Shrink from_raw_parts's MIR so that Vec::deref MIR-inlines again

library/core/src/ptr/internal_repr.rs

@@ -0,0 +1,74 @@
+//! This encapsulates the layout knowledge for pointers, only exporting the two
+//! safe functions that can be used to interact with the metadata directly.
+
+use super::{NonNull, Pointee, Thin};
+
+#[inline]
+pub(super) const fn metadata<P: RawPointer>(ptr: P) -> <P::Pointee as Pointee>::Metadata {
+    // SAFETY: Transmuting like this is safe since `P` and `PtrComponents<P, _>`
+    // have the same memory layouts. Only std can make this guarantee.
+    unsafe {
+        crate::intrinsics::transmute_unchecked::<
+            P,
+            PtrComponents<P::Family, <P::Pointee as Pointee>::Metadata>,
+        >(ptr)
+        .metadata
+    }
+}
+
+/// Just like [`from_raw_parts`] and [`from_raw_parts_mut`], but more flexible
+/// in terms of which types it can take, allowing smaller MIR.
+// See <https://github.com/rust-lang/rust/issues/123174>
+#[rustc_const_unstable(feature = "ptr_metadata", issue = "81513")]
+#[inline]
+pub(super) const fn from_raw_parts<P: RawPointer>(
+    data_pointer: impl RawPointer<Pointee: Thin, Family = P::Family>,
+    metadata: <P::Pointee as Pointee>::Metadata,
+) -> P {
+    // SAFETY: Transmuting like this is safe since `P` and `PtrComponents<P, _>`
+    // have the same memory layouts. Only std can make this guarantee.
+    unsafe {
+        crate::intrinsics::transmute_unchecked::<
+            PtrComponents<_, <P::Pointee as Pointee>::Metadata>,
+            P,
+        >(PtrComponents { data_pointer, metadata })
+    }
+}
+
+// Intentionally private with no derives, as it's only used via transmuting.
+// This layout is not stable; only std can rely on it.
+// (And should only do so in the two functions in this module.)
+#[repr(C)]
+struct PtrComponents<P, M = ()> {
+    data_pointer: P,
+    metadata: M,
+}
+
+/// Internal trait to avoid bad instantiations of [`PtrComponents`]
+///
+/// # Safety
+///
+/// Must have the same layout as `*const Self::Pointee` and be able to hold provenance.
+///
+/// Every type with the same associated `Family` must be soundly transmutable
+/// between each other when the metadata is the same.
+pub unsafe trait RawPointer: Copy {
+    type Pointee: ?Sized + super::Pointee;
+    type Family: RawPointer<Pointee = (), Family = Self::Family>;
+}
+
+// SAFETY: `*const T` is obviously a raw pointer
+unsafe impl<T: ?Sized> RawPointer for *const T {
+    type Pointee = T;
+    type Family = *const ();
+}
+// SAFETY: `*mut T` is obviously a raw pointer
+unsafe impl<T: ?Sized> RawPointer for *mut T {
+    type Pointee = T;
+    type Family = *mut ();
+}
+// SAFETY: `NonNull<T>` is a transparent newtype around a `*const T`.
+unsafe impl<T: ?Sized> RawPointer for NonNull<T> {
+    type Pointee = T;
+    type Family = NonNull<()>;
+}

library/core/src/ptr/metadata.rs

@@ -2,6 +2,7 @@
 
 use crate::fmt;
 use crate::hash::{Hash, Hasher};
+use crate::ptr::internal_repr;
 
 /// Provides the pointer metadata type of any pointed-to type.
 ///
@@ -92,10 +93,7 @@ pub trait Thin = Pointee<Metadata = ()>;
 #[rustc_const_unstable(feature = "ptr_metadata", issue = "81513")]
 #[inline]
 pub const fn metadata<T: ?Sized>(ptr: *const T) -> <T as Pointee>::Metadata {
-    // SAFETY: Accessing the value from the `PtrRepr` union is safe since *const T
-    // and PtrComponents<T> have the same memory layouts. Only std can make this
-    // guarantee.
-    unsafe { PtrRepr { const_ptr: ptr }.components.metadata }
+    internal_repr::metadata(ptr)
 }
 
 /// Forms a (possibly-wide) raw pointer from a data pointer and metadata.
@@ -112,10 +110,7 @@ pub const fn from_raw_parts<T: ?Sized>(
     data_pointer: *const (),
     metadata: <T as Pointee>::Metadata,
 ) -> *const T {
-    // SAFETY: Accessing the value from the `PtrRepr` union is safe since *const T
-    // and PtrComponents<T> have the same memory layouts. Only std can make this
-    // guarantee.
-    unsafe { PtrRepr { components: PtrComponents { data_pointer, metadata } }.const_ptr }
+    internal_repr::from_raw_parts(data_pointer, metadata)
 }
 
 /// Performs the same functionality as [`from_raw_parts`], except that a
@@ -129,33 +124,7 @@ pub const fn from_raw_parts_mut<T: ?Sized>(
     data_pointer: *mut (),
     metadata: <T as Pointee>::Metadata,
 ) -> *mut T {
-    // SAFETY: Accessing the value from the `PtrRepr` union is safe since *const T
-    // and PtrComponents<T> have the same memory layouts. Only std can make this
-    // guarantee.
-    unsafe { PtrRepr { components: PtrComponents { data_pointer, metadata } }.mut_ptr }
-}
-
-#[repr(C)]
-union PtrRepr<T: ?Sized> {
-    const_ptr: *const T,
-    mut_ptr: *mut T,
-    components: PtrComponents<T>,
-}
-
-#[repr(C)]
-struct PtrComponents<T: ?Sized> {
-    data_pointer: *const (),
-    metadata: <T as Pointee>::Metadata,
-}
-
-// Manual impl needed to avoid `T: Copy` bound.
-impl<T: ?Sized> Copy for PtrComponents<T> {}
-
-// Manual impl needed to avoid `T: Clone` bound.
-impl<T: ?Sized> Clone for PtrComponents<T> {
-    fn clone(&self) -> Self {
-        *self
-    }
+    internal_repr::from_raw_parts(data_pointer, metadata)
 }
 
 /// The metadata for a `Dyn = dyn SomeTrait` trait object type.

library/core/src/ptr/mod.rs

@@ -410,6 +410,8 @@ pub use crate::intrinsics::copy;
 #[doc(inline)]
 pub use crate::intrinsics::write_bytes;
 
+mod internal_repr;
+
 mod metadata;
 #[unstable(feature = "ptr_metadata", issue = "81513")]
 pub use metadata::{from_raw_parts, from_raw_parts_mut, metadata, DynMetadata, Pointee, Thin};
@@ -812,7 +814,7 @@ pub const fn from_mut<T: ?Sized>(r: &mut T) -> *mut T {
 #[rustc_allow_const_fn_unstable(ptr_metadata)]
 #[rustc_diagnostic_item = "ptr_slice_from_raw_parts"]
 pub const fn slice_from_raw_parts<T>(data: *const T, len: usize) -> *const [T] {
-    from_raw_parts(data.cast(), len)
+    internal_repr::from_raw_parts(data, len)
 }
 
 /// Forms a raw mutable slice from a pointer and a length.
@@ -858,7 +860,7 @@ pub const fn slice_from_raw_parts<T>(data: *const T, len: usize) -> *const [T] {
 #[rustc_const_unstable(feature = "const_slice_from_raw_parts_mut", issue = "67456")]
 #[rustc_diagnostic_item = "ptr_slice_from_raw_parts_mut"]
 pub const fn slice_from_raw_parts_mut<T>(data: *mut T, len: usize) -> *mut [T] {
-    from_raw_parts_mut(data.cast(), len)
+    internal_repr::from_raw_parts(data, len)
 }
 
 /// Swaps the values at two mutable locations of the same type, without

library/core/src/ptr/non_null.rs

@@ -6,8 +6,7 @@ use crate::marker::Unsize;
 use crate::mem::{MaybeUninit, SizedTypeProperties};
 use crate::num::NonZero;
 use crate::ops::{CoerceUnsized, DispatchFromDyn};
-use crate::ptr;
-use crate::ptr::Unique;
+use crate::ptr::{self, internal_repr, Unique};
 use crate::slice::{self, SliceIndex};
 use crate::ub_checks::assert_unsafe_precondition;
 
@@ -265,10 +264,7 @@ impl<T: ?Sized> NonNull<T> {
         data_pointer: NonNull<()>,
         metadata: <T as super::Pointee>::Metadata,
     ) -> NonNull<T> {
-        // SAFETY: The result of `ptr::from::raw_parts_mut` is non-null because `data_pointer` is.
-        unsafe {
-            NonNull::new_unchecked(super::from_raw_parts_mut(data_pointer.as_ptr(), metadata))
-        }
+        internal_repr::from_raw_parts(data_pointer, metadata)
     }
 
     /// Decompose a (possibly wide) pointer into its data pointer and metadata components.

tests/mir-opt/pre-codegen/vec_deref.vec_deref_to_slice.PreCodegen.after.panic-abort.mir

@@ -3,12 +3,73 @@
 fn vec_deref_to_slice(_1: &Vec<u8>) -> &[u8] {
     debug v => _1;
     let mut _0: &[u8];
-
-    bb0: {
-        _0 = <Vec<u8> as Deref>::deref(move _1) -> [return: bb1, unwind unreachable];
+    scope 1 (inlined <Vec<u8> as Deref>::deref) {
+        debug self => _1;
+        let mut _4: *const u8;
+        let mut _5: usize;
+        scope 2 {
+            scope 3 (inlined Vec::<u8>::as_ptr) {
+                debug self => _1;
+                let mut _2: &alloc::raw_vec::RawVec<u8>;
+                scope 4 (inlined alloc::raw_vec::RawVec::<u8>::ptr) {
+                    debug self => _2;
+                    let mut _3: std::ptr::NonNull<u8>;
+                    scope 5 (inlined Unique::<u8>::as_ptr) {
+                        debug ((self: Unique<u8>).0: std::ptr::NonNull<u8>) => _3;
+                        debug ((self: Unique<u8>).1: std::marker::PhantomData<u8>) => const PhantomData::<u8>;
+                        scope 6 (inlined NonNull::<u8>::as_ptr) {
+                            debug self => _3;
+                        }
+                    }
+                }
+            }
+            scope 7 (inlined std::slice::from_raw_parts::<'_, u8>) {
+                debug data => _4;
+                debug len => _5;
+                let _7: *const [u8];
+                scope 8 {
+                    scope 9 (inlined core::ub_checks::check_language_ub) {
+                        scope 10 (inlined core::ub_checks::check_language_ub::runtime) {
+                        }
+                    }
+                    scope 11 (inlined std::mem::size_of::<u8>) {
+                    }
+                    scope 12 (inlined align_of::<u8>) {
+                    }
+                    scope 13 (inlined slice_from_raw_parts::<u8>) {
+                        debug data => _4;
+                        debug len => _5;
+                        scope 14 (inlined std::ptr::internal_repr::from_raw_parts::<*const [u8], *const u8>) {
+                            debug data_pointer => _4;
+                            debug metadata => _5;
+                            let mut _6: std::ptr::internal_repr::PtrComponents<*const u8, usize>;
+                            scope 15 {
+                            }
+                        }
+                    }
+                }
+            }
+        }
     }
 
-    bb1: {
+    bb0: {
+        StorageLive(_4);
+        StorageLive(_2);
+        _2 = &((*_1).0: alloc::raw_vec::RawVec<u8>);
+        StorageLive(_3);
+        _3 = ((((*_1).0: alloc::raw_vec::RawVec<u8>).0: std::ptr::Unique<u8>).0: std::ptr::NonNull<u8>);
+        _4 = (_3.0: *const u8);
+        StorageDead(_3);
+        StorageDead(_2);
+        StorageLive(_5);
+        _5 = ((*_1).1: usize);
+        StorageLive(_6);
+        _6 = std::ptr::internal_repr::PtrComponents::<*const u8, usize> { data_pointer: _4, metadata: _5 };
+        _7 = move _6 as *const [u8] (Transmute);
+        StorageDead(_6);
+        StorageDead(_5);
+        StorageDead(_4);
+        _0 = &(*_7);
         return;
     }
 }

tests/mir-opt/pre-codegen/vec_deref.vec_deref_to_slice.PreCodegen.after.panic-unwind.mir

@@ -3,12 +3,73 @@
 fn vec_deref_to_slice(_1: &Vec<u8>) -> &[u8] {
     debug v => _1;
     let mut _0: &[u8];
-
-    bb0: {
-        _0 = <Vec<u8> as Deref>::deref(move _1) -> [return: bb1, unwind continue];
+    scope 1 (inlined <Vec<u8> as Deref>::deref) {
+        debug self => _1;
+        let mut _4: *const u8;
+        let mut _5: usize;
+        scope 2 {
+            scope 3 (inlined Vec::<u8>::as_ptr) {
+                debug self => _1;
+                let mut _2: &alloc::raw_vec::RawVec<u8>;
+                scope 4 (inlined alloc::raw_vec::RawVec::<u8>::ptr) {
+                    debug self => _2;
+                    let mut _3: std::ptr::NonNull<u8>;
+                    scope 5 (inlined Unique::<u8>::as_ptr) {
+                        debug ((self: Unique<u8>).0: std::ptr::NonNull<u8>) => _3;
+                        debug ((self: Unique<u8>).1: std::marker::PhantomData<u8>) => const PhantomData::<u8>;
+                        scope 6 (inlined NonNull::<u8>::as_ptr) {
+                            debug self => _3;
+                        }
+                    }
+                }
+            }
+            scope 7 (inlined std::slice::from_raw_parts::<'_, u8>) {
+                debug data => _4;
+                debug len => _5;
+                let _7: *const [u8];
+                scope 8 {
+                    scope 9 (inlined core::ub_checks::check_language_ub) {
+                        scope 10 (inlined core::ub_checks::check_language_ub::runtime) {
+                        }
+                    }
+                    scope 11 (inlined std::mem::size_of::<u8>) {
+                    }
+                    scope 12 (inlined align_of::<u8>) {
+                    }
+                    scope 13 (inlined slice_from_raw_parts::<u8>) {
+                        debug data => _4;
+                        debug len => _5;
+                        scope 14 (inlined std::ptr::internal_repr::from_raw_parts::<*const [u8], *const u8>) {
+                            debug data_pointer => _4;
+                            debug metadata => _5;
+                            let mut _6: std::ptr::internal_repr::PtrComponents<*const u8, usize>;
+                            scope 15 {
+                            }
+                        }
+                    }
+                }
+            }
+        }
     }
 
-    bb1: {
+    bb0: {
+        StorageLive(_4);
+        StorageLive(_2);
+        _2 = &((*_1).0: alloc::raw_vec::RawVec<u8>);
+        StorageLive(_3);
+        _3 = ((((*_1).0: alloc::raw_vec::RawVec<u8>).0: std::ptr::Unique<u8>).0: std::ptr::NonNull<u8>);
+        _4 = (_3.0: *const u8);
+        StorageDead(_3);
+        StorageDead(_2);
+        StorageLive(_5);
+        _5 = ((*_1).1: usize);
+        StorageLive(_6);
+        _6 = std::ptr::internal_repr::PtrComponents::<*const u8, usize> { data_pointer: _4, metadata: _5 };
+        _7 = move _6 as *const [u8] (Transmute);
+        StorageDead(_6);
+        StorageDead(_5);
+        StorageDead(_4);
+        _0 = &(*_7);
         return;
     }
 }