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, 2 insertions, 1 deletions

diff --git a/lib/Transforms/Instrumentation/ProfilePaths/ProfilePaths.cpp b/lib/Transforms/Instrumentation/ProfilePaths/ProfilePaths.cpp
index d0d0f550fe..ce9e328c27 100644
--- a/lib/Transforms/Instrumentation/ProfilePaths/ProfilePaths.cpp
+++ b/lib/Transforms/Instrumentation/ProfilePaths/ProfilePaths.cpp
@@ -189,7 +189,8 @@ bool ProfilePaths::runOnFunction(Function &F){
     // IN THEIR INITIALIZE METHOD!!
     Function *initialize =
       F.getParent()->getOrInsertFunction("reoptimizerInitialize", Type::VoidTy,
-                                         PointerType::get(Type::IntTy), 0);
+                                         PointerType::get(Type::IntTy),
+                                         (Type *)0);
 
     std::vector<Value *> trargs;
     trargs.push_back(threshold);