1
0
Fork 0
mirror of https://github.com/sockspls/badfish synced 2025-07-11 19:49:14 +00:00

Don't fear races when are harmless

Actually race conditions do exist in an engine, just
think for a moment to TT concurrent access. Racy code
is not a problem per se, if the consequences are well
known and correctly handled.

In case of TT access we ensure that the TT move is validated
before to be tried, here we just retry the same move in less
that 1 case out of a million: this is totally harmless considering
that very probably the second time the move is tried we get
immediately a TT hit and search quickly returns.

So we simplify the code for no harm.

No fuctional change (in single thread case)
This commit is contained in:
Marco Costalba 2014-02-08 13:07:57 +01:00
parent 399968f1d0
commit 0a1092f64e

View file

@ -233,15 +233,15 @@ void MovePicker::generate_next() {
killers[2].move = killers[3].move = MOVE_NONE;
killers[4].move = killers[5].move = MOVE_NONE;
// Please note that following code is racy and could yield to rare (less
// than 1 out of a million) duplicated entries in SMP case. This is harmless.
// Be sure countermoves are different from killers
for (int i = 0; i < 2; ++i)
if ( countermoves[i] != (cur+0)->move
&& countermoves[i] != (cur+1)->move)
(end++)->move = countermoves[i];
if (countermoves[1] && countermoves[1] == countermoves[0]) // Due to SMP races
killers[3].move = MOVE_NONE;
// Be sure followupmoves are different from killers and countermoves
for (int i = 0; i < 2; ++i)
if ( followupmoves[i] != (cur+0)->move
@ -249,10 +249,6 @@ void MovePicker::generate_next() {
&& followupmoves[i] != (cur+2)->move
&& followupmoves[i] != (cur+3)->move)
(end++)->move = followupmoves[i];
if (followupmoves[1] && followupmoves[1] == followupmoves[0]) // Due to SMP races
(--end)->move = MOVE_NONE;
return;
case QUIETS_1_S1: