From 478f8287c0e2c35cda511fd3ac01b7ac78ee7cfe Mon Sep 17 00:00:00 2001 From: Nicholas Nethercote Date: Mon, 2 Jun 2025 08:59:29 +1000 Subject: Introduce `ByteSymbol`. It's like `Symbol` but for byte strings. The interner is now used for both `Symbol` and `ByteSymbol`. E.g. if you intern `"dog"` and `b"dog"` you'll get a `Symbol` and a `ByteSymbol` with the same index and the characters will only be stored once. The motivation for this is to eliminate the `Arc`s in `ast::LitKind`, to make `ast::LitKind` impl `Copy`, and to avoid the need to arena-allocate `ast::LitKind` in HIR. The latter change reduces peak memory by a non-trivial amount on literal-heavy benchmarks such as `deep-vector` and `tuple-stress`. `Encoder`, `Decoder`, `SpanEncoder`, and `SpanDecoder` all get some changes so that they can handle normal strings and byte strings. This change does slow down compilation of programs that use `include_bytes!` on large files, because the contents of those files are now interned (hashed). This makes `include_bytes!` more similar to `include_str!`, though `include_bytes!` contents still aren't escaped, and hashing is still much cheaper than escaping. --- src/tools/clippy/clippy_utils/src/consts.rs | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'src/tools/clippy/clippy_utils') diff --git a/src/tools/clippy/clippy_utils/src/consts.rs b/src/tools/clippy/clippy_utils/src/consts.rs index aaa071fd5c9..09299c869dc 100644 --- a/src/tools/clippy/clippy_utils/src/consts.rs +++ b/src/tools/clippy/clippy_utils/src/consts.rs @@ -4,8 +4,6 @@ //! executable MIR bodies, so we have to do this instead. #![allow(clippy::float_cmp)] -use std::sync::Arc; - use crate::source::{SpanRangeExt, walk_span_to_context}; use crate::{clip, is_direct_expn_of, sext, unsext}; @@ -38,7 +36,7 @@ pub enum Constant<'tcx> { /// A `String` (e.g., "abc"). Str(String), /// A binary string (e.g., `b"abc"`). - Binary(Arc<[u8]>), + Binary(Vec), /// A single `char` (e.g., `'a'`). Char(char), /// An integer's bit representation. @@ -306,7 +304,9 @@ pub fn lit_to_mir_constant<'tcx>(lit: &LitKind, ty: Option>) -> Constan match *lit { LitKind::Str(ref is, _) => Constant::Str(is.to_string()), LitKind::Byte(b) => Constant::Int(u128::from(b)), - LitKind::ByteStr(ref s, _) | LitKind::CStr(ref s, _) => Constant::Binary(Arc::clone(s)), + LitKind::ByteStr(ref s, _) | LitKind::CStr(ref s, _) => { + Constant::Binary(s.as_byte_str().to_vec()) + } LitKind::Char(c) => Constant::Char(c), LitKind::Int(n, _) => Constant::Int(n.get()), LitKind::Float(ref is, LitFloatType::Suffixed(fty)) => match fty { @@ -568,7 +568,9 @@ impl<'tcx> ConstEvalCtxt<'tcx> { } else { match &lit.node { LitKind::Str(is, _) => Some(is.is_empty()), - LitKind::ByteStr(s, _) | LitKind::CStr(s, _) => Some(s.is_empty()), + LitKind::ByteStr(s, _) | LitKind::CStr(s, _) => { + Some(s.as_byte_str().is_empty()) + } _ => None, } } -- cgit 1.4.1-3-g733a5