[clang] Check AuxTarget exists when creating target in CompilerInstance.

[oToToT]

Summary

D97493 separate target creation out to a single function CompilerInstance::createTarget. However, it would overwrite AuxTarget even if it has been set.
As @kadircet recommended in D98128, this patch check the existence of AuxTarget and not overwrite it when it has been set.

[kadircet]

thanks, lgtm!

clang/lib/Frontend/CompilerInstance.cpp

109

can we just do !getAuxTarget() and not introduce hasAuxTarget() ? i don't think the new api in the interface provides much value.

[oToToT]

clang/lib/Frontend/CompilerInstance.cpp

109

After some thinking, I’m wondering might it be more flexible if we add an option to let the user determine whether overwrite AuxTarget by themselves?

Like changing the prototype to
bool CompilerInstance::createTarget(bool OverwriteAuxTarget = true);
and simply check if (OverwriteAuxTarget && ... here.

[kadircet]

clang/lib/Frontend/CompilerInstance.cpp

109

i think it is better to let the clients clear the target before hand via setAuxTarget(nullptr) (or not set it at all), then confuse them with (defaulted) extra options when calling some function.

[oToToT]

Simply use !getAuxTarget() to check the existence of AuxTarget.

clang/lib/Frontend/CompilerInstance.cpp

 
 bool CompilerInstance::createTarget() {
   // Create the target instance.
   setTarget(TargetInfo::CreateTargetInfo(getDiagnostics(),
                                          getInvocation().TargetOpts));
   if (!hasTarget())
     return false;
 
-  // Create TargetInfo for the other side of CUDA/OpenMP/SYCL compilation.
-  if ((getLangOpts().CUDA || getLangOpts().OpenMPIsDevice ||
+  // Check whether AuxTarget exists, if not, then create TargetInfo for the
+  // other side of CUDA/OpenMP/SYCL compilation.
+  if (!getAuxTarget() &&
+      (getLangOpts().CUDA || getLangOpts().OpenMPIsDevice ||
        getLangOpts().SYCLIsDevice) &&
       !getFrontendOpts().AuxTriple.empty()) {
     auto TO = std::make_shared<TargetOptions>();
     TO->Triple = llvm::Triple::normalize(getFrontendOpts().AuxTriple);
     if (getFrontendOpts().AuxTargetCPU)
       TO->CPU = getFrontendOpts().AuxTargetCPU.getValue();
     if (getFrontendOpts().AuxTargetFeatures)
       TO->FeaturesAsWritten = getFrontendOpts().AuxTargetFeatures.getValue();