fs: copy directory trees for fs.cp() on the thread pool

fs.cp() and fs.promises.cp() walked the tree in JavaScript with several
thread pool round trips per entry (opendir batches, two stat()s, the
copyFile(), a chmod()), all awaited in sequence: a 2 100-file tree took
~215 ms with ~110 ms of that on the main thread, against ~36 ms for
fs.cpSync(), which copies the tree in C++ when no filter is given.

Factor that C++ walk into CopyDirRecursive(), which records the error
instead of throwing so that it can run on any thread, and run it as one
ThreadPoolWork request (CpDirJob) for fs.cp()/fs.promises.cp() where
its behavior is the same as the JavaScript walk's: no filter function,
no dereference, not errorOnExist without force, permission model off.
For that job it also rejects sockets, FIFOs and unknown entries inside
the tree as the JavaScript walk does (cpSync keeps skipping them). The
same tree now takes ~28 ms with under 1 ms on the main thread.

Errors raised inside the walk carry the same codes as before but are
the plain Error objects cpSync produces rather than SystemErrors.

Signed-off-by: Shelley Vohr <shelley.vohr@gmail.com>

Author: codebytere

benchmark/fs/bench-cp.js

@@ -0,0 +1,33 @@
+'use strict';
+
+// fs.promises.cp() of a directory tree.
+
+const common = require('../common');
+const fs = require('fs');
+const path = require('path');
+const tmpdir = require('../../test/common/tmpdir');
+
+const bench = common.createBenchmark(main, {
+  files: [500],
+  n: [3],
+});
+
+function prepareSource(files) {
+  const src = tmpdir.resolve('cp-src');
+  for (let i = 0; i < files; i++) {
+    const dir = path.join(src, `dir-${i % 10}`, `sub-${i % 7}`);
+    fs.mkdirSync(dir, { recursive: true });
+    fs.writeFileSync(path.join(dir, `file-${i}.js`), 'x'.repeat(1024 + (i % 512)));
+  }
+  return src;
+}
+
+async function main({ files, n }) {
+  tmpdir.refresh();
+  const src = prepareSource(files);
+  bench.start();
+  for (let i = 0; i < n; i++) {
+    await fs.promises.cp(src, tmpdir.resolve(`cp-dest-${i}`), { recursive: true });
+  }
+  bench.end(n);
+}

lib/internal/fs/cp/cp.js

@@ -6,6 +6,8 @@ const {
   ArrayPrototypeEvery,
   ArrayPrototypeFilter,
   Boolean,
+  ErrorCaptureStackTrace,
+  Promise,
   PromisePrototypeThen,
   PromiseReject,
   SafePromiseAll,
@@ -55,6 +57,7 @@ const {
   sep,
 } = require('path');
 const fsBinding = internalBinding('fs');
+const permission = require('internal/process/permission');
 
 async function cpFn(src, dest, opts) {
   // Warn about using preserveTimestamps on 32-bit node
@@ -315,12 +318,45 @@ async function onDir(srcStat, destStat, src, dest, opts) {
 }
 
 async function mkDirAndCopy(srcMode, src, dest, opts) {
+  if (canCopyDirNatively(opts)) {
+    // Creates dest with the mode of src itself.
+    return copyDirNative(src, dest, opts);
+  }
   await mkdir(dest);
   await copyDir(src, dest, opts);
   return setDestMode(dest, srcMode);
 }
 
+// Without a filter there is no JavaScript to call per entry, so the tree is
+// copied in one thread pool request by the implementation fs.cpSync() uses.
+// The per-entry walk is kept where that implementation behaves differently:
+// it merges into existing nested directories (errorOnExist without force
+// rejects them here) and does not dereference symbolic links inside the tree.
+function canCopyDirNatively(opts) {
+  return !opts.filter && !opts.dereference && (opts.force || !opts.errorOnExist) &&
+    !permission.isEnabled();
+}
+
+function copyDirNative(src, dest, opts) {
+  return new Promise((resolve, reject) => {
+    const job = new fsBinding.CpDirJob(src, dest, opts.force, opts.dereference, opts.errorOnExist,
+                                       opts.verbatimSymlinks, opts.preserveTimestamps);
+    job.ondone = (err) => {
+      if (err != null) {
+        ErrorCaptureStackTrace(err, copyDirNative);
+        reject(err);
+      } else {
+        resolve();
+      }
+    };
+    job.run();
+  });
+}
+
 async function copyDir(src, dest, opts) {
+  if (canCopyDirNatively(opts)) {
+    return copyDirNative(src, dest, opts);
+  }
   const dir = await opendir(src);
 
   for await (const { name } of dir) {