Fix several MP/qsearch correctness bugs found by auditing ctx->uPly

indexing and per-thread state across MakeMove/recursion boundaries;
add a position-signature stamp so full-eval-only data can be validated
cheaply instead of approximated.

- chess.h/searchsup.c/search.c/root.c: add NODE_TYPE (PV/CUT/ALL)
  classification, set centrally in CommonSearchInit, replacing the old
  binary fPvNode flag. Fixes split.c's node-type check reading stale
  data on helper threads (they never call CommonSearchInit for the
  split-point ply) and removes a redundant, always-true parent-type
  check in search.c's own LMR gate.

- split.c: StartParallelSearch now copies the split point's eNodeType
  into SPLIT_INFO for helper threads to read (mirroring the existing
  iEval/iImprovement pattern); a split-node fail-high is now detected
  explicitly in search.c's move loop instead of falling through to the
  generic no_more_moves classifier, which misread it as a fail-low
  (skipping killer/history updates and storing an inverted-sense upper
  bound instead of a lower bound); ThreadUnderTerminatingSplit now
  checks all nested split pointers instead of just the first, so a
  thread notices when an inner split terminates while it's still
  helping an outer one; MAX_SPLITS bumped 20->64 for higher thread
  counts.

- chess.h/searchsup.c/search.c: convert
  SEARCHER_THREAD_CONTEXT.uLastPositionalSource (a single per-thread
  scalar) to PLY_INFO.uLastPositionalSource (per-ply). The scalar
  version was clobbered whenever QSearch's own move loop recursed
  before a later iteration read it back, corrupting which qsearch
  futility margin got applied; also fixes the same hazard in the
  newly-added check-evasion futility prune.

- chess.h/generate.c/searchsup.c: generalize move-scoring interpretation
  via a new MOVE_SCORING_SCHEME tag (per-ply, set by whichever
  generate.c routine last scored that ply's moves) and a single shared
  DecodeMoveValueEstimate, replacing scheme-blind decode logic in
  ComputeMoveScore and _ShouldWeConsiderThisMove.

- chess.h/eval.c/dynamic.c: stamp PLY_INFO.u64FullEvalSig with the
  position signature whenever a node completes a full (non-lazy) Eval()
  call. cEnprise/pEnprise/cTrapped/pTrapped/iKingScore/the eval-scoped
  bb*Attacks bitboards are only valid when this matches the position's
  live signature; _EnpriseSlotValid/_TrappedSlotValid now compare sigs
  directly instead of re-checking piece identity on the live board.

- searchsup.c: CommonSearchInit clears pi->mvBest on both its IsDraw and
  too-deep early-exit paths to prevent a later sibling node from reading
  stale mvBest data left by an unrelated earlier visit to the same ply.

Co-Authored-By: Claude Sonnet 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01C9r36BydPnfpyYLFhkTo6d

1 files changed, 20 insertions, 4 deletions

diff --git a/src/split.c b/src/split.c
index f8da7ed..3a69ab4 100644
--- a/src/split.c
+++ b/src/split.c
@@ -81,7 +81,13 @@ Revision History:
 
 extern ULONG g_uIterateDepth;
 ULONG g_uNumHelperThreads = 0;
-#define MAX_SPLITS          (20)
+// Sized for up to 64 threads (--cpus' documented max, main.c) each
+// holding a handful of nested split pointers (NUM_SPLIT_PTRS_IN_CONTEXT,
+// chess.h) at once -- exhausting this pool just degrades gracefully
+// (StartParallelSearch falls back to serial search), but a ceiling
+// tuned for a handful of threads bites harder the more of them are
+// actually configured, and each SPLIT_INFO is only a few KB.
+#define MAX_SPLITS          (64)
 #define IDLE                ((ULONG)-1)
 
 void
@@ -570,6 +576,7 @@ Return value:
                    sizeof(POSITION));
             g_SplitInfo[u].iEval = ctx->sPlyInfo[ctx->uPly].iEval;
             g_SplitInfo[u].iImprovement = iImprovement;
+            g_SplitInfo[u].eNodeType = ctx->sPlyInfo[ctx->uPly].eNodeType;
 #if DEBUG
             g_SplitInfo[u].uSplitPly = ctx->uPly;
             memcpy(&(g_SplitInfo[u].sSplitPosition),
@@ -886,7 +893,17 @@ Return value:
             if (iScore >= g_SplitInfo[u].iBeta)
             {
                 //
-                // We failed high so we want to update the split node
+                // We failed high so we want to update the split node.
+                // Do NOT touch g_SplitInfo[u].iAlpha here -- the caller
+                // (StartParallelSearch's caller in search.c) asserts
+                // iAlpha stays < iBeta after a split returns, and
+                // detects fail-high purely via iBestScore/iScore, not
+                // via alpha having moved. (See search.c's handling of
+                // this return for the other half of this fix: it must
+                // special-case iBestScore >= iBeta itself rather than
+                // falling into the generic no_more_moves classifier,
+                // which infers "exact score" vs. "fail-high" purely
+                // from whether alpha moved.)
                 //
                 g_SplitInfo[u].fTerminate = TRUE;
                 g_SplitInfo[u].PV[0] = NULLMOVE;
@@ -1152,8 +1169,7 @@ Return value:
             // the deterministic missing-gate bug it actually was.
             //
             if ((uDepth > TWO_PLY) &&
-                !(ctx->sPlyInfo[ctx->uPly].fPvNode) &&
-                !(ctx->sPlyInfo[ctx->uPly - 1].fPvNode) &&
+                (NODE_PV != g_SplitInfo[u].eNodeType) &&
                 ((g_SplitInfo[u].uAlreadyDone + uMoveNum + 1) > 5) &&
                 (0 == iExtend) &&
                 (!IS_ESCAPING_CHECK(mv)) &&