Add note about PhantomData and dropck

src/libs/maintaining-std.md

@@ -187,7 +187,41 @@ Changes to collection internals may affect the order their items are dropped in.
 
 ### Is there a manual `Drop` implementation?
 
-Generic types that manually implement `Drop` should consider whether a `#[may_dangle]` attribute is appropriate. The [Nomicon][dropck] has some details on what `#[may_dangle]` is all about.
+A generic `Type<T>` that manually implements `Drop` should consider whether a `#[may_dangle]` attribute is appropriate on `T`. The [Nomicon][dropck] has some details on what `#[may_dangle]` is all about.
+
+If a generic `Type<T>` has a manual drop implementation that may also involve dropping `T` then dropck needs to know about it. If `Type<T>`'s ownership of `T` is expressed through types that don't drop `T` themselves such as `ManuallyDrop<T>`, `*mut T`, or `MaybeUninit<T>` then `Type<T>` also [needs a `PhantomData<T>` field][RFC 0769 PhantomData] to tell dropck that `T` may be dropped. Types in the standard library that use the internal `Unique<T>` pointer type don't need a `PhantomData<T>` marker field. That's taken care of for them by `Unique<T>`.
+
+As a real-world example of where this can go wrong, consider an `OptionCell<T>` that looks something like this:
+
+```rust
+struct OptionCell<T> {
+    is_init: bool,
+    value: MaybeUninit<T>,
+}
+
+impl<T> Drop for OptionCell<T> {
+    fn drop(&mut self) {
+        if self.is_init {
+            // Safety: `value` is guaranteed to be fully initialized when `is_init` is true.
+            // Safety: The cell is being dropped, so it can't be accessed again.
+            drop(unsafe { self.value.read() });
+        }
+    }
+}
+```
+
+Adding a `#[may_dangle]` attribute to this `OptionCell<T>` that didn't have a `PhantomData<T>` marker field opened up [a soundness hole][rust/issues/76367] for `T`'s that didn't strictly outlive the `OptionCell<T>`, and so could be accessed after being dropped in their own `Drop` implementations. The correct application of `#[may_dangle]` also required a `PhantomData<T>` field:
+
+```diff
+struct OptionCell<T> {
+    is_init: bool,
+    value: MaybeUninit<T>,
++   _marker: PhantomData<T>,
+}
+
+- impl<T> Drop for OptionCell<T> {
++ unsafe impl<#[may_dangle] T> Drop for OptionCell<T> {
+```
 
 ### How could `mem` break assumptions?
 
@@ -260,7 +294,9 @@ Where `unsafe` and `const` is involved, e.g., for operations which are "unconst"
 [Forge]: https://forge.rust-lang.org/
 [RFC 1023]: https://rust-lang.github.io/rfcs/1023-rebalancing-coherence.html
 [RFC 1105]: https://rust-lang.github.io/rfcs/1105-api-evolution.html
+[RFC 0769 PhantomData]: https://github.com/rust-lang/rfcs/blob/master/text/0769-sound-generic-drop.md#phantom-data
 [Everyone Poops]: http://cglab.ca/~abeinges/blah/everyone-poops
 [rust/pull/46799]: https://github.com/rust-lang/rust/pull/46799
+[rust/issues/76367]: https://github.com/rust-lang/rust/issues/76367
 [hashbrown/pull/119]: https://github.com/rust-lang/hashbrown/pull/119
 [rollup guidelines]: ../compiler/reviews.md#rollups