diff --git a/compiler/rustc_codegen_gcc/src/lib.rs b/compiler/rustc_codegen_gcc/src/lib.rs index 2a6b642782dfd..04ac0254a81ba 100644 --- a/compiler/rustc_codegen_gcc/src/lib.rs +++ b/compiler/rustc_codegen_gcc/src/lib.rs @@ -239,6 +239,10 @@ impl WriteBackendMethods for GccCodegenBackend { unimplemented!(); } + fn print_statistics(&self) { + unimplemented!() + } + unsafe fn optimize(_cgcx: &CodegenContext, _diag_handler: &Handler, module: &ModuleCodegen, config: &ModuleConfig) -> Result<(), FatalError> { module.module_llvm.context.set_optimization_level(to_gcc_opt_level(config.opt_level)); Ok(()) diff --git a/compiler/rustc_codegen_llvm/src/lib.rs b/compiler/rustc_codegen_llvm/src/lib.rs index 24ba28bbc82c2..5aab06febe44b 100644 --- a/compiler/rustc_codegen_llvm/src/lib.rs +++ b/compiler/rustc_codegen_llvm/src/lib.rs @@ -40,12 +40,13 @@ use rustc_metadata::EncodedMetadata; use rustc_middle::dep_graph::{WorkProduct, WorkProductId}; use rustc_middle::query::Providers; use rustc_middle::ty::TyCtxt; -use rustc_session::config::{OptLevel, OutputFilenames, PrintRequest}; +use rustc_session::config::{OptLevel, OutputFilenames, PrintKind, PrintRequest}; use rustc_session::Session; use rustc_span::symbol::Symbol; use std::any::Any; use std::ffi::CStr; +use std::io::Write; mod back { pub mod archive; @@ -178,7 +179,28 @@ impl WriteBackendMethods for LlvmCodegenBackend { type ThinBuffer = back::lto::ThinBuffer; fn print_pass_timings(&self) { unsafe { - llvm::LLVMRustPrintPassTimings(); + let mut size = 0; + let cstr = llvm::LLVMRustPrintPassTimings(&mut size as *mut usize); + if cstr.is_null() { + println!("failed to get pass timings"); + } else { + let timings = std::slice::from_raw_parts(cstr as *const u8, size); + std::io::stdout().write_all(timings).unwrap(); + libc::free(cstr as *mut _); + } + } + } + fn print_statistics(&self) { + unsafe { + let mut size = 0; + let cstr = llvm::LLVMRustPrintStatistics(&mut size as *mut usize); + if cstr.is_null() { + println!("failed to get pass stats"); + } else { + let stats = std::slice::from_raw_parts(cstr as *const u8, size); + std::io::stdout().write_all(stats).unwrap(); + libc::free(cstr as *mut _); + } } } fn run_link( @@ -262,10 +284,10 @@ impl CodegenBackend for LlvmCodegenBackend { |tcx, ()| llvm_util::global_llvm_features(tcx.sess, true) } - fn print(&self, req: PrintRequest, sess: &Session) { - match req { - PrintRequest::RelocationModels => { - println!("Available relocation models:"); + fn print(&self, req: &PrintRequest, out: &mut dyn PrintBackendInfo, sess: &Session) { + match req.kind { + PrintKind::RelocationModels => { + writeln!(out, "Available relocation models:"); for name in &[ "static", "pic", @@ -276,26 +298,27 @@ impl CodegenBackend for LlvmCodegenBackend { "ropi-rwpi", "default", ] { - println!(" {}", name); + writeln!(out, " {}", name); } - println!(); + writeln!(out); } - PrintRequest::CodeModels => { - println!("Available code models:"); + PrintKind::CodeModels => { + writeln!(out, "Available code models:"); for name in &["tiny", "small", "kernel", "medium", "large"] { - println!(" {}", name); + writeln!(out, " {}", name); } - println!(); + writeln!(out); } - PrintRequest::TlsModels => { - println!("Available TLS models:"); + PrintKind::TlsModels => { + writeln!(out, "Available TLS models:"); for name in &["global-dynamic", "local-dynamic", "initial-exec", "local-exec"] { - println!(" {}", name); + writeln!(out, " {}", name); } - println!(); + writeln!(out); } - PrintRequest::StackProtectorStrategies => { - println!( + PrintKind::StackProtectorStrategies => { + writeln!( + out, r#"Available stack protector strategies: all Generate stack canaries in all functions. @@ -319,7 +342,7 @@ impl CodegenBackend for LlvmCodegenBackend { "# ); } - req => llvm_util::print(req, sess), + _other => llvm_util::print(req, out, sess), } } diff --git a/compiler/rustc_codegen_llvm/src/llvm/ffi.rs b/compiler/rustc_codegen_llvm/src/llvm/ffi.rs index 605f0154773a7..5fe6407ac4fe8 100644 --- a/compiler/rustc_codegen_llvm/src/llvm/ffi.rs +++ b/compiler/rustc_codegen_llvm/src/llvm/ffi.rs @@ -1868,7 +1868,10 @@ extern "C" { pub fn LLVMRustGetLastError() -> *const c_char; /// Print the pass timings since static dtors aren't picking them up. - pub fn LLVMRustPrintPassTimings(); + pub fn LLVMRustPrintPassTimings(size: *const size_t) -> *const c_char; + + /// Print the statistics since static dtors aren't picking them up. + pub fn LLVMRustPrintStatistics(size: *const size_t) -> *const c_char; pub fn LLVMStructCreateNamed(C: &Context, Name: *const c_char) -> &Type; @@ -2280,7 +2283,12 @@ extern "C" { pub fn LLVMRustHasFeature(T: &TargetMachine, s: *const c_char) -> bool; - pub fn LLVMRustPrintTargetCPUs(T: &TargetMachine, cpu: *const c_char); + pub fn LLVMRustPrintTargetCPUs( + T: &TargetMachine, + cpu: *const c_char, + print: unsafe extern "C" fn(out: *mut c_void, string: *const c_char, len: usize), + out: *mut c_void, + ); pub fn LLVMRustGetTargetFeaturesCount(T: &TargetMachine) -> size_t; pub fn LLVMRustGetTargetFeature( T: &TargetMachine, diff --git a/compiler/rustc_codegen_llvm/src/llvm_util.rs b/compiler/rustc_codegen_llvm/src/llvm_util.rs index 03be0654b50bb..4f20fbf2045a9 100644 --- a/compiler/rustc_codegen_llvm/src/llvm_util.rs +++ b/compiler/rustc_codegen_llvm/src/llvm_util.rs @@ -8,16 +8,17 @@ use libc::c_int; use rustc_codegen_ssa::target_features::{ supported_target_features, tied_target_features, RUSTC_SPECIFIC_FEATURES, }; +use rustc_codegen_ssa::traits::PrintBackendInfo; use rustc_data_structures::fx::{FxHashMap, FxHashSet}; use rustc_data_structures::small_c_str::SmallCStr; use rustc_fs_util::path_to_c_string; use rustc_middle::bug; -use rustc_session::config::PrintRequest; +use rustc_session::config::{PrintKind, PrintRequest}; use rustc_session::Session; use rustc_span::symbol::Symbol; use rustc_target::spec::{MergeFunctions, PanicStrategy}; -use std::ffi::{CStr, CString}; +use std::ffi::{c_char, c_void, CStr, CString}; use std::path::Path; use std::ptr; use std::slice; @@ -110,6 +111,10 @@ unsafe fn configure_llvm(sess: &Session) { // Use non-zero `import-instr-limit` multiplier for cold callsites. add("-import-cold-multiplier=0.1", false); + if sess.print_llvm_stats() { + add("-stats", false); + } + for arg in sess_args { add(&(*arg), true); } @@ -350,7 +355,7 @@ fn llvm_target_features(tm: &llvm::TargetMachine) -> Vec<(&str, &str)> { ret } -fn print_target_features(sess: &Session, tm: &llvm::TargetMachine) { +fn print_target_features(out: &mut dyn PrintBackendInfo, sess: &Session, tm: &llvm::TargetMachine) { let mut llvm_target_features = llvm_target_features(tm); let mut known_llvm_target_features = FxHashSet::<&'static str>::default(); let mut rustc_target_features = supported_target_features(sess) @@ -383,36 +388,48 @@ fn print_target_features(sess: &Session, tm: &llvm::TargetMachine) { .max() .unwrap_or(0); - println!("Features supported by rustc for this target:"); + writeln!(out, "Features supported by rustc for this target:"); for (feature, desc) in &rustc_target_features { - println!(" {1:0$} - {2}.", max_feature_len, feature, desc); + writeln!(out, " {1:0$} - {2}.", max_feature_len, feature, desc); } - println!("\nCode-generation features supported by LLVM for this target:"); + writeln!(out, "\nCode-generation features supported by LLVM for this target:"); for (feature, desc) in &llvm_target_features { - println!(" {1:0$} - {2}.", max_feature_len, feature, desc); + writeln!(out, " {1:0$} - {2}.", max_feature_len, feature, desc); } if llvm_target_features.is_empty() { - println!(" Target features listing is not supported by this LLVM version."); + writeln!(out, " Target features listing is not supported by this LLVM version."); } - println!("\nUse +feature to enable a feature, or -feature to disable it."); - println!("For example, rustc -C target-cpu=mycpu -C target-feature=+feature1,-feature2\n"); - println!("Code-generation features cannot be used in cfg or #[target_feature],"); - println!("and may be renamed or removed in a future version of LLVM or rustc.\n"); + writeln!(out, "\nUse +feature to enable a feature, or -feature to disable it."); + writeln!(out, "For example, rustc -C target-cpu=mycpu -C target-feature=+feature1,-feature2\n"); + writeln!(out, "Code-generation features cannot be used in cfg or #[target_feature],"); + writeln!(out, "and may be renamed or removed in a future version of LLVM or rustc.\n"); } -pub(crate) fn print(req: PrintRequest, sess: &Session) { +pub(crate) fn print(req: &PrintRequest, mut out: &mut dyn PrintBackendInfo, sess: &Session) { require_inited(); let tm = create_informational_target_machine(sess); - match req { - PrintRequest::TargetCPUs => { + match req.kind { + PrintKind::TargetCPUs => { // SAFETY generate a C compatible string from a byte slice to pass // the target CPU name into LLVM, the lifetime of the reference is // at least as long as the C function let cpu_cstring = CString::new(handle_native(sess.target.cpu.as_ref())) .unwrap_or_else(|e| bug!("failed to convert to cstring: {}", e)); - unsafe { llvm::LLVMRustPrintTargetCPUs(tm, cpu_cstring.as_ptr()) }; + unsafe extern "C" fn callback(out: *mut c_void, string: *const c_char, len: usize) { + let out = &mut *(out as *mut &mut dyn PrintBackendInfo); + let bytes = slice::from_raw_parts(string as *const u8, len); + write!(out, "{}", String::from_utf8_lossy(bytes)); + } + unsafe { + llvm::LLVMRustPrintTargetCPUs( + tm, + cpu_cstring.as_ptr(), + callback, + &mut out as *mut &mut dyn PrintBackendInfo as *mut c_void, + ); + } } - PrintRequest::TargetFeatures => print_target_features(sess, tm), + PrintKind::TargetFeatures => print_target_features(out, sess, tm), _ => bug!("rustc_codegen_llvm can't handle print request: {:?}", req), } } diff --git a/compiler/rustc_codegen_ssa/messages.ftl b/compiler/rustc_codegen_ssa/messages.ftl index f73080182bfcd..b6c70c6224977 100644 --- a/compiler/rustc_codegen_ssa/messages.ftl +++ b/compiler/rustc_codegen_ssa/messages.ftl @@ -197,6 +197,8 @@ codegen_ssa_specify_libraries_to_link = use the `-l` flag to specify native libr codegen_ssa_static_library_native_artifacts = Link against the following native artifacts when linking against this static library. The order and any duplication can be significant on some platforms. +codegen_ssa_static_library_native_artifacts_to_file = Native artifacts to link against have been written to {$path}. The order and any duplication can be significant on some platforms. + codegen_ssa_stripping_debug_info_failed = stripping debug info with `{$util}` failed: {$status} .note = {$output} diff --git a/compiler/rustc_codegen_ssa/src/back/link.rs b/compiler/rustc_codegen_ssa/src/back/link.rs index 0dfb41f42f0dc..eefa4ac34dde2 100644 --- a/compiler/rustc_codegen_ssa/src/back/link.rs +++ b/compiler/rustc_codegen_ssa/src/back/link.rs @@ -12,8 +12,8 @@ use rustc_metadata::fs::{copy_to_stdout, emit_wrapper_file, METADATA_FILENAME}; use rustc_middle::middle::debugger_visualizer::DebuggerVisualizerFile; use rustc_middle::middle::dependency_format::Linkage; use rustc_middle::middle::exported_symbols::SymbolExportKind; -use rustc_session::config::{self, CFGuard, CrateType, DebugInfo, Strip}; -use rustc_session::config::{OutputFilenames, OutputType, PrintRequest, SplitDwarfKind}; +use rustc_session::config::{self, CFGuard, CrateType, DebugInfo, OutFileName, Strip}; +use rustc_session::config::{OutputFilenames, OutputType, PrintKind, SplitDwarfKind}; use rustc_session::cstore::DllImport; use rustc_session::output::{check_file_is_writeable, invalid_output_for_target, out_filename}; use rustc_session::search_paths::PathKind; @@ -596,8 +596,10 @@ fn link_staticlib<'a>( all_native_libs.extend_from_slice(&codegen_results.crate_info.used_libraries); - if sess.opts.prints.contains(&PrintRequest::NativeStaticLibs) { - print_native_static_libs(sess, &all_native_libs, &all_rust_dylibs); + for print in &sess.opts.prints { + if print.kind == PrintKind::NativeStaticLibs { + print_native_static_libs(sess, &print.out, &all_native_libs, &all_rust_dylibs); + } } Ok(()) @@ -744,8 +746,11 @@ fn link_natively<'a>( cmd.env_remove(k.as_ref()); } - if sess.opts.prints.contains(&PrintRequest::LinkArgs) { - println!("{:?}", &cmd); + for print in &sess.opts.prints { + if print.kind == PrintKind::LinkArgs { + let content = format!("{:?}", cmd); + print.out.overwrite(&content, sess); + } } // May have not found libraries in the right formats. @@ -1386,6 +1391,7 @@ enum RlibFlavor { fn print_native_static_libs( sess: &Session, + out: &OutFileName, all_native_libs: &[NativeLib], all_rust_dylibs: &[&Path], ) { @@ -1459,11 +1465,22 @@ fn print_native_static_libs( lib_args.push(format!("-l{}", lib)); } } - if !lib_args.is_empty() { - sess.emit_note(errors::StaticLibraryNativeArtifacts); - // Prefix for greppability - // Note: This must not be translated as tools are allowed to depend on this exact string. - sess.note_without_error(format!("native-static-libs: {}", &lib_args.join(" "))); + + match out { + OutFileName::Real(path) => { + out.overwrite(&lib_args.join(" "), sess); + if !lib_args.is_empty() { + sess.emit_note(errors::StaticLibraryNativeArtifactsToFile { path }); + } + } + OutFileName::Stdout => { + if !lib_args.is_empty() { + sess.emit_note(errors::StaticLibraryNativeArtifacts); + // Prefix for greppability + // Note: This must not be translated as tools are allowed to depend on this exact string. + sess.note_without_error(format!("native-static-libs: {}", &lib_args.join(" "))); + } + } } } diff --git a/compiler/rustc_codegen_ssa/src/back/write.rs b/compiler/rustc_codegen_ssa/src/back/write.rs index 1c5d7a7c68e92..6c781c6f0820d 100644 --- a/compiler/rustc_codegen_ssa/src/back/write.rs +++ b/compiler/rustc_codegen_ssa/src/back/write.rs @@ -1945,6 +1945,10 @@ impl OngoingCodegen { self.backend.print_pass_timings() } + if sess.print_llvm_stats() { + self.backend.print_statistics() + } + ( CodegenResults { metadata: self.metadata, diff --git a/compiler/rustc_codegen_ssa/src/errors.rs b/compiler/rustc_codegen_ssa/src/errors.rs index 056b4abd23533..c72d37be74815 100644 --- a/compiler/rustc_codegen_ssa/src/errors.rs +++ b/compiler/rustc_codegen_ssa/src/errors.rs @@ -455,6 +455,12 @@ pub struct LinkerFileStem; #[diag(codegen_ssa_static_library_native_artifacts)] pub struct StaticLibraryNativeArtifacts; +#[derive(Diagnostic)] +#[diag(codegen_ssa_static_library_native_artifacts_to_file)] +pub struct StaticLibraryNativeArtifactsToFile<'a> { + pub path: &'a Path, +} + #[derive(Diagnostic)] #[diag(codegen_ssa_link_script_unavailable)] pub struct LinkScriptUnavailable; diff --git a/compiler/rustc_codegen_ssa/src/traits/backend.rs b/compiler/rustc_codegen_ssa/src/traits/backend.rs index b3c9ecf8b938b..1991b55f19141 100644 --- a/compiler/rustc_codegen_ssa/src/traits/backend.rs +++ b/compiler/rustc_codegen_ssa/src/traits/backend.rs @@ -23,6 +23,8 @@ use rustc_span::symbol::Symbol; use rustc_target::abi::call::FnAbi; use rustc_target::spec::Target; +use std::fmt; + pub trait BackendTypes { type Value: CodegenObject; type Function: CodegenObject; @@ -61,7 +63,7 @@ pub trait CodegenBackend { fn locale_resource(&self) -> &'static str; fn init(&self, _sess: &Session) {} - fn print(&self, _req: PrintRequest, _sess: &Session) {} + fn print(&self, _req: &PrintRequest, _out: &mut dyn PrintBackendInfo, _sess: &Session) {} fn target_features(&self, _sess: &Session, _allow_unstable: bool) -> Vec { vec![] } @@ -162,3 +164,19 @@ pub trait ExtraBackendMethods: std::thread::Builder::new().name(name).spawn(f) } } + +pub trait PrintBackendInfo { + fn infallible_write_fmt(&mut self, args: fmt::Arguments<'_>); +} + +impl PrintBackendInfo for String { + fn infallible_write_fmt(&mut self, args: fmt::Arguments<'_>) { + fmt::Write::write_fmt(self, args).unwrap(); + } +} + +impl dyn PrintBackendInfo + '_ { + pub fn write_fmt(&mut self, args: fmt::Arguments<'_>) { + self.infallible_write_fmt(args); + } +} diff --git a/compiler/rustc_codegen_ssa/src/traits/mod.rs b/compiler/rustc_codegen_ssa/src/traits/mod.rs index 8cb58bd4c704d..728c2bc8c49bc 100644 --- a/compiler/rustc_codegen_ssa/src/traits/mod.rs +++ b/compiler/rustc_codegen_ssa/src/traits/mod.rs @@ -30,7 +30,9 @@ mod write; pub use self::abi::AbiBuilderMethods; pub use self::asm::{AsmBuilderMethods, AsmMethods, GlobalAsmOperandRef, InlineAsmOperandRef}; -pub use self::backend::{Backend, BackendTypes, CodegenBackend, ExtraBackendMethods}; +pub use self::backend::{ + Backend, BackendTypes, CodegenBackend, ExtraBackendMethods, PrintBackendInfo, +}; pub use self::builder::{BuilderMethods, OverflowOp}; pub use self::consts::ConstMethods; pub use self::coverageinfo::CoverageInfoBuilderMethods; diff --git a/compiler/rustc_codegen_ssa/src/traits/write.rs b/compiler/rustc_codegen_ssa/src/traits/write.rs index 9826256a4c5d5..a3b11ed4fe8f6 100644 --- a/compiler/rustc_codegen_ssa/src/traits/write.rs +++ b/compiler/rustc_codegen_ssa/src/traits/write.rs @@ -35,6 +35,7 @@ pub trait WriteBackendMethods: 'static + Sized + Clone { cached_modules: Vec<(SerializedModule, WorkProduct)>, ) -> Result<(Vec>, Vec), FatalError>; fn print_pass_timings(&self); + fn print_statistics(&self); unsafe fn optimize( cgcx: &CodegenContext, diag_handler: &Handler, diff --git a/compiler/rustc_driver_impl/messages.ftl b/compiler/rustc_driver_impl/messages.ftl index 9b2f2c3386078..d3bd3244a52aa 100644 --- a/compiler/rustc_driver_impl/messages.ftl +++ b/compiler/rustc_driver_impl/messages.ftl @@ -19,5 +19,3 @@ driver_impl_rlink_rustc_version_mismatch = .rlink file was produced by rustc ver driver_impl_rlink_unable_to_read = failed to read rlink file: `{$err}` driver_impl_rlink_wrong_file_type = The input does not look like a .rlink file - -driver_impl_unpretty_dump_fail = pretty-print failed to write `{$path}` due to error `{$err}` diff --git a/compiler/rustc_driver_impl/src/lib.rs b/compiler/rustc_driver_impl/src/lib.rs index 11303e7d09e8d..8722979cb5efa 100644 --- a/compiler/rustc_driver_impl/src/lib.rs +++ b/compiler/rustc_driver_impl/src/lib.rs @@ -37,9 +37,7 @@ use rustc_interface::{interface, Queries}; use rustc_lint::LintStore; use rustc_metadata::locator; use rustc_session::config::{nightly_options, CG_OPTIONS, Z_OPTIONS}; -use rustc_session::config::{ - ErrorOutputType, Input, OutFileName, OutputType, PrintRequest, TrimmedDefPaths, -}; +use rustc_session::config::{ErrorOutputType, Input, OutFileName, OutputType, TrimmedDefPaths}; use rustc_session::cstore::MetadataLoader; use rustc_session::getopts::{self, Matches}; use rustc_session::lint::{Lint, LintId}; @@ -53,6 +51,7 @@ use std::cmp::max; use std::collections::BTreeMap; use std::env; use std::ffi::OsString; +use std::fmt::Write as _; use std::fs; use std::io::{self, IsTerminal, Read, Write}; use std::panic::{self, catch_unwind}; @@ -72,6 +71,11 @@ macro do_not_use_print($($t:tt)*) { ) } +#[allow(unused_macros)] +macro do_not_use_safe_print($($t:tt)*) { + std::compile_error!("Don't use `safe_print` or `safe_println` here, use `println_info` instead") +} + // This import blocks the use of panicking `print` and `println` in all the code // below. Please use `safe_print` and `safe_println` to avoid ICE when // encountering an I/O error during print. @@ -720,10 +724,17 @@ fn print_crate_info( sess: &Session, parse_attrs: bool, ) -> Compilation { - use rustc_session::config::PrintRequest::*; + use rustc_session::config::PrintKind::*; + + // This import prevents the following code from using the printing macros + // used by the rest of the module. Within this function, we only write to + // the output specified by `sess.io.output_file`. + #[allow(unused_imports)] + use {do_not_use_safe_print as safe_print, do_not_use_safe_print as safe_println}; + // NativeStaticLibs and LinkArgs are special - printed during linking // (empty iterator returns true) - if sess.opts.prints.iter().all(|&p| p == NativeStaticLibs || p == LinkArgs) { + if sess.opts.prints.iter().all(|p| p.kind == NativeStaticLibs || p.kind == LinkArgs) { return Compilation::Continue; } @@ -739,17 +750,23 @@ fn print_crate_info( } else { None }; + for req in &sess.opts.prints { - match *req { + let mut crate_info = String::new(); + macro println_info($($arg:tt)*) { + crate_info.write_fmt(format_args!("{}\n", format_args!($($arg)*))).unwrap() + } + + match req.kind { TargetList => { let mut targets = rustc_target::spec::TARGETS.to_vec(); targets.sort_unstable(); - safe_println!("{}", targets.join("\n")); + println_info!("{}", targets.join("\n")); } - Sysroot => safe_println!("{}", sess.sysroot.display()), - TargetLibdir => safe_println!("{}", sess.target_tlib_path.dir.display()), + Sysroot => println_info!("{}", sess.sysroot.display()), + TargetLibdir => println_info!("{}", sess.target_tlib_path.dir.display()), TargetSpec => { - safe_println!("{}", serde_json::to_string_pretty(&sess.target.to_json()).unwrap()); + println_info!("{}", serde_json::to_string_pretty(&sess.target.to_json()).unwrap()); } AllTargetSpecs => { let mut targets = BTreeMap::new(); @@ -758,26 +775,30 @@ fn print_crate_info( let target = Target::expect_builtin(&triple); targets.insert(name, target.to_json()); } - safe_println!("{}", serde_json::to_string_pretty(&targets).unwrap()); + println_info!("{}", serde_json::to_string_pretty(&targets).unwrap()); } - FileNames | CrateName => { + FileNames => { let Some(attrs) = attrs.as_ref() else { // no crate attributes, print out an error and exit return Compilation::Continue; }; let t_outputs = rustc_interface::util::build_output_filenames(attrs, sess); let id = rustc_session::output::find_crate_name(sess, attrs); - if *req == PrintRequest::CrateName { - safe_println!("{id}"); - continue; - } let crate_types = collect_crate_types(sess, attrs); for &style in &crate_types { let fname = rustc_session::output::filename_for_input(sess, style, id, &t_outputs); - safe_println!("{}", fname.as_path().file_name().unwrap().to_string_lossy()); + println_info!("{}", fname.as_path().file_name().unwrap().to_string_lossy()); } } + CrateName => { + let Some(attrs) = attrs.as_ref() else { + // no crate attributes, print out an error and exit + return Compilation::Continue; + }; + let id = rustc_session::output::find_crate_name(sess, attrs); + println_info!("{id}"); + } Cfg => { let mut cfgs = sess .parse_sess @@ -809,13 +830,13 @@ fn print_crate_info( cfgs.sort(); for cfg in cfgs { - safe_println!("{cfg}"); + println_info!("{cfg}"); } } CallingConventions => { let mut calling_conventions = rustc_target::spec::abi::all_names(); calling_conventions.sort_unstable(); - safe_println!("{}", calling_conventions.join("\n")); + println_info!("{}", calling_conventions.join("\n")); } RelocationModels | CodeModels @@ -823,7 +844,7 @@ fn print_crate_info( | TargetCPUs | StackProtectorStrategies | TargetFeatures => { - codegen_backend.print(*req, sess); + codegen_backend.print(req, &mut crate_info, sess); } // Any output here interferes with Cargo's parsing of other printed output NativeStaticLibs => {} @@ -833,7 +854,7 @@ fn print_crate_info( for split in &[Off, Packed, Unpacked] { if sess.target.options.supported_split_debuginfo.contains(split) { - safe_println!("{split}"); + println_info!("{split}"); } } } @@ -841,7 +862,7 @@ fn print_crate_info( use rustc_target::spec::current_apple_deployment_target; if sess.target.is_like_osx { - safe_println!( + println_info!( "deployment_target={}", current_apple_deployment_target(&sess.target) .expect("unknown Apple target OS") @@ -852,6 +873,8 @@ fn print_crate_info( } } } + + req.out.overwrite(&crate_info, sess); } Compilation::Stop } diff --git a/compiler/rustc_driver_impl/src/pretty.rs b/compiler/rustc_driver_impl/src/pretty.rs index 24a5f4030b88d..222c7b5d6a72a 100644 --- a/compiler/rustc_driver_impl/src/pretty.rs +++ b/compiler/rustc_driver_impl/src/pretty.rs @@ -1,6 +1,5 @@ //! The various pretty-printing routines. -use crate::session_diagnostics::UnprettyDumpFail; use rustc_ast as ast; use rustc_ast_pretty::pprust; use rustc_errors::ErrorGuaranteed; @@ -358,17 +357,7 @@ fn get_source(sess: &Session) -> (String, FileName) { } fn write_or_print(out: &str, sess: &Session) { - match &sess.io.output_file { - None | Some(OutFileName::Stdout) => print!("{out}"), - Some(OutFileName::Real(p)) => { - if let Err(e) = std::fs::write(p, out) { - sess.emit_fatal(UnprettyDumpFail { - path: p.display().to_string(), - err: e.to_string(), - }); - } - } - } + sess.io.output_file.as_ref().unwrap_or(&OutFileName::Stdout).overwrite(out, sess); } pub fn print_after_parsing(sess: &Session, krate: &ast::Crate, ppm: PpMode) { diff --git a/compiler/rustc_driver_impl/src/session_diagnostics.rs b/compiler/rustc_driver_impl/src/session_diagnostics.rs index f7f06b7d0f2b8..5eb587c54d990 100644 --- a/compiler/rustc_driver_impl/src/session_diagnostics.rs +++ b/compiler/rustc_driver_impl/src/session_diagnostics.rs @@ -32,13 +32,6 @@ pub(crate) struct RLinkRustcVersionMismatch<'a> { #[diag(driver_impl_rlink_no_a_file)] pub(crate) struct RlinkNotAFile; -#[derive(Diagnostic)] -#[diag(driver_impl_unpretty_dump_fail)] -pub(crate) struct UnprettyDumpFail { - pub path: String, - pub err: String, -} - #[derive(Diagnostic)] #[diag(driver_impl_ice)] pub(crate) struct Ice; diff --git a/compiler/rustc_interface/src/tests.rs b/compiler/rustc_interface/src/tests.rs index 5c6c3491b388c..12124f14a821f 100644 --- a/compiler/rustc_interface/src/tests.rs +++ b/compiler/rustc_interface/src/tests.rs @@ -715,6 +715,7 @@ fn test_unstable_options_tracking_hash() { untracked!(perf_stats, true); // `pre_link_arg` is omitted because it just forwards to `pre_link_args`. untracked!(pre_link_args, vec![String::from("abc"), String::from("def")]); + untracked!(print_codegen_stats, true); untracked!(print_llvm_passes, true); untracked!(print_mono_items, Some(String::from("abc"))); untracked!(print_type_sizes, true); diff --git a/compiler/rustc_llvm/llvm-wrapper/PassWrapper.cpp b/compiler/rustc_llvm/llvm-wrapper/PassWrapper.cpp index eb3d67e720f2d..e5fb6b0953f5a 100644 --- a/compiler/rustc_llvm/llvm-wrapper/PassWrapper.cpp +++ b/compiler/rustc_llvm/llvm-wrapper/PassWrapper.cpp @@ -1,5 +1,6 @@ #include +#include #include #include @@ -306,44 +307,55 @@ static size_t getLongestEntryLength(ArrayRef Table) { return MaxLen; } -extern "C" void LLVMRustPrintTargetCPUs(LLVMTargetMachineRef TM, const char* TargetCPU) { +using PrintBackendInfo = void(void*, const char* Data, size_t Len); + +extern "C" void LLVMRustPrintTargetCPUs(LLVMTargetMachineRef TM, + const char* TargetCPU, + PrintBackendInfo Print, + void* Out) { const TargetMachine *Target = unwrap(TM); const MCSubtargetInfo *MCInfo = Target->getMCSubtargetInfo(); const Triple::ArchType HostArch = Triple(sys::getDefaultTargetTriple()).getArch(); const Triple::ArchType TargetArch = Target->getTargetTriple().getArch(); + std::ostringstream Buf; + #if LLVM_VERSION_GE(17, 0) const ArrayRef CPUTable = MCInfo->getAllProcessorDescriptions(); #elif defined(LLVM_RUSTLLVM) const ArrayRef CPUTable = MCInfo->getCPUTable(); #else - printf("Full target CPU help is not supported by this LLVM version.\n\n"); + Buf << "Full target CPU help is not supported by this LLVM version.\n\n"; SubtargetSubTypeKV TargetCPUKV = { TargetCPU, {{}}, {{}} }; const ArrayRef CPUTable = TargetCPUKV; #endif unsigned MaxCPULen = getLongestEntryLength(CPUTable); - printf("Available CPUs for this target:\n"); + Buf << "Available CPUs for this target:\n"; // Don't print the "native" entry when the user specifies --target with a // different arch since that could be wrong or misleading. if (HostArch == TargetArch) { MaxCPULen = std::max(MaxCPULen, (unsigned) std::strlen("native")); const StringRef HostCPU = sys::getHostCPUName(); - printf(" %-*s - Select the CPU of the current host (currently %.*s).\n", - MaxCPULen, "native", (int)HostCPU.size(), HostCPU.data()); + Buf << " " << std::left << std::setw(MaxCPULen) << "native" + << " - Select the CPU of the current host " + "(currently " << HostCPU.str() << ").\n"; } for (auto &CPU : CPUTable) { // Compare cpu against current target to label the default if (strcmp(CPU.Key, TargetCPU) == 0) { - printf(" %-*s - This is the default target CPU" - " for the current build target (currently %s).", - MaxCPULen, CPU.Key, Target->getTargetTriple().str().c_str()); + Buf << " " << std::left << std::setw(MaxCPULen) << CPU.Key + << " - This is the default target CPU for the current build target " + "(currently " << Target->getTargetTriple().str() << ")."; } else { - printf(" %-*s", MaxCPULen, CPU.Key); + Buf << " " << CPU.Key; } - printf("\n"); + Buf << "\n"; } + + const auto &BufString = Buf.str(); + Print(Out, BufString.data(), BufString.size()); } extern "C" size_t LLVMRustGetTargetFeaturesCount(LLVMTargetMachineRef TM) { diff --git a/compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp b/compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp index ab5fa961b95a4..870a2e02cbaa5 100644 --- a/compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp +++ b/compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp @@ -1,4 +1,5 @@ #include "LLVMWrapper.h" +#include "llvm/ADT/Statistic.h" #include "llvm/IR/DebugInfoMetadata.h" #include "llvm/IR/DiagnosticHandler.h" #include "llvm/IR/DiagnosticInfo.h" @@ -111,9 +112,26 @@ extern "C" void LLVMRustSetNormalizedTarget(LLVMModuleRef M, unwrap(M)->setTargetTriple(Triple::normalize(Triple)); } -extern "C" void LLVMRustPrintPassTimings() { - raw_fd_ostream OS(2, false); // stderr. - TimerGroup::printAll(OS); +extern "C" const char *LLVMRustPrintPassTimings(size_t *Len) { + std::string buf; + raw_string_ostream SS(buf); + TimerGroup::printAll(SS); + SS.flush(); + *Len = buf.length(); + char *CStr = (char *)malloc(*Len); + memcpy(CStr, buf.c_str(), *Len); + return CStr; +} + +extern "C" const char *LLVMRustPrintStatistics(size_t *Len) { + std::string buf; + raw_string_ostream SS(buf); + llvm::PrintStatistics(SS); + SS.flush(); + *Len = buf.length(); + char *CStr = (char *)malloc(*Len); + memcpy(CStr, buf.c_str(), *Len); + return CStr; } extern "C" LLVMValueRef LLVMRustGetNamedValue(LLVMModuleRef M, const char *Name, diff --git a/compiler/rustc_session/messages.ftl b/compiler/rustc_session/messages.ftl index 4897bd8d5daec..ee24c6d902fa5 100644 --- a/compiler/rustc_session/messages.ftl +++ b/compiler/rustc_session/messages.ftl @@ -26,6 +26,8 @@ session_feature_gate_error = {$explain} session_file_is_not_writeable = output file {$file} is not writeable -- check its permissions +session_file_write_fail = failed to write `{$path}` due to error `{$err}` + session_hexadecimal_float_literal_not_supported = hexadecimal float literal is not supported session_incompatible_linker_flavor = linker flavor `{$flavor}` is incompatible with the current target diff --git a/compiler/rustc_session/src/config.rs b/compiler/rustc_session/src/config.rs index 593983f117d03..a8147ede970c3 100644 --- a/compiler/rustc_session/src/config.rs +++ b/compiler/rustc_session/src/config.rs @@ -3,6 +3,7 @@ pub use crate::options::*; +use crate::errors::FileWriteFail; use crate::search_paths::SearchPath; use crate::utils::{CanonicalizedPath, NativeLib, NativeLibKind}; use crate::{lint, HashStableContext}; @@ -31,6 +32,7 @@ use std::collections::btree_map::{ use std::collections::{BTreeMap, BTreeSet}; use std::ffi::OsStr; use std::fmt; +use std::fs; use std::hash::Hash; use std::iter; use std::path::{Path, PathBuf}; @@ -710,8 +712,14 @@ impl ExternEntry { } } +#[derive(Clone, PartialEq, Debug)] +pub struct PrintRequest { + pub kind: PrintKind, + pub out: OutFileName, +} + #[derive(Copy, Clone, PartialEq, Eq, Debug)] -pub enum PrintRequest { +pub enum PrintKind { FileNames, Sysroot, TargetLibdir, @@ -855,6 +863,17 @@ impl OutFileName { OutFileName::Stdout => outputs.temp_path(flavor, codegen_unit_name), } } + + pub fn overwrite(&self, content: &str, sess: &Session) { + match self { + OutFileName::Stdout => print!("{content}"), + OutFileName::Real(path) => { + if let Err(e) = fs::write(path, content) { + sess.emit_fatal(FileWriteFail { path, err: e.to_string() }); + } + } + } + } } #[derive(Clone, Hash, Debug, HashStable_Generic)] @@ -2005,13 +2024,7 @@ fn parse_output_types( if !unstable_opts.parse_only { for list in matches.opt_strs("emit") { for output_type in list.split(',') { - let (shorthand, path) = match output_type.split_once('=') { - None => (output_type, None), - Some((shorthand, "-")) => (shorthand, Some(OutFileName::Stdout)), - Some((shorthand, path)) => { - (shorthand, Some(OutFileName::Real(PathBuf::from(path)))) - } - }; + let (shorthand, path) = split_out_file_name(output_type); let output_type = OutputType::from_shorthand(shorthand).unwrap_or_else(|| { handler.early_error(format!( "unknown emission type: `{shorthand}` - expected one of: {display}", @@ -2028,6 +2041,14 @@ fn parse_output_types( OutputTypes(output_types) } +fn split_out_file_name(arg: &str) -> (&str, Option) { + match arg.split_once('=') { + None => (arg, None), + Some((kind, "-")) => (kind, Some(OutFileName::Stdout)), + Some((kind, path)) => (kind, Some(OutFileName::Real(PathBuf::from(path)))), + } +} + fn should_override_cgus_and_disable_thinlto( handler: &EarlyErrorHandler, output_types: &OutputTypes, @@ -2091,41 +2112,49 @@ fn collect_print_requests( ) -> Vec { let mut prints = Vec::::new(); if cg.target_cpu.as_ref().is_some_and(|s| s == "help") { - prints.push(PrintRequest::TargetCPUs); + prints.push(PrintRequest { kind: PrintKind::TargetCPUs, out: OutFileName::Stdout }); cg.target_cpu = None; }; if cg.target_feature == "help" { - prints.push(PrintRequest::TargetFeatures); + prints.push(PrintRequest { kind: PrintKind::TargetFeatures, out: OutFileName::Stdout }); cg.target_feature = String::new(); } - const PRINT_REQUESTS: &[(&str, PrintRequest)] = &[ - ("crate-name", PrintRequest::CrateName), - ("file-names", PrintRequest::FileNames), - ("sysroot", PrintRequest::Sysroot), - ("target-libdir", PrintRequest::TargetLibdir), - ("cfg", PrintRequest::Cfg), - ("calling-conventions", PrintRequest::CallingConventions), - ("target-list", PrintRequest::TargetList), - ("target-cpus", PrintRequest::TargetCPUs), - ("target-features", PrintRequest::TargetFeatures), - ("relocation-models", PrintRequest::RelocationModels), - ("code-models", PrintRequest::CodeModels), - ("tls-models", PrintRequest::TlsModels), - ("native-static-libs", PrintRequest::NativeStaticLibs), - ("stack-protector-strategies", PrintRequest::StackProtectorStrategies), - ("target-spec-json", PrintRequest::TargetSpec), - ("all-target-specs-json", PrintRequest::AllTargetSpecs), - ("link-args", PrintRequest::LinkArgs), - ("split-debuginfo", PrintRequest::SplitDebuginfo), - ("deployment-target", PrintRequest::DeploymentTarget), + const PRINT_KINDS: &[(&str, PrintKind)] = &[ + ("crate-name", PrintKind::CrateName), + ("file-names", PrintKind::FileNames), + ("sysroot", PrintKind::Sysroot), + ("target-libdir", PrintKind::TargetLibdir), + ("cfg", PrintKind::Cfg), + ("calling-conventions", PrintKind::CallingConventions), + ("target-list", PrintKind::TargetList), + ("target-cpus", PrintKind::TargetCPUs), + ("target-features", PrintKind::TargetFeatures), + ("relocation-models", PrintKind::RelocationModels), + ("code-models", PrintKind::CodeModels), + ("tls-models", PrintKind::TlsModels), + ("native-static-libs", PrintKind::NativeStaticLibs), + ("stack-protector-strategies", PrintKind::StackProtectorStrategies), + ("target-spec-json", PrintKind::TargetSpec), + ("all-target-specs-json", PrintKind::AllTargetSpecs), + ("link-args", PrintKind::LinkArgs), + ("split-debuginfo", PrintKind::SplitDebuginfo), + ("deployment-target", PrintKind::DeploymentTarget), ]; + // We disallow reusing the same path in multiple prints, such as `--print + // cfg=output.txt --print link-args=output.txt`, because outputs are printed + // by disparate pieces of the compiler, and keeping track of which files + // need to be overwritten vs appended to is annoying. + let mut printed_paths = FxHashSet::default(); + prints.extend(matches.opt_strs("print").into_iter().map(|req| { - match PRINT_REQUESTS.iter().find(|&&(name, _)| name == req) { - Some((_, PrintRequest::TargetSpec)) => { + let (req, out) = split_out_file_name(&req); + + let kind = match PRINT_KINDS.iter().find(|&&(name, _)| name == req) { + Some((_, PrintKind::TargetSpec)) => { if unstable_opts.unstable_options { - PrintRequest::TargetSpec + PrintKind::TargetSpec } else { handler.early_error( "the `-Z unstable-options` flag must also be passed to \ @@ -2133,9 +2162,9 @@ fn collect_print_requests( ); } } - Some((_, PrintRequest::AllTargetSpecs)) => { + Some((_, PrintKind::AllTargetSpecs)) => { if unstable_opts.unstable_options { - PrintRequest::AllTargetSpecs + PrintKind::AllTargetSpecs } else { handler.early_error( "the `-Z unstable-options` flag must also be passed to \ @@ -2143,16 +2172,28 @@ fn collect_print_requests( ); } } - Some(&(_, print_request)) => print_request, + Some(&(_, print_kind)) => print_kind, None => { let prints = - PRINT_REQUESTS.iter().map(|(name, _)| format!("`{name}`")).collect::>(); + PRINT_KINDS.iter().map(|(name, _)| format!("`{name}`")).collect::>(); let prints = prints.join(", "); handler.early_error(format!( "unknown print request `{req}`. Valid print requests are: {prints}" )); } + }; + + let out = out.unwrap_or(OutFileName::Stdout); + if let OutFileName::Real(path) = &out { + if !printed_paths.insert(path.clone()) { + handler.early_error(format!( + "cannot print multiple outputs to the same path: {}", + path.display(), + )); + } } + + PrintRequest { kind, out } })); prints diff --git a/compiler/rustc_session/src/errors.rs b/compiler/rustc_session/src/errors.rs index 4a3e668da111a..dd15ad45145f1 100644 --- a/compiler/rustc_session/src/errors.rs +++ b/compiler/rustc_session/src/errors.rs @@ -163,6 +163,13 @@ pub struct FileIsNotWriteable<'a> { pub file: &'a std::path::Path, } +#[derive(Diagnostic)] +#[diag(session_file_write_fail)] +pub(crate) struct FileWriteFail<'a> { + pub path: &'a std::path::Path, + pub err: String, +} + #[derive(Diagnostic)] #[diag(session_crate_name_does_not_match)] pub struct CrateNameDoesNotMatch { diff --git a/compiler/rustc_session/src/options.rs b/compiler/rustc_session/src/options.rs index 87d67c099cedb..39efe9abeecdb 100644 --- a/compiler/rustc_session/src/options.rs +++ b/compiler/rustc_session/src/options.rs @@ -1668,6 +1668,9 @@ options! { "use a more precise version of drop elaboration for matches on enums (default: yes). \ This results in better codegen, but has caused miscompilations on some tier 2 platforms. \ See #77382 and #74551."), + #[rustc_lint_opt_deny_field_access("use `Session::print_codegen_stats` instead of this field")] + print_codegen_stats: bool = (false, parse_bool, [UNTRACKED], + "print codegen statistics (default: no)"), print_fuel: Option = (None, parse_opt_string, [TRACKED], "make rustc print the total optimization fuel used by a crate"), print_llvm_passes: bool = (false, parse_bool, [UNTRACKED], diff --git a/compiler/rustc_session/src/session.rs b/compiler/rustc_session/src/session.rs index 146bb11bd3a7b..8844fd28cbbae 100644 --- a/compiler/rustc_session/src/session.rs +++ b/compiler/rustc_session/src/session.rs @@ -1057,6 +1057,10 @@ impl Session { self.opts.unstable_opts.verbose } + pub fn print_llvm_stats(&self) -> bool { + self.opts.unstable_opts.print_codegen_stats + } + pub fn verify_llvm_ir(&self) -> bool { self.opts.unstable_opts.verify_llvm_ir || option_env!("RUSTC_VERIFY_LLVM_IR").is_some() } diff --git a/library/alloc/src/rc.rs b/library/alloc/src/rc.rs index b3ec830a7d7c6..bf01b2082eda2 100644 --- a/library/alloc/src/rc.rs +++ b/library/alloc/src/rc.rs @@ -661,10 +661,14 @@ impl Rc { impl Rc { /// Returns a reference to the underlying allocator. + /// + /// Note: this is an associated function, which means that you have + /// to call it as `Rc::allocator(&r)` instead of `r.allocator()`. This + /// is so that there is no conflict with a method on the inner type. #[inline] #[unstable(feature = "allocator_api", issue = "32838")] - pub fn allocator(&self) -> &A { - &self.alloc + pub fn allocator(this: &Self) -> &A { + &this.alloc } /// Constructs a new `Rc` in the provided allocator. /// diff --git a/library/alloc/src/sync.rs b/library/alloc/src/sync.rs index e00850eb5d8a1..c2202f2fce553 100644 --- a/library/alloc/src/sync.rs +++ b/library/alloc/src/sync.rs @@ -678,10 +678,14 @@ impl Arc { impl Arc { /// Returns a reference to the underlying allocator. + /// + /// Note: this is an associated function, which means that you have + /// to call it as `Arc::allocator(&a)` instead of `a.allocator()`. This + /// is so that there is no conflict with a method on the inner type. #[inline] #[unstable(feature = "allocator_api", issue = "32838")] - pub fn allocator(&self) -> &A { - &self.alloc + pub fn allocator(this: &Self) -> &A { + &this.alloc } /// Constructs a new `Arc` in the provided allocator. /// diff --git a/library/std/src/sys/windows/thread_local_dtor.rs b/library/std/src/sys/windows/thread_local_dtor.rs index 9707a95dff21b..cf542d2bfb838 100644 --- a/library/std/src/sys/windows/thread_local_dtor.rs +++ b/library/std/src/sys/windows/thread_local_dtor.rs @@ -4,29 +4,4 @@ #![unstable(feature = "thread_local_internals", issue = "none")] #![cfg(target_thread_local)] -// Using a per-thread list avoids the problems in synchronizing global state. -#[thread_local] -static mut DESTRUCTORS: Vec<(*mut u8, unsafe extern "C" fn(*mut u8))> = Vec::new(); - -// Ensure this can never be inlined because otherwise this may break in dylibs. -// See #44391. -#[inline(never)] -pub unsafe fn register_dtor(t: *mut u8, dtor: unsafe extern "C" fn(*mut u8)) { - DESTRUCTORS.push((t, dtor)); -} - -#[inline(never)] // See comment above -/// Runs destructors. This should not be called until thread exit. -pub unsafe fn run_keyless_dtors() { - // Drop all the destructors. - // - // Note: While this is potentially an infinite loop, it *should* be - // the case that this loop always terminates because we provide the - // guarantee that a TLS key cannot be set after it is flagged for - // destruction. - while let Some((ptr, dtor)) = DESTRUCTORS.pop() { - (dtor)(ptr); - } - // We're done so free the memory. - DESTRUCTORS = Vec::new(); -} +pub use super::thread_local_key::register_keyless_dtor as register_dtor; diff --git a/library/std/src/sys/windows/thread_local_key.rs b/library/std/src/sys/windows/thread_local_key.rs index 17628b7579b8d..036d96596e9c0 100644 --- a/library/std/src/sys/windows/thread_local_key.rs +++ b/library/std/src/sys/windows/thread_local_key.rs @@ -1,7 +1,7 @@ use crate::cell::UnsafeCell; use crate::ptr; use crate::sync::atomic::{ - AtomicPtr, AtomicU32, + AtomicBool, AtomicPtr, AtomicU32, Ordering::{AcqRel, Acquire, Relaxed, Release}, }; use crate::sys::c; @@ -9,6 +9,41 @@ use crate::sys::c; #[cfg(test)] mod tests; +/// An optimization hint. The compiler is often smart enough to know if an atomic +/// is never set and can remove dead code based on that fact. +static HAS_DTORS: AtomicBool = AtomicBool::new(false); + +// Using a per-thread list avoids the problems in synchronizing global state. +#[thread_local] +#[cfg(target_thread_local)] +static mut DESTRUCTORS: Vec<(*mut u8, unsafe extern "C" fn(*mut u8))> = Vec::new(); + +// Ensure this can never be inlined because otherwise this may break in dylibs. +// See #44391. +#[inline(never)] +#[cfg(target_thread_local)] +pub unsafe fn register_keyless_dtor(t: *mut u8, dtor: unsafe extern "C" fn(*mut u8)) { + DESTRUCTORS.push((t, dtor)); + HAS_DTORS.store(true, Relaxed); +} + +#[inline(never)] // See comment above +#[cfg(target_thread_local)] +/// Runs destructors. This should not be called until thread exit. +unsafe fn run_keyless_dtors() { + // Drop all the destructors. + // + // Note: While this is potentially an infinite loop, it *should* be + // the case that this loop always terminates because we provide the + // guarantee that a TLS key cannot be set after it is flagged for + // destruction. + while let Some((ptr, dtor)) = DESTRUCTORS.pop() { + (dtor)(ptr); + } + // We're done so free the memory. + DESTRUCTORS = Vec::new(); +} + type Key = c::DWORD; type Dtor = unsafe extern "C" fn(*mut u8); @@ -156,6 +191,8 @@ static DTORS: AtomicPtr = AtomicPtr::new(ptr::null_mut()); /// Should only be called once per key, otherwise loops or breaks may occur in /// the linked list. unsafe fn register_dtor(key: &'static StaticKey) { + // Ensure this is never run when native thread locals are available. + assert_eq!(false, cfg!(target_thread_local)); let this = <*const StaticKey>::cast_mut(key); // Use acquire ordering to pass along the changes done by the previously // registered keys when we store the new head with release ordering. @@ -167,6 +204,7 @@ unsafe fn register_dtor(key: &'static StaticKey) { Err(new) => head = new, } } + HAS_DTORS.store(true, Release); } // ------------------------------------------------------------------------- @@ -240,10 +278,14 @@ pub static p_thread_callback: unsafe extern "system" fn(c::LPVOID, c::DWORD, c:: #[allow(dead_code, unused_variables)] unsafe extern "system" fn on_tls_callback(h: c::LPVOID, dwReason: c::DWORD, pv: c::LPVOID) { + if !HAS_DTORS.load(Acquire) { + return; + } if dwReason == c::DLL_THREAD_DETACH || dwReason == c::DLL_PROCESS_DETACH { + #[cfg(not(target_thread_local))] run_dtors(); #[cfg(target_thread_local)] - super::thread_local_dtor::run_keyless_dtors(); + run_keyless_dtors(); } // See comments above for what this is doing. Note that we don't need this diff --git a/library/std/src/sys/windows/thread_local_key/tests.rs b/library/std/src/sys/windows/thread_local_key/tests.rs index c95f383fb90e3..c739f0caf3ec0 100644 --- a/library/std/src/sys/windows/thread_local_key/tests.rs +++ b/library/std/src/sys/windows/thread_local_key/tests.rs @@ -1,3 +1,7 @@ +// This file only tests the thread local key fallback. +// Windows targets with native thread local support do not use this. +#![cfg(not(target_thread_local))] + use super::StaticKey; use crate::ptr; diff --git a/src/doc/rustc/src/command-line-arguments.md b/src/doc/rustc/src/command-line-arguments.md index 2c7c05c0c4b88..4d32897cc14c7 100644 --- a/src/doc/rustc/src/command-line-arguments.md +++ b/src/doc/rustc/src/command-line-arguments.md @@ -260,6 +260,10 @@ The valid types of print values are: This returns rustc's minimum supported deployment target if no `*_DEPLOYMENT_TARGET` variable is present in the environment, or otherwise returns the variable's parsed value. +A filepath may optionally be specified for each requested information kind, in +the format `--print KIND=PATH`, just like for `--emit`. When a path is +specified, information will be written there instead of to stdout. + [conditional compilation]: ../reference/conditional-compilation.html [deployment target]: https://developer.apple.com/library/archive/documentation/DeveloperTools/Conceptual/cross_development/Configuring/configuring.html diff --git a/src/doc/style-guide/src/README.md b/src/doc/style-guide/src/README.md index b8aa64ba14f6f..29204603e58d3 100644 --- a/src/doc/style-guide/src/README.md +++ b/src/doc/style-guide/src/README.md @@ -40,7 +40,6 @@ options. * Each level of indentation must be four spaces (that is, all indentation outside of string literals and comments must be a multiple of four). * The maximum width for a line is 100 characters. -* A tool may choose to make some of these configurable. #### Block indent @@ -63,7 +62,8 @@ example) and less rightward drift. ### Trailing commas -Lists should have a trailing comma when followed by a newline: +In comma-separated lists of any kind, use a trailing comma when followed by a +newline: ```rust function_call( @@ -99,8 +99,6 @@ fn bar() {} fn baz() {} ``` -Formatting tools may wish to make the bounds on blank lines configurable. - ### [Module-level items](items.md) ### [Statements](statements.md) ### [Expressions](expressions.md) @@ -114,17 +112,17 @@ formatter might skip formatting of comments. Prefer line comments (`//`) to block comments (`/* ... */`). -When using line comments there should be a single space after the opening sigil. +When using line comments, put a single space after the opening sigil. -When using single-line block comments there should be a single space after the -opening sigil and before the closing sigil. Multi-line block comments should -have a newline after the opening sigil and before the closing sigil. +When using single-line block comments, put a single space after the opening +sigil and before the closing sigil. For multi-line block comments, put a +newline after the opening sigil, and a newline before the closing sigil. -Prefer to put a comment on its own line. Where a comment follows code, there -should be a single space before it. Where a block comment is inline, there -should be surrounding whitespace as if it were an identifier or keyword. There -should be no trailing whitespace after a comment or at the end of any line in a -multi-line comment. Examples: +Prefer to put a comment on its own line. Where a comment follows code, put a +single space before it. Where a block comment appears inline, use surrounding +whitespace as if it were an identifier or keyword. Do not include trailing +whitespace after a comment or at the end of any line in a multi-line comment. +Examples: ```rust // A comment on an item. @@ -173,7 +171,7 @@ Prefer line comments (`///`) to block comments (`/** ... */`). Prefer outer doc comments (`///` or `/** ... */`), only use inner doc comments (`//!` and `/*! ... */`) to write module-level or crate-level documentation. -Doc comments should come before attributes. +Put doc comments before attributes. ### Attributes @@ -198,18 +196,20 @@ struct CRepr { } ``` -For attributes with an equal sign, there should be a single space before and -after the `=`, e.g., `#[foo = 42]`. +For attributes with an equal sign, put a single space before and after the `=`, +e.g., `#[foo = 42]`. There must only be a single `derive` attribute. Note for tool authors: if combining multiple `derive` attributes into a single attribute, the ordering of -the derived names should be preserved. E.g., `#[derive(bar)] #[derive(foo)] -struct Baz;` should be formatted to `#[derive(bar, foo)] struct Baz;`. +the derived names must generally be preserved for correctness: +`#[derive(Foo)] #[derive(Bar)] struct Baz;` must be formatted to +`#[derive(Foo, Bar)] struct Baz;`. ### *small* items -In many places in this guide we specify that a formatter may format an item -differently if it is *small*, for example struct literals: +In many places in this guide we specify formatting that depends on a code +construct being *small*. For example, single-line vs multi-line struct +literals: ```rust // Normal formatting @@ -218,7 +218,7 @@ Foo { f2: another_expression(), } -// *small* formatting +// "small" formatting Foo { f1, f2 } ``` @@ -231,10 +231,6 @@ complexity of an item (for example, that all components must be simple names, not more complex sub-expressions). For more discussion on suitable heuristics, see [this issue](https://github.com/rust-lang-nursery/fmt-rfcs/issues/47). -Tools should give the user an option to ignore such heuristics and always use -the normal formatting. - - ## [Non-formatting conventions](advice.md) ## [Cargo.toml conventions](cargo.md) diff --git a/src/doc/style-guide/src/expressions.md b/src/doc/style-guide/src/expressions.md index 41a9eb82cb5af..0aed8763a0628 100644 --- a/src/doc/style-guide/src/expressions.md +++ b/src/doc/style-guide/src/expressions.md @@ -2,10 +2,13 @@ ### Blocks -A block expression should have a newline after the initial `{` and before the -terminal `}`. Any qualifier before the block (e.g., `unsafe`) should always be -on the same line as the opening brace, and separated with a single space. The -contents of the block should be block indented: +A block expression must have a newline after the initial `{` and before the +terminal `}`, unless it qualifies to be written as a single line based on +another style rule. + +A keyword before the block (such as `unsafe` or `async`) must be on the same +line as the opening brace, with a single space between the keyword and the +opening brace. Indent the contents of the block. ```rust fn block_as_stmt() { @@ -40,7 +43,7 @@ fn unsafe_block_as_stmt() { } ``` -If a block has an attribute, it should be on its own line: +If a block has an attribute, put it on its own line before the block: ```rust fn block_as_stmt() { @@ -54,18 +57,18 @@ fn block_as_stmt() { } ``` -Avoid writing comments on the same line as the braces. +Avoid writing comments on the same lines as either of the braces. -An empty block should be written as `{}`. +Write an empty block as `{}`. -A block may be written on a single line if: +Write a block on a single line if: * it is either used in expression position (not statement position) or is an unsafe block in statement position -* contains a single-line expression and no statements -* contains no comments +* it contains a single-line expression and no statements +* it contains no comments -A single line block should have spaces after the opening brace and before the +For a single-line block, put spaces after the opening brace and before the closing brace. Examples: @@ -117,9 +120,9 @@ fn main() { ### Closures Don't put any extra spaces before the first `|` (unless the closure is prefixed -by `move`); put a space between the second `|` and the expression of the -closure. Between the `|`s, you should use function definition syntax, however, -elide types where possible. +by a keyword such as `move`); put a space between the second `|` and the +expression of the closure. Between the `|`s, use function definition syntax, +but elide types where possible. Use closures without the enclosing `{}`, if possible. Add the `{}` when you have a return type, when there are statements, there are comments in the body, or the @@ -155,13 +158,14 @@ move |arg1: i32, arg2: i32| -> i32 { ### Struct literals -If a struct literal is *small* it may be formatted on a single line. If not, -each field should be on it's own, block-indented line. There should be a -trailing comma in the multi-line form only. There should be a space after the -colon only. +If a struct literal is *small*, format it on a single line, and do not use a +trailing comma. If not, split it across multiple lines, with each field on its +own block-indented line, and use a trailing comma. + +For each `field: value` entry, put a space after the colon only. -There should be a space before the opening brace. In the single-line form there -should be spaces after the opening brace and before the closing brace. +Put a space before the opening brace. In the single-line form, put spaces after +the opening brace and before the closing brace. ```rust Foo { field1, field2: 0 } @@ -172,7 +176,7 @@ let f = Foo { ``` Functional record update syntax is treated like a field, but it must never have -a trailing comma. There should be no space after `..`. +a trailing comma. Do not put a space after `..`. let f = Foo { field1, @@ -182,9 +186,13 @@ let f = Foo { ### Tuple literals -Use a single-line form where possible. There should not be spaces around the -parentheses. Where a single-line form is not possible, each element of the tuple -should be on its own block-indented line and there should be a trailing comma. +Use a single-line form where possible. Do not put spaces between the opening +parenthesis and the first element, or between the last element and the closing +parenthesis. Separate elements with a comma followed by a space. + +Where a single-line form is not possible, write the tuple across +multiple lines, with each element of the tuple on its own block-indented line, +and use a trailing comma. ```rust (a, b, c) @@ -198,14 +206,23 @@ let x = ( ### Tuple struct literals -There should be no space between the identifier and the opening parenthesis. -Otherwise, follow the rules for tuple literals, e.g., `Foo(a, b)`. +Do not put space between the identifier and the opening parenthesis. Otherwise, +follow the rules for tuple literals: + +```rust +Foo(a, b, c) + +let x = Foo( + a_long_expr, + another_very_long_expr, +); +``` ### Enum literals Follow the formatting rules for the various struct literals. Prefer using the -name of the enum as a qualifying name, unless the enum is in the prelude. E.g., +name of the enum as a qualifying name, unless the enum is in the prelude: ```rust Foo::Bar(a, b) @@ -219,24 +236,29 @@ Ok(an_expr) ### Array literals -For simple array literals, avoid line breaking, no spaces around square -brackets, contents of the array should be separated by commas and spaces. If -using the repeating initialiser, there should be a space after the semicolon -only. Apply the same rules if using the `vec!` or similar macros (always use -square brackets here). Examples: +Write small array literals on a single line. Do not put spaces between the opening +square bracket and the first element, or between the last element and the closing +square bracket. Separate elements with a comma followed by a space. + +If using the repeating initializer, put a space after the semicolon +only. + +Apply the same rules if using `vec!` or similar array-like macros; always use +square brackets with such macros. Examples: ```rust fn main() { - [1, 2, 3]; - vec![a, b, c, d]; + let x = [1, 2, 3]; + let y = vec![a, b, c, d]; let a = [42; 10]; } ``` -If a line must be broken, prefer breaking only after the `;`, if possible. -Otherwise, follow the rules below for function calls. In any case, the contents -of the initialiser should be block indented and there should be line breaks -after the opening bracket and before the closing bracket: +For arrays that have to be broken across lines, if using the repeating +initializer, break after the `;`, not before. Otherwise, follow the rules below +for function calls. In any case, block-indent the contents of the initializer, +and put line breaks after the opening square bracket and before the closing +square bracket: ```rust fn main() { @@ -255,11 +277,12 @@ fn main() { ### Array accesses, indexing, and slicing. -No spaces around the square brackets, avoid breaking lines if possible, never -break a line between the target expression and the opening bracket. If the -indexing expression covers multiple lines, then it should be block indented and -there should be newlines after the opening brackets and before the closing -bracket. However, this should be avoided where possible. +Don't put spaces around the square brackets. Avoid breaking lines if possible. +Never break a line between the target expression and the opening square +bracket. If the indexing expression must be broken onto a subsequent line, or +spans multiple lines itself, then block-indent the indexing expression, and put +newlines after the opening square bracket and before the closing square +bracket: Examples: @@ -291,7 +314,7 @@ if you have `t: &T`, and `u: U`, prefer `*t op u` to `t op &u`. In general, within expressions, prefer dereferencing to taking references, unless necessary (e.g. to avoid an unnecessarily expensive operation). -Use parentheses liberally, do not necessarily elide them due to precedence. +Use parentheses liberally; do not necessarily elide them due to precedence. Tools should not automatically insert or remove parentheses. Do not use spaces to indicate precedence. @@ -353,10 +376,10 @@ foo(x, y, z) #### Multi-line calls If the function call is not *small*, it would otherwise over-run the max width, -or any argument or the callee is multi-line, then the call should be formatted -across multiple lines. In this case, each argument should be on it's own block- -indented line, there should be a newline after the opening parenthesis and -before the closing parenthesis, and there should be a trailing comma. E.g., +or any argument or the callee is multi-line, then format the call across +multiple lines. In this case, put each argument on its own block-indented line, +break after the opening parenthesis and before the closing parenthesis, +and use a trailing comma: ```rust a_function_call( @@ -379,17 +402,18 @@ x.foo().bar().baz(x, y, z); ### Macro uses -Macros which can be parsed like other constructs should be formatted like those +If a macro can be parsed like other constructs, format it like those constructs. For example, a macro use `foo!(a, b, c)` can be parsed like a -function call (ignoring the `!`), therefore it should be formatted following the -rules for function calls. +function call (ignoring the `!`), so format it using the rules for function +calls. #### Special case macros -Macros which take a format string and where all other arguments are *small* may -be formatted with arguments before and after the format string on a single line -and the format string on its own line, rather than putting each argument on its -own line. For example, +For macros which take a format string, if all other arguments are *small*, +format the arguments before the format string on a single line if they fit, and +format the arguments after the format string on a single line if they fit, with +the format string on its own line. If the arguments are not small or do not +fit, put each on its own line as with a function. For example: ```rust println!( @@ -416,13 +440,13 @@ let cstr = "Hi\0" as *const str as *const [u8] as *const std::os::raw::c_char; ### Chains of fields and method calls -A chain is a sequence of field accesses and/or method calls. A chain may also -include the try operator ('?'). E.g., `a.b.c().d` or `foo?.bar().baz?`. +A chain is a sequence of field accesses, method calls, and/or uses of the try +operator `?`. E.g., `a.b.c().d` or `foo?.bar().baz?`. -Prefer formatting on one line if possible, and the chain is *small*. If -formatting on multiple lines, each field access or method call in the chain -should be on its own line with the line-break before the `.` and after any `?`. -Each line should be block-indented. E.g., +Format the chain on one line if it is "small" and otherwise possible to do so. +If formatting on multiple lines, put each field access or method call in the +chain on its own line, with the line-break before the `.` and after any `?`. +Block-indent each subsequent line: ```rust let foo = bar @@ -431,13 +455,16 @@ let foo = bar ``` If the length of the last line of the first element plus its indentation is -less than or equal to the indentation of the second line (and there is space), -then combine the first and second lines, e.g., +less than or equal to the indentation of the second line, then combine the +first and second lines if they fit. Apply this rule recursively. ```rust x.baz? .qux() +x.y.z + .qux() + let foo = x .baz? .qux(); @@ -489,13 +516,13 @@ self.pre_comment.as_ref().map_or( This section covers `if`, `if let`, `loop`, `while`, `while let`, and `for` expressions. -The keyword, any initial clauses, and the opening brace of the block should be -on a single line. The usual rules for [block formatting](#blocks) should be -applied to the block. +Put the keyword, any initial clauses, and the opening brace of the block all on +a single line, if they fit. Apply the usual rules for [block +formatting](#blocks) to the block. -If there is an `else` component, then the closing brace, `else`, any following -clause, and the opening brace should all be on the same line. There should be a -single space before and after the `else` keyword. For example: +If there is an `else` component, then put the closing brace, `else`, any +following clause, and the opening brace all on the same line, with a single +space before and after the `else` keyword: ```rust if ... { @@ -513,10 +540,10 @@ if let ... { } ``` -If the control line needs to be broken, then prefer to break before the `=` in -`* let` expressions and before `in` in a `for` expression; the following line -should be block indented. If the control line is broken for any reason, then the -opening brace should be on its own line and not indented. Examples: +If the control line needs to be broken, prefer to break before the `=` in `* +let` expressions and before `in` in a `for` expression; block-indent the +following line. If the control line is broken for any reason, put the opening +brace on its own line, not indented. Examples: ```rust while let Some(foo) @@ -539,10 +566,10 @@ if a_long_expression } ``` -Where the initial clause is multi-lined and ends with one or more closing -parentheses, square brackets, or braces, and there is nothing else on that line, -and that line is not indented beyond the indent on the first line of the control -flow expression, then the opening brace of the block should be put on the same +Where the initial clause spans multiple lines and ends with one or more closing +parentheses, square brackets, or braces, and there is nothing else on that +line, and that line is not indented beyond the indent on the first line of the +control flow expression, then put the opening brace of the block on the same line with a preceding space. For example: ```rust @@ -558,9 +585,9 @@ if !self.config.file_lines().intersects( #### Single line `if else` -Formatters may place an `if else` or `if let else` on a single line if it occurs -in expression context (i.e., is not a standalone statement), it contains a -single `else` clause, and is *small*. For example: +Put an `if else` or `if let else` on a single line if it occurs in expression +context (i.e., is not a standalone statement), it contains a single `else` +clause, and is *small*: ```rust let y = if x { 0 } else { 1 }; @@ -582,9 +609,9 @@ if x { ### Match -Prefer not to line-break inside the discriminant expression. There must always -be a line break after the opening brace and before the closing brace. The match -arms must be block indented once: +Prefer not to line-break inside the discriminant expression. Always break after +the opening brace and before the closing brace. Block-indent the match arms +once: ```rust match foo { @@ -598,7 +625,7 @@ let x = match foo.bar.baz() { Use a trailing comma for a match arm if and only if not using a block. -Never start a match arm pattern with `|`, e.g., +Never start a match arm pattern with `|`: ```rust match foo { @@ -608,14 +635,13 @@ match foo { | a_very_long_pattern | another_pattern | yet_another_pattern - | a_forth_pattern => { + | a_fourth_pattern => { ... } } ``` -Prefer - +Prefer: ```rust match foo { @@ -623,7 +649,7 @@ match foo { a_very_long_pattern | another_pattern | yet_another_pattern - | a_forth_pattern => { + | a_fourth_pattern => { ... } } @@ -633,11 +659,11 @@ Avoid splitting the left-hand side (before the `=>`) of a match arm where possible. If the right-hand side of the match arm is kept on the same line, never use a block (unless the block is empty). -If the right-hand side consists of multiple statements or has line comments or -the start of the line cannot be fit on the same line as the left-hand side, use -a block. +If the right-hand side consists of multiple statements, or has line comments, +or the start of the line does not fit on the same line as the left-hand side, +use a block. -The body of a block arm should be block indented once. +Block-indent the body of a block arm. Examples: @@ -662,8 +688,8 @@ match foo { ``` If the body is a single expression with no line comments and not a control flow -expression, then it may be started on the same line as the right-hand side. If -not, then it must be in a block. Example, +expression, start it on the same line as the left-hand side. If not, then it +must be in a block. Example: ```rust match foo { @@ -687,8 +713,8 @@ match foo { #### Line-breaking -Where it is possible to use a block form on the right-hand side and avoid -breaking the left-hand side, do that. E.g. +If using a block form on the right-hand side of a match arm makes it possible +to avoid breaking on the left-hand side, do that: ```rust // Assuming the following line does not fit in the max width @@ -720,7 +746,7 @@ body on a new line: If required to break the pattern, put each clause of the pattern on its own line with no additional indent, breaking before the `|`. If there is an `if` -clause, then you must use the above form: +clause, use the above form: ```rust a_very_long_pattern @@ -740,7 +766,7 @@ clause, then you must use the above form: ``` If the pattern is multi-line, and the last line is less wide than the indent, do -not put the `if` clause on a newline. E.g., +not put the `if` clause on a new line. E.g., ```rust Token::Dimension { @@ -753,8 +779,8 @@ not put the `if` clause on a newline. E.g., ``` If every clause in a pattern is *small*, but the whole pattern does not fit on -one line, then the pattern may be formatted across multiple lines with as many -clauses per line as possible. Again break before a `|`: +one line, then format the pattern across multiple lines with as many clauses +per line as possible. Again, break before a `|`: ```rust foo | bar | baz @@ -783,8 +809,8 @@ E.g., `&&Some(foo)` matches, `Foo(4, Bar)` does not. ### Combinable expressions Where a function call has a single argument, and that argument is formatted -across multiple-lines, the outer call may be formatted as if it were a single- -line call. The same combining behaviour may be applied to any similar +across multiple-lines, format the outer call as if it were a single-line call, +if the result fits. Apply the same combining behaviour to any similar expressions which have multi-line, block-indented lists of sub-expressions delimited by parentheses (e.g., macros or tuple struct literals). E.g., @@ -814,13 +840,12 @@ let arr = [combinable( )]; ``` -Such behaviour should extend recursively, however, tools may choose to limit the -depth of nesting. +Apply this behavior recursively. -Only where the multi-line sub-expression is a closure with an explicit block, -this combining behaviour may be used where there are other arguments, as long as -all the arguments and the first line of the closure fit on the first line, the -closure is the last argument, and there is only one closure argument: +For a function with multiple arguments, if the last argument is a multi-line +closure with an explicit block, there are no other closure arguments, and all +the arguments and the first line of the closure fit on the first line, use the +same combining behavior: ```rust foo(first_arg, x, |param| { @@ -835,16 +860,17 @@ foo(first_arg, x, |param| { Do not put spaces in ranges, e.g., `0..10`, `x..=y`, `..x.len()`, `foo..`. When writing a range with both upper and lower bounds, if the line must be -broken, break before the range operator and block indent the second line: +broken within the range, break before the range operator and block indent the +second line: ```rust a_long_expression ..another_long_expression ``` -For the sake of indicating precedence, we recommend that if either bound is a -compound expression, then use parentheses around it, e.g., `..(x + 1)`, -`(x.f)..(x.f.len())`, or `0..(x - 10)`. +For the sake of indicating precedence, if either bound is a compound +expression, use parentheses around it, e.g., `..(x + 1)`, `(x.f)..(x.f.len())`, +or `0..(x - 10)`. ### Hexadecimal literals @@ -852,11 +878,8 @@ compound expression, then use parentheses around it, e.g., `..(x + 1)`, Hexadecimal literals may use upper- or lower-case letters, but they must not be mixed within the same literal. Projects should use the same case for all literals, but we do not make a recommendation for either lower- or upper-case. -Tools should have an option to convert mixed case literals to upper-case, and -may have an option to convert all literals to either lower- or upper-case. - ## Patterns -Patterns should be formatted like their corresponding expressions. See the -section on `match` for additional formatting for patterns in match arms. +Format patterns like their corresponding expressions. See the section on +`match` for additional formatting for patterns in match arms. diff --git a/src/doc/style-guide/src/items.md b/src/doc/style-guide/src/items.md index de31e8a8799b3..6e5ea335e6ab2 100644 --- a/src/doc/style-guide/src/items.md +++ b/src/doc/style-guide/src/items.md @@ -9,19 +9,17 @@ an item appears at module level or within another item. alphabetically. `use` statements, and module *declarations* (`mod foo;`, not `mod { ... }`) -must come before other items. We recommend that imports come before module -declarations; if imports and modules are separated, then they should be ordered -alphabetically. When sorting, `self` and `super` must come before any other -names. Module declarations should not be moved if they are annotated with -`#[macro_use]`, since that may be semantics changing. - -Tools should make the above ordering optional. +must come before other items. Put imports before module declarations. Sort each +alphabetically, except that `self` and `super` must come before any other +names. +Don't automatically move module declarations annotated with `#[macro_use]`, +since that might change semantics. ### Function definitions In Rust, people often find functions by searching for `fn [function-name]`, so -the formatting of function definitions shold enable this. +the formatting of function definitions must enable this. The proper ordering and spacing is: @@ -83,9 +81,9 @@ enum FooBar { } ``` -If a struct variant is [*small*](index.html#small-items), it may be formatted on -one line. In this case, do not use a trailing comma for the field list, but do -put spaces around each brace: +If a struct variant is [*small*](index.html#small-items), format it on one +line. In this case, do not use a trailing comma for the field list, but do put +spaces around each brace: ```rust enum FooBar { @@ -94,9 +92,9 @@ enum FooBar { ``` In an enum with multiple struct variants, if any struct variant is written on -multiple lines, then the multi-line formatting should be used for all struct -variants. However, such a situation might be an indication that you should -factor out the fields of the variant into their own struct. +multiple lines, use the multi-line formatting for all struct variants. However, +such a situation might be an indication that you should factor out the fields +of the variant into their own struct. ### Structs and Unions @@ -143,9 +141,9 @@ union Foo { ### Tuple structs -Put the whole struct on one line if possible. Types in the parentheses should be -separated by a comma and space with no trailing comma. No spaces around the -parentheses or semicolon: +Put the whole struct on one line if possible. Separate types within the +parentheses using a comma and space. Don't use a trailing comma for a +single-line tuple struct. Don't put spaces around the parentheses or semicolon: ```rust pub struct Foo(String, u8); @@ -154,9 +152,11 @@ pub struct Foo(String, u8); Prefer unit structs to empty tuple structs (these only exist to simplify code generation), e.g., `struct Foo;` rather than `struct Foo();`. -For more than a few fields, prefer a proper struct with named fields. Given -this, a tuple struct should always fit on one line. If it does not, block format -the fields with a field on each line and a trailing comma: +For more than a few fields (in particular if the tuple struct does not fit on +one line), prefer a proper struct with named fields. + +For a multi-line tuple struct, block-format the fields with a field on each +line and a trailing comma: ```rust pub struct Foo( @@ -168,9 +168,9 @@ pub struct Foo( ### Traits -Trait items should be block-indented. If there are no items, the trait may be -formatted on a single line. Otherwise there should be line-breaks after the -opening brace and before the closing brace: +Use block-indent for trait items. If there are no items, format the trait (including its `{}`) +on a single line. Otherwise, break after the opening brace and before +the closing brace: ```rust trait Foo {} @@ -180,8 +180,8 @@ pub trait Bar { } ``` -If the trait has bounds, there should be a space after the colon but not before -and before and after each `+`, e.g., +If the trait has bounds, put a space after the colon but not before, +and put spaces around each `+`, e.g., ```rust trait Foo: Debug + Bar {} @@ -207,9 +207,9 @@ pub trait IndexRanges: ### Impls -Impl items should be block indented. If there are no items, the impl may be -formatted on a single line. Otherwise there should be line-breaks after the -opening brace and before the closing brace: +Use block-indent for impl items. If there are no items, format the impl +(including its `{}`) on a single line. Otherwise, break after the opening brace +and before the closing brace: ```rust impl Foo {} @@ -267,12 +267,12 @@ macro_rules! foo { Prefer to put a generics clause on one line. Break other parts of an item declaration rather than line-breaking a generics clause. If a generics clause is -large enough to require line-breaking, you should prefer to use a `where` clause -instead. +large enough to require line-breaking, prefer a `where` clause instead. -Do not put spaces before or after `<` nor before `>`. Only put a space after `>` -if it is followed by a word or opening brace, not an opening parenthesis. There -should be a space after each comma and no trailing comma. +Do not put spaces before or after `<` nor before `>`. Only put a space after +`>` if it is followed by a word or opening brace, not an opening parenthesis. +Put a space after each comma. Do not use a trailing comma for a single-line +generics clause. ```rust fn foo(x: Vec, y: Vec) ... @@ -280,10 +280,9 @@ fn foo(x: Vec, y: Vec) ... impl SomeType { ... ``` -If the generics clause must be formatted across multiple lines, each parameter -should have its own block-indented line, there should be newlines after the -opening bracket and before the closing bracket, and there should be a trailing -comma. +If the generics clause must be formatted across multiple lines, put each +parameter on its own block-indented line, break after the opening `<` and +before the closing `>`, and use a trailing comma. ```rust fn foo< @@ -292,8 +291,7 @@ fn foo< >(x: Vec, y: Vec) ... ``` -If an associated type is bound in a generic type, then there should be spaces on -either side of the `=`: +If an associated type is bound in a generic type, put spaces around the `=`: ```rust > @@ -306,12 +304,14 @@ Prefer to use single-letter names for generic parameters. These rules apply for `where` clauses on any item. -A `where` clause may immediately follow a closing bracket of any kind. -Otherwise, it must start a new line, with no indent. Each component of a `where` -clause must be on its own line and be block indented. There should be a trailing +If immediately following a closing bracket of any kind, write the keyword +`where` on the same line, with a space before it. + +Otherwise, put `where` on a new line at the same indentation level. Put each +component of a `where` clause on its own line, block-indented. Use a trailing comma, unless the clause is terminated with a semicolon. If the `where` clause -is followed by a block (or assignment), the block should be started on a new -line. Examples: +is followed by a block (or assignment), start that block on a new line. +Examples: ```rust fn function(args) @@ -355,12 +355,12 @@ where = Bar; ``` -If a `where` clause is very short, we recommend using an inline bound on the -type parameter. +If a `where` clause is very short, prefer using an inline bound on the type +parameter. - -If a component of a `where` clause is long, it may be broken before `+` and -further block indented. Each bound should go on its own line. E.g., +If a component of a `where` clause does not fit and contains `+`, break it +before each `+` and block-indent the continuation lines. Put each bound on its +own line. E.g., ```rust impl IndexRanges for T @@ -369,40 +369,14 @@ where + Index, Output = Self::Output> + Index, Output = Self::Output> + Index, Output = Self::Output> - + Index, Output = Self::Output> + Index -``` - -#### Option - `where_single_line` - -`where_single_line` is `false` by default. If `true`, then a where clause with -exactly one component may be formatted on a single line if the rest of the -item's signature is also kept on one line. In this case, there is no need for a -trailing comma and if followed by a block, no need for a newline before the -block. E.g., - -```rust -// May be single-lined. -fn foo(args) -> ReturnType -where T: Bound { - body -} - -// Must be multi-lined. -fn foo( - args -) -> ReturnType -where - T: Bound, -{ - body -} + + Index, Output = Self::Output> + + Index, ``` - ### Type aliases -Type aliases should generally be kept on one line. If necessary to break the -line, do so after the `=`; the right-hand-side should be block indented: +Keep type aliases on one line when they fit. If necessary to break the line, do +so after the `=`, and block-indent the right-hand side: ```rust pub type Foo = Bar; @@ -427,9 +401,8 @@ where ### Associated types -Associated types should follow the guidelines above for type aliases. Where an -associated type has a bound, there should be a space after the colon but not -before: +Format associated types like type aliases. Where an associated type has a +bound, put a space after the colon but not before: ```rust pub type Foo: Bar; @@ -438,15 +411,14 @@ pub type Foo: Bar; ### extern items -When writing extern items (such as `extern "C" fn`), always be explicit about -the ABI. For example, write `extern "C" fn foo ...`, not `extern fn foo ...`, or +When writing extern items (such as `extern "C" fn`), always specify the ABI. +For example, write `extern "C" fn foo ...`, not `extern fn foo ...`, or `extern "C" { ... }`. ### Imports (`use` statements) -If an import can be formatted on one line, do so. There should be no spaces -around braces. +Format imports on one line where possible. Don't put spaces around braces. ```rust use a::b::c; @@ -458,7 +430,7 @@ use a::b::{foo, bar, baz}; #### Large list imports Prefer to use multiple imports rather than a multi-line import. However, tools -should not split imports by default (they may offer this as an option). +should not split imports by default. If an import does require multiple lines (either because a list of single names does not fit within the max width, or because of the rules for nested imports @@ -521,14 +493,12 @@ example, `a::*` comes before `b::a` but `a::b` comes before `a::*`. E.g., #### Normalisation -Tools must make the following normalisations: +Tools must make the following normalisations, recursively: * `use a::self;` -> `use a;` * `use a::{};` -> (nothing) * `use a::{b};` -> `use a::b;` -And must apply these recursively. - Tools must not otherwise merge or un-merge import lists or adjust glob imports (without an explicit option). diff --git a/src/doc/style-guide/src/statements.md b/src/doc/style-guide/src/statements.md index a5cd6da10616d..8c8f893fbe904 100644 --- a/src/doc/style-guide/src/statements.md +++ b/src/doc/style-guide/src/statements.md @@ -2,8 +2,8 @@ ### Let statements -There should be spaces after the `:` and on both sides of the `=` (if they are -present). No space before the semicolon. +Put a space after the `:` and on both sides of the `=` (if they are present). +Don't put a space before the semicolon. ```rust // A comment. @@ -14,19 +14,19 @@ let pattern: Type; let pattern = expr; ``` -If possible the declaration should be formatted on a single line. If this is not -possible, then try splitting after the `=`, if the declaration can fit on two -lines. The expression should be block indented. +If possible, format the declaration on a single line. If not possible, then try +splitting after the `=`, if the declaration fits on two lines. Block-indent the +expression. ```rust let pattern: Type = expr; ``` -If the first line does not fit on a single line, then split after the colon, -using block indentation. If the type covers multiple lines, even after line- -breaking after the `:`, then the first line may be placed on the same line as -the `:`, subject to the [combining rules](https://github.com/rust-lang-nursery/fmt-rfcs/issues/61) (WIP). +If the first line still does not fit on a single line, split after the `:`, and +use block indentation. If the type requires multiple lines, even after +line-breaking after the `:`, then place the first line on the same line as the +`:`, subject to the [combining rules](expressions.html#combinable-expressions). ```rust @@ -51,12 +51,12 @@ let (abcd, ``` If the expression covers multiple lines, if the first line of the expression -fits in the remaining space, it stays on the same line as the `=`, the rest of the -expression is not indented. If the first line does not fit, then it should start -on the next lines, and should be block indented. If the expression is a block -and the type or pattern cover multiple lines, then the opening brace should be -on a new line and not indented (this provides separation for the interior of the -block from the type), otherwise the opening brace follows the `=`. +fits in the remaining space, it stays on the same line as the `=`, and the rest +of the expression is not further indented. If the first line does not fit, then +put the expression on subsequent lines, block indented. If the expression is a +block and the type or pattern cover multiple lines, put the opening brace on a +new line and not indented (this provides separation for the interior of the +block from the type); otherwise, the opening brace follows the `=`. Examples: @@ -108,8 +108,8 @@ In this case, always apply the same formatting rules to the components preceding the `else` block (i.e. the `let pattern: Type = initializer_expr` portion) as described [for other let statements](#let-statements). -The entire let-else statement may be formatted on a single line if all the -following are true: +Format the entire let-else statement on a single line if all the following are +true: * the entire statement is *short* * the `else` block contains only a single-line expression and no statements @@ -120,9 +120,6 @@ following are true: let Some(1) = opt else { return }; ``` -Formatters may allow users to configure the value of the threshold -used to determine whether a let-else statement is *short*. - Otherwise, the let-else statement requires some line breaks. If breaking a let-else statement across multiple lines, never break between the @@ -157,9 +154,9 @@ before the `else`. }; ``` -If the initializer expression is multi-line, the `else` keyword and opening -brace of the block (i.e. `else {`) should be put on the same line as the end of -the initializer expression, with a space between them, if and only if all the +If the initializer expression is multi-line, put the `else` keyword and opening +brace of the block (i.e. `else {`) on the same line as the end of the +initializer expression, with a space between them, if and only if all the following are true: * The initializer expression ends with one or more closing @@ -182,9 +179,9 @@ let Some(x) = y.foo( } ``` -Otherwise, the `else` keyword and opening brace should be placed on the next -line after the end of the initializer expression, and the `else` keyword should -have the same indentation level as the `let` keyword. +Otherwise, put the `else` keyword and opening brace on the next line after the +end of the initializer expression, with the `else` keyword at the same +indentation level as the `let` keyword. For example: @@ -236,9 +233,9 @@ fn main() { ### Macros in statement position -A macro use in statement position should use parentheses or square brackets as -delimiters and should be terminated with a semicolon. There should be no spaces -between the name, `!`, the delimiters, or the `;`. +For a macro use in statement position, use parentheses or square brackets as +delimiters, and terminate it with a semicolon. Do not put spaces around the +name, `!`, the delimiters, or the `;`. ```rust // A comment. @@ -248,14 +245,14 @@ a_macro!(...); ### Expressions in statement position -There should be no space between the expression and the semicolon. +Do not put space between the expression and the semicolon. ``` ; ``` -All expressions in statement position should be terminated with a semicolon, -unless they end with a block or are used as the value for a block. +Terminate all expressions in statement position with a semicolon, unless they +end with a block or are used as the value for a block. E.g., diff --git a/src/doc/style-guide/src/types.md b/src/doc/style-guide/src/types.md index ae456ef21c8d4..15b001d4f2ea5 100644 --- a/src/doc/style-guide/src/types.md +++ b/src/doc/style-guide/src/types.md @@ -7,14 +7,14 @@ * `*const T`, `*mut T` (no space after `*`, space before type) * `&'a T`, `&T`, `&'a mut T`, `&mut T` (no space after `&`, single spaces separating other words) * `unsafe extern "C" fn<'a, 'b, 'c>(T, U, V) -> W` or `fn()` (single spaces around keywords and sigils, and after commas, no trailing commas, no spaces around brackets) -* `!` should be treated like any other type name, `Name` +* `!` gets treated like any other type name, `Name` * `(A, B, C, D)` (spaces after commas, no spaces around parens, no trailing comma unless it is a one-tuple) * ` as SomeTrait>::Foo::Bar` or `Foo::Bar` or `::Foo::Bar` (no spaces around `::` or angle brackets, single spaces around `as`) * `Foo::Bar` (spaces after commas, no trailing comma, no spaces around angle brackets) * `T + T + T` (single spaces between types, and `+`). * `impl T + T + T` (single spaces between keyword, types, and `+`). -Parentheses used in types should not be surrounded by whitespace, e.g., `(Foo)` +Do not put space around parentheses used in types, e.g., `(Foo)` ### Line breaks @@ -37,13 +37,17 @@ Foo> ``` -`[T; expr]` may be broken after the `;` if necessary. +If a type requires line-breaks in order to fit, this section outlines where to +break such types if necessary. -Function types may be broken following the rules for function declarations. +Break `[T; expr]` after the `;` if necessary. -Generic types may be broken following the rules for generics. +Break function types following the rules for function declarations. -Types with `+` may be broken after any `+` using block indent and breaking before the `+`. When breaking such a type, all `+`s should be line broken, e.g., +Break generic types following the rules for generics. + +Break types with `+` by breaking before the `+` and block-indenting the +subsequent lines. When breaking such a type, break before *every* `+`: ```rust impl Clone diff --git a/tests/run-make/print-cfg/Makefile b/tests/run-make/print-cfg/Makefile index 126f5768c90f2..6b153e5b54edd 100644 --- a/tests/run-make/print-cfg/Makefile +++ b/tests/run-make/print-cfg/Makefile @@ -2,7 +2,7 @@ include ../tools.mk -all: default +all: default output_to_file $(RUSTC) --target x86_64-pc-windows-gnu --print cfg | $(CGREP) windows $(RUSTC) --target x86_64-pc-windows-gnu --print cfg | $(CGREP) x86_64 $(RUSTC) --target i686-pc-windows-msvc --print cfg | $(CGREP) msvc @@ -11,6 +11,23 @@ all: default $(RUSTC) --target arm-unknown-linux-gnueabihf --print cfg | $(CGREP) target_abi= $(RUSTC) --target arm-unknown-linux-gnueabihf --print cfg | $(CGREP) eabihf +output_to_file: + # Backend-independent, printed by rustc_driver_impl/src/lib.rs + $(RUSTC) --target x86_64-pc-windows-gnu --print cfg=$(TMPDIR)/cfg.txt + $(CGREP) windows < $(TMPDIR)/cfg.txt + + # Printed from CodegenBackend trait impl in rustc_codegen_llvm/src/lib.rs + $(RUSTC) --print relocation-models=$(TMPDIR)/relocation-models.txt + $(CGREP) dynamic-no-pic < $(TMPDIR)/relocation-models.txt + + # Printed by compiler/rustc_codegen_llvm/src/llvm_util.rs + $(RUSTC) --target wasm32-unknown-unknown --print target-features=$(TMPDIR)/target-features.txt + $(CGREP) reference-types < $(TMPDIR)/target-features.txt + + # Printed by C++ code in rustc_llvm/llvm-wrapper/PassWrapper.cpp + $(RUSTC) --target wasm32-unknown-unknown --print target-cpus=$(TMPDIR)/target-cpus.txt + $(CGREP) generic < $(TMPDIR)/target-cpus.txt + ifdef IS_WINDOWS default: $(RUSTC) --print cfg | $(CGREP) windows diff --git a/tests/ui/codegen/target-cpus.rs b/tests/ui/codegen/target-cpus.rs new file mode 100644 index 0000000000000..1dff3ee6011bf --- /dev/null +++ b/tests/ui/codegen/target-cpus.rs @@ -0,0 +1,4 @@ +// needs-llvm-components: webassembly +// min-llvm-version: 17 +// compile-flags: --print=target-cpus --target=wasm32-unknown-unknown +// check-pass diff --git a/tests/ui/codegen/target-cpus.stdout b/tests/ui/codegen/target-cpus.stdout new file mode 100644 index 0000000000000..f60ba0f5034ba --- /dev/null +++ b/tests/ui/codegen/target-cpus.stdout @@ -0,0 +1,4 @@ +Available CPUs for this target: + bleeding-edge + generic - This is the default target CPU for the current build target (currently wasm32-unknown-unknown). + mvp diff --git a/tests/ui/unpretty/avoid-crash.stderr b/tests/ui/unpretty/avoid-crash.stderr index 11cd3866fa869..15bcc277e6491 100644 --- a/tests/ui/unpretty/avoid-crash.stderr +++ b/tests/ui/unpretty/avoid-crash.stderr @@ -1,4 +1,4 @@ -error: pretty-print failed to write `/tmp/` due to $ERROR_MESSAGE +error: failed to write `/tmp/` due to $ERROR_MESSAGE error: aborting due to previous error