[analyzer] When creating a trimmed graph, preserve whether a node is a sink.

This is important because sometimes two nodes are identical, except the
second one is a sink.

This bug has probably been around for a while, but it wouldn't have been an
issue in the old report graph algorithm. I'm ashamed to say I actually looked
at this the first time around and thought it would never be a problem...and
then didn't include an assertion to back that up.

PR15684

llvm-svn: 178944

clang/lib/StaticAnalyzer/Core/BugReporter.cpp

@@ -2010,7 +2010,8 @@ bool TrimmedGraph::popNextReportGraph(ReportGraph &GraphWrapper) {
   while (true) {
     // Create the equivalent node in the new graph with the same state
     // and location.
-    ExplodedNode *NewN = GNew->getNode(OrigN->getLocation(), OrigN->getState());
+    ExplodedNode *NewN = GNew->getNode(OrigN->getLocation(), OrigN->getState(),
+                                       OrigN->isSink());
 
     // Store the mapping to the original node.
     InterExplodedGraphMap::const_iterator IMitr = InverseMap.find(OrigN);

clang/test/Analysis/misc-ps.c

@@ -163,3 +163,15 @@ int PR14634(int x) {
   return !y;
 }
 
+
+// PR15684: If a checker generates a sink node after generating a regular node
+// and no state changes between the two, graph trimming would consider the two
+// the same node, forming a loop.
+struct PR15684 {
+  void (*callback)(int);
+};
+void sinkAfterRegularNode(struct PR15684 *context) {
+  int uninitialized;
+  context->callback(uninitialized); // expected-warning {{uninitialized}}
+}
+