[clang] [clang] LazyOffsetPtr: Use native pointer width (PR #111995)

[Jessica Clarke via cfe-commits]

================
@@ -326,25 +326,25 @@ struct LazyOffsetPtr {
   ///
   /// If the low bit is clear, a pointer to the AST node. If the low
   /// bit is set, the upper 63 bits are the offset.
-  mutable uint64_t Ptr = 0;
+  mutable uintptr_t Ptr = 0;
 
 public:
   LazyOffsetPtr() = default;
-  explicit LazyOffsetPtr(T *Ptr) : Ptr(reinterpret_cast<uint64_t>(Ptr)) {}
+  explicit LazyOffsetPtr(T *Ptr) : Ptr(reinterpret_cast<uintptr_t>(Ptr)) {}
 
-  explicit LazyOffsetPtr(uint64_t Offset) : Ptr((Offset << 1) | 0x01) {
-    assert((Offset << 1 >> 1) == Offset && "Offsets must require < 63 bits");
+  explicit LazyOffsetPtr(uintptr_t Offset) : Ptr((Offset << 1) | 0x01) {
+    assert((Offset << 1 >> 1) == Offset && "Offsets must fit in addressable bits");
     if (Offset == 0)
       Ptr = 0;
   }
 
   LazyOffsetPtr &operator=(T *Ptr) {
-    this->Ptr = reinterpret_cast<uint64_t>(Ptr);
+    this->Ptr = reinterpret_cast<uintptr_t>(Ptr);
     return *this;
   }
 
-  LazyOffsetPtr &operator=(uint64_t Offset) {
-    assert((Offset << 1 >> 1) == Offset && "Offsets must require < 63 bits");
+  LazyOffsetPtr &operator=(uintptr_t Offset) {
----------------
jrtc27 wrote:

No, it's not. Please re-read the problem description. Casts in the LazyOffsetPtr implementation aren't going to fix anything. The problem is that getAddressOfPointer pretends that the underlying storage is a pointer, when it's really a uint64_t, and on a 32-bit big-endian system `*(void *)&my_uint64 = p` and `my_uint64 = (uint64_t)(uintptr_t)p` store the contents of `p` in different halves of `my_uint64` (and zero the other half in the latter, but that isn't what's causing a problem). If you wanted to do something *really* disgusting you could implement `getAddressOfPointer` as `return (T **)&Ptr + 1`, I suppose. But ideally we wouldn't be doing anything as highly dodgy when it comes to strict aliasing as that.

https://github.com/llvm/llvm-project/pull/111995