[NFC] Use llvm::SmallVector to workaround XL compiler problem on AIX

[Xiangling_L]

Summary

LLVM is recommending to use SmallVector (that is, omitting the N), in the absence of a well-motivated choice for the number of inlined elements N.

However, this doesn't work well with XL compiler on AIX since some header(s) aren't properly picked up with it. We need to take a further look into the real issue underneath and fix it in a later patch.

But currently we'd like to use this patch to unblock the build compiler issue first.

Note:
The first build failure was spot here: http://lab.llvm.org:8014/#/builders/126/builds/71

[craig.topper]

Maybe this needs to be addressed in the forward declaration in clang/include/clang/Basic/LLVM.h ?

[Xiangling_L]

In D98265#2614245, @craig.topper wrote:

Maybe this needs to be addressed in the forward declaration in clang/include/clang/Basic/LLVM.h ?

Yeah, adding an additional llvm namespace is a certain another way to workaround this issue. But I think the fix we apply is also a quick way to workaround the build issue. And as I mentioned, we would expect a real fix in a later patch.

[hubert.reinterpretcast]

LGTM; thanks.

clang/lib/CodeGen/CGStmtOpenMP.cpp

static EmittedClosureTy emitCapturedStmtFunc(CodeGenFunction &ParentCGF,
   return {F, CapStruct.getPointer(ParentCGF)};
 }
 
 /// Emit a call to a previously captured closure.
 static llvm::CallInst *
 emitCapturedStmtCall(CodeGenFunction &ParentCGF, EmittedClosureTy Cap,
                      llvm::ArrayRef<llvm::Value *> Args) {
   // Append the closure context to the argument.
-  SmallVector<llvm::Value *> EffectiveArgs;
+  llvm::SmallVector<llvm::Value *> EffectiveArgs;
   EffectiveArgs.reserve(Args.size() + 1);
   llvm::append_range(EffectiveArgs, Args);
   EffectiveArgs.push_back(Cap.second);
 
   return ParentCGF.Builder.CreateCall(Cap.first, EffectiveArgs);
 }
 
 llvm::CanonicalLoopInfo *