndexType OutputChunkSize = 256 / 8;
+ static_assert((HalfDimensions / 2) % OutputChunkSize == 0);
+ constexpr IndexType NumOutputChunks = HalfDimensions / 2 / OutputChunkSize;
- for (IndexType p = 0; p < 2; ++p)
- {
- const IndexType offset = HalfDimensions * p;
- auto out = reinterpret_cast<__m128i*>(&output[offset]);
- for (IndexType j = 0; j < NumChunks; ++j)
+ const __m256i Zero = _mm256_setzero_si256();
+ const __m256i One = _mm256_set1_epi16(127);
+ constexpr int Control = 0b11011000;
+
+ const __m256i* in0 = reinterpret_cast<const __m256i*>(&(accumulation[perspectives[p]][0]));
+ const __m256i* in1 = reinterpret_cast<const __m256i*>(&(accumulation[perspectives[p]][HalfDimensions / 2]));
+ __m256i* out = reinterpret_cast< __m256i*>(output + offset);
+
+ for (IndexType j = 0; j < NumOutputChunks; j += 1)
{
- __m128i sum0 = _mm_load_si128(&reinterpret_cast<const __m128i*>
- (accumulation[perspectives[p]])[j * 2 + 0]);
- __m128i sum1 = _mm_load_si128(&reinterpret_cast<const __m128i*>
- (accumulation[perspectives[p]])[j * 2 + 1]);
- const __m128i packedbytes = _mm_packs_epi16(sum0, sum1);
-
- #ifdef USE_SSE41
- _mm_store_si128(&out[j], _mm_max_epi8(packedbytes, Zero));
- #else
- _mm_store_si128(&out[j], _mm_subs_epi8(_mm_adds_epi8(packedbytes, k0x80s), k0x80s));
- #endif
+ const __m256i sum0a = _mm256_max_epi16(_mm256_min_epi16(in0[j * 2 + 0], One), Zero);
+ const __m256i sum0b = _mm256_max_epi16(_mm256_min_epi16(in0[j * 2 + 1], One), Zero);
+ const __m256i sum1a = _mm256_max_epi16(_mm256_min_epi16(in1[j * 2 + 0], One), Zero);
+ const __m256i sum1b = _mm256_max_epi16(_mm256_min_epi16(in1[j * 2 + 1], One), Zero);
+
+ const __m256i pa = _mm256_srli_epi16(_mm256_mullo_epi16(sum0a, sum1a), 7);
+ const __m256i pb = _mm256_srli_epi16(_mm256_mullo_epi16(sum0b, sum1b), 7);
+
+ out[j] = _mm256_permute4x64_epi64(_mm256_packs_epi16(pa, pb), Control);
}
- }
- return psqt;
- #elif defined(USE_MMX)
+#elif defined(USE_SSE2)
- constexpr IndexType NumChunks = HalfDimensions / SimdWidth;
- const __m64 k0x80s = _mm_set1_pi8(-128);
+ constexpr IndexType OutputChunkSize = 128 / 8;
+ static_assert((HalfDimensions / 2) % OutputChunkSize == 0);
+ constexpr IndexType NumOutputChunks = HalfDimensions / 2 / OutputChunkSize;
- for (IndexType p = 0; p < 2; ++p)
- {
- const IndexType offset = HalfDimensions * p;
- auto out = reinterpret_cast<__m64*>(&output[offset]);
- for (IndexType j = 0; j < NumChunks; ++j)
+ const __m128i Zero = _mm_setzero_si128();
+ const __m128i One = _mm_set1_epi16(127);
+
+ const __m128i* in0 = reinterpret_cast<const __m128i*>(&(accumulation[perspectives[p]][0]));
+ const __m128i* in1 = reinterpret_cast<const __m128i*>(&(accumulation[perspectives[p]][HalfDimensions / 2]));
+ __m128i* out = reinterpret_cast< __m128i*>(output + offset);
+
+ for (IndexType j = 0; j < NumOutputChunks; j += 1)
{
- __m64 sum0 = *(&reinterpret_cast<const __m64*>(accumulation[perspectives[p]])[j * 2 + 0]);
- __m64 sum1 = *(&reinterpret_cast<const __m64*>(accumulation[perspectives[p]])[j * 2 + 1]);
- const __m64 packedbytes = _mm_packs_pi16(sum0, sum1);
- out[j] = _mm_subs_pi8(_mm_adds_pi8(packedbytes, k0x80s), k0x80s);
+ const __m128i sum0a = _mm_max_epi16(_mm_min_epi16(in0[j * 2 + 0], One), Zero);
+ const __m128i sum0b = _mm_max_epi16(_mm_min_epi16(in0[j * 2 + 1], One), Zero);
+ const __m128i sum1a = _mm_max_epi16(_mm_min_epi16(in1[j * 2 + 0], One), Zero);
+ const __m128i sum1b = _mm_max_epi16(_mm_min_epi16(in1[j * 2 + 1], One), Zero);
+
+ const __m128i pa = _mm_srli_epi16(_mm_mullo_epi16(sum0a, sum1a), 7);
+ const __m128i pb = _mm_srli_epi16(_mm_mullo_epi16(sum0b, sum1b), 7);
+
+ out[j] = _mm_packs_epi16(pa, pb);
}
- }
- _mm_empty();
- return psqt;
- #elif defined(USE_NEON)
+#elif defined(USE_NEON)
- constexpr IndexType NumChunks = HalfDimensions / (SimdWidth / 2);
- const int8x8_t Zero = {0};
+ constexpr IndexType OutputChunkSize = 128 / 8;
+ static_assert((HalfDimensions / 2) % OutputChunkSize == 0);
+ constexpr IndexType NumOutputChunks = HalfDimensions / 2 / OutputChunkSize;
- for (IndexType p = 0; p < 2; ++p)
- {
- const IndexType offset = HalfDimensions * p;
- const auto out = reinterpret_cast<int8x8_t*>(&output[offset]);
- for (IndexType j = 0; j < NumChunks; ++j)
+ const int16x8_t Zero = vdupq_n_s16(0);
+ const int16x8_t One = vdupq_n_s16(127);
+
+ const int16x8_t* in0 = reinterpret_cast<const int16x8_t*>(&(accumulation[perspectives[p]][0]));
+ const int16x8_t* in1 = reinterpret_cast<const int16x8_t*>(&(accumulation[perspectives[p]][HalfDimensions / 2]));
+ int8x16_t* out = reinterpret_cast< int8x16_t*>(output + offset);
+
+ for (IndexType j = 0; j < NumOutputChunks; j += 1)
{
- int16x8_t sum = reinterpret_cast<const int16x8_t*>(accumulation[perspectives[p]])[j];
- out[j] = vmax_s8(vqmovn_s16(sum), Zero);
+ const int16x8_t sum0a = vmaxq_s16(vminq_s16(in0[j * 2 + 0], One), Zero);
+ const int16x8_t sum0b = vmaxq_s16(vminq_s16(in0[j * 2 + 1], One), Zero);
+ const int16x8_t sum1a = vmaxq_s16(vminq_s16(in1[j * 2 + 0], One), Zero);
+ const int16x8_t sum1b = vmaxq_s16(vminq_s16(in1[j * 2 + 1], One), Zero);
+
+ const int8x8_t pa = vshrn_n_s16(vmulq_s16(sum0a, sum1a), 7);
+ const int8x8_t pb = vshrn_n_s16(vmulq_s16(sum0b, sum1b), 7);
+
+ out[j] = vcombine_s8(pa, pb);
}
- }
- return psqt;
- #else
+#else
- for (IndexType p = 0; p < 2; ++p)
- {
- const IndexType offset = HalfDimensions * p;
- for (IndexType j = 0; j < HalfDimensions; ++j)
- {
- BiasType sum = accumulation[perspectives[p]][j];
- output[offset + j] = static_cast<OutputType>(std::max<int>(0, std::min<int>(127, sum)));
+ for (IndexType j = 0; j < HalfDimensions / 2; ++j) {
+ BiasType sum0 = accumulation[static_cast<int>(perspectives[p])][j + 0];
+ BiasType sum1 = accumulation[static_cast<int>(perspectives[p])][j + HalfDimensions / 2];
+ sum0 = std::max<int>(0, std::min<int>(127, sum0));
+ sum1 = std::max<int>(0, std::min<int>(127, sum1));
+ output[offset + j] = static_cast<OutputType>(sum0 * sum1 / 128);
}
+
+#endif
}
- return psqt;
- #endif
+ return psqt;
} // end of function transform()
// That might depend on the feature set and generally relies on the
// feature set's update cost calculation to be correct and never
// allow updates with more added/removed features than MaxActiveDimensions.
- using IndexList = ValueList<IndexType, FeatureSet::MaxActiveDimensions>;
#ifdef VECTOR
// Gcc-10.2 unnecessarily spills AVX2 registers if this array
// of the estimated gain in terms of features to be added/subtracted.
StateInfo *st = pos.state(), *next = nullptr;
int gain = FeatureSet::refresh_cost(pos);
- while (st->accumulator.state[perspective] == EMPTY)
+ while (st->previous && !st->accumulator.computed[perspective])
{
// This governs when a full feature refresh is needed and how many
// updates are better than just one full refresh.
st = st->previous;
}
- if (st->accumulator.state[perspective] == COMPUTED)
+ if (st->accumulator.computed[perspective])
{
if (next == nullptr)
return;
// Gather all features to be updated.
const Square ksq = pos.square<KING>(perspective);
- IndexList removed[2], added[2];
+ FeatureSet::IndexList removed[2], added[2];
FeatureSet::append_changed_indices(
- ksq, next, perspective, removed[0], added[0]);
+ ksq, next->dirtyPiece, perspective, removed[0], added[0]);
for (StateInfo *st2 = pos.state(); st2 != next; st2 = st2->previous)
FeatureSet::append_changed_indices(
- ksq, st2, perspective, removed[1], added[1]);
+ ksq, st2->dirtyPiece, perspective, removed[1], added[1]);
// Mark the accumulators as computed.
- next->accumulator.state[perspective] = COMPUTED;
- pos.state()->accumulator.state[perspective] = COMPUTED;
+ next->accumulator.computed[perspective] = true;
+ pos.state()->accumulator.computed[perspective] = true;
// Now update the accumulators listed in states_to_update[], where the last element is a sentinel.
StateInfo *states_to_update[3] =
{
// Refresh the accumulator
auto& accumulator = pos.state()->accumulator;
- accumulator.state[perspective] = COMPUTED;
- IndexList active;
+ accumulator.computed[perspective] = true;
+ FeatureSet::IndexList active;
FeatureSet::append_active_indices(pos, perspective, active);
#ifdef VECTOR
#endif
}
- using BiasType = std::int16_t;
- using WeightType = std::int16_t;
- using PSQTWeightType = std::int32_t;
-
alignas(CacheLineSize) BiasType biases[HalfDimensions];
alignas(CacheLineSize) WeightType weights[HalfDimensions * InputDimensions];
alignas(CacheLineSize) PSQTWeightType psqtWeights[InputDimensions * PSQTBuckets];