When a function takes a variable number of pointer arguments, with a zero

pointer marking the end of the list, the zero *must* be cast to the pointer
type.  An un-cast zero is a 32-bit int, and at least on x86_64, gcc will
not extend the zero to 64 bits, thus allowing the upper 32 bits to be
random junk.

The new END_WITH_NULL macro may be used to annotate a such a function
so that GCC (version 4 or newer) will detect the use of un-casted zero
at compile time.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23888 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 1 insertions, 1 deletions

diff --git a/lib/Transforms/Instrumentation/ProfilePaths/EdgeCode.cpp b/lib/Transforms/Instrumentation/ProfilePaths/EdgeCode.cpp
index 95822df002..bf94943788 100644
--- a/lib/Transforms/Instrumentation/ProfilePaths/EdgeCode.cpp
+++ b/lib/Transforms/Instrumentation/ProfilePaths/EdgeCode.cpp
@@ -40,7 +40,7 @@ static void getTriggerCode(Module *M, BasicBlock *BB, int MethNo, Value *pathNo,
   const Type *PIntTy = PointerType::get(Type::IntTy);
   Function *trigMeth = M->getOrInsertFunction("trigger", Type::VoidTy,
                                               Type::IntTy, Type::IntTy,
-                                              PIntTy, PIntTy, 0);
+                                              PIntTy, PIntTy, (Type *)0);
   assert(trigMeth && "trigger method could not be inserted!");
 
   vector<Value *> trargs;