jirayu_sri/stockfish
1
0
Fork 0
mirror of https://github.com/official-stockfish/Stockfish.git synced 2026-07-22 20:57:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

Prefetch threat weight rows during append_changed_indices

Browse Source 
Passed STC (first run):
https://tests.stockfishchess.org/tests/view/698030ee6362aee5c8a552c7
LLR: 2.96 (-2.94,2.94) <0.00,2.00>
Total: 120800 W: 31336 L: 30912 D: 58552
Ptnml(0-2): 330, 13084, 33172, 13460, 354

Passed LTC:
https://tests.stockfishchess.org/tests/view/6983946d473df9d1d24a916d
LLR: 2.94 (-2.94,2.94) <0.50,2.50>
Total: 86850 W: 22352 L: 21951 D: 42547
Ptnml(0-2): 45, 8506, 25934, 8883, 57

Passed STC SMP (after changes made after opening the PR):
https://tests.stockfishchess.org/tests/view/6987b4c1b0f3ca5200aaf9da
LLR: 2.95 (-2.94,2.94) <0.00,2.00>
Total: 84480 W: 21994 L: 21627 D: 40859
Ptnml(0-2): 98, 9378, 22932, 9723, 109

Passed STC (second run, as a sanity check made after changes made after opening the PR -- these changed, however, didn't change assembly code produced by the compiler):
https://tests.stockfishchess.org/tests/view/6987ad21b0f3ca5200aaf9c9
LLR: 2.93 (-2.94,2.94) <0.00,2.00>
Total: 132192 W: 34526 L: 34083 D: 63583
Ptnml(0-2): 421, 14511, 35759, 15014, 391

GCC maps __builtin_prefetch locality=1 to PREFETCHT2 instruction, targeting L2
cache. To make things clear, added template-based approach as suggested by
Disservin.

PREFETCHT2 should be the appropriate choice for the 78 MB threat weight table:
avoids L1d pollution, parks data in L2 where it is promoted to L1d on actual
load.

With 200-500 cycle lead time from make_index, data has time to arrive in L2.

suggestions by Disservin and anematode included

closes https://github.com/official-stockfish/Stockfish/pull/6602

No functional change
This commit is contained in:
Maxim Masiutin
2026-02-13 07:01:15 +01:00
committed by Joost VandeVondele
parent 99221e5ee6
commit 238ef05bb0
5 changed files with 108 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/nnue/nnue_accumulator.cpp
+21 -5
View File
@@ -554,10 +554,12 @@ void double_inc_update(Color p
fusedData.dp2removed = dp2.remove_sq;
ThreatFeatureSet::IndexList removed, added;
const auto* pfBase = &featureTransformer.threatWeights[0];
auto pfStride = static_cast<IndexType>(TransformedFeatureDimensions);
ThreatFeatureSet::append_changed_indices(perspective, ksq, middle_state.diff, removed, added,
&fusedData, true);
&fusedData, true, pfBase, pfStride);
ThreatFeatureSet::append_changed_indices(perspective, ksq, target_state.diff, removed, added,
&fusedData, false);
&fusedData, false, pfBase, pfStride);
auto updateContext =
make_accumulator_update_context(perspective, featureTransformer, computed, target_state);
@@ -585,10 +587,24 @@ void update_accumulator_incremental(
// In this case, the maximum size of both feature addition and removal
// is 2, since we are incrementally updating one move at a time.
typename FeatureSet::IndexList removed, added;
if constexpr (Forward)
FeatureSet::append_changed_indices(perspective, ksq, target_state.diff, removed, added);
if constexpr (std::is_same_v<FeatureSet, ThreatFeatureSet>)
{
const auto* pfBase = &featureTransformer.threatWeights[0];
auto pfStride = static_cast<IndexType>(TransformedFeatureDimensions);
if constexpr (Forward)
FeatureSet::append_changed_indices(perspective, ksq, target_state.diff, removed, added,
nullptr, false, pfBase, pfStride);
else
FeatureSet::append_changed_indices(perspective, ksq, computed.diff, added, removed,
nullptr, false, pfBase, pfStride);
}
else
FeatureSet::append_changed_indices(perspective, ksq, computed.diff, added, removed);
{
if constexpr (Forward)
FeatureSet::append_changed_indices(perspective, ksq, target_state.diff, removed, added);
else
FeatureSet::append_changed_indices(perspective, ksq, computed.diff, added, removed);
}
auto updateContext =
make_accumulator_update_context(perspective, featureTransformer, computed, target_state);