Merge tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Don't destroy user event fields when removal fails

   User event fields are destroyed before the event is removed from
   visibility. But that can fail leaving the still visible event with no
   fields. Move the destroying of the fields to after the event is
   successfully removed from visibility.

 - Initialize function graph state is fork before calling
   copy_exec_state()

   For non-CLONE_VM forks, copy_exec_state() allocates a new
   task_exec_state. If that allocation fails, ftrace_graph_exit_task()
   will free the tasks ret_stack pointer. Since that pointer is still
   using the parent's ret_stack, it mistakenly frees the parent's
   pointer too.

   Call ftrace_graph_init() on the task first which will NULL out the
   new tasks's ret_stack and if the copy fails, it will not free
   anything.

 - Remove FGRAPH_MAX_INDEX

   The macro FGRAPH_MAX_INDEX was added but never used. Remove it.

 - Save ent_size in function graph printing of nested functions

   The function graph tracer needs to look at the next event to see if
   the next event is the return of the current function entry. If it is,
   it prints a single line:

	ktime_get();

   Otherwise it prints it like a nested function:

	tick_nohz_irq_exit() {
	    ktime_get();
	    kcpustat_irq_exit();
	}

   In order to look at the next event, it must save the current event so
   that it has the information to print from it. It saves the event in
   the iterator descriptor called "ent". What it doesn't save is the
   ent_size of the event which is now used to know if the function graph
   arguments are to be printed. The peek doesn't save the size so the
   size used happens to be that of the size of the last event that was
   seen.

   Save the entry event size in the iterator descriptor so that the
   correct size is used.

 - Fix several errors with freeing data in the histogram code

   The histogram code had a lot of leaked or or incorrect accounting
   when failures happen. Correct them.

 - Fix histogram regression of .percent and .graph modifiers

   Up until 6.3 histogram values could have "percent" or "graph"
   modifiers that changed how they were printed. But a change that added
   restricting histograms values from being strings, stack traces and
   other modifiers inadvertently prevented them from using the percent
   and graph modifiers, which were legal use cases for values.

   Put back the percent and graph modifiers.

 - Fix various typos in the comments

 - Set the trace_clock before initializing a histogram with clock
   argument

   The histogram API allows the user to specific which trace clock to
   use via a "clock=" string. The histogram is set up first before the
   clock is checked. If the passed in clock is not valid, it exits
   without fully fixing up the histogram leaving it on the list and a
   use-after-free can trigger.

   Update the clock argument first and if it fails then exit gracefully
   before the histogram trigger is placed on any lists.

 - Restore :mod: trailer after parsing in ftrace_set_clr_event

   The function ftrace_set_clr_event() modifies the parse string and
   needs to put it back to what was passed in. It searches for ":mod:"
   via a strsep() but fails to put back the first ':' in the string.

   Add back the ':' in the passed in string.

 - Take trace_array reference when opening a tracer options file

   The options files are dynamically created and some tracers add their
   own options. When a tracer adds their own list of options, the
   trace_array holding them has an array to hold the list of options for
   each tracer. This array increases in size via a krealloc(), and the
   new entry gets a newly allocated array to hold the options of the new
   tracer being added.

   The element in each entry of the tracer's option array holds a
   pointer back to the trace_array, a pointer to the tracer it is
   associated to, a pointer to the flags of the option.

   The issue is that these arrays are freed when the trace_array is
   freed when its instance it represents is removed from the instances
   directory. There's a race that an open of one of these options files
   can happen when the instance is being removed.

   Add a new helper function to be called by the open function of the
   options file to iterate all existing trace_arrays under a lock and
   find the one that has the given option element in one of it's tracer
   arrays. If found, then update the associated trace_array's reference
   counter to keep it from being freed. If not found, have the open call
   return -ENODEV.

 - Disable interrupts when acquiring the lock in rb_wake_up_waiters()

   The function rb_wake_up_waiters() assumes it will be called in
   interrupt context and does not disable irqs when taking
   cpu_buffer->reader_lock, which can be called in hard interrupt
   context. The issue is in PREEMPT_RT, this function is called in
   thread context leaving this lock open to a deadlock.

   Take the lock with interrupts disabled.

 - Use rcu_assign_pointer() for tmp_ops filter hash

   The tmp_ops used in update_ftrace_direct_mod() assigns its
   filter_hash field directly, but that field is annotated as __rcu and
   sparse complains. Assign it with rcu_assign_pointer()

 - Fix use-after-free in enable_trigger_private_data_free()

   The trace_event_call is accessed through the event_trigger_data's
   trace_event_file pointer to put the trace_event_call on freeing. The
   issue is that the trace_event_file data may have been freed already
   causing a use-after-free. Add a field to the event_trigger_data that
   points directly to the trace_event_call so that it can decrement its
   reference directly without needing to go through the
   trace_event_file.

 - Fix accounting of buffer data remote headers

   trace_buffer_desc_size() and trace_remote_alloc_buffer() undercount
   the number of pages is needed for the asked for size as it doesn't
   take into account the meta data on each page. Add a helper function
   to do the calculation properly and use that in these functions.

 - Catch nr_page_va overflow in ring_buffer_desc sizing

   The number of pages per remote ring buffer is capped by
   ring_buffer_desc::nr_page_va (32 bits). A buffer_size large enough to
   overflow that field would silently allocate a descriptor smaller than
   what was asked for.

 - Do not resize the subbuf order if any per_cpu buffer is disabled

   The mmapping of ring buffers disables resizing the subbuffers, but it
   is done per-cpu whereas the subbuf size change is done for all the
   per_cpu buffers under the buffer->mutex. It could change the size of
   some while the mapping is happening on others. Have the resize of the
   subbuf order check all the per_cpu buffers under the lock to see if
   any of them is disabled before starting and causing an inconsistency
   between buffers that are being mapped.

* tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (25 commits)
  ring-buffer: Check resize_disabled before publishing the new subbuf order
  tracing/remotes: Catch nr_page_va overflow in ring_buffer_desc sizing
  tracing/remotes: Account for ring buffer page header in size calculation
  tracing: Don't dereference trace_event_file in deferred trigger free
  ftrace: Use rcu_assign_pointer() for tmp_ops filter hash
  ring-buffer: Acquire the lock with irqsave in rb_wake_up_waiters()
  tracing: Take trace_array reference when opening a tracer options file
  tracing: Fix ring_buffer_read_page_size() kernel-doc
  tracing: Restore :mod: trailer after parsing in ftrace_set_clr_event()
  tracing: Fix memory corruption from a "STACKTRACE" histogram key
  tracing: Fix memory corruption from the histogram stacktrace modifier
  tracing: Undo the registration when enabling the histogram trigger fails
  tracing: Take the reference before publishing the named histogram trigger
  tracing: Set the trace clock before registering the histogram trigger
  tracing: Fix typo "preceeded" in comment
  tracing: Fix typo "availabe" in comment
  tracing: Let histogram values keep the percent and graph modifiers
  tracing: Keep the entry count when the histogram stats allocation fails
  tracing: Free histogram the field rejected for a bad modifier
  tracing: Free histogram the var ref when its initialization fails
  ...

1 files changed, 1862 insertions, 0 deletions

diff --git a/rust/syn/lit.rs b/rust/syn/lit.rs
new file mode 100644
index 000000000..1d2d8bdcc
--- /dev/null
+++ b/rust/syn/lit.rs
@@ -0,0 +1,1862 @@
+// SPDX-License-Identifier: Apache-2.0 OR MIT
+
+#[cfg(feature = "parsing")]
+use crate::lookahead;
+#[cfg(feature = "parsing")]
+use crate::parse::{Parse, Parser};
+use crate::{Error, Result};
+use proc_macro2::{Ident, Literal, Span};
+#[cfg(feature = "parsing")]
+use proc_macro2::{TokenStream, TokenTree};
+use std::ffi::{CStr, CString};
+use std::fmt::{self, Display};
+#[cfg(feature = "extra-traits")]
+use std::hash::{Hash, Hasher};
+use std::str::{self, FromStr};
+
+ast_enum_of_structs! {
+    /// A Rust literal such as a string or integer or boolean.
+    ///
+    /// # Syntax tree enum
+    ///
+    /// This type is a [syntax tree enum].
+    ///
+    /// [syntax tree enum]: crate::expr::Expr#syntax-tree-enums
+    #[non_exhaustive]
+    pub enum Lit {
+        /// A UTF-8 string literal: `"foo"`.
+        Str(LitStr),
+
+        /// A byte string literal: `b"foo"`.
+        ByteStr(LitByteStr),
+
+        /// A nul-terminated C-string literal: `c"foo"`.
+        CStr(LitCStr),
+
+        /// A byte literal: `b'f'`.
+        Byte(LitByte),
+
+        /// A character literal: `'a'`.
+        Char(LitChar),
+
+        /// An integer literal: `1` or `1u16`.
+        Int(LitInt),
+
+        /// A floating point literal: `1f64` or `1.0e10f64`.
+        ///
+        /// Must be finite. May not be infinite or NaN.
+        Float(LitFloat),
+
+        /// A boolean literal: `true` or `false`.
+        Bool(LitBool),
+
+        /// A raw token literal not interpreted by Syn.
+        Verbatim(Literal),
+    }
+}
+
+ast_struct! {
+    /// A UTF-8 string literal: `"foo"`.
+    pub struct LitStr {
+        repr: Box<LitRepr>,
+    }
+}
+
+ast_struct! {
+    /// A byte string literal: `b"foo"`.
+    pub struct LitByteStr {
+        repr: Box<LitRepr>,
+    }
+}
+
+ast_struct! {
+    /// A nul-terminated C-string literal: `c"foo"`.
+    pub struct LitCStr {
+        repr: Box<LitRepr>,
+    }
+}
+
+ast_struct! {
+    /// A byte literal: `b'f'`.
+    pub struct LitByte {
+        repr: Box<LitRepr>,
+    }
+}
+
+ast_struct! {
+    /// A character literal: `'a'`.
+    pub struct LitChar {
+        repr: Box<LitRepr>,
+    }
+}
+
+struct LitRepr {
+    token: Literal,
+    suffix: Box<str>,
+}
+
+ast_struct! {
+    /// An integer literal: `1` or `1u16`.
+    pub struct LitInt {
+        repr: Box<LitIntRepr>,
+    }
+}
+
+struct LitIntRepr {
+    token: Literal,
+    digits: Box<str>,
+    suffix: Box<str>,
+}
+
+ast_struct! {
+    /// A floating point literal: `1f64` or `1.0e10f64`.
+    ///
+    /// Must be finite. May not be infinite or NaN.
+    pub struct LitFloat {
+        repr: Box<LitFloatRepr>,
+    }
+}
+
+struct LitFloatRepr {
+    token: Literal,
+    digits: Box<str>,
+    suffix: Box<str>,
+}
+
+ast_struct! {
+    /// A boolean literal: `true` or `false`.
+    pub struct LitBool {
+        pub value: bool,
+        pub span: Span,
+    }
+}
+
+impl LitStr {
+    pub fn new(value: &str, span: Span) -> Self {
+        let mut token = Literal::string(value);
+        token.set_span(span);
+        LitStr {
+            repr: Box::new(LitRepr {
+                token,
+                suffix: Box::<str>::default(),
+            }),
+        }
+    }
+
+    pub fn value(&self) -> String {
+        let repr = self.repr.token.to_string();
+        let (value, _suffix) = value::parse_lit_str(&repr);
+        String::from(value)
+    }
+
+    /// Parse a syntax tree node from the content of this string literal.
+    ///
+    /// All spans in the syntax tree will point to the span of this `LitStr`.
+    ///
+    /// # Example
+    ///
+    /// ```
+    /// use syn::{Attribute, Error, Expr, Lit, Meta, Path, Result};
+    ///
+    /// // Parses the path from an attribute that looks like:
+    /// //
+    /// //     #[path = "a::b::c"]
+    /// //
+    /// // or returns `None` if the input is some other attribute.
+    /// fn get_path(attr: &Attribute) -> Result<Option<Path>> {
+    ///     if !attr.path().is_ident("path") {
+    ///         return Ok(None);
+    ///     }
+    ///
+    ///     if let Meta::NameValue(meta) = &attr.meta {
+    ///         if let Expr::Lit(expr) = &meta.value {
+    ///             if let Lit::Str(lit_str) = &expr.lit {
+    ///                 return lit_str.parse().map(Some);
+    ///             }
+    ///         }
+    ///     }
+    ///
+    ///     let message = "expected #[path = \"...\"]";
+    ///     Err(Error::new_spanned(attr, message))
+    /// }
+    /// ```
+    #[cfg(feature = "parsing")]
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    pub fn parse<T: Parse>(&self) -> Result<T> {
+        self.parse_with(T::parse)
+    }
+
+    /// Invoke parser on the content of this string literal.
+    ///
+    /// All spans in the syntax tree will point to the span of this `LitStr`.
+    ///
+    /// # Example
+    ///
+    /// ```
+    /// # use proc_macro2::Span;
+    /// # use syn::{LitStr, Result};
+    /// #
+    /// # fn main() -> Result<()> {
+    /// #     let lit_str = LitStr::new("a::b::c", Span::call_site());
+    /// #
+    /// #     const IGNORE: &str = stringify! {
+    /// let lit_str: LitStr = /* ... */;
+    /// #     };
+    ///
+    /// // Parse a string literal like "a::b::c" into a Path, not allowing
+    /// // generic arguments on any of the path segments.
+    /// let basic_path = lit_str.parse_with(syn::Path::parse_mod_style)?;
+    /// #
+    /// #     Ok(())
+    /// # }
+    /// ```
+    #[cfg(feature = "parsing")]
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    pub fn parse_with<F: Parser>(&self, parser: F) -> Result<F::Output> {
+        use proc_macro2::Group;
+
+        // Token stream with every span replaced by the given one.
+        fn respan_token_stream(stream: TokenStream, span: Span) -> TokenStream {
+            stream
+                .into_iter()
+                .map(|token| respan_token_tree(token, span))
+                .collect()
+        }
+
+        // Token tree with every span replaced by the given one.
+        fn respan_token_tree(mut token: TokenTree, span: Span) -> TokenTree {
+            match &mut token {
+                TokenTree::Group(g) => {
+                    let stream = respan_token_stream(g.stream(), span);
+                    *g = Group::new(g.delimiter(), stream);
+                    g.set_span(span);
+                }
+                other => other.set_span(span),
+            }
+            token
+        }
+
+        // Parse string literal into a token stream with every span equal to the
+        // original literal's span.
+        let span = self.span();
+        let mut tokens = TokenStream::from_str(&self.value())?;
+        tokens = respan_token_stream(tokens, span);
+
+        let result = crate::parse::parse_scoped(parser, span, tokens)?;
+
+        let suffix = self.suffix();
+        if !suffix.is_empty() {
+            return Err(Error::new(
+                self.span(),
+                format!("unexpected suffix `{}` on string literal", suffix),
+            ));
+        }
+
+        Ok(result)
+    }
+
+    pub fn span(&self) -> Span {
+        self.repr.token.span()
+    }
+
+    pub fn set_span(&mut self, span: Span) {
+        self.repr.token.set_span(span);
+    }
+
+    pub fn suffix(&self) -> &str {
+        &self.repr.suffix
+    }
+
+    pub fn token(&self) -> Literal {
+        self.repr.token.clone()
+    }
+}
+
+impl LitByteStr {
+    pub fn new(value: &[u8], span: Span) -> Self {
+        let mut token = Literal::byte_string(value);
+        token.set_span(span);
+        LitByteStr {
+            repr: Box::new(LitRepr {
+                token,
+                suffix: Box::<str>::default(),
+            }),
+        }
+    }
+
+    pub fn value(&self) -> Vec<u8> {
+        let repr = self.repr.token.to_string();
+        let (value, _suffix) = value::parse_lit_byte_str(&repr);
+        value
+    }
+
+    pub fn span(&self) -> Span {
+        self.repr.token.span()
+    }
+
+    pub fn set_span(&mut self, span: Span) {
+        self.repr.token.set_span(span);
+    }
+
+    pub fn suffix(&self) -> &str {
+        &self.repr.suffix
+    }
+
+    pub fn token(&self) -> Literal {
+        self.repr.token.clone()
+    }
+}
+
+impl LitCStr {
+    pub fn new(value: &CStr, span: Span) -> Self {
+        let mut token = Literal::c_string(value);
+        token.set_span(span);
+        LitCStr {
+            repr: Box::new(LitRepr {
+                token,
+                suffix: Box::<str>::default(),
+            }),
+        }
+    }
+
+    pub fn value(&self) -> CString {
+        let repr = self.repr.token.to_string();
+        let (value, _suffix) = value::parse_lit_c_str(&repr);
+        value
+    }
+
+    pub fn span(&self) -> Span {
+        self.repr.token.span()
+    }
+
+    pub fn set_span(&mut self, span: Span) {
+        self.repr.token.set_span(span);
+    }
+
+    pub fn suffix(&self) -> &str {
+        &self.repr.suffix
+    }
+
+    pub fn token(&self) -> Literal {
+        self.repr.token.clone()
+    }
+}
+
+impl LitByte {
+    pub fn new(value: u8, span: Span) -> Self {
+        let mut token = Literal::u8_suffixed(value);
+        token.set_span(span);
+        LitByte {
+            repr: Box::new(LitRepr {
+                token,
+                suffix: Box::<str>::default(),
+            }),
+        }
+    }
+
+    pub fn value(&self) -> u8 {
+        let repr = self.repr.token.to_string();
+        let (value, _suffix) = value::parse_lit_byte(&repr);
+        value
+    }
+
+    pub fn span(&self) -> Span {
+        self.repr.token.span()
+    }
+
+    pub fn set_span(&mut self, span: Span) {
+        self.repr.token.set_span(span);
+    }
+
+    pub fn suffix(&self) -> &str {
+        &self.repr.suffix
+    }
+
+    pub fn token(&self) -> Literal {
+        self.repr.token.clone()
+    }
+}
+
+impl LitChar {
+    pub fn new(value: char, span: Span) -> Self {
+        let mut token = Literal::character(value);
+        token.set_span(span);
+        LitChar {
+            repr: Box::new(LitRepr {
+                token,
+                suffix: Box::<str>::default(),
+            }),
+        }
+    }
+
+    pub fn value(&self) -> char {
+        let repr = self.repr.token.to_string();
+        let (value, _suffix) = value::parse_lit_char(&repr);
+        value
+    }
+
+    pub fn span(&self) -> Span {
+        self.repr.token.span()
+    }
+
+    pub fn set_span(&mut self, span: Span) {
+        self.repr.token.set_span(span);
+    }
+
+    pub fn suffix(&self) -> &str {
+        &self.repr.suffix
+    }
+
+    pub fn token(&self) -> Literal {
+        self.repr.token.clone()
+    }
+}
+
+impl LitInt {
+    pub fn new(repr: &str, span: Span) -> Self {
+        let (digits, suffix) = match value::parse_lit_int(repr) {
+            Some(parse) => parse,
+            None => panic!("not an integer literal: `{}`", repr),
+        };
+
+        let mut token: Literal = repr.parse().unwrap();
+        token.set_span(span);
+        LitInt {
+            repr: Box::new(LitIntRepr {
+                token,
+                digits,
+                suffix,
+            }),
+        }
+    }
+
+    pub fn base10_digits(&self) -> &str {
+        &self.repr.digits
+    }
+
+    /// Parses the literal into a selected number type.
+    ///
+    /// This is equivalent to `lit.base10_digits().parse()` except that the
+    /// resulting errors will be correctly spanned to point to the literal token
+    /// in the macro input.
+    ///
+    /// ```
+    /// use syn::LitInt;
+    /// use syn::parse::{Parse, ParseStream, Result};
+    ///
+    /// struct Port {
+    ///     value: u16,
+    /// }
+    ///
+    /// impl Parse for Port {
+    ///     fn parse(input: ParseStream) -> Result<Self> {
+    ///         let lit: LitInt = input.parse()?;
+    ///         let value = lit.base10_parse::<u16>()?;
+    ///         Ok(Port { value })
+    ///     }
+    /// }
+    /// ```
+    pub fn base10_parse<N>(&self) -> Result<N>
+    where
+        N: FromStr,
+        N::Err: Display,
+    {
+        self.base10_digits()
+            .parse()
+            .map_err(|err| Error::new(self.span(), err))
+    }
+
+    pub fn suffix(&self) -> &str {
+        &self.repr.suffix
+    }
+
+    pub fn span(&self) -> Span {
+        self.repr.token.span()
+    }
+
+    pub fn set_span(&mut self, span: Span) {
+        self.repr.token.set_span(span);
+    }
+
+    pub fn token(&self) -> Literal {
+        self.repr.token.clone()
+    }
+}
+
+impl From<Literal> for LitInt {
+    fn from(token: Literal) -> Self {
+        let repr = token.to_string();
+        if let Some((digits, suffix)) = value::parse_lit_int(&repr) {
+            LitInt {
+                repr: Box::new(LitIntRepr {
+                    token,
+                    digits,
+                    suffix,
+                }),
+            }
+        } else {
+            panic!("not an integer literal: `{}`", repr);
+        }
+    }
+}
+
+impl Display for LitInt {
+    fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+        self.repr.token.fmt(formatter)
+    }
+}
+
+impl LitFloat {
+    pub fn new(repr: &str, span: Span) -> Self {
+        let (digits, suffix) = match value::parse_lit_float(repr) {
+            Some(parse) => parse,
+            None => panic!("not a float literal: `{}`", repr),
+        };
+
+        let mut token: Literal = repr.parse().unwrap();
+        token.set_span(span);
+        LitFloat {
+            repr: Box::new(LitFloatRepr {
+                token,
+                digits,
+                suffix,
+            }),
+        }
+    }
+
+    pub fn base10_digits(&self) -> &str {
+        &self.repr.digits
+    }
+
+    pub fn base10_parse<N>(&self) -> Result<N>
+    where
+        N: FromStr,
+        N::Err: Display,
+    {
+        self.base10_digits()
+            .parse()
+            .map_err(|err| Error::new(self.span(), err))
+    }
+
+    pub fn suffix(&self) -> &str {
+        &self.repr.suffix
+    }
+
+    pub fn span(&self) -> Span {
+        self.repr.token.span()
+    }
+
+    pub fn set_span(&mut self, span: Span) {
+        self.repr.token.set_span(span);
+    }
+
+    pub fn token(&self) -> Literal {
+        self.repr.token.clone()
+    }
+}
+
+impl From<Literal> for LitFloat {
+    fn from(token: Literal) -> Self {
+        let repr = token.to_string();
+        if let Some((digits, suffix)) = value::parse_lit_float(&repr) {
+            LitFloat {
+                repr: Box::new(LitFloatRepr {
+                    token,
+                    digits,
+                    suffix,
+                }),
+            }
+        } else {
+            panic!("not a float literal: `{}`", repr);
+        }
+    }
+}
+
+impl Display for LitFloat {
+    fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+        self.repr.token.fmt(formatter)
+    }
+}
+
+impl LitBool {
+    pub fn new(value: bool, span: Span) -> Self {
+        LitBool { value, span }
+    }
+
+    pub fn value(&self) -> bool {
+        self.value
+    }
+
+    pub fn span(&self) -> Span {
+        self.span
+    }
+
+    pub fn set_span(&mut self, span: Span) {
+        self.span = span;
+    }
+
+    pub fn token(&self) -> Ident {
+        let s = if self.value { "true" } else { "false" };
+        Ident::new(s, self.span)
+    }
+}
+
+#[cfg(feature = "extra-traits")]
+mod debug_impls {
+    use crate::lit::{LitBool, LitByte, LitByteStr, LitCStr, LitChar, LitFloat, LitInt, LitStr};
+    use std::fmt::{self, Debug};
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+    impl Debug for LitStr {
+        fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+            self.debug(formatter, "LitStr")
+        }
+    }
+
+    impl LitStr {
+        pub(crate) fn debug(&self, formatter: &mut fmt::Formatter, name: &str) -> fmt::Result {
+            formatter
+                .debug_struct(name)
+                .field("token", &format_args!("{}", self.repr.token))
+                .finish()
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+    impl Debug for LitByteStr {
+        fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+            self.debug(formatter, "LitByteStr")
+        }
+    }
+
+    impl LitByteStr {
+        pub(crate) fn debug(&self, formatter: &mut fmt::Formatter, name: &str) -> fmt::Result {
+            formatter
+                .debug_struct(name)
+                .field("token", &format_args!("{}", self.repr.token))
+                .finish()
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+    impl Debug for LitCStr {
+        fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+            self.debug(formatter, "LitCStr")
+        }
+    }
+
+    impl LitCStr {
+        pub(crate) fn debug(&self, formatter: &mut fmt::Formatter, name: &str) -> fmt::Result {
+            formatter
+                .debug_struct(name)
+                .field("token", &format_args!("{}", self.repr.token))
+                .finish()
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+    impl Debug for LitByte {
+        fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+            self.debug(formatter, "LitByte")
+        }
+    }
+
+    impl LitByte {
+        pub(crate) fn debug(&self, formatter: &mut fmt::Formatter, name: &str) -> fmt::Result {
+            formatter
+                .debug_struct(name)
+                .field("token", &format_args!("{}", self.repr.token))
+                .finish()
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+    impl Debug for LitChar {
+        fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+            self.debug(formatter, "LitChar")
+        }
+    }
+
+    impl LitChar {
+        pub(crate) fn debug(&self, formatter: &mut fmt::Formatter, name: &str) -> fmt::Result {
+            formatter
+                .debug_struct(name)
+                .field("token", &format_args!("{}", self.repr.token))
+                .finish()
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+    impl Debug for LitInt {
+        fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+            self.debug(formatter, "LitInt")
+        }
+    }
+
+    impl LitInt {
+        pub(crate) fn debug(&self, formatter: &mut fmt::Formatter, name: &str) -> fmt::Result {
+            formatter
+                .debug_struct(name)
+                .field("token", &format_args!("{}", self.repr.token))
+                .finish()
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+    impl Debug for LitFloat {
+        fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+            self.debug(formatter, "LitFloat")
+        }
+    }
+
+    impl LitFloat {
+        pub(crate) fn debug(&self, formatter: &mut fmt::Formatter, name: &str) -> fmt::Result {
+            formatter
+                .debug_struct(name)
+                .field("token", &format_args!("{}", self.repr.token))
+                .finish()
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+    impl Debug for LitBool {
+        fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+            self.debug(formatter, "LitBool")
+        }
+    }
+
+    impl LitBool {
+        pub(crate) fn debug(&self, formatter: &mut fmt::Formatter, name: &str) -> fmt::Result {
+            formatter
+                .debug_struct(name)
+                .field("value", &self.value)
+                .finish()
+        }
+    }
+}
+
+#[cfg(feature = "clone-impls")]
+#[cfg_attr(docsrs, doc(cfg(feature = "clone-impls")))]
+impl Clone for LitRepr {
+    fn clone(&self) -> Self {
+        LitRepr {
+            token: self.token.clone(),
+            suffix: self.suffix.clone(),
+        }
+    }
+}
+
+#[cfg(feature = "clone-impls")]
+#[cfg_attr(docsrs, doc(cfg(feature = "clone-impls")))]
+impl Clone for LitIntRepr {
+    fn clone(&self) -> Self {
+        LitIntRepr {
+            token: self.token.clone(),
+            digits: self.digits.clone(),
+            suffix: self.suffix.clone(),
+        }
+    }
+}
+
+#[cfg(feature = "clone-impls")]
+#[cfg_attr(docsrs, doc(cfg(feature = "clone-impls")))]
+impl Clone for LitFloatRepr {
+    fn clone(&self) -> Self {
+        LitFloatRepr {
+            token: self.token.clone(),
+            digits: self.digits.clone(),
+            suffix: self.suffix.clone(),
+        }
+    }
+}
+
+macro_rules! lit_extra_traits {
+    ($ty:ident) => {
+        #[cfg(feature = "clone-impls")]
+        #[cfg_attr(docsrs, doc(cfg(feature = "clone-impls")))]
+        impl Clone for $ty {
+            fn clone(&self) -> Self {
+                $ty {
+                    repr: self.repr.clone(),
+                }
+            }
+        }
+
+        #[cfg(feature = "extra-traits")]
+        #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+        impl PartialEq for $ty {
+            fn eq(&self, other: &Self) -> bool {
+                self.repr.token.to_string() == other.repr.token.to_string()
+            }
+        }
+
+        #[cfg(feature = "extra-traits")]
+        #[cfg_attr(docsrs, doc(cfg(feature = "extra-traits")))]
+        impl Hash for $ty {
+            fn hash<H>(&self, state: &mut H)
+            where
+                H: Hasher,
+            {
+                self.repr.token.to_string().hash(state);
+            }
+        }
+
+        #[cfg(feature = "parsing")]
+        pub_if_not_doc! {
+            #[doc(hidden)]
+            #[allow(non_snake_case)]
+            pub fn $ty(marker: lookahead::TokenMarker) -> $ty {
+                match marker {}
+            }
+        }
+    };
+}
+
+lit_extra_traits!(LitStr);
+lit_extra_traits!(LitByteStr);
+lit_extra_traits!(LitCStr);
+lit_extra_traits!(LitByte);
+lit_extra_traits!(LitChar);
+lit_extra_traits!(LitInt);
+lit_extra_traits!(LitFloat);
+
+#[cfg(feature = "parsing")]
+pub_if_not_doc! {
+    #[doc(hidden)]
+    #[allow(non_snake_case)]
+    pub fn LitBool(marker: lookahead::TokenMarker) -> LitBool {
+        match marker {}
+    }
+}
+
+/// The style of a string literal, either plain quoted or a raw string like
+/// `r##"data"##`.
+#[doc(hidden)] // https://github.com/dtolnay/syn/issues/1566
+pub enum StrStyle {
+    /// An ordinary string like `"data"`.
+    Cooked,
+    /// A raw string like `r##"data"##`.
+    ///
+    /// The unsigned integer is the number of `#` symbols used.
+    Raw(usize),
+}
+
+#[cfg(feature = "parsing")]
+pub_if_not_doc! {
+    #[doc(hidden)]
+    #[allow(non_snake_case)]
+    pub fn Lit(marker: lookahead::TokenMarker) -> Lit {
+        match marker {}
+    }
+}
+
+#[cfg(feature = "parsing")]
+pub(crate) mod parsing {
+    use crate::buffer::Cursor;
+    use crate::error::Result;
+    use crate::lit::{
+        value, Lit, LitBool, LitByte, LitByteStr, LitCStr, LitChar, LitFloat, LitFloatRepr, LitInt,
+        LitIntRepr, LitStr,
+    };
+    use crate::parse::{Parse, ParseStream, Unexpected};
+    use crate::token::{self, Token};
+    use proc_macro2::{Literal, Punct, Span};
+    use std::cell::Cell;
+    use std::rc::Rc;
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    impl Parse for Lit {
+        fn parse(input: ParseStream) -> Result<Self> {
+            input.step(|cursor| {
+                if let Some((lit, rest)) = cursor.literal() {
+                    return Ok((Lit::new(lit), rest));
+                }
+
+                if let Some((ident, rest)) = cursor.ident() {
+                    let value = ident == "true";
+                    if value || ident == "false" {
+                        let lit_bool = LitBool {
+                            value,
+                            span: ident.span(),
+                        };
+                        return Ok((Lit::Bool(lit_bool), rest));
+                    }
+                }
+
+                if let Some((punct, rest)) = cursor.punct() {
+                    if punct.as_char() == '-' {
+                        if let Some((lit, rest)) = parse_negative_lit(punct, rest) {
+                            return Ok((lit, rest));
+                        }
+                    }
+                }
+
+                Err(cursor.error("expected literal"))
+            })
+        }
+    }
+
+    fn parse_negative_lit(neg: Punct, cursor: Cursor) -> Option<(Lit, Cursor)> {
+        let (lit, rest) = cursor.literal()?;
+
+        let mut span = neg.span();
+        span = span.join(lit.span()).unwrap_or(span);
+
+        let mut repr = lit.to_string();
+        repr.insert(0, '-');
+
+        if let Some((digits, suffix)) = value::parse_lit_int(&repr) {
+            let mut token: Literal = repr.parse().unwrap();
+            token.set_span(span);
+            return Some((
+                Lit::Int(LitInt {
+                    repr: Box::new(LitIntRepr {
+                        token,
+                        digits,
+                        suffix,
+                    }),
+                }),
+                rest,
+            ));
+        }
+
+        let (digits, suffix) = value::parse_lit_float(&repr)?;
+        let mut token: Literal = repr.parse().unwrap();
+        token.set_span(span);
+        Some((
+            Lit::Float(LitFloat {
+                repr: Box::new(LitFloatRepr {
+                    token,
+                    digits,
+                    suffix,
+                }),
+            }),
+            rest,
+        ))
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    impl Parse for LitStr {
+        fn parse(input: ParseStream) -> Result<Self> {
+            let head = input.fork();
+            match input.parse() {
+                Ok(Lit::Str(lit)) => Ok(lit),
+                _ => Err(head.error("expected string literal")),
+            }
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    impl Parse for LitByteStr {
+        fn parse(input: ParseStream) -> Result<Self> {
+            let head = input.fork();
+            match input.parse() {
+                Ok(Lit::ByteStr(lit)) => Ok(lit),
+                _ => Err(head.error("expected byte string literal")),
+            }
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    impl Parse for LitCStr {
+        fn parse(input: ParseStream) -> Result<Self> {
+            let head = input.fork();
+            match input.parse() {
+                Ok(Lit::CStr(lit)) => Ok(lit),
+                _ => Err(head.error("expected C string literal")),
+            }
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    impl Parse for LitByte {
+        fn parse(input: ParseStream) -> Result<Self> {
+            let head = input.fork();
+            match input.parse() {
+                Ok(Lit::Byte(lit)) => Ok(lit),
+                _ => Err(head.error("expected byte literal")),
+            }
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    impl Parse for LitChar {
+        fn parse(input: ParseStream) -> Result<Self> {
+            let head = input.fork();
+            match input.parse() {
+                Ok(Lit::Char(lit)) => Ok(lit),
+                _ => Err(head.error("expected character literal")),
+            }
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    impl Parse for LitInt {
+        fn parse(input: ParseStream) -> Result<Self> {
+            let head = input.fork();
+            match input.parse() {
+                Ok(Lit::Int(lit)) => Ok(lit),
+                _ => Err(head.error("expected integer literal")),
+            }
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    impl Parse for LitFloat {
+        fn parse(input: ParseStream) -> Result<Self> {
+            let head = input.fork();
+            match input.parse() {
+                Ok(Lit::Float(lit)) => Ok(lit),
+                _ => Err(head.error("expected floating point literal")),
+            }
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "parsing")))]
+    impl Parse for LitBool {
+        fn parse(input: ParseStream) -> Result<Self> {
+            let head = input.fork();
+            match input.parse() {
+                Ok(Lit::Bool(lit)) => Ok(lit),
+                _ => Err(head.error("expected boolean literal")),
+            }
+        }
+    }
+
+    fn peek_impl(cursor: Cursor, peek: fn(ParseStream) -> bool) -> bool {
+        let scope = Span::call_site();
+        let unexpected = Rc::new(Cell::new(Unexpected::None));
+        let buffer = crate::parse::new_parse_buffer(scope, cursor, unexpected);
+        peek(&buffer)
+    }
+
+    macro_rules! impl_token {
+        ($display:literal $name:ty) => {
+            impl Token for $name {
+                fn peek(cursor: Cursor) -> bool {
+                    fn peek(input: ParseStream) -> bool {
+                        <$name as Parse>::parse(input).is_ok()
+                    }
+                    peek_impl(cursor, peek)
+                }
+
+                fn display() -> &'static str {
+                    $display
+                }
+            }
+
+            impl token::private::Sealed for $name {}
+        };
+    }
+
+    impl_token!("literal" Lit);
+    impl_token!("string literal" LitStr);
+    impl_token!("byte string literal" LitByteStr);
+    impl_token!("C-string literal" LitCStr);
+    impl_token!("byte literal" LitByte);
+    impl_token!("character literal" LitChar);
+    impl_token!("integer literal" LitInt);
+    impl_token!("floating point literal" LitFloat);
+    impl_token!("boolean literal" LitBool);
+}
+
+#[cfg(feature = "printing")]
+mod printing {
+    use crate::lit::{LitBool, LitByte, LitByteStr, LitCStr, LitChar, LitFloat, LitInt, LitStr};
+    use proc_macro2::TokenStream;
+    use quote::{ToTokens, TokenStreamExt};
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "printing")))]
+    impl ToTokens for LitStr {
+        fn to_tokens(&self, tokens: &mut TokenStream) {
+            self.repr.token.to_tokens(tokens);
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "printing")))]
+    impl ToTokens for LitByteStr {
+        fn to_tokens(&self, tokens: &mut TokenStream) {
+            self.repr.token.to_tokens(tokens);
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "printing")))]
+    impl ToTokens for LitCStr {
+        fn to_tokens(&self, tokens: &mut TokenStream) {
+            self.repr.token.to_tokens(tokens);
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "printing")))]
+    impl ToTokens for LitByte {
+        fn to_tokens(&self, tokens: &mut TokenStream) {
+            self.repr.token.to_tokens(tokens);
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "printing")))]
+    impl ToTokens for LitChar {
+        fn to_tokens(&self, tokens: &mut TokenStream) {
+            self.repr.token.to_tokens(tokens);
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "printing")))]
+    impl ToTokens for LitInt {
+        fn to_tokens(&self, tokens: &mut TokenStream) {
+            self.repr.token.to_tokens(tokens);
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "printing")))]
+    impl ToTokens for LitFloat {
+        fn to_tokens(&self, tokens: &mut TokenStream) {
+            self.repr.token.to_tokens(tokens);
+        }
+    }
+
+    #[cfg_attr(docsrs, doc(cfg(feature = "printing")))]
+    impl ToTokens for LitBool {
+        fn to_tokens(&self, tokens: &mut TokenStream) {
+            tokens.append(self.token());
+        }
+    }
+}
+
+mod value {
+    use crate::bigint::BigInt;
+    use crate::lit::{
+        Lit, LitBool, LitByte, LitByteStr, LitCStr, LitChar, LitFloat, LitFloatRepr, LitInt,
+        LitIntRepr, LitRepr, LitStr,
+    };
+    use proc_macro2::{Literal, Span};
+    use std::ascii;
+    use std::char;
+    use std::ffi::CString;
+    use std::ops::{Index, RangeFrom};
+
+    impl Lit {
+        /// Interpret a Syn literal from a proc-macro2 literal.
+        pub fn new(token: Literal) -> Self {
+            let repr = token.to_string();
+
+            match byte(&repr, 0) {
+                // "...", r"...", r#"..."#
+                b'"' | b'r' => {
+                    let (_, suffix) = parse_lit_str(&repr);
+                    return Lit::Str(LitStr {
+                        repr: Box::new(LitRepr { token, suffix }),
+                    });
+                }
+                b'b' => match byte(&repr, 1) {
+                    // b"...", br"...", br#"...#"
+                    b'"' | b'r' => {
+                        let (_, suffix) = parse_lit_byte_str(&repr);
+                        return Lit::ByteStr(LitByteStr {
+                            repr: Box::new(LitRepr { token, suffix }),
+                        });
+                    }
+                    // b'...'
+                    b'\'' => {
+                        let (_, suffix) = parse_lit_byte(&repr);
+                        return Lit::Byte(LitByte {
+                            repr: Box::new(LitRepr { token, suffix }),
+                        });
+                    }
+                    _ => {}
+                },
+                // c"...", cr"...", cr#"..."#
+                b'c' => {
+                    let (_, suffix) = parse_lit_c_str(&repr);
+                    return Lit::CStr(LitCStr {
+                        repr: Box::new(LitRepr { token, suffix }),
+                    });
+                }
+                // '...'
+                b'\'' => {
+                    let (_, suffix) = parse_lit_char(&repr);
+                    return Lit::Char(LitChar {
+                        repr: Box::new(LitRepr { token, suffix }),
+                    });
+                }
+                b'0'..=b'9' | b'-' => {
+                    // 0, 123, 0xFF, 0o77, 0b11
+                    if let Some((digits, suffix)) = parse_lit_int(&repr) {
+                        return Lit::Int(LitInt {
+                            repr: Box::new(LitIntRepr {
+                                token,
+                                digits,
+                                suffix,
+                            }),
+                        });
+                    }
+                    // 1.0, 1e-1, 1e+1
+                    if let Some((digits, suffix)) = parse_lit_float(&repr) {
+                        return Lit::Float(LitFloat {
+                            repr: Box::new(LitFloatRepr {
+                                token,
+                                digits,
+                                suffix,
+                            }),
+                        });
+                    }
+                }
+                // true, false
+                b't' | b'f' => {
+                    if repr == "true" || repr == "false" {
+                        return Lit::Bool(LitBool {
+                            value: repr == "true",
+                            span: token.span(),
+                        });
+                    }
+                }
+                b'(' if repr == "(/*ERROR*/)" => return Lit::Verbatim(token),
+                _ => {}
+            }
+
+            panic!("unrecognized literal: `{}`", repr);
+        }
+
+        pub fn suffix(&self) -> &str {
+            match self {
+                Lit::Str(lit) => lit.suffix(),
+                Lit::ByteStr(lit) => lit.suffix(),
+                Lit::CStr(lit) => lit.suffix(),
+                Lit::Byte(lit) => lit.suffix(),
+                Lit::Char(lit) => lit.suffix(),
+                Lit::Int(lit) => lit.suffix(),
+                Lit::Float(lit) => lit.suffix(),
+                Lit::Bool(_) | Lit::Verbatim(_) => "",
+            }
+        }
+
+        pub fn span(&self) -> Span {
+            match self {
+                Lit::Str(lit) => lit.span(),
+                Lit::ByteStr(lit) => lit.span(),
+                Lit::CStr(lit) => lit.span(),
+                Lit::Byte(lit) => lit.span(),
+                Lit::Char(lit) => lit.span(),
+                Lit::Int(lit) => lit.span(),
+                Lit::Float(lit) => lit.span(),
+                Lit::Bool(lit) => lit.span,
+                Lit::Verbatim(lit) => lit.span(),
+            }
+        }
+
+        pub fn set_span(&mut self, span: Span) {
+            match self {
+                Lit::Str(lit) => lit.set_span(span),
+                Lit::ByteStr(lit) => lit.set_span(span),
+                Lit::CStr(lit) => lit.set_span(span),
+                Lit::Byte(lit) => lit.set_span(span),
+                Lit::Char(lit) => lit.set_span(span),
+                Lit::Int(lit) => lit.set_span(span),
+                Lit::Float(lit) => lit.set_span(span),
+                Lit::Bool(lit) => lit.span = span,
+                Lit::Verbatim(lit) => lit.set_span(span),
+            }
+        }
+    }
+
+    /// Get the byte at offset idx, or a default of `b'\0'` if we're looking
+    /// past the end of the input buffer.
+    pub(crate) fn byte<S: AsRef<[u8]> + ?Sized>(s: &S, idx: usize) -> u8 {
+        let s = s.as_ref();
+        if idx < s.len() {
+            s[idx]
+        } else {
+            0
+        }
+    }
+
+    fn next_chr(s: &str) -> char {
+        s.chars().next().unwrap_or('\0')
+    }
+
+    // Returns (content, suffix).
+    pub(crate) fn parse_lit_str(s: &str) -> (Box<str>, Box<str>) {
+        match byte(s, 0) {
+            b'"' => parse_lit_str_cooked(s),
+            b'r' => parse_lit_str_raw(s),
+            _ => unreachable!(),
+        }
+    }
+
+    fn parse_lit_str_cooked(mut s: &str) -> (Box<str>, Box<str>) {
+        assert_eq!(byte(s, 0), b'"');
+        s = &s[1..];
+
+        let mut content = String::new();
+        'outer: loop {
+            let ch = match byte(s, 0) {
+                b'"' => break,
+                b'\\' => {
+                    let b = byte(s, 1);
+                    s = &s[2..];
+                    match b {
+                        b'x' => {
+                            let (byte, rest) = backslash_x(s);
+                            s = rest;
+                            assert!(byte <= 0x7F, "invalid \\x byte in string literal");
+                            char::from_u32(u32::from(byte)).unwrap()
+                        }
+                        b'u' => {
+                            let (ch, rest) = backslash_u(s);
+                            s = rest;
+                            ch
+                        }
+                        b'n' => '\n',
+                        b'r' => '\r',
+                        b't' => '\t',
+                        b'\\' => '\\',
+                        b'0' => '\0',
+                        b'\'' => '\'',
+                        b'"' => '"',
+                        b'\r' | b'\n' => loop {
+                            let b = byte(s, 0);
+                            match b {
+                                b' ' | b'\t' | b'\n' | b'\r' => s = &s[1..],
+                                _ => continue 'outer,
+                            }
+                        },
+                        b => panic!(
+                            "unexpected byte '{}' after \\ character in string literal",
+                            ascii::escape_default(b),
+                        ),
+                    }
+                }
+                b'\r' => {
+                    assert_eq!(byte(s, 1), b'\n', "bare CR not allowed in string");
+                    s = &s[2..];
+                    '\n'
+                }
+                _ => {
+                    let ch = next_chr(s);
+                    s = &s[ch.len_utf8()..];
+                    ch
+                }
+            };
+            content.push(ch);
+        }
+
+        assert!(s.starts_with('"'));
+        let content = content.into_boxed_str();
+        let suffix = s[1..].to_owned().into_boxed_str();
+        (content, suffix)
+    }
+
+    fn parse_lit_str_raw(mut s: &str) -> (Box<str>, Box<str>) {
+        assert_eq!(byte(s, 0), b'r');
+        s = &s[1..];
+
+        let mut pounds = 0;
+        while byte(s, pounds) == b'#' {
+            pounds += 1;
+        }
+        assert_eq!(byte(s, pounds), b'"');
+        let close = s.rfind('"').unwrap();
+        for end in s[close + 1..close + 1 + pounds].bytes() {
+            assert_eq!(end, b'#');
+        }
+
+        let content = s[pounds + 1..close].to_owned().into_boxed_str();
+        let suffix = s[close + 1 + pounds..].to_owned().into_boxed_str();
+        (content, suffix)
+    }
+
+    // Returns (content, suffix).
+    pub(crate) fn parse_lit_byte_str(s: &str) -> (Vec<u8>, Box<str>) {
+        assert_eq!(byte(s, 0), b'b');
+        match byte(s, 1) {
+            b'"' => parse_lit_byte_str_cooked(s),
+            b'r' => parse_lit_byte_str_raw(s),
+            _ => unreachable!(),
+        }
+    }
+
+    fn parse_lit_byte_str_cooked(mut s: &str) -> (Vec<u8>, Box<str>) {
+        assert_eq!(byte(s, 0), b'b');
+        assert_eq!(byte(s, 1), b'"');
+        s = &s[2..];
+
+        // We're going to want to have slices which don't respect codepoint boundaries.
+        let mut v = s.as_bytes();
+
+        let mut out = Vec::new();
+        'outer: loop {
+            let byte = match byte(v, 0) {
+                b'"' => break,
+                b'\\' => {
+                    let b = byte(v, 1);
+                    v = &v[2..];
+                    match b {
+                        b'x' => {
+                            let (b, rest) = backslash_x(v);
+                            v = rest;
+                            b
+                        }
+                        b'n' => b'\n',
+                        b'r' => b'\r',
+                        b't' => b'\t',
+                        b'\\' => b'\\',
+                        b'0' => b'\0',
+                        b'\'' => b'\'',
+                        b'"' => b'"',
+                        b'\r' | b'\n' => loop {
+                            let byte = byte(v, 0);
+                            if matches!(byte, b' ' | b'\t' | b'\n' | b'\r') {
+                                v = &v[1..];
+                            } else {
+                                continue 'outer;
+                            }
+                        },
+                        b => panic!(
+                            "unexpected byte '{}' after \\ character in byte-string literal",
+                            ascii::escape_default(b),
+                        ),
+                    }
+                }
+                b'\r' => {
+                    assert_eq!(byte(v, 1), b'\n', "bare CR not allowed in string");
+                    v = &v[2..];
+                    b'\n'
+                }
+                b => {
+                    v = &v[1..];
+                    b
+                }
+            };
+            out.push(byte);
+        }
+
+        assert_eq!(byte(v, 0), b'"');
+        let suffix = s[s.len() - v.len() + 1..].to_owned().into_boxed_str();
+        (out, suffix)
+    }
+
+    fn parse_lit_byte_str_raw(s: &str) -> (Vec<u8>, Box<str>) {
+        assert_eq!(byte(s, 0), b'b');
+        let (value, suffix) = parse_lit_str_raw(&s[1..]);
+        (String::from(value).into_bytes(), suffix)
+    }
+
+    // Returns (content, suffix).
+    pub(crate) fn parse_lit_c_str(s: &str) -> (CString, Box<str>) {
+        assert_eq!(byte(s, 0), b'c');
+        match byte(s, 1) {
+            b'"' => parse_lit_c_str_cooked(s),
+            b'r' => parse_lit_c_str_raw(s),
+            _ => unreachable!(),
+        }
+    }
+
+    fn parse_lit_c_str_cooked(mut s: &str) -> (CString, Box<str>) {
+        assert_eq!(byte(s, 0), b'c');
+        assert_eq!(byte(s, 1), b'"');
+        s = &s[2..];
+
+        // We're going to want to have slices which don't respect codepoint boundaries.
+        let mut v = s.as_bytes();
+
+        let mut out = Vec::new();
+        'outer: loop {
+            let byte = match byte(v, 0) {
+                b'"' => break,
+                b'\\' => {
+                    let b = byte(v, 1);
+                    v = &v[2..];
+                    match b {
+                        b'x' => {
+                            let (b, rest) = backslash_x(v);
+                            assert!(b != 0, "\\x00 is not allowed in C-string literal");
+                            v = rest;
+                            b
+                        }
+                        b'u' => {
+                            let (ch, rest) = backslash_u(v);
+                            assert!(ch != '\0', "\\u{{0}} is not allowed in C-string literal");
+                            v = rest;
+                            out.extend_from_slice(ch.encode_utf8(&mut [0u8; 4]).as_bytes());
+                            continue 'outer;
+                        }
+                        b'n' => b'\n',
+                        b'r' => b'\r',
+                        b't' => b'\t',
+                        b'\\' => b'\\',
+                        b'\'' => b'\'',
+                        b'"' => b'"',
+                        b'\r' | b'\n' => loop {
+                            let byte = byte(v, 0);
+                            if matches!(byte, b' ' | b'\t' | b'\n' | b'\r') {
+                                v = &v[1..];
+                            } else {
+                                continue 'outer;
+                            }
+                        },
+                        b => panic!(
+                            "unexpected byte '{}' after \\ character in byte literal",
+                            ascii::escape_default(b),
+                        ),
+                    }
+                }
+                b'\r' => {
+                    assert_eq!(byte(v, 1), b'\n', "bare CR not allowed in string");
+                    v = &v[2..];
+                    b'\n'
+                }
+                b => {
+                    v = &v[1..];
+                    b
+                }
+            };
+            out.push(byte);
+        }
+
+        assert_eq!(byte(v, 0), b'"');
+        let suffix = s[s.len() - v.len() + 1..].to_owned().into_boxed_str();
+        (CString::new(out).unwrap(), suffix)
+    }
+
+    fn parse_lit_c_str_raw(s: &str) -> (CString, Box<str>) {
+        assert_eq!(byte(s, 0), b'c');
+        let (value, suffix) = parse_lit_str_raw(&s[1..]);
+        (CString::new(String::from(value)).unwrap(), suffix)
+    }
+
+    // Returns (value, suffix).
+    pub(crate) fn parse_lit_byte(s: &str) -> (u8, Box<str>) {
+        assert_eq!(byte(s, 0), b'b');
+        assert_eq!(byte(s, 1), b'\'');
+
+        // We're going to want to have slices which don't respect codepoint boundaries.
+        let mut v = &s.as_bytes()[2..];
+
+        let b = match byte(v, 0) {
+            b'\\' => {
+                let b = byte(v, 1);
+                v = &v[2..];
+                match b {
+                    b'x' => {
+                        let (b, rest) = backslash_x(v);
+                        v = rest;
+                        b
+                    }
+                    b'n' => b'\n',
+                    b'r' => b'\r',
+                    b't' => b'\t',
+                    b'\\' => b'\\',
+                    b'0' => b'\0',
+                    b'\'' => b'\'',
+                    b'"' => b'"',
+                    b => panic!(
+                        "unexpected byte '{}' after \\ character in byte literal",
+                        ascii::escape_default(b),
+                    ),
+                }
+            }
+            b => {
+                v = &v[1..];
+                b
+            }
+        };
+
+        assert_eq!(byte(v, 0), b'\'');
+        let suffix = s[s.len() - v.len() + 1..].to_owned().into_boxed_str();
+        (b, suffix)
+    }
+
+    // Returns (value, suffix).
+    pub(crate) fn parse_lit_char(mut s: &str) -> (char, Box<str>) {
+        assert_eq!(byte(s, 0), b'\'');
+        s = &s[1..];
+
+        let ch = match byte(s, 0) {
+            b'\\' => {
+                let b = byte(s, 1);
+                s = &s[2..];
+                match b {
+                    b'x' => {
+                        let (byte, rest) = backslash_x(s);
+                        s = rest;
+                        assert!(byte <= 0x7F, "invalid \\x byte in character literal");
+                        char::from_u32(u32::from(byte)).unwrap()
+                    }
+                    b'u' => {
+                        let (ch, rest) = backslash_u(s);
+                        s = rest;
+                        ch
+                    }
+                    b'n' => '\n',
+                    b'r' => '\r',
+                    b't' => '\t',
+                    b'\\' => '\\',
+                    b'0' => '\0',
+                    b'\'' => '\'',
+                    b'"' => '"',
+                    b => panic!(
+                        "unexpected byte '{}' after \\ character in character literal",
+                        ascii::escape_default(b),
+                    ),
+                }
+            }
+            _ => {
+                let ch = next_chr(s);
+                s = &s[ch.len_utf8()..];
+                ch
+            }
+        };
+        assert_eq!(byte(s, 0), b'\'');
+        let suffix = s[1..].to_owned().into_boxed_str();
+        (ch, suffix)
+    }
+
+    fn backslash_x<S>(s: &S) -> (u8, &S)
+    where
+        S: Index<RangeFrom<usize>, Output = S> + AsRef<[u8]> + ?Sized,
+    {
+        let mut ch = 0;
+        let b0 = byte(s, 0);
+        let b1 = byte(s, 1);
+        ch += 0x10
+            * match b0 {
+                b'0'..=b'9' => b0 - b'0',
+                b'a'..=b'f' => 10 + (b0 - b'a'),
+                b'A'..=b'F' => 10 + (b0 - b'A'),
+                _ => panic!("unexpected non-hex character after \\x"),
+            };
+        ch += match b1 {
+            b'0'..=b'9' => b1 - b'0',
+            b'a'..=b'f' => 10 + (b1 - b'a'),
+            b'A'..=b'F' => 10 + (b1 - b'A'),
+            _ => panic!("unexpected non-hex character after \\x"),
+        };
+        (ch, &s[2..])
+    }
+
+    fn backslash_u<S>(mut s: &S) -> (char, &S)
+    where
+        S: Index<RangeFrom<usize>, Output = S> + AsRef<[u8]> + ?Sized,
+    {
+        if byte(s, 0) != b'{' {
+            panic!("{}", "expected { after \\u");
+        }
+        s = &s[1..];
+
+        let mut ch = 0;
+        let mut digits = 0;
+        loop {
+            let b = byte(s, 0);
+            let digit = match b {
+                b'0'..=b'9' => b - b'0',
+                b'a'..=b'f' => 10 + b - b'a',
+                b'A'..=b'F' => 10 + b - b'A',
+                b'_' if digits > 0 => {
+                    s = &s[1..];
+                    continue;
+                }
+                b'}' if digits == 0 => panic!("invalid empty unicode escape"),
+                b'}' => break,
+                _ => panic!("unexpected non-hex character after \\u"),
+            };
+            if digits == 6 {
+                panic!("overlong unicode escape (must have at most 6 hex digits)");
+            }
+            ch *= 0x10;
+            ch += u32::from(digit);
+            digits += 1;
+            s = &s[1..];
+        }
+        assert!(byte(s, 0) == b'}');
+        s = &s[1..];
+
+        if let Some(ch) = char::from_u32(ch) {
+            (ch, s)
+        } else {
+            panic!("character code {:x} is not a valid unicode character", ch);
+        }
+    }
+
+    // Returns base 10 digits and suffix.
+    pub(crate) fn parse_lit_int(mut s: &str) -> Option<(Box<str>, Box<str>)> {
+        let negative = byte(s, 0) == b'-';
+        if negative {
+            s = &s[1..];
+        }
+
+        let base = match (byte(s, 0), byte(s, 1)) {
+            (b'0', b'x') => {
+                s = &s[2..];
+                16
+            }
+            (b'0', b'o') => {
+                s = &s[2..];
+                8
+            }
+            (b'0', b'b') => {
+                s = &s[2..];
+                2
+            }
+            (b'0'..=b'9', _) => 10,
+            _ => return None,
+        };
+
+        let mut value = BigInt::new();
+        let mut has_digit = false;
+        'outer: loop {
+            let b = byte(s, 0);
+            let digit = match b {
+                b'0'..=b'9' => b - b'0',
+                b'a'..=b'f' if base > 10 => b - b'a' + 10,
+                b'A'..=b'F' if base > 10 => b - b'A' + 10,
+                b'_' => {
+                    s = &s[1..];
+                    continue;
+                }
+                // If looking at a floating point literal, we don't want to
+                // consider it an integer.
+                b'.' if base == 10 => return None,
+                b'e' | b'E' if base == 10 => {
+                    let mut has_exp = false;
+                    for (i, b) in s[1..].bytes().enumerate() {
+                        match b {
+                            b'_' => {}
+                            b'-' | b'+' => return None,
+                            b'0'..=b'9' => has_exp = true,
+                            _ => {
+                                let suffix = &s[1 + i..];
+                                if has_exp && crate::ident::xid_ok(suffix) {
+                                    return None;
+                                } else {
+                                    break 'outer;
+                                }
+                            }
+                        }
+                    }
+                    if has_exp {
+                        return None;
+                    } else {
+                        break;
+                    }
+                }
+                _ => break,
+            };
+
+            if digit >= base {
+                return None;
+            }
+
+            has_digit = true;
+            value *= base;
+            value += digit;
+            s = &s[1..];
+        }
+
+        if !has_digit {
+            return None;
+        }
+
+        let suffix = s;
+        if suffix.is_empty() || crate::ident::xid_ok(suffix) {
+            let mut repr = value.to_string();
+            if negative {
+                repr.insert(0, '-');
+            }
+            Some((repr.into_boxed_str(), suffix.to_owned().into_boxed_str()))
+        } else {
+            None
+        }
+    }
+
+    // Returns base 10 digits and suffix.
+    pub(crate) fn parse_lit_float(input: &str) -> Option<(Box<str>, Box<str>)> {
+        // Rust's floating point literals are very similar to the ones parsed by
+        // the standard library, except that rust's literals can contain
+        // ignorable underscores. Let's remove those underscores.
+
+        let mut bytes = input.to_owned().into_bytes();
+
+        let start = (*bytes.first()? == b'-') as usize;
+        match bytes.get(start)? {
+            b'0'..=b'9' => {}
+            _ => return None,
+        }
+
+        let mut read = start;
+        let mut write = start;
+        let mut has_dot = false;
+        let mut has_e = false;
+        let mut has_sign = false;
+        let mut has_exponent = false;
+        while read < bytes.len() {
+            match bytes[read] {
+                b'_' => {
+                    // Don't increase write
+                    read += 1;
+                    continue;
+                }
+                b'0'..=b'9' => {
+                    if has_e {
+                        has_exponent = true;
+                    }
+                    bytes[write] = bytes[read];
+                }
+                b'.' => {
+                    if has_e || has_dot {
+                        return None;
+                    }
+                    has_dot = true;
+                    bytes[write] = b'.';
+                }
+                b'e' | b'E' => {
+                    match bytes[read + 1..]
+                        .iter()
+                        .find(|b| **b != b'_')
+                        .unwrap_or(&b'\0')
+                    {
+                        b'-' | b'+' | b'0'..=b'9' => {}
+                        _ => break,
+                    }
+                    if has_e {
+                        if has_exponent {
+                            break;
+                        } else {
+                            return None;
+                        }
+                    }
+                    has_e = true;
+                    bytes[write] = b'e';
+                }
+                b'-' | b'+' => {
+                    if has_sign || has_exponent || !has_e {
+                        return None;
+                    }
+                    has_sign = true;
+                    if bytes[read] == b'-' {
+                        bytes[write] = bytes[read];
+                    } else {
+                        // Omit '+'
+                        read += 1;
+                        continue;
+                    }
+                }
+                _ => break,
+            }
+            read += 1;
+            write += 1;
+        }
+
+        if has_e && !has_exponent {
+            return None;
+        }
+
+        let mut digits = String::from_utf8(bytes).unwrap();
+        let suffix = digits.split_off(read);
+        digits.truncate(write);
+        if suffix.is_empty() || crate::ident::xid_ok(&suffix) {
+            Some((digits.into_boxed_str(), suffix.into_boxed_str()))
+        } else {
+            None
+        }
+    }
+}