[CodeGen][ObjC] Initial WebAssembly Support for GNUstep #169043

hmelder · 2025-11-21T14:13:15Z

This pull request adds initial support for compiling Objective-C to WebAssembly. I tested my changes with libobjc2 and the swift-corelibs-blocksruntime.

There are two outstanding issues, which I cannot fix as deeper knowledge of the subsystems is required:

Symbols marked as explicitly hidden in code generation are exported
Clang crashes in SelectionDAG when compiling an Objective-C try/catch block with -fwasm-exceptions

First Issue

Emscripten is processing the generated .wasm file in emscripten.py and checks if all exported symbols are valid javascript identifiers (tools/js_manipulation.py#L104). However, hidden symbols such as .objc_init are intentionally an invalid C identifier.

The core of the problem is that symbols with the WASM_SYMBOL_NO_STRIP attribute are exported when targeting Emscripten (https://reviews.llvm.org/D62542). This attribute is added to the symbol during relocation in WasmObjectWriter::recordRelocation. So we are accidentally exporting a lot of hidden symbols and not only ones generated by ObjC CG...

I'm currently hacking around this by not exporting no-strip symbols. This is the default behaviour for Wasm.

Second Issue

Here is a minimal example that triggers the crash.

#include<stdio.h>

int main(void) {
	int ret = 0;
	@try {
	}
	@catch (id a)
	{
		ret = 1;
                 puts("abc");
	}

	return ret;
}

The following assertion is triggered:

clang: /home/vm/llvm-project/llvm/lib/Target/WebAssembly/WebAssemblyExceptionInfo.cpp:124: void llvm::WebAssemblyExceptionInfo::recalculate(MachineFunction &, MachineDominatorTree &, const MachineDominanceFrontier &): Assertion `EHInfo' failed.

Here is the crash report main-c3884.zip.

You can use emcc with a modified LLVM build by exporting EM_LLVM_ROOT before sourcing emsdk/emsdk_env.sh:

emcc -fobjc-runtime=gnustep-2.2 -fwasm-exceptions -c main.m

or just invoke clang directly:

/home/vm/llvm-build-wasm/bin/clang -target wasm32-unknown-emscripten -mllvm -combiner-global-alias-analysis=false -mllvm -wasm-enable-sjlj -mllvm -wasm-use-legacy-eh=false -mllvm -disable-lsr --sysroot=/home/vm/emsdk/upstream/emscripten/cache/sysroot -DEMSCRIPTEN -fobjc-runtime=gnustep-2.2 -fwasm-exceptions -c main.m

Building libobjc2 and the BlocksRuntime

Building the BlocksRuntime

cmake -DCMAKE_TOOLCHAIN_FILE=$EMSDK/upstream/emscripten/cmake/Modules/Platform/Emscripten.cmake   -DCMAKE_INSTALL_PREFIX=/home/vm/demo-install -DCMAKE_BUILD_TYPE=Debug -B build -G Ninja

Building libobjc2

cmake -DCMAKE_TOOLCHAIN_FILE=$EMSDK/upstream/emscripten/cmake/Modules/Platform/Emscripten.cmake   -DCMAKE_INSTALL_PREFIX=/home/vm/demo-install -DBlocksRuntime_LIBRARIES=/home/vm/demo-install/lib/libBlocksRuntime.a -DBlocksRuntime_INCLUDE_DIR=/home/vm/demo-install/include/BlocksRuntime -DEMBEDDED_BLOCKS_RUNTIME=OFF -DTESTS=OFF  -B build  -DCMAKE_BUILD_TYPE=Debug  -G Ninja

llvmbot · 2025-11-21T14:13:45Z

@llvm/pr-subscribers-clang
@llvm/pr-subscribers-clang-codegen
@llvm/pr-subscribers-clang-driver

@llvm/pr-subscribers-backend-webassembly

Author: Hugo Melder (hmelder)

Changes

This pull request adds initial support for compiling Objective-C to WebAssembly. I tested my changes with libobjc2 and the swift-corelibs-blocksruntime.

There are two outstanding issues, which I cannot fix as deeper knowledge of the subsystems is required:

Symbols marked as explicitly hidden in code generation are exported
Clang crashes in SelectionDAG when compiling an Objective-C try/catch block with -fwasm-exceptions

First Issue

Emscripten is processing the generated .wasm file in emscripten.py and checks if all exported symbols are valid javascript identifiers (tools/js_manipulation.py#L104). However, hidden symbols such as .objc_init are intentionally an invalid C identifier.

The core of the problem is that symbols with the WASM_SYMBOL_NO_STRIP attribute are exported when targeting Emscripten (https://reviews.llvm.org/D62542). This attribute is added to the symbol during relocation in WasmObjectWriter::recordRelocation. So we are accidentally exporting a lot of hidden symbols and not only ones generated by ObjC CG...

I'm currently hacking around this by not exporting no-strip symbols. This is the default behaviour for Wasm.

Second Issue

Here is a minimal example that triggers the crash.

#include&lt;stdio.h&gt;

int main(void) {
	int ret = 0;
	@<!-- -->try {
	}
	@<!-- -->catch (id a)
	{
		ret = 1;
                 puts("abc");
	}

	return ret;
}

The following assertion is triggered:

clang: /home/vm/llvm-project/llvm/lib/Target/WebAssembly/WebAssemblyExceptionInfo.cpp:124: void llvm::WebAssemblyExceptionInfo::recalculate(MachineFunction &amp;, MachineDominatorTree &amp;, const MachineDominanceFrontier &amp;): Assertion `EHInfo' failed.

Here is the crash report main-c3884.zip.

You can use emcc with a modified LLVM build by exporting EM_LLVM_ROOT before sourcing emsdk/emsdk_env.sh:

emcc -fobjc-runtime=gnustep-2.2 -fwasm-exceptions -c main.m

or just invoke clang directly:

/home/vm/llvm-build-wasm/bin/clang -target wasm32-unknown-emscripten -mllvm -combiner-global-alias-analysis=false -mllvm -wasm-enable-sjlj -mllvm -wasm-use-legacy-eh=false -mllvm -disable-lsr --sysroot=/home/vm/emsdk/upstream/emscripten/cache/sysroot -DEMSCRIPTEN -fobjc-runtime=gnustep-2.2 -fwasm-exceptions -c main.m

Building libobjc2 and the BlocksRuntime

Building the BlocksRuntime

cmake -DCMAKE_TOOLCHAIN_FILE=$EMSDK/upstream/emscripten/cmake/Modules/Platform/Emscripten.cmake   -DCMAKE_INSTALL_PREFIX=/home/vm/demo-install -DCMAKE_BUILD_TYPE=Debug -B build -G Ninja

Building libobjc2

cmake -DCMAKE_TOOLCHAIN_FILE=$EMSDK/upstream/emscripten/cmake/Modules/Platform/Emscripten.cmake   -DCMAKE_INSTALL_PREFIX=/home/vm/demo-install -DBlocksRuntime_LIBRARIES=/home/vm/demo-install/lib/libBlocksRuntime.a -DBlocksRuntime_INCLUDE_DIR=/home/vm/demo-install/include/BlocksRuntime -DEMBEDDED_BLOCKS_RUNTIME=OFF -DTESTS=OFF  -B build  -DCMAKE_BUILD_TYPE=Debug  -G Ninja

Full diff: https://github.com/llvm/llvm-project/pull/169043.diff

3 Files Affected:

(modified) clang/lib/CodeGen/CGObjCGNU.cpp (+10-4)
(modified) clang/lib/Driver/ToolChains/Clang.cpp (+2-1)
(modified) llvm/lib/MC/WasmObjectWriter.cpp (-3)

diff --git a/clang/lib/CodeGen/CGObjCGNU.cpp b/clang/lib/CodeGen/CGObjCGNU.cpp
index 06643d4bdc211..3b9f9f306829d 100644
--- a/clang/lib/CodeGen/CGObjCGNU.cpp
+++ b/clang/lib/CodeGen/CGObjCGNU.cpp
@@ -179,8 +179,15 @@ class CGObjCGNU : public CGObjCRuntime {
       (R.getVersion() >= VersionTuple(major, minor));
   }
 
-  std::string ManglePublicSymbol(StringRef Name) {
-    return (StringRef(CGM.getTriple().isOSBinFormatCOFF() ? "$_" : "._") + Name).str();
+  const std::string ManglePublicSymbol(StringRef Name) {
+    auto triple = CGM.getTriple();
+
+    // Exported symbols in Emscripten must be a valid Javascript identifier.
+    if (triple.isOSBinFormatCOFF() || triple.isOSBinFormatWasm()) {
+      return (StringRef("$_") + Name).str();
+    } else {
+      return (StringRef("._") + Name).str();
+    }
   }
 
   std::string SymbolForProtocol(Twine Name) {
@@ -4106,8 +4113,7 @@ llvm::Function *CGObjCGNU::ModuleInitFunction() {
   if (!ClassAliases.empty()) {
     llvm::Type *ArgTypes[2] = {PtrTy, PtrToInt8Ty};
     llvm::FunctionType *RegisterAliasTy =
-      llvm::FunctionType::get(Builder.getVoidTy(),
-                              ArgTypes, false);
+        llvm::FunctionType::get(BoolTy, ArgTypes, false);
     llvm::Function *RegisterAlias = llvm::Function::Create(
       RegisterAliasTy,
       llvm::GlobalValue::ExternalWeakLinkage, "class_registerAlias_np",
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
index 30d3e5293a31b..6cbec5e17ae1a 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -8001,7 +8001,8 @@ ObjCRuntime Clang::AddObjCRuntimeArgs(const ArgList &args,
     if ((runtime.getKind() == ObjCRuntime::GNUstep) &&
         (runtime.getVersion() >= VersionTuple(2, 0)))
       if (!getToolChain().getTriple().isOSBinFormatELF() &&
-          !getToolChain().getTriple().isOSBinFormatCOFF()) {
+          !getToolChain().getTriple().isOSBinFormatCOFF() &&
+          !getToolChain().getTriple().isOSBinFormatWasm()) {
         getToolChain().getDriver().Diag(
             diag::err_drv_gnustep_objc_runtime_incompatible_binary)
           << runtime.getVersion().getMajor();
diff --git a/llvm/lib/MC/WasmObjectWriter.cpp b/llvm/lib/MC/WasmObjectWriter.cpp
index 15590b31fd07f..d882146e21b8a 100644
--- a/llvm/lib/MC/WasmObjectWriter.cpp
+++ b/llvm/lib/MC/WasmObjectWriter.cpp
@@ -1794,9 +1794,6 @@ uint64_t WasmObjectWriter::writeOneObject(MCAssembler &Asm,
       Flags |= wasm::WASM_SYMBOL_UNDEFINED;
     if (WS.isNoStrip()) {
       Flags |= wasm::WASM_SYMBOL_NO_STRIP;
-      if (isEmscripten()) {
-        Flags |= wasm::WASM_SYMBOL_EXPORTED;
-      }
     }
     if (WS.hasImportName())
       Flags |= wasm::WASM_SYMBOL_EXPLICIT_NAME;

llvmbot · 2025-11-21T14:13:46Z

@llvm/pr-subscribers-llvm-mc

Author: Hugo Melder (hmelder)

Changes

This pull request adds initial support for compiling Objective-C to WebAssembly. I tested my changes with libobjc2 and the swift-corelibs-blocksruntime.

There are two outstanding issues, which I cannot fix as deeper knowledge of the subsystems is required:

Symbols marked as explicitly hidden in code generation are exported
Clang crashes in SelectionDAG when compiling an Objective-C try/catch block with -fwasm-exceptions

First Issue

Emscripten is processing the generated .wasm file in emscripten.py and checks if all exported symbols are valid javascript identifiers (tools/js_manipulation.py#L104). However, hidden symbols such as .objc_init are intentionally an invalid C identifier.

The core of the problem is that symbols with the WASM_SYMBOL_NO_STRIP attribute are exported when targeting Emscripten (https://reviews.llvm.org/D62542). This attribute is added to the symbol during relocation in WasmObjectWriter::recordRelocation. So we are accidentally exporting a lot of hidden symbols and not only ones generated by ObjC CG...

I'm currently hacking around this by not exporting no-strip symbols. This is the default behaviour for Wasm.

Second Issue

Here is a minimal example that triggers the crash.

#include&lt;stdio.h&gt;

int main(void) {
	int ret = 0;
	@<!-- -->try {
	}
	@<!-- -->catch (id a)
	{
		ret = 1;
                 puts("abc");
	}

	return ret;
}

The following assertion is triggered:

clang: /home/vm/llvm-project/llvm/lib/Target/WebAssembly/WebAssemblyExceptionInfo.cpp:124: void llvm::WebAssemblyExceptionInfo::recalculate(MachineFunction &amp;, MachineDominatorTree &amp;, const MachineDominanceFrontier &amp;): Assertion `EHInfo' failed.

Here is the crash report main-c3884.zip.

You can use emcc with a modified LLVM build by exporting EM_LLVM_ROOT before sourcing emsdk/emsdk_env.sh:

emcc -fobjc-runtime=gnustep-2.2 -fwasm-exceptions -c main.m

or just invoke clang directly:

/home/vm/llvm-build-wasm/bin/clang -target wasm32-unknown-emscripten -mllvm -combiner-global-alias-analysis=false -mllvm -wasm-enable-sjlj -mllvm -wasm-use-legacy-eh=false -mllvm -disable-lsr --sysroot=/home/vm/emsdk/upstream/emscripten/cache/sysroot -DEMSCRIPTEN -fobjc-runtime=gnustep-2.2 -fwasm-exceptions -c main.m

Building libobjc2 and the BlocksRuntime

Building the BlocksRuntime

cmake -DCMAKE_TOOLCHAIN_FILE=$EMSDK/upstream/emscripten/cmake/Modules/Platform/Emscripten.cmake   -DCMAKE_INSTALL_PREFIX=/home/vm/demo-install -DCMAKE_BUILD_TYPE=Debug -B build -G Ninja

Building libobjc2

cmake -DCMAKE_TOOLCHAIN_FILE=$EMSDK/upstream/emscripten/cmake/Modules/Platform/Emscripten.cmake   -DCMAKE_INSTALL_PREFIX=/home/vm/demo-install -DBlocksRuntime_LIBRARIES=/home/vm/demo-install/lib/libBlocksRuntime.a -DBlocksRuntime_INCLUDE_DIR=/home/vm/demo-install/include/BlocksRuntime -DEMBEDDED_BLOCKS_RUNTIME=OFF -DTESTS=OFF  -B build  -DCMAKE_BUILD_TYPE=Debug  -G Ninja

Full diff: https://github.com/llvm/llvm-project/pull/169043.diff

3 Files Affected:

(modified) clang/lib/CodeGen/CGObjCGNU.cpp (+10-4)
(modified) clang/lib/Driver/ToolChains/Clang.cpp (+2-1)
(modified) llvm/lib/MC/WasmObjectWriter.cpp (-3)

diff --git a/clang/lib/CodeGen/CGObjCGNU.cpp b/clang/lib/CodeGen/CGObjCGNU.cpp
index 06643d4bdc211..3b9f9f306829d 100644
--- a/clang/lib/CodeGen/CGObjCGNU.cpp
+++ b/clang/lib/CodeGen/CGObjCGNU.cpp
@@ -179,8 +179,15 @@ class CGObjCGNU : public CGObjCRuntime {
       (R.getVersion() >= VersionTuple(major, minor));
   }
 
-  std::string ManglePublicSymbol(StringRef Name) {
-    return (StringRef(CGM.getTriple().isOSBinFormatCOFF() ? "$_" : "._") + Name).str();
+  const std::string ManglePublicSymbol(StringRef Name) {
+    auto triple = CGM.getTriple();
+
+    // Exported symbols in Emscripten must be a valid Javascript identifier.
+    if (triple.isOSBinFormatCOFF() || triple.isOSBinFormatWasm()) {
+      return (StringRef("$_") + Name).str();
+    } else {
+      return (StringRef("._") + Name).str();
+    }
   }
 
   std::string SymbolForProtocol(Twine Name) {
@@ -4106,8 +4113,7 @@ llvm::Function *CGObjCGNU::ModuleInitFunction() {
   if (!ClassAliases.empty()) {
     llvm::Type *ArgTypes[2] = {PtrTy, PtrToInt8Ty};
     llvm::FunctionType *RegisterAliasTy =
-      llvm::FunctionType::get(Builder.getVoidTy(),
-                              ArgTypes, false);
+        llvm::FunctionType::get(BoolTy, ArgTypes, false);
     llvm::Function *RegisterAlias = llvm::Function::Create(
       RegisterAliasTy,
       llvm::GlobalValue::ExternalWeakLinkage, "class_registerAlias_np",
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
index 30d3e5293a31b..6cbec5e17ae1a 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -8001,7 +8001,8 @@ ObjCRuntime Clang::AddObjCRuntimeArgs(const ArgList &args,
     if ((runtime.getKind() == ObjCRuntime::GNUstep) &&
         (runtime.getVersion() >= VersionTuple(2, 0)))
       if (!getToolChain().getTriple().isOSBinFormatELF() &&
-          !getToolChain().getTriple().isOSBinFormatCOFF()) {
+          !getToolChain().getTriple().isOSBinFormatCOFF() &&
+          !getToolChain().getTriple().isOSBinFormatWasm()) {
         getToolChain().getDriver().Diag(
             diag::err_drv_gnustep_objc_runtime_incompatible_binary)
           << runtime.getVersion().getMajor();
diff --git a/llvm/lib/MC/WasmObjectWriter.cpp b/llvm/lib/MC/WasmObjectWriter.cpp
index 15590b31fd07f..d882146e21b8a 100644
--- a/llvm/lib/MC/WasmObjectWriter.cpp
+++ b/llvm/lib/MC/WasmObjectWriter.cpp
@@ -1794,9 +1794,6 @@ uint64_t WasmObjectWriter::writeOneObject(MCAssembler &Asm,
       Flags |= wasm::WASM_SYMBOL_UNDEFINED;
     if (WS.isNoStrip()) {
       Flags |= wasm::WASM_SYMBOL_NO_STRIP;
-      if (isEmscripten()) {
-        Flags |= wasm::WASM_SYMBOL_EXPORTED;
-      }
     }
     if (WS.hasImportName())
       Flags |= wasm::WASM_SYMBOL_EXPLICIT_NAME;

github-actions · 2025-11-21T14:40:50Z

🐧 Linux x64 Test Results

111606 tests passed
4467 tests skipped

hmelder · 2025-11-28T11:50:24Z

@sunfishcode, I see that you are the original author of https://reviews.llvm.org/D62542. As @dschuff said in the review back then:

I'm hoping we can make that export behavior nicer soon; I find the attribute(used) -> export behavior a bit odd too. Once we drop fastcomp it will be easier to redefine EMSCRIPTEN_KEEPALIVE and other things.

This was in 2019, is this hack still required now that fastcomp is deprecated? Then problem with the current behaviour is that hidden no-strip symbols, added during codegen, are exported.

hmelder · 2025-11-28T11:54:04Z

@davidchisnall the changes in codegen are trivial:

Mangle public symbols with '$' instead of '.' as the latter is not a valid javascript identifier.
Fix the function signature of class_registerAlias_np to return a bool instead of void.

hmelder · 2025-11-28T11:57:47Z

Assuming that the new WASM exception implementation implements the mandatory functions and data structure of the Itanium EH ABI correctly, not much needs to be done to get EH working with libobjc2. I just need to find the root course of the crash...

davidchisnall

The Objective-C bits look fine to me, the MC bit possibly should be a separate PR.

clang/lib/CodeGen/CGObjCGNU.cpp

github-actions · 2025-11-28T13:11:47Z

✅ With the latest revision this PR passed the C/C++ code formatter.

davidchisnall

Code LGTM once clang-format is happy.

We probably should have a test in tests/CodeGenObjC checking that the mangling is correct for WAsm.

clang/lib/CodeGen/CGObjCGNU.cpp

Emscripten requires that exported symbols. See emscripten-core/emscripten#23563.

[clang] Whitelist wasm when targeting GNUstep 2.x

a07542b

llvmbot added backend:WebAssembly clang:driver 'clang' and 'clang++' user-facing binaries. Not 'clang-cl' clang:codegen IR generation bugs: mangling, exceptions, etc. llvm:mc Machine (object) code labels Nov 21, 2025

efriedma-quic requested review from davidchisnall and rjmccall November 21, 2025 18:58

davidchisnall reviewed Nov 28, 2025

View reviewed changes

clang/lib/CodeGen/CGObjCGNU.cpp Outdated Show resolved Hide resolved

hmelder force-pushed the objc-wasm branch from 1b37875 to 3974abf Compare November 28, 2025 13:11

llvmbot added the clang Clang issues not falling into any other category label Nov 28, 2025

hmelder force-pushed the objc-wasm branch from 3974abf to ba4cbb0 Compare November 28, 2025 13:18

hmelder requested a review from davidchisnall November 28, 2025 13:19

davidchisnall reviewed Nov 28, 2025

View reviewed changes

clang/lib/CodeGen/CGObjCGNU.cpp Outdated Show resolved Hide resolved

hmelder force-pushed the objc-wasm branch from ba4cbb0 to 002c4e0 Compare November 28, 2025 16:29

hmelder added 3 commits November 28, 2025 16:39

[CodeGen][ObjC] Mangle public symbols for wasm

0ee0356

Emscripten requires that exported symbols. See emscripten-core/emscripten#23563.

[CodeGen][ObjC] Fix class_registerAlias_np signature

45e7f44

[CodeGen][ObjC] Add WASM symbol mangling test

8ad5d24

davidchisnall approved these changes Nov 28, 2025

View reviewed changes

hmelder force-pushed the objc-wasm branch from 002c4e0 to 8ad5d24 Compare November 28, 2025 16:44

[CodeGen][ObjC] Initial WebAssembly Support for GNUstep #169043

Are you sure you want to change the base?

[CodeGen][ObjC] Initial WebAssembly Support for GNUstep #169043

Conversation

hmelder commented Nov 21, 2025

First Issue

Second Issue

Building libobjc2 and the BlocksRuntime

Building the BlocksRuntime

Building libobjc2

Uh oh!

llvmbot commented Nov 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

First Issue

Second Issue

Building libobjc2 and the BlocksRuntime

Building the BlocksRuntime

Building libobjc2

Uh oh!

llvmbot commented Nov 21, 2025

First Issue

Second Issue

Building libobjc2 and the BlocksRuntime

Building the BlocksRuntime

Building libobjc2

Uh oh!

github-actions bot commented Nov 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🐧 Linux x64 Test Results

Uh oh!

hmelder commented Nov 28, 2025

Uh oh!

hmelder commented Nov 28, 2025

Uh oh!

hmelder commented Nov 28, 2025

Uh oh!

davidchisnall left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

github-actions bot commented Nov 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

davidchisnall left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

llvmbot commented Nov 21, 2025 •

edited

Loading

github-actions bot commented Nov 21, 2025 •

edited

Loading

github-actions bot commented Nov 28, 2025 •

edited

Loading