Revert "[Coro] [async] Disable inlining in async coroutine splitting (llvm#80904)"

This reverts commit b1ac052.

This commit breaks coroutine splitting for non-swift calling convention
functions. In this example:

```ll
; ModuleID = 'repro.ll'
source_filename = "stdlib/test/runtime/test_llcl.mojo"
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

@0 = internal constant { i32, i32 } { i32 trunc (i64 sub (i64 ptrtoint (ptr @craSH to i64), i64 ptrtoint (ptr getelementptr inbounds ({ i32, i32 }, ptr @0, i32 0, i32 1) to i64)) to i32), i32 64 }

define dso_local void @af_suspend_fn(ptr %0, i64 %1, ptr %2) #0 {
  ret void
}

define dso_local void @craSH(ptr %0) #0 {
  %2 = call token @llvm.coro.id.async(i32 64, i32 8, i32 0, ptr @0)
  %3 = call ptr @llvm.coro.begin(token %2, ptr null)
  %4 = getelementptr inbounds { ptr, { ptr, ptr }, i64, { ptr, i1 }, i64, i64 }, ptr poison, i32 0, i32 0
  %5 = call ptr @llvm.coro.async.resume()
  store ptr %5, ptr %4, align 8
  %6 = call { ptr, ptr, ptr } (i32, ptr, ptr, ...) @llvm.coro.suspend.async.sl_p0p0p0s(i32 0, ptr %5, ptr @ctxt_proj_fn, ptr @af_suspend_fn, ptr poison, i64 -1, ptr poison)
  ret void
}

define dso_local ptr @ctxt_proj_fn(ptr %0) #0 {
  ret ptr %0
}

; Function Attrs: nomerge nounwind
declare { ptr, ptr, ptr } @llvm.coro.suspend.async.sl_p0p0p0s(i32, ptr, ptr, ...) #1

; Function Attrs: nounwind
declare token @llvm.coro.id.async(i32, i32, i32, ptr) #2

; Function Attrs: nounwind
declare ptr @llvm.coro.begin(token, ptr writeonly) #2

; Function Attrs: nomerge nounwind
declare ptr @llvm.coro.async.resume() #1

attributes #0 = { "target-features"="+adx,+aes,+avx,+avx2,+bmi,+bmi2,+clflushopt,+clwb,+clzero,+crc32,+cx16,+cx8,+f16c,+fma,+fsgsbase,+fxsr,+invpcid,+lzcnt,+mmx,+movbe,+mwaitx,+pclmul,+pku,+popcnt,+prfchw,+rdpid,+rdpru,+rdrnd,+rdseed,+sahf,+sha,+sse,+sse2,+sse3,+sse4.1,+sse4.2,+sse4a,+ssse3,+vaes,+vpclmulqdq,+wbnoinvd,+x87,+xsave,+xsavec,+xsaveopt,+xsaves" }
attributes #1 = { nomerge nounwind }
attributes #2 = { nounwind }
```

This verifier crashes after the `coro-split` pass with

```
cannot guarantee tail call due to mismatched parameter counts
  musttail call void @af_suspend_fn(ptr poison, i64 -1, ptr poison)
LLVM ERROR: Broken function
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace.
Stack dump:
0.      Program arguments: opt ../../../reduced.ll -O0
 #0 0x00007f1d89645c0e __interceptor_backtrace.part.0 /build/gcc-11-XeT9lY/gcc-11-11.4.0/build/x86_64-linux-gnu/libsanitizer/asan/../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4193:28
 #1 0x0000556d94d254f7 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Support/Unix/Signals.inc:723:22
 #2 0x0000556d94d19a2f llvm::sys::RunSignalHandlers() /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Support/Signals.cpp:105:20
 #3 0x0000556d94d1aa42 SignalHandler(int) /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Support/Unix/Signals.inc:371:36
 llvm#4 0x00007f1d88e42520 (/lib/x86_64-linux-gnu/libc.so.6+0x42520)
 llvm#5 0x00007f1d88e969fc __pthread_kill_implementation ./nptl/pthread_kill.c:44:76
 llvm#6 0x00007f1d88e969fc __pthread_kill_internal ./nptl/pthread_kill.c:78:10
 llvm#7 0x00007f1d88e969fc pthread_kill ./nptl/pthread_kill.c:89:10
 llvm#8 0x00007f1d88e42476 gsignal ./signal/../sysdeps/posix/raise.c:27:6
 llvm#9 0x00007f1d88e287f3 abort ./stdlib/abort.c:81:7
 llvm#10 0x0000556d8944be01 std::vector<llvm::json::Value, std::allocator<llvm::json::Value>>::size() const /usr/include/c++/11/bits/stl_vector.h:919:40
 llvm#11 0x0000556d8944be01 bool std::operator==<llvm::json::Value, std::allocator<llvm::json::Value>>(std::vector<llvm::json::Value, std::allocator<llvm::json::Value>> const&, std::vector<llvm::json::Value, std::allocator<llvm::json::Value>> const&) /usr/include/c++/11/bits/stl_vector.h:1893:23
 llvm#12 0x0000556d8944be01 llvm::json::operator==(llvm::json::Array const&, llvm::json::Array const&) /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/Support/JSON.h:572:69
 llvm#13 0x0000556d8944be01 llvm::json::operator==(llvm::json::Value const&, llvm::json::Value const&) (.cold) /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Support/JSON.cpp:204:28
 llvm#14 0x0000556d949ed2bd llvm::report_fatal_error(char const*, bool) /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Support/ErrorHandling.cpp:82:70
 llvm#15 0x0000556d8e37e876 llvm::SmallVectorBase<unsigned int>::size() const /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/ADT/SmallVector.h:91:32
 llvm#16 0x0000556d8e37e876 llvm::SmallVectorTemplateCommon<llvm::DiagnosticInfoOptimizationBase::Argument, void>::end() /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/ADT/SmallVector.h:282:41
 llvm#17 0x0000556d8e37e876 llvm::SmallVector<llvm::DiagnosticInfoOptimizationBase::Argument, 4u>::~SmallVector() /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/ADT/SmallVector.h:1215:24
 llvm#18 0x0000556d8e37e876 llvm::DiagnosticInfoOptimizationBase::~DiagnosticInfoOptimizationBase() /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/DiagnosticInfo.h:413:7
 llvm#19 0x0000556d8e37e876 llvm::DiagnosticInfoIROptimization::~DiagnosticInfoIROptimization() /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/DiagnosticInfo.h:622:7
 llvm#20 0x0000556d8e37e876 llvm::OptimizationRemark::~OptimizationRemark() /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/DiagnosticInfo.h:689:7
 llvm#21 0x0000556d8e37e876 operator() /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Transforms/Coroutines/CoroSplit.cpp:2213:14
 llvm#22 0x0000556d8e37e876 emit<llvm::CoroSplitPass::run(llvm::LazyCallGraph::SCC&, llvm::CGSCCAnalysisManager&, llvm::LazyCallGraph&, llvm::CGSCCUpdateResult&)::<lambda()> > /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/Analysis/OptimizationRemarkEmitter.h:83:12
 llvm#23 0x0000556d8e37e876 llvm::CoroSplitPass::run(llvm::LazyCallGraph::SCC&, llvm::AnalysisManager<llvm::LazyCallGraph::SCC, llvm::LazyCallGraph&>&, llvm::LazyCallGraph&, llvm::CGSCCUpdateResult&) /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Transforms/Coroutines/CoroSplit.cpp:2212:13
 llvm#24 0x0000556d8c36ecb1 llvm::detail::PassModel<llvm::LazyCallGraph::SCC, llvm::CoroSplitPass, llvm::AnalysisManager<llvm::LazyCallGraph::SCC, llvm::LazyCallGraph&>, llvm::LazyCallGraph&, llvm::CGSCCUpdateResult&>::run(llvm::LazyCallGraph::SCC&, llvm::AnalysisManager<llvm::LazyCallGraph::SCC, llvm::LazyCallGraph&>&, llvm::LazyCallGraph&, llvm::CGSCCUpdateResult&) /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/PassManagerInternal.h:91:3
 llvm#25 0x0000556d91c1a84f llvm::PassManager<llvm::LazyCallGraph::SCC, llvm::AnalysisManager<llvm::LazyCallGraph::SCC, llvm::LazyCallGraph&>, llvm::LazyCallGraph&, llvm::CGSCCUpdateResult&>::run(llvm::LazyCallGraph::SCC&, llvm::AnalysisManager<llvm::LazyCallGraph::SCC, llvm::LazyCallGraph&>&, llvm::LazyCallGraph&, llvm::CGSCCUpdateResult&) /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Analysis/CGSCCPassManager.cpp:90:12
 llvm#26 0x0000556d8c3690d1 llvm::detail::PassModel<llvm::LazyCallGraph::SCC, llvm::PassManager<llvm::LazyCallGraph::SCC, llvm::AnalysisManager<llvm::LazyCallGraph::SCC, llvm::LazyCallGraph&>, llvm::LazyCallGraph&, llvm::CGSCCUpdateResult&>, llvm::AnalysisManager<llvm::LazyCallGraph::SCC, llvm::LazyCallGraph&>, llvm::LazyCallGraph&, llvm::CGSCCUpdateResult&>::run(llvm::LazyCallGraph::SCC&, llvm::AnalysisManager<llvm::LazyCallGraph::SCC, llvm::LazyCallGraph&>&, llvm::LazyCallGraph&, llvm::CGSCCUpdateResult&) /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/PassManagerInternal.h:91:3
 llvm#27 0x0000556d91c2162d llvm::ModuleToPostOrderCGSCCPassAdaptor::run(llvm::Module&, llvm::AnalysisManager<llvm::Module>&) /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Analysis/CGSCCPassManager.cpp:278:18
 llvm#28 0x0000556d8c369035 llvm::detail::PassModel<llvm::Module, llvm::ModuleToPostOrderCGSCCPassAdaptor, llvm::AnalysisManager<llvm::Module>>::run(llvm::Module&, llvm::AnalysisManager<llvm::Module>&) /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/PassManagerInternal.h:91:3
 llvm#29 0x0000556d9457abc5 llvm::PassManager<llvm::Module, llvm::AnalysisManager<llvm::Module>>::run(llvm::Module&, llvm::AnalysisManager<llvm::Module>&) /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/PassManager.h:247:20
 llvm#30 0x0000556d8e30979e llvm::CoroConditionalWrapper::run(llvm::Module&, llvm::AnalysisManager<llvm::Module>&) /home/ubuntu/modular/third-party/llvm-project/llvm/lib/Transforms/Coroutines/CoroConditionalWrapper.cpp:19:74
 llvm#31 0x0000556d8c365755 llvm::detail::PassModel<llvm::Module, llvm::CoroConditionalWrapper, llvm::AnalysisManager<llvm::Module>>::run(llvm::Module&, llvm::AnalysisManager<llvm::Module>&) /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/PassManagerInternal.h:91:3
 llvm#32 0x0000556d9457abc5 llvm::PassManager<llvm::Module, llvm::AnalysisManager<llvm::Module>>::run(llvm::Module&, llvm::AnalysisManager<llvm::Module>&) /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/PassManager.h:247:20
 llvm#33 0x0000556d89818556 llvm::SmallPtrSetImplBase::isSmall() const /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/ADT/SmallPtrSet.h:196:33
 llvm#34 0x0000556d89818556 llvm::SmallPtrSetImplBase::~SmallPtrSetImplBase() /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/ADT/SmallPtrSet.h:84:17
 llvm#35 0x0000556d89818556 llvm::SmallPtrSetImpl<llvm::AnalysisKey*>::~SmallPtrSetImpl() /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/ADT/SmallPtrSet.h:321:7
 llvm#36 0x0000556d89818556 llvm::SmallPtrSet<llvm::AnalysisKey*, 2u>::~SmallPtrSet() /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/ADT/SmallPtrSet.h:427:7
 llvm#37 0x0000556d89818556 llvm::PreservedAnalyses::~PreservedAnalyses() /home/ubuntu/modular/third-party/llvm-project/llvm/include/llvm/IR/Analysis.h:109:7
 llvm#38 0x0000556d89818556 llvm::runPassPipeline(llvm::StringRef, llvm::Module&, llvm::TargetMachine*, llvm::TargetLibraryInfoImpl*, llvm::ToolOutputFile*, llvm::ToolOutputFile*, llvm::ToolOutputFile*, llvm::StringRef, llvm::ArrayRef<llvm::PassPlugin>, llvm::ArrayRef<std::function<void (llvm::PassBuilder&)>>, llvm::opt_tool::OutputKind, llvm::opt_tool::VerifierKind, bool, bool, bool, bool, bool, bool, bool) /home/ubuntu/modular/third-party/llvm-project/llvm/tools/opt/NewPMDriver.cpp:532:10
 llvm#39 0x0000556d897e3939 optMain /home/ubuntu/modular/third-party/llvm-project/llvm/tools/opt/optdriver.cpp:737:27
 llvm#40 0x0000556d89455461 main /home/ubuntu/modular/third-party/llvm-project/llvm/tools/opt/opt.cpp:25:33
 llvm#41 0x00007f1d88e29d90 __libc_start_call_main ./csu/../sysdeps/nptl/libc_start_call_main.h:58:16
 llvm#42 0x00007f1d88e29e40 call_init ./csu/../csu/libc-start.c:128:20
 llvm#43 0x00007f1d88e29e40 __libc_start_main ./csu/../csu/libc-start.c:379:5
 llvm#44 0x0000556d897b6335 _start (/home/ubuntu/modular/.derived/third-party/llvm-project/build-relwithdebinfo-asan/bin/opt+0x150c335)
Aborted (core dumped)

Author: Mogball

llvm/lib/Transforms/Coroutines/CoroSplit.cpp

@@ -117,8 +117,8 @@ class CoroCloner {
   /// Create a cloner for a switch lowering.
   CoroCloner(Function &OrigF, const Twine &Suffix, coro::Shape &Shape,
              Kind FKind)
-    : OrigF(OrigF), NewF(nullptr), Suffix(Suffix), Shape(Shape),
-      FKind(FKind), Builder(OrigF.getContext()) {
+      : OrigF(OrigF), NewF(nullptr), Suffix(Suffix), Shape(Shape), FKind(FKind),
+        Builder(OrigF.getContext()) {
     assert(Shape.ABI == coro::ABI::Switch);
   }
 
@@ -170,8 +170,7 @@ class CoroCloner {
 static void maybeFreeRetconStorage(IRBuilder<> &Builder,
                                    const coro::Shape &Shape, Value *FramePtr,
                                    CallGraph *CG) {
-  assert(Shape.ABI == coro::ABI::Retcon ||
-         Shape.ABI == coro::ABI::RetconOnce);
+  assert(Shape.ABI == coro::ABI::Retcon || Shape.ABI == coro::ABI::RetconOnce);
   if (Shape.RetconLowering.IsFrameInlineInStorage)
     return;
 
@@ -208,12 +207,17 @@ static bool replaceCoroEndAsync(AnyCoroEndInst *End) {
   // Insert the return instruction.
   Builder.SetInsertPoint(End);
   Builder.CreateRetVoid();
+  InlineFunctionInfo FnInfo;
 
   // Remove the rest of the block, by splitting it into an unreachable block.
   auto *BB = End->getParent();
   BB->splitBasicBlock(End);
   BB->getTerminator()->eraseFromParent();
 
+  auto InlineRes = InlineFunction(*MustTailCall, FnInfo);
+  assert(InlineRes.isSuccess() && "Expected inlining to succeed");
+  (void)InlineRes;
+
   // We have cleaned up the coro.end block above.
   return false;
 }
@@ -264,7 +268,7 @@ static void replaceFallthroughCoroEnd(AnyCoroEndInst *End,
 
     if (auto *RetStructTy = dyn_cast<StructType>(RetTy)) {
       assert(RetStructTy->getNumElements() == NumReturns &&
-           "numbers of returns should match resume function singature");
+             "numbers of returns should match resume function singature");
       Value *ReturnValue = UndefValue::get(RetStructTy);
       unsigned Idx = 0;
       for (Value *RetValEl : CoroResults->return_values())
@@ -277,7 +281,8 @@ static void replaceFallthroughCoroEnd(AnyCoroEndInst *End,
       assert(NumReturns == 1);
       Builder.CreateRet(*CoroResults->retval_begin());
     }
-    CoroResults->replaceAllUsesWith(ConstantTokenNone::get(CoroResults->getContext()));
+    CoroResults->replaceAllUsesWith(
+        ConstantTokenNone::get(CoroResults->getContext()));
     CoroResults->eraseFromParent();
     break;
   }
@@ -291,7 +296,7 @@ static void replaceFallthroughCoroEnd(AnyCoroEndInst *End,
     auto RetTy = Shape.getResumeFunctionType()->getReturnType();
     auto RetStructTy = dyn_cast<StructType>(RetTy);
     PointerType *ContinuationTy =
-      cast<PointerType>(RetStructTy ? RetStructTy->getElementType(0) : RetTy);
+        cast<PointerType>(RetStructTy ? RetStructTy->getElementType(0) : RetTy);
 
     Value *ReturnValue = ConstantPointerNull::get(ContinuationTy);
     if (RetStructTy) {
@@ -480,11 +485,12 @@ void CoroCloner::replaceRetconOrAsyncSuspendUses() {
          Shape.ABI == coro::ABI::Async);
 
   auto NewS = VMap[ActiveSuspend];
-  if (NewS->use_empty()) return;
+  if (NewS->use_empty())
+    return;
 
   // Copy out all the continuation arguments after the buffer pointer into
   // an easily-indexed data structure for convenience.
-  SmallVector<Value*, 8> Args;
+  SmallVector<Value *, 8> Args;
   // The async ABI includes all arguments -- including the first argument.
   bool IsAsyncABI = Shape.ABI == coro::ABI::Async;
   for (auto I = IsAsyncABI ? NewF->arg_begin() : std::next(NewF->arg_begin()),
@@ -511,7 +517,8 @@ void CoroCloner::replaceRetconOrAsyncSuspendUses() {
   }
 
   // If we have no remaining uses, we're done.
-  if (NewS->use_empty()) return;
+  if (NewS->use_empty())
+    return;
 
   // Otherwise, we need to create an aggregate.
   Value *Agg = PoisonValue::get(NewS->getType());
@@ -549,7 +556,8 @@ void CoroCloner::replaceCoroSuspends() {
 
   for (AnyCoroSuspendInst *CS : Shape.CoroSuspends) {
     // The active suspend was handled earlier.
-    if (CS == ActiveSuspend) continue;
+    if (CS == ActiveSuspend)
+      continue;
 
     auto *MappedCS = cast<AnyCoroSuspendInst>(VMap[CS]);
     MappedCS->replaceAllUsesWith(SuspendResult);
@@ -707,7 +715,7 @@ void CoroCloner::replaceEntryBlock() {
     // In switch-lowering, we built a resume-entry block in the original
     // function.  Make the entry block branch to this.
     auto *SwitchBB =
-      cast<BasicBlock>(VMap[Shape.SwitchLowering.ResumeEntryBlock]);
+        cast<BasicBlock>(VMap[Shape.SwitchLowering.ResumeEntryBlock]);
     Builder.CreateBr(SwitchBB);
     break;
   }
@@ -1055,7 +1063,7 @@ void CoroCloner::create() {
   // to suppress deallocation code.
   if (Shape.ABI == coro::ABI::Switch)
     coro::replaceCoroFree(cast<CoroIdInst>(VMap[Shape.CoroBegin->getId()]),
-                          /*Elide=*/ FKind == CoroCloner::Kind::SwitchCleanup);
+                          /*Elide=*/FKind == CoroCloner::Kind::SwitchCleanup);
 }
 
 static void updateAsyncFuncPointerContextSize(coro::Shape &Shape) {
@@ -1842,8 +1850,13 @@ static void splitAsyncCoroutine(Function &F, coro::Shape &Shape,
     SmallVector<Value *, 8> Args(Suspend->args());
     auto FnArgs = ArrayRef<Value *>(Args).drop_front(
         CoroSuspendAsyncInst::MustTailCallFuncArg + 1);
-    coro::createMustTailCall(Suspend->getDebugLoc(), Fn, TTI, FnArgs, Builder);
+    auto *TailCall = coro::createMustTailCall(Suspend->getDebugLoc(), Fn, TTI,
+                                              FnArgs, Builder);
     Builder.CreateRetVoid();
+    InlineFunctionInfo FnInfo;
+    auto InlineRes = InlineFunction(*TailCall, FnInfo);
+    assert(InlineRes.isSuccess() && "Expected inlining to succeed");
+    (void)InlineRes;
 
     // Replace the lvm.coro.async.resume intrisic call.
     replaceAsyncResumeFunction(Suspend, Continuation);
@@ -1860,8 +1873,7 @@ static void splitAsyncCoroutine(Function &F, coro::Shape &Shape,
 
 static void splitRetconCoroutine(Function &F, coro::Shape &Shape,
                                  SmallVectorImpl<Function *> &Clones) {
-  assert(Shape.ABI == coro::ABI::Retcon ||
-         Shape.ABI == coro::ABI::RetconOnce);
+  assert(Shape.ABI == coro::ABI::Retcon || Shape.ABI == coro::ABI::RetconOnce);
   assert(Clones.empty());
 
   // Reset various things that the optimizer might have decided it
@@ -1887,7 +1899,7 @@ static void splitRetconCoroutine(Function &F, coro::Shape &Shape,
     // FIXME: pass the required alignment
     RawFramePtr = Shape.emitAlloc(Builder, Builder.getInt64(Size), nullptr);
     RawFramePtr =
-      Builder.CreateBitCast(RawFramePtr, Shape.CoroBegin->getType());
+        Builder.CreateBitCast(RawFramePtr, Shape.CoroBegin->getType());
 
     // Stash the allocated frame pointer in the continuation storage.
     Builder.CreateStore(RawFramePtr, Id->getStorage());
@@ -1927,8 +1939,8 @@ static void splitRetconCoroutine(Function &F, coro::Shape &Shape,
     // Create the unified return block.
     if (!ReturnBB) {
       // Place it before the first suspend.
-      ReturnBB = BasicBlock::Create(F.getContext(), "coro.return", &F,
-                                    NewSuspendBB);
+      ReturnBB =
+          BasicBlock::Create(F.getContext(), "coro.return", &F, NewSuspendBB);
       Shape.RetconLowering.ReturnBlock = ReturnBB;
 
       IRBuilder<> Builder(ReturnBB);
@@ -1942,8 +1954,8 @@ static void splitRetconCoroutine(Function &F, coro::Shape &Shape,
 
       // Next, all the directly-yielded values.
       for (auto *ResultTy : Shape.getRetconResultTypes())
-        ReturnPHIs.push_back(Builder.CreatePHI(ResultTy,
-                                               Shape.CoroSuspends.size()));
+        ReturnPHIs.push_back(
+            Builder.CreatePHI(ResultTy, Shape.CoroSuspends.size()));
 
       // Build the return value.
       auto RetTy = F.getReturnType();
@@ -1952,9 +1964,9 @@ static void splitRetconCoroutine(Function &F, coro::Shape &Shape,
       // We can't rely on the types matching up because that type would
       // have to be infinite.
       auto CastedContinuationTy =
-        (ReturnPHIs.size() == 1 ? RetTy : RetTy->getStructElementType(0));
+          (ReturnPHIs.size() == 1 ? RetTy : RetTy->getStructElementType(0));
       auto *CastedContinuation =
-        Builder.CreateBitCast(ReturnPHIs[0], CastedContinuationTy);
+          Builder.CreateBitCast(ReturnPHIs[0], CastedContinuationTy);
 
       Value *RetV;
       if (ReturnPHIs.size() == 1) {
@@ -1988,17 +2000,18 @@ static void splitRetconCoroutine(Function &F, coro::Shape &Shape,
 }
 
 namespace {
-  class PrettyStackTraceFunction : public PrettyStackTraceEntry {
-    Function &F;
-  public:
-    PrettyStackTraceFunction(Function &F) : F(F) {}
-    void print(raw_ostream &OS) const override {
-      OS << "While splitting coroutine ";
-      F.printAsOperand(OS, /*print type*/ false, F.getParent());
-      OS << "\n";
-    }
-  };
-}
+class PrettyStackTraceFunction : public PrettyStackTraceEntry {
+  Function &F;
+
+public:
+  PrettyStackTraceFunction(Function &F) : F(F) {}
+  void print(raw_ostream &OS) const override {
+    OS << "While splitting coroutine ";
+    F.printAsOperand(OS, /*print type*/ false, F.getParent());
+    OS << "\n";
+  }
+};
+} // namespace
 
 static coro::Shape
 splitCoroutine(Function &F, SmallVectorImpl<Function *> &Clones,

llvm/test/Transforms/Coroutines/coro-async-addr-lifetime-infinite-loop-bug.ll

@@ -22,8 +22,8 @@ declare void @my_other_async_function(ptr %async.ctxt)
      i32 128    ; Initial async context size without space for frame
 }>
 
-define swifttailcc void @my_other_async_function_fp.apply(ptr %fnPtr, ptr %async.ctxt) alwaysinline {
-  tail call swifttailcc void %fnPtr(ptr %async.ctxt)
+define swiftcc void @my_other_async_function_fp.apply(ptr %fnPtr, ptr %async.ctxt) {
+  tail call swiftcc void %fnPtr(ptr %async.ctxt)
   ret void
 }
 
@@ -37,12 +37,12 @@ entry:
 
 ; The address of alloca escapes but the analysis based on lifetimes fails to see
 ; that it can't localize this alloca.
-; CHECK: define swifttailcc void @my_async_function(ptr swiftasync %async.ctxt) {
+; CHECK: define swiftcc void @my_async_function(ptr swiftasync %async.ctxt) {
 ; CHECK: entry:
 ; CHECK-NOT: ret
 ; CHECK-NOT:   [[ESCAPED_ADDR:%.*]] = alloca i64, align 8
 ; CHECK: ret
-define swifttailcc void @my_async_function(ptr swiftasync %async.ctxt) {
+define swiftcc void @my_async_function(ptr swiftasync %async.ctxt) {
 entry:
   %escaped_addr = alloca i64
 

llvm/test/Transforms/Coroutines/coro-async-addr-lifetime-start-bug.ll

@@ -22,8 +22,8 @@ declare void @my_other_async_function(ptr %async.ctxt)
      i32 128    ; Initial async context size without space for frame
 }>
 
-define swifttailcc void @my_other_async_function_fp.apply(ptr %fnPtr, ptr %async.ctxt) alwaysinline {
-  tail call swifttailcc void %fnPtr(ptr %async.ctxt)
+define swiftcc void @my_other_async_function_fp.apply(ptr %fnPtr, ptr %async.ctxt) {
+  tail call swiftcc void %fnPtr(ptr %async.ctxt)
   ret void
 }
 
@@ -36,7 +36,7 @@ entry:
   ret ptr %resume_ctxt
 }
 
-define swifttailcc void @my_async_function(ptr swiftasync %async.ctxt) {
+define swiftcc void @my_async_function(ptr swiftasync %async.ctxt) {
 entry:
   %escaped_addr = alloca i64
 

llvm/test/Transforms/Coroutines/coro-async-dyn-align.ll

@@ -33,20 +33,20 @@ declare swiftcc void @asyncReturn(ptr)
 declare swiftcc void @asyncSuspend(ptr)
 declare {ptr} @llvm.coro.suspend.async(i32, ptr, ptr, ...)
 
-define swifttailcc void @my_async_function.my_other_async_function_fp.apply(ptr %fnPtr, ptr %async.ctxt) alwaysinline {
-  musttail call swifttailcc void %fnPtr(ptr %async.ctxt)
+define swiftcc void @my_async_function.my_other_async_function_fp.apply(ptr %fnPtr, ptr %async.ctxt) {
+  tail call swiftcc void %fnPtr(ptr %async.ctxt)
   ret void
 }
 
-define ptr @__swift_async_resume_project_context(ptr %ctxt) alwaysinline {
+define ptr @__swift_async_resume_project_context(ptr %ctxt) {
 entry:
   %resume_ctxt = load ptr, ptr %ctxt, align 8
   ret ptr %resume_ctxt
 }
 
 
 ; CHECK: %my_async_function.Frame = type { i64, [48 x i8], i64, i64, [16 x i8], ptr, i64, ptr }
-; CHECK: define swifttailcc void @my_async_function
+; CHECK: define swiftcc void @my_async_function
 ; CHECK:  [[T0:%.*]] = getelementptr inbounds %my_async_function.Frame, ptr %async.ctx.frameptr, i32 0, i32 3
 ; CHECK:  [[T1:%.*]] = ptrtoint ptr [[T0]] to i64
 ; CHECK:  [[T2:%.*]] = add i64 [[T1]], 31
@@ -60,7 +60,7 @@ entry:
 ; CHECK:  store i64 2, ptr [[T4]]
 ; CHECK:  store i64 3, ptr [[T9]]
 
-define swifttailcc void @my_async_function(ptr swiftasync %async.ctxt) presplitcoroutine {
+define swiftcc void @my_async_function(ptr swiftasync %async.ctxt) presplitcoroutine {
 entry:
   %tmp = alloca i64, align 8
   %tmp2 = alloca i64, align 16