From 4b3df0a4260a77922739b4bd27d381ffee05d81d Mon Sep 17 00:00:00 2001
From: Andrew Noyes <andrew@weaselab.dev>
Date: Thu, 1 Aug 2024 10:46:08 -0700
Subject: [PATCH] Avoid dispatching on node type twice in nextPhysical

---
 ConflictSet.cpp | 117 ++++++++++++++++++++++++++++++++++++++----------
 1 file changed, 94 insertions(+), 23 deletions(-)

diff --git a/ConflictSet.cpp b/ConflictSet.cpp
index 51780e6..0c20680 100644
--- a/ConflictSet.cpp
+++ b/ConflictSet.cpp
@@ -844,28 +844,41 @@ template <class NodeT> int getNodeIndex(NodeT *self, uint8_t index) {
 #endif
 }
 
+// Precondition - an entry for index must exist in the node
+Node *&getChildExists(Node3 *self, uint8_t index) {
+  return self->children[getNodeIndex(self, index)];
+}
+// Precondition - an entry for index must exist in the node
+Node *&getChildExists(Node16 *self, uint8_t index) {
+  return self->children[getNodeIndex(self, index)];
+}
+// Precondition - an entry for index must exist in the node
+Node *&getChildExists(Node48 *self, uint8_t index) {
+  assert(self->bitSet.test(index));
+  return self->children[self->index[index]];
+}
+// Precondition - an entry for index must exist in the node
+Node *&getChildExists(Node256 *self, uint8_t index) {
+  assert(self->bitSet.test(index));
+  return self->children[index];
+}
+
 // Precondition - an entry for index must exist in the node
 Node *&getChildExists(Node *self, uint8_t index) {
   switch (self->getType()) {
   case Type_Node0:           // GCOVR_EXCL_LINE
     __builtin_unreachable(); // GCOVR_EXCL_LINE
   case Type_Node3: {
-    auto *self3 = static_cast<Node3 *>(self);
-    return self3->children[getNodeIndex(self3, index)];
+    return getChildExists(static_cast<Node3 *>(self), index);
   }
   case Type_Node16: {
-    auto *self16 = static_cast<Node16 *>(self);
-    return self16->children[getNodeIndex(self16, index)];
+    return getChildExists(static_cast<Node16 *>(self), index);
   }
   case Type_Node48: {
-    auto *self48 = static_cast<Node48 *>(self);
-    assert(self48->bitSet.test(index));
-    return self48->children[self48->index[index]];
+    return getChildExists(static_cast<Node48 *>(self), index);
   }
   case Type_Node256: {
-    auto *self256 = static_cast<Node256 *>(self);
-    assert(self256->bitSet.test(index));
-    return self256->children[index];
+    return getChildExists(static_cast<Node256 *>(self), index);
   }
   default:                   // GCOVR_EXCL_LINE
     __builtin_unreachable(); // GCOVR_EXCL_LINE
@@ -979,24 +992,47 @@ template <class NodeT> int getChildGeqSimd(NodeT *self, int child) {
 #endif
 }
 
-int getChildGeq(Node *self, int child) {
+int getChildGeq(Node0 *, int child) {
   if (child > 255) {
     return -1;
   }
+  return -1;
+}
+int getChildGeq(Node3 *self, int child) {
+  if (child > 255) {
+    return -1;
+  }
+  return getChildGeqSimd(self, child);
+}
+int getChildGeq(Node16 *self, int child) {
+  if (child > 255) {
+    return -1;
+  }
+  return getChildGeqSimd(self, child);
+}
+int getChildGeq(Node48 *self, int child) {
+  if (child > 255) {
+    return -1;
+  }
+  return self->bitSet.firstSetGeq(child);
+}
+int getChildGeq(Node256 *self, int child) {
+  static_assert(offsetof(Node48, bitSet) == offsetof(Node256, bitSet));
+  return getChildGeq(reinterpret_cast<Node48 *>(self), child);
+}
+
+int getChildGeq(Node *self, int child) {
   switch (self->getType()) {
   case Type_Node0:
-    return -1;
+    return getChildGeq(static_cast<Node0 *>(self), child);
   case Type_Node3:
-    return getChildGeqSimd(static_cast<Node3 *>(self), child);
+    return getChildGeq(static_cast<Node3 *>(self), child);
   case Type_Node16:
-    return getChildGeqSimd(static_cast<Node16 *>(self), child);
+    return getChildGeq(static_cast<Node16 *>(self), child);
   case Type_Node48:
-    [[fallthrough]];
-  case Type_Node256: {
-    static_assert(offsetof(Node48, bitSet) == offsetof(Node256, bitSet));
-    auto *self48 = static_cast<Node48 *>(self);
-    return self48->bitSet.firstSetGeq(child);
-  }
+    return getChildGeq(static_cast<Node48 *>(self), child);
+  case Type_Node256:
+    return getChildGeq(static_cast<Node256 *>(self), child);
   default:                   // GCOVR_EXCL_LINE
     __builtin_unreachable(); // GCOVR_EXCL_LINE
   }
@@ -1201,9 +1237,44 @@ Node *&getOrCreateChild(Node *&self, uint8_t index, WriteContext *tls) {
 Node *nextPhysical(Node *node) {
   int index = -1;
   for (;;) {
-    auto nextChild = getChildGeq(node, index + 1);
-    if (nextChild >= 0) {
-      return getChildExists(node, nextChild);
+    switch (node->getType()) {
+    case Type_Node0: {
+      auto *n = static_cast<Node0 *>(node);
+      auto nextChild = getChildGeq(n, index + 1);
+      if (nextChild >= 0) {
+        return getChildExists(n, nextChild);
+      }
+    } break;
+    case Type_Node3: {
+      auto *n = static_cast<Node3 *>(node);
+      auto nextChild = getChildGeq(n, index + 1);
+      if (nextChild >= 0) {
+        return getChildExists(n, nextChild);
+      }
+    } break;
+    case Type_Node16: {
+      auto *n = static_cast<Node16 *>(node);
+      auto nextChild = getChildGeq(n, index + 1);
+      if (nextChild >= 0) {
+        return getChildExists(n, nextChild);
+      }
+    } break;
+    case Type_Node48: {
+      auto *n = static_cast<Node48 *>(node);
+      auto nextChild = getChildGeq(n, index + 1);
+      if (nextChild >= 0) {
+        return getChildExists(n, nextChild);
+      }
+    } break;
+    case Type_Node256: {
+      auto *n = static_cast<Node256 *>(node);
+      auto nextChild = getChildGeq(n, index + 1);
+      if (nextChild >= 0) {
+        return getChildExists(n, nextChild);
+      }
+    } break;
+    default:                   // GCOVR_EXCL_LINE
+      __builtin_unreachable(); // GCOVR_EXCL_LINE
     }
     index = node->parentsIndex;
     node = node->parent;