Auto merge of #98638 - bjorn3:less_string_interning, r=tmiasko

Use less string interning

This removes string interning in a couple of places where doing so won't result in perf improvements. I also switched one place to use pre-interned symbols.

Author: bors

compiler/rustc_codegen_gcc/src/common.rs

@@ -12,7 +12,6 @@ use rustc_middle::mir::Mutability;
 use rustc_middle::ty::ScalarInt;
 use rustc_middle::ty::layout::{TyAndLayout, LayoutOf};
 use rustc_middle::mir::interpret::{ConstAllocation, GlobalAlloc, Scalar};
-use rustc_span::Symbol;
 use rustc_target::abi::{self, HasDataLayout, Pointer, Size};
 
 use crate::consts::const_alloc_to_gcc;
@@ -125,12 +124,15 @@ impl<'gcc, 'tcx> ConstMethods<'tcx> for CodegenCx<'gcc, 'tcx> {
         self.context.new_rvalue_from_double(typ, val)
     }
 
-    fn const_str(&self, s: Symbol) -> (RValue<'gcc>, RValue<'gcc>) {
-        let s_str = s.as_str();
-        let str_global = *self.const_str_cache.borrow_mut().entry(s).or_insert_with(|| {
-            self.global_string(s_str)
-        });
-        let len = s_str.len();
+    fn const_str(&self, s: &str) -> (RValue<'gcc>, RValue<'gcc>) {
+        let str_global = *self
+            .const_str_cache
+            .borrow_mut()
+            .raw_entry_mut()
+            .from_key(s)
+            .or_insert_with(|| (s.to_owned(), self.global_string(s)))
+            .1;
+        let len = s.len();
         let cs = self.const_ptrcast(str_global.get_address(None),
             self.type_ptr_to(self.layout_of(self.tcx.types.str_).gcc_type(self, true)),
         );

compiler/rustc_codegen_gcc/src/context.rs

@@ -13,7 +13,7 @@ use rustc_middle::mir::mono::CodegenUnit;
 use rustc_middle::ty::{self, Instance, ParamEnv, PolyExistentialTraitRef, Ty, TyCtxt};
 use rustc_middle::ty::layout::{FnAbiError, FnAbiOfHelpers, FnAbiRequest, HasParamEnv, HasTyCtxt, LayoutError, TyAndLayout, LayoutOfHelpers};
 use rustc_session::Session;
-use rustc_span::{Span, Symbol};
+use rustc_span::Span;
 use rustc_target::abi::{call::FnAbi, HasDataLayout, PointeeInfo, Size, TargetDataLayout, VariantIdx};
 use rustc_target::spec::{HasTargetSpec, Target, TlsModel};
 
@@ -101,7 +101,7 @@ pub struct CodegenCx<'gcc, 'tcx> {
     pub global_lvalues: RefCell<FxHashMap<RValue<'gcc>, LValue<'gcc>>>,
 
     /// Cache of constant strings,
-    pub const_str_cache: RefCell<FxHashMap<Symbol, LValue<'gcc>>>,
+    pub const_str_cache: RefCell<FxHashMap<String, LValue<'gcc>>>,
 
     /// Cache of globals.
     pub globals: RefCell<FxHashMap<String, RValue<'gcc>>>,

compiler/rustc_codegen_gcc/src/lib.rs

@@ -6,7 +6,14 @@
  * TODO(antoyo): remove the patches.
  */
 
-#![feature(rustc_private, decl_macro, associated_type_bounds, never_type, trusted_len)]
+#![feature(
+    rustc_private,
+    decl_macro,
+    associated_type_bounds,
+    never_type,
+    trusted_len,
+    hash_raw_entry
+)]
 #![allow(broken_intra_doc_links)]
 #![recursion_limit="256"]
 #![warn(rust_2018_idioms)]

compiler/rustc_codegen_llvm/src/common.rs

@@ -14,7 +14,6 @@ use rustc_middle::bug;
 use rustc_middle::mir::interpret::{ConstAllocation, GlobalAlloc, Scalar};
 use rustc_middle::ty::layout::{LayoutOf, TyAndLayout};
 use rustc_middle::ty::ScalarInt;
-use rustc_span::symbol::Symbol;
 use rustc_target::abi::{self, AddressSpace, HasDataLayout, Pointer, Size};
 
 use libc::{c_char, c_uint};
@@ -180,22 +179,27 @@ impl<'ll, 'tcx> ConstMethods<'tcx> for CodegenCx<'ll, 'tcx> {
         unsafe { llvm::LLVMConstReal(t, val) }
     }
 
-    fn const_str(&self, s: Symbol) -> (&'ll Value, &'ll Value) {
-        let s_str = s.as_str();
-        let str_global = *self.const_str_cache.borrow_mut().entry(s).or_insert_with(|| {
-            let sc = self.const_bytes(s_str.as_bytes());
-            let sym = self.generate_local_symbol_name("str");
-            let g = self.define_global(&sym, self.val_ty(sc)).unwrap_or_else(|| {
-                bug!("symbol `{}` is already defined", sym);
-            });
-            unsafe {
-                llvm::LLVMSetInitializer(g, sc);
-                llvm::LLVMSetGlobalConstant(g, True);
-                llvm::LLVMRustSetLinkage(g, llvm::Linkage::InternalLinkage);
-            }
-            g
-        });
-        let len = s_str.len();
+    fn const_str(&self, s: &str) -> (&'ll Value, &'ll Value) {
+        let str_global = *self
+            .const_str_cache
+            .borrow_mut()
+            .raw_entry_mut()
+            .from_key(s)
+            .or_insert_with(|| {
+                let sc = self.const_bytes(s.as_bytes());
+                let sym = self.generate_local_symbol_name("str");
+                let g = self.define_global(&sym, self.val_ty(sc)).unwrap_or_else(|| {
+                    bug!("symbol `{}` is already defined", sym);
+                });
+                unsafe {
+                    llvm::LLVMSetInitializer(g, sc);
+                    llvm::LLVMSetGlobalConstant(g, True);
+                    llvm::LLVMRustSetLinkage(g, llvm::Linkage::InternalLinkage);
+                }
+                (s.to_owned(), g)
+            })
+            .1;
+        let len = s.len();
         let cs = consts::ptrcast(
             str_global,
             self.type_ptr_to(self.layout_of(self.tcx.types.str_).llvm_type(self)),

compiler/rustc_codegen_llvm/src/context.rs

@@ -26,7 +26,6 @@ use rustc_session::config::{BranchProtection, CFGuard, CFProtection};
 use rustc_session::config::{CrateType, DebugInfo, PAuthKey, PacRet};
 use rustc_session::Session;
 use rustc_span::source_map::Span;
-use rustc_span::symbol::Symbol;
 use rustc_target::abi::{
     call::FnAbi, HasDataLayout, PointeeInfo, Size, TargetDataLayout, VariantIdx,
 };
@@ -56,7 +55,7 @@ pub struct CodegenCx<'ll, 'tcx> {
     pub vtables:
         RefCell<FxHashMap<(Ty<'tcx>, Option<ty::PolyExistentialTraitRef<'tcx>>), &'ll Value>>,
     /// Cache of constant strings,
-    pub const_str_cache: RefCell<FxHashMap<Symbol, &'ll Value>>,
+    pub const_str_cache: RefCell<FxHashMap<String, &'ll Value>>,
 
     /// Reverse-direction for const ptrs cast from globals.
     ///

compiler/rustc_codegen_llvm/src/lib.rs

@@ -5,6 +5,7 @@
 //! This API is completely unstable and subject to change.
 
 #![doc(html_root_url = "https://doc.rust-lang.org/nightly/nightly-rustc/")]
+#![feature(hash_raw_entry)]
 #![feature(let_chains)]
 #![feature(let_else)]
 #![feature(extern_types)]

compiler/rustc_codegen_ssa/src/back/archive.rs

@@ -1,13 +1,12 @@
 use rustc_data_structures::temp_dir::MaybeTempDir;
 use rustc_session::cstore::DllImport;
 use rustc_session::Session;
-use rustc_span::symbol::Symbol;
 
 use std::io;
 use std::path::{Path, PathBuf};
 
 pub(super) fn find_library(
-    name: Symbol,
+    name: &str,
     verbatim: bool,
     search_paths: &[PathBuf],
     sess: &Session,

compiler/rustc_codegen_ssa/src/back/command.rs

@@ -7,7 +7,6 @@ use std::io;
 use std::mem;
 use std::process::{self, Output};
 
-use rustc_span::symbol::Symbol;
 use rustc_target::spec::LldFlavor;
 
 #[derive(Clone)]
@@ -47,11 +46,6 @@ impl Command {
         self
     }
 
-    pub fn sym_arg(&mut self, arg: Symbol) -> &mut Command {
-        self.arg(arg.as_str());
-        self
-    }
-
     pub fn args<I>(&mut self, args: I) -> &mut Command
     where
         I: IntoIterator<Item: AsRef<OsStr>>,

compiler/rustc_codegen_ssa/src/back/link.rs

@@ -358,7 +358,7 @@ fn link_rlib<'a, B: ArchiveBuilder<'a>>(
         }
         if let Some(name) = lib.name {
             let location =
-                find_library(name, lib.verbatim.unwrap_or(false), &lib_search_paths, sess);
+                find_library(name.as_str(), lib.verbatim.unwrap_or(false), &lib_search_paths, sess);
             ab.add_archive(&location, |_| false).unwrap_or_else(|e| {
                 sess.fatal(&format!(
                     "failed to add native library {}: {}",
@@ -1122,7 +1122,7 @@ fn link_sanitizer_runtime(sess: &Session, linker: &mut dyn Linker, name: &str) {
         let path = find_sanitizer_runtime(&sess, &filename);
         let rpath = path.to_str().expect("non-utf8 component in path");
         linker.args(&["-Wl,-rpath", "-Xlinker", rpath]);
-        linker.link_dylib(Symbol::intern(&filename), false, true);
+        linker.link_dylib(&filename, false, true);
     } else {
         let filename = format!("librustc{}_rt.{}.a", channel, name);
         let path = find_sanitizer_runtime(&sess, &filename).join(&filename);
@@ -2204,6 +2204,7 @@ fn add_local_native_libraries(
         let Some(name) = lib.name else {
             continue;
         };
+        let name = name.as_str();
 
         // Skip if this library is the same as the last.
         last = if (lib.name, lib.kind, lib.verbatim) == last {
@@ -2367,6 +2368,7 @@ fn add_upstream_rust_crates<'a, B: ArchiveBuilder<'a>>(
                         let Some(name) = lib.name else {
                             continue;
                         };
+                        let name = name.as_str();
                         if !relevant_lib(sess, lib) {
                             continue;
                         }
@@ -2524,7 +2526,7 @@ fn add_upstream_rust_crates<'a, B: ArchiveBuilder<'a>>(
         }
         let filestem = cratepath.file_stem().unwrap().to_str().unwrap();
         cmd.link_rust_dylib(
-            Symbol::intern(&unlib(&sess.target, filestem)),
+            &unlib(&sess.target, filestem),
             parent.unwrap_or_else(|| Path::new("")),
         );
     }
@@ -2556,6 +2558,7 @@ fn add_upstream_native_libraries(
             let Some(name) = lib.name else {
                 continue;
             };
+            let name = name.as_str();
             if !relevant_lib(sess, &lib) {
                 continue;
             }