Auto merge of #126930 - Xaeroxe:file-checksum-hint, r=<try>

Add unstable support for outputting file checksums for use in cargo

Adds an unstable option that appends file checksums and expected lengths to the end of the dep-info file such that `cargo` can read and use these values as an alternative to file mtimes.

This PR powers the changes made in this cargo PR rust-lang/cargo#14137

Here's the tracking issue for the cargo feature rust-lang/cargo#14136.

Cargo.lock

@@ -237,6 +237,12 @@ dependencies = [
  "object 0.35.0",
 ]
 
+[[package]]
+name = "arrayref"
+version = "0.3.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6b4930d2cb77ce62f89ee5d5289b4ac049559b1c45539271f5ed4fdc7db34545"
+
 [[package]]
 name = "arrayvec"
 version = "0.7.4"
@@ -356,6 +362,19 @@ version = "2.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "cf4b9d6a944f767f8e5e0db018570623c85f3d925ac718db4e06d0187adb21c1"
 
+[[package]]
+name = "blake3"
+version = "1.5.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3d08263faac5cde2a4d52b513dadb80846023aade56fcd8fc99ba73ba8050e92"
+dependencies = [
+ "arrayref",
+ "arrayvec",
+ "cc",
+ "cfg-if",
+ "constant_time_eq",
+]
+
 [[package]]
 name = "block-buffer"
 version = "0.10.4"
@@ -842,6 +861,12 @@ dependencies = [
  "windows-sys 0.52.0",
 ]
 
+[[package]]
+name = "constant_time_eq"
+version = "0.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f7144d30dcf0fafbce74250a3963025d8d52177934239851c917d29f1df280c2"
+
 [[package]]
 name = "core"
 version = "0.0.0"
@@ -4754,6 +4779,7 @@ dependencies = [
 name = "rustc_span"
 version = "0.0.0"
 dependencies = [
+ "blake3",
  "derivative",
  "indexmap",
  "itoa",

compiler/rustc_codegen_llvm/src/debuginfo/metadata.rs

@@ -636,6 +636,7 @@ pub fn file_metadata<'ll>(cx: &CodegenCx<'ll, '_>, source_file: &SourceFile) ->
             rustc_span::SourceFileHashAlgorithm::Md5 => llvm::ChecksumKind::MD5,
             rustc_span::SourceFileHashAlgorithm::Sha1 => llvm::ChecksumKind::SHA1,
             rustc_span::SourceFileHashAlgorithm::Sha256 => llvm::ChecksumKind::SHA256,
+            rustc_span::SourceFileHashAlgorithm::Blake3 => llvm::ChecksumKind::None,
         };
         let hash_value = hex_encode(source_file.src_hash.hash_bytes());
 

compiler/rustc_interface/src/interface.rs

@@ -387,12 +387,13 @@ pub fn run_compiler<R: Send>(config: Config, f: impl FnOnce(&Compiler) -> R + Se
     let file_loader = config.file_loader.unwrap_or_else(|| Box::new(RealFileLoader));
     let path_mapping = config.opts.file_path_mapping();
     let hash_kind = config.opts.unstable_opts.src_hash_algorithm(&target);
+    let checksum_hash_kind = config.opts.unstable_opts.checksum_hash_algorithm();
 
     util::run_in_thread_pool_with_globals(
         &early_dcx,
         config.opts.edition,
         config.opts.unstable_opts.threads,
-        SourceMapInputs { file_loader, path_mapping, hash_kind },
+        SourceMapInputs { file_loader, path_mapping, hash_kind, checksum_hash_kind },
         |current_gcx| {
             // The previous `early_dcx` can't be reused here because it doesn't
             // impl `Send`. Creating a new one is fine.

compiler/rustc_interface/src/lib.rs

@@ -1,5 +1,6 @@
 // tidy-alphabetical-start
 #![feature(decl_macro)]
+#![feature(iter_intersperse)]
 #![feature(let_chains)]
 #![feature(thread_spawn_unchecked)]
 #![feature(try_blocks)]

compiler/rustc_interface/src/passes.rs

@@ -32,12 +32,13 @@ use rustc_session::output::{collect_crate_types, find_crate_name};
 use rustc_session::search_paths::PathKind;
 use rustc_session::{Limit, Session};
 use rustc_span::symbol::{sym, Symbol};
-use rustc_span::FileName;
+use rustc_span::{FileName, SourceFileHash, SourceFileHashAlgorithm};
 use rustc_target::spec::PanicStrategy;
 use rustc_trait_selection::traits;
 
 use std::any::Any;
 use std::ffi::OsString;
+use std::fs::File;
 use std::io::{self, BufWriter, Write};
 use std::path::{Path, PathBuf};
 use std::sync::{Arc, LazyLock};
@@ -420,15 +421,23 @@ fn write_out_deps(tcx: TyCtxt<'_>, outputs: &OutputFilenames, out_filenames: &[P
     let result: io::Result<()> = try {
         // Build a list of files used to compile the output and
         // write Makefile-compatible dependency rules
-        let mut files: Vec<String> = sess
+        let mut files: Vec<(String, u64, Option<SourceFileHash>)> = sess
             .source_map()
             .files()
             .iter()
             .filter(|fmap| fmap.is_real_file())
             .filter(|fmap| !fmap.is_imported())
-            .map(|fmap| escape_dep_filename(&fmap.name.prefer_local().to_string()))
+            .map(|fmap| {
+                (
+                    escape_dep_filename(&fmap.name.prefer_local().to_string()),
+                    fmap.source_len.0 as u64,
+                    fmap.checksum_hash,
+                )
+            })
             .collect();
 
+        let checksum_hash_algo = sess.opts.unstable_opts.checksum_hash_algorithm;
+
         // Account for explicitly marked-to-track files
         // (e.g. accessed in proc macros).
         let file_depinfo = sess.psess.file_depinfo.borrow();
@@ -438,58 +447,115 @@ fn write_out_deps(tcx: TyCtxt<'_>, outputs: &OutputFilenames, out_filenames: &[P
             escape_dep_filename(&file.prefer_local().to_string())
         };
 
+        fn hash_iter_files<P: AsRef<Path>>(
+            it: impl Iterator<Item = P>,
+            checksum_hash_algo: Option<SourceFileHashAlgorithm>,
+        ) -> impl Iterator<Item = (P, u64, Option<SourceFileHash>)> {
+            it.map(move |path| {
+                match checksum_hash_algo.and_then(|algo| {
+                    File::open(path.as_ref())
+                        .and_then(|mut file| {
+                            SourceFileHash::new(algo, &mut file).map(|h| (file, h))
+                        })
+                        .and_then(|(file, h)| file.metadata().map(|m| (m.len(), h)))
+                        .map_err(|e| {
+                            tracing::error!(
+                                "failed to compute checksum, omitting it from dep-info {} {e}",
+                                path.as_ref().display()
+                            )
+                        })
+                        .ok()
+                }) {
+                    Some((file_len, checksum)) => (path, file_len, Some(checksum)),
+                    None => (path, 0, None),
+                }
+            })
+        }
+
         // The entries will be used to declare dependencies beween files in a
         // Makefile-like output, so the iteration order does not matter.
         #[allow(rustc::potential_query_instability)]
-        let extra_tracked_files =
-            file_depinfo.iter().map(|path_sym| normalize_path(PathBuf::from(path_sym.as_str())));
+        let extra_tracked_files = hash_iter_files(
+            file_depinfo.iter().map(|path_sym| normalize_path(PathBuf::from(path_sym.as_str()))),
+            checksum_hash_algo,
+        );
         files.extend(extra_tracked_files);
 
         // We also need to track used PGO profile files
         if let Some(ref profile_instr) = sess.opts.cg.profile_use {
-            files.push(normalize_path(profile_instr.as_path().to_path_buf()));
+            files.extend(hash_iter_files(
+                iter::once(normalize_path(profile_instr.as_path().to_path_buf())),
+                checksum_hash_algo,
+            ));
         }
         if let Some(ref profile_sample) = sess.opts.unstable_opts.profile_sample_use {
-            files.push(normalize_path(profile_sample.as_path().to_path_buf()));
+            files.extend(hash_iter_files(
+                iter::once(normalize_path(profile_sample.as_path().to_path_buf())),
+                checksum_hash_algo,
+            ));
         }
 
         // Debugger visualizer files
         for debugger_visualizer in tcx.debugger_visualizers(LOCAL_CRATE) {
-            files.push(normalize_path(debugger_visualizer.path.clone().unwrap()));
+            files.extend(hash_iter_files(
+                iter::once(normalize_path(debugger_visualizer.path.clone().unwrap())),
+                checksum_hash_algo,
+            ));
         }
 
         if sess.binary_dep_depinfo() {
             if let Some(ref backend) = sess.opts.unstable_opts.codegen_backend {
                 if backend.contains('.') {
                     // If the backend name contain a `.`, it is the path to an external dynamic
                     // library. If not, it is not a path.
-                    files.push(backend.to_string());
+                    files.extend(hash_iter_files(
+                        iter::once(backend.to_string()),
+                        checksum_hash_algo,
+                    ));
                 }
             }
 
             for &cnum in tcx.crates(()) {
                 let source = tcx.used_crate_source(cnum);
                 if let Some((path, _)) = &source.dylib {
-                    files.push(escape_dep_filename(&path.display().to_string()));
+                    files.extend(hash_iter_files(
+                        iter::once(escape_dep_filename(&path.display().to_string())),
+                        checksum_hash_algo,
+                    ));
                 }
                 if let Some((path, _)) = &source.rlib {
-                    files.push(escape_dep_filename(&path.display().to_string()));
+                    files.extend(hash_iter_files(
+                        iter::once(escape_dep_filename(&path.display().to_string())),
+                        checksum_hash_algo,
+                    ));
                 }
                 if let Some((path, _)) = &source.rmeta {
-                    files.push(escape_dep_filename(&path.display().to_string()));
+                    files.extend(hash_iter_files(
+                        iter::once(escape_dep_filename(&path.display().to_string())),
+                        checksum_hash_algo,
+                    ));
                 }
             }
         }
 
         let write_deps_to_file = |file: &mut dyn Write| -> io::Result<()> {
             for path in out_filenames {
-                writeln!(file, "{}: {}\n", path.display(), files.join(" "))?;
+                writeln!(
+                    file,
+                    "{}: {}\n",
+                    path.display(),
+                    files
+                        .iter()
+                        .map(|(path, _file_len, _checksum_hash_algo)| path.as_str())
+                        .intersperse(" ")
+                        .collect::<String>()
+                )?;
             }
 
             // Emit a fake target for each input file to the compilation. This
             // prevents `make` from spitting out an error if a file is later
             // deleted. For more info see #28735
-            for path in files {
+            for (path, _file_len, _checksum_hash_algo) in &files {
                 writeln!(file, "{path}:")?;
             }
 
@@ -513,6 +579,18 @@ fn write_out_deps(tcx: TyCtxt<'_>, outputs: &OutputFilenames, out_filenames: &[P
                 }
             }
 
+            // If caller requested this information, add special comments about source file checksums.
+            // These are not necessarily the same checksums as was used in the debug files.
+            if sess.opts.unstable_opts.checksum_hash_algorithm().is_some() {
+                for (path, file_len, checksum_hash) in
+                    files.iter().filter_map(|(path, file_len, hash_algo)| {
+                        hash_algo.map(|hash_algo| (path, file_len, hash_algo))
+                    })
+                {
+                    writeln!(file, "# checksum:{checksum_hash} file_len:{file_len} {path}")?;
+                }
+            }
+
             Ok(())
         };
 

compiler/rustc_interface/src/tests.rs

@@ -48,10 +48,12 @@ where
     let sysroot = filesearch::materialize_sysroot(sessopts.maybe_sysroot.clone());
     let target = rustc_session::config::build_target_config(&early_dcx, &sessopts, &sysroot);
     let hash_kind = sessopts.unstable_opts.src_hash_algorithm(&target);
+    let checksum_hash_kind = sessopts.unstable_opts.checksum_hash_algorithm();
     let sm_inputs = Some(SourceMapInputs {
         file_loader: Box::new(RealFileLoader) as _,
         path_mapping: sessopts.file_path_mapping(),
         hash_kind,
+        checksum_hash_kind,
     });
 
     rustc_span::create_session_globals_then(DEFAULT_EDITION, sm_inputs, || {

compiler/rustc_metadata/src/rmeta/decoder.rs

@@ -1723,6 +1723,7 @@ impl<'a, 'tcx> CrateMetadataRef<'a> {
                 let rustc_span::SourceFile {
                     mut name,
                     src_hash,
+                    checksum_hash,
                     start_pos: original_start_pos,
                     source_len,
                     lines,
@@ -1774,6 +1775,7 @@ impl<'a, 'tcx> CrateMetadataRef<'a> {
                 let local_version = sess.source_map().new_imported_source_file(
                     name,
                     src_hash,
+                    checksum_hash,
                     stable_id,
                     source_len.to_u32(),
                     self.cnum,

compiler/rustc_query_system/src/ich/impls_syntax.rs

@@ -68,6 +68,8 @@ impl<'a> HashStable<StableHashingContext<'a>> for SourceFile {
             // Do not hash the source as it is not encoded
             src: _,
             ref src_hash,
+            // Already includes src_hash, this is redundant
+            checksum_hash: _,
             external_src: _,
             start_pos: _,
             source_len: _,

compiler/rustc_session/src/config.rs

@@ -1219,6 +1219,10 @@ impl UnstableOptions {
             }
         })
     }
+
+    pub fn checksum_hash_algorithm(&self) -> Option<SourceFileHashAlgorithm> {
+        self.checksum_hash_algorithm
+    }
 }
 
 // The type of entry function, so users can have their own entry functions

compiler/rustc_session/src/options.rs

@@ -413,7 +413,8 @@ mod desc {
     pub const parse_merge_functions: &str = "one of: `disabled`, `trampolines`, or `aliases`";
     pub const parse_symbol_mangling_version: &str =
         "one of: `legacy`, `v0` (RFC 2603), or `hashed`";
-    pub const parse_src_file_hash: &str = "either `md5` or `sha1`";
+    pub const parse_cargo_src_file_hash: &str = "one of `sha256`";
+    pub const parse_src_file_hash: &str = "one of `md5`, `sha1`, or `sha256`";
     pub const parse_relocation_model: &str =
         "one of supported relocation models (`rustc --print relocation-models`)";
     pub const parse_code_model: &str = "one of supported code models (`rustc --print code-models`)";
@@ -1261,6 +1262,23 @@ mod parse {
         true
     }
 
+    pub(crate) fn parse_cargo_src_file_hash(
+        slot: &mut Option<SourceFileHashAlgorithm>,
+        v: Option<&str>,
+    ) -> bool {
+        match v.and_then(|s| SourceFileHashAlgorithm::from_str(s).ok()) {
+            Some(hash_kind) => {
+                if hash_kind.supported_in_cargo() {
+                    *slot = Some(hash_kind);
+                } else {
+                    return false;
+                }
+            }
+            _ => return false,
+        }
+        true
+    }
+
     pub(crate) fn parse_target_feature(slot: &mut String, v: Option<&str>) -> bool {
         match v {
             Some(s) => {
@@ -1647,6 +1665,8 @@ options! {
         "instrument control-flow architecture protection"),
     check_cfg_all_expected: bool = (false, parse_bool, [UNTRACKED],
         "show all expected values in check-cfg diagnostics (default: no)"),
+    checksum_hash_algorithm: Option<SourceFileHashAlgorithm> = (None, parse_cargo_src_file_hash, [TRACKED],
+        "hash algorithm of source files used to check freshness in cargo (`sha256`)"),
     codegen_backend: Option<String> = (None, parse_opt_string, [TRACKED],
         "the backend to use"),
     combine_cgu: bool = (false, parse_bool, [TRACKED],

compiler/rustc_span/Cargo.toml

@@ -5,6 +5,7 @@ edition = "2021"
 
 [dependencies]
 # tidy-alphabetical-start
+blake3 = "1.5.2"
 derivative = "2.2.0"
 indexmap = { version = "2.0.0" }
 itoa = "1.0"