jirayu_sri/stockfish
1
0
Fork 0
mirror of https://github.com/official-stockfish/Stockfish.git synced 2026-07-22 12:47:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

Optimize make_index() using templates and lookup tables.

Browse Source 
https://tests.stockfishchess.org/tests/view/634517e54bc7650f07542f99
LLR: 2.94 (-2.94,2.94) <0.00,2.00>
Total: 642672 W: 171819 L: 170658 D: 300195
Ptnml(0-2): 2278, 68077, 179416, 69336, 2229

this also introduces `-flto-partition=one` as suggested by MinetaS (Syine Mineta)
to avoid linking errors due to LTO on 32 bit mingw. This change was tested in isolation as well

https://tests.stockfishchess.org/tests/view/634aacf84bc7650f0755188b
LLR: 2.94 (-2.94,2.94) <-1.75,0.25>
Total: 119352 W: 31986 L: 31862 D: 55504
Ptnml(0-2): 439, 12624, 33400, 12800, 413

closes https://github.com/official-stockfish/Stockfish/pull/4199

No functional change
This commit is contained in:
mstembera
2022-10-16 11:42:19 +02:00
committed by Joost VandeVondele
parent e90341f9c9
commit 93f71ecfe1
4 changed files with 96 additions and 68 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/nnue/nnue_feature_transformer.h
+33 -32
View File
@@ -271,8 +271,8 @@ namespace Stockfish::Eval::NNUE {
// Convert input features
std::int32_t transform(const Position& pos, OutputType* output, int bucket) const {
update_accumulator(pos, WHITE);
update_accumulator(pos, BLACK);
update_accumulator<WHITE>(pos);
update_accumulator<BLACK>(pos);
const Color perspectives[2] = {pos.side_to_move(), ~pos.side_to_move()};
const auto& accumulation = pos.state()->accumulator.accumulation;
@@ -338,7 +338,8 @@ namespace Stockfish::Eval::NNUE {
private:
void update_accumulator(const Position& pos, const Color perspective) const {
template<Color Perspective>
void update_accumulator(const Position& pos) const {
// The size must be enough to contain the largest possible update.
// That might depend on the feature set and generally relies on the
@@ -356,18 +357,18 @@ namespace Stockfish::Eval::NNUE {
// 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->previous && !st->accumulator.computed[perspective])
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.
if ( FeatureSet::requires_refresh(st, perspective)
if ( FeatureSet::requires_refresh(st, Perspective)
|| (gain -= FeatureSet::update_cost(st) + 1) < 0)
break;
next = st;
st = st->previous;
}
if (st->accumulator.computed[perspective])
if (st->accumulator.computed[Perspective])
{
if (next == nullptr)
return;
@@ -376,17 +377,17 @@ namespace Stockfish::Eval::NNUE {
// accumulator. Then, we update the current accumulator (pos.state()).
// Gather all features to be updated.
const Square ksq = pos.square<KING>(perspective);
const Square ksq = pos.square<KING>(Perspective);
FeatureSet::IndexList removed[2], added[2];
FeatureSet::append_changed_indices(
ksq, next->dirtyPiece, perspective, removed[0], added[0]);
FeatureSet::append_changed_indices<Perspective>(
ksq, next->dirtyPiece, removed[0], added[0]);
for (StateInfo *st2 = pos.state(); st2 != next; st2 = st2->previous)
FeatureSet::append_changed_indices(
ksq, st2->dirtyPiece, perspective, removed[1], added[1]);
FeatureSet::append_changed_indices<Perspective>(
ksq, st2->dirtyPiece, removed[1], added[1]);
// Mark the accumulators as computed.
next->accumulator.computed[perspective] = true;
pos.state()->accumulator.computed[perspective] = true;
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] =
@@ -396,7 +397,7 @@ namespace Stockfish::Eval::NNUE {
{
// Load accumulator
auto accTile = reinterpret_cast<vec_t*>(
&st->accumulator.accumulation[perspective][j * TileHeight]);
&st->accumulator.accumulation[Perspective][j * TileHeight]);
for (IndexType k = 0; k < NumRegs; ++k)
acc[k] = vec_load(&accTile[k]);
@@ -422,7 +423,7 @@ namespace Stockfish::Eval::NNUE {
// Store accumulator
accTile = reinterpret_cast<vec_t*>(
&states_to_update[i]->accumulator.accumulation[perspective][j * TileHeight]);
&states_to_update[i]->accumulator.accumulation[Perspective][j * TileHeight]);
for (IndexType k = 0; k < NumRegs; ++k)
vec_store(&accTile[k], acc[k]);
}
@@ -432,7 +433,7 @@ namespace Stockfish::Eval::NNUE {
{
// Load accumulator
auto accTilePsqt = reinterpret_cast<psqt_vec_t*>(
&st->accumulator.psqtAccumulation[perspective][j * PsqtTileHeight]);
&st->accumulator.psqtAccumulation[Perspective][j * PsqtTileHeight]);
for (std::size_t k = 0; k < NumPsqtRegs; ++k)
psqt[k] = vec_load_psqt(&accTilePsqt[k]);
@@ -458,7 +459,7 @@ namespace Stockfish::Eval::NNUE {
// Store accumulator
accTilePsqt = reinterpret_cast<psqt_vec_t*>(
&states_to_update[i]->accumulator.psqtAccumulation[perspective][j * PsqtTileHeight]);
&states_to_update[i]->accumulator.psqtAccumulation[Perspective][j * PsqtTileHeight]);
for (std::size_t k = 0; k < NumPsqtRegs; ++k)
vec_store_psqt(&accTilePsqt[k], psqt[k]);
}
@@ -467,12 +468,12 @@ namespace Stockfish::Eval::NNUE {
#else
for (IndexType i = 0; states_to_update[i]; ++i)
{
std::memcpy(states_to_update[i]->accumulator.accumulation[perspective],
st->accumulator.accumulation[perspective],
std::memcpy(states_to_update[i]->accumulator.accumulation[Perspective],
st->accumulator.accumulation[Perspective],
HalfDimensions * sizeof(BiasType));
for (std::size_t k = 0; k < PSQTBuckets; ++k)
states_to_update[i]->accumulator.psqtAccumulation[perspective][k] = st->accumulator.psqtAccumulation[perspective][k];
states_to_update[i]->accumulator.psqtAccumulation[Perspective][k] = st->accumulator.psqtAccumulation[Perspective][k];
st = states_to_update[i];
@@ -482,10 +483,10 @@ namespace Stockfish::Eval::NNUE {
const IndexType offset = HalfDimensions * index;
for (IndexType j = 0; j < HalfDimensions; ++j)
st->accumulator.accumulation[perspective][j] -= weights[offset + j];
st->accumulator.accumulation[Perspective][j] -= weights[offset + j];
for (std::size_t k = 0; k < PSQTBuckets; ++k)
st->accumulator.psqtAccumulation[perspective][k] -= psqtWeights[index * PSQTBuckets + k];
st->accumulator.psqtAccumulation[Perspective][k] -= psqtWeights[index * PSQTBuckets + k];
}
// Difference calculation for the activated features
@@ -494,10 +495,10 @@ namespace Stockfish::Eval::NNUE {
const IndexType offset = HalfDimensions * index;
for (IndexType j = 0; j < HalfDimensions; ++j)
st->accumulator.accumulation[perspective][j] += weights[offset + j];
st->accumulator.accumulation[Perspective][j] += weights[offset + j];
for (std::size_t k = 0; k < PSQTBuckets; ++k)
st->accumulator.psqtAccumulation[perspective][k] += psqtWeights[index * PSQTBuckets + k];
st->accumulator.psqtAccumulation[Perspective][k] += psqtWeights[index * PSQTBuckets + k];
}
}
#endif
@@ -506,9 +507,9 @@ namespace Stockfish::Eval::NNUE {
{
// Refresh the accumulator
auto& accumulator = pos.state()->accumulator;
accumulator.computed[perspective] = true;
accumulator.computed[Perspective] = true;
FeatureSet::IndexList active;
FeatureSet::append_active_indices(pos, perspective, active);
FeatureSet::append_active_indices<Perspective>(pos, active);
#ifdef VECTOR
for (IndexType j = 0; j < HalfDimensions / TileHeight; ++j)
@@ -528,7 +529,7 @@ namespace Stockfish::Eval::NNUE {
}
auto accTile = reinterpret_cast<vec_t*>(
&accumulator.accumulation[perspective][j * TileHeight]);
&accumulator.accumulation[Perspective][j * TileHeight]);
for (unsigned k = 0; k < NumRegs; k++)
vec_store(&accTile[k], acc[k]);
}
@@ -548,27 +549,27 @@ namespace Stockfish::Eval::NNUE {
}
auto accTilePsqt = reinterpret_cast<psqt_vec_t*>(
&accumulator.psqtAccumulation[perspective][j * PsqtTileHeight]);
&accumulator.psqtAccumulation[Perspective][j * PsqtTileHeight]);
for (std::size_t k = 0; k < NumPsqtRegs; ++k)
vec_store_psqt(&accTilePsqt[k], psqt[k]);
}
#else
std::memcpy(accumulator.accumulation[perspective], biases,
std::memcpy(accumulator.accumulation[Perspective], biases,
HalfDimensions * sizeof(BiasType));
for (std::size_t k = 0; k < PSQTBuckets; ++k)
accumulator.psqtAccumulation[perspective][k] = 0;
accumulator.psqtAccumulation[Perspective][k] = 0;
for (const auto index : active)
{
const IndexType offset = HalfDimensions * index;
for (IndexType j = 0; j < HalfDimensions; ++j)
accumulator.accumulation[perspective][j] += weights[offset + j];
accumulator.accumulation[Perspective][j] += weights[offset + j];
for (std::size_t k = 0; k < PSQTBuckets; ++k)
accumulator.psqtAccumulation[perspective][k] += psqtWeights[index * PSQTBuckets + k];
accumulator.psqtAccumulation[Perspective][k] += psqtWeights[index * PSQTBuckets + k];
}
#endif
}