Rollup merge of #145460 - Kobzol:bootstrap-speedup-copy-src-dirs, r=jieyouxu

Speedup `copy_src_dirs` in bootstrap

I was kinda offended by how slow it was. Just the `copy_src_dirs` part took ~3s locally in the `x dist rustc-src` step. In release mode it was just 1s, but that's kind of cheating (I wonder if we should build bootstrap in release mode on CI though...).

Did some basic optimizations to bring it down to ~1s also in debug mode.

Maybe it's overkill, due to #145455. Up to you whether we should merge it or close it :)

r? `````````@jieyouxu`````````

Author: Zalathar

src/bootstrap/src/core/build_steps/dist.rs

@@ -916,79 +916,74 @@ fn copy_src_dirs(
     exclude_dirs: &[&str],
     dst_dir: &Path,
 ) {
+    // The src directories should be relative to `base`, we depend on them not being absolute
+    // paths below.
+    for src_dir in src_dirs {
+        assert!(Path::new(src_dir).is_relative());
+    }
+
     // Iterating, filtering and copying a large number of directories can be quite slow.
     // Avoid doing it in dry run (and thus also tests).
     if builder.config.dry_run() {
         return;
     }
 
     fn filter_fn(exclude_dirs: &[&str], dir: &str, path: &Path) -> bool {
+        // The paths are relative, e.g. `llvm-project/...`.
         let spath = match path.to_str() {
             Some(path) => path,
             None => return false,
         };
         if spath.ends_with('~') || spath.ends_with(".pyc") {
             return false;
         }
+        // Normalize slashes
+        let spath = spath.replace("\\", "/");
 
-        const LLVM_PROJECTS: &[&str] = &[
+        static LLVM_PROJECTS: &[&str] = &[
             "llvm-project/clang",
-            "llvm-project\\clang",
             "llvm-project/libunwind",
-            "llvm-project\\libunwind",
             "llvm-project/lld",
-            "llvm-project\\lld",
             "llvm-project/lldb",
-            "llvm-project\\lldb",
             "llvm-project/llvm",
-            "llvm-project\\llvm",
             "llvm-project/compiler-rt",
-            "llvm-project\\compiler-rt",
             "llvm-project/cmake",
-            "llvm-project\\cmake",
             "llvm-project/runtimes",
-            "llvm-project\\runtimes",
             "llvm-project/third-party",
-            "llvm-project\\third-party",
         ];
-        if spath.contains("llvm-project")
-            && !spath.ends_with("llvm-project")
-            && !LLVM_PROJECTS.iter().any(|path| spath.contains(path))
-        {
-            return false;
-        }
+        if spath.starts_with("llvm-project") && spath != "llvm-project" {
+            if !LLVM_PROJECTS.iter().any(|path| spath.starts_with(path)) {
+                return false;
+            }
 
-        // Keep only these third party libraries
-        const LLVM_THIRD_PARTY: &[&str] =
-            &["llvm-project/third-party/siphash", "llvm-project\\third-party\\siphash"];
-        if (spath.starts_with("llvm-project/third-party")
-            || spath.starts_with("llvm-project\\third-party"))
-            && !(spath.ends_with("llvm-project/third-party")
-                || spath.ends_with("llvm-project\\third-party"))
-            && !LLVM_THIRD_PARTY.iter().any(|path| spath.contains(path))
-        {
-            return false;
-        }
+            // Keep siphash third-party dependency
+            if spath.starts_with("llvm-project/third-party")
+                && spath != "llvm-project/third-party"
+                && !spath.starts_with("llvm-project/third-party/siphash")
+            {
+                return false;
+            }
 
-        const LLVM_TEST: &[&str] = &["llvm-project/llvm/test", "llvm-project\\llvm\\test"];
-        if LLVM_TEST.iter().any(|path| spath.contains(path))
-            && (spath.ends_with(".ll") || spath.ends_with(".td") || spath.ends_with(".s"))
-        {
-            return false;
+            if spath.starts_with("llvm-project/llvm/test")
+                && (spath.ends_with(".ll") || spath.ends_with(".td") || spath.ends_with(".s"))
+            {
+                return false;
+            }
         }
 
         // Cargo tests use some files like `.gitignore` that we would otherwise exclude.
-        const CARGO_TESTS: &[&str] = &["tools/cargo/tests", "tools\\cargo\\tests"];
-        if CARGO_TESTS.iter().any(|path| spath.contains(path)) {
+        if spath.starts_with("tools/cargo/tests") {
             return true;
         }
 
-        let full_path = Path::new(dir).join(path);
-        if exclude_dirs.iter().any(|excl| full_path == Path::new(excl)) {
-            return false;
+        if !exclude_dirs.is_empty() {
+            let full_path = Path::new(dir).join(path);
+            if exclude_dirs.iter().any(|excl| full_path == Path::new(excl)) {
+                return false;
+            }
         }
 
-        let excludes = [
+        static EXCLUDES: &[&str] = &[
             "CVS",
             "RCS",
             "SCCS",
@@ -1011,7 +1006,15 @@ fn copy_src_dirs(
             ".hgrags",
             "_darcs",
         ];
-        !path.iter().map(|s| s.to_str().unwrap()).any(|s| excludes.contains(&s))
+
+        // We want to check if any component of `path` doesn't contain the strings in `EXCLUDES`.
+        // However, since we traverse directories top-down in `Builder::cp_link_filtered`,
+        // it is enough to always check only the last component:
+        // - If the path is a file, we will iterate to it and then check it's filename
+        // - If the path is a dir, if it's dir name contains an excluded string, we will not even
+        //   recurse into it.
+        let last_component = path.iter().next_back().map(|s| s.to_str().unwrap()).unwrap();
+        !EXCLUDES.contains(&last_component)
     }
 
     // Copy the directories using our filter

src/bootstrap/src/lib.rs

@@ -1862,7 +1862,6 @@ impl Build {
                     self.create_dir(&dst);
                     self.cp_link_filtered_recurse(&path, &dst, &relative, filter);
                 } else {
-                    let _ = fs::remove_file(&dst);
                     self.copy_link(&path, &dst, FileType::Regular);
                 }
             }