Merge from rustc

Author: RalfJung

Cargo.lock

@@ -1,3 +1,5 @@
+cargo-features = ["public-dependency"]
+
 [package]
 name = "alloc"
 version = "0.0.0"
@@ -9,8 +11,8 @@ autobenches = false
 edition = "2021"
 
 [dependencies]
-core = { path = "../core" }
-compiler_builtins = { version = "=0.1.146", features = ['rustc-dep-of-std'] }
+core = { path = "../core", public = true }
+compiler_builtins = { version = "=0.1.147", features = ['rustc-dep-of-std'] }
 
 [dev-dependencies]
 rand = { version = "0.9.0", default-features = false, features = ["alloc"] }

alloc/Cargo.toml

@@ -237,23 +237,13 @@ pub struct Box<
 /// the newly allocated memory. This is an intrinsic to avoid unnecessary copies.
 ///
 /// This is the surface syntax for `box <expr>` expressions.
-#[cfg(not(bootstrap))]
 #[rustc_intrinsic]
 #[rustc_intrinsic_must_be_overridden]
 #[unstable(feature = "liballoc_internals", issue = "none")]
 pub fn box_new<T>(_x: T) -> Box<T> {
     unreachable!()
 }
 
-/// Transition function for the next bootstrap bump.
-#[cfg(bootstrap)]
-#[unstable(feature = "liballoc_internals", issue = "none")]
-#[inline(always)]
-pub fn box_new<T>(x: T) -> Box<T> {
-    #[rustc_box]
-    Box::new(x)
-}
-
 impl<T> Box<T> {
     /// Allocates memory on the heap and then places `x` into it.
     ///
@@ -1689,7 +1679,20 @@ impl<T: Default> Default for Box<T> {
     /// Creates a `Box<T>`, with the `Default` value for T.
     #[inline]
     fn default() -> Self {
-        Box::write(Box::new_uninit(), T::default())
+        let mut x: Box<mem::MaybeUninit<T>> = Box::new_uninit();
+        unsafe {
+            // SAFETY: `x` is valid for writing and has the same layout as `T`.
+            // If `T::default()` panics, dropping `x` will just deallocate the Box as `MaybeUninit<T>`
+            // does not have a destructor.
+            //
+            // We use `ptr::write` as `MaybeUninit::write` creates
+            // extra stack copies of `T` in debug mode.
+            //
+            // See https://github.com/rust-lang/rust/issues/136043 for more context.
+            ptr::write(&raw mut *x as *mut T, T::default());
+            // SAFETY: `x` was just initialized above.
+            x.assume_init()
+        }
     }
 }
 

alloc/src/boxed.rs

@@ -105,6 +105,7 @@
 #![feature(box_uninit_write)]
 #![feature(bstr)]
 #![feature(bstr_internals)]
+#![feature(char_max_len)]
 #![feature(clone_to_uninit)]
 #![feature(coerce_unsized)]
 #![feature(const_eval_select)]
@@ -136,7 +137,6 @@
 #![feature(pointer_like_trait)]
 #![feature(ptr_internals)]
 #![feature(ptr_metadata)]
-#![feature(ptr_sub_ptr)]
 #![feature(set_ptr_value)]
 #![feature(sized_type_properties)]
 #![feature(slice_from_ptr_range)]

alloc/src/lib.rs

@@ -27,7 +27,7 @@ pub use core::slice::ArrayChunksMut;
 pub use core::slice::ArrayWindows;
 #[stable(feature = "inherent_ascii_escape", since = "1.60.0")]
 pub use core::slice::EscapeAscii;
-#[stable(feature = "get_many_mut", since = "CURRENT_RUSTC_VERSION")]
+#[stable(feature = "get_many_mut", since = "1.86.0")]
 pub use core::slice::GetDisjointMutError;
 #[stable(feature = "slice_get_slice", since = "1.28.0")]
 pub use core::slice::SliceIndex;

alloc/src/slice.rs

@@ -1419,7 +1419,9 @@ impl String {
     pub fn push(&mut self, ch: char) {
         match ch.len_utf8() {
             1 => self.vec.push(ch as u8),
-            _ => self.vec.extend_from_slice(ch.encode_utf8(&mut [0; 4]).as_bytes()),
+            _ => {
+                self.vec.extend_from_slice(ch.encode_utf8(&mut [0; char::MAX_LEN_UTF8]).as_bytes())
+            }
         }
     }
 
@@ -1716,7 +1718,7 @@ impl String {
     #[rustc_confusables("set")]
     pub fn insert(&mut self, idx: usize, ch: char) {
         assert!(self.is_char_boundary(idx));
-        let mut bits = [0; 4];
+        let mut bits = [0; char::MAX_LEN_UTF8];
         let bits = ch.encode_utf8(&mut bits).as_bytes();
 
         unsafe {
@@ -2797,7 +2799,7 @@ impl SpecToString for core::ascii::Char {
 impl SpecToString for char {
     #[inline]
     fn spec_to_string(&self) -> String {
-        String::from(self.encode_utf8(&mut [0; 4]))
+        String::from(self.encode_utf8(&mut [0; char::MAX_LEN_UTF8]))
     }
 }
 
@@ -3155,6 +3157,24 @@ impl From<String> for Vec<u8> {
     }
 }
 
+#[stable(feature = "try_from_vec_u8_for_string", since = "CURRENT_RUSTC_VERSION")]
+impl TryFrom<Vec<u8>> for String {
+    type Error = FromUtf8Error;
+    /// Converts the given [`Vec<u8>`] into a  [`String`] if it contains valid UTF-8 data.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// let s1 = b"hello world".to_vec();
+    /// let v1 = String::try_from(s1).unwrap();
+    /// assert_eq!(v1, "hello world");
+    ///
+    /// ```
+    fn try_from(bytes: Vec<u8>) -> Result<Self, Self::Error> {
+        Self::from_utf8(bytes)
+    }
+}
+
 #[cfg(not(no_global_oom_handling))]
 #[stable(feature = "rust1", since = "1.0.0")]
 impl fmt::Write for String {

alloc/src/string.rs

@@ -232,7 +232,7 @@ impl<T, A: Allocator> Drop for Drain<'_, T, A> {
             // it from the original vec but also avoid creating a &mut to the front since that could
             // invalidate raw pointers to it which some unsafe code might rely on.
             let vec_ptr = vec.as_mut().as_mut_ptr();
-            let drop_offset = drop_ptr.sub_ptr(vec_ptr);
+            let drop_offset = drop_ptr.offset_from_unsigned(vec_ptr);
             let to_drop = ptr::slice_from_raw_parts_mut(vec_ptr.add(drop_offset), drop_len);
             ptr::drop_in_place(to_drop);
         }

alloc/src/vec/drain.rs

@@ -379,7 +379,7 @@ where
         let sink =
             self.try_fold::<_, _, Result<_, !>>(sink, write_in_place_with_drop(end)).into_ok();
         // iteration succeeded, don't drop head
-        unsafe { ManuallyDrop::new(sink).dst.sub_ptr(dst_buf) }
+        unsafe { ManuallyDrop::new(sink).dst.offset_from_unsigned(dst_buf) }
     }
 }
 

alloc/src/vec/in_place_collect.rs

@@ -14,7 +14,7 @@ pub(super) struct InPlaceDrop<T> {
 
 impl<T> InPlaceDrop<T> {
     fn len(&self) -> usize {
-        unsafe { self.dst.sub_ptr(self.inner) }
+        unsafe { self.dst.offset_from_unsigned(self.inner) }
     }
 }
 

alloc/src/vec/in_place_drop.rs

@@ -179,7 +179,7 @@ impl<T, A: Allocator> IntoIter<T, A> {
                 // say that they're all at the beginning of the "allocation".
                 0..this.len()
             } else {
-                this.ptr.sub_ptr(this.buf)..this.end.sub_ptr(buf)
+                this.ptr.offset_from_unsigned(this.buf)..this.end.offset_from_unsigned(buf)
             };
             let cap = this.cap;
             let alloc = ManuallyDrop::take(&mut this.alloc);
@@ -230,7 +230,7 @@ impl<T, A: Allocator> Iterator for IntoIter<T, A> {
         let exact = if T::IS_ZST {
             self.end.addr().wrapping_sub(self.ptr.as_ptr().addr())
         } else {
-            unsafe { non_null!(self.end, T).sub_ptr(self.ptr) }
+            unsafe { non_null!(self.end, T).offset_from_unsigned(self.ptr) }
         };
         (exact, Some(exact))
     }

alloc/src/vec/into_iter.rs

@@ -2526,7 +2526,7 @@ impl<T, A: Allocator> Vec<T, A> {
     /// assert_eq!(vec, [1, 2, 3]);
     /// assert_eq!(vec.pop_if(pred), None);
     /// ```
-    #[stable(feature = "vec_pop_if", since = "CURRENT_RUSTC_VERSION")]
+    #[stable(feature = "vec_pop_if", since = "1.86.0")]
     pub fn pop_if(&mut self, predicate: impl FnOnce(&mut T) -> bool) -> Option<T> {
         let last = self.last_mut()?;
         if predicate(last) { self.pop() } else { None }

alloc/src/vec/mod.rs

@@ -3,6 +3,7 @@
 #![feature(iter_array_chunks)]
 #![feature(assert_matches)]
 #![feature(btree_extract_if)]
+#![feature(char_max_len)]
 #![feature(cow_is_borrowed)]
 #![feature(core_intrinsics)]
 #![feature(downcast_unchecked)]

alloc/tests/lib.rs

@@ -2,6 +2,7 @@
 
 use std::assert_matches::assert_matches;
 use std::borrow::Cow;
+use std::char::MAX_LEN_UTF8;
 use std::cmp::Ordering::{Equal, Greater, Less};
 use std::str::{from_utf8, from_utf8_unchecked};
 
@@ -1231,7 +1232,7 @@ fn test_to_uppercase_rev_iterator() {
 #[test]
 #[cfg_attr(miri, ignore)] // Miri is too slow
 fn test_chars_decoding() {
-    let mut bytes = [0; 4];
+    let mut bytes = [0; MAX_LEN_UTF8];
     for c in (0..0x110000).filter_map(std::char::from_u32) {
         let s = c.encode_utf8(&mut bytes);
         if Some(c) != s.chars().next() {
@@ -1243,7 +1244,7 @@ fn test_chars_decoding() {
 #[test]
 #[cfg_attr(miri, ignore)] // Miri is too slow
 fn test_chars_rev_decoding() {
-    let mut bytes = [0; 4];
+    let mut bytes = [0; MAX_LEN_UTF8];
     for c in (0..0x110000).filter_map(std::char::from_u32) {
         let s = c.encode_utf8(&mut bytes);
         if Some(c) != s.chars().rev().next() {

alloc/tests/str.rs

@@ -23,14 +23,15 @@ optimize_for_size = []
 # Make `RefCell` store additional debugging information, which is printed out when
 # a borrow error occurs
 debug_refcell = []
+# Make `TypeId` store a reference to the name of the type, so that it can print that name.
+debug_typeid = []
 
 [lints.rust.unexpected_cfgs]
 level = "warn"
 check-cfg = [
     'cfg(bootstrap)',
     'cfg(no_fp_fmt_parse)',
     'cfg(stdarch_intel_sde)',
-    'cfg(target_arch, values("xtensa"))',
     # core use #[path] imports to portable-simd `core_simd` crate
     # and to stdarch `core_arch` crate which messes-up with Cargo list
     # of declared features, we therefor expect any feature cfg

core/Cargo.toml

@@ -711,6 +711,8 @@ pub struct TypeId {
     // We avoid using `u128` because that imposes higher alignment requirements on many platforms.
     // See issue #115620 for more information.
     t: (u64, u64),
+    #[cfg(feature = "debug_typeid")]
+    name: &'static str,
 }
 
 #[stable(feature = "rust1", since = "1.0.0")]
@@ -741,10 +743,14 @@ impl TypeId {
     #[rustc_const_unstable(feature = "const_type_id", issue = "77125")]
     pub const fn of<T: ?Sized + 'static>() -> TypeId {
         let t: u128 = intrinsics::type_id::<T>();
-
         let t1 = (t >> 64) as u64;
         let t2 = t as u64;
-        TypeId { t: (t1, t2) }
+
+        TypeId {
+            t: (t1, t2),
+            #[cfg(feature = "debug_typeid")]
+            name: type_name::<T>(),
+        }
     }
 
     fn as_u128(self) -> u128 {
@@ -775,7 +781,15 @@ impl hash::Hash for TypeId {
 #[stable(feature = "rust1", since = "1.0.0")]
 impl fmt::Debug for TypeId {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> Result<(), fmt::Error> {
-        write!(f, "TypeId({:#034x})", self.as_u128())
+        #[cfg(feature = "debug_typeid")]
+        {
+            write!(f, "TypeId({:#034x} = {})", self.as_u128(), self.name)?;
+        }
+        #[cfg(not(feature = "debug_typeid"))]
+        {
+            write!(f, "TypeId({:#034x})", self.as_u128())?;
+        }
+        Ok(())
     }
 }
 

core/src/any.rs

@@ -71,6 +71,16 @@ impl char {
     #[stable(feature = "assoc_char_consts", since = "1.52.0")]
     pub const MAX: char = '\u{10FFFF}';
 
+    /// The maximum number of bytes required to [encode](char::encode_utf8) a `char` to
+    /// UTF-8 encoding.
+    #[unstable(feature = "char_max_len", issue = "121714")]
+    pub const MAX_LEN_UTF8: usize = 4;
+
+    /// The maximum number of two-byte units required to [encode](char::encode_utf16) a `char`
+    /// to UTF-16 encoding.
+    #[unstable(feature = "char_max_len", issue = "121714")]
+    pub const MAX_LEN_UTF16: usize = 2;
+
     /// `U+FFFD REPLACEMENT CHARACTER` (�) is used in Unicode to represent a
     /// decoding error.
     ///

core/src/char/methods.rs

@@ -95,6 +95,16 @@ const MAX_THREE_B: u32 = 0x10000;
 #[stable(feature = "rust1", since = "1.0.0")]
 pub const MAX: char = char::MAX;
 
+/// The maximum number of bytes required to [encode](char::encode_utf8) a `char` to
+/// UTF-8 encoding.
+#[unstable(feature = "char_max_len", issue = "121714")]
+pub const MAX_LEN_UTF8: usize = char::MAX_LEN_UTF8;
+
+/// The maximum number of two-byte units required to [encode](char::encode_utf16) a `char`
+/// to UTF-16 encoding.
+#[unstable(feature = "char_max_len", issue = "121714")]
+pub const MAX_LEN_UTF16: usize = char::MAX_LEN_UTF16;
+
 /// `U+FFFD REPLACEMENT CHARACTER` (�) is used in Unicode to represent a
 /// decoding error. Use [`char::REPLACEMENT_CHARACTER`] instead.
 #[stable(feature = "decode_utf16", since = "1.9.0")]

core/src/char/mod.rs

@@ -1,5 +1,4 @@
 //! Unstable module containing the unstable contracts lang items and attribute macros.
-#![cfg(not(bootstrap))]
 
 pub use crate::macros::builtin::{contracts_ensures as ensures, contracts_requires as requires};
 

core/src/contracts.rs

@@ -1099,5 +1099,5 @@ impl Error for crate::time::TryFromFloatSecsError {}
 #[stable(feature = "cstr_from_bytes_until_nul", since = "1.69.0")]
 impl Error for crate::ffi::FromBytesUntilNulError {}
 
-#[stable(feature = "get_many_mut", since = "CURRENT_RUSTC_VERSION")]
+#[stable(feature = "get_many_mut", since = "1.86.0")]
 impl Error for crate::slice::GetDisjointMutError {}

core/src/error.rs

@@ -3,7 +3,7 @@
 #![stable(feature = "rust1", since = "1.0.0")]
 
 use crate::cell::{Cell, Ref, RefCell, RefMut, SyncUnsafeCell, UnsafeCell};
-use crate::char::EscapeDebugExtArgs;
+use crate::char::{EscapeDebugExtArgs, MAX_LEN_UTF8};
 use crate::marker::PhantomData;
 use crate::num::fmt as numfmt;
 use crate::ops::Deref;
@@ -187,7 +187,7 @@ pub trait Write {
     /// ```
     #[stable(feature = "fmt_write_char", since = "1.1.0")]
     fn write_char(&mut self, c: char) -> Result {
-        self.write_str(c.encode_utf8(&mut [0; 4]))
+        self.write_str(c.encode_utf8(&mut [0; MAX_LEN_UTF8]))
     }
 
     /// Glue for usage of the [`write!`] macro with implementors of this trait.
@@ -2768,7 +2768,7 @@ impl Display for char {
         if f.options.width.is_none() && f.options.precision.is_none() {
             f.write_char(*self)
         } else {
-            f.pad(self.encode_utf8(&mut [0; 4]))
+            f.pad(self.encode_utf8(&mut [0; MAX_LEN_UTF8]))
         }
     }
 }

core/src/fmt/mod.rs

@@ -472,7 +472,7 @@ pub fn spin_loop() {
 /// During constant evaluation, `black_box` is treated as a no-op.
 #[inline]
 #[stable(feature = "bench_black_box", since = "1.66.0")]
-#[rustc_const_stable(feature = "const_black_box", since = "CURRENT_RUSTC_VERSION")]
+#[rustc_const_stable(feature = "const_black_box", since = "1.86.0")]
 pub const fn black_box<T>(dummy: T) -> T {
     crate::intrinsics::black_box(dummy)
 }