/*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1
- Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
- Copyright (C) 2008-2015 Marco Costalba, Joona Kiiski, Tord Romstad
- Copyright (C) 2015-2019 Marco Costalba, Joona Kiiski, Gary Linscott, Tord Romstad
+ Copyright (C) 2004-2023 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
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
-#include <cstring> // For std::memset
+#include "tt.h"
+
+#include <cassert>
+#include <cstdlib>
+#include <cstring>
#include <iostream>
#include <thread>
+#include <vector>
-#include "bitboard.h"
#include "misc.h"
#include "thread.h"
-#include "tt.h"
#include "uci.h"
+namespace Stockfish {
+
TranspositionTable TT; // Our global transposition table
-/// TTEntry::save populates the TTEntry with a new node's data, possibly
-/// overwriting an old position. Update is not atomic and can be racy.
+// TTEntry::save() populates the TTEntry with a new node's data, possibly
+// overwriting an old position. The update is not atomic and can be racy.
void TTEntry::save(Key k, Value v, bool pv, Bound b, Depth d, Move m, Value ev) {
- assert(d / ONE_PLY * ONE_PLY == d);
-
// Preserve any existing move for the same position
- if (m || (k >> 48) != key16)
- move16 = (uint16_t)m;
+ if (m || uint16_t(k) != key16)
+ move16 = uint16_t(m);
- // Overwrite less valuable entries
- if ( (k >> 48) != key16
- ||(d - DEPTH_OFFSET) / ONE_PLY > depth8 - 4
- || b == BOUND_EXACT)
+ // Overwrite less valuable entries (cheapest checks first)
+ if ( b == BOUND_EXACT
+ || uint16_t(k) != key16
+ || d - DEPTH_OFFSET + 2 * pv > depth8 - 4)
{
- assert((d - DEPTH_OFFSET) / ONE_PLY >= 0);
-
- key16 = (uint16_t)(k >> 48);
- value16 = (int16_t)v;
- eval16 = (int16_t)ev;
- genBound8 = (uint8_t)(TT.generation8 | uint8_t(pv) << 2 | b);
- depth8 = (uint8_t)((d - DEPTH_OFFSET) / ONE_PLY);
+ assert(d > DEPTH_OFFSET);
+ assert(d < 256 + DEPTH_OFFSET);
+
+ key16 = uint16_t(k);
+ depth8 = uint8_t(d - DEPTH_OFFSET);
+ genBound8 = uint8_t(TT.generation8 | uint8_t(pv) << 2 | b);
+ value16 = int16_t(v);
+ eval16 = int16_t(ev);
}
}
-/// TranspositionTable::resize() sets the size of the transposition table,
-/// measured in megabytes. Transposition table consists of a power of 2 number
-/// of clusters and each cluster consists of ClusterSize number of TTEntry.
+// TranspositionTable::resize() sets the size of the transposition table,
+// measured in megabytes. Transposition table consists of a power of 2 number
+// of clusters and each cluster consists of ClusterSize number of TTEntry.
void TranspositionTable::resize(size_t mbSize) {
Threads.main()->wait_for_search_finished();
- clusterCount = mbSize * 1024 * 1024 / sizeof(Cluster);
+ aligned_large_pages_free(table);
- free(mem);
- mem = malloc(clusterCount * sizeof(Cluster) + CacheLineSize - 1);
+ clusterCount = mbSize * 1024 * 1024 / sizeof(Cluster);
- if (!mem)
+ table = static_cast<Cluster*>(aligned_large_pages_alloc(clusterCount * sizeof(Cluster)));
+ if (!table)
{
std::cerr << "Failed to allocate " << mbSize
<< "MB for transposition table." << std::endl;
exit(EXIT_FAILURE);
}
- table = (Cluster*)((uintptr_t(mem) + CacheLineSize - 1) & ~(CacheLineSize - 1));
clear();
}
-/// TranspositionTable::clear() initializes the entire transposition table to zero,
+// TranspositionTable::clear() initializes the entire transposition table to zero,
// in a multi-threaded way.
void TranspositionTable::clear() {
std::vector<std::thread> threads;
- for (size_t idx = 0; idx < Options["Threads"]; ++idx)
+ for (size_t idx = 0; idx < size_t(Options["Threads"]); ++idx)
{
threads.emplace_back([this, idx]() {
WinProcGroup::bindThisThread(idx);
// Each thread will zero its part of the hash table
- const size_t stride = clusterCount / Options["Threads"],
- start = stride * idx,
- len = idx != Options["Threads"] - 1 ?
+ const size_t stride = size_t(clusterCount / Options["Threads"]),
+ start = size_t(stride * idx),
+ len = idx != size_t(Options["Threads"]) - 1 ?
stride : clusterCount - start;
std::memset(&table[start], 0, len * sizeof(Cluster));
});
}
- for (std::thread& th: threads)
+ for (std::thread& th : threads)
th.join();
}
-/// TranspositionTable::probe() looks up the current position in the transposition
-/// table. It returns true and a pointer to the TTEntry if the position is found.
-/// Otherwise, it returns false and a pointer to an empty or least valuable TTEntry
-/// to be replaced later. The replace value of an entry is calculated as its depth
-/// minus 8 times its relative age. TTEntry t1 is considered more valuable than
-/// TTEntry t2 if its replace value is greater than that of t2.
+
+// TranspositionTable::probe() looks up the current position in the transposition
+// table. It returns true and a pointer to the TTEntry if the position is found.
+// Otherwise, it returns false and a pointer to an empty or least valuable TTEntry
+// to be replaced later. The replace value of an entry is calculated as its depth
+// minus 8 times its relative age. TTEntry t1 is considered more valuable than
+// TTEntry t2 if its replace value is greater than that of t2.
TTEntry* TranspositionTable::probe(const Key key, bool& found) const {
TTEntry* const tte = first_entry(key);
- const uint16_t key16 = key >> 48; // Use the high 16 bits as key inside the cluster
+ const uint16_t key16 = uint16_t(key); // Use the low 16 bits as key inside the cluster
for (int i = 0; i < ClusterSize; ++i)
- if (!tte[i].key16 || tte[i].key16 == key16)
+ if (tte[i].key16 == key16 || !tte[i].depth8)
{
- tte[i].genBound8 = uint8_t(generation8 | (tte[i].genBound8 & 0x7)); // Refresh
+ tte[i].genBound8 = uint8_t(generation8 | (tte[i].genBound8 & (GENERATION_DELTA - 1))); // Refresh
- return found = (bool)tte[i].key16, &tte[i];
+ return found = bool(tte[i].depth8), &tte[i];
}
// Find an entry to be replaced according to the replacement strategy
TTEntry* replace = tte;
for (int i = 1; i < ClusterSize; ++i)
// Due to our packed storage format for generation and its cyclic
- // nature we add 263 (256 is the modulus plus 7 to keep the unrelated
- // lowest three bits from affecting the result) to calculate the entry
- // age correctly even after generation8 overflows into the next cycle.
- if ( replace->depth8 - ((263 + generation8 - replace->genBound8) & 0xF8)
- > tte[i].depth8 - ((263 + generation8 - tte[i].genBound8) & 0xF8))
+ // nature we add GENERATION_CYCLE (256 is the modulus, plus what
+ // is needed to keep the unrelated lowest n bits from affecting
+ // the result) to calculate the entry age correctly even after
+ // generation8 overflows into the next cycle.
+ if ( replace->depth8 - ((GENERATION_CYCLE + generation8 - replace->genBound8) & GENERATION_MASK)
+ > tte[i].depth8 - ((GENERATION_CYCLE + generation8 - tte[i].genBound8) & GENERATION_MASK))
replace = &tte[i];
return found = false, replace;
}
-/// TranspositionTable::hashfull() returns an approximation of the hashtable
-/// occupation during a search. The hash is x permill full, as per UCI protocol.
+// TranspositionTable::hashfull() returns an approximation of the hashtable
+// occupation during a search. The hash is x permill full, as per UCI protocol.
int TranspositionTable::hashfull() const {
int cnt = 0;
- for (int i = 0; i < 1000 / ClusterSize; ++i)
+ for (int i = 0; i < 1000; ++i)
for (int j = 0; j < ClusterSize; ++j)
- cnt += (table[i].entry[j].genBound8 & 0xF8) == generation8;
+ cnt += table[i].entry[j].depth8 && (table[i].entry[j].genBound8 & GENERATION_MASK) == generation8;
- return cnt * 1000 / (ClusterSize * (1000 / ClusterSize));
+ return cnt / ClusterSize;
}
+
+} // namespace Stockfish
projects / stockfish / blobdiff