Reuse 5 slots instead of 4

[stockfish] / src / tt.cpp

diff --git a/src/tt.cpp b/src/tt.cpp
index f2313eabf1c72eee350de8cdda7038519944d222..5ea6a808acf86b2ad256f8715bf5c59772e015b1 100644 (file)
--- a/src/tt.cpp
+++ b/src/tt.cpp
@@ -25,15 +25,16 @@
 #include <cassert>
 #include <cmath>
 #include <cstring>
-#include <xmmintrin.h>
 
 #include "movegen.h"
 #include "tt.h"
 
+#if defined(_MSC_VER)
+#include <xmmintrin.h>
+#endif
 
-/// This is the number of TTEntry slots for each position
-static const int ClusterSize = 5;
-
+// The main transposition table
+TranspositionTable TT;
 
 ////
 //// Functions
@@ -63,14 +64,14 @@ void TranspositionTable::set_size(unsigned mbSize) {
 
   // We store a cluster of ClusterSize number of TTEntry for each position
   // and newSize is the maximum number of storable positions.
-  while ((2 * newSize) * ClusterSize * (sizeof(TTEntry)) <= (mbSize << 20))
+  while ((2 * newSize) * sizeof(TTCluster) <= (mbSize << 20))
       newSize *= 2;
 
   if (newSize != size)
   {
       size = newSize;
       delete [] entries;
-      entries = new TTEntry[size * ClusterSize];
+      entries = new TTCluster[size];
       if (!entries)
       {
           std::cerr << "Failed to allocate " << mbSize
@@ -89,7 +90,17 @@ void TranspositionTable::set_size(unsigned mbSize) {
 
 void TranspositionTable::clear() {
 
-  memset(entries, 0, size * ClusterSize * sizeof(TTEntry));
+  memset(entries, 0, size * sizeof(TTCluster));
+}
+
+
+/// TranspositionTable::first_entry returns a pointer to the first
+/// entry of a cluster given a position. The low 32 bits of the key
+/// are used to get the index in the table.
+
+inline TTEntry* TranspositionTable::first_entry(const Key posKey) const {
+
+  return entries[uint32_t(posKey) & (size - 1)].data;
 }
 
 
@@ -145,7 +156,7 @@ void TranspositionTable::store(const Key posKey, Value v, ValueType t, Depth d,
 TTEntry* TranspositionTable::retrieve(const Key posKey) const {
 
   uint32_t posKey32 = posKey >> 32;
-  TTEntry *tte = first_entry(posKey);
+  TTEntry* tte = first_entry(posKey);
 
   for (int i = 0; i < ClusterSize; i++, tte++)
       if (tte->key() == posKey32)
@@ -154,26 +165,28 @@ TTEntry* TranspositionTable::retrieve(const Key posKey) const {
   return NULL;
 }
 
+
 /// TranspositionTable::prefetch looks up the current position in the
 /// transposition table and load it in L1/L2 cache. This is a non
 /// blocking function and do not stalls the CPU waiting for data
 /// to be loaded from RAM, that can be very slow. When we will
 /// subsequently call retrieve() the TT data will be already
-/// quickly accessible in L1/l2 CPU cache.
+/// quickly accessible in L1/L2 CPU cache.
 
 void TranspositionTable::prefetch(const Key posKey) const {
 
-  _mm_prefetch((char*)first_entry(posKey), _MM_HINT_T0);
+#if defined(_MSC_VER)
+   char* addr = (char*)first_entry(posKey);
+  _mm_prefetch(addr, _MM_HINT_T0);
+  _mm_prefetch(addr+64, _MM_HINT_T0);
+#else
+  // We need to force an asm volatile here because gcc builtin
+  // is optimized away by Intel compiler.
+  char* addr = (char*)first_entry(posKey);
+  asm volatile("prefetcht0 %0" :: "m" (addr));
+#endif
 }
 
-/// TranspositionTable::first_entry returns a pointer to the first
-/// entry of a cluster given a position. The low 32 bits of the key
-/// are used to get the index in the table.
-
-inline TTEntry* TranspositionTable::first_entry(const Key posKey) const {
-
-  return entries + ((uint32_t(posKey) & (size - 1)) * ClusterSize);
-}
 
 /// TranspositionTable::new_search() is called at the beginning of every new
 /// search. It increments the "generation" variable, which is used to