Avoid unnecessary string interning for const_str

Author: bjorn3

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};
@@ -181,22 +180,27 @@ 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/mir/block.rs

@@ -481,8 +481,7 @@ impl<'a, 'tcx, Bx: BuilderMethods<'a, 'tcx>> FunctionCx<'a, 'tcx, Bx> {
                 (LangItem::PanicBoundsCheck, vec![index, len, location])
             }
             _ => {
-                let msg_str = Symbol::intern(msg.description());
-                let msg = bx.const_str(msg_str);
+                let msg = bx.const_str(msg.description());
                 // It's `pub fn panic(expr: &str)`, with the wide reference being passed
                 // as two arguments, and `#[track_caller]` adds an implicit third argument.
                 (LangItem::Panic, vec![msg.0, msg.1, location])
@@ -563,7 +562,7 @@ impl<'a, 'tcx, Bx: BuilderMethods<'a, 'tcx>> FunctionCx<'a, 'tcx, Bx> {
                         }
                     })
                 });
-                let msg = bx.const_str(Symbol::intern(&msg_str));
+                let msg = bx.const_str(&msg_str);
                 let location = self.get_caller_location(bx, source_info).immediate();
 
                 // Obtain the panic entry point.

compiler/rustc_codegen_ssa/src/traits/consts.rs

@@ -2,7 +2,6 @@ use super::BackendTypes;
 use crate::mir::place::PlaceRef;
 use rustc_middle::mir::interpret::{ConstAllocation, Scalar};
 use rustc_middle::ty::layout::TyAndLayout;
-use rustc_span::Symbol;
 use rustc_target::abi::{self, Size};
 
 pub trait ConstMethods<'tcx>: BackendTypes {
@@ -21,7 +20,7 @@ pub trait ConstMethods<'tcx>: BackendTypes {
     fn const_u8(&self, i: u8) -> Self::Value;
     fn const_real(&self, t: Self::Type, val: f64) -> Self::Value;
 
-    fn const_str(&self, s: Symbol) -> (Self::Value, Self::Value);
+    fn const_str(&self, s: &str) -> (Self::Value, Self::Value);
     fn const_struct(&self, elts: &[Self::Value], packed: bool) -> Self::Value;
 
     fn const_to_opt_uint(&self, v: Self::Value) -> Option<u64>;