[PATCH] D15410: AnalysisConsumer: use canonical decl for both lookup and store of visited decls

[Aleksei Sidorin via cfe-commits]

a.sidorin removed rL LLVM as the repository for this revision.
a.sidorin updated this revision to Diff 43723.
a.sidorin added a comment.

Added a comment.


http://reviews.llvm.org/D15410

Files:
  lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp
  test/Analysis/inlining/analysis-order.c

Index: test/Analysis/inlining/analysis-order.c
===================================================================
--- /dev/null
+++ test/Analysis/inlining/analysis-order.c
@@ -0,0 +1,15 @@
+// RUN: %clang_cc1 -analyze -analyzer-checker=core.builtin.NoReturnFunctions -analyzer-display-progress %s 2>&1 | FileCheck %s
+
+// Do not analyze test1() again because it was inlined
+void test1();
+
+void test2() {
+  test1();
+}
+
+void test1() {
+}
+
+// CHECK: analysis-order.c test2
+// CHECK-NEXT: analysis-order.c test1
+// CHECK-NEXT: analysis-order.c test2
Index: lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp
===================================================================
--- lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp
+++ lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp
@@ -494,7 +494,9 @@
     // Add the visited callees to the global visited set.
     for (SetOfConstDecls::iterator I = VisitedCallees.begin(),
                                    E = VisitedCallees.end(); I != E; ++I) {
-        Visited.insert(*I);
+      // Decls from CallGraph are already canonical. But Decls coming from
+      // CallExprs may be not. We should canonicalize them manually.
+      Visited.insert(isa<ObjCMethodDecl>(*I) ? *I : (*I)->getCanonicalDecl());
     }
     VisitedAsTopLevel.insert(D);
   }


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D15410.43723.patch
Type: text/x-patch
Size: 1310 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20151229/b78cfac5/attachment.bin>