[WebAssembly] Make SjLj lowering globals thread-local

[tlively]

Summary

Emscripten's longjump and exception mechanism depends on two global variables,
__THREW__ and __threwValue, which are changed to be defined as thread-local
in https://github.com/emscripten-core/emscripten/pull/12056. This patch updates
the corresponding code in the WebAssembly backend to properly declare these
globals as thread-local as well.

[kripken]

Just to be sure, is it ok to do this even when atomics are not enabled?

[sbc100]

Awesome thanks!

llvm/lib/Target/WebAssembly/WebAssemblyLowerEmscriptenEHSjLj.cpp

322

I'm surprised this is the cleanest way to do make the global as TLS... I still find lambdas a little harder to parse that straight line code. It kind of makes me what to write a little named functions called makeTLSGlobal to make this easier tfor me to read, but maybe thats just me.

[tlively]

In D88262#2293764, @kripken wrote:

Just to be sure, is it ok to do this even when atomics are not enabled?

Yes, because I moved the CoalesceFeaturesAndStripAtomics and LowerAtomics passes to run after this changed pass.

[tlively]

llvm/lib/Target/WebAssembly/WebAssemblyLowerEmscriptenEHSjLj.cpp

322

Yeah, although I think the readability problem is mostly due to the huge number of arguments the GlobalVariable constructor takes. It would be nice if they had some more specific factory methods or constructor overloads.

[tlively]

@kripken I'm going to land this. Will you take care of the combined roll with https://github.com/emscripten-core/emscripten/pull/12056?

[kripken]

I can do a roll if needed, sure. (Is one needed? Let's see on the bots I guess...)

llvm/lib/Target/WebAssembly/WebAssemblyLowerEmscriptenEHSjLj.cpp

static bool canThrow(const Value *V) {
   // not a function, so an indirect call - can throw, we can't tell
   return true;
 }
 
 // Get a global variable with the given name.  If it doesn't exist declare it,
 // which will generate an import and asssumes that it will exist at link time.
 static GlobalVariable *getGlobalVariableI32(Module &M, IRBuilder<> &IRB,
                                             const char *Name) {
-
-  auto *GV =
-      dyn_cast<GlobalVariable>(M.getOrInsertGlobal(Name, IRB.getInt32Ty()));
+  auto Int32Ty = IRB.getInt32Ty();
+  auto *GV = dyn_cast<GlobalVariable>(M.getOrInsertGlobal(Name, Int32Ty, [&]() {
+    return new GlobalVariable(M, Int32Ty, false,
+                              GlobalVariable::ExternalLinkage, nullptr, Name,
+                              nullptr, GlobalValue::LocalExecTLSModel);
+  }));
   if (!GV)
     report_fatal_error(Twine("unable to create global: ") + Name);
 
   return GV;
 }
 
 // Simple function name mangler.
 // This function simply takes LLVM's string representation of parameter types

llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp

 }
 
 //===----------------------------------------------------------------------===//
 // The following functions are called from lib/CodeGen/Passes.cpp to modify
 // the CodeGen pass sequence.
 //===----------------------------------------------------------------------===//
 
 void WebAssemblyPassConfig::addIRPasses() {
-  // Runs LowerAtomicPass if necessary
-  addPass(new CoalesceFeaturesAndStripAtomics(&getWebAssemblyTargetMachine()));
-
-  // This is a no-op if atomics are not used in the module
-  addPass(createAtomicExpandPass());
-
   // Add signatures to prototype-less function declarations
   addPass(createWebAssemblyAddMissingPrototypes());
 
   // Lower .llvm.global_dtors into .llvm_global_ctors with __cxa_atexit calls.
   addPass(createWebAssemblyLowerGlobalDtors());
 
   // Fix function bitcasts, as WebAssembly requires caller and callee signatures
   // to match.
void WebAssemblyPassConfig::addIRPasses() {
   // Handle exceptions and setjmp/longjmp if enabled.
   if (EnableEmException || EnableEmSjLj)
     addPass(createWebAssemblyLowerEmscriptenEHSjLj(EnableEmException,
                                                    EnableEmSjLj));
 
   // Expand indirectbr instructions to switches.
   addPass(createIndirectBrExpandPass());
 
+  // Lower atomics and TLS if necessary
+  addPass(new CoalesceFeaturesAndStripAtomics(&getWebAssemblyTargetMachine()));
+
+  // This is a no-op if atomics are not used in the module
+  addPass(createAtomicExpandPass());
+
   TargetPassConfig::addIRPasses();
 }
 
 bool WebAssemblyPassConfig::addInstSelector() {
   (void)TargetPassConfig::addInstSelector();
   addPass(
       createWebAssemblyISelDag(getWebAssemblyTargetMachine(), getOptLevel()));
   // Run the argument-move pass immediately after the ScheduleDAG scheduler

llvm/test/CodeGen/WebAssembly/lower-em-exceptions.ll

 ; RUN: opt < %s -wasm-lower-em-ehsjlj -S | FileCheck %s
 
 target datalayout = "e-m:e-p:32:32-i64:64-n32:64-S128"
 target triple = "wasm32-unknown-unknown"
 
 @_ZTIi = external constant i8*
 @_ZTIc = external constant i8*
-; CHECK-DAG: __THREW__ = external global i32
-; CHECK-DAG: __threwValue = external global i32
+; CHECK-DAG: __THREW__ = external thread_local(localexec) global i32
+; CHECK-DAG: __threwValue = external thread_local(localexec) global i32
 
 ; Test invoke instruction with clauses (try-catch block)
 define void @clause() personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) {
 ; CHECK-LABEL: @clause(
 entry:
   invoke void @foo(i32 3)
           to label %invoke.cont unwind label %lpad
 ; CHECK: entry:

llvm/test/CodeGen/WebAssembly/lower-em-sjlj.ll

 ; RUN: opt < %s -wasm-lower-em-ehsjlj -S | FileCheck %s
 
 target datalayout = "e-m:e-p:32:32-i64:64-n32:64-S128"
 target triple = "wasm32-unknown-unknown"
 
 %struct.__jmp_buf_tag = type { [6 x i32], i32, [32 x i32] }
 
 @global_var = global i32 0, align 4
-; CHECK-DAG: __THREW__ = external global i32
-; CHECK-DAG: __threwValue = external global i32
+; CHECK-DAG: __THREW__ = external thread_local(localexec) global i32
+; CHECK-DAG: __threwValue = external thread_local(localexec) global i32
 
 ; Test a simple setjmp - longjmp sequence
 define void @setjmp_longjmp() {
 ; CHECK-LABEL: @setjmp_longjmp
 entry:
   %buf = alloca [1 x %struct.__jmp_buf_tag], align 16
   %arraydecay = getelementptr inbounds [1 x %struct.__jmp_buf_tag], [1 x %struct.__jmp_buf_tag]* %buf, i32 0, i32 0
   %call = call i32 @setjmp(%struct.__jmp_buf_tag* %arraydecay) #0