gh-95853: Add script to automate WASM build (GH-95828)

tiran · brettcannon · web-flow · commit 32ac98e8992c · 2022-08-13T21:56:08.000+02:00
Automate WASM build with a new Python script. The script provides
several build profiles with configure flags for Emscripten flavors
and WASI. The script can detect and use Emscripten SDK and WASI SDK from
default locations or env vars.

``configure`` now detects Node arguments and creates HOSTRUNNER
arguments for Node 16. It also sets correct arguments for
``wasm64-emscripten``.

Co-authored-by: Brett Cannon &lt;brett@python.org&gt;
diff --git a/Lib/test/test_unicode_file_functions.py b/Lib/test/test_unicode_file_functions.py
@@ -6,6 +6,7 @@
 import warnings
 from unicodedata import normalize
 from test.support import os_helper
+from test import support
 
 
 filenames = [
@@ -123,6 +124,10 @@ def test_open(self):
     # NFKD in Python is useless, because darwin will normalize it later and so
     # open(), os.stat(), etc. don't raise any exception.
     @unittest.skipIf(sys.platform == 'darwin', 'irrelevant test on Mac OS X')
+    @unittest.skipIf(
+        support.is_emscripten or support.is_wasi,
+        "test fails on Emscripten/WASI when host platform is macOS."
+    )
     def test_normalize(self):
         files = set(self.files)
         others = set()
diff --git a/Lib/test/test_warnings/__init__.py b/Lib/test/test_warnings/__init__.py
@@ -487,7 +487,14 @@ def test_warn_explicit_non_ascii_filename(self):
                 module=self.module) as w:
             self.module.resetwarnings()
             self.module.filterwarnings("always", category=UserWarning)
-            for filename in ("nonascii\xe9\u20ac", "surrogate\udc80"):
+            filenames = ["nonascii\xe9\u20ac"]
+            if not support.is_emscripten:
+                # JavaScript does not like surrogates.
+                # Invalid UTF-8 leading byte 0x80 encountered when
+                # deserializing a UTF-8 string in wasm memory to a JS
+                # string!
+                filenames.append("surrogate\udc80")
+            for filename in filenames:
                 try:
                     os.fsencode(filename)
                 except UnicodeEncodeError:
diff --git a/Misc/NEWS.d/next/Tools-Demos/2022-08-10-17-08-43.gh-issue-95853.HCjC2m.rst b/Misc/NEWS.d/next/Tools-Demos/2022-08-10-17-08-43.gh-issue-95853.HCjC2m.rst
@@ -0,0 +1,2 @@
+The new tool ``Tools/wasm/wasm_builder.py`` automates configure, compile, and
+test steps for building CPython on WebAssembly platforms.
diff --git a/Python/sysmodule.c b/Python/sysmodule.c
@@ -2789,14 +2789,18 @@ EM_JS(char *, _Py_emscripten_runtime, (void), {
     if (typeof navigator == 'object') {
         info = navigator.userAgent;
     } else if (typeof process == 'object') {
-        info = "Node.js ".concat(process.version)
+        info = "Node.js ".concat(process.version);
     } else {
-        info = "UNKNOWN"
+        info = "UNKNOWN";
     }
     var len = lengthBytesUTF8(info) + 1;
     var res = _malloc(len);
-    stringToUTF8(info, res, len);
+    if (res) stringToUTF8(info, res, len);
+#if __wasm64__
+    return BigInt(res);
+#else
     return res;
+#endif
 });
 
 static PyObject *
diff --git a/Tools/wasm/README.md b/Tools/wasm/README.md
@@ -35,7 +35,13 @@ docker run --rm -ti -v $(pwd):/python-wasm/cpython -w /python-wasm/cpython quay.
 
 ### Compile a build Python interpreter
 
-From within the container, run the following commands:
+From within the container, run the following command:
+
+```shell
+./Tools/wasm/wasm_build.py build
+```
+
+The command is roughly equivalent to:
 
 ```shell
 mkdir -p builddir/build
@@ -45,13 +51,13 @@ make -j$(nproc)
 popd
 ```
 
-### Fetch and build additional emscripten ports
+### Cross-compile to wasm32-emscripten for browser
 
 ```shell
-embuilder build zlib bzip2
+./Tools/wasm/wasm_build.py emscripten-browser
 ```
 
-### Cross compile to wasm32-emscripten for browser
+The command is roughly equivalent to:
 
 ```shell
 mkdir -p builddir/emscripten-browser
@@ -85,22 +91,29 @@ and header files with debug builds.
 ### Cross compile to wasm32-emscripten for node
 
 ```shell
-mkdir -p builddir/emscripten-node
-pushd builddir/emscripten-node
+./Tools/wasm/wasm_build.py emscripten-browser-dl
+```
+
+The command is roughly equivalent to:
+
+```shell
+mkdir -p builddir/emscripten-node-dl
+pushd builddir/emscripten-node-dl
 
 CONFIG_SITE=../../Tools/wasm/config.site-wasm32-emscripten \
   emconfigure ../../configure -C \
     --host=wasm32-unknown-emscripten \
     --build=$(../../config.guess) \
     --with-emscripten-target=node \
+    --enable-wasm-dynamic-linking \
     --with-build-python=$(pwd)/../build/python
 
 emmake make -j$(nproc)
 popd
 ```
 
 ```shell
-node --experimental-wasm-threads --experimental-wasm-bulk-memory --experimental-wasm-bigint builddir/emscripten-node/python.js
+node --experimental-wasm-threads --experimental-wasm-bulk-memory --experimental-wasm-bigint builddir/emscripten-node-dl/python.js
 ```
 
 (``--experimental-wasm-bigint`` is not needed with recent NodeJS versions)
@@ -199,6 +212,15 @@ Node builds use ``NODERAWFS``.
 - Node RawFS allows direct access to the host file system without need to
   perform ``FS.mount()`` call.
 
+## wasm64-emscripten
+
+- wasm64 requires recent NodeJS and ``--experimental-wasm-memory64``.
+- ``EM_JS`` functions must return ``BigInt()``.
+- ``Py_BuildValue()`` format strings must match size of types. Confusing 32
+  and 64 bits types leads to memory corruption, see
+  [gh-95876](https://github.com/python/cpython/issues/95876) and
+  [gh-95878](https://github.com/python/cpython/issues/95878).
+
 # Hosting Python WASM builds
 
 The simple REPL terminal uses SharedArrayBuffer. For security reasons
@@ -234,6 +256,12 @@ The script ``wasi-env`` sets necessary compiler and linker flags as well as
 ``pkg-config`` overrides. The script assumes that WASI-SDK is installed in
 ``/opt/wasi-sdk`` or ``$WASI_SDK_PATH``.
 
+```shell
+./Tools/wasm/wasm_build.py wasi
+```
+
+The command is roughly equivalent to:
+
 ```shell
 mkdir -p builddir/wasi
 pushd builddir/wasi
diff --git a/Tools/wasm/wasi-env b/Tools/wasm/wasi-env
@@ -72,4 +72,5 @@ export CFLAGS LDFLAGS
 export PKG_CONFIG_PATH PKG_CONFIG_LIBDIR PKG_CONFIG_SYSROOT_DIR
 export PATH
 
-exec "$@"
+# no exec, it makes arvg[0] path absolute.
+"$@"
diff --git a/Tools/wasm/wasm_build.py b/Tools/wasm/wasm_build.py
diff --git a/configure b/configure
diff --git a/configure.ac b/configure.ac

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,2 @@`
	`1`	+The new tool ``Tools/wasm/wasm_builder.py`` automates configure, compile, and
	`2`	`+test steps for building CPython on WebAssembly platforms.`