1 files changed, 11 insertions, 0 deletions
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
index d361f51bca1..6643c1ae7eb 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
@@ -130,6 +130,12 @@ static cl::opt<bool> EnableLibCallSimplify(
   cl::init(true),
   cl::Hidden);
 
+static cl::opt<bool> EnableLowerKernelArguments(
+  "amdgpu-ir-lower-kernel-arguments",
+  cl::desc("Lower kernel argument loads in IR pass"),
+  cl::init(true),
+  cl::Hidden);
+
 extern "C" void LLVMInitializeAMDGPUTarget() {
   // Register the target
   RegisterTargetMachine<R600TargetMachine> X(getTheAMDGPUTarget());
@@ -155,6 +161,7 @@ extern "C" void LLVMInitializeAMDGPUTarget() {
   initializeAMDGPUAnnotateKernelFeaturesPass(*PR);
   initializeAMDGPUAnnotateUniformValuesPass(*PR);
   initializeAMDGPUArgumentUsageInfoPass(*PR);
+  initializeAMDGPULowerKernelArgumentsPass(*PR);
   initializeAMDGPULowerKernelAttributesPass(*PR);
   initializeAMDGPULowerIntrinsicsPass(*PR);
   initializeAMDGPUOpenCLEnqueuedBlockLoweringPass(*PR);
@@ -669,6 +676,10 @@ void AMDGPUPassConfig::addIRPasses() {
 }
 
 void AMDGPUPassConfig::addCodeGenPrepare() {
+  if (TM->getTargetTriple().getArch() == Triple::amdgcn &&
+      EnableLowerKernelArguments)
+    addPass(createAMDGPULowerKernelArgumentsPass());
+
   TargetPassConfig::addCodeGenPrepare();
 
   if (EnableLoadStoreVectorizer)