[WasmFS] Add wasmfs_before_preload hook to run code before preloading (#16967)

Preloaded files are added during startup, but sometimes the user wants to
load them under a custom backend and not the default. In the old FS, the
user could mount directories in a Module.preRun, but with WasmFS that
has to be done in compiled code. To allow that, this defines a hook that
users can implement:

wasmfs_before_preload

If that the symbol is defined then it is called right before preloading.

To make this work the file preloading code will now skip any directories
that already exist, instead of erroring on them.

Author: kripken

system/include/emscripten/wasmfs.h

@@ -35,7 +35,7 @@ int wasmfs_create_directory(const char* path, long mode, backend_t backend);
 // Backend creation
 
 // Creates a JSFile Backend in the new file system.
-backend_t wasmfs_create_js_file_backend();
+backend_t wasmfs_create_js_file_backend(void);
 
 // A function that receives a void* and returns a backend.
 typedef backend_t (*backend_constructor_t)(void*);
@@ -50,6 +50,15 @@ backend_t wasmfs_create_node_backend(const char* root);
 
 backend_t wasmfs_create_opfs_backend(void);
 
+// Hooks
+
+// A hook users can do to run code during WasmFS startup. This hook happens
+// before file preloading, so user code could create backends and mount them,
+// which would then affect in which backend the preloaded files are loaded (the
+// preloaded files have paths, and so they are added to that path and whichever
+// backend is present there).
+void wasmfs_before_preload(void);
+
 #ifdef __cplusplus
 }
 #endif

system/lib/wasmfs/wasmfs.cpp

@@ -39,6 +39,15 @@ void _wasmfs_get_preloaded_path_name(int index, char* fileName);
 void _wasmfs_get_preloaded_child_path(int index, char* childName);
 }
 
+// If the user does not implement this hook, do nothing.
+__attribute__((weak)) extern "C" void wasmfs_before_preload(void) {}
+
+// Set up global data structures and preload files.
+WasmFS::WasmFS() : rootDirectory(initRootDirectory()), cwd(rootDirectory) {
+  wasmfs_before_preload();
+  preloadFiles();
+}
+
 WasmFS::~WasmFS() {
   // Flush musl libc streams.
   // TODO: Integrate musl exit() which would call this for us. That might also
@@ -92,9 +101,6 @@ void WasmFS::preloadFiles() {
   assert(timesCalled == 1);
 #endif
 
-  // Obtain the backend of the root directory.
-  auto rootBackend = getRootDirectory()->getBackend();
-
   // Ensure that files are preloaded from the main thread.
   assert(emscripten_is_main_runtime_thread());
 
@@ -124,8 +130,14 @@ void WasmFS::preloadFiles() {
     char childName[PATH_MAX] = {};
     _wasmfs_get_preloaded_child_path(i, childName);
 
+    auto lockedParentDir = parentDir->locked();
+    if (lockedParentDir.getChild(childName)) {
+      // The child already exists, so we don't need to do anything here.
+      continue;
+    }
+
     auto inserted =
-      parentDir->locked().insertDirectory(childName, S_IRUGO | S_IXUGO);
+      lockedParentDir.insertDirectory(childName, S_IRUGO | S_IXUGO);
     assert(inserted && "TODO: handle preload insertion errors");
   }
 

system/lib/wasmfs/wasmfs.h

@@ -37,11 +37,7 @@ class WasmFS {
   void preloadFiles();
 
 public:
-  // Set up global data structures and preload files.
-  WasmFS() : rootDirectory(initRootDirectory()), cwd(rootDirectory) {
-    preloadFiles();
-  }
-
+  WasmFS();
   ~WasmFS();
 
   FileTable& getFileTable() { return fileTable; }

tests/test_other.py

@@ -11721,6 +11721,13 @@ def test_wasmfs_jsfile_proxying_backend(self):
     self.set_setting('EXIT_RUNTIME')
     self.test_wasmfs_jsfile()
 
+  def test_wasmfs_before_preload(self):
+    self.set_setting('WASMFS')
+    os.mkdir('js_backend_files')
+    create_file('js_backend_files/file.dat', 'data')
+    self.emcc_args += ['--preload-file', 'js_backend_files/file.dat']
+    self.do_run_in_out_file_test('wasmfs/wasmfs_before_preload.c')
+
   @disabled('Running with initial >2GB heaps is not currently supported on the CI version of Node')
   def test_hello_world_above_2gb(self):
     self.run_process([EMCC, test_file('hello_world.c'), '-sGLOBAL_BASE=2147483648', '-sINITIAL_MEMORY=3GB'])

tests/wasmfs/wasmfs_before_preload.c

@@ -0,0 +1,41 @@
+/*
+ * Copyright 2022 The Emscripten Authors.  All rights reserved.
+ * Emscripten is available under two separate licenses, the MIT license and the
+ * University of Illinois/NCSA Open Source License.  Both these licenses can be
+ * found in the LICENSE file.
+ */
+
+#include <assert.h>
+#include <emscripten/console.h>
+#include <emscripten/wasmfs.h>
+#include <stdio.h>
+
+static backend_t my_js_file_backend;
+
+void wasmfs_before_preload(void) {
+  emscripten_console_log("before_preload");
+
+  // Make a JS file backend during startup, by implementing this hook function,
+  // so that the preloaded files end up stored there (instead of in the default
+  // backend).
+  my_js_file_backend = wasmfs_create_js_file_backend();
+  assert(my_js_file_backend);
+  int result =
+    wasmfs_create_directory("/js_backend_files", 0777, my_js_file_backend);
+  assert(result == 0);
+}
+
+int main() {
+  emscripten_console_log("main");
+
+  // The root uses the default backend.
+  backend_t default_backend = wasmfs_get_backend_by_path("/");
+  assert(default_backend != my_js_file_backend);
+
+  // The preloaded file uses the js backend we created during startup.
+  backend_t file_backend =
+    wasmfs_get_backend_by_path("/js_backend_files/file.dat");
+  assert(file_backend == my_js_file_backend);
+
+  emscripten_console_log("success");
+}

tests/wasmfs/wasmfs_before_preload.out

@@ -0,0 +1,3 @@
+before_preload
+main
+success