diff --git a/library/alloc/src/lib.rs b/library/alloc/src/lib.rs
index 73197d021f1a3..bd0ee36383b87 100644
--- a/library/alloc/src/lib.rs
+++ b/library/alloc/src/lib.rs
@@ -99,7 +99,6 @@
 #![feature(cast_maybe_uninit)]
 #![feature(cell_get_cloned)]
 #![feature(char_internals)]
-#![feature(char_max_len)]
 #![feature(clone_to_uninit)]
 #![feature(coerce_unsized)]
 #![feature(const_convert)]
diff --git a/library/alloc/src/wtf8/mod.rs b/library/alloc/src/wtf8/mod.rs
index 047994adc4486..e4834a24bf430 100644
--- a/library/alloc/src/wtf8/mod.rs
+++ b/library/alloc/src/wtf8/mod.rs
@@ -14,7 +14,7 @@
 #[cfg(test)]
 mod tests;
 
-use core::char::{MAX_LEN_UTF8, encode_utf8_raw};
+use core::char::encode_utf8_raw;
 use core::hash::{Hash, Hasher};
 pub use core::wtf8::{CodePoint, Wtf8};
 #[cfg(not(test))]
@@ -166,7 +166,7 @@ impl Wtf8Buf {
     /// This does **not** include the WTF-8 concatenation check or `is_known_utf8` check.
     /// Copied from String::push.
     unsafe fn push_code_point_unchecked(&mut self, code_point: CodePoint) {
-        let mut bytes = [0; MAX_LEN_UTF8];
+        let mut bytes = [0; char::MAX_LEN_UTF8];
         let bytes = encode_utf8_raw(code_point.to_u32(), &mut bytes);
         self.bytes.extend_from_slice(bytes)
     }
diff --git a/library/alloctests/lib.rs b/library/alloctests/lib.rs
index efdcb893bfeef..8c1942432533e 100644
--- a/library/alloctests/lib.rs
+++ b/library/alloctests/lib.rs
@@ -19,7 +19,6 @@
 #![feature(array_into_iter_constructors)]
 #![feature(assert_matches)]
 #![feature(char_internals)]
-#![feature(char_max_len)]
 #![feature(copied_into_inner)]
 #![feature(core_intrinsics)]
 #![feature(exact_size_is_empty)]
diff --git a/library/alloctests/tests/lib.rs b/library/alloctests/tests/lib.rs
index c2649be0558a1..7ad16a953e9b2 100644
--- a/library/alloctests/tests/lib.rs
+++ b/library/alloctests/tests/lib.rs
@@ -3,7 +3,6 @@
 #![feature(iter_array_chunks)]
 #![feature(assert_matches)]
 #![feature(wtf8_internals)]
-#![feature(char_max_len)]
 #![feature(cow_is_borrowed)]
 #![feature(core_intrinsics)]
 #![feature(deque_extend_front)]
diff --git a/library/alloctests/tests/str.rs b/library/alloctests/tests/str.rs
index 906fa2d425e77..fbb3b01fd67f9 100644
--- a/library/alloctests/tests/str.rs
+++ b/library/alloctests/tests/str.rs
@@ -2,7 +2,6 @@
 
 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};
 
@@ -1232,7 +1231,7 @@ fn test_to_uppercase_rev_iterator() {
 #[test]
 #[cfg_attr(miri, ignore)] // Miri is too slow
 fn test_chars_decoding() {
-    let mut bytes = [0; MAX_LEN_UTF8];
+    let mut bytes = [0; char::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() {
@@ -1244,7 +1243,7 @@ fn test_chars_decoding() {
 #[test]
 #[cfg_attr(miri, ignore)] // Miri is too slow
 fn test_chars_rev_decoding() {
-    let mut bytes = [0; MAX_LEN_UTF8];
+    let mut bytes = [0; char::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() {
diff --git a/library/core/src/char/methods.rs b/library/core/src/char/methods.rs
index 76f54db287079..d1de2c5606154 100644
--- a/library/core/src/char/methods.rs
+++ b/library/core/src/char/methods.rs
@@ -74,12 +74,12 @@ impl char {
 
     /// The maximum number of bytes required to [encode](char::encode_utf8) a `char` to
     /// UTF-8 encoding.
-    #[unstable(feature = "char_max_len", issue = "121714")]
+    #[stable(feature = "char_max_len_assoc", since = "CURRENT_RUSTC_VERSION")]
     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")]
+    #[stable(feature = "char_max_len_assoc", since = "CURRENT_RUSTC_VERSION")]
     pub const MAX_LEN_UTF16: usize = 2;
 
     /// `U+FFFD REPLACEMENT CHARACTER` (�) is used in Unicode to represent a
diff --git a/library/core/src/fmt/mod.rs b/library/core/src/fmt/mod.rs
index 502ca4aefe10d..38811a099aa1e 100644
--- a/library/core/src/fmt/mod.rs
+++ b/library/core/src/fmt/mod.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, MAX_LEN_UTF8};
+use crate::char::EscapeDebugExtArgs;
 use crate::marker::{PhantomData, PointeeSized};
 use crate::num::fmt as numfmt;
 use crate::ops::Deref;
@@ -179,7 +179,7 @@ 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; MAX_LEN_UTF8]))
+        self.write_str(c.encode_utf8(&mut [0; char::MAX_LEN_UTF8]))
     }
 
     /// Glue for usage of the [`write!`] macro with implementors of this trait.
@@ -2775,7 +2775,7 @@ impl Display for char {
         if f.options.flags & (flags::WIDTH_FLAG | flags::PRECISION_FLAG) == 0 {
             f.write_char(*self)
         } else {
-            f.pad(self.encode_utf8(&mut [0; MAX_LEN_UTF8]))
+            f.pad(self.encode_utf8(&mut [0; char::MAX_LEN_UTF8]))
         }
     }
 }
diff --git a/library/core/src/str/pattern.rs b/library/core/src/str/pattern.rs
index e116b13838323..b54522fcc886f 100644
--- a/library/core/src/str/pattern.rs
+++ b/library/core/src/str/pattern.rs
@@ -38,7 +38,6 @@
     issue = "27721"
 )]
 
-use crate::char::MAX_LEN_UTF8;
 use crate::cmp::Ordering;
 use crate::convert::TryInto as _;
 use crate::slice::memchr;
@@ -563,7 +562,7 @@ impl Pattern for char {
 
     #[inline]
     fn into_searcher<'a>(self, haystack: &'a str) -> Self::Searcher<'a> {
-        let mut utf8_encoded = [0; MAX_LEN_UTF8];
+        let mut utf8_encoded = [0; char::MAX_LEN_UTF8];
         let utf8_size = self
             .encode_utf8(&mut utf8_encoded)
             .len()
diff --git a/library/core/src/wtf8.rs b/library/core/src/wtf8.rs
index 11cd2b8776f22..b64fccedc19cd 100644
--- a/library/core/src/wtf8.rs
+++ b/library/core/src/wtf8.rs
@@ -19,7 +19,7 @@
 // implementations, so, we'll have to add more doc(hidden)s anyway
 #![doc(hidden)]
 
-use crate::char::{MAX_LEN_UTF16, encode_utf16_raw};
+use crate::char::encode_utf16_raw;
 use crate::clone::CloneToUninit;
 use crate::fmt::{self, Write};
 use crate::hash::{Hash, Hasher};
@@ -541,7 +541,7 @@ impl Iterator for EncodeWide<'_> {
             return Some(tmp);
         }
 
-        let mut buf = [0; MAX_LEN_UTF16];
+        let mut buf = [0; char::MAX_LEN_UTF16];
         self.code_points.next().map(|code_point| {
             let n = encode_utf16_raw(code_point.to_u32(), &mut buf).len();
             if n == 2 {
diff --git a/library/coretests/tests/char.rs b/library/coretests/tests/char.rs
index 6f94065b2d927..f0f6a24429284 100644
--- a/library/coretests/tests/char.rs
+++ b/library/coretests/tests/char.rs
@@ -1,4 +1,3 @@
-use std::char::MAX_LEN_UTF8;
 use std::str::FromStr;
 use std::{char, str};
 
@@ -259,7 +258,7 @@ fn test_escape_unicode() {
 #[test]
 fn test_encode_utf8() {
     fn check(input: char, expect: &[u8]) {
-        let mut buf = [0; MAX_LEN_UTF8];
+        let mut buf = [0; char::MAX_LEN_UTF8];
         let ptr = buf.as_ptr();
         let s = input.encode_utf8(&mut buf);
         assert_eq!(s.as_ptr() as usize, ptr as usize);
diff --git a/library/std/src/fs/tests.rs b/library/std/src/fs/tests.rs
index 0517760c35501..9fd87e119906e 100644
--- a/library/std/src/fs/tests.rs
+++ b/library/std/src/fs/tests.rs
@@ -9,7 +9,6 @@ use rand::RngCore;
     target_vendor = "apple",
 ))]
 use crate::assert_matches::assert_matches;
-use crate::char::MAX_LEN_UTF8;
 #[cfg(any(
     windows,
     target_os = "freebsd",
@@ -174,7 +173,7 @@ fn file_test_io_non_positional_read() {
 #[test]
 fn file_test_io_seek_and_tell_smoke_test() {
     let message = "ten-four";
-    let mut read_mem = [0; MAX_LEN_UTF8];
+    let mut read_mem = [0; char::MAX_LEN_UTF8];
     let set_cursor = 4 as u64;
     let tell_pos_pre_read;
     let tell_pos_post_read;
@@ -405,7 +404,7 @@ fn file_test_io_seek_shakedown() {
     let chunk_one: &str = "qwer";
     let chunk_two: &str = "asdf";
     let chunk_three: &str = "zxcv";
-    let mut read_mem = [0; MAX_LEN_UTF8];
+    let mut read_mem = [0; char::MAX_LEN_UTF8];
     let tmpdir = tmpdir();
     let filename = &tmpdir.join("file_rt_io_file_test_seek_shakedown.txt");
     {
@@ -782,7 +781,7 @@ fn file_test_directoryinfo_readdir() {
         check!(w.write(msg));
     }
     let files = check!(fs::read_dir(dir));
-    let mut mem = [0; MAX_LEN_UTF8];
+    let mut mem = [0; char::MAX_LEN_UTF8];
     for f in files {
         let f = f.unwrap().path();
         {
diff --git a/library/std/src/lib.rs b/library/std/src/lib.rs
index a8c50cec01e0b..ee7a4ec3c14b4 100644
--- a/library/std/src/lib.rs
+++ b/library/std/src/lib.rs
@@ -276,7 +276,6 @@
 #![feature(cfg_sanitizer_cfi)]
 #![feature(cfg_target_thread_local)]
 #![feature(cfi_encoding)]
-#![feature(char_max_len)]
 #![feature(const_trait_impl)]
 #![feature(core_float_math)]
 #![feature(decl_macro)]
diff --git a/library/std/src/sys/stdio/windows.rs b/library/std/src/sys/stdio/windows.rs
index 9b27f76b9dd1a..62ec115d7b0cb 100644
--- a/library/std/src/sys/stdio/windows.rs
+++ b/library/std/src/sys/stdio/windows.rs
@@ -1,6 +1,5 @@
 #![unstable(issue = "none", feature = "windows_stdio")]
 
-use core::char::MAX_LEN_UTF8;
 use core::str::utf8_char_width;
 
 use crate::mem::MaybeUninit;
@@ -427,7 +426,7 @@ fn utf16_to_utf8(utf16: &[u16], utf8: &mut [u8]) -> io::Result<usize> {
 
 impl IncompleteUtf8 {
     pub const fn new() -> IncompleteUtf8 {
-        IncompleteUtf8 { bytes: [0; MAX_LEN_UTF8], len: 0 }
+        IncompleteUtf8 { bytes: [0; char::MAX_LEN_UTF8], len: 0 }
     }
 }